• Center on Health Equity and Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Willingness to Recommend a Health Plan: Who Is Dissatisfied and What Don't They Like?

Publication
Article
The American Journal of Managed CareJune 2004
Volume 10
Issue 6

Objectives: To explore the characteristics of individuals who are dissatisfied with their health plan, assess what aspects of the medical care experience are associated with plan dissatisfaction, and examine how this association varies according to plan type.

Study Design: Retrospective observational study using the 1996 Medical Expenditure Panel Survey.

Methods: Unwillingness to recommend a health plan to others was used as a measure of overall plan dissatisfaction. Descriptive statistics were calculated to characterize the dissatisfied population. Logistic regressions and predicted probabilities controlling for personal characteristics were calculated to determine the association between plan or provider attribute and unwillingness to recommend a plan.

Results: We found no personal characteristics that significantly differentiated individuals who reported their family was unwilling to recommend a health plan from those who were willing. The largest predictors of unwillingness to recommend a plan were dissatisfaction with choice of providers and preventive services coverage; in contrast, provider and personal characteristics were not significant predictors of dissatisfaction. We estimated the probability of being unwilling to recommend a health plan was 38% for individuals dissatisfied with the choice of providers and 34% for those dissatisfied with preventive services coverage. Although provider attributes were not found to be predictors of plan dissatisfaction for the entire sample, they were predictors of dissatisfaction for HMO and multiple plan members.

Conclusions: Enrollee dissatisfaction with the choice of providers and preventive services coverage are major predictors of health plan dissatisfaction. Managers concerned about plan satisfaction may want to examine enrollee assessments of these measures.

(Am J Manag Care. 2004;10:393-400)

Contrary to popular perception, empirical studies show that people are fairly satisfied with their health plans. For example, survey results from more than 135 000 respondents in 270 commercial health plans reported by the Consumer Assessment of Health Plans Survey (CAHPS) Benchmarking Database in 2000 indicated that more than three fourths of the respondents gave their health plan a rating of 7 or higher on a 10-point scale. More than a third of the respondents rated their plan as a 9 or 10.1 Despite this broad perception of satisfaction, published reports have indicated that a nontrivial portion of privately insured respondents are dissatisfied with their health plan.2 This group of dissatisfied enrollees may be cause for concern because policy makers, researchers, and managers commonly use measures of patient satisfaction, or dissatisfaction, as an indicator of the quality of care.3,4 Patient satisfaction has been shown to be associated with enrollment and disenrollment activities of plan members5,6 and to affect the financial outcomes of health insurers, healthcare providers and hospitals.7,8

The goal of this study was to empirically evaluate 3 questions:

  • Do individuals reporting that their family is dissatisfied with their health plan share any common characteristics?

What aspects of the process of health services delivery are associated with overall health plan dissatisfaction?

  • Do these aspects vary according to plan type?

Although a goal of 100% satisfaction may be unrealistic, understanding where plans are falling short will help managers and policy makers to identify areas to focus on when exploring corrective actions to improve enrollee satisfaction.

There are a number of gaps in the existing literature. Despite a broad patient satisfaction literature base, a large portion of the existing work exploring the determinants of patient satisfaction concentrates solely on the delivery of healthcare services, ignoring the role of insurers in the system. The literature that examines the determinants of health plan satisfaction tends either to look at a single plan or to examine respondents in a limited geographic locale,6,9,10 concentrates on a specific population such as children or Medicare recipients,11 or explores variations in satisfaction according to a single dimension (eg, looking for differences according to plan type, patient sex, or use of a specific service).12-15 In addition, much of the existing literature considers the interactions with insurers and providers independently when attempting to determine what consumers value when assessing their satisfaction with their healthcare providers and their health plans. That framework may have been appropriate under the old solo-practitioner fee-for-service (FFS) healthcare system. In today's complex marketplace, where the roles and responsibilities of providers and insurers tend to merge, it is unclear how patients integrate their perceptions of their experiences with their health plans and their providers with their overall assessment of satisfaction.

