Despite increased enrollment, the role of Medicare managed care in explaining declines in preventable hospitalization rates diminished over time.
To examine the association between preventable hospitalization rates and proportions of managed care enrollment at the primary care service area level.
The study used the Healthcare Cost and Utilization Project State Inpatient Data from the Agency for Healthcare Research and Quality for Arizona, Massachusetts, and New York for the years 1995 and 2005 to examine the association between preventable hospitalization rates and proportions of managed care enrollment in 1995 and 2005. The period 1995-2005 was marked by the beginning and end of several legislative and policy initiatives causing changes in elderly hospitalization patterns as well as Medicare managed care enrollment patterns. The study used ordinary least squares regressions, adjusting for heteroscedasticity. A cross-sectional analysis was used to examine the association each year. A pooled sample analysis over years tested the changes in relative contributions of managed care over time.
Preventable hospitalization rates were inversely associated with Medicare managed enrollment in both years. This association was, however, found to be weaker in 2005 than in 1995. The decline in contributions of managed care was also statistically significant.
Despite increased managed care enrollment, the role of Medicare managed care in explaining declines in preventable hospitalization rates diminished over time. The results could be explained by the growth of private fee-for-service types of managed care plans and the resultant decline in emphasis on care coordination relative to health maintenance organization plans.
(Am J Manag Care. 2012;18(8):e280-e290)Small area analysis was used to examine the association between preventable hospitalization rates and proportions of Medicare managed care enrollment.
Analysis of preventable hospitalizations has become an established tool for assessment of primary care access and quality. Alternatively known as ambulatory care sensitive condition (ACSC) admissions, a lower rate of preventable hospitalization has become an accepted indicator of access to quality primary care. Previous research found Medicare managed care to be inversely related to preventable hospitalization rates. In this study, we examine this association in 2 time periods, 1995 and 2005, in 3 US states: Arizona, Massachusetts, and New York. The period 1995-2005 is particularly significant in terms of changes in hospitalization patterns for the elderly. Two important pieces of Medicare legislation, the Balanced Budget Act (BBA) of 1997 and Medicare Modernization Act (MMA) of 2003, became effective during this period, both affecting elderly patients’ use of preventive services. While Medicare managed care enrollment faced a setback in the post-BBA period, it soared in the post-MMA period as a result of changes in payment structure. A major part of the increase occurred in private fee-for-service managed care (PFFS) plans. These PFFS plans resembled the fee-for-service (FFS) plans and provided fewer incentives for care coordination than health maintenance organizations (HMOs), a model that dominated the market in the earlier years. Studying the association of Medicare managed care with preventable hospitalization rates in the 2 time periods enables examination of how these changes affected the beneficiaries’ access to quality primary care.
Managed care plans are defined as health insurance plan types that actively use utilization controls and care management practices, in contrast to traditional indemnity plans, which pay providers on an FFS basis.1 Although private managed care health plans have been an option for Medicare beneficiaries for many years, legislative changes have been enacted to encourage greater plan and beneficiary participation. The first major changes were mandated in the BBA of 1997, which created the Medicare+ Choice program. In the MMA of 2003, the Medicare+ Choice program was replaced with the Medicare Advantage program, modifying the determination of plan payments and expanding the types of private health plans eligible to participate.
