Abstract: OBJECTIVE: To explore whether geographic variations in Medicare hospital utilization rates are due to differences in local hospital capacity, after controlling for socioeconomic status and disease burden, and to determine whether greater hospital capacity is associated with lower Medicare mortality rates. DATA SOURCES/STUDY SETTING: The study population: a 20 percent sample of 1989 Medicare enrollees. Measures of resources were based on a national small area analysis of 313 Hospital Referral Regions (HRR). Demographic and socioeconomic data were obtained from the 1990 U.S. Census. Measures of local disease burden were developed using Medicare claims files. STUDY DESIGN: The study was a cross-sectional analysis of the relationship between per capita measures of hospital resources in each region and hospital utilization and mortality rates among Medicare enrollees. Regression techniques were used to control for differences in sociodemographic characteristics and disease burden across areas. DATA COLLECTION/EXTRACTION METHODS: Data on the study population were obtained from Medicare enrollment (Denominator File) and hospital claims files (MedPAR) and U.S. Census files. PRINCIPAL FINDINGS: The per capita supply of hospital beds varied by more than twofold across U.S. regions. Residents of areas with more beds were up to 30 percent more likely to be hospitalized, controlling for ecologic measures of socioeconomic characteristics and disease burden. A greater proportion of the population was hospitalized at least once during the year in areas with more beds; death was also more likely to take place in an inpatient setting. All effects were consistent across racial and income groups. Residence in areas with greater levels of hospital resources was not associated with a decreased risk of death. CONCLUSIONS: Residence in areas of greater hospital capacity is associated with substantially increased use of the hospital, even after controlling for socioeconomic characteristics and illness burden. This increased use provides no detectable mortality benefit.

Abstract: OBJECTIVE: To examine the pathways between income inequality, self-rated health, and mortality in the United States. DATA SOURCE: The first National Health and Nutrition Examination Survey and Epidemiologic Follow-up Study. DESIGN: This was a longitudinal, multilevel study. DATA COLLECTION: Baseline data were collected on county income inequality, individual income, age, sex, self-rated health, level of depressive symptoms, and severity of biomedical morbidity from physical examination. Follow-up data included self-rated health assessed in 1982 through 1984 and mortality through 1987. PRINCIPAL FINDINGS: After adjustment for age and sex, income inequality had a modest independent effect on the level of depressive symptoms, and on baseline and follow-up self-rated health, but no independent effect on biomedical morbidity or subsequent mortality. Individual income had a larger effect on severity of biomedical morbidity, level of depressive symptoms, baseline and follow-up self-rated health, and mortality. CONCLUSION: Income inequality appears to have a small effect on self-rated health but not mortality; the effect is mediated in part by psychological, but not biomedical pathways. Individual income has a much larger effect on all of the health pathways.

Abstract: OBJECTIVE: To examine the service experiences and outcomes of low-income Medicaid beneficiaries with disabilities under two different models for organizing home-based personal assistance services: agency-directed and consumer-directed. DATA SOURCE: A survey of a random sample of 1,095 clients, age 18 and over, who receive services in California's In-Home Supportive Services (IHSS) program funded primarily by Medicaid. Other data were obtained from the California Management and Payrolling System (CMIPS). STUDY DESIGN: The sample was stratified by service model (agency-directed or consumer-directed), client age (over or under age 65), and severity. Data were collected on client demographics, condition/functional status, and supportive service experience. Outcome measures were developed in three areas: safety, unmet need, and service satisfaction. Factor analysis was used to reduce multiple outcome measures to nine dimensions. Multiple regression analysis was used to assess the effect of service model on each outcome dimension, taking into account the client-provider relationship, client demographics, and case mix. DATA COLLECTION: Recipients of IHSS services as of mid-1996 were interviewed by telephone. The survey was conducted in late 1996 and early 1997. PRINCIPAL FINDINGS: On various outcomes, recipients in the consumer-directed model report more positive outcomes than those in the agency model, or they report no difference. Statistically significant differences emerge on recipient safety, unmet needs, and service satisfaction. A family member present as a paid provider is also associated with more positive reported outcomes within the consumer-directed model, but model differences persist even when this is taken into account. Although both models have strengths and weaknesses, from a recipient perspective the consumer-directed model is associated with more positive outcomes. CONCLUSIONS: Although health professionals have expressed concerns about the capacity of consumer direction to assure quality, particularly with respect to safety, meeting unmet needs, and technical quality, our findings suggest that the consumer-directed service model is a viable alternative to the agency model. Because public programs are under growing pressure to address the long-term care needs of low-income people of all ages with disabilities, the Medicaid personal assistance benefit needs to be reassessed in light of these findings. Consumer-directed models may offer a less elaborate and possibly less costly option for organizing supportive services at home. Study limitations may limit the generalizability of these findings. This was a natural experiment, in which only some counties offered both service models and counties assigned recipients to a service model. The use of a telephone survey excluded important recipient subsets, notably people with severe cognitive impairments. A more definitive study would include direct observations as well as survey approaches.