This study will fill these gaps in the literature by using a nationally representative sample across a broad spectrum of health plans and plan types. This sample allows us to characterize the population that is dissatisfied with their health plan and examine factors that are associated with consumer perceptions of plan dissatisfaction. This information on which aspects of the healthcare experience have the strongest association with overall plan dissatisfaction will allow health plan managers and healthcare providers to concentrate on the aspects that can have the greatest impact, presumably improving the overall quality of care for their members and their own financial outcomes.

This paper adds to the body of literature exploring the influence of managed care and insurance characteristics on consumer behavior and the US healthcare system (see, eg, references 11, 12, and 16-22). Because the data for this study are derived from a nationally representative sample, they provide customer assessments of a wide variety of health plans, enabling us to infer generalizations about the predictive effects of a broad spectrum of variables on private insurance plan dissatisfaction. In addition, studies have suggested that individuals with higher satisfaction are more likely to respond to patient satisfaction surveys, resulting in overestimation of satisfaction.23 The broad focus and ongoing, in-person interview format of the survey that is the source of our data may reduce this potential nonresponse bias.

CONCEPTUAL FRAMEWORK

A significant body of literature has been dedicated to developing theoretical models of the factors influencing patient satisfaction with healthcare services and healthcare plans, tools to measure these determinants, and assessments of these measures.24 The conceptual model framing this study is based on the "attribute basis of satisfaction" (Figure 1).25-27 This framework recognizes that delivery of healthcare services and provision of health insurance are complex services. Complex services are comprised of a range of individual experiences, which can be grouped into attributes of the overall good or service. For example, attributes of health insurance may include the design of the plan, what it covers, restrictions on where services may be obtained, cost, administrative requirements for getting reimbursed, and so forth. The literature has shown that individuals form perceptions of their satisfaction with specific attributes of the service based on the performance of that attribute and personal value judgments of that performance. These attribute perceptions then are aggregated into overall perceptions of satisfaction with the complex service, in this case assessments of health plan satisfaction or dissatisfaction.24

METHODS

Data

We used data from the 1996 Medical Expenditure Panel Survey (MEPS), the most recent MEPS dataset addressing plan satisfaction that was available in March 2003, to examine the characteristics of people dissatisfied with their health plans and to explore the association between various plan and provider experiences and overall plan dissatisfaction. MEPS is a nationally representative survey conducted by the Agency for Health Care Policy and Research (now the Agency for Healthcare Research and Quality). This broad, ongoing survey collects detailed information on demographic characteristics, health conditions, health status, use of medical care services, charges and payments, access to care, health insurance coverage, income, and employment from respondents.28

The population of interest for this study was privately insured US adults between the ages of 18 and 64 years. Respondents age 65 years and older were excluded from our study to avoid the potential complicating factor of dual coverage with Medicare. To clarify the role of plan type, we also eliminated from the sample respondents classifying their health insurance plan as something other than FFS, HMO, or preferred provider organization (PPO). The 1996 MEPS contains full-year data on 22 601 eligible individuals; 4104 individuals between the ages of 18 and 64 years who were covered by private insurance reported on their family's satisfaction with their health plan(s). Because of missing data for some of our explanatory measures, 1224 observations were available for logistic regression calculations. One third of our sample respondents were covered by more than 1 health plan during the reporting period. In those cases, we used an average satisfaction rating across all of the individual's health plans for each of the plan-related attributes and overall plan satisfaction measures. Sensitivity analyses comparing estimation results using individual plan and attribute satisfaction ratings with these average measures found no significant differences in the predictions generated.