Under Medicare Advantage, many types of health plans are eligible to participate in Medicare. The types of plans that had been available include HMOs, preferred provider organizations, PFFS plans, and medical savings account plans. Regional preferred provider organizations and special needs plans are new options. With the exception of PFFS plans and medical savings account plans, Medicare Advantage plans must now offer at least 1 benefit package that includes prescription drug coverage (Medicare Part D) in each area they serve.2
Although the HMOs were the most dominant form of managed care up until at least 2005, PFFS plans saw very rapid growth between 2001 and 2005. PFFS enrollment and plan availability have grown rapidly since enactment of the MMA and implementation of the Part D Medicare drug benefit.3 The number of PFFS enrollees grew from 25,587 to 208,990 between 2003 and 2005.4 An analysis of the Centers for Medicare & Medicaid Services Geographic Service Area file finds that the HMO enrollment dropped by 10%, while PFFS enrollment increased by almost 8-fold between 2001 and 2005. Although HMOs continued to dominate the market, their market share dropped from 99% to 94% within the 5-year period, while the market share of PFFS plans increased from 0.3% to 2.7%. PFFS plans most closely resemble a privately administered version of traditional FFS Medicare and share few characteristics with Medicare managed care plans such as HMOs and preferred provider organizations.3 It has been noted that the firms that offer PFFS plans are not required to provide a plan with a Medicare Part D drug benefit, nor are they required to have quality and utilization review and reporting procedures. Moreover, unlike other Medicare Advantage plans such as HMOs and preferred provider organizations, PFFS plans are unable to guarantee access to physicians and other providers for their enrollees, and have limited ability to coordinate or to manage care.4,5
Because of the multitude of changes in the socioeconomic and policy environment, 1995-2005 is an eventful time span. That period marks the pre-post era in which 2 pieces of Medicare legislation were passed, both significantly impacting Medicare managed care enrollment. The year 1995 marks the early stage of Medicare managed care growth when Medicare managed care market composition was dominated by HMOs, the strongest form of care management along the spectrum of managed care plans. The year 2005 represents a strategic time when the managed care markets were still dominated by HMOs, but other managed care products also saw rapid growth. The study uses these 2 critical years to test the association of Medicare managed care with preventable hospitalizations, with the purposes of highlighting the effect of care coordination provided in Medicare managed care plans and identifying changes in this effect over time.
Objectives and Conceptual Hypothesis
A growing body of literature on the elderly has found that Medicare beneficiaries in managed care receive more preventive services and have better outcomes than their FFS counterparts.1,6-9 In general, managed care plans can directly reduce preventable hospitalizations by making more primary and preventive services available to their constituents. Accordingly, a lower rate of preventable hospitalization was proposed as an indicator of better health plan performance.10 Although the role of managed care market penetration and the spillover effects on other providers could be mixed, previous research using individual discharge data found Medicare managed care enrollment to be associated with less use of hospitals for preventable conditions.1,9
This study departs from earlier approaches that examined individual-level discharge data to examine the likelihood of preventable hospitalization associated with Medicare managed care enrollment. In this study, we developed an area-level estimate of preventable hospitalization rates and examined variations in these rates. The unit of analysis is primary care service areas (PCSAs), which are small primary care market areas, defined by FFS Medicare patient flow to physician offices.11 The factors underlying variations in admission rates across small areas have been the focus of many scientific and policy-related studies.12-16 Although a few prior studies and reports17,18 examined the national trends in preventable hospitalization rates, no studies examined the small area variation in these rates and the factors that contributed to changes in preventable hospitalization rates over time.
This study investigated the following questions:
1. Whether and how the proportions of Medicare managed care enrollment in a market area were associated with rates of preventable hospitalizations at each cross-section of time.
2. Whether and how the marginal contributions of managed care to the preventable hospitalization rates changed over time.
The following hypotheses were made to address these research questions: (1) The proportions of Medicare managed care enrollment will be negatively associated with preventable hospitalization rates in an area at each point in time in our study. (2) However, a decline in the marginal contributions of Medicare managed care enrollment to the preventable hospitalization rates is anticipated over time.
The first hypothesis is consistent with the conceptual framework and findings reported above. It is expected that the managed care climate in which the elderly seek care can support greater accessibility of primary care, which subsequently could affect preventable hospitalization rates among elderly patients. The second hypothesis follows from an analysis of changes in the elderly managed care market during 1995-2005 as described in the Background section.