Abstract: OBJECTIVE: To evaluate the validity of racial/ethnic information in California birth certificate data. DATA SOURCES: Computerized birth certificate data and postpartum interviews with California mothers. STUDY DESIGN AND DATA COLLECTION: Birth certificates were matched with face-to-face structured postpartum interviews with 7,428 mothers to compare racial/ethnic information between the two data sources. Interviews were conducted in Spanish or English during delivery stays at 16 California hospitals, 1994-1995. PRINCIPAL FINDINGS: The sensitivity of racial/ethnic classification in birth certificate data was very high (94 percent to 99 percent) for African Americans, Asians/Pacific Islanders, Europeans/Middle Easterners, and Latinas (Hispanics). For Native Americans, however, the sensitivity was only 54 percent. The positive predictive value of birth certificate classification of race/ethnicity was high for all racial/ethnic groups (96 percent to 97 percent). CONCLUSIONS: Despite limited training of birth clerks, the maternal racial/ethnic information in California birth certificate data appears to be a valid measure of self-identified race and Hispanic ethnicity for groups other than Native Americans.

Abstract: OBJECTIVE: To provide an empirical examination of the effect that chronic illicit drug use has on emergency room (ER) utilization, controlling for the potential biases introduced by correlation between unobservable determinants of chronic illicit drug use and ER utilization. DATA SOURCES/STUDY SETTING: From the National Household Survey on Drug Abuse 1994 (NHSDA94). STUDY DESIGN: Chronic illicit drug use and ER utilization are analyzed for 5,384 females and 4,177 males in 1994. The study uses a two-stage estimation technique. In the first stage, sociodemographic, drug use history, and drug use risk variables are used to estimate the probability that the subject is a chronic illicit drug user (CDU). In the second stage, the first-stage estimates provide information needed to test for the possibility of bias in the estimation of ER utilization. This bias is the result of the correlation between unobservable influences on the probability that the person is a CDU and the probability that he or she uses an ER. DATA COLLECTION/EXTRACTION METHODS: The data were collected through a multistage stratified sampling design. With the use of this methodology, the resulting data set provides the most comprehensive information on household drug use. PRINCIPAL FINDINGS: Without a correction for the possibility of endogeneity bias, chronic illicit drug use is a positive (for both males and females) and a significant (for females only) determinant of the probability of using an ER for medical treatment. After a correction for endogeneity, the influence of chronic drug use remains positive and significant for females and becomes significant for males. The corresponding change in probability for females is from 6 percent to 30 percent, while for males the increase is from an insignificant 0.1 percent to a significant 36 percent change. CONCLUSIONS: We estimate that chronic drug-using females and males, after adjustments for bias, increase the probability that they use an ER by more than 30 percent compared to their casual or non-drug-using counterparts. Therefore, policymakers and health services providers may consider designing programs to bring primary care and prevention services to facilities where drug users are more likely to seek access to care, within an ER setting.

Abstract: OBJECTIVE: To examine the extent to which linkage mechanisms (on-site delivery, external arrangements, case management, and transportation assistance) are associated with increased utilization of medical and psychosocial services in outpatient drug abuse treatment units. DATA SOURCES: Survey of administrative directors and clinical supervisors from a nationally representative sample of 597 outpatient drug abuse treatment units in 1995. STUDY DESIGN: We generated separate two-stage multivariate generalized linear models to evaluate the correlation of on-site service delivery, formal external arrangements (joint program/venture or contract), referral agreements, case management, and transportation with the percentage of clients reported to have utilized eight services: physical examinations, routine medical care, tuberculosis screening, HIV treatment, mental health care, employment counseling, housing assistance, and financial counseling services. PRINCIPAL FINDINGS: On-site service delivery and transportation assistance were significantly associated with higher levels of client utilization of ancillary services. Referral agreements and formal external arrangements had no detectable relationship to most service utilization. On-site case management was related to increased clients' use of routine medical care, financial counseling, and housing assistance, but off-site case management was not correlated with utilization of most services. CONCLUSIONS: On-site service delivery appears to be the most reliable mechanism to link drug abuse treatment clients to ancillary services, while referral agreements and formal external mechanisms offer little detectable advantage over ad hoc referral. On-site case management might facilitate utilization of some services, but transportation seems a more important linkage mechanism overall. These findings imply that initiatives and policies to promote linkage of such clients to medical and psychosocial services should emphasize on-site service delivery, transportation and, for some services, on-site case management.

