Summarize Research Articles on template attached

Thornton et al. BMC Health Services Research (2017) 17:361 DOI 10.1186/s12913-017-2307-z RESEARCH ARTICLE Open Access Influences on patient satisfaction in healthcare centers: a semi-quantitative study over 5 years Ruth D. Thornton1, Nicole Nurse2, Laura Snavely3, Stacey Hackett-Zahler4, Kenice Frank5 and Robert A. DiTomasso1* Abstract Background: Knowledge of ambulatory patients’ satisfaction with clinic visits help improve communication and delivery of healthcare. The goal was to examine patient satisfaction in a primary care setting, identify how selected patient and physician setting and characteristics affected satisfaction, and determine if feedback provided to medical directors over time impacted patient satisfaction. Methods: A three-phase, semi-quantitative analysis was performed using anonymous, validated patient satisfaction surveys collected from 889 ambulatory outpatients in 6 healthcare centers over 5-years. Patients’ responses to 21 questions were analyzed by principal components varimax rotated factor analysis. Three classifiable components emerged: Satisfaction with Physician, Availability/Convenience, and Orderly/Time. To study the effects of several independent variables (location of clinics, patients’ and physicians’ age, education level and duration at the clinic), data were subjected to multivariate analysis of variance (MANOVA).. Results: Changes in the healthcare centers over time were not significantly related to patient satisfaction. However, location of the center did affect satisfaction. Urban patients were more satisfied with their physicians than rural, and inner city patients were less satisfied than urban or rural on Availability/Convenience and less satisfied than urban patients on Orderly/Time. How long a patient attended a center most affected satisfaction, with patients attending >10 years more satisfied in all three components than those attending 60 years old. Patients were significantly more satisfied with their 30–40 year-old physicians compared with those over 60. On Orderly/Time, patients were more satisfied with physicians who were in their 50′s than physicians >60. Conclusions: Improvement in patient satisfaction includes a need for immediate, specific feedback. Although Medical Directors received feedback yearly, we found no significant changes in patient satisfaction over time. Our results suggest that, to increase satisfaction, patients with lower education, those who are sicker, and those who are new to the center likely would benefit from additional high quality interactions with their physicians. Keywords: Patient satisfaction, Health care delivery, Community health * Correspondence: RobertD@pcom.edu 1 Department of Psychology, Philadelphia College of Osteopathic Medicine, 4170 City Ave., Philadelphia, PA 19131, USA Full list of author information is available at the end of the article © The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Thornton et al. BMC Health Services Research (2017) 17:361 Background Patient satisfaction surveys are often used to understand patients’ concerns and determine areas for improvement, including improving communication between physicians and patients. Survey results document progress and allow physicians and staff to maintain high standards. Although results of patient satisfaction surveys are used by payer systems to profile individual physicians and guide physician compensation, one study showed that < 25% of primary care physicians found profiles useful for improving patient care and fewer used the profiles to change [1]. Improvements are more likely to occur if staff receives more immediate feedback [2]. Data collection methods play a role in outcomes. Onsite surveys provide an immediate outlet for patients who are experiencing problems, although higher ratings for on-site surveys may also relate directly to doctorpatient communication. Surveys administered later after a clinic visit may yield lower ratings, possibly due to the course of treatment [2, 3]. Many factors influence patient satisfaction. Patient demographics such as age, gender, income, socioeconomic and general health status impact patients’ responses [3, 4]. Characteristics of the medical provider, including demographics and experience, also affect their interactions with patients [5–9]. Other factors include the type of setting the patient is in [10] and the amount of time patients had to wait [11]. However, Anderson found that complaints about wait time can be moderated by time spent with the physician [12]. Physician characteristics extend beyond the obvious. Physician-patient concordance in race, gender or age may be important in patient satisfaction, but many other factors such as primary language, parental status, sexual orientation, values, beliefs, or communication style may be associated [13, 14]. How long the patient has been with this physician and the degree to which the physicians’ communication is patientcentered are significant [13]. A physician’s experience plays a role, with lowest patient satisfaction with firstyear residents; interestingly, residents with some more Page 2 of 9 experience attained similar satisfaction ratings to those of the faculty attendings, suggesting that the requisite skills are acquired during the first year of training [7]. Whether to administer patient satisfaction surveys depends on the overall goals of the medical facility and on physician buy-in to change [1, 15]. The views of the medical director and administrator are key components as to whether the surveys are taken seriously and acted upon by physicians [16]. Patient satisfaction can become a success criterion of the center when physicians and staff meet regularly to discuss improvements in a context of cooperation and mutual support. Methods We initiated this study of patient satisfaction to help physicians better understand their patients at the healthcare centers (HCCs) of a not-for-profit medical school’s outpatient primary care centers on the east coast. Physicians were provided raw data and results of open-ended questions very soon after each year’s study. However, we decided to statistically analyze the overall data in order to understand where patients were most and least satisfied and what influenced their satisfaction. Our goal was to provide information which could help focus physician directors’ changes to improve patient satisfaction. The research was under the auspices of a medical college (Philadelphia College of Osteopathic Medicine, PCOM) which owns and operates five outpatient HCCs, four of which are located within the city limits of Philadelphia and the fifth HCC located in a rural area. [17] Two within Philadelphia are considered urban, while two are in the inner city [18]. An additional nonaffiliated, inner city HCC located within Philadelphia was also used in the research. We considered the nonaffiliated HCC as a control, but expected it to likely agree with data from the affiliated inner city HCCs. The quantity of surveys administered are listed in Table 1. This research arose from a need to quickly and inexpensively conduct patient satisfaction surveys in the Healthcare Centers, incorporating a research component Table 1 Numbers of patients surveyed from each Healthcare center during year 1 (Fall, 2005), year 2 (Summer, 2007), and year 3 (Summer 2010) HCC Location Year 1 # surveyed Year 2 # surveyed Year 3 # surveyed TOTAL surveyed 1 Inner City (PCOM) 40 68 90 198 2 Urban (PCOM) 34 54 69 157 3 Inner City (PCOM) 21 43 70 134 4 Rural (PCOM) 30 19 25 74 5 Urban (PCOM) 25 51 45 121 6 Inner City (non-PCOM) TOTALS 50 75 80 205 200 310 379 889 Thornton et al. BMC Health Services Research (2017) 17:361 involving graduate students interested in health related careers. Surveys were administered to patients at the five HCCs. Patient questions were adapted from the validated DiTomasso-Willard Patient Satisfaction Questionnaire [19] (questions are listed in Table 2). Demographic information and responses to open-ended questions were also collected. In 2005 (year 1), 2007 (year 2), and 2010 (year 3), students in a master’s program at the medical school approached patients in the waiting areas at each HCC asking them to complete a survey. Patients could take the surveys with them into the examination room, but they returned the survey before leaving the HCC. If requested, the student helped a patient read the questions. Each surveying period was conducted over an approximately one month of time. Students varied their sampling by time of day and day of week. Therefore, the sample was comprised of a random representation of patients attending each HCC during each one-month period of surveying. The students approached anyone who was in the waiting room during sampling times, but Page 3 of 9 patients were free to refuse if they wished. The goal was to obtain approximately 10% of the average number of patients seen by each HCC in a month. The protocol (Protocol #H05-022X) was approved by the Institutional Review Board (IRB) of PCOM that determined it to be exempt from informed consent requirements under 45 CFR 46.101(b)(2)–survey research in which the responses will be recorded in such a manner that the human subjects cannot be identified, directly or through identifiers linked to the subjects (e.g., name, Social Security number). Further, no master list existed linking such identifiers to the subjects. Approximately 5–15% of the average numbers of patients coming to each HCC in a month were surveyed. Inclusion criteria included patients willing to respond, patient age of at least 18 years, and patients who spoke English. Patients were assured the questionnaire was confidential without any identifying information, the results would be presented in aggregate form, and that their responses would not affect their specific care at the HCC. In order to maintain anonymity, a patient’s medical status was Table 2 Grouping of the 21 survey questions using factor analysis, Rotated Component Matrix Component: Question: 1 2 3 Q1. During a typical visit, my doctor spends enough time explaining my medical condition to me. 0.773 0.151 0.162 Q2. My doctor gives me the best quality of care. 0.869 0.190 0.133 Q3. I would recommend my doctor to friends. 0.827 0.191 0.112 Q4. The staff are helpful to the patients. 0.311 0.562 0.127 Q5. My doctor uses technical terms that confuse me. 0.109 −0.200 0.628 Q6. My doctor is available when I need him/her.b 0.412 0.491 0.076 Q7. The waiting room time is too long. −0.084 0.402 0.442 Q8. My doctor really follows through. 0.751 0.230 0.082 Q9. I plan to return to this center for care. 0.713 0.370 0.148 Q10. It’s easy to get an appointment when I need one. 0.223 0.665 0.122 Q11. My doctor wastes time talking about things that don’t really matter to me.a 0.271 −0.014 0.702 Q12. My doctor treats the “whole” person. 0.640 0.300 0.143 Q13. The staff accommodates my needs over the phone. 0.241 0.677 0.070 Q14. I am satisfied with the quality of the medical care I receive here. 0.724 0.398 0.171 a b Q15. I receive prompt attention while waiting in this facility. 0.285 0.658 0.132 Q16. I have to tell my story several times before getting an appointment.a 0.001 0.409 0.631 Q17. I am treated the same as other people who get care here. 0.366 0.511 0.105 Q18. Check-out time at the front desk is too time-consuming.a −0.030 0.338 0.648 Q19. I would not recommend this center to a friend. 0.250 0.101 0.530 Q20. Everything seems so confusing at this center.a 0.199 0.160 0.731 Q21. When I’m sick I can get an appointment pretty quickly. 0.229 0.712 0.057 a Component 1: Satisfaction with Doctor (Questions 1, 2, 3, 8, 9, 12, 14) Component 2: Availability/Convenience (Questions 4, 10, 13, 15, 17, 20) Component 3: Orderly/Time (Questions 5, 11, 16, 18, 19, 20) a Questions worded in the negative were reversed for statistical analysis b Question not classified by component Thornton et al. BMC Health Services Research (2017) 17:361 not requested, although in retrospect, it may have been helpful. From observation, students reported that those with acute medical issues were less inclined to participate. Although an absolute count was not performed, students who administered surveys consistently estimated that only about 5% of the patients in the waiting room refused to participate. Survey results were entered into IBM’s Statistical Package for the Social Sciences (SPSS 18.0) for analysis. Missing data were filled in using Linear Interpolation, and any negative questions were transformed to the positive on the Likert scale, so that, for all questions, 5 (strongly agree) meant “most satisfied.” All 21 survey statements were subjected to a principal components varimax rotated factor analysis according to Kaiser’s criterion [20] which ultimately allowed for a reduction of statements into three classifiable components, Satisfaction with Physician, Availability/Convenience, and Orderly/Time (Table 2). Following each survey period, the data were analyzed in SPSS to collapse the questions into three classifiable components/categories. These three categories did not vary during the 3 data collection periods. After each survey period, study staff attended face-to-face meetings with Medical Directors of each healthcare center, the Dean of the Medical School, and the Chair of Family Medicine to present the results. HCC staff were provided with mean scores for each question for their HCC compared with a composite of all HCC’s. They also received the data collapsed into the three categories for their HCC compared with a composite of all HCC’s, but without statistical analysis. For analysis of the composite data, multivariate analysis of variance (MANOVA) was performed for groups of data, using post hoc Tukey to distinguish specific significance between groups. Independent t-test was used for gender analysis, and Chi square analysis was done to compare the observed gender data from patients who completed surveys with patient demographics of each HCC. See Additional Data for more specific information. In using factor analysis, it is common practice to require 10 subjects per number of items. In the present case, this criterion was far exceeded. For the separate MANOVA analyses using 3 dependent variables, setting power at 95% for a medium effect size at the 0.05 level of significance comparing 2 levels (male vs. female) of the independent variable, 3 levels (3 locations) and 5 levels (physician age groups), the required number of subjects was 280, 171, and 145 respectively. In all cases there was sufficient power. Results Surveys were administered to a total of 889 patients who visited one of the HCCs for treatment (Table 1). These Page 4 of 9 numbers represented between 5–15% of the average number of patients seen monthly in the affiliated HCCs, and comparable numbers of surveys were obtained from the much larger, non-affiliated HCC. Applying principal components varimax rotated factor analysis to the survey responses resulted in groups of identifiable questions that constituted factors (Rotated component matrix for all questions is shown on Table 2). Three classifiable factors, Satisfaction with Physician, Availability/Convenience, and Orderly/Time, emerged from the analysis and are used throughout this research. Two questions (Q6 and Q7) were not included as the items did not load on any of the factors (Table 2). Using the survey questions that constituted each factor (Table 2), the three factors have the following characteristics: Satisfaction with Physician involves being satisfied with the quality of medical care received, as well as the physician spending enough time with the patient. Availability/Convenience involves being satisfied with the staff and their helpfulness in making appointments, whether in person or by phone. Orderly/Time has to do with patients’ time being respected, and interactions with staff and physicians being clear and to the point, avoiding confusion. Overall, patients were quite satisfied with their HCCs, as evidenced by overall mean scores greater than 3.89 on a Likert scale of 1–5 (see Additional file 1: Table S3A). Mean scores were highest in Satisfaction with Physician (4.27 ± 0.65), while Availability/Convenience (3.92 ± 0.69) and Orderly/Time (3.89 ± 0.66) were somewhat lower. Even so, a score of 3.9 represents the top 20–25% of satisfaction. The open-ended responses emphasized the importance of patients’ satisfaction with their physician, even if patients were somewhat less satisfied with other aspects of their visit (see Additional file 2: Table S6). The goal of this research was to identify areas found to be statistically significant. More complete data can be found in the Additional files 1, 2, 3,and 4. Based on MANOVA, there was no significance over time in any of the three categories (see Additional file 1: Table S3B). This points to a consistency over time in the operations and functioning of these HCC’s. The following areas were found to be statistically significant by MANOVA: Analyzing satisfaction in inner city, urban and rural HCCs (Fig. 1), significance was observed in the following area.: Patients in inner city HCCs were less satisfied than those in urban or rural HCC’s on Availability/Convenience, and those in inner city HCCs were less satisfied than urban patients in the area of Orderly/Time. Urban patients were more satisfied with their Physician than were rural patients while inner city patients’ satisfaction with Thornton et al. BMC Health Services Research (2017) 17:361 Fig. 1 Satisfaction by location (Inner City, Urban and Rural). Lines/ Brackets indicate comparisons by color that were significantly different in each of the categories their Physician was not significantly different from the other localities (See Additional file 1: Table S3C, for more detail). When individual HCCs were analyzed (Fig. 2), one urban HCC (#5) had significantly higher satisfaction with their Physician than the other urban HCC (#2) or one inner city HCC (#6). The other urban HCC (#2) had more satisfaction in the category of Orderly/Time than two of the three inner city HCCs (#3 and #6). Two inner city HCCs (#1 and #6) had significantly lower satisfaction in the category of Availability/Convenience than the rural HCC (#4). (See Additional file 1: Table S3D, for details.) Patients’ demographics appear to play a role in the level of satisfaction. Patients over 60 years old were more satisfied with the Availability/Convenience of the HCC than patients who were in their 40′s (Fig. 3). Those with more education (in the range from graduating high school through graduate Fig. 2 Satisfaction by individual HCCs. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories Page 5 of 9 school) were more satisfied with the Orderly/Time category than those with less than a high school diploma (Fig. 4). Finally, patients who had been with their HCC for longer periods of time were more satisfied than those who had been there less than 5 years in all three categories of satisfaction with Physician, Availability/Convenience, and Orderly/ Time (Fig. 5) (See Additional file 2: Table S4C, for details). Physicians in these centers tended to longevity in their positions. Patients were more satisfied with their Physicians who were in their 30′s and 40′s than with physicians in their 50′s (Fig. 6). Also, physicians in their 50′s were perceived to be more Available than those in their 60′s. Patients rated male physicians as more Available than female physicians, and in the Inner City HCCs, patients rated their Caucasian physicians higher on Availability than African American physicians (see Additional file 3: Table S5B and C). Open ended responses were overall very positive, with the exception of the rural HCC4 during year 1. After personnel replacements at this HCC, more positive responses were also seen there. Wait times were seen as a problem in some HCCs, particularly in the inner city centers. The majority of patients were very satisfied with the convenience of their HCC (See Additional file 4: Table S6). Discussion In the examination of changes over time, patient satisfaction at the HCCs in the study remained overall quite high in all three categories of Satisfaction with Physician, Availability/Convenience, and Orderly/Time. Meanwhile, notable changes at the affiliated centers during this time period included a new telephone system installed between years 1–2, major renovations of one of the centers in year 2, and installation of a system of Electronic Fig. 3 The effect of patient age on satisfaction. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories Thornton et al. BMC Health Services Research (2017) 17:361 Fig. 4 The effect of patient education on satisfaction. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories Medical Records (EMR) in all affiliated HCCs between years 2 and 3; the non-affiliated HCC #6 also introduced EMR prior to year 3. While telephone changes would likely affect staff-patient interactions, instituting EMR represented a major change in the physician-patient interactions, with the addition of computers to each examination room. We were surprised that these seemingly “major” changes did not significantly affect the satisfaction levels over this time period. De Leon et al. found generally higher patient satisfaction with a center after EMR were introduced [21], while we found no significant differences after EMR was installed. Results of the patient satisfaction surveys were presented to Medical Directors and staff in a timely manner after each survey period, but without statistical analysis. From the initial data given to each HCC, medical staff could compare their mean results with a composite mean result for all the centers. However, they did not Fig. 5 The effect of length of time at a HCC on satisfaction. Lines/ Brackets indicate comparisons by color that were significantly different in each of the categories Page 6 of 9 Fig. 6 The effect of physician’s age on satisfaction. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories have access to comparisons of individual HCCs (see Additional file 1, Table S3D). Nor did they have access to figures such as Fig. 2, comparing individual HCCs. It is not surprising that each HCC is unique. An example is HCC 5 with a significantly higher level of satisfaction with Physician compared with two other HCCs, one urban and inner city (Fig. 2, and Additional file 1: Table S3D). This merits more in-depth analysis of the physician practices at this outstanding urban HCC as a positive example for others. We projected that HCCs sharing similar locations (inner city, urban or rural) would be more alike and this proved to be the case. In the components of Availability/ Convenience and Orderly/Time, patients in the inner city HCCs were less satisfied than those in urban or rural settings, consistent with findings of the individual HCCs. There could be several reasons for differences between inner city and other HCCs. Fan et al. found that functional status (disease severity, physical limitation) was only weakly associated with general satisfaction, while education, coping skills and disease perception were more important to patient’s satisfaction [4]. Patients in the inner city may be sicker due to overall inadequate health knowledge or reluctance to visit a doctor, possibly due to lack of insurance. These findings suggest that physician-patient interactions with the goal of improved disease understanding might help as much as actual improvement in health. We did not ask for the health status of individual patients in our survey, so we can only guess the health status of patients at different locations. Comparing locations (Fig. 1) with individual HCCs (Fig. 2) reveals the sources of these differences. For example, in Fig. 1, inner city patients were statistically less satisfied in the component of Orderly/Time than were patients in urban settings. Fig. 2 shows that the differences were primarily with one urban HCC #2 (but not with urban HCC #5), compared only with 2 inner city Thornton et al. BMC Health Services Research (2017) 17:361 HCC #3 and #6 (but not with inner city HCC #1). So generalizations require examining the individual HCCs as well. Education level of the patient can also be reflective of location. Approximately 80% of inner city respondents reported having high school education or less, similar to rural patients (76%), while only 58% of urban patients had a high school education or less (data not shown). Other issues facing patients, such as availability of public transportation, may be more of an obstacle in the inner city than in either urban or rural settings. While public transportation is also not widely available in rural settings, it is likely most patients have access to a vehicle. Inner city respondents also were less satisfied in the component of Orderly/Time than respondents in urban settings, and this is confirmed in the open-ended questions (see Additional file 4: Table S6) where a larger number of patients specifically mentioned the wait time as a problem in the inner city HCCs than in the urban or rural HCCs. Although we wondered if dissatisfaction with wait time could be directly attributable to student participation in the examination room, that seems not to be the case, as a very small percent of responders mentioned students in the open-ended questions and half of those were positive. Mol et al. found that patients generally felt neutral or positive about the presence of students, and in that study, between 83 and 98% of patients consented to student participation [22]. Our only finding of differences associated with education level in satisfaction were in the area of Orderly/ Time; patients with less than a high school education were less satisfied in the component of Orderly/Time than any other group. This could be due to their inability to understand the medical parlance or the protocols involved in their care. However, one study also found that the converse-a physician’s satisfaction with a patient-was associated with their patients’ higher education level [23], suggesting that the responsibility may be reciprocal between the physician and the patient. Another patient demographic of age can also contribute to patient satisfaction. Our finding that patients over 60 years old had a higher degree of satisfaction in Availability/Convenience is not surprising. This finding agrees with Jackson who reported that patients over 65 years old and with higher functional status were more satisfied [3]. Peck found that physicians were more likely to have patient-centered encounters with patients over age 65, which in turn meant that older patients were more satisfied [24]. Although there was no impact of patients’ gender on level of satisfaction, we did find that, in general, more female patients agreed to fill out the surveys than were actually represented as patients in the HCCs. Not surprisingly, the most significant differences were found in the length of time a patient had been attending their Page 7 of 9 HCC. This is undoubtedly a self-selection, where either the physician or the location suits the patient who continues to visit that center. Pelletier calls this “sampling bias,” citing that “those who stay with a program…may be those who are most satisfied” [23]. Another explanation is through “visit continuity,” where respondents rated the quality of physician-patient interaction as being more important during the early stages of continuity or when the patient reported worse self-rated health [25]. This suggests that physicians who focus on those newer patients or sicker patients who would benefit the most from additional interactions may have the most positive results over time. Demographics of the physician may also be important to patient satisfaction. The physicians at the affiliated HCCs were all osteopathic (DO) physicians, who self-reported that they used Osteopathic Manipulative Treatment at their clinics about 20% of the time. In the open-ended questions, some patients did express a preference for DO physicians. On age of physicians, it appears that more patients prefer a physician younger than 50 years old in the component of Satisfaction with Physician, but in Orderly/ Time, they prefer a physician in their 50′s rather than in their 60′s. We speculate that physicians in their 50′s are likely to be at the pinnacle of their profession, although other considerations may also be important, such as humor or degree of connection that the patient perceives with that physician. In the variable of Orderly/Time, it is possible that physicians in their 50′s may be more efficient, having a well-run visit, while the slower, possibly more thorough pace of older physicians may not be as appreciated. The statistical significances found in this data enhance the details which were presented to the medical directors after each surveying period and provide additional measures of patient satisfaction. Presenting the data to medical directors in figure form rather than as graphs is likely to enhance understanding. Finally, presenting the data of each individual HCC rather than as a composite may help medical directors to see the larger picture. The present study has several limitations: In retrospect from patients’ written responses, an additional choice under the education demographic would have better captured any additional education received, such as technical certificates or Associate degrees. Also, the severity of the patient’s medical condition should have been noted, as this has been shown to influence patient satisfaction [4]. In addition, the questions that fell under the component Orderly/Time in the factor analysis fortuitously contained all questions which had been originally stated in the negative and then were reversed for analysis. Finally, presentation of the data to the medical directors in a timely fashion could be improved by presenting figures in addition to tables, and showing results of each individual HCC. Thornton et al. BMC Health Services Research (2017) 17:361 Conclusions This study was designed to provide feedback to Medical Directors on patient satisfaction in their HCCs. Our findings point to a consistency in the operations and functioning of these HCCs over time, even when renovations or installation of EMR were performed. Differences in locality (inner city, urban, rural) were found, as well as differences in satisfaction by patient demographics (age, education level, length of time with a HCC) and by physician demographics (age, gender). However, uniqueness of individual HCCs contributes to these differences. Physicians from each HCC regularly meet together, and they can use these meetings to help better understand and build on their strengths and individuality. Results of this study can be used to increase satisfaction if physicians help their patients benefit from their services and increase their satisfaction. Particularly, physicians can concentrate on providing additional high-quality interactions for patients with less education, those who are sicker, and those who are new to the HCC. Additional files Additional file 1: Table S3. Comparisons overall and by time, location, individual HCCs vs. 3 components. (DOC 41 kb) Additional file 2: Table S4. Patient demographics vs. three factors. *refers to higher mean score; ns, not significant. (DOC 44 kb) Additional file 3: Table S5. Physician demographics vs. three factors. *refers to higher mean score; ns, not significant. (DOC 40 kb) Additional file 4: Table S6. Open-ended questions by healthcare center and year. (+) refers to positive statements, what did you like most? (−) refers to negative statements, what did you like least? (DOC 48 kb) Abbreviations DO: Doctor of Osteopathic Medicine; EMR: Electronic Medical Records; HCCs: Healthcare Centers; IRB: Institutional Review Board; MANOVA: Multivariate analysis of variants; PCOM: Philadelphia College of Osteopathic Medicine; SPSS: Statistical Package for the Social Sciences Acknowledgements The authors wish to acknowledge Audrey Rossowski, M.S. for collecting surveys at the healthcare centers and Barbara A. Mitchell, Ph.D. for helpful editing. We also sincerely thank PCOM Kenneth Veit, DO, MBA, FACOFP, Provost, Senior Vice President for Academic Affairs and Dean. We also thank PCOM Family Medicine Chair and Director Harry Morris, DO; PCOM Healthcare Center Physicians and Medical Directors, Oliver Bullock, DO., Michael Becker, DO, Izola David, DO, Larry Finklestein, DO, David Kuo, DO, Marta Motel, DO, Barbara Williams-Page, DO, and David Wood, DO; and A. Scott McNeal, DO, Fairmount Primary Care Center, for allowing our participation. An abstract of this research was presented by Kenice Frank, M.S., at the Student National Medical Association’s National Meeting, May, 2012, Atlanta, GA. Funding No funding sources were used for this research project. Availability of data and materials The SPSS data and subsequent analysis is available from RDT and the corresponding author, RAD, on reasonable request. A large part of the data is already shown in Additional files 1, 2, 3, and 4, associated with this publication. Page 8 of 9 Authors’ contributions RDT conceived of the research, wrote the paper, and, with the help of RAD, analyzed the data. NN received her M.S. degree from PCOM for her research on the first year (2005) of the study. She gathered the data, analyzed it, and presented a thesis, “A measure of patient satisfaction of PCOM healthcare centers based upon geographic settings.” LS received her M.S. degree from PCOM for her research on the first year (2005) of the study. She gathered the data, analyzed it, and presented a thesis, “The effects of demographic factors on patient satisfaction in an ambulatory setting.” SH-Z received her M.S. degree from PCOM for her research on the second year (2007) of the study. She gathered the data for 2007, analyzed the 2007 data and compared it with 2005 data, and presented a thesis, “Examining improvement levels in measured patient satisfaction in an academic primary care setting.” KF gathered the data on the third year (2010) of the study. She analyzed the 2010 data, compared it with the 2005 and 2007 data, and presented a poster on the study at the Student National Medical Association in 2011. RAD gave guidance to the entire study, providing the original survey questions [17], analyzing statistically all of the data, serving on each M.S. student’s thesis committee, and editing the paper. All authors read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Consent for publication Not applicable. Ethics approval and consent to participate The protocol was approved by the Institutional Review Board (IRB) at Philadelphia College of Osteopathic Medicine that determined it to be exempt from informed consent requirements under 45 CFR 46.101(b)(2)survey research in which the responses will be recorded in such a manner that the human subjects cannot be identified, directly or through identifiers linked to the subjects (e.g., name, Social Security number). Further, no master list existed linking such identifiers to the subjects. Inclusion criteria included patients willing to respond, patient age of at least 18 years, and patients who spoke English. Patients were assured the questionnaire was confidential without any identifying information, the results would be presented in aggregate form, and that their responses would not affect their specific care at the HCC. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 Department of Psychology, Philadelphia College of Osteopathic Medicine, 4170 City Ave., Philadelphia, PA 19131, USA. 2North Philadelphia Health System/St. Joseph’s Hospital, Philadelphia, PA, USA. 3Geisinger Medical Center, Danville, PA, USA. 4Maria Fareri Children’s Hospital, Westchester, NY, USA. 5College of Podiatric Medicine & Surgery, Des Moines University, Des Moines, IO, USA. Received: 3 April 2017 Accepted: 11 May 2017 References 1. Rider EA, Perrin JM. Performance profiles: the influence of patient satisfaction data on physicians’ practice. Pediatrics. 2002;109(5):752–7. 2. Burroughs TE, Waterman BM, Gilin D, Adams D, McCollegan J, Cira J. Do onsite patient satisfaction surveys bias results? Jt Comm J Qual Pt Safety. 2005; 31(3):158–66. 3. Jackson JL, Chamberlin J, Kroenke K. Predictors of patient satisfaction. Soc Sci Med. 2001;52(4):609–20. 4. Fan VS, Reiber GE, Diehr P, Burman M, McDonell MB, Fihn SD. Functional status and patient satisfaction: a comparison of ischemic heart disease, obstructive lung disease, and diabetes mellitus. J Gen Intern Med. 2005; 20(5):452–9. 5. Fan VS, Burman M, McDonell MB, Fihn SD. Continuity of care and other determinants of patient satisfaction with primary care. J Gen Intern Med. 2005;20(3):226–33. Thornton et al. BMC Health Services Research (2017) 17:361 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Page 9 of 9 Fairchild DG, McLoughlin KS, Gharib S, Horsky J, Portnow M, Richter J, Gagliano N, Bates DW. Productivity, quality, and patient satisfaction: comparison of part-time and full-time primary care physicians. J Gen Intern Med. 2001;16(10):663–7. Monk SM, Nanagas MT, Fitch JL, Stolfi A, Pickoff AS. Comparison of resident and faculty patient satisfaction surveys in a pediatric ambulatory clinic. Teach Learn Med. 2006;18(4):343–7. Rodriguez HP, von Glahn T, Chang H, Rogers WH, Safran DG. Measuring patients’ experiences with individual specialist physicians and their practices. Am J Med Qual. 2009;24(1):35–44. Ware Jr JE, Davies AR. Behavioral consequences of consumer dissatisfaction with medical care. Eval Prog Plan. 1983;6(3–4):291–7. Perneger TV, Etter JF, Raetzo MA, Schaller P, Stalder H. Comparison of patient satisfaction with ambulatory visits in competing health care delivery settings in Geneva, Switzerland. J Epidemiol Comm Health. 1996;50(4):463–8. Feddock CA, Hoellein AR, Griffith 3rd CH, Wilson JF, Bowerman JL, Becker NS, Caudill TS. Can physicians improve patient satisfaction with long waiting times? Eval Health Prof. 2005;28(1):40–52. Anderson RT, Camacho FT, Balkrishnan R. Willing to wait?: the influence of patient wait time on satisfaction with primary care. BMC Health Serv Res. 2007;7:31. Street Jr RL, O’Malley KJ, Cooper LA, Haidet P. Understanding concordance in patient-physician relationships: personal and ethnic dimensions of shared identity. Ann Fam Med. 2008;6(3):198–205. Thornton RLJ, Powe NR, Roter D, Cooper LA. Patient-physician social concordance, medical visit communication and patients’ perceptions of health care quality. Pat Educ Couns. 2011;85:e201–8. Gooding TD, Newcomb L, Mertens K. Patient-centered measurement at an academic medical center. Jt Comm J Qual Improv. 1999;25(7):343–51. Conner DR. Managing at the Speed of Change. 1st ed. Ed Anonymous New York: Villard Books; 1993. p. 105–24. US Dept of Agriculture. Econ Res Service. Rural Classification. http://www. ers.usda.gov/topics/rural-economy-population/rural-classifications/what-isrural.aspx. Accessed 2016. Merriam-Webster Dictionary. Urban & Inner City. http://www.merriamwebster.com/dictionary. Accessed 2016. DiTomasso RA, Willard M. The development of a patient satisfaction questionnaire in the ambulatory setting. Fam Med. 1991;23(2):127–31. Field A. Discovering statistics using SPSS. 2nd ed. 2005. De Leon DF, Silfen SL, Wang JJ, Kamara TS, Wu WY, Shih SC. Patient experiences at primary care practices using electronic health records. J Med Pract Manage. 2012;28(3):169–76. Mol SSL, Peelen JH, Kuyvenhoven MM. Patients’ view on student participation in general practice consultations: A comprehensive review. Med Teacher. 2011;33:e397–400. Pelletier M. Client satisfaction surveys: variables to watch out for. Dimens Health Serv. 1985;62(1):37–9. Peck BM. Age-related differences in doctor-patient interaction and patient satisfaction. Curr Gerontol Geriatr Res. 2011;2011:137492–501. Rodriguez HP, Rogers WH, Marshall RE, Safran DG. The effects of primary care physician visit continuity on patients’ experiences with care. J Gen Intern Med. 2007;22(6):787–93. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit 2 3 Quick Guide to Bivariate Statistical Tests 4 5 6 Acquisitions Editor: Christina Burns Product Development Editor: Katherine Burland Editorial Assistant: Cassie Berube Marketing Manager: Dean Karampelas Production Project Manager: Cynthia Rudy Design Coordinator: Joan Wendt Manufacturing Coordinator: Karin Duffield Prepress Vendor: Absolute Service, Inc. Tenth edition Copyright © 2017 Wolters Kluwer. Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins. Copyright © 2008, 2004, 1999 by Lippincott Williams & Wilkins. Copyright © 1995, 1991, 1987, 1983, 1978 by J. B. Lippincott Company. All rights reserved. This book is protected by copyright. No part of this book may be reproduced or transmitted in any form or by any means, including as photocopies or scanned-in or other electronic copies, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotations embodied in critical articles and reviews. Materials appearing in this book prepared by individuals as part of their official duties as U.S. government employees are not covered by the above-mentioned copyright. To request permission, please contact Wolters Kluwer at Two Commerce Square, 2001 Market Street, Philadelphia, PA 19103, via email at permissions@lww.com, or via our website at lww.com (products and services). 987654321 Printed in China Library of Congress Cataloging-in-Publication Data Polit, Denise F., author. Nursing research : generating and assessing evidence for nursing practice / Denise F. Polit, Cheryl Tatano Beck. — Tenth edition. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4963-0023-2 I. Beck, Cheryl Tatano, author. II. Title. [DNLM: 1. Nursing Research—methods. WY 20.5] RT81.5 610.73072—dc23 7 2015033543 This work is provided “as is,” and the publisher disclaims any and all warranties, express or implied, including any warranties as to accuracy, comprehensiveness, or currency of the content of this work. This work is no substitute for individual patient assessment based on healthcare professionals’ examination of each patient and consideration of, among other things, age, weight, gender, current or prior medical conditions, medication history, laboratory data, and other factors unique to the patient. The publisher does not provide medical advice or guidance, and this work is merely a reference tool. Healthcare professionals, and not the publisher, are solely responsible for the use of this work including all medical judgments and for any resulting diagnosis and treatments. Given continuous, rapid advances in medical science and health information, independent professional verification of medical diagnoses, indications, appropriate pharmaceutical selections and dosages, and treatment options should be made and healthcare professionals should consult a variety of sources. When prescribing medication, healthcare professionals are advised to consult the product information sheet (the manufacturer’s package insert) accompanying each drug to verify, among other things, conditions of use, warnings, and side effects and identify any changes in dosage schedule or contraindications, particularly if the medication to be administered is new, infrequently used, or has a narrow therapeutic range. To the maximum extent permitted under applicable law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property, as a matter of products liability, negligence law or otherwise, or from any reference to or use by any person of this work. LWW.com 8 TO Our Beloved Family: Our Husbands, Our Children (Spouses/Fiancés), and Our Grandchildren Husbands: Alan Janosy and Chuck Beck Children: Alex (Maryanna), Alaine (Jeff), Lauren (Vadim), and Norah (Chris); and Curt and Lisa Grandchildren: Cormac, Julia, Maren, and Ronan 9 Acknowledgments This 10th edition, like the previous nine editions, depended on the contribution of dozens of people. Many faculty and students who used the text have made invaluable suggestions for its improvement, and to all of you we are very grateful. In addition to all those who assisted us during the past 35 years with the earlier editions, the following individuals deserve special mention. We would like to acknowledge the comments of reviewers of the previous edition of this book, anonymous to us initially, whose feedback influenced our revisions. Faculty at Griffith University in Australia made useful suggestions and also inspired the inclusion of some new content. Valori Banfi, reference librarian at the University of Connecticut, provided ongoing assistance. Dr. Deborah Dillon McDonald was extraordinarily generous in giving us access to her NINR grant application and related material for the Resource Manual. We also extend our thanks to those who helped to turn the manuscript into a finished product. The staff at Wolters Kluwer has been of great assistance to us over the years. We are indebted to Christina Burns, Kate Burland, Cynthia Rudy, and all the others behind the scenes for their fine contributions. Finally, we thank our family and friends. Our husbands Alan and Chuck have become accustomed to our demanding schedules, but we recognize that their support involves a lot of patience and many sacrifices. 