Our primary dependent variable to measure overall plan satisfaction was responses to the MEPS question: "How likely is anyone in the family to recommend your health plan to family and friends?" We chose this measure because it has been reported that stated behavioral intentions are indicators of dissatisfaction, regardless of whether those intentions are acted on or not.24 In addition, it has been theorized that recommendation levels are more sensitive indicators of overall plan satisfaction because they are less influenced by financial considerations of plan choice.29

MEPS codes the responses for plan recommendation on a 4-point Likert scale ranging from 1 = very likely to 4 = not at all likely. Attribute satisfaction measures are similarly coded, ranging from 1 = very satisfied to 4 = not at all satisfied. For ease of interpretation, we collapsed the 4-point Likert scale measures into dichotomous satisfied/dissatisfied or willing/unwilling measures. For the purposes of this study, we categorized respondents as being unwilling to recommend their plan if they responded either 3 or 4 (not too likely or not at all likely), and we considered persons as being dissatisfied with an attribute if they answered either not too satisfied or not at all satisfied. These less-stringent cut points were set to minimize estimation difficulties arising from the small proportion of respondents reporting extreme dissatisfaction and to account for any potential "upcoding" or positive skewness effects that may bias the results.30

Statistical Analysis

Who Is Dissatisfied?

We used chi-square tests to assess which personal characteristics were associated with plan dissatisfaction. The personal and insurance measures that were examined were chosen in accordance with the existing determinants-of-satisfaction literature.6,31,32 Personal characteristics examined were age, sex, marital status, education, income, race, and self-reported assessments of physical and mental health status. In addition, 2 measures of insurance characteristics were examined: a self-reported classification of the type of plan enrolled in and whether a gatekeeper was required.

What Don't They Like?

Bivariate analysis and chi-square tests were performed to assess which specific aspects of the healthcare experience were associated with enrollees' unwillingness to recommend their health plan. Health plan-related variables examined satisfaction with these measures: the amount paid for the plan, the choice of providers allowed by the plan, hospital coverage, preventive services coverage, prescription drug coverage, the amount and difficulty of paperwork, and difficulty in getting a referral to a specialist. Primary healthcare provider-related attributes were a physician availability performance index, respondents' perceptions of whether their provider listens to their concerns, and confidence in their usual healthcare provider's abilities. The physician availability performance index was created to minimize the effects of multicollinearity and is the average of categorical measures of respondents' usual wait to see their usual healthcare provider, difficulty in contacting the usual provider's office on the telephone, and difficulty in getting an appointment with their usual provider.

P

Multivariate logistic regressions were used to examine the predictive effect of the specific aspects of the healthcare experience on unwillingness to recommend a health plan, controlling for personal characteristics. Only the variables that showed at least a minor level of statistical significance of association in the bivariate analyses ( ¡Ü .40) were included in these logistic regression models. Sampling weights were used and standard error corrections were performed to reflect the noninstitutionalized civilian US population and to account for the complex survey design in all bivariate and regression analyses. A Hosmer-Lemeshow goodness-of-fit test was performed to ensure our model fit the data adequately. Predicted probabilities were calculated using the plan and provider attribute coefficients from the logistic regressions and controlling for personal and insurance characteristics.

Do the Predictors of Dissatisfaction Vary According to Plan Type?

Logistic regression models were run for subgroups based on the type of plan (FFS or PPO plans, HMOs, or enrollment in multiple plans) because previous studies have hypothesized that consumers in different types of health plans may react to different things.6

FINDINGS

Our sample was primarily white, middle or upper income, moderately educated individuals in very good or excellent health. Sixty percent were enrolled in plans that had some gatekeeper requirements, and respondents were fairly evenly divided between the plan type categories: FFS or PPO plans, HMOs, or multiple plans. Of the 4104 respondents in our sample reporting the likelihood of recommending their health plan to friends and family, 42% indicated they were very likely to recommend, 34% somewhat likely, 13% not too likely, and 11% not at all likely.

Who Is Dissatisfied?