Several changes occurred in the elderly managed care market during the period of study that could impact the preventable hospitalization rates: (1) the enrollment in staff/ group model HMOs peaked in 1995-1996 and commercial premiums actually went down in absolute value, which could have lowered Medicare preventable admissions in that year; (2) after the 1997 BBA legislation, some plans dropped out of market areas, and one would expect preventable admissions to increase if other factors held constant; (3) the big growth in Medicare Advantage enrollment came after the 2003 MMA legislation, but that was mostly growth in PFFS Medicare Advantage plans, which had less stringent controls on inpatient utilization and less emphasis than HMO models on care coordination and care management.3-5 Thus, perhaps the bulk of PFFS enrollment came from FFS Medicare. If so, then PFFS plans would have to have inpatient utilization controls that only were marginally better than FFS Medicare in order to generate a downward trend. On the other hand, the Medicare HMO plans in 1995 were probably more organized and effective in delivering primary care.
Hypotheses About Covariates
The study uses a multivariate model to test main hypotheses, using covariates based on the past research concerned with ACSC or preventable hospitalizations.13,19-21 Some of the factors affecting the demand for outpatient care, and hence associated with the hospitalization for preventable conditions, include poverty, education level, public and private insurance,22,23 and disease severity.24 A number of these determinants could confound with race and ethnicity. The supply factors associated with preventable hospitalizations are inpatient bed capacity,20,25 supplies of primary care physicians, and physician practice patterns.19,26 Area characteristics such as the proportions of elderly patients living in poverty or in a rural location are expected to exert important influences on increased rates of preventable admissions.13,27 Studies have found the degree of remoteness and rural residence to be positively associated28 and population density to be negatively associated29 with preventable admissions.
The study used hospital discharge data (Healthcare Cost and Utilization Project State Inpatient Data from the Agency for Healthcare Research and Quality) for Arizona, Massachusetts, and New York for the years 1995 and 2005.30 Many of the states with high managed care penetration in 2005 (eg, California) either did not have or did not report hospital discharge data on Medicare HMOs back in 1995. The criteria used in selecting the 3 states included the availability of Medicare managed care data for both 1995 and 2005 in the Healthcare Cost and Utilization Project State Inpatient Data, as well as the representativeness of these 3 states for the entire US population and market conditions.
The sample statistics on means over 1995-2005 for individual states (data not shown) indicate wide diversities in racial/ ethnic mix, socioeconomic conditions, and geographic spread of patients and hospitals over the study period. For example, New York had the highest proportion of African Americans (6%), Arizona had the highest proportion of Hispanics (13%) and other races (13%), and Massachusetts had the highest proportion of whites (94%) in the total hospitalized elderly population in these 3 states. Arizona and New York both had higher percentages of patients in rural areas (39% and 41%, respectively) than Massachusetts (24%), whereas elderly Massachusetts patients were less likely to have income under the federal poverty limit (8% vs 14% and 10% in Arizona and New York, respectively) and had a higher average household income ($49,928 vs $35,822 and $43,420 in Arizona and New York, respectively). Massachusetts had more primary care physicians per 1000 population (0.68 vs 0.57 and 0.61 in Arizona and New York, respectively) and a higher proportion of hospitalized elderly individuals in the 85+ age group (22% vs 14% and 20% in Arizona and New York, respectively).
Geographic characteristics also varied across states, with the average distance to hospitals significantly higher in Arizona (55 miles) than in either New York or Massachusetts (16 miles and 14.5 miles, respectively). The 3 states diverged in terms of population densities as well, with Arizona having the lowest and New York having the highest density.
These 3 states also were characterized by higher-thanaverage Medicare managed care penetration rates in 2005. The enrollment in Medicare Advantage plans as a share of all Medicare beneficiaries was 18%, 26%, and 15.9% in New York, Arizona, and Massachusetts, respectively, compared with the national rate of 12.7% in 2005, with a consistent across-the-board decline over 2000-2005.31 More recent data show that, collectively, the Medicare Advantage to elderly (>65 years) population ratio in these 3 states was slightly higher than the US average (33% vs 29%) in 2011.31 The Medicare managed care growth rate, however, showed different patterns across states over 1995-2005 in our study sample. For example, while 0.8% of all hospitalized elderly patients in New York were Medicare managed care enrollees in 1995, the corresponding rates for Massachusetts and Arizona were, respectively, 3% and 9%. By 2005, New York, Arizona, and Massachusetts had comparable rates: 10%, 12%, and 13%, respectively, of hospitalized elderly patients had Medicare managed care insurance, with New York showing the most rapid growth and Arizona showing the least. The 3 states had both similar and different characteristics to collectively represent the sampling framework for our analysis.