10 Reviewers Ellise D. Adams, PhD, CNM 11 Associate Professor The University of Alabama in Huntsville Huntsville, Alabama Jennifer Bellot, PhD, RN, MHSA Associate Professor and Director, DNP Program Thomas Jefferson University Philadelphia, Pennsylvania Kathleen D. Black, PhD, RNC Assistant Professor, Jefferson College of Nursing Thomas Jefferson University Philadelphia, Pennsylvania Dee Campbell, PhD, APRN, NE-BC, CNL Professor, Graduate Department Felician College, School of Nursing Lodi, New Jersey Patricia Cannistraci, DNS, RN, CNE 12 Assistant Dean 13 Excelsior College Albany, New York Julie L. Daniels, DNP, CNM 14 Assistant Professor Frontier Nursing University Hyden, Kentucky Rebecca Fountain, PhD, RN 15 Associate Professor University of Texas at Tyler Tyler, Texas Teresa S. Johnson, PhD, RN Associate Professor, College of Nursing University of Wisconsin—Milwaukee Milwaukee, Wisconsin Jacqueline Jones, PhD, RN, FAAN Associate Professor, College of Nursing University of Colorado, Anschutz Medical Campus Aurora, Colorado Mary Lopez, PhD, RN Associate Dean, Research Western University of Health Sciences Pomona, California Audra Malone, DNP, FNP-BC 16 Assistant Professor Frontier Nursing University Hyden, Kentucky Sharon R. Rainer, PhD, CRNP Assistant Professor, Jefferson College of Nursing Thomas Jefferson University Philadelphia, Pennsylvania Maria A. Revell, PhD, RN 17 Professor of Nursing Middle Tennessee State University Murfreesboro, Tennessee Stephanie Vaughn, PhD, RN, CRRN Interim Director, School of Nursing California State University, Fullerton Fullerton, California 18 Preface Research methodology is not a static enterprise. Even after writing nine editions of this book, we continue to draw inspiration and new material from groundbreaking advances in research methods and in nurse researchers’ use of those methods. It is exciting and uplifting to share many of those advances in this new edition. We expect that many of the new methodologic and technologic advances will be translated into powerful evidence for nursing practice. Five years ago, we considered the ninth edition as a watershed edition of a classic textbook. We are persuaded, however, that this 10th edition is even better. We have retained many features that made this book a classic textbook and resource, including its focus on research as a support for evidence-based nursing, but have introduced important innovations that will help to shape the future of nursing research. N E W TO T H I S E D I T I O N New Chapters We have added two new chapters on “cutting-edge” topics that are not well covered in any major research methods textbook, regardless of discipline. The first is a chapter on an issue of critical importance to health professionals and yet inadequately addressed in the nursing literature: the clinical significance of research findings. In Chapter 20, we discuss various conceptualizations of clinical significance and present methods of operationalizing those conceptualizations so that clinical significance can be assessed at both the individual and group level. We believe that this is a “must-read” chapter for nurses whose research is designed to inform clinical practice. The second new chapter in this edition concerns the design and conduct of pilot studies. In recent years, experts have written at length about the poor quality of many pilot studies. Chapter 28 provides guidance on how to develop pilot study objectives and draw conclusions about the appropriate next step—that is, whether to proceed to a full-scale study, make major revisions, or 19 abandon the project. This chapter is included in Part 5 of this book, which is devoted to mixed methods research, because pilots can benefit from both qualitative and quantitative evidence. New Content Throughout the book, we have included material on methodologic innovations that have arisen in nursing, medicine, and the social sciences during the past 4 to 5 years. The many additions and changes are too numerous to describe here, but a few deserve special mention. In particular, we have totally revised the chapters on measurement (Chapter 14) and scale development (Chapter 15) to reflect emerging ideas about key measurement properties and the assessment of newly developed instruments. The inclusion of two new chapters made it challenging to keep the textbook to a manageable length. Our solution was to move some content in the ninth edition to supplements that are available online. In fact, every chapter has an online supplement, which gave us the opportunity to add a considerable amount of new content. For example, one supplement is devoted to evidence-based methods to recruit and retain study participants. Other supplements include a description of various randomization methods, an overview of item response theory, guidance on wording proposals to conduct pilot studies, and a discussion of quality improvement studies. Following is a complete list of the supplements for the 31 chapters of this textbook: 1. The History of Nursing Research 2. Evaluating Clinical Practice Guidelines—AGREE II 3. Deductive and Inductive Reasoning 4. Complex Relationships and Hypotheses 5. Literature Review Matrices 6. Prominent Conceptual Models of Nursing Used by Nurse Researchers, and a Guide to Middle-Range Theories 7. Historical Background on Unethical Research Conduct 8. Research Control 9. Randomization Strategies 10. The RE-AIM Framework 11. Other Specific Types of Research 12. Sample Recruitment and Retention 13. Other Types of Structured Self-Reports 14. Cross-Cultural Validity and the Adaptation/Translation of Measures 15. Overview of Item Response Theory 16. SPSS Analysis of Descriptive Statistics 17. SPSS Analysis of Inferential Statistics 18. SPSS Analysis and Multivariate Statistics 19. Some Preliminary Steps in Quantitative Analysis Using SPSS 20. Clinical Significance Assessment with the Jacobson-Truax Approach 21. Historical Nursing Research 22. Generalizability and Qualitative Research 23. Additional Types of Unstructured Self-Reports 24. Transcribing Qualitative Data 25. Whittemore and Colleagues’ Framework of Quality Criteria in Qualitative Research 26. Converting Quantitative and Qualitative Data 27. 20 Complex Intervention Development: Exploratory Questions 28. Examples of Various Pilot Study Objectives 29. Publication Bias in Meta-Analyses 30. Tips for Publishing Reports on Pilot Intervention Studies 31. Proposals for Pilot Intervention Studies Another new feature of this edition concerns our interest in readers’ access to references we cited. To the extent possible, the studies we have chosen as examples of particular research methods are published as openaccess articles. These studies are identified with an asterisk in the reference list at the end of each chapter, and a link to the article is included in the Toolkit section of the Resource Manual. We hope that these revisions will help users of this book to maximize their learning experience. O R G A N I Z AT I O N O F T H E T E X T The content of this edition is organized into six main parts. • Part I—Foundations of Nursing Research and Evidence-Based Practice introduces fundamental concepts in nursing research. Chapter 1 briefly summarizes the history and future of nursing research, discusses the philosophical underpinnings of qualitative research versus quantitative research, and describes major purposes of nursing research. Chapter 2 offers guidance on utilizing research to build an evidence-based practice. Chapter 3 introduces readers to key research terms and presents an overview of steps in the research process for both qualitative and quantitative studies. • Part II—Conceptualizing and Planning a Study to Generate Evidence further sets the stage for learning about the research process by discussing issues relating to a study’s conceptualization: the formulation of research questions and hypotheses (Chapter 4), the review of relevant research (Chapter 5), the development of theoretical and conceptual contexts (Chapter 6), and the fostering of ethically sound approaches in doing research (Chapter 7). Chapter 8 provides an overview of important issues that researchers must attend to during the planning of any type of study. • Part III—Designing and Conducting Quantitative Studies to Generate Evidence presents material on undertaking quantitative nursing studies. Chapter 9 describes fundamental principles and applications of quantitative research design, and Chapter 10 focuses on methods to enhance the rigor of a quantitative study, including mechanisms of research control. Chapter 11 examines research with different and distinct purposes, including surveys, outcomes research, and evaluations. Chapter 12 presents strategies for sampling study participants in quantitative research. Chapter 13 describes using structured data collection methods that yield quantitative information. Chapter 14 discusses the concept of measurement and then focuses on methods of assessing 21 the quality of formal measuring instruments. In this edition, we describe methods to assess the properties of point-in-time measurements (reliability and validity) and longitudinal measurements—change scores (reliability of change scores and responsiveness). Chapter 15 presents material on how to develop high-quality self-report instruments. Chapters 16, 17, and 18 present an overview of univariate, bivariate, and multivariate statistical analyses, respectively. Chapter 19 describes the development of an overall analytic strategy for quantitative studies, including material on handling missing data. Chapter 20, a new chapter, discusses the issue of interpreting results and making inferences about clinical significance. • Part IV—Designing and Conducting Qualitative Studies to Generate Evidence presents material on undertaking qualitative nursing studies. Chapter 21 is devoted to research designs and approaches for qualitative studies, including material on critical theory, feminist, and participatory action research. Chapter 22 discusses strategies for sampling study participants in qualitative inquiries. Chapter 23 describes methods of gathering unstructured self-report and observational data for qualitative studies. Chapter 24 discusses methods of analyzing qualitative data, with specific information on grounded theory, phenomenologic, and ethnographic analyses. Chapter 25 elaborates on methods qualitative researchers can use to enhance (and assess) integrity and quality throughout their inquiries. • Part V—Designing and Conducting Mixed Methods Studies to Generate Evidence presents material on mixed methods nursing studies. Chapter 26 discusses a broad range of issues, including asking mixed methods questions, designing a study to address the questions, sampling participants in mixed methods research, and analyzing and integrating qualitative and quantitative data. Chapter 27 presents innovative information about using mixed methods approaches in the development of nursing interventions. In Chapter 28, a new chapter, we provide guidance for designing and conducting a pilot study and using data from the pilot to draw conclusions about how best to proceed. • Part VI—Building an Evidence Base for Nursing Practice provides additional guidance on linking research and clinical practice. Chapter 29 offers an overview of methods of conducting systematic reviews that support EBP, with an emphasis on meta-analyses, metasyntheses, and mixed studies reviews. Chapter 30 discusses dissemination of evidence—how to prepare a research report (including theses and dissertations) and how to publish research findings. The concluding chapter (Chapter 31) offers suggestions and guidelines on developing research proposals and getting financial support and includes information about applying for NIH grants and interpreting scores from NIH’s 22 new scoring system. K E Y FE AT U R E S This textbook was designed to be helpful to those who are learning how to do research as well as to those who are learning to appraise research reports critically and to use research findings in practice. Many of the features successfully used in previous editions have been retained in this 10th edition. Among the basic principles that helped to shape this and earlier editions of this book are (1) an unswerving conviction that the development of research skills is critical to the nursing profession, (2) a fundamental belief that research is intellectually and professionally rewarding, and (3) a steadfast opinion that learning about research methods need be neither intimidating nor dull. Consistent with these principles, we have tried to present the fundamentals of research methods in a way that both facilitates understanding and arouses curiosity and interest. Key features of our approach include the following: • Research Examples. Each chapter concludes with one or two actual research examples designed to highlight critical points made in the chapter and to sharpen the reader’s critical thinking skills. In addition, many research examples are used to illustrate key points in the text and to stimulate ideas for a study. Many of the examples used in this edition are open-access articles that can be used for further learning and classroom discussions. • Critiquing Guidelines. Most chapters include guidelines for conducting a critique of each aspect of a research report. These guidelines provide a list of questions that draw attention to specific aspects of a report that are amenable to appraisal. • Clear, “user-friendly” style. Our writing style is designed to be easily digestible and nonintimidating. Concepts are introduced carefully and systematically, difficult ideas are presented clearly, and readers are assumed to have no prior exposure to technical terms. • Specific practical tips on doing research. This textbook is filled with practical guidance on how to translate the abstract notions of research methods into realistic strategies for conducting research. Every chapter includes several tips for applying the chapter’s lessons to real-life situations. These suggestions are in recognition of the fact that there is often a large gap between what gets taught in research methods textbooks and what a researcher needs to know to conduct a study. • Aids to student learning. Several features are used to enhance and reinforce learning and to help focus the student’s attention on specific areas of text content, including the following: succinct, bulleted summaries at the end of each 23 chapter; tables and figures that provide examples and graphic materials in support of the text discussion; study suggestions at the end of each chapter; a detailed glossary; and a comprehensive index for accessing information quickly. T E A C H I N G – L E A R N I N G PA C K A G E Nursing Research: Generating and Assessing Evidence for Nursing Practice, 10th edition, has an ancillary package designed with both students and instructors in mind. • The Resource Manual augments the textbook in important ways. The manual itself provides students with exercises that correspond to each text chapter, with a focus on opportunities to critique actual studies. The appendix includes 12 research journal articles in their entirety, plus a successful grant application for a study funded by the National Institute of Nursing Research. The 12 reports cover a range of nursing research ventures, including qualitative, quantitative, and mixed methods studies, an instrument development study, an evidencebased practice translation project, and two systematic reviews. Full critiques of two of the reports are also included and can serve as models for a comprehensive research critique. • The Toolkit to the Resource Manual is a “must-have” innovation that will save considerable time for both students and seasoned researchers. Included on thePoint, the Toolkit offers dozens of research resources in Word documents that can be downloaded and used directly or adapted. The resources reflect bestpractice research material, most of which have been pretested and refined in our own research. The Toolkit originated with our realization that in our technologically advanced environment, it is possible to not only illustrate methodologic tools as graphics in the textbook but also to make them directly available for use and adaptation. Thus, we have included dozens of documents in Word files that can readily be used in research projects, without requiring researchers to “reinvent the wheel” or tediously retype material from this textbook. Examples include informed consent forms, a demographic questionnaire, content validity forms, and a coding sheet for a meta-analysis— to name only a few. The Toolkit also has lists of relevant and useful websites for each chapter, which can be “clicked” on directly without having to retype the URL and risk a typographical error. Links to open-access articles cited in the textbook, as well as other open-access articles relevant to each chapter, are included in the Toolkit. • The Instructor’s Resources on the Point include PowerPoint slides summarizing key points in each chapter, test questions that have been placed into a program that allows instructors to automatically generate a test, and an 24 image bank. It is our hope that the content, style, and organization of this book continue to meet the needs of a broad spectrum of nursing students and nurse researchers. We also hope that this book will help to foster enthusiasm for the kinds of discoveries that research can produce and for the knowledge that will help support an evidence-based nursing practice. DENISE F. POLIT, PhD, FAAN CHERYL TATANO BECK, DNSc, CNM, FAAN 25 26 Contents PART 1: FOUNDATIONS OF NURSING RESEARCH Chapter 1: Introduction to Nursing Research in an Evidence-Based Practice Environment Chapter 2: Evidence-Based Nursing: Translating Research Evidence into Practice Chapter 3: Key Concepts and Steps in Qualitative and Quantitative Research PART 2: CONCEPTUALIZING AND PLANNING A STUDY TO GENERATE EVIDENCE FOR NURSING Chapter 4: Research Problems, Research Questions, and Hypotheses Chapter 5: Literature Reviews: Finding and Critiquing Evidence Chapter 6: Theoretical Frameworks Chapter 7: Ethics in Nursing Research Chapter 8: Planning a Nursing Study PART 3: DESIGNING AND CONDUCTING QUANTITATIVE STUDIES TO GENERATE EVIDENCE FOR NURSING Chapter 9: Quantitative Research Design Chapter 10: Rigor and Validity in Quantitative Research Chapter 11: Specific Types of Quantitative Research Chapter 12: Sampling in Quantitative Research Chapter 13: Data Collection in Quantitative Research Chapter 14: Measurement and Data Quality Chapter 15: Developing and Testing Self-Report Scales Chapter 16: Descriptive Statistics Chapter 17: Inferential Statistics 27 Chapter 18: Multivariate Statistics Chapter 19: Processes of Quantitative Data Analysis Chapter 20: Clinical Significance and Interpretation of Quantitative Results PART 4: DESIGNING AND CONDUCTING QUALITATIVE STUDIES TO GENERATE EVIDENCE FOR NURSING Chapter 21: Qualitative Research Design and Approaches Chapter 22: Sampling in Qualitative Research Chapter 23: Data Collection in Qualitative Research Chapter 24: Qualitative Data Analysis Chapter 25: Trustworthiness and Integrity in Qualitative Research PART 5: DESIGNING AND CONDUCTING MIXED METHODS STUDIES TO GENERATE EVIDENCE FOR NURSING Chapter 26: Basics of Mixed Methods Research Chapter 27: Developing Complex Nursing Interventions Using Mixed Methods Research Chapter 28: Feasibility Assessments and Pilot Tests of Interventions Using Mixed Methods PART 6: BUILDING AN EVIDENCE BASE FOR NURSING PRACTICE Chapter 29: Systematic Reviews of Research Evidence: Meta-Analysis, Metasynthesis, and Mixed Studies Review Chapter 30: Disseminating Evidence: Reporting Research Findings Chapter 31: Writing Proposals to Generate Evidence Glossary Appendix: Statistical Tables Index 28 Check Out the Latest Book Authored by Research Expert Dr. Polit If you want to make thoughtful but practical decisions about the measurement of health constructs, check out Dr. Polit and Dr. Yang’s latest book, a “gentle” introduction to and overview of complex measurement content, called Measurement and the Measurement of Change. This book is for researchers and clinicians from all health disciplines because measurement is vital to high-quality science and to excellence in clinical practice. The text focuses on the measurement of health constructs, particularly those constructs that are not amenable to quantification by means of laboratory analysis or technical instrumentation. These health constructs include a wide range of human attributes, such as quality of life, functional ability, self-efficacy, depression, and pain. Measures of such constructs are proliferating at a rapid rate and often without adequate attention paid to ensuring that standards of scientific rigor are met. 29 In this book, the authors offer guidance to those who develop new instruments, adapt existing ones, select instruments for use in a clinical trial or in clinical practice, interpret information from measurements and changes in scores, or undertake a systematic review on instruments. This book offers guidance on how to develop new instruments using both “classical” and “modern” approaches from psychometrics as well as methods used in clinimetrics. Much of this book, however, concerns the evaluation of instruments in relation to three key measurement domains: reliability, validity, and responsiveness. This text was designed to be useful in graduate-level courses on measurement or research methods and will also serve as an important reference and resource for researchers and clinicians. 30 PART 1 FOUNDATIONS OF NURSING RESEARCH 31 1 Introduction to Nursing Research in an Evidence-Based Practice Environment 32 NURSING RESEARCH IN PERSPECTIVE In all parts of the world, nursing has experienced a profound culture change. Nurses are increasingly expected to understand and conduct research and to base their professional practice on research evidence—that is, to adopt an evidencebased practice (EBP). EBP involves using the best evidence (as well as clinical judgment and patient preferences) in making patient care decisions, and “best evidence” typically comes from research conducted by nurses and other health care professionals. What Is Nursing Research? Research is systematic inquiry that uses disciplined methods to answer questions or solve problems. The ultimate goal of research is to develop and expand knowledge. Nurses are increasingly engaged in disciplined studies that benefit nursing and its clients. Nursing research is systematic inquiry designed to generate trustworthy evidence about issues of importance to the nursing profession, including nursing practice, education, administration, and informatics. In this book, we emphasize clinical nursing research, that is, research to guide nursing practice and to improve the health and quality of life of nurses’ clients. Nursing research has experienced remarkable growth in the past three decades, providing nurses with a growing evidence base from which to practice. Yet many questions endure and much remains to be done to incorporate research innovations into nursing practice. Examples of Nursing Research Questions: • How effective is pressurized irrigation, compared to a swabbing method, in cleansing wounds, in terms of time to wound healing, pain, patients’ satisfaction with comfort, and costs? (Mak et al., 2015) • What are the experiences of women in Zimbabwe who are living with advanced HIV infection? (Gona & DeMarco, 2015) The Importance of Research in Nursing Research findings from rigorous studies provide especially strong evidence for informing nurses’ decisions and actions. Nurses are accepting the need to base specific nursing actions on research evidence indicating that the actions are clinically appropriate, cost-effective, and result in positive outcomes for clients. In the United States, research plays an important role in nursing in terms of cred 33 entialing and status. The American Nurses Credentialing Center (ANCC)—an arm of the American Nurses Association and the largest and most prestigious credentialing organization in the United States—developed a Magnet Recognition Program to acknowledge health care organizations that provide high-quality nursing care. As Reigle and her colleagues (2008) noted, “the road to Magnet Recognition is paved with EBP” (p. 102) and the 2014 Magnet application manual incorporated revisions that strengthened evidence-based requirements (Drenkard, 2013). The good news is that there is growing confirmation that the focus on research and evidence-based practice may have important payoffs. For example, McHugh and co-researchers (2013) found that Magnet hospitals have lower riskadjusted mortality and failure to rescue than non-Magnet hospitals, even when differences among the hospitals in nursing credentials and patient characteristics are taken into account. Changes to nursing practice now occur regularly because of EBP efforts. Practice changes often are local initiatives that are not publicized, but broader clinical changes are also occurring based on accumulating research evidence about beneficial practice innovations. Example of Evidence-Based Practice: Numerous clinical practice changes reflect the impact of research. For example, “kangaroo care” (the holding of diaper-clad infants skin to skin by parents) is now practiced in many neonatal intensive care units (NICUs), but this is a relatively new trend. As recently as the 1990s, only a minority of NICUs offered kangaroo care options. Expanded adoption of this practice reflects mounting evidence that early skin-to-skin contact has benefits without negative side effects (e.g., Ludington-Hoe, 2011; Moore et al., 2012). Some of that evidence came from rigorous studies conducted by nurse researchers in several countries (e.g., Chwo et al., 2002; Cong et al., 2009; Cong et al., 2011; Hake-Brooks & Anderson, 2008). Nurses continue to study the potential benefits of kangaroo care in important clinical trials (e.g., Campbell-Yeo et al., 2013). The Consumer–Producer Continuum in Nursing Research In our current environment, all nurses are likely to engage in activities along a continuum of research participation. At one end of the continuum are consumers of nursing research, who read research reports or research summaries to keep up-to-date on findings that might affect their practice. EBP depends on well-informed nursing research consumers. At the other end of the continuum are the producers of nursing research: nurses who design and conduct research. At one time, most nurse researchers were 34 academics who taught in schools of nursing, but research is increasingly being conducted by nurses in health care settings who want to find solutions to recurring problems in patient care. Between these end points on the continuum lie a variety of research activities that are undertaken by nurses. Even if you never personally undertake a study, you may (1) contribute to an idea or a plan for a clinical study; (2) gather data for a study; (3) advise clients about participating in research; (4) solve a clinical problem by searching for research evidence; or (5) discuss the implications of a new study in a journal club in your practice setting, which involves meetings (in groups or online) to discuss research articles. In all possible research001-related activities, nurses who have some research skills are better able than those without them to make a contribution to nursing and to EBP. An understanding of nursing research can improve the depth and breadth of every nurse’s professional practice. Nursing Research in Historical Perspective Table 1.1 summarizes some of the key events in the historical evolution of nursing research. (An expanded summary of the history of nursing research appears in the Supplement to this chapter on ). 