We found no personal characteristics that significantly differentiated individuals who reported their family was unwilling to recommend a health plan from willing individuals (Table 1). Twenty-four percent of the study population reported they would not be willing to recommend their health plan. The only personal or insurance characteristics that were associated with plan dissatisfaction were age, physical and mental health status, plan type, and gatekeeper requirements. Even though these variables had a statistically significant association, the difference in likelihood of recommending a plan among the subgroups was small. There was no difference in the proportion of individuals in our sample who reported they were unwilling to recommend their plan according to sex, marital status, race, or income.

What Don't They Like?

In general, the respondents were satisfied with the various plan-related attributes, with dissatisfaction rates ranging between 5% and 15% (Figure 2). The vast majority of the respondents also indicated they were satisfied with the quality of care from their usual healthcare provider. Access to healthcare providers may be more of a concern, however. One quarter of the respondents indicated they had difficulty getting an appointment or contacting their usual healthcare provider's office by telephone, and 14% reported difficulties in getting a referral to a specialist.

All of the plan- and provider-related attributes examined had statistically significant associations with plan dissatisfaction (Table 2). For all of these measures, a greater proportion of individuals reporting dissatisfaction with a specific attribute indicated their family was unlikely to recommend the plan. The size of this effect, however, varied according to the attribute measured. A much larger proportion of individuals reporting dissatisfaction with plan-related attributes would not recommend their plan than those reporting poorer provider-related attribute performance.

What Are the Predictors of Unwillingness to Recommend a Health Plan?

Dissatisfaction with any of the plan-related attributes other than hospital and prescription drug coverage increased the odds of being unwilling to recommend a health plan; in contrast, none of the provider-related or personal characteristics were significant predictors of dissatisfaction (Table 3). The largest predictor of unwillingness to recommend a plan was dissatisfaction with the choice of providers, closely followed by dissatisfaction with preventive services coverage. The odds of individuals dissatisfied with these measures to be unlikely to recommend their plan were 4 times the odds of those who were satisfied. The odds of individuals expressing difficulties in getting specialist referrals, dissatisfied with the amount paid, or dissatisfied with the amount of paperwork to be unwilling to recommend their plan were between 2 and 3 times the odds of individuals satisfied with those attributes. Personal and insurance characteristics were not significant predictors of unwillingness to recommend a plan after controlling for plan attribute satisfaction and provider attribute performance.

Predictions of Unwillingness to Recommend a Plan

Figure 3 summarizes the predicted probability of being unwilling to recommend a health plan based on dissatisfaction with a specific attribute, controlling for personal characteristics. This figure shows that dissatisfaction with the choice of providers has the largest impact on the probability of unwillingness to recommend the plan, followed closely by dissatisfaction with preventive services coverage. These predictions assumed the individuals were age 18-39 years, had some college education, and were in a gatekeeper plan; and the predictions were based on the mean marital status and health status of the sample. We estimated that if a person were dissatisfied with the choice of providers or preventive services coverage, the probability that he or she would be unwilling to recommend the plan would be 38% or 34%, respectively. The probability of FFS or PPO enrollees being unwilling to recommend their plan when dissatisfied with any of the plan or provider attributes was higher than that of HMO members. Personal characteristics had little effect on these predicted probability calculations.

Do the Predictors of Dissatisfaction Vary According to Plan Type?

The attributes that were significant predictors of plan dissatisfaction and the size of their impact varied according to the type of plan. The odds that FFS and PPO enrollees would be unwilling to recommend their health plan increased if they were dissatisfied with the amount paid for the policy, the choice of providers available, drug coverage, or the amount of paperwork. Dissatisfaction with the choice of providers, hospital and preventive services coverage, and ease of referral to specialists increased the odds that HMO enrollees would be unwilling to recommend their plan, but the effect of hospital coverage was 3 to 4 times the size of the effect of the other plan-related attributes. Dissatisfaction with the choice of providers, preventive services coverage, and amount of paperwork were the significant predictors of increased odds of unwillingness to recommend the plan for individuals enrolled in multiple plans. Lack of confidence in the provider's abilities increased the odds of unwillingness to recommend the plan for HMO members, but reduced the odds for enrollees in multiple plans.