We first selected all elderly (aged >65 years) patients who were hospitalized for any condition in these 3 states. We then selected 20 ACSCs based on criteria developed by Billings et al23 and selected all elderly patients who were hospitalized with any of these conditions in the above states to use in our analysis of preventable hospitalization (alternatively called ACSC) rates over time. ACSCs include congenital syphilis; immunization-related and preventable conditions; severe ear, nose, and throat infections; chronic obstructive pulmonary disease; diabetes; convulsions; gastroenteritis requiring hospitalization; asthma; congestive heart failure; angina; tuberculosis; hypertension; cellulitis; hypoglycemia; kidney/urinary tract infection; dehydration—volume depletion; iron deficiency anemia; nutritional deficiencies; failure to thrive; pelvic inflammatory disease; and certain dental conditions. Although these conditions were originally developed for the under-65 population, these were validated and used for the elderly age group in subsequent research.9,10,16
We classified patients by their area of residence and used PCSA as the natural market area in which patients receive preventive care. PCSAs were developed using Medicare utilization data to represent geographic approximations of markets for primary care services received by the elderly. These regions have been validated in previous research on elderly patients’ access to preventive care services.10,13 PCSAs are generally smaller and more numerous than counties.
We constructed 2 multivariate models, 1 for each year, to understand the relative importance of Medicare managed care enrollment, while controlling for other predictors of preventable admission rates. Since the unit of observation is PCSAs in each state, the data for 3 states were combined to get a higher sample size. We focused on all ACSCs combined for the multivariate models and selected all elderly patients hospitalized with all ACSCs in the 3 sample states. The outcome variable was ACSC hospitalization rates by PCSAs, derived as the number of ACSC admissions (ie, those admitted with ACSCs) of elderly (aged >65 years) residents of each PCSA, divided by the number of all hospitalizations of elderly patients living in that PCSA. To better capture the atrisk population, we used total hospital discharges by resident population as denominators in lieu of resident population in the respective PCSAs.
As control variables, we used 2 sets of factors: compositional factors describing hospitalized patients and environmental factors defining community characteristics. All factors were aggregated to the PCSA level. The compositional factors represent characteristics of all elderly hospitalized patients from the area and include insurance coverage, disease severity (using a proxy measure of average distance traveled), and socioeconomic composition (race, age, income) of all elderly hospitalized patients who are residents of the PCSA. Community characteristics include availability of health service resources (including inpatient hospital capacity and primary care physician supply), population density, rural location, and poverty levels in the PCSA. These variables were selected based on the conceptual framework described above.
shows a complete list of the independent variables, methods of calculation, and sources of data used in the regression models. The primary explanatory variable for this study is the proportion of hospitalized elderly beneficiaries from a PCSA who were enrolled in Medicare managed care plans. The rest of the insurance coverage for the elderly included Medicare FFS, private managed care, and all other payers. These were not separately identified in the study. It should be noted that Medicare FFS accounted for the major share of the elderly patients’ insurance, 87% and 79% in 1995 and 2005, respectively, in these 3 states. The share of Medicare managed care increased from 2% to 11% in these states between 1995 and 2005 ().
We used the ordinary least squares method (OLS) to estimate multivariate regression models for each year. The initial tests showed that preventable hospitalization rates closely followed a normally distributed (bell-shaped) curve, so the OLS analysis should be appropriate. We conducted additional tests for heteroscedasticity and corrected standard errors in our models where appropriate, using Stata (version 8) commands. The results are presented in . In order to further assess whether Medicare managed care enrollment had incremental effects suggestive of influences on changes in preventable admission rates over time, additional analysis was done by pooling the sample for 2 years to analyze whether Medicare managed care had statistically significant changes in effects. To perform this analysis, we used time (2005 = 1, 1995 = 0) as the fixed effect and examined the interactions of all variables with time, adjusting for heteroscedasticity (using Stata version 8 commands). The selected results from this model showing interaction coefficients are reported in .