35 Most people would agree that research in nursing began with Florence Nightingale in the 1850s. Her most well-known research contribution involved an analysis of factors affecting soldier mortality and morbidity during the Crimean War. Based on skillful analyses, she was successful in effecting changes in nursing care and, more generally, in public health. After Nightingale’s work, research was absent from the nursing literature until the early 1900s, but most early studies concerned nurses’ education rather than clinical issues. In the 1950s, research by nurses began to accelerate. For example, a nursing research center was established at the Walter Reed Army Institute of Research. Also, the American Nurses Foundation, which is devoted to the promotion of nursing research, was founded. The surge in the number of studies conducted in the 1950s created the need for a new journal; Nursing Research came into being in 1952. As shown in Table 1.1, dissemination opportunities in professional journals grew steadily thereafter. In the 1960s, nursing leaders expressed concern about the shortage of research 36 on practice issues. Professional nursing organizations, such as the Western Interstate Council for Higher Education in Nursing, established research priorities, and practice-oriented research on various clinical topics began to emerge in the literature. During the 1970s, improvements in client care became a more visible research priority and nurses also began to pay attention to the clinical utilization of research findings. Guidance on assessing research for application in practice settings became available. Several journals that focus on nursing research were established in the 1970s, including Advances in Nursing Science, Research in Nursing & Health, and the Western Journal of Nursing Research. Nursing research also expanded internationally. For example, the Workgroup of European Nurse Researchers was established in 1978 to develop greater communication and opportunities for partnerships among 25 European National Nurses Associations. Nursing research continued to expand in the 1980s. In the United States, the National Center for Nursing Research (NCNR) at the National Institutes of Health (NIH) was established in 1986. Several forces outside of nursing also helped to shape the nursing research landscape. A group from the McMaster Medical School in Canada designed a clinical learning strategy that was called evidence-based medicine (EBM). EBM, which promulgated the view that research findings were far superior to the opinions of authorities as a basis for clinical decisions, constituted a profound shift for medical education and practice, and has had a major effect on all health care professions. Nursing research was strengthened and given more visibility when NCNR was promoted to full institute status within the NIH. In 1993, the National Institute of Nursing Research (NINR) was established, helping to put nursing research more into the mainstream of health research. Funding opportunities for nursing research expanded in other countries as well. Current and Future Directions for Nursing Research Nursing research continues to develop at a rapid pace and will undoubtedly flourish in the 21st century. Funding continues to grow. For example, NINR funding in fiscal year 2014 was more than $140 million compared to $70 million in 1999—and the competition for available funding is increasingly vigorous as more nurses seek support for testing innovative ideas for practice improvements. Broadly speaking, the priority for future nursing research will be the promotion of excellence in nursing science. Toward this end, nurse researchers and practicing nurses will be sharpening their research skills and using those skills to address 37 emerging issues of importance to the profession and its clientele. Among the trends we foresee for the early 21st century are the following: • Continued focus on EBP. Encouragement for nurses to engage in evidence-based patient care is sure to continue. In turn, improvements will be needed both in the quality of studies and in nurses’ skills in locating, understanding, critiquing, and using relevant study results. Relatedly, there is an emerging interest in translational research— research on how findings from studies can best be translated into practice. Translation potential will require researchers to think more strategically about long-term feasibility, scalability, and sustainability when they test solutions to problems. • Development of a stronger evidence base through confirmatory strategies. Practicing nurses are unlikely to adopt an innovation based on weakly designed or isolated studies. Strong research designs are essential, and confirmation is usually needed through the replication (i.e., the repeating) of studies with different clients, in different clinical settings, and at different times to ensure that the findings are robust. • Greater emphasis on systematic reviews. Systematic reviews are a cornerstone of EBP and will take on increased importance in all health disciplines. Systematic reviews rigorously integrate research information on a topic so that conclusions about the state of evidence can be reached. Best practice clinical guidelines typically rely on such systematic reviews. • Innovation. There is currently a major push for creative and innovative solutions to recurring practice problems. “Innovation” has become an important buzzword throughout NIH and in nursing associations. For example, the 2013 annual conference of the Council for the Advancement of Nursing Science was “Innovative Approaches to Symptom Science.” Innovative interventions—and new methods for studying nursing questions—are sure to be part of the future research landscape in nursing. • Expanded local research in health care settings. Small studies designed to solve local problems will likely increase. This trend will be reinforced as more hospitals apply for (and are recertified for) Magnet status in the United States and in other countries. Mechanisms will need to be developed to ensure that evidence from these small projects becomes available to others facing similar problems, such as communication within and between regional nursing research alliances. • Strengthening of interdisciplinary collaboration. Collaboration of nurses with researchers in related fields is likely to expand in the 21st century as researchers address fundamental health care problems. In turn, such collaborative efforts 38 could lead to nurse researchers playing a more prominent role in national and international health care policies. One of four major recommendations in a 2010 report on the future of nursing by the Institute of Medicine was that nurses should be full partners with physicians and other health care professionals in redesigning health care. • Expanded dissemination of research findings. The Internet and other electronic communication have a big impact on disseminating research information, which in turn helps to promote EBP. Through technologic advances, information about innovations can be communicated more widely and more quickly than ever before. • Increased focus on cultural issues and health disparities. The issue of health disparities has emerged as a central concern in nursing and other health disciplines; this in turn has raised consciousness about the cultural sensitivity of health interventions and the cultural competence of health care workers. There is growing awareness that research must be sensitive to the health beliefs, behaviors, and values of culturally and linguistically diverse populations. • Clinical significance and patient input. Research findings increasingly must meet the test of being clinically significant, and patients have taken center stage in efforts to define clinical significance. A major challenge in the years ahead will involve getting both research evidence and patient preferences into clinical decisions, and designing research to study the process and the outcomes. Broad research priorities for the future have been articulated by many nursing organizations, including NINR and Sigma Theta Tau International. Expert panels and research working groups help NINR to identify gaps in current knowledge that require research. The primary areas of research funded by NINR in 2014 were health promotion/disease prevention, eliminating health disparities, caregiving, symptom management, and self-management. Research priorities that have been expressed by Sigma Theta Tau International include advancing healthy communities through health promotion; preventing disease and recognizing social, economic, and political determinants; implementation of evidence-based practice; targeting the needs of vulnerable populations such as the poor and chronically ill; and developing nurses’ capacity for research. Priorities also have been developed for several nursing specialties and for nurses in several countries—for example, Ireland (Brenner et al., 2014; Drennan et al., 2007), Sweden (Bäck-Pettersson et al., 2008), Australia (Wynaden et al., 2014), and Korea (Kim et al., 2002). S O U R C E S O F E V I D E N C E FO R N U R S I N G PR A C T I C E Nurses make clinical decisions based on knowledge from many sources, including 39 coursework, textbooks, and their own clinical experience. Because evidence is constantly evolving, learning about best practice nursing perseveres throughout a nurse’s career. Some of what nurses learn is based on systematic research, but much of it is not. What are the sources of evidence for nursing practice? Where does knowledge for practice come from? Until fairly recently, knowledge primarily was handed down from one generation to the next based on experience, trial and error, tradition, and expert opinion. Information sources for clinical practice vary in dependability, giving rise to what is called an evidence hierarchy, which acknowledges that certain types of evidence are better than others. A brief discussion of some alternative sources of evidence shows how research001-based information is different. Tradition and Authority Decisions are sometimes based on custom or tradition. Certain “truths” are accepted as given, and such “knowledge” is so much a part of a common heritage that few seek verification. Tradition facilitates communication by providing a common foundation of accepted truth, but many traditions have never been evaluated for their validity. There is concern that some nursing interventions are based on tradition, custom, and “unit culture” rather than on sound evidence. Indeed, a recent analysis suggests that some “sacred cows” (ineffective traditional habits) persist even in a health care center recognized as a leader in evidence-based practice (Hanrahan et al., 2015). Another common source of information is an authority, a person with specialized expertise. We often make decisions about problems with which we have little experience; it seems natural to place our trust in the judgment of people with specialized training or experience. As a source of evidence, however, authority has shortcomings. Authorities are not infallible, particularly if their expertise is based primarily on personal experience; yet, like tradition, their knowledge often goes unchallenged. Example of “Myths” in Nursing Textbooks: A study suggests that even nursing textbooks may contain “myths.” In their analysis of 23 widely used undergraduate psychiatric nursing textbooks, Holman and colleagues (2010) found that all books contained at least one unsupported assumption (myth) about loss and grief—that is, assumptions not supported by research evidence. Moreover, many evidence-based findings about grief and loss failed to be included in the textbooks. Clinical Experience, Trial and Error, and Intuition Clinical experience is a familiar, functional source of knowledge. The ability to 40 generalize, to recognize regularities, and to make predictions is an important characteristic of the human mind. Nevertheless, personal experience is limited as a knowledge source because each nurse’s experience is too narrow to be generally useful. A second limitation is that the same objective event is often experienced and perceived differently by two nurses. A related method is trial and error in which alternatives are tried successively until a solution to a problem is found. We likely have all used this method in our professional work. For example, many patients dislike the taste of potassium chloride solution. Nurses try to disguise the taste of the medication in various ways until one method meets with the approval of the patient. Trial and error may offer a practical means of securing knowledge, but the method tends to be haphazard and solutions may be idiosyncratic. Intuition is a knowledge source that cannot be explained based on reasoning or prior instruction. Although intuition and hunches undoubtedly play a role in nursing—as they do in the conduct of research—it is difficult to develop nursing policies and practices based on intuition. Logical Reasoning Solutions to some problems are developed by logical thought processes. As a problem-solving method, logical reasoning combines experience, intellectual faculties, and formal systems of thought. Inductive reasoning involves developing generalizations from specific observations. For example, a nurse may observe the anxious behavior of (specific) hospitalized children and conclude that (in general) children’s separation from their parents is stressful. Deductive reasoning involves developing specific predictions from general principles. For example, if we assume that separation anxiety occurs in hospitalized children (in general), then we might predict that (specific) children in a hospital whose parents do not room-in will manifest symptoms of stress. Both systems of reasoning are useful for understanding and organizing phenomena, and both play a role in research. Logical reasoning in and of itself, however, is limited because the validity of reasoning depends on the accuracy of the premises with which one starts. Assembled Information In making clinical decisions, health care professionals rely on information that has been assembled for a variety of purposes. For example, local, national, and international benchmarking data provide information on such issues as infection rates or the rates of using various procedures (e.g., cesarean births) and can 41 facilitate evaluations of clinical practices. Cost data—information on the costs associated with certain procedures, policies, or practices—are sometimes used as a factor in clinical decision making. Quality improvement and risk data, such as medication error reports, can be used to assess the need for practice changes. Such sources are useful, but they do not provide a good mechanism for determining whether improvements in patient outcomes result from their use. Disciplined Research Research conducted in a disciplined framework is the most sophisticated method of acquiring knowledge. Nursing research combines logical reasoning with other features to create evidence that, although fallible, tends to yield the most reliable evidence. Carefully synthesized findings from rigorous research are at the pinnacle of most evidence hierarchies. The current emphasis on EBP requires nurses to base their clinical practice to the greatest extent possible on rigorous research001-based findings rather than on tradition, authority, intuition, or personal experience— although nursing will always remain a rich blend of art and science. PA R A D I G M S A N D M E T H O D S FO R N U R S I N G RESEARCH A paradigm is a worldview, a general perspective on the complexities of the world. Paradigms for human inquiry are often characterized in terms of the ways in which they respond to basic philosophical questions, such as, What is the nature of reality? (ontologic) and What is the relationship between the inquirer and those being studied? (epistemologic). Disciplined inquiry in nursing has been conducted mainly within two broad paradigms, positivism and constructivism. This section describes these two paradigms and outlines the research methods associated with them. In later chapters, we describe the transformative paradigm that involves critical theory research (Chapter 21), and a pragmatism paradigm that involves mixed methods research (Chapter 26). The Positivist Paradigm The paradigm that dominated nursing research for decades is known as positivism (also called logical positivism). Positivism is rooted in 19th century thought, guided by such philosophers as Mill, Newton, and Locke. Positivism reflects a broader cultural phenomenon that, in the humanities, is referred to as modernism, which emphasizes the rational and the scientific. As shown in Table 1.2, a fundamental assumption of positivists is that there is a reality out there that can be studied and known (an assumption is a basic principle 42 that is believed to be true without proof or verification). Adherents of positivism assume that nature is basically ordered and regular and that reality exists independent of human observation. In other words, the world is assumed not to be merely a creation of the human mind. The related assumption of determinism refers to the positivists’ belief that phenomena are not haphazard but rather have antecedent causes. If a person has a cerebrovascular accident, the researcher in a positivist tradition assumes that there must be one or more reasons that can be potentially identified. Within the positivist paradigm, much research activity is directed at understanding the underlying causes of phenomena. Positivists value objectivity and attempt to hold personal beliefs and biases in check to avoid contaminating the phenomena under study. The positivists’ scientific approach involves using orderly, disciplined procedures with tight controls of the research situation to test hunches about the phenomena being studied. Strict positivist thinking has been challenged, and few researchers adhere to the 43 tenets of pure positivism. In the postpositivist paradigm, there is still a belief in reality and a desire to understand it, but postpositivists recognize the impossibility of total objectivity. They do, however, see objectivity as a goal and strive to be as neutral as possible. Postpositivists also appreciate the impediments to knowing reality with certainty and therefore seek probabilistic evidence—that is, learning what the true state of a phenomenon probably is, with a high degree of likelihood. This modified positivist position remains a dominant force in nursing research. For the sake of simplicity, we refer to it as positivism. The Constructivist Paradigm The constructivist paradigm (often called the naturalistic paradigm) began as a countermovement to positivism with writers such as Weber and Kant. Just as positivism reflects the cultural phenomenon of modernism that burgeoned after the industrial revolution, naturalism is an outgrowth of the cultural transformation called postmodernism. Postmodern thinking emphasizes the value of deconstruction—taking apart old ideas and structures—and reconstruction—putting ideas and structures together in new ways. The constructivist paradigm represents a major alternative system for conducting disciplined research in nursing. Table 1.2 compares the major assumptions of the positivist and constructivist paradigms. For the naturalistic inquirer, reality is not a fixed entity but rather is a construction of the individuals participating in the research; reality exists within a context, and many constructions are possible. Naturalists thus take the position of relativism: If there are multiple interpretations of reality that exist in people’s minds, then there is no process by which the ultimate truth or falsity of the constructions can be determined. The constructivist paradigm assumes that knowledge is maximized when the distance between the inquirer and those under study is minimized. The voices and interpretations of study participants are crucial to understanding the phenomenon of interest, and subjective interactions are the primary way to access them. Findings from a constructivist inquiry are the product of the interaction between the inquirer and the participants. Paradigms and Methods: Quantitative and Qualitative Research Research methods are the techniques researchers use to structure a study and to gather and analyze information relevant to the research question. The two alternative paradigms correspond to different 44 methods for developing evidence. A key methodologic distinction is between quantitative research, which is most closely allied with positivism, and qualitative research, which is associated with constructivist inquiry—although positivists sometimes undertake qualitative studies, and constructivist researchers sometimes collect quantitative information. This section provides an overview of the methods associated with the two paradigms. The Scientific Method and Quantitative Research The traditional, positivist scientific method refers to a set of orderly, disciplined procedures used to acquire information. Quantitative researchers use deductive reasoning to generate predictions that are tested in the real world. They typically move in a systematic fashion from the definition of a problem and the selection of concepts on which to focus to the solution of the problem. By systematic, we mean that the investigator progresses logically through a series of steps, according to a specified plan of action. Quantitative researchers use various control strategies. Control involves imposing conditions on the research situation so that biases are minimized and precision and validity are maximized. Control mechanisms are discussed at length in this book. Quantitative researchers gather empirical evidence—evidence that is rooted in objective reality and gathered through the senses. Empirical evidence, then, consists of observations gathered through sight, hearing, taste, touch, or smell. Observations of the presence or absence of skin inflammation, patients’ anxiety level, or infant birth weight are all examples of empirical observations. The requirement to use empirical evidence means that findings are grounded in reality rather than in researchers’ personal beliefs. Evidence for a study in the positivist paradigm is gathered according to an established plan, using structured methods to collect needed information. Usually (but not always) the information gathered is quantitative—that is, numeric information that is obtained from a formal measurement and is analyzed statistically.