Abstract: OBJECTIVE: To measure the effect of body weight on employment disability. DATA SOURCES: Female respondents to the National Longitudinal Survey of Youth (NLSY), a nationally representative sample of American youth, surveyed from 1979 to 1998, merged with data from the child sample of the NLSY. STUDY DESIGN: A series of probit models and probit models with instrumental variables is estimated with the goal of measuring the effect of body weight on employment disability. The two outcomes of interest are whether a woman reports that her health limits the amount of work that she can do for pay, and whether she reports that her health limits the kind of work that she can do for pay. The models control for factors that affect the probability of health limitations on employment, such as education, cognitive ability, income of other family members, and characteristics of children in the household. Self-reports of height and weight are corrected for reporting error. PRINCIPAL FINDINGS: All else being equal, heavier women are more likely to report employment disability. However, this overall correlation may be due to any or all of the following factors: weight causing disability, disability causing weight gain, or unobserved factors causing both. Instrumental variables estimates provide no evidence that body weight affects the probability of either type of employment disability. CONCLUSIONS: This study finds no evidence that body weight causes employment disability. Instead, the observed correlation between heaviness and disability may be due to disability causing weight gain or unobservable factors causing both disability and weight gain.

Abstract: OBJECTIVE: To examine the effect of reengineering on the competitive position of hospitals. Although many promises have been made regarding outcomes of process reengineering, little or no research has examined this issue. This article provides an initial exploration of the direct effects of reengineering on the competitive cost position of hospitals and the modifying effects of implementation factors. DATA SOURCES/STUDY SETTING: Obtained for primary data from a 1996/1997 national survey of hospital restructuring and reengineering sponsored by the American Hospital Association and the Leonard Davis Institute for Health Economics. Responses from approximately 30 percent of all U.S. acute care hospitals with 100 or more inpatient beds in metropolitan service areas were combined with American Hospital Association annual survey and InterStudy HMO data in this study. STUDY DESIGN: A first-difference multivariate regression was utilized to examine the effects of reengineering and other explanatory variables on the change in the cost position of a hospital's expenses per adjusted patient day relative to its market's costs per adjusted patient day. DATA COLLECTION/EXTRACTION METHODS: The survey of hospital restructuring and reengineering was mailed to hospital chief executive officers. The CEOs identified reengineering and restructuring hospital activities over the previous five years. The extensiveness and components of reengineering and internal restructuring were identified and used in the empirical analysis. PRINCIPAL FINDINGS: Results suggest that reengineering without integrative and coordinative efforts may damage an organization's cost position. The use of steering committees, project teams, codification of the change process, and executive involvement in core changes modifies the results of reengineering to improve an organization's competitive position. CONCLUSIONS: In a national sample of hospitals, reengineering alone was not found to improve the relative cost-competitive position. Organizations attempting to improve their cost competitiveness must consider the way in which change is implemented. This research suggests that the process of change may be as important as the change instrument. Additional research is needed to explore differences between early and late adopters.

Abstract: Objective To estimate the differences in functional outcomes attributable to discharge to one of four different venues for post-hospital care for each of five different types of illness associated with post-hospital care: stroke, chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), hip procedures, and hip fracture, and to estimate the costs and benefits associated with discharge to the type of care that was estimated to produce the greatest improvement. Study setting/data sources Consecutive patients with any of the target diagnoses were enrolled from 52 hospitals in three cities. Data sources included interviews with patients or their proxies, medical record reviews, and the Medicare Automated Data Retrieval System. Analysis A two-stage regression model looked first at the factors associated with discharge to each type of post-hospital care and then at the outcomes associated with each location. An instrumental variables technique was used to adjust for selection bias. A predictive model was created for each patient to estimate how that person would have fared had she or he been discharged to each type of care. The optimal discharge location was determined as that which produced the greatest improvement in function after adjusting for patients' baseline characteristics. The costs of discharge to the optimal type of care was based on the differences in mean costs for each location. Data collection/extraction methods Data were collected from patients or their proxies at discharge from hospital and at three post-discharge follow-up times: six weeks, six months, and one year. In addition, the medical records for each participant were abstracted by trained abstractors, using a modification of the Medisgroups method, and Medicare data were summarized for the years before and after the hospitalization. Principal findings In general, patients discharged to nursing homes fared worst and those sent home with home health care or to rehabilitation did best. Because the cost of rehabilitation is high, greater use of home care could result in improved outcomes at modest or no additional cost. Conclusions Better decisions about where to discharge patients could improve the course of many patients. It is possible to save money by making wiser discharge planning decisions. Nursing homes are generally associated with poorer outcomes and higher costs than the other post-hospital care modalities.