Table 1 reports the mean (unweighted) values of all the variables used in the regression analysis for 1995 and 2005. The table also reports the changes over 1995-2005 that were statistically significant, based on results of 2-sample t tests conducted for each variable. Table 1 indicates that ACSC or preventable admission rates for the elderly declined significantly as a whole in the 3 states. A few major significant changes occurred during the 11-year period in the demographic, contextual, and policy environments. First, significant increase in the Medicare managed care enrollment occurred over the 11- year period. The proportion of the elderly hospitalized Medicare beneficiaries who were enrolled in managed care plans increased from 2% in 1995 to 11% overall in these 3 states. Further analysis by state (data not shown) indicates that these proportions increased from 9.1% to 11.79% in Arizona, 0.75% to 10.37% in New York, and 3.02% to 12.96% in Massachusetts between 1995 and 2005. Among other notable changes, the average number of primary care physicians per 1000 population increased from 0.48 to 0.75, while the proportions of elderly patients with income under the federal poverty level declined. The sociodemographic composition also changed, with greater increase in hospitalizations of minority and aged elderly. Average household income increased, and population density declined. In addition, the average proportion of rural patients declined, and the average inpatient hospital capacity per 1000 population increased.
Table 2 shows the OLS regression results for 1995 and 2005 using the same set of variables for comparison. The results reported in Table 2 support our first hypothesis that the proportion of enrollment under Medicare managed care plans was inversely associated with rates of preventable hospitalization in 1995. The relationship was not found statistically significant in 2005, although the expected negative association prevailed. Thus, the results also support our second hypothesis that marginal benefits of managed care may have declined over time. Table 2 shows that marginal contributions toward declines in preventable hospitalization rates dropped in 2005 (β = —0.015, P >.05) relative to 1995 (β = —0.128, P <.01) for managed care enrollees.
shows the results of the multivariate regression model that examined the interactions of the independent variables with time to discern the changes attributed to a shift in time. The table indicates that there was a substantial and significant decline over time in the magnitude of the negative association. The results show that the decline in effects of Medicare managed care in these 3 states as a whole was statistically significant at a probability level less than 1%.
Among other factors reported in Tables 2 and 3, younger elderly age groups (aged 65-74 years) had a negative association with ACSC hospitalization rates (relative to those aged >85 years). The 2 racial and ethnic groups, African Americans and Hispanics, were both associated with increased ACSC hospitalizations in 1995, with both experiencing relative declines in effects over time. The primary care physician density was a significant predictor in 2005, being associated with reductions in ACSC rates. The area socioeconomic indicators, measured by both average area income and proportions below the federal poverty level, were found to have played an important role in ACSC hospitalizations over time, the latter becoming a stronger predictor of preventable hospitalization rates over time.
Preventable hospitalization rates for the elderly declined significantly as a whole in the 3 states. By conducting 2 crosssectional studies as well as testing for the effects of the shifts in the time parameters, we were able to assess whether the Medicare managed care enrollment played a significant role in driving these hospitalization rates down over time. This question was of policy interest because the Medicare managed care enrollment expanded in size dramatically over this period, and this increase was expected to result in lower rates of preventable hospitalizations, according to the expected negative association reported in previous literature.1,9
The study indeed reported an inverse relationship between Medicare managed care and preventable hospital admission rates at each cross-section of time: 1995 and 2005. However, our cross-sectional analysis shows that although the Medicare managed care enrollment increased substantially between 1995 and 2005, its marginal contributions to preventable hospitalization rates were lower in 2005 than in 1995 (were actuallynonsignificant in 2005), and the decline in effects over time was also statistically significant. To test the robustness of this finding, a sensitivity test was conducted using a restricted model, pooling samples over years, forcing all coefficient estimates to be the same over time, and introducing time internactions with Medicare managed care alone. Since the years are far apart, the assumption of constant coefficients over time appeared less tenable than the model used in the main body of the text. However, the results support our previous finding that the association of Medicare managed care with preventable hospitalization rates significantly declined over time (). As hypothesized, a likely reason could be the growth in PFFS plans after 2003. These plans have less emphasis than HMO models do on care coordination and prevention, which might have made them less willing to provide primary care and more inefficient in doing so.