ORDER A PLAGIARISM FREE PAPER NOW

A traditional scientific study strives to go beyond the specifics of a research situation. For example, quantitative researchers are typically not as interested in understanding why a particular person has a stroke as in understanding what factors influence its occurrence in people generally. The degree to which research 45 findings can be generalized to individuals other than those who participated in the study is called the study’s generalizability. The scientific method has enjoyed considerable stature as a method of inquiry and has been used productively by nurse researchers studying a range of nursing problems. This is not to say, however, that this approach can solve all nursing problems. One important limitation—common to both quantitative and qualitative research—is that research cannot be used to answer moral or ethical questions. Many persistent, intriguing questions about human beings fall into this area— questions such as whether euthanasia should be practiced or abortion should be legal. The traditional resear…

Summarize Research Articles on template attached

Thornton et al. BMC Health Services Research (2017) 17:361 DOI 10.1186/s12913-017-2307-z RESEARCH ARTICLE Open Access Influences on patient satisfaction in healthcare centers: a semi-quantitative study over 5 years Ruth D. Thornton1, Nicole Nurse2, Laura Snavely3, Stacey Hackett-Zahler4, Kenice Frank5 and Robert A. DiTomasso1* Abstract Background: Knowledge of ambulatory patients’ satisfaction

ORDER A PLAGIARISM FREE PAPER NOW

with clinic visits help improve communication and delivery of healthcare. The goal was to examine patient satisfaction in a primary care setting, identify how selected patient and physician setting and characteristics affected satisfaction, and determine if feedback provided to medical directors over time impacted patient satisfaction. Methods: A three-phase, semi-quantitative analysis was performed using anonymous, validated patient satisfaction surveys collected from 889 ambulatory outpatients in 6 healthcare centers over 5-years. Patients’ responses to 21 questions were analyzed by principal components varimax rotated factor analysis. Three classifiable components emerged: Satisfaction with Physician, Availability/Convenience, and Orderly/Time. To study the effects of several independent variables (location of clinics, patients’ and physicians’ age, education level and duration at the clinic), data were subjected to multivariate analysis of variance (MANOVA).. Results: Changes in the healthcare centers over time were not significantly related to patient satisfaction. However, location of the center did affect satisfaction. Urban patients were more satisfied with their physicians than rural, and inner city patients were less satisfied than urban or rural on Availability/Convenience and less satisfied than urban patients on Orderly/Time. How long a patient attended a center most affected satisfaction, with patients attending >10 years more satisfied in all three components than those attending 60 years old. Patients were significantly more satisfied with their 30–40 year-old physicians compared with those over 60. On Orderly/Time, patients were more satisfied with physicians who were in their 50′s than physicians >60. Conclusions: Improvement in patient satisfaction includes a need for immediate, specific feedback. Although Medical Directors received feedback yearly, we found no significant changes in patient satisfaction over time. Our results suggest that, to increase satisfaction, patients with lower education, those who are sicker, and those who are new to the center likely would benefit from additional high quality interactions with their physicians. Keywords: Patient satisfaction, Health care delivery, Community health * Correspondence: RobertD@pcom.edu 1 Department of Psychology, Philadelphia College of Osteopathic Medicine, 4170 City Ave., Philadelphia, PA 19131, USA Full list of author information is available at the end of the article © The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Thornton et al. BMC Health Services Research (2017) 17:361 Background Patient satisfaction surveys are often used to understand patients’ concerns and determine areas for improvement, including improving communication between physicians and patients. Survey results document progress and allow physicians and staff to maintain high standards. Although results of patient satisfaction surveys are used by payer systems to profile individual physicians and guide physician compensation, one study showed that < 25% of primary care physicians found profiles useful for improving patient care and fewer used the profiles to change [1]. Improvements are more likely to occur if staff receives more immediate feedback [2]. Data collection methods play a role in outcomes. Onsite surveys provide an immediate outlet for patients who are experiencing problems, although higher ratings for on-site surveys may also relate directly to doctorpatient communication. Surveys administered later after a clinic visit may yield lower ratings, possibly due to the course of treatment [2, 3]. Many factors influence patient satisfaction. Patient demographics such as age, gender, income, socioeconomic and general health status impact patients’ responses [3, 4]. Characteristics of the medical provider, including demographics and experience, also affect their interactions with patients [5–9]. Other factors include the type of setting the patient is in [10] and the amount of time patients had to wait [11]. However, Anderson found that complaints about wait time can be moderated by time spent with the physician [12]. Physician characteristics extend beyond the obvious. Physician-patient concordance in race, gender or age may be important in patient satisfaction, but many other factors such as primary language, parental status, sexual orientation, values, beliefs, or communication style may be associated [13, 14]. How long the patient has been with this physician and the degree to which the physicians’ communication is patientcentered are significant [13]. A physician’s experience plays a role, with lowest patient satisfaction with firstyear residents; interestingly, residents with some more Page 2 of 9 experience attained similar satisfaction ratings to those of the faculty attendings, suggesting that the requisite skills are acquired during the first year of training [7]. Whether to administer patient satisfaction surveys depends on the overall goals of the medical facility and on physician buy-in to change [1, 15]. The views of the medical director and administrator are key components as to whether the surveys are taken seriously and acted upon by physicians [16]. Patient satisfaction can become a success criterion of the center when physicians and staff meet regularly to discuss improvements in a context of cooperation and mutual support. Methods We initiated this study of patient satisfaction to help physicians better understand their patients at the healthcare centers (HCCs) of a not-for-profit medical school’s outpatient primary care centers on the east coast. Physicians were provided raw data and results of open-ended questions very soon after each year’s study. However, we decided to statistically analyze the overall data in order to understand where patients were most and least satisfied and what influenced their satisfaction. Our goal was to provide information which could help focus physician directors’ changes to improve patient satisfaction. The research was under the auspices of a medical college (Philadelphia College of Osteopathic Medicine, PCOM) which owns and operates five outpatient HCCs, four of which are located within the city limits of Philadelphia and the fifth HCC located in a rural area. [17] Two within Philadelphia are considered urban, while two are in the inner city [18]. An additional nonaffiliated, inner city HCC located within Philadelphia was also used in the research. We considered the nonaffiliated HCC as a control, but expected it to likely agree with data from the affiliated inner city HCCs. The quantity of surveys administered are listed in Table 1. This research arose from a need to quickly and inexpensively conduct patient satisfaction surveys in the Healthcare Centers, incorporating a research component Table 1 Numbers of patients surveyed from each Healthcare center during year 1 (Fall, 2005), year 2 (Summer, 2007), and year 3 (Summer 2010) HCC Location Year 1 # surveyed Year 2 # surveyed Year 3 # surveyed TOTAL surveyed 1 Inner City (PCOM) 40 68 90 198 2 Urban (PCOM) 34 54 69 157 3 Inner City (PCOM) 21 43 70 134 4 Rural (PCOM) 30 19 25 74 5 Urban (PCOM) 25 51 45 121 6 Inner City (non-PCOM) TOTALS 50 75 80 205 200 310 379 889 Thornton et al. BMC Health Services Research (2017) 17:361 involving graduate students interested in health related careers. Surveys were administered to patients at the five HCCs. Patient questions were adapted from the validated DiTomasso-Willard Patient Satisfaction Questionnaire [19] (questions are listed in Table 2). Demographic information and responses to open-ended questions were also collected. In 2005 (year 1), 2007 (year 2), and 2010 (year 3), students in a master’s program at the medical school approached patients in the waiting areas at each HCC asking them to complete a survey. Patients could take the surveys with them into the examination room, but they returned the survey before leaving the HCC. If requested, the student helped a patient read the questions. Each surveying period was conducted over an approximately one month of time. Students varied their sampling by time of day and day of week. Therefore, the sample was comprised of a random representation of patients attending each HCC during each one-month period of surveying. The students approached anyone who was in the waiting room during sampling times, but Page 3 of 9 patients were free to refuse if they wished. The goal was to obtain approximately 10% of the average number of patients seen by each HCC in a month. The protocol (Protocol #H05-022X) was approved by the Institutional Review Board (IRB) of PCOM that determined it to be exempt from informed consent requirements under 45 CFR 46.101(b)(2)–survey research in which the responses will be recorded in such a manner that the human subjects cannot be identified, directly or through identifiers linked to the subjects (e.g., name, Social Security number). Further, no master list existed linking such identifiers to the subjects. Approximately 5–15% of the average numbers of patients coming to each HCC in a month were surveyed. Inclusion criteria included patients willing to respond, patient age of at least 18 years, and patients who spoke English. Patients were assured the questionnaire was confidential without any identifying information, the results would be presented in aggregate form, and that their responses would not affect their specific care at the HCC. In order to maintain anonymity, a patient’s medical status was Table 2 Grouping of the 21 survey questions using factor analysis, Rotated Component Matrix Component: Question: 1 2 3 Q1. During a typical visit, my doctor spends enough time explaining my medical condition to me. 0.773 0.151 0.162 Q2. My doctor gives me the best quality of care. 0.869 0.190 0.133 Q3. I would recommend my doctor to friends. 0.827 0.191 0.112 Q4. The staff are helpful to the patients. 0.311 0.562 0.127 Q5. My doctor uses technical terms that confuse me. 0.109 −0.200 0.628 Q6. My doctor is available when I need him/her.b 0.412 0.491 0.076 Q7. The waiting room time is too long. −0.084 0.402 0.442 Q8. My doctor really follows through. 0.751 0.230 0.082 Q9. I plan to return to this center for care. 0.713 0.370 0.148 Q10. It’s easy to get an appointment when I need one. 0.223 0.665 0.122 Q11. My doctor wastes time talking about things that don’t really matter to me.a 0.271 −0.014 0.702 Q12. My doctor treats the “whole” person. 0.640 0.300 0.143 Q13. The staff accommodates my needs over the phone. 0.241 0.677 0.070 Q14. I am satisfied with the quality of the medical care I receive here. 0.724 0.398 0.171 a b Q15. I receive prompt attention while waiting in this facility. 0.285 0.658 0.132 Q16. I have to tell my story several times before getting an appointment.a 0.001 0.409 0.631 Q17. I am treated the same as other people who get care here. 0.366 0.511 0.105 Q18. Check-out time at the front desk is too time-consuming.a −0.030 0.338 0.648 Q19. I would not recommend this center to a friend. 0.250 0.101 0.530 Q20. Everything seems so confusing at this center.a 0.199 0.160 0.731 Q21. When I’m sick I can get an appointment pretty quickly. 0.229 0.712 0.057 a Component 1: Satisfaction with Doctor (Questions 1, 2, 3, 8, 9, 12, 14) Component 2: Availability/Convenience (Questions 4, 10, 13, 15, 17, 20) Component 3: Orderly/Time (Questions 5, 11, 16, 18, 19, 20) a Questions worded in the negative were reversed for statistical analysis b Question not classified by component Thornton et al. BMC Health Services Research (2017) 17:361 not requested, although in retrospect, it may have been helpful. From observation, students reported that those with acute medical issues were less inclined to participate. Although an absolute count was not performed, students who administered surveys consistently estimated that only about 5% of the patients in the waiting room refused to participate. Survey results were entered into IBM’s Statistical Package for the Social Sciences (SPSS 18.0) for analysis. Missing data were filled in using Linear Interpolation, and any negative questions were transformed to the positive on the Likert scale, so that, for all questions, 5 (strongly agree) meant “most satisfied.” All 21 survey statements were subjected to a principal components varimax rotated factor analysis according to Kaiser’s criterion [20] which ultimately allowed for a reduction of statements into three classifiable components, Satisfaction with Physician, Availability/Convenience, and Orderly/Time (Table 2). Following each survey period, the data were analyzed in SPSS to collapse the questions into three classifiable components/categories. These three categories did not vary during the 3 data collection periods. After each survey period, study staff attended face-to-face meetings with Medical Directors of each healthcare center, the Dean of the Medical School, and the Chair of Family Medicine to present the results. HCC staff were provided with mean scores for each question for their HCC compared with a composite of all HCC’s. They also received the data collapsed into the three categories for their HCC compared with a composite of all HCC’s, but without statistical analysis. For analysis of the composite data, multivariate analysis of variance (MANOVA) was performed for groups of data, using post hoc Tukey to distinguish specific significance between groups. Independent t-test was used for gender analysis, and Chi square analysis was done to compare the observed gender data from patients who completed surveys with patient demographics of each HCC. See Additional Data for more specific information. In using factor analysis, it is common practice to require 10 subjects per number of items. In the present case, this criterion was far exceeded. For the separate MANOVA analyses using 3 dependent variables, setting power at 95% for a medium effect size at the 0.05 level of significance comparing 2 levels (male vs. female) of the independent variable, 3 levels (3 locations) and 5 levels (physician age groups), the required number of subjects was 280, 171, and 145 respectively. In all cases there was sufficient power. Results Surveys were administered to a total of 889 patients who visited one of the HCCs for treatment (Table 1). These Page 4 of 9 numbers represented between 5–15% of the average number of patients seen monthly in the affiliated HCCs, and comparable numbers of surveys were obtained from the much larger, non-affiliated HCC. Applying principal components varimax rotated factor analysis to the survey responses resulted in groups of identifiable questions that constituted factors (Rotated component matrix for all questions is shown on Table 2). Three classifiable factors, Satisfaction with Physician, Availability/Convenience, and Orderly/Time, emerged from the analysis and are used throughout this research. Two questions (Q6 and Q7) were not included as the items did not load on any of the factors (Table 2). Using the survey questions that constituted each factor (Table 2), the three factors have the following characteristics: Satisfaction with Physician involves being satisfied with the quality of medical care received, as well as the physician spending enough time with the patient. Availability/Convenience involves being satisfied with the staff and their helpfulness in making appointments, whether in person or by phone. Orderly/Time has to do with patients’ time being respected, and interactions with staff and physicians being clear and to the point, avoiding confusion. Overall, patients were quite satisfied with their HCCs, as evidenced by overall mean scores greater than 3.89 on a Likert scale of 1–5 (see Additional file 1: Table S3A). Mean scores were highest in Satisfaction with Physician (4.27 ± 0.65), while Availability/Convenience (3.92 ± 0.69) and Orderly/Time (3.89 ± 0.66) were somewhat lower. Even so, a score of 3.9 represents the top 20–25% of satisfaction. The open-ended responses emphasized the importance of patients’ satisfaction with their physician, even if patients were somewhat less satisfied with other aspects of their visit (see Additional file 2: Table S6). The goal of this research was to identify areas found to be statistically significant. More complete data can be found in the Additional files 1, 2, 3,and 4. Based on MANOVA, there was no significance over time in any of the three categories (see Additional file 1: Table S3B). This points to a consistency over time in the operations and functioning of these HCC’s. The following areas were found to be statistically significant by MANOVA: Analyzing satisfaction in inner city, urban and rural HCCs (Fig. 1), significance was observed in the following area.: Patients in inner city HCCs were less satisfied than those in urban or rural HCC’s on Availability/Convenience, and those in inner city HCCs were less satisfied than urban patients in the area of Orderly/Time. Urban patients were more satisfied with their Physician than were rural patients while inner city patients’ satisfaction with Thornton et al. BMC Health Services Research (2017) 17:361 Fig. 1 Satisfaction by location (Inner City, Urban and Rural). Lines/ Brackets indicate comparisons by color that were significantly different in each of the categories their Physician was not significantly different from the other localities (See Additional file 1: Table S3C, for more detail). When individual HCCs were analyzed (Fig. 2), one urban HCC (#5) had significantly higher satisfaction with their Physician than the other urban HCC (#2) or one inner city HCC (#6). The other urban HCC (#2) had more satisfaction in the category of Orderly/Time than two of the three inner city HCCs (#3 and #6). Two inner city HCCs (#1 and #6) had significantly lower satisfaction in the category of Availability/Convenience than the rural HCC (#4). (See Additional file 1: Table S3D, for details.) Patients’ demographics appear to play a role in the level of satisfaction. Patients over 60 years old were more satisfied with the Availability/Convenience of the HCC than patients who were in their 40′s (Fig. 3). Those with more education (in the range from graduating high school through graduate Fig. 2 Satisfaction by individual HCCs. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories Page 5 of 9 school) were more satisfied with the Orderly/Time category than those with less than a high school diploma (Fig. 4). Finally, patients who had been with their HCC for longer periods of time were more satisfied than those who had been there less than 5 years in all three categories of satisfaction with Physician, Availability/Convenience, and Orderly/ Time (Fig. 5) (See Additional file 2: Table S4C, for details). Physicians in these centers tended to longevity in their positions. Patients were more satisfied with their Physicians who were in their 30′s and 40′s than with physicians in their 50′s (Fig. 6). Also, physicians in their 50′s were perceived to be more Available than those in their 60′s. Patients rated male physicians as more Available than female physicians, and in the Inner City HCCs, patients rated their Caucasian physicians higher on Availability than African American physicians (see Additional file 3: Table S5B and C). Open ended responses were overall very positive, with the exception of the rural HCC4 during year 1. After personnel replacements at this HCC, more positive responses were also seen there. Wait times were seen as a problem in some HCCs, particularly in the inner city centers. The majority of patients were very satisfied with the convenience of their HCC (See Additional file 4: Table S6). Discussion In the examination of changes over time, patient satisfaction at the HCCs in the study remained overall quite high in all three categories of Satisfaction with Physician, Availability/Convenience, and Orderly/Time. Meanwhile, notable changes at the affiliated centers during this time period included a new telephone system installed between years 1–2, major renovations of one of the centers in year 2, and installation of a system of Electronic Fig. 3 The effect of patient age on satisfaction. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories Thornton et al. BMC Health Services Research (2017) 17:361 Fig. 4 The effect of patient education on satisfaction. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories Medical Records (EMR) in all affiliated HCCs between years 2 and 3; the non-affiliated HCC #6 also introduced EMR prior to year 3. While telephone changes would likely affect staff-patient interactions, instituting EMR represented a major change in the physician-patient interactions, with the addition of computers to each examination room. We were surprised that these seemingly “major” changes did not significantly affect the satisfaction levels over this time period. De Leon et al. found generally higher patient satisfaction with a center after EMR were introduced [21], while we found no significant differences after EMR was installed. Results of the patient satisfaction surveys were presented to Medical Directors and staff in a timely manner after each survey period, but without statistical analysis. From the initial data given to each HCC, medical staff could compare their mean results with a composite mean result for all the centers. However, they did not Fig. 5 The effect of length of time at a HCC on satisfaction. Lines/ Brackets indicate comparisons by color that were significantly different in each of the categories Page 6 of 9 Fig. 6 The effect of physician’s age on satisfaction. Lines/Brackets indicate comparisons by color that were significantly different in each of the categories have access to comparisons of individual HCCs (see Additional file 1, Table S3D). Nor did they have access to figures such as Fig. 2, comparing individual HCCs. It is not surprising that each HCC is unique. An example is HCC 5 with a significantly higher level of satisfaction with Physician compared with two other HCCs, one urban and inner city (Fig. 2, and Additional file 1: Table S3D). This merits more in-depth analysis of the physician practices at this outstanding urban HCC as a positive example for others. We projected that HCCs sharing similar locations (inner city, urban or rural) would be more alike and this proved to be the case. In the components of Availability/ Convenience and Orderly/Time, patients in the inner city HCCs were less satisfied than those in urban or rural settings, consistent with findings of the individual HCCs. There could be several reasons for differences between inner city and other HCCs. Fan et al. found that functional status (disease severity, physical limitation) was only weakly associated with general satisfaction, while education, coping skills and disease perception were more important to patient’s satisfaction [4]. Patients in the inner city may be sicker due to overall inadequate health knowledge or reluctance to visit a doctor, possibly due to lack of insurance. These findings suggest that physician-patient interactions with the goal of improved disease understanding might help as much as actual improvement in health. We did not ask for the health status of individual patients in our survey, so we can only guess the health status of patients at different locations. Comparing locations (Fig. 1) with individual HCCs (Fig. 2) reveals the sources of these differences. For example, in Fig. 1, inner city patients were statistically less satisfied in the component of Orderly/Time than were patients in urban settings. Fig. 2 shows that the differences were primarily with one urban HCC #2 (but not with urban HCC #5), compared only with 2 inner city Thornton et al. BMC Health Services Research (2017) 17:361 HCC #3 and #6 (but not with inner city HCC #1). So generalizations require examining the individual HCCs as well. Education level of the patient can also be reflective of location. Approximately 80% of inner city respondents reported having high school education or less, similar to rural patients (76%), while only 58% of urban patients had a high school education or less (data not shown). Other issues facing patients, such as availability of public transportation, may be more of an obstacle in the inner city than in either urban or rural settings. While public transportation is also not widely available in rural settings, it is likely most patients have access to a vehicle. Inner city respondents also were less satisfied in the component of Orderly/Time than respondents in urban settings, and this is confirmed in the open-ended questions (see Additional file 4: Table S6) where a larger number of patients specifically mentioned the wait time as a problem in the inner city HCCs than in the urban or rural HCCs. Although we wondered if dissatisfaction with wait time could be directly attributable to student participation in the examination room, that seems not to be the case, as a very small percent of responders mentioned students in the open-ended questions and half of those were positive. Mol et al. found that patients generally felt neutral or positive about the presence of students, and in that study, between 83 and 98% of patients consented to student participation [22]. Our only finding of differences associated with education level in satisfaction were in the area of Orderly/ Time; patients with less than a high school education were less satisfied in the component of Orderly/Time than any other group. This could be due to their inability to understand the medical parlance or the protocols involved in their care. However, one study also found that the converse-a physician’s satisfaction with a patient-was associated with their patients’ higher education level [23], suggesting that the responsibility may be reciprocal between the physician and the patient. Another patient demographic of age can also contribute to patient satisfaction. Our finding that patients over 60 years old had a higher degree of satisfaction in Availability/Convenience is not surprising. This finding agrees with Jackson who reported that patients over 65 years old and with higher functional status were more satisfied [3]. Peck found that physicians were more likely to have patient-centered encounters with patients over age 65, which in turn meant that older patients were more satisfied [24]. Although there was no impact of patients’ gender on level of satisfaction, we did find that, in general, more female patients agreed to fill out the surveys than were actually represented as patients in the HCCs. Not surprisingly, the most significant differences were found in the length of time a patient had been attending their Page 7 of 9 HCC. This is undoubtedly a self-selection, where either the physician or the location suits the patient who continues to visit that center. Pelletier calls this “sampling bias,” citing that “those who stay with a program…may be those who are most satisfied” [23]. Another explanation is through “visit continuity,” where respondents rated the quality of physician-patient interaction as being more important during the early stages of continuity or when the patient reported worse self-rated health [25]. This suggests that physicians who focus on those newer patients or sicker patients who would benefit the most from additional interactions may have the most positive results over time. Demographics of the physician may also be important to patient satisfaction. The physicians at the affiliated HCCs were all osteopathic (DO) physicians, who self-reported that they used Osteopathic Manipulative Treatment at their clinics about 20% of the time. In the open-ended questions, some patients did express a preference for DO physicians. On age of physicians, it appears that more patients prefer a physician younger than 50 years old in the component of Satisfaction with Physician, but in Orderly/ Time, they prefer a physician in their 50′s rather than in their 60′s. We speculate that physicians in their 50′s are likely to be at the pinnacle of their profession, although other considerations may also be important, such as humor or degree of connection that the patient perceives with that physician. In the variable of Orderly/Time, it is possible that physicians in their 50′s may be more efficient, having a well-run visit, while the slower, possibly more thorough pace of older physicians may not be as appreciated. The statistical significances found in this data enhance the details which were presented to the medical directors after each surveying period and provide additional measures of patient satisfaction. Presenting the data to medical directors in figure form rather than as graphs is likely to enhance understanding. Finally, presenting the data of each individual HCC rather than as a composite may help medical directors to see the larger picture. The present study has several limitations: In retrospect from patients’ written responses, an additional choice under the education demographic would have better captured any additional education received, such as technical certificates or Associate degrees. Also, the severity of the patient’s medical condition should have been noted, as this has been shown to influence patient satisfaction [4]. In addition, the questions that fell under the component Orderly/Time in the factor analysis fortuitously contained all questions which had been originally stated in the negative and then were reversed for analysis. Finally, presentation of the data to the medical directors in a timely fashion could be improved by presenting figures in addition to tables, and showing results of each individual HCC. Thornton et al. BMC Health Services Research (2017) 17:361 Conclusions This study was designed to provide feedback to Medical Directors on patient satisfaction in their HCCs. Our findings point to a consistency in the operations and functioning of these HCCs over time, even when renovations or installation of EMR were performed. Differences in locality (inner city, urban, rural) were found, as well as differences in satisfaction by patient demographics (age, education level, length of time with a HCC) and by physician demographics (age, gender). However, uniqueness of individual HCCs contributes to these differences. Physicians from each HCC regularly meet together, and they can use these meetings to help better understand and build on their strengths and individuality. Results of this study can be used to increase satisfaction if physicians help their patients benefit from their services and increase their satisfaction. Particularly, physicians can concentrate on providing additional high-quality interactions for patients with less education, those who are sicker, and those who are new to the HCC. Additional files Additional file 1: Table S3. Comparisons overall and by time, location, individual HCCs vs. 3 components. (DOC 41 kb) Additional file 2: Table S4. Patient demographics vs. three factors. *refers to higher mean score; ns, not significant. (DOC 44 kb) Additional file 3: Table S5. Physician demographics vs. three factors. *refers to higher mean score; ns, not significant. (DOC 40 kb) Additional file 4: Table S6. Open-ended questions by healthcare center and year. (+) refers to positive statements, what did you like most? (−) refers to negative statements, what did you like least? (DOC 48 kb) Abbreviations DO: Doctor of Osteopathic Medicine; EMR: Electronic Medical Records; HCCs: Healthcare Centers; IRB: Institutional Review Board; MANOVA: Multivariate analysis of variants; PCOM: Philadelphia College of Osteopathic Medicine; SPSS: Statistical Package for the Social Sciences Acknowledgements The authors wish to acknowledge Audrey Rossowski, M.S. for collecting surveys at the healthcare centers and Barbara A. Mitchell, Ph.D. for helpful editing. We also sincerely thank PCOM Kenneth Veit, DO, MBA, FACOFP, Provost, Senior Vice President for Academic Affairs and Dean. We also thank PCOM Family Medicine Chair and Director Harry Morris, DO; PCOM Healthcare Center Physicians and Medical Directors, Oliver Bullock, DO., Michael Becker, DO, Izola David, DO, Larry Finklestein, DO, David Kuo, DO, Marta Motel, DO, Barbara Williams-Page, DO, and David Wood, DO; and A. Scott McNeal, DO, Fairmount Primary Care Center, for allowing our participation. An abstract of this research was presented by Kenice Frank, M.S., at the Student National Medical Association’s National Meeting, May, 2012, Atlanta, GA. Funding No funding sources were used for this research project. Availability of data and materials The SPSS data and subsequent analysis is available from RDT and the corresponding author, RAD, on reasonable request. A large part of the data is already shown in Additional files 1, 2, 3, and 4, associated with this publication. Page 8 of 9 Authors’ contributions RDT conceived of the research, wrote the paper, and, with the help of RAD, analyzed the data. NN received her M.S. degree from PCOM for her research on the first year (2005) of the study. She gathered the data, analyzed it, and presented a thesis, “A measure of patient satisfaction of PCOM healthcare centers based upon geographic settings.” LS received her M.S. degree from PCOM for her research on the first year (2005) of the study. She gathered the data, analyzed it, and presented a thesis, “The effects of demographic factors on patient satisfaction in an ambulatory setting.” SH-Z received her M.S. degree from PCOM for her research on the second year (2007) of the study. She gathered the data for 2007, analyzed the 2007 data and compared it with 2005 data, and presented a thesis, “Examining improvement levels in measured patient satisfaction in an academic primary care setting.” KF gathered the data on the third year (2010) of the study. She analyzed the 2010 data, compared it with the 2005 and 2007 data, and presented a poster on the study at the Student National Medical Association in 2011. RAD gave guidance to the entire study, providing the original survey questions [17], analyzing statistically all of the data, serving on each M.S. student’s thesis committee, and editing the paper. All authors read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Consent for publication Not applicable. Ethics approval and consent to participate The protocol was approved by the Institutional Review Board (IRB) at Philadelphia College of Osteopathic Medicine that determined it to be exempt from informed consent requirements under 45 CFR 46.101(b)(2)survey research in which the responses will be recorded in such a manner that the human subjects cannot be identified, directly or through identifiers linked to the subjects (e.g., name, Social Security number). Further, no master list existed linking such identifiers to the subjects. Inclusion criteria included patients willing to respond, patient age of at least 18 years, and patients who spoke English. Patients were assured the questionnaire was confidential without any identifying information, the results would be presented in aggregate form, and that their responses would not affect their specific care at the HCC. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 Department of Psychology, Philadelphia College of Osteopathic Medicine, 4170 City Ave., Philadelphia, PA 19131, USA. 2North Philadelphia Health System/St. Joseph’s Hospital, Philadelphia, PA, USA. 3Geisinger Medical Center, Danville, PA, USA. 4Maria Fareri Children’s Hospital, Westchester, NY, USA. 5College of Podiatric Medicine & Surgery, Des Moines University, Des Moines, IO, USA. Received: 3 April 2017 Accepted: 11 May 2017 References 1. Rider EA, Perrin JM. Performance profiles: the influence of patient satisfaction data on physicians’ practice. Pediatrics. 2002;109(5):752–7. 2. Burroughs TE, Waterman BM, Gilin D, Adams D, McCollegan J, Cira J. Do onsite patient satisfaction surveys bias results? Jt Comm J Qual Pt Safety. 2005; 31(3):158–66. 3. Jackson JL, Chamberlin J, Kroenke K. Predictors of patient satisfaction. Soc Sci Med. 2001;52(4):609–20. 4. Fan VS, Reiber GE, Diehr P, Burman M, McDonell MB, Fihn SD. Functional status and patient satisfaction: a comparison of ischemic heart disease, obstructive lung disease, and diabetes mellitus. J Gen Intern Med. 2005; 20(5):452–9. 5. Fan VS, Burman M, McDonell MB, Fihn SD. Continuity of care and other determinants of patient satisfaction with primary care. J Gen Intern Med. 2005;20(3):226–33. Thornton et al. BMC Health Services Research (2017) 17:361 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Page 9 of 9 Fairchild DG, McLoughlin KS, Gharib S, Horsky J, Portnow M, Richter J, Gagliano N, Bates DW. Productivity, quality, and patient satisfaction: comparison of part-time and full-time primary care physicians. J Gen Intern Med. 2001;16(10):663–7. Monk SM, Nanagas MT, Fitch JL, Stolfi A, Pickoff AS. Comparison of resident and faculty patient satisfaction surveys in a pediatric ambulatory clinic. Teach Learn Med. 2006;18(4):343–7. Rodriguez HP, von Glahn T, Chang H, Rogers WH, Safran DG. Measuring patients’ experiences with individual specialist physicians and their practices. Am J Med Qual. 2009;24(1):35–44. Ware Jr JE, Davies AR. Behavioral consequences of consumer dissatisfaction with medical care. Eval Prog Plan. 1983;6(3–4):291–7. Perneger TV, Etter JF, Raetzo MA, Schaller P, Stalder H. Comparison of patient satisfaction with ambulatory visits in competing health care delivery settings in Geneva, Switzerland. J Epidemiol Comm Health. 1996;50(4):463–8. Feddock CA, Hoellein AR, Griffith 3rd CH, Wilson JF, Bowerman JL, Becker NS, Caudill TS. Can physicians improve patient satisfaction with long waiting times? Eval Health Prof. 2005;28(1):40–52. Anderson RT, Camacho FT, Balkrishnan R. Willing to wait?: the influence of patient wait time on satisfaction with primary care. BMC Health Serv Res. 2007;7:31. Street Jr RL, O’Malley KJ, Cooper LA, Haidet P. Understanding concordance in patient-physician relationships: personal and ethnic dimensions of shared identity. Ann Fam Med. 2008;6(3):198–205. Thornton RLJ, Powe NR, Roter D, Cooper LA. Patient-physician social concordance, medical visit communication and patients’ perceptions of health care quality. Pat Educ Couns. 2011;85:e201–8. Gooding TD, Newcomb L, Mertens K. Patient-centered measurement at an academic medical center. Jt Comm J Qual Improv. 1999;25(7):343–51. Conner DR. Managing at the Speed of Change. 1st ed. Ed Anonymous New York: Villard Books; 1993. p. 105–24. US Dept of Agriculture. Econ Res Service. Rural Classification. http://www. ers.usda.gov/topics/rural-economy-population/rural-classifications/what-isrural.aspx. Accessed 2016. Merriam-Webster Dictionary. Urban & Inner City. http://www.merriamwebster.com/dictionary. Accessed 2016. DiTomasso RA, Willard M. The development of a patient satisfaction questionnaire in the ambulatory setting. Fam Med. 1991;23(2):127–31. Field A. Discovering statistics using SPSS. 2nd ed. 2005. De Leon DF, Silfen SL, Wang JJ, Kamara TS, Wu WY, Shih SC. Patient experiences at primary care practices using electronic health records. J Med Pract Manage. 2012;28(3):169–76. Mol SSL, Peelen JH, Kuyvenhoven MM. Patients’ view on student participation in general practice consultations: A comprehensive review. Med Teacher. 2011;33:e397–400. Pelletier M. Client satisfaction surveys: variables to watch out for. Dimens Health Serv. 1985;62(1):37–9. Peck BM. Age-related differences in doctor-patient interaction and patient satisfaction. Curr Gerontol Geriatr Res. 2011;2011:137492–501. Rodriguez HP, Rogers WH, Marshall RE, Safran DG. The effects of primary care physician visit continuity on patients’ experiences with care. J Gen Intern Med. 2007;22(6):787–93. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit 2 3 Quick Guide to Bivariate Statistical Tests 4 5 6 Acquisitions Editor: Christina Burns Product Development Editor: Katherine Burland Editorial Assistant: Cassie Berube Marketing Manager: Dean Karampelas Production Project Manager: Cynthia Rudy Design Coordinator: Joan Wendt Manufacturing Coordinator: Karin Duffield Prepress Vendor: Absolute Service, Inc. Tenth edition Copyright © 2017 Wolters Kluwer. Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins. Copyright © 2008, 2004, 1999 by Lippincott Williams & Wilkins. Copyright © 1995, 1991, 1987, 1983, 1978 by J. B. Lippincott Company. All rights reserved. This book is protected by copyright. No part of this book may be reproduced or transmitted in any form or by any means, including as photocopies or scanned-in or other electronic copies, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotations embodied in critical articles and reviews. Materials appearing in this book prepared by individuals as part of their official duties as U.S. government employees are not covered by the above-mentioned copyright. To request permission, please contact Wolters Kluwer at Two Commerce Square, 2001 Market Street, Philadelphia, PA 19103, via email at permissions@lww.com, or via our website at lww.com (products and services). 987654321 Printed in China Library of Congress Cataloging-in-Publication Data Polit, Denise F., author. Nursing research : generating and assessing evidence for nursing practice / Denise F. Polit, Cheryl Tatano Beck. — Tenth edition. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4963-0023-2 I. Beck, Cheryl Tatano, author. II. Title. [DNLM: 1. Nursing Research—methods. WY 20.5] RT81.5 610.73072—dc23 7 2015033543 This work is provided “as is,” and the publisher disclaims any and all warranties, express or implied, including any warranties as to accuracy, comprehensiveness, or currency of the content of this work. This work is no substitute for individual patient assessment based on healthcare professionals’ examination of each patient and consideration of, among other things, age, weight, gender, current or prior medical conditions, medication history, laboratory data, and other factors unique to the patient. The publisher does not provide medical advice or guidance, and this work is merely a reference tool. Healthcare professionals, and not the publisher, are solely responsible for the use of this work including all medical judgments and for any resulting diagnosis and treatments. Given continuous, rapid advances in medical science and health information, independent professional verification of medical diagnoses, indications, appropriate pharmaceutical selections and dosages, and treatment options should be made and healthcare professionals should consult a variety of sources. When prescribing medication, healthcare professionals are advised to consult the product information sheet (the manufacturer’s package insert) accompanying each drug to verify, among other things, conditions of use, warnings, and side effects and identify any changes in dosage schedule or contraindications, particularly if the medication to be administered is new, infrequently used, or has a narrow therapeutic range. To the maximum extent permitted under applicable law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property, as a matter of products liability, negligence law or otherwise, or from any reference to or use by any person of this work. LWW.com 8 TO Our Beloved Family: Our Husbands, Our Children (Spouses/Fiancés), and Our Grandchildren Husbands: Alan Janosy and Chuck Beck Children: Alex (Maryanna), Alaine (Jeff), Lauren (Vadim), and Norah (Chris); and Curt and Lisa Grandchildren: Cormac, Julia, Maren, and Ronan 9 Acknowledgments This 10th edition, like the previous nine editions, depended on the contribution of dozens of people. Many faculty and students who used the text have made invaluable suggestions for its improvement, and to all of you we are very grateful. In addition to all those who assisted us during the past 35 years with the earlier editions, the following individuals deserve special mention. We would like to acknowledge the comments of reviewers of the previous edition of this book, anonymous to us initially, whose feedback influenced our revisions. Faculty at Griffith University in Australia made useful suggestions and also inspired the inclusion of some new content. Valori Banfi, reference librarian at the University of Connecticut, provided ongoing assistance. Dr. Deborah Dillon McDonald was extraordinarily generous in giving us access to her NINR grant application and related material for the Resource Manual. We also extend our thanks to those who helped to turn the manuscript into a finished product. The staff at Wolters Kluwer has been of great assistance to us over the years. We are indebted to Christina Burns, Kate Burland, Cynthia Rudy, and all the others behind the scenes for their fine contributions. Finally, we thank our family and friends. Our husbands Alan and Chuck have become accustomed to our demanding schedules, but we recognize that their support involves a lot of patience and many sacrifices. 