Abstract: OBJECTIVE: To develop a mathematical model for the location of trauma care resources. DATA SOURCES/STUDY SETTING: Severely injured patients queried from Maryland hospital discharge and vital statistics data. A spatial injury profile was created by parsing these patients into ZIP codes. STUDY DESIGN: The Trauma Resource Allocation Model for Ambulances and Hospitals (TRAMAH) was formulated using integer and heuristic programming. To maximize coverage of severely injured patients, trauma centers and aeromedical depots were simultaneously sited using TRAMAH. A severe injury was considered covered if at least one trauma center was sited within a time standard by ground, or if an aeromedical depot-trauma center pair was sited in such a way that the sum of the flying time from the aeromedical depot to the scene of injury plus the flying time from the scene of injury to the trauma center was within the same time standard. PRINCIPAL FINDINGS: From 1992 to 1994, 26,774 severe injuries were considered for coverage. Across Maryland, 94.8 percent of severely injured residents had access to trauma system resources within 30 minutes and 70.3 percent had access within 15 minutes. For the same number of resources as the existing Maryland Trauma System, TRAMAH achieved a coverage objective of 99.97 percent within 30 minutes. This translated into an additional 461 severely injured people covered each year. Holding in place the trauma centers of the existing system, approximately the same percentage of coverage as that of the existing system was achieved within 15 minutes by optimally locating six fewer aeromedical depots. CONCLUSIONS: TRAMAH will allow trauma systems planners to better locate their resources with respect to spatial needs and response times.

Abstract: OBJECTIVE: To learn whether consumer reports of health plan quality can affect health plan selection. DATA SOURCES: A sample of 311 privately insured adults from Los Angeles County. STUDY DESIGN: The design was a fractional factorial experiment. Consumers reviewed materials on four hypothetical health plans and selected one. The health plans varied as to cost, coverage, type of plan, ability to keep one's doctor, and quality, as measured by the Consumer Assessment of Health Plans Study (CAHPS) survey. DATA ANALYSIS: We used multinomial logistic regression to model each consumer's choice among health plans. PRINCIPAL FINDINGS: In the absence of CAHPS information, 86 percent of consumers preferred plans that covered more services, even though they cost more. When CAHPS information was provided, consumers shifted to less expensive plans covering fewer services if CAHPS ratings identified those plans as higher quality (59 percent of consumers preferred plans covering more services). Consumer choices were unaffected when CAHPS ratings identified the more expensive plans covering more services as higher quality (89 percent of consumers preferred plans covering more services). CONCLUSIONS: This study establishes that, under certain realistic conditions, CAHPS ratings could affect consumer selection of health plans and ultimately contain costs. Other studies are needed to learn how to enhance exposure and use of CAHPS information in the real world as well as to identify other conditions in which CAHPS ratings could make a difference.

Abstract: OBJECTIVE: To compare outcome and cost-effectiveness of the two primary addiction treatment options, day hospitals (DH) and traditional outpatient programs (OP) in a managed care organization, in a population large enough to examine patient subgroups. DATA SOURCES: Interviews with new admissions to a large HMO's chemical dependency program in Sacramento, California between April 1994 and April 1996, with follow-up interviews eight months later. Computerized utilization and cost data were collected from 1993 to 1997. STUDY DESIGN: Design was a randomized control trial of adult patients entering the HMO's alcohol and drug treatment program (N = 668). To examine the generalizability of findings as well as self-selection factors, we also studied patients presenting during the same period who were unable or unwilling to be randomized (N = 405). Baseline interviews characterized type of substance use, addiction severity, psychiatric status, and motivation. Follow-up interviews were conducted at eight months following intake. Breathanalysis and urinalysis were conducted. Program costs were calculated. DATA COLLECTION: Interview data were merged with computerized utilization and cost data. PRINCIPAL FINDINGS: Among randomized subjects, both study arms showed significant improvement in all drug and alcohol measures. There were no differences overall in outcomes between DH and OP, but DH subjects with midlevel psychiatric severity had significantly better outcomes, particularly in regard to alcohol abstinence (OR = 2.4; 95% CI = 1.2, 4.9). The average treatment costs were $1,640 and $895 for DH and OP programs, respectively. In the midlevel psychiatric severity group, the cost of obtaining an additional person abstinent from alcohol in the DH cohort was approximately $5,464. Among the 405 self-selected subjects, DH was related to abstinence (OR = 2.1; 95% CI = 1.3, 3.5). CONCLUSIONS: Although significant benefits of the DH program were not found in the randomized study, DH treatment was associated with better outcomes in the self-selected group. However, for subjects with mid-level psychiatric severity in both the randomized and self-selected samples, the DH program produced higher rates of abstention and was more cost-effective. Self-selection in studies that randomize patients to services requiring very different levels of commitment may be important in interpreting findings for clinical practice.