CONCLUSIONS AND DISCUSSION

This study used a nationally representative sample of individuals across a broad spectrum of health plans to (1) empirically explore the characteristics of privately insured individuals who report they are dissatisfied with their health plan, (2) assess what aspects of the healthcare delivery process individuals are dissatisfied with, and (3) determine what aspects of the healthcare experience are most associated with overall health plan dissatisfaction. We found several plan attributes that are predictors of plan dissatisfaction. In contrast, provider-related attributes and personal characteristics were relatively unimportant in predicting unwillingness to recommend a plan.

We found that the largest contributor to the probability of a respondent expressing unwillingness to recommend his or her plan was dissatisfaction with plan-related attributes such as the choice of providers, the extent of preventive services coverage, the amount paid for the policy, and the amount of paperwork involved. Plan type was not found to be a predictor of enrollee dissatisfaction, but the predictors of plan dissatisfaction varied depending on whether the individual was a member of a HMO or enrolled in a FFS or PPO plan.

Our findings on the size and composition of the dissatisfied population were fairly consistent with what has been previously reported (eg, by the Agency for Healthcare Research and Quality,1 Weisman and Henderson,13 Dellana and Glascoff,15 and Hall and Dornan31). However, our findings on the predictive effect of certain attributes on an enrollee's perceptions of his or her health plan deviate from what has been previously reported. Previous studies have indicated that access to specialists was an important predictor of plan dissatisfaction (or disenrollment).5,33 Those studies, however, tended to be conducted with a date from a single health plan, usually an HMO. Although consistent with our findings regarding HMO members' willingness to recommend a plan, this effect disappeared when we examined enrollees in FFS, PPO, or multiple plans. This discrepancy highlights the importance of understanding the study setting and the generalizability of the data when examining patient satisfaction results.

Variable measurement issues limit the conclusions that can be drawn from this study. Many of the attribute measures available through MEPS were determined on a "big picture" level of abstraction and were quite broad. Our findings can be useful to plan administrators, policy makers, and healthcare practitioners by pointing them to key areas associated with plan dissatisfaction. However, the lack of specificity of the MEPS measures offers limited guidance for the development of specific actions to rectify the situation. For example, although dissatisfaction with choice of providers was shown to be a significant issue, we have no insight regarding specifically what is unsatisfactory with the available choice. Should the provider roster include a wider variety of provider specialties, a broader geographic distribution, or a larger selection within a specific specialty? Evaluation of the attributes shown to be predictors of dissatisfaction using more detailed questions would allow us to narrow down the set of corrective actions that might be warranted.

One way to facilitate the attainment of detailed satisfaction data could be through broad application of the measures available through CAHPS. CAHPS is a standardized instrument for health plans to measure members' perceptions of their healthcare provider, interactions with specialists, and their health plans. The CAHPS questionnaires address a broad array of specific healthcare experiences such as getting an appointment and communication with the respondent's healthcare providers and their staff.1 Despite the benefits of having a single, widely distributed, standardized instrument for collecting comprehensive enrollee satisfaction data, the decision to administer the survey and submit survey results to the National CAHPS Benchmarking Database is up to the individual health plan. Any measurement error or bias resulting from an insurer's decision to administer the instrument or not may affect the generalizability of the findings. This is a concern because, as this study found, the conclusions about the predictors of plan dissatisfaction can vary according to the study setting and generalizability of the data. The voluntary nature of participation in the National CAHPS Benchmarking Database limits the representativeness of the reported results across plans. Of the 240 commercial health plans reporting satisfaction results to the National CAHPS Benchmarking Database in 2000, only 2 were FFS or indemnity plans. The remaining 238 were HMO, PPO, and point-of-service plans.1 A wealth of information could be catalogued if policies were enacted so that the CAHPS instrument was universally and uniformly administered and the results warehoused at a central repository (eg, the National CAHPS Benchmarking Database). The detailed data that would be available would enable the linkage between plan attributes, provider behavior, and overall health plan dissatisfaction to be understood more precisely.