Minorities, aged elderly patients, and poor patients had higher preventable hospitalization rates, which is consistent with previous research findings related to socioeconomic status and preventable hospitalizations.10,23,32 Analysis of these findings could highlight important factors that might be associated with the observed reductions in preventable hospitalization rates over time in the elderly. For example, although the percentage of hospitalizations increased among minority elderly patients between 1995 and 2005, the association between minorities and preventable admission rates did not commensurately increase. The association between proportions of Hispanics and African Americans with preventable hospitalizations actually dropped between 1995 and 2005. There was a decline over time in the proportion of elderly individuals with income under the federal poverty level, which could have mitigated the growth of preventable admissions attributable to poverty. The proportion of the elderly population living in rural areas significantly declined, as did the marginal contribution of rural status to preventable admissions, indicating better geographic access over time. Despite increased primary care physician density, the marginal effects on the preventable hospitalization rate did not significantly increase, which may be explained by an increased supply and distribution of primary care physicians over the last decade and a potential decline in additional benefits from further increase.
Several limitations of the study should be acknowledged. The 3 states examined are not fully representative of the entire US population and market conditions, although they were selected to represent diverse socioeconomic and geographic characteristics. All 3 states had a well-established Medicare managed care presence, so the findings should be interpreted in that context. Another limitation could be that the study examined only 1 outcome indicator for the quality of primary care. Although our quality indicator is quite important regarding total social cost of care and a signal of effective management of chronic illness, other measures could be examined.
Data limitations did not allow us to adjust well for the severity of morbidity in the elderly population in the comparative analysis for 1995-2005, which may cause omittedvariables bias in the impact of managed care penetration, to the extent that managed care plans adversely or beneficially select patients by illness severity. The findings regarding favorable selection into HMOs, however, vary across studies. For example, a recent study did not find evidence that Medicare Advantage plans benefited from positive selection of healthier enrollees.33 Mello et al report that Medicare HMO enrollees were not markedly healthier than nonenrollees on most measures and that the extent of favorable HMO selection found in different studies may be influenced by the way in which the effects are reported.34 It has also been reported that favorable selection into Medicare managed care may decline as the managed care market matures and market share increases.35 To the extent that is true, Medicare managed care enrollees will consist of sicker patients as the Medicare market matures over time. However, another study reports lack of strong evidence that an association exists between the degree of favorable HMO selection and the extent of HMO market penetration.34 Thus, existing evidence does not strongly suggest that results of the current study would be subject to bias due to favorable selection into Medicare managed care, or that the amount of bias, if any, would have varied over time.
To account for possible health status differences between managed care and non—managed care enrollees, the current study adjusted directly for severity-of-illness differences using a number of variables such as poverty level, age composition, racial composition, and average income. In addition, following previous research that reported sicker patients more often getting care in distant hospitals,36 a PCSA-specific index measuring average distance from patient’s zip code to hospital zip code was used. This index, used in conjunction with additional measures such as population density and rurality, could appropriately serve as a proxy for disease severity.24,36 The index was found to be inversely related to preventable hospitalization rates in an area, consistent with the fact that a significant proportion of preventable admissions are due to chronic conditions.
Despite increased managed care enrollment, the role of Medicare managed care in explaining declines in preventable hospitalization rates significantly diminished over time in 3 states with high managed care penetration. The results could be explained by various changes in the Medicare managed care market during this period, including the growth of PFFS plans and the resultant potential decline in emphasis on care coordination relative to HMO plans. Using small area analysis, the study indicates that although HMO models of managed care predicted declines in preventable hospitalization rates among the elderly, the association was found weaker as incentive systems changed and other managed care products grew after 2003. The study’s findings are important in light of an accelerated enrollment in PFFS plans after 2005 and a growing debate on Medicare managed care plan performance in recent years.3,4 Future studies should use more recent data to examine the evolving role of various managed care plans in providing primary care to the elderly. The study indicates that improvements in socioeconomic conditions and geographic access may have helped improve the quality of primary care received by the elderly over the last decade.