Abstract: OBJECTIVE: To determine if a widely used proprietary risk-adjustment system, APR-DRGs, misadjusts for severity of illness and misclassifies provider performance. DATA SOURCES: (1) Discharge abstracts for 116,174 noninstitutionalized adults with acute myocardial infarction (AMI) admitted to nonfederal California hospitals in 1991-1993; (2) inpatient medical records for a stratified probability sample of 974 patients with AMIs admitted to 30 California hospitals between July 31, 1990 and May 31, 1991. STUDY DESIGN: Using the 1991-1993 data set, we evaluated the predictive performance of APR-DRGs Version 12. Using the 1990/1991 validation sample, we assessed the effect of assigning APR-DRGs based on different sources of ICD-9-CM data. DATA COLLECTION/EXTRACTION METHODS: Trained, blinded coders reabstracted all ICD-9-CM diagnoses and procedures, and established the timing of each diagnosis. APR-DRG Risk of Mortality and Severity of Illness classes were assigned based on (1) all hospital-reported diagnoses, (2) all reabstracted diagnoses, and (3) reabstracted diagnoses present at admission. The outcome variables were 30-day mortality in the 1991-1993 data set and 30-day inpatient mortality in the 1990/1991 validation sample. PRINCIPAL FINDINGS: The APR-DRG Risk of Mortality class was a strong predictor of death (c = .831-.847), but was further enhanced by adding age and sex. Reabstracting diagnoses improved the apparent performance of APR-DRGs (c = .93 versus c = .87), while using only the diagnoses present at admission decreased apparent performance (c = .74). Reabstracting diagnoses had less effect on hospitals' expected mortality rates (r = .83-.85) than using diagnoses present at admission instead of all reabstracted diagnoses (r = .72-.77). There was fair agreement in classifying hospital performance based on these three sets of diagnostic data (K = 0.35-0.38). CONCUSIONS: The APR-DRG Risk of Mortality system is a powerful risk-adjustment tool, largely because it includes all relevant diagnoses, regardless of timing. Although some late diagnoses may not be preventable, APR-DRGs appear suitable only if one assumes that none is preventable.

Abstract: OBJECTIVE: To examine the effects of selective contracting on California hospital costs and revenues over the 1983-1997 period. DATA SOURCES/STUDY SETTING: Annual disclosure data and discharge data sets for 421 California general acute care hospitals from 1980 to 1997. ANALYSIS: Using measures of competition developed from patient-level discharge data, and financial and utilization measures from the disclosure data, we estimated a fixed effect multivariate regression model of hospital costs and revenues. FINDINGS: We found that hospitals in more competitive areas had a substantially lower rate of increase in both costs and revenues over this extended period of time. For-profit hospitals lowered their costs and revenues after selective contracting was initiated relative to the cost and revenue levels of not-for-profit hospitals. The Medicare PPS has also led high-cost hospitals to lower their costs. CONCLUSIONS: The more competitive the hospital's market, the greater degree to which it has had to lower the rate of increase in costs. A similar pattern exists with regard to hospital revenues. Both of these trends appear to result from the growth of selective contracting. It remains unclear to what extent these cost reductions were the result of increased efficiency or of reduced quality. Since hospital cost growth is sensitive to the competitiveness of its market, antitrust enforcement is a critical element in any cost containment policy.

Abstract: OBJECTIVE: To examine the determinants of postsurgery length of stay (LOS) and inpatient mortality in the United States (California and Massachusetts) and Canada (Manitoba and Quebec). DATA SOURCES/STUDY SETTING: Patient discharge abstracts from the Agency for Health Care Policy and Research Nationwide Inpatient Sample and from provincial health ministries. STUDY DESIGN: Descriptive statistics by state or province, pooled competing risks hazards models (which control for censoring of LOS and inpatient mortality data), and instrumental variables (which control for confounding in observational data) were used to analyze the effect of wait time for hip fracture surgery on postsurgery outcomes. DATA EXTRACTIONS: Data were extracted for patients admitted to an acute care hospital with a primary diagnosis of hip fracture who received hip fracture surgery, were admitted from home or the emergency room, were age 45 or older, stayed in the hospital 365 days or less, and were not trauma patients. PRINCIPAL FINDINGS: The descriptive data indicate that wait times for surgery are longer in the two Canadian provinces than in the two U.S. states. Canadians also have longer postsurgery LOS and higher inpatient mortality. Yet the competing risks hazards model indicates that the effect of wait time on postsurgery LOS is small in magnitude. Instrumental variables analysis reveals that wait time for surgery is not a significant predictor of postsurgery length of stay. The hazards model reveals significant differences in mortality across regions. However, both the regressions and the instrumental variables indicate that these differences are not attributable to wait time for surgery. CONCLUSIONS: Statistical models that account for censoring and confounding yield conclusions that differ from those implied by descriptive statistics in administrative data. Longer wait time for hip fracture surgery does not explain the difference in postsurgery outcomes across countries.