This paper identifies enrollee dissatisfaction with the choice of providers and preventive services coverage as the major predictors of being dissatisfied with a health plan, as measured by an unwillingness to recommend it to people they know. This information points managers and policy makers to areas to concentrate on when undertaking efforts to improve satisfaction. The availability of a generalizable database containing more detailed patient satisfaction information would allow suggested actions to be refined, possibly indicating what types of providers to include in an expanded provider roster, or what additional services should be included in an enhanced coverage policy.

From the Department of Psychiatry (JAS), the Department of Clinical Pharmacy (KAP, SYL), and the Institute of Health Policy Studies (KAP), University of California San Francisco, Calif; and the Division of General Internal Medicine, Brigham and Women's Hospital, Boston, Mass (JSH).

Dr Sakowski was supported by an institutional research training fellowship sponsored by the National Institute of Mental Health while working on this study. This work also was partially supported by funding from research grants from the National Cancer Institute (R01 CA81130) and the Agency for Healthcare Research and Quality (P01 HS10771 and P01 HS10856).

Address correspondence to: Julie Sakowski, PhD, Sutter Health Institute for Research and Education, 345 California St., Suite 2000, San Francisco, CA 94104. E-mail: sakowsj@sutterhealth.org.

Annual Report of the National CAHPS Benchmarking Database.

1. Agency for Healthcare Research and Quality. Rockville, Md: Agency for Healthcare Research and Quality; 2000.

Health Care Strategic Manage.

2. Harris poll shows most people like their health plans, if not managed care in general. 2002;20(2):8.

Inquiry.

3. Cleary P, McNeil B. Patient satisfaction as an indicator of quality care. 1988;25:25-36.

Health Care Financ Rev.

4. Gold M, Wooldridge J. Surveying consumer satisfaction to assess managed-care quality: current practices. 1995;16(4):155-173.

Health Care Manage Rev.

5. Gamble JE, Icenogle ML, Bryan NB, Rickert DA. The effects of open access on member satisfaction and intentions to remain in an HMO. 2000;25(4):24-47.

J Bus Res.

6. Mummalaneni V, Gopalakrishna P. Access, resource, and cost impacts on consumer satisfaction with health care: a comparison across alternative health care modes and time. 1997;39(3):173-186.

J Health Care Marketing.

7. Mascarenhas OAJ. Marketing health care to employees: the structure of employee health care plan satisfaction. 1993;13(3):34-46.

Health Care Marketing.

8. Nelson EC. Do patient perceptions of quality relate to hospital financial performance? 1992;12(4):6-13.

Health Serv Res.

9. Roohan PJ, Franko SJ, Anarella JP, Dellehunt LK, Gesten FC. Do commercial managed care members rate their health plans differently than Medicaid managed care members? 2003;38(4):1121-1134.

J Health Care Marketing.

10. Rutledge R, Nascimento P. Satisfaction with HMOs. 1996;16(1):22-27.

Health Serv Res.

11. Newacheck PW, Hung YY, Marchi KS, Hughes DC, Pitter C, Stoddard JJ. The impact of managed care on children's access, satisfaction, use, and quality of care. 2001;36(2):315-334.

Health Serv Res

12. Escarce JJ, Kapur K, Solomon MD, et al. Practice characteristics and HMO enrollee satisfaction with specialty care: an analysis of patients with glaucoma and diabetic retinopathy. . 2003;38(4):1135-1155.

Womens Health Issues.

13. Weisman CS, Henderson JT. Managed care and women's health: access, preventive services, and satisfaction. 2001;11(3):201-215.

Womens Health Issues.