The author would like to acknowledge the state data organizations in Massachusetts, Arizona, and New York that participated in the Healthcare Cost and Utilization Project State Inpatient Databases.
Author Affiliation: From the Agency for Healthcare Research and Quality, Rockville, MD.
Funding Source: This research was funded wholly by the author’s employer, the Agency for Healthcare Research and Quality.
Author Disclosures: The author reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis; and obtaining funding.
Address correspondence to: Jayasree Basu, PhD, Agency for Healthcare Research and Quality, 540 Gaither Rd, Rockville, MD 20850. E-mail: Jayasree.email@example.com. Basu J, Mobley LR. Do HMOs reduce preventable hospitalizations for Medicare beneficiaries? Med Care Res Rev. 2007;64(5):544-567.
2. National Health Policy Forum website. http://www.nhpf.org. https://www.nhpf.org/library/details.cfm/2542. Published December 6, 2006. Accessed July 30, 2012.
3. Blum J, Brown R, Frieder M. An Examination of Medicare Private Fee-for-Service Plans. Issue brief. Prepared for Henry J. Kaiser Family Foundation. http://www.allhealth.org/briefingmaterials/examinationof medicareprivatefee-for-serviceplans-824.pdf. Published March 2007. Accessed January 10, 2012.
4. Neuman P. Private Fee-for-Service Plans in Medicare: Rapid Growth and Future Implications. http://www.kff.org/medicare/upload/7648.pdf. Published May 22, 2007. Accessed September 10, 2012.
5. Biles B, Adrian E, Guterman S. Medicare Advantage’s Private Fee-for- Service Plans: Paying for Coordinated Care Without the Coordination. Issue brief. Vol 44. The Commonwealth Fund. http://www.commonwealthfund.org/Publications/Issue-Briefs/2008/Oct/Medicare- Advantages-Private-Fee-for-Service-Plans--Paying-for-Coordinated-Care-Without-the-Coordinat.aspx. Published October 21, 2008. Accessed February 8, 2012.
6. Riley GF, Potosky AL, Lubitz JD, Brown JD. Stage of cancer at diagnosis for Medicare HMO and fee-for-service enrollees. Am J Public Health. 1994;84(10):1598-604.
7. Riley GF, Potosky AL, Klabunde CN, Warren JL, Ballard-Barbash R. Stage at diagnosis and treatment patterns among older women with breast cancer: an HMO and fee-for-service comparison. JAMA. 1999;281(8): 720-726.
8. Rizzo J. Are HMOs bad for health maintenance? Health Econ. 2005;14 (11):1117-1131.
9. Basu J. Medicare managed care and primary care quality: examining racial/ethnic effects across states. Health Care Manage Sci. 2012;15(1): 15-28.
10. Blustein J, Hanson K, Shea S. Preventable hospitalizations and socioeconomic status. Health Aff (Millwood). 1998;17(2):177-189.
11. Goodman DC, Mick SS, Bott D, et al. Primary care service areas: a new tool for the evaluation of primary care services. Health Serv Res. 2003;38(1, pt 1):287-309.
12. Restuccia J, Shwartz M, Ash A, Payne S. High hospital admission rates and inappropriate care. Health Aff (Millwood). 1996;15(4): 156-163.
13. Mobley L, Root E, Anselin E, Gracia N, Koschinsky J. Spatial analysis of elderly access to primary care services. Int J Health Geogr. 2006;5:19.
14. Pathman DE, Ricketts TC 3rd, Konrad TR. How adults’ access to outpatient physician services relates to the local supply of primary care physicians in the rural southeast. Health Serv Res. 2006;41(1):79-102.
15. Hossain MM, Laditka JN. Using hospitalization for ambulatory care sensitive conditions to measure access to primary health care: an application of spatial structural equation modeling. Int J Health Geogr. 2009;8:51.