Abstract: OBJECTIVE: To assess the effects of payment methods on the costs of care in medical group practices. DATA SOURCES: Eighty-six clinics providing services for a Blue Cross managed care program during 1995. The clinics were analyzed to determine the relationship between payment methods and cost of care. Cost and patient data were obtained from Blue Cross records, and medical group practice clinic data were obtained by a survey of those organizations. STUDY DESIGN: The effects of clinic and physician payment methods on per member per year (PMPY) adjusted patient costs are evaluated using a two-stage regression model. Patient costs are adjusted for differences in payment schedules; patient age, gender, and ACG; clinic organizational variables are included as explanatory variables. DATA COLLECTION: Patient cost data were extracted from Blue Cross claims files, and patient and physician data from their enrollee and provider data banks. Medical group practice data were obtained by a mailed survey with telephone follow-up. PRINCIPAL FINDINGS: Capitation payment is correlated with lower patient care costs. When combined with fee-for-service with withhold provisions, this effect is smaller indicating that these two clinic payment methods are not interchangeable. Clinics with more physician compensation based on measures of resource use or based on some share of the net revenue of the clinic have lower patient care costs than those with more compensation related to productivity or based on salary. Salary compensation is strongly associated with higher costs. The use of physician profiles and clinical guidelines is associated with lower costs, but referral management systems have no such effect. The lower cost clinics are the smaller, multispecialty clinics. CONCLUSIONS: This study indicates that payment methods at both the medical group practice and physician levels influence the cost of care. However, the methods by which that influence is manifest is not clear. Although the organizational structure of clinics and their use of managed care programs appear to play a role, this influence is less than expected.

Abstract: OBJECTIVE: To determine whether state-level parity legislation has led to an increase in utilization of mental health services. DATA SOURCES: Healthcare For Communities (HCC), a multi-site nationally representative study sponsored by the Robert Wood Johnson Foundation that tracks health care system changes for mental health and substance abuse treatment. Information on state-level parity legislation was provided by state offices of the National Alliance for the Mentally Ill (NAMI); local and state market data come from the Area Resource File; information on other health mandates from Blue Cross/Blue Shield. STUDY DESIGN: Two-stage regressions are used to estimate the effect of state parity legislation on use of any mental health services, use of specialty mental health services, and number of specialty visits in the past year. In the first stage, we predicted the probability that a state decides to pass parity legislation as a function of state health care market indicators and previous legislative activity. The fitted probability is used in the second stage to determine the effect of this legislation on access and utilization. PRINCIPAL FINDINGS: State parity legislation is not associated with a significant increase in any of our measures of mental health services utilization. These results are robust to various specifications of the models. CONCLUSIONS: Those states that are able to pass parity legislation do not experience significant increases in the utilization of mental health services. This may be due in part to a loss of coverage for those people most at risk for mental health disorders. The results could be very different, however, if strong federal legislation were passed.

Abstract: OBJECTIVE: To examine the relationship of in-hospital and 30-day mortality rates and the association between in-hospital mortality and hospital discharge practices. DATA SOURCES/STUDY SETTING: A secondary analysis of data for 13,834 patients with congestive heart failure who were admitted to 30 hospitals in northeast Ohio in 1992-1994. DESIGN: A retrospective cohort study was conducted. DATA COLLECTION: Demographic and clinical data were collected from patients' medical records and were used to develop multivariable models that estimated the risk of in-hospital and 30-day (post-admission) mortality. Standardized mortality ratios (SMRs) for in-hospital and 30-day mortality were determined by dividing observed death rates by predicted death rates. PRINCIPAL FINDINGS: In-hospital SMRs ranged from 0.54 to 1.42, and six hospitals were classified as statistical outliers (p <.05); 30-day SMRs ranged from 0.63 to 1.73, and seven hospitals were outliers. Although the correlation between in-hospital SMRs and 30-day SMRs was substantial (R = 0.78, p < .001), outlier status changed for seven of the 30 hospitals. Nonetheless, changes in outlier status reflected relatively small differences between in-hospital and 30-day SMRs. Rates of discharge to nursing homes or other inpatient facilities varied from 5.4 percent to 34.2 percent across hospitals. However, relationships between discharge rates to such facilities and in-hospital SMRs (R = 0.08; p = .65) and early post-discharge mortality rates (R = 0.23; p = .21) were not significant. CONCLUSIONS: SMRs based on in-hospital and 30-day mortality were relatively similar, although classification of hospitals as statistical outliers often differed. However, there was no evidence that in-hospital SMRs were biased by differences in post-discharge mortality or discharge practices.