14. Weisman CS, Henderson JT, Schifrin E, Romans M, Clancy CM. Gender and patient satisfaction in managed care plans: analysis of the 1999 HEDIS/CAHPS 2.0H Adult Survey. 2001;11(5):401-415.

Health Care Manage Rev.

15. Dellana SA, Glascoff DW. The impact of health insurance plan type on satisfaction with health care. 2001;26(2):33-46.

Health Serv Res.

16. Reschovsky JD, Kemper P, Tu H. Does type of health insurance affect health care use and assessments of care among the privately insured? 2000;35(1 pt 2):219-237.

Issue Brief Cent Stud Health Syst Change.

17. Reschovsky JD, Kemper P, Tu HT, Lake T, Wong HJ. Do HMOs make a difference? March 2000;(28):1-8.

Issue Brief Cent Stud Health Syst Change

18. Reschovsky JD, Hargraves JL . Health care perceptions and experiences. . September 2000;(30):1-6.

Health Serv Res.

19. Tye S, Phillips K, Liang S. Moving beyond the typologies of managed care: the example of health plan predictors of screening mammography. 2004;39(1):179-206.

Health Serv Res

20. Phillips KA, Haas JS, Liang SY, et al. Are gatekeeper requirements associated with cancer screening utilization? . 2004;39(1):153-178.

Am J Manag Care.

21. Liang S, Phillips KA, Tye S, Haas JS, Sakowski J. Does patient cost sharing matter? Its impact on recommended versus controversial cancer screening services. 2004;10:99-107.

Med Care.

22. Haas JS, Phillips KA, Sonneborn D, McCulloch CE, Liang SY. The effect of managed care insurance on the use of preventative care for specific ethnic groups in the United States. 2002; 40(9):743-751.

Health Serv Res

23. Mazor KM, Clauser BE, Field T, Yood RA, Gurwitz JH. A demonstration of the impact of response bias on the results of patient satisfaction surveys. . 2002;37(5):1403-1417.

Health Technol Assess.

24. Crow R, Gage H, Hampson S, et al. The measurement of satisfaction with health care: implications for practice from a systematic review of the literature. 2002;6(32):1-244.

Service Quality: New Directions in Theory and Practice.

25. Bolton RN. Linking customer satisfaction to service operations and outcomes. In: Rust RT, Oliver R, eds. Thousand Oaks, Calif: Sage Publications; 1994:173-200.

J Consumer Res.

26. Oliver R. Cognitive, affective and attribute bases of the satisfaction response. 1993;20(3):418-430.

Service Quality: New Directions in Theory and Practice.

27. Rust RT, Oliver R. Service quality: insights and managerial implications from the frontier. In: Rust RT, Oliver R, eds. Thousand Oaks, Calif: Sage Publications; 1994:1-20.

Design and Methods of the Medical Expenditure Panel Survey Household Component: MEPS Methodology Report No. 1.

28. Cohen J. Rockville, Md: Agency for Health Care Policy and Research (now the Agency for Healthcare Research and Quality); 1997.

Employee Benefit Plan Rev.

29. CareData survey yields mixed reviews on plan satisfaction. 1997;51(8):24-28.

Designing and Conducting Health Surveys: A Comprehensive Guide

30. Aday LA. . San Francisco, Calif: Jossey Bass; 1996.

Soc Sci Med

31. Hall JA, Dornan MC. Patient sociodemographic characteristics as predictors of satisfaction with medical care: a meta-analysis. . 1990;30(7)811-818.

Health Care Manage Rev.

32. Bowers M, Swan J, Koehler W. What attributes determine quality and satisfaction in health care delivery? 1994;19(4):49-55.

Med Care Res Rev.

33. Kerr EA, Hays RD, Lee ML, Siu AL. Does dissatisfaction with access to specialists affect the desire to leave a managed care plan? 1998;55(1):59-77.

Related Videos
Related Content
© 2024 MJH Life Sciences
AJMC®
All rights reserved.