10 Reviewers Ellise D. Adams, PhD, CNM 11 Associate Professor The University of Alabama in Huntsville Huntsville, Alabama Jennifer Bellot, PhD, RN, MHSA Associate Professor and Director, DNP Program Thomas Jefferson University Philadelphia, Pennsylvania Kathleen D. Black, PhD, RNC Assistant Professor, Jefferson College of Nursing Thomas Jefferson University Philadelphia, Pennsylvania Dee Campbell, PhD, APRN, NE-BC, CNL Professor, Graduate Department Felician College, School of Nursing Lodi, New Jersey Patricia Cannistraci, DNS, RN, CNE 12 Assistant Dean 13 Excelsior College Albany, New York Julie L. Daniels, DNP, CNM 14 Assistant Professor Frontier Nursing University Hyden, Kentucky Rebecca Fountain, PhD, RN 15 Associate Professor University of Texas at Tyler Tyler, Texas Teresa S. Johnson, PhD, RN Associate Professor, College of Nursing University of Wisconsin—Milwaukee Milwaukee, Wisconsin Jacqueline Jones, PhD, RN, FAAN Associate Professor, College of Nursing University of Colorado, Anschutz Medical Campus Aurora, Colorado Mary Lopez, PhD, RN Associate Dean, Research Western University of Health Sciences Pomona, California Audra Malone, DNP, FNP-BC 16 Assistant Professor Frontier Nursing University Hyden, Kentucky Sharon R. Rainer, PhD, CRNP Assistant Professor, Jefferson College of Nursing Thomas Jefferson University Philadelphia, Pennsylvania Maria A. Revell, PhD, RN 17 Professor of Nursing Middle Tennessee State University Murfreesboro, Tennessee Stephanie Vaughn, PhD, RN, CRRN Interim Director, School of Nursing California State University, Fullerton Fullerton, California 18 Preface Research methodology is not a static enterprise. Even after writing nine editions of this book, we continue to draw inspiration and new material from groundbreaking advances in research methods and in nurse researchers’ use of those methods. It is exciting and uplifting to share many of those advances in this new edition. We expect that many of the new methodologic and technologic advances will be translated into powerful evidence for nursing practice. Five years ago, we considered the ninth edition as a watershed edition of a classic textbook. We are persuaded, however, that this 10th edition is even better. We have retained many features that made this book a classic textbook and resource, including its focus on research as a support for evidence-based nursing, but have introduced important innovations that will help to shape the future of nursing research. N E W TO T H I S E D I T I O N New Chapters We have added two new chapters on “cutting-edge” topics that are not well covered in any major research methods textbook, regardless of discipline. The first is a chapter on an issue of critical importance to health professionals and yet inadequately addressed in the nursing literature: the clinical significance of research findings. In Chapter 20, we discuss various conceptualizations of clinical significance and present methods of operationalizing those conceptualizations so that clinical significance can be assessed at both the individual and group level. We believe that this is a “must-read” chapter for nurses whose research is designed to inform clinical practice. The second new chapter in this edition concerns the design and conduct of pilot studies. In recent years, experts have written at length about the poor quality of many pilot studies. Chapter 28 provides guidance on how to develop pilot study objectives and draw conclusions about the appropriate next step—that is, whether to proceed to a full-scale study, make major revisions, or 19 abandon the project. This chapter is included in Part 5 of this book, which is devoted to mixed methods research, because pilots can benefit from both qualitative and quantitative evidence. New Content Throughout the book, we have included material on methodologic innovations that have arisen in nursing, medicine, and the social sciences during the past 4 to 5 years. The many additions and changes are too numerous to describe here, but a few deserve special mention. In particular, we have totally revised the chapters on measurement (Chapter 14) and scale development (Chapter 15) to reflect emerging ideas about key measurement properties and the assessment of newly developed instruments. The inclusion of two new chapters made it challenging to keep the textbook to a manageable length. Our solution was to move some content in the ninth edition to supplements that are available online. In fact, every chapter has an online supplement, which gave us the opportunity to add a considerable amount of new content. For example, one supplement is devoted to evidence-based methods to recruit and retain study participants. Other supplements include a description of various randomization methods, an overview of item response theory, guidance on wording proposals to conduct pilot studies, and a discussion of quality improvement studies. Following is a complete list of the supplements for the 31 chapters of this textbook: 1. The History of Nursing Research 2. Evaluating Clinical Practice Guidelines—AGREE II 3. Deductive and Inductive Reasoning 4. Complex Relationships and Hypotheses 5. Literature Review Matrices 6. Prominent Conceptual Models of Nursing Used by Nurse Researchers, and a Guide to Middle-Range Theories 7. Historical Background on Unethical Research Conduct 8. Research Control 9. Randomization Strategies 10. The RE-AIM Framework 11. Other Specific Types of Research 12. Sample Recruitment and Retention 13. Other Types of Structured Self-Reports 14. Cross-Cultural Validity and the Adaptation/Translation of Measures 15. Overview of Item Response Theory 16. SPSS Analysis of Descriptive Statistics 17. SPSS Analysis of Inferential Statistics 18. SPSS Analysis and Multivariate Statistics 19. Some Preliminary Steps in Quantitative Analysis Using SPSS 20. Clinical Significance Assessment with the Jacobson-Truax Approach 21. Historical Nursing Research 22. Generalizability and Qualitative Research 23. Additional Types of Unstructured Self-Reports 24. Transcribing Qualitative Data 25. Whittemore and Colleagues’ Framework of Quality Criteria in Qualitative Research 26. Converting Quantitative and Qualitative Data 27. 20 Complex Intervention Development: Exploratory Questions 28. Examples of Various Pilot Study Objectives 29. Publication Bias in Meta-Analyses 30. Tips for Publishing Reports on Pilot Intervention Studies 31. Proposals for Pilot Intervention Studies Another new feature of this edition concerns our interest in readers’ access to references we cited. To the extent possible, the studies we have chosen as examples of particular research methods are published as openaccess articles. These studies are identified with an asterisk in the reference list at the end of each chapter, and a link to the article is included in the Toolkit section of the Resource Manual. We hope that these revisions will help users of this book to maximize their learning experience. O R G A N I Z AT I O N O F T H E T E X T The content of this edition is organized into six main parts. • Part I—Foundations of Nursing Research and Evidence-Based Practice introduces fundamental concepts in nursing research. Chapter 1 briefly summarizes the history and future of nursing research, discusses the philosophical underpinnings of qualitative research versus quantitative research, and describes major purposes of nursing research. Chapter 2 offers guidance on utilizing research to build an evidence-based practice. Chapter 3 introduces readers to key research terms and presents an overview of steps in the research process for both qualitative and quantitative studies. • Part II—Conceptualizing and Planning a Study to Generate Evidence further sets the stage for learning about the research process by discussing issues relating to a study’s conceptualization: the formulation of research questions and hypotheses (Chapter 4), the review of relevant research (Chapter 5), the development of theoretical and conceptual contexts (Chapter 6), and the fostering of ethically sound approaches in doing research (Chapter 7). Chapter 8 provides an overview of important issues that researchers must attend to during the planning of any type of study. • Part III—Designing and Conducting Quantitative Studies to Generate Evidence presents material on undertaking quantitative nursing studies. Chapter 9 describes fundamental principles and applications of quantitative research design, and Chapter 10 focuses on methods to enhance the rigor of a quantitative study, including mechanisms of research control. Chapter 11 examines research with different and distinct purposes, including surveys, outcomes research, and evaluations. Chapter 12 presents strategies for sampling study participants in quantitative research. Chapter 13 describes using structured data collection methods that yield quantitative information. Chapter 14 discusses the concept of measurement and then focuses on methods of assessing 21 the quality of formal measuring instruments. In this edition, we describe methods to assess the properties of point-in-time measurements (reliability and validity) and longitudinal measurements—change scores (reliability of change scores and responsiveness). Chapter 15 presents material on how to develop high-quality self-report instruments. Chapters 16, 17, and 18 present an overview of univariate, bivariate, and multivariate statistical analyses, respectively. Chapter 19 describes the development of an overall analytic strategy for quantitative studies, including material on handling missing data. Chapter 20, a new chapter, discusses the issue of interpreting results and making inferences about clinical significance. • Part IV—Designing and Conducting Qualitative Studies to Generate Evidence presents material on undertaking qualitative nursing studies. Chapter 21 is devoted to research designs and approaches for qualitative studies, including material on critical theory, feminist, and participatory action research. Chapter 22 discusses strategies for sampling study participants in qualitative inquiries. Chapter 23 describes methods of gathering unstructured self-report and observational data for qualitative studies. Chapter 24 discusses methods of analyzing qualitative data, with specific information on grounded theory, phenomenologic, and ethnographic analyses. Chapter 25 elaborates on methods qualitative researchers can use to enhance (and assess) integrity and quality throughout their inquiries. • Part V—Designing and Conducting Mixed Methods Studies to Generate Evidence presents material on mixed methods nursing studies. Chapter 26 discusses a broad range of issues, including asking mixed methods questions, designing a study to address the questions, sampling participants in mixed methods research, and analyzing and integrating qualitative and quantitative data. Chapter 27 presents innovative information about using mixed methods approaches in the development of nursing interventions. In Chapter 28, a new chapter, we provide guidance for designing and conducting a pilot study and using data from the pilot to draw conclusions about how best to proceed. • Part VI—Building an Evidence Base for Nursing Practice provides additional guidance on linking research and clinical practice. Chapter 29 offers an overview of methods of conducting systematic reviews that support EBP, with an emphasis on meta-analyses, metasyntheses, and mixed studies reviews. Chapter 30 discusses dissemination of evidence—how to prepare a research report (including theses and dissertations) and how to publish research findings. The concluding chapter (Chapter 31) offers suggestions and guidelines on developing research proposals and getting financial support and includes information about applying for NIH grants and interpreting scores from NIH’s 22 new scoring system. K E Y FE AT U R E S This textbook was designed to be helpful to those who are learning how to do research as well as to those who are learning to appraise research reports critically and to use research findings in practice. Many of the features successfully used in previous editions have been retained in this 10th edition. Among the basic principles that helped to shape this and earlier editions of this book are (1) an unswerving conviction that the development of research skills is critical to the nursing profession, (2) a fundamental belief that research is intellectually and professionally rewarding, and (3) a steadfast opinion that learning about research methods need be neither intimidating nor dull. Consistent with these principles, we have tried to present the fundamentals of research methods in a way that both facilitates understanding and arouses curiosity and interest. Key features of our approach include the following: • Research Examples. Each chapter concludes with one or two actual research examples designed to highlight critical points made in the chapter and to sharpen the reader’s critical thinking skills. In addition, many research examples are used to illustrate key points in the text and to stimulate ideas for a study. Many of the examples used in this edition are open-access articles that can be used for further learning and classroom discussions. • Critiquing Guidelines. Most chapters include guidelines for conducting a critique of each aspect of a research report. These guidelines provide a list of questions that draw attention to specific aspects of a report that are amenable to appraisal. • Clear, “user-friendly” style. Our writing style is designed to be easily digestible and nonintimidating. Concepts are introduced carefully and systematically, difficult ideas are presented clearly, and readers are assumed to have no prior exposure to technical terms. • Specific practical tips on doing research. This textbook is filled with practical guidance on how to translate the abstract notions of research methods into realistic strategies for conducting research. Every chapter includes several tips for applying the chapter’s lessons to real-life situations. These suggestions are in recognition of the fact that there is often a large gap between what gets taught in research methods textbooks and what a researcher needs to know to conduct a study. • Aids to student learning. Several features are used to enhance and reinforce learning and to help focus the student’s attention on specific areas of text content, including the following: succinct, bulleted summaries at the end of each 23 chapter; tables and figures that provide examples and graphic materials in support of the text discussion; study suggestions at the end of each chapter; a detailed glossary; and a comprehensive index for accessing information quickly. T E A C H I N G – L E A R N I N G PA C K A G E Nursing Research: Generating and Assessing Evidence for Nursing Practice, 10th edition, has an ancillary package designed with both students and instructors in mind. • The Resource Manual augments the textbook in important ways. The manual itself provides students with exercises that correspond to each text chapter, with a focus on opportunities to critique actual studies. The appendix includes 12 research journal articles in their entirety, plus a successful grant application for a study funded by the National Institute of Nursing Research. The 12 reports cover a range of nursing research ventures, including qualitative, quantitative, and mixed methods studies, an instrument development study, an evidencebased practice translation project, and two systematic reviews. Full critiques of two of the reports are also included and can serve as models for a comprehensive research critique. • The Toolkit to the Resource Manual is a “must-have” innovation that will save considerable time for both students and seasoned researchers. Included on thePoint, the Toolkit offers dozens of research resources in Word documents that can be downloaded and used directly or adapted. The resources reflect bestpractice research material, most of which have been pretested and refined in our own research. The Toolkit originated with our realization that in our technologically advanced environment, it is possible to not only illustrate methodologic tools as graphics in the textbook but also to make them directly available for use and adaptation. Thus, we have included dozens of documents in Word files that can readily be used in research projects, without requiring researchers to “reinvent the wheel” or tediously retype material from this textbook. Examples include informed consent forms, a demographic questionnaire, content validity forms, and a coding sheet for a meta-analysis— to name only a few. The Toolkit also has lists of relevant and useful websites for each chapter, which can be “clicked” on directly without having to retype the URL and risk a typographical error. Links to open-access articles cited in the textbook, as well as other open-access articles relevant to each chapter, are included in the Toolkit. • The Instructor’s Resources on the Point include PowerPoint slides summarizing key points in each chapter, test questions that have been placed into a program that allows instructors to automatically generate a test, and an 24 image bank. It is our hope that the content, style, and organization of this book continue to meet the needs of a broad spectrum of nursing students and nurse researchers. We also hope that this book will help to foster enthusiasm for the kinds of discoveries that research can produce and for the knowledge that will help support an evidence-based nursing practice. DENISE F. POLIT, PhD, FAAN CHERYL TATANO BECK, DNSc, CNM, FAAN 25 26 Contents PART 1: FOUNDATIONS OF NURSING RESEARCH Chapter 1: Introduction to Nursing Research in an Evidence-Based Practice Environment Chapter 2: Evidence-Based Nursing: Translating Research Evidence into Practice Chapter 3: Key Concepts and Steps in Qualitative and Quantitative Research PART 2: CONCEPTUALIZING AND PLANNING A STUDY TO GENERATE EVIDENCE FOR NURSING Chapter 4: Research Problems, Research Questions, and Hypotheses Chapter 5: Literature Reviews: Finding and Critiquing Evidence Chapter 6: Theoretical Frameworks Chapter 7: Ethics in Nursing Research Chapter 8: Planning a Nursing Study PART 3: DESIGNING AND CONDUCTING QUANTITATIVE STUDIES TO GENERATE EVIDENCE FOR NURSING Chapter 9: Quantitative Research Design Chapter 10: Rigor and Validity in Quantitative Research Chapter 11: Specific Types of Quantitative Research Chapter 12: Sampling in Quantitative Research Chapter 13: Data Collection in Quantitative Research Chapter 14: Measurement and Data Quality Chapter 15: Developing and Testing Self-Report Scales Chapter 16: Descriptive Statistics Chapter 17: Inferential Statistics 27 Chapter 18: Multivariate Statistics Chapter 19: Processes of Quantitative Data Analysis Chapter 20: Clinical Significance and Interpretation of Quantitative Results PART 4: DESIGNING AND CONDUCTING QUALITATIVE STUDIES TO GENERATE EVIDENCE FOR NURSING Chapter 21: Qualitative Research Design and Approaches Chapter 22: Sampling in Qualitative Research Chapter 23: Data Collection in Qualitative Research Chapter 24: Qualitative Data Analysis Chapter 25: Trustworthiness and Integrity in Qualitative Research PART 5: DESIGNING AND CONDUCTING MIXED METHODS STUDIES TO GENERATE EVIDENCE FOR NURSING Chapter 26: Basics of Mixed Methods Research Chapter 27: Developing Complex Nursing Interventions Using Mixed Methods Research Chapter 28: Feasibility Assessments and Pilot Tests of Interventions Using Mixed Methods PART 6: BUILDING AN EVIDENCE BASE FOR NURSING PRACTICE Chapter 29: Systematic Reviews of Research Evidence: Meta-Analysis, Metasynthesis, and Mixed Studies Review Chapter 30: Disseminating Evidence: Reporting Research Findings Chapter 31: Writing Proposals to Generate Evidence Glossary Appendix: Statistical Tables Index 28 Check Out the Latest Book Authored by Research Expert Dr. Polit If you want to make thoughtful but practical decisions about the measurement of health constructs, check out Dr. Polit and Dr. Yang’s latest book, a “gentle” introduction to and overview of complex measurement content, called Measurement and the Measurement of Change. This book is for researchers and clinicians from all health disciplines because measurement is vital to high-quality science and to excellence in clinical practice. The text focuses on the measurement of health constructs, particularly those constructs that are not amenable to quantification by means of laboratory analysis or technical instrumentation. These health constructs include a wide range of human attributes, such as quality of life, functional ability, self-efficacy, depression, and pain. Measures of such constructs are proliferating at a rapid rate and often without adequate attention paid to ensuring that standards of scientific rigor are met. 29 In this book, the authors offer guidance to those who develop new instruments, adapt existing ones, select instruments for use in a clinical trial or in clinical practice, interpret information from measurements and changes in scores, or undertake a systematic review on instruments. This book offers guidance on how to develop new instruments using both “classical” and “modern” approaches from psychometrics as well as methods used in clinimetrics. Much of this book, however, concerns the evaluation of instruments in relation to three key measurement domains: reliability, validity, and responsiveness. This text was designed to be useful in graduate-level courses on measurement or research methods and will also serve as an important reference and resource for researchers and clinicians. 30 PART 1 FOUNDATIONS OF NURSING RESEARCH 31 1 Introduction to Nursing Research in an Evidence-Based Practice Environment 32 NURSING RESEARCH IN PERSPECTIVE In all parts of the world, nursing has experienced a profound culture change. Nurses are increasingly expected to understand and conduct research and to base their professional practice on research evidence—that is, to adopt an evidencebased practice (EBP). EBP involves using the best evidence (as well as clinical judgment and patient preferences) in making patient care decisions, and “best evidence” typically comes from research conducted by nurses and other health care professionals. What Is Nursing Research? Research is systematic inquiry that uses disciplined methods to answer questions or solve problems. The ultimate goal of research is to develop and expand knowledge. Nurses are increasingly engaged in disciplined studies that benefit nursing and its clients. Nursing research is systematic inquiry designed to generate trustworthy evidence about issues of importance to the nursing profession, including nursing practice, education, administration, and informatics. In this book, we emphasize clinical nursing research, that is, research to guide nursing practice and to improve the health and quality of life of nurses’ clients. Nursing research has experienced remarkable growth in the past three decades, providing nurses with a growing evidence base from which to practice. Yet many questions endure and much remains to be done to incorporate research innovations into nursing practice. Examples of Nursing Research Questions: • How effective is pressurized irrigation, compared to a swabbing method, in cleansing wounds, in terms of time to wound healing, pain, patients’ satisfaction with comfort, and costs? (Mak et al., 2015) • What are the experiences of women in Zimbabwe who are living with advanced HIV infection? (Gona & DeMarco, 2015) The Importance of Research in Nursing Research findings from rigorous studies provide especially strong evidence for informing nurses’ decisions and actions. Nurses are accepting the need to base specific nursing actions on research evidence indicating that the actions are clinically appropriate, cost-effective, and result in positive outcomes for clients. In the United States, research plays an important role in nursing in terms of cred 33 entialing and status. The American Nurses Credentialing Center (ANCC)—an arm of the American Nurses Association and the largest and most prestigious credentialing organization in the United States—developed a Magnet Recognition Program to acknowledge health care organizations that provide high-quality nursing care. As Reigle and her colleagues (2008) noted, “the road to Magnet Recognition is paved with EBP” (p. 102) and the 2014 Magnet application manual incorporated revisions that strengthened evidence-based requirements (Drenkard, 2013). The good news is that there is growing confirmation that the focus on research and evidence-based practice may have important payoffs. For example, McHugh and co-researchers (2013) found that Magnet hospitals have lower riskadjusted mortality and failure to rescue than non-Magnet hospitals, even when differences among the hospitals in nursing credentials and patient characteristics are taken into account. Changes to nursing practice now occur regularly because of EBP efforts. Practice changes often are local initiatives that are not publicized, but broader clinical changes are also occurring based on accumulating research evidence about beneficial practice innovations. Example of Evidence-Based Practice: Numerous clinical practice changes reflect the impact of research. For example, “kangaroo care” (the holding of diaper-clad infants skin to skin by parents) is now practiced in many neonatal intensive care units (NICUs), but this is a relatively new trend. As recently as the 1990s, only a minority of NICUs offered kangaroo care options. Expanded adoption of this practice reflects mounting evidence that early skin-to-skin contact has benefits without negative side effects (e.g., Ludington-Hoe, 2011; Moore et al., 2012). Some of that evidence came from rigorous studies conducted by nurse researchers in several countries (e.g., Chwo et al., 2002; Cong et al., 2009; Cong et al., 2011; Hake-Brooks & Anderson, 2008). Nurses continue to study the potential benefits of kangaroo care in important clinical trials (e.g., Campbell-Yeo et al., 2013). The Consumer–Producer Continuum in Nursing Research In our current environment, all nurses are likely to engage in activities along a continuum of research participation. At one end of the continuum are consumers of nursing research, who read research reports or research summaries to keep up-to-date on findings that might affect their practice. EBP depends on well-informed nursing research consumers. At the other end of the continuum are the producers of nursing research: nurses who design and conduct research. At one time, most nurse researchers were 34 academics who taught in schools of nursing, but research is increasingly being conducted by nurses in health care settings who want to find solutions to recurring problems in patient care. Between these end points on the continuum lie a variety of research activities that are undertaken by nurses. Even if you never personally undertake a study, you may (1) contribute to an idea or a plan for a clinical study; (2) gather data for a study; (3) advise clients about participating in research; (4) solve a clinical problem by searching for research evidence; or (5) discuss the implications of a new study in a journal club in your practice setting, which involves meetings (in groups or online) to discuss research articles. In all possible research001-related activities, nurses who have some research skills are better able than those without them to make a contribution to nursing and to EBP. An understanding of nursing research can improve the depth and breadth of every nurse’s professional practice. Nursing Research in Historical Perspective Table 1.1 summarizes some of the key events in the historical evolution of nursing research. (An expanded summary of the history of nursing research appears in the Supplement to this chapter on ). 