Abstract: OBJECTIVE: To investigate the effect of price on the health insurance decisions of Medicare-eligible retirees in a managed competition setting. DATA SOURCE: The study is based on four years of administrative data from the University of California (UC) Retiree Health Benefits Program, which closely resembles the managed competition model upon which several leading Medicare reform proposals are based. STUDY DESIGN: A change in UC's premium contribution policy between 1993 and 1994 created a unique natural experiment for investigating the effect of price on retirees' health insurance decisions. This study consists of two related analyses. First, I estimate the effect of changes in out-of-pocket premiums between 1993 and 1994 on the decision to switch plans during open enrollment. Second, using data from 1993 to 1996, I examine the extent to which rising premiums for fee-for-service Medigap coverage increased HMO enrollment among Medicare-eligible UC retirees. PRINCIPLE FINDINGS: Price is a significant factor affecting the health plan decisions of Medicare-eligible UC retirees. However, these retirees are substantially less price sensitive than active UC employees and the non-elderly in other similar programs. This result is likely attributable to higher nonpecuniary switching costs facing older individuals. CONCLUSIONS: Although it is not clear exactly how price sensitive enrollees must be in order to generate price competition among health plans, the behavioral differences between retirees and active employees suggest that caution should be taken in extrapolating from research on the non-elderly to the Medicare program.

Abstract: OBJECTIVE: To assess the impact of HMO market structure on the formation of physician-hospital strategic alliances from 1993 through 1995 The two trends, managed care and physician-hospital integration have been prominent in reshaping insurance and provider markets over the past decade STUDY DESIGN: Pooled cross-sectional data from the InterStudy HMO Census and the Annual Survey conducted by the American Hospital Association (AHA) between 1993 and the end of 1995 to examine the effects of HMO penetration and HMO numbers in a market on the formation of hospital-sponsored alliances with physicians Because prior research has found nonlinear effects of HMOs on a variety of dependent variables, we operationalized HMO market structure two ways: using a Taylor series expansion and cross-classifying quartile distributions of HMO penetration and numbers into 16 dummy indicators Alliance formation was operationalized using the presence of any alliance model (IPA, PHO, MSO, and foundation) and the sum of the four models present in the hospital Because managed care and physician-hospital integration are endogenous (eg, some hospitals also sponsor HMOs), we used an instrumental variables approach to model the determinants of HMO penetration and HMO numbers These instruments were then used with other predictors of alliance formation: physician supply characteristics, the extent of hospital competition, hospital-level descriptors, population size and demographic characteristics, and indicators for each year All equations were estimated at the MSA level using mixed linear models and first-difference models PRINCIPAL FINDINGS: Contrary to conventional wisdom, alliance formation is shaped by the number of HMOs in the market rather than by HMO penetration This confirms a growing perception that hospital-sponsored alliances with physicians are contracting vehicles for managed care: the greater the number of HMOs to contract with, the greater the development of alliances The models also show that alliance formation is low in markets where a small number of HMOs have deeply penetrated the market First-difference models further show that alliance formation is linked to HMO consolidation (drop in the number of HMOs in a market) and hospital downsizing Alliance formation is not linked to changes in hospital costs, profitability, or market competition with other hospitals CONCLUSIONS: Hospitals appear to form alliances with physicians for several reasons Alliances serve to contract with the growing number of HMOs, to pose a countervailing bargaining force of providers in the face of HMO consolidation, and to accompany hospital downsizing and restructuring efforts IMPLICATIONS FOR POLICY, DELIVERY, OR PRACTICE: Physician-hospital integration is often mentioned as a provider response to increasing cost-containment pressures due to rising managed care penetration Our findings do not support this view Alliances appear to serve the hospital's interest in bargaining with managed care plans on a more even basis

Abstract: OBJECTIVE: To determine the effect of postpartum length of stay on newborn readmission. DATA SOURCES: Secondary data set consisting of newborns born in Washington state in 1989 and 1990. The data set contains information about the characteristics of the newborn and its parents, physician, hospital, and insurance status. STUDY DESIGN: Analysis of the effect of length of stay on the probability of newborn readmission using hour of birth and method of delivery as instrumental variables (IVs) to account for unobserved heterogeneity. Of approximately 150,000 newborns born in Washington in 1989 and 1990, 108,551 (72 percent) were included in our analysis. PRINCIPAL FINDINGS: Newborns with different lengths of stay differ in unmeasured characteristics, biasing estimates based on standard statistical methods. The results of our analyses show that a 12-hour increase in length of stay is associated with a reduction in the newborn readmission rate of 0.6 percentage points. This is twice as large as the estimate obtained using standard statistical (non-IV) methods. CONCLUSION: An increase in the length of postpartum hospital stays may result in a decline in newborn readmissions. The magnitude of this decline in readmissions may be larger than previously thought.