16. Basu J, Mobley LR. Impact of local resources on hospitalization patterns of Medicare beneficiaries and propensity to travel outside local markets. J Rural Health. 2010;26(1):20-29.
17. Kozak LJ, Hall MJ, Owings MF. Trends in avoidable hospitalizations, 1980-1998, Health Aff (Millwood). 2001;20(2):225-232.
18. Stranges E, Friedman B. Potentially Preventable Hospitalization Rates Declined for Older Adults, 2003—2007. 2009. H-CUP Statistical Brief #83. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb83.pdf. Published December 2009. Accessed February 1, 2012.
19. Parchman ML, Culler S. Primary care physicians and avoidable hospitalization. J Fam Pract. 1994;39(2):123-128.
20. Krakauer H, Jacoby I, Millman M, Lukomnik JE. Physician impact on hospital admission and on mortality rates in the Medicare population. Health Serv Res. 1996;31(2):191-211.
21. Friedman B, Basu J. Health insurance, primary care, and preventable hospitalization of children in a large state. Am J Manag Care. 2001; 7(5):473-481.
22. Weissman JS, Gatsonis C, Epstein AM. Rates of avoidable hospitalization by insurance status in Massachusetts and Maryland. JAMA. 1992;268(17):2388-2394.
23. Billings J, Zeitel L, Lukomnik J, Carey TS, Blank AE, Newman L. Impact of socioeconomic status on hospital use in New York City. Health Aff (Millwood). 1993;12(1):162-173.
24. Basu J, Friedman B. A re-examination of distance as a proxy for severity of illness and implications for differences in utilization by race/ ethnicity. Health Econ. 2007;16(7):687-701.
25. Goodman DC, Fisher ES, Gittelsohn A, Chang CH, Fleming C. Whyare children hospitalized? the role of non-clinical factors in pediatric hospitalizations. Pediatrics. 1994;93(6, pt 1):896-902.
26. Homer CJ, Szilagyi P, Rodewald L, et al. Does quality of care affect rates of hospitalization for childhood asthma? Pediatrics. 1996;98(1): 18-23.
27. Laditka JN, Laditka SB, Probst JC. Health care access in rural areas: evidence that hospitalization for ambulatory care-sensitive conditions in the United States may increase with the level of rurality. Health Place. 2009;15(3):731-740.
28. Ansari Z, Laditka JN, Laditka SB. Access to health care and hospitalization for ambulatory care sensitive conditions. Med Care Res Rev. 2006;63(6):719-741.
29. Schreiber S, Zielinsky T. The meaning of ambulatory care sensitive admissions: urban and rural perspectives. J Rural Health. 1997;13(4): 276-284.
30. Agency for Healthcare Research and Quality. HCUP SID Database Documentation. Healthcare Cost and Utilization Project (HCUP). http:// www.hcup-us.ahrq.gov/db/state/siddbdocumentation.jsp. Published 2004. Accessed December 2005.
31. Kaiser Family Foundation. Medicare health and prescription drug plan tracker. http://healthplantracker.kff.org/. Published 2011. Accessed June 27, 2012.
32. Chang CF, Mirvis DM, Waters TM. The effects of race and insurance on potentially avoidable hospitalizations in Tennessee. Med Care Res Rev. 2008;65(5):596-616.
33. Basu J, Mobley LR. Medicare managed care plan performance: a comparison across hospitalization types. Medicare Medicaid Res Rev. 2012;2(1):E1-E20. http://www.cms.gov/Research-Statistics-Data-and-Systems/Research/MMRR/MMRR-Published-Articles-Items/CMS1255460.html.
34. Mello MM, Stearns SC, Norton EC, Ricketts TC 3rd. Understanding biased selection in Medicare HMOs. Health Serv Res. 2003;38(3): 961-992.
35. Call KT, Dowd B, Feldman R, et al. Selection experiences in Medicare HMOs: pre-enrollment expenditures. Health Care Financ Rev. 1999;20(4):197-209.
36. Welch HG, Larson EB, Welch WP. Could distance be a proxy for severity-of-illness? a comparison of hospital costs in distant and local patients. Health Serv Res. 1993;28(4):441-458.