35 Most people would agree that research in nursing began with Florence Nightingale in the 1850s. Her most well-known research contribution involved an analysis of factors affecting soldier mortality and morbidity during the Crimean War. Based on skillful analyses, she was successful in effecting changes in nursing care and, more generally, in public health. After Nightingale’s work, research was absent from the nursing literature until the early 1900s, but most early studies concerned nurses’ education rather than clinical issues. In the 1950s, research by nurses began to accelerate. For example, a nursing research center was established at the Walter Reed Army Institute of Research. Also, the American Nurses Foundation, which is devoted to the promotion of nursing research, was founded. The surge in the number of studies conducted in the 1950s created the need for a new journal; Nursing Research came into being in 1952. As shown in Table 1.1, dissemination opportunities in professional journals grew steadily thereafter. In the 1960s, nursing leaders expressed concern about the shortage of research 36 on practice issues. Professional nursing organizations, such as the Western Interstate Council for Higher Education in Nursing, established research priorities, and practice-oriented research on various clinical topics began to emerge in the literature. During the 1970s, improvements in client care became a more visible research priority and nurses also began to pay attention to the clinical utilization of research findings. Guidance on assessing research for application in practice settings became available. Several journals that focus on nursing research were established in the 1970s, including Advances in Nursing Science, Research in Nursing & Health, and the Western Journal of Nursing Research. Nursing research also expanded internationally. For example, the Workgroup of European Nurse Researchers was established in 1978 to develop greater communication and opportunities for partnerships among 25 European National Nurses Associations. Nursing research continued to expand in the 1980s. In the United States, the National Center for Nursing Research (NCNR) at the National Institutes of Health (NIH) was established in 1986. Several forces outside of nursing also helped to shape the nursing research landscape. A group from the McMaster Medical School in Canada designed a clinical learning strategy that was called evidence-based medicine (EBM). EBM, which promulgated the view that research findings were far superior to the opinions of authorities as a basis for clinical decisions, constituted a profound shift for medical education and practice, and has had a major effect on all health care professions. Nursing research was strengthened and given more visibility when NCNR was promoted to full institute status within the NIH. In 1993, the National Institute of Nursing Research (NINR) was established, helping to put nursing research more into the mainstream of health research. Funding opportunities for nursing research expanded in other countries as well. Current and Future Directions for Nursing Research Nursing research continues to develop at a rapid pace and will undoubtedly flourish in the 21st century. Funding continues to grow. For example, NINR funding in fiscal year 2014 was more than $140 million compared to $70 million in 1999—and the competition for available funding is increasingly vigorous as more nurses seek support for testing innovative ideas for practice improvements. Broadly speaking, the priority for future nursing research will be the promotion of excellence in nursing science. Toward this end, nurse researchers and practicing nurses will be sharpening their research skills and using those skills to address 37 emerging issues of importance to the profession and its clientele. Among the trends we foresee for the early 21st century are the following: • Continued focus on EBP. Encouragement for nurses to engage in evidence-based patient care is sure to continue. In turn, improvements will be needed both in the quality of studies and in nurses’ skills in locating, understanding, critiquing, and using relevant study results. Relatedly, there is an emerging interest in translational research— research on how findings from studies can best be translated into practice. Translation potential will require researchers to think more strategically about long-term feasibility, scalability, and sustainability when they test solutions to problems. • Development of a stronger evidence base through confirmatory strategies. Practicing nurses are unlikely to adopt an innovation based on weakly designed or isolated studies. Strong research designs are essential, and confirmation is usually needed through the replication (i.e., the repeating) of studies with different clients, in different clinical settings, and at different times to ensure that the findings are robust. • Greater emphasis on systematic reviews. Systematic reviews are a cornerstone of EBP and will take on increased importance in all health disciplines. Systematic reviews rigorously integrate research information on a topic so that conclusions about the state of evidence can be reached. Best practice clinical guidelines typically rely on such systematic reviews. • Innovation. There is currently a major push for creative and innovative solutions to recurring practice problems. “Innovation” has become an important buzzword throughout NIH and in nursing associations. For example, the 2013 annual conference of the Council for the Advancement of Nursing Science was “Innovative Approaches to Symptom Science.” Innovative interventions—and new methods for studying nursing questions—are sure to be part of the future research landscape in nursing. • Expanded local research in health care settings. Small studies designed to solve local problems will likely increase. This trend will be reinforced as more hospitals apply for (and are recertified for) Magnet status in the United States and in other countries. Mechanisms will need to be developed to ensure that evidence from these small projects becomes available to others facing similar problems, such as communication within and between regional nursing research alliances. • Strengthening of interdisciplinary collaboration. Collaboration of nurses with researchers in related fields is likely to expand in the 21st century as researchers address fundamental health care problems. In turn, such collaborative efforts 38 could lead to nurse researchers playing a more prominent role in national and international health care policies. One of four major recommendations in a 2010 report on the future of nursing by the Institute of Medicine was that nurses should be full partners with physicians and other health care professionals in redesigning health care. • Expanded dissemination of research findings. The Internet and other electronic communication have a big impact on disseminating research information, which in turn helps to promote EBP. Through technologic advances, information about innovations can be communicated more widely and more quickly than ever before. • Increased focus on cultural issues and health disparities. The issue of health disparities has emerged as a central concern in nursing and other health disciplines; this in turn has raised consciousness about the cultural sensitivity of health interventions and the cultural competence of health care workers. There is growing awareness that research must be sensitive to the health beliefs, behaviors, and values of culturally and linguistically diverse populations. • Clinical significance and patient input. Research findings increasingly must meet the test of being clinically significant, and patients have taken center stage in efforts to define clinical significance. A major challenge in the years ahead will involve getting both research evidence and patient preferences into clinical decisions, and designing research to study the process and the outcomes. Broad research priorities for the future have been articulated by many nursing organizations, including NINR and Sigma Theta Tau International. Expert panels and research working groups help NINR to identify gaps in current knowledge that require research. The primary areas of research funded by NINR in 2014 were health promotion/disease prevention, eliminating health disparities, caregiving, symptom management, and self-management. Research priorities that have been expressed by Sigma Theta Tau International include advancing healthy communities through health promotion; preventing disease and recognizing social, economic, and political determinants; implementation of evidence-based practice; targeting the needs of vulnerable populations such as the poor and chronically ill; and developing nurses’ capacity for research. Priorities also have been developed for several nursing specialties and for nurses in several countries—for example, Ireland (Brenner et al., 2014; Drennan et al., 2007), Sweden (Bäck-Pettersson et al., 2008), Australia (Wynaden et al., 2014), and Korea (Kim et al., 2002). S O U R C E S O F E V I D E N C E FO R N U R S I N G PR A C T I C E Nurses make clinical decisions based on knowledge from many sources, including 39 coursework, textbooks, and their own clinical experience. Because evidence is constantly evolving, learning about best practice nursing perseveres throughout a nurse’s career. Some of what nurses learn is based on systematic research, but much of it is not. What are the sources of evidence for nursing practice? Where does knowledge for practice come from? Until fairly recently, knowledge primarily was handed down from one generation to the next based on experience, trial and error, tradition, and expert opinion. Information sources for clinical practice vary in dependability, giving rise to what is called an evidence hierarchy, which acknowledges that certain types of evidence are better than others. A brief discussion of some alternative sources of evidence shows how research001-based information is different. Tradition and Authority Decisions are sometimes based on custom or tradition. Certain “truths” are accepted as given, and such “knowledge” is so much a part of a common heritage that few seek verification. Tradition facilitates communication by providing a common foundation of accepted truth, but many traditions have never been evaluated for their validity. There is concern that some nursing interventions are based on tradition, custom, and “unit culture” rather than on sound evidence. Indeed, a recent analysis suggests that some “sacred cows” (ineffective traditional habits) persist even in a health care center recognized as a leader in evidence-based practice (Hanrahan et al., 2015). Another common source of information is an authority, a person with specialized expertise. We often make decisions about problems with which we have little experience; it seems natural to place our trust in the judgment of people with specialized training or experience. As a source of evidence, however, authority has shortcomings. Authorities are not infallible, particularly if their expertise is based primarily on personal experience; yet, like tradition, their knowledge often goes unchallenged. Example of “Myths” in Nursing Textbooks: A study suggests that even nursing textbooks may contain “myths.” In their analysis of 23 widely used undergraduate psychiatric nursing textbooks, Holman and colleagues (2010) found that all books contained at least one unsupported assumption (myth) about loss and grief—that is, assumptions not supported by research evidence. Moreover, many evidence-based findings about grief and loss failed to be included in the textbooks. Clinical Experience, Trial and Error, and Intuition Clinical experience is a familiar, functional source of knowledge. The ability to 40 generalize, to recognize regularities, and to make predictions is an important characteristic of the human mind. Nevertheless, personal experience is limited as a knowledge source because each nurse’s experience is too narrow to be generally useful. A second limitation is that the same objective event is often experienced and perceived differently by two nurses. A related method is trial and error in which alternatives are tried successively until a solution to a problem is found. We likely have all used this method in our professional work. For example, many patients dislike the taste of potassium chloride solution. Nurses try to disguise the taste of the medication in various ways until one method meets with the approval of the patient. Trial and error may offer a practical means of securing knowledge, but the method tends to be haphazard and solutions may be idiosyncratic. Intuition is a knowledge source that cannot be explained based on reasoning or prior instruction. Although intuition and hunches undoubtedly play a role in nursing—as they do in the conduct of research—it is difficult to develop nursing policies and practices based on intuition. Logical Reasoning Solutions to some problems are developed by logical thought processes. As a problem-solving method, logical reasoning combines experience, intellectual faculties, and formal systems of thought. Inductive reasoning involves developing generalizations from specific observations. For example, a nurse may observe the anxious behavior of (specific) hospitalized children and conclude that (in general) children’s separation from their parents is stressful. Deductive reasoning involves developing specific predictions from general principles. For example, if we assume that separation anxiety occurs in hospitalized children (in general), then we might predict that (specific) children in a hospital whose parents do not room-in will manifest symptoms of stress. Both systems of reasoning are useful for understanding and organizing phenomena, and both play a role in research. Logical reasoning in and of itself, however, is limited because the validity of reasoning depends on the accuracy of the premises with which one starts. Assembled Information In making clinical decisions, health care professionals rely on information that has been assembled for a variety of purposes. For example, local, national, and international benchmarking data provide information on such issues as infection rates or the rates of using various procedures (e.g., cesarean births) and can 41 facilitate evaluations of clinical practices. Cost data—information on the costs associated with certain procedures, policies, or practices—are sometimes used as a factor in clinical decision making. Quality improvement and risk data, such as medication error reports, can be used to assess the need for practice changes. Such sources are useful, but they do not provide a good mechanism for determining whether improvements in patient outcomes result from their use. Disciplined Research Research conducted in a disciplined framework is the most sophisticated method of acquiring knowledge. Nursing research combines logical reasoning with other features to create evidence that, although fallible, tends to yield the most reliable evidence. Carefully synthesized findings from rigorous research are at the pinnacle of most evidence hierarchies. The current emphasis on EBP requires nurses to base their clinical practice to the greatest extent possible on rigorous research001-based findings rather than on tradition, authority, intuition, or personal experience— although nursing will always remain a rich blend of art and science. PA R A D I G M S A N D M E T H O D S FO R N U R S I N G RESEARCH A paradigm is a worldview, a general perspective on the complexities of the world. Paradigms for human inquiry are often characterized in terms of the ways in which they respond to basic philosophical questions, such as, What is the nature of reality? (ontologic) and What is the relationship between the inquirer and those being studied? (epistemologic). Disciplined inquiry in nursing has been conducted mainly within two broad paradigms, positivism and constructivism. This section describes these two paradigms and outlines the research methods associated with them. In later chapters, we describe the transformative paradigm that involves critical theory research (Chapter 21), and a pragmatism paradigm that involves mixed methods research (Chapter 26). The Positivist Paradigm The paradigm that dominated nursing research for decades is known as positivism (also called logical positivism). Positivism is rooted in 19th century thought, guided by such philosophers as Mill, Newton, and Locke. Positivism reflects a broader cultural phenomenon that, in the humanities, is referred to as modernism, which emphasizes the rational and the scientific. As shown in Table 1.2, a fundamental assumption of positivists is that there is a reality out there that can be studied and known (an assumption is a basic principle 42 that is believed to be true without proof or verification). Adherents of positivism assume that nature is basically ordered and regular and that reality exists independent of human observation. In other words, the world is assumed not to be merely a creation of the human mind. The related assumption of determinism refers to the positivists’ belief that phenomena are not haphazard but rather have antecedent causes. If a person has a cerebrovascular accident, the researcher in a positivist tradition assumes that there must be one or more reasons that can be potentially identified. Within the positivist paradigm, much research activity is directed at understanding the underlying causes of phenomena. Positivists value objectivity and attempt to hold personal beliefs and biases in check to avoid contaminating the phenomena under study. The positivists’ scientific approach involves using orderly, disciplined procedures with tight controls of the research situation to test hunches about the phenomena being studied. Strict positivist thinking has been challenged, and few researchers adhere to the 43 tenets of pure positivism. In the postpositivist paradigm, there is still a belief in reality and a desire to understand it, but postpositivists recognize the impossibility of total objectivity. They do, however, see objectivity as a goal and strive to be as neutral as possible. Postpositivists also appreciate the impediments to knowing reality with certainty and therefore seek probabilistic evidence—that is, learning what the true state of a phenomenon probably is, with a high degree of likelihood. This modified positivist position remains a dominant force in nursing research. For the sake of simplicity, we refer to it as positivism. The Constructivist Paradigm The constructivist paradigm (often called the naturalistic paradigm) began as a countermovement to positivism with writers such as Weber and Kant. Just as positivism reflects the cultural phenomenon of modernism that burgeoned after the industrial revolution, naturalism is an outgrowth of the cultural transformation called postmodernism. Postmodern thinking emphasizes the value of deconstruction—taking apart old ideas and structures—and reconstruction—putting ideas and structures together in new ways. The constructivist paradigm represents a major alternative system for conducting disciplined research in nursing. Table 1.2 compares the major assumptions of the positivist and constructivist paradigms. For the naturalistic inquirer, reality is not a fixed entity but rather is a construction of the individuals participating in the research; reality exists within a context, and many constructions are possible. Naturalists thus take the position of relativism: If there are multiple interpretations of reality that exist in people’s minds, then there is no process by which the ultimate truth or falsity of the constructions can be determined. The constructivist paradigm assumes that knowledge is maximized when the distance between the inquirer and those under study is minimized. The voices and interpretations of study participants are crucial to understanding the phenomenon of interest, and subjective interactions are the primary way to access them. Findings from a constructivist inquiry are the product of the interaction between the inquirer and the participants. Paradigms and Methods: Quantitative and Qualitative Research Research methods are the techniques researchers use to structure a study and to gather and analyze information relevant to the research question. The two alternative paradigms correspond to different 44 methods for developing evidence. A key methodologic distinction is between quantitative research, which is most closely allied with positivism, and qualitative research, which is associated with constructivist inquiry—although positivists sometimes undertake qualitative studies, and constructivist researchers sometimes collect quantitative information. This section provides an overview of the methods associated with the two paradigms. The Scientific Method and Quantitative Research The traditional, positivist scientific method refers to a set of orderly, disciplined procedures used to acquire information. Quantitative researchers use deductive reasoning to generate predictions that are tested in the real world. They typically move in a systematic fashion from the definition of a problem and the selection of concepts on which to focus to the solution of the problem. By systematic, we mean that the investigator progresses logically through a series of steps, according to a specified plan of action. Quantitative researchers use various control strategies. Control involves imposing conditions on the research situation so that biases are minimized and precision and validity are maximized. Control mechanisms are discussed at length in this book. Quantitative researchers gather empirical evidence—evidence that is rooted in objective reality and gathered through the senses. Empirical evidence, then, consists of observations gathered through sight, hearing, taste, touch, or smell. Observations of the presence or absence of skin inflammation, patients’ anxiety level, or infant birth weight are all examples of empirical observations. The requirement to use empirical evidence means that findings are grounded in reality rather than in researchers’ personal beliefs. Evidence for a study in the positivist paradigm is gathered according to an established plan, using structured methods to collect needed information. Usually (but not always) the information gathered is quantitative—that is, numeric information that is obtained from a formal measurement and is analyzed statistically. A traditional scientific study strives to go beyond the specifics of a research situation. For example, quantitative researchers are typically not as interested in understanding why a particular person has a stroke as in understanding what factors influence its occurrence in people generally. The degree to which research 45 findings can be generalized to individuals other than those who participated in the study is called the study’s generalizability. The scientific method has enjoyed considerable stature as a method of inquiry and has been used productively by nurse researchers studying a range of nursing problems. This is not to say, however, that this approach can solve all nursing problems. One important limitation—common to both quantitative and qualitative research—is that research cannot be used to answer moral or ethical questions. Many persistent, intriguing questions about human beings fall into this area— questions such as whether euthanasia should be practiced or abortion should be legal. The traditional resear…
Purchase answer to see full attachment