Content provided in partnership with
Health Care Industry
Industry: Email Alert RSS FeedUsing public reports of patient satisfaction for hospital quality improvement
Health Services Research, June, 2006 by Judith K. Barr, Tierney E. Giannotti, Shoshanna Sofaer, Cathy E. Duquette, William J. Waters, Marcia K. Petrillo
Sample
- More Articles of Interest
- Patient satisfaction survey.(Brief Article)
- Study shows link between career satisfaction, patient satisfaction
- Measuring patient satisfaction.(Guest Editorial)
- Role of Clinical and Process Quality in Achieving Patient Satisfaction in...
- How should we measure patient satisfaction?
The study used a purposive sample for interviewing four key executives in each hospital (Chief Executive Officer, Medical Director, Nurse Executive, and Patient Satisfaction Coordinator). The intention was to capture multiple perspectives from hospital staff, including top administration, clinical areas (medicine and nursing), and the person most familiar with the implementation of the patient satisfaction survey. A letter was sent by e-mail from the Hospital Association of RI to the individual in these positions at each hospital, informing them of the evaluation effort and encouraging their participation in the interview. Of the 52 positions identified, a total of 42 people were interviewed at the 11 general and two specialty hospitals. This sample included: 13 CEOs; 16 clinical staff (eight Medical Directors and eight Nurse Executives); and 13 Patient Satisfaction Coordinators. The latter included administrative positions ranging from manager to director and vice president, and nearly all had "quality" or "performance improvement" in the rifles. The overall response rate was 81 percent with at least three executives interviewed at each hospital, except for one hospital where two executives were interviewed (mean = 3.2 respondents per hospital). Most interviews (70 percent) were completed by telephone, and 30 percent were conducted in person at the respondent's office. No differences in amount of time spent or responsiveness of those interviewed were noted for these two methods.
Data Analysis
Interviews were audio recorded (except in a few instances because of equipment malfunction) and transcribed to electronic format to provide a written transcript that then could be electronically analyzed. A preliminary set of a priori codes was developed by the authors and, after review of 15 randomly selected transcripts, the team developed additional codes that emerged from the interviews. The revised codes were applied to the transcripts (Miles and Huberman 1994) using QSR International's NVivo, a software package used for analyzing qualitative data, such as interviews. For this study, NVivo was used not only to apply the a priori and emergent codes to the transcripts but also to aggregate the detailed codes into general themes. It was also used to report on the weight of the evidence, i.e., how often a given code was mentioned by respondents. Coding of a 20 percent random sample by a second researcher yielded a high degree of reliability (agreement rate = 93 percent), and discrepancies were resolved by consensus resulting in a refinement to the existing list of codes. Using an iterative approach to move from specific code categories to more general themes, a process of "functional reduction" combined infrequent code categories and added new ones (Becker 1998), resulting in clusters of common responses and consistencies in the data (Miles and Huberman 1994).
The hospital was the unit of analysis for data related to QI activities, QI structure, and barriers to QI. Because we sought to capture multiple perspectives, responses were ascribed to a hospital regardless of how many respondents cited a view or activity within each hospital. For example, if one person described a QI activity, we did not require agreement to count the activity; also, we counted the activity only once even if mentioned by more than one person in the hospital. However, when the analysis focused on attitudes about the public reporting process, the individual respondent was the unit of analysis, and responses were analyzed across hospitals (Hibbard, Stockard, and Tusler 2003). Recognizing that responses were voluntarily given and not always asked directly or in discrete response categories, we sought to identify responses related to the same category or theme. The analysis assessed the extent of agreement among respondents to the interview questions, as well as the range of opinions that describe the multiple perspectives and richness of interpretation of the public reporting process in RI (Sofaer 1999, 2002).
Find Research Guides for:
Find Featured Titles for: Autos
Most Popular Articles in Health
- Fuel your workout: exercisers who eat before they work out have more energy ...
- Soothe a dry, itchy scalp: 5 easy expert solutions
- Cocktails and calories: Beer, wine and liquor calories can really add up. ...
- The sour truth about apple cider vinegar - evaluation of therapeutic use
- The, six best supplements you've never heard of: these secret weapons can ...
- More »