Abstract: OBJECTIVE: To describe the long-term productivity costs of occupational assaults. DATA SOURCES/STUDY SETTING: All incidents of physical assaults that resulted in indemnity payments, identified from the Minnesota Department of Labor and Industry (DLI) Workers' Compensation system in 1992. Medical expenditures were obtained from insurers, and data on lost wages, legal fees, and permanency ratings were collected from DLI records. Insurance administrative expenses were estimated. Lost fringe benefits and household production losses were imputed. STUDY DESIGN: The human capital approach was used to describe the long-term costs of occupational assaults. Economic software was used to apply a modified version of Rice, MacKenzie, and Associates' (1989) model for estimating the present value of past losses from 1992 through 1995 for all cases, and the future losses for cases open in 1996. PRINCIPAL FINDINGS: The total costs for 344 nonfatal work-related assaults were estimated at $5,885,448 (1996 dollars). Calculation of injury incidence and average costs per case and per employee identified populations with an elevated risk of assault. An analysis by industry revealed an elevated risk for workers employed in justice and safety (incidence: 198/100,000; $19,251 per case; $38 per employee), social service (incidence: 127/100,000; $24,210 per case; $31 per employee), and health care (incidence: 76/100,000; $13,197 per case; $10 per employee). CONCLUSIONS: Identified subgroups warrant attention for risk factor identification and prevention efforts. Cost estimates can serve as the basis for business calculations on the potential value of risk management interventions.

Abstract: OBJECTIVE: To answer two related questions: (1) Do managed care organizations (MCOs) in New York State (NYS) consider quality when they choose cardiac surgeons? (2) Do they use information about risk-adjusted mortality rates (RAMR) provided in the New York State Cardiac Surgery Reports? DATA SOURCES: (1) Telephone interviews with and contracting data from the majority of MCOs licensed in NYS; (2) RAMR, quality outlier designation, and procedure volume for all cardiac surgeons, as reported in the Cardiac Surgery Reports. STUDY DESIGN: Interview data were analyzed in conjunction with patterns revealed by contracting data. Null hypotheses that MCOs' contracting choices were random with respect to the information published in the Cardiac Surgery Reports were tested. PRINCIPAL FINDINGS: Sixty percent of MCOs ranked the quality of surgeons as most important in their contracting considerations. Although 64 percent of MCOs indicated some knowledge of the NYS Cardiac Surgery Reports, only 20 percent indicated that the reports were a major factor in their contracting decision. Analyses of actual contracting patterns show that in aggregate, the hypothesis of random choice could be rejected with respect to high-quality outlier status and high procedure volume but not for RAMR or poor-quality outlier status. The panel composition of the majority of MCOs (80.2 percent) was within two standard deviations of the expected mean under the null hypothesis. CONCLUSIONS: Despite a professed preference for high-quality surgeons, the use of publicly available quality reports by MCOs is currently low, and contracting practices for the majority of MCOs do not indicate a systematic selection either for or against surgeons based on their reported mortality scores. This study suggests that policy initiatives to increase the effective use of report cards should be encouraged.

Abstract: OBJECTIVE: To measure the impact of patient waiting time on the probability of graft failure following liver transplantation. DATA SOURCES: Observations on all patients receiving transplants between 1995 and 1997 collected by the United Network for Organ Sharing. STUDY DESIGN: This study uses a two-stage probit algorithm. The first stage is an ordinary least squares regression of controls and indicators of patient blood type (the instrument set) on waiting time. The second uses the predicted value from the first regression in a probit analysis where the dependent variable is graft failure. PRINCIPAL FINDINGS: Waiting time has a statistically and clinically significant effect on the probability of graft failure following transplantation. Waiting 50 additional days for a transplant increases the probability of graft failure at one year by between one and two percentage points. CONCLUSIONS: Policies to reduce waiting times will yield clinical benefits beyond decreasing the length of time patients spend in poor health.

Abstract: OBJECTIVES: To present results from an outcome evaluation of the Henry J. Kaiser Family Foundation's Community Health Promotion Grants Program (CHPGP) in the West, which represented a major community-based initiative designed to promote improved health by changing community norms, environmental conditions, and individual behavior in 11 western communities. METHODS: The evaluation design: 14 randomly assigned intervention and control communities, 4 intervention communities selected on special merit, and 4 matched controls. Data for the outcome evaluation were obtained from surveys, administered every two years at three points in time, of community leaders and representative adults and adolescents, and from specially designed surveys of grocery stores. Outcomes for each of the 11 intervention communities were compared with outcomes in control communities. RESULTS: With the exception of two intervention communities-a largely Hispanic community and a Native American reservation-we found little evidence of positive changes in the outcomes targeted by the 11 intervention communities. The programs that demonstrated positive outcomes targeted dietary behavior and adolescent substance abuse. CONCLUSIONS: Improvement of health through community-based interventions remains a critical public health challenge. The CHPGP, like other prominent community-based initiatives, generally failed to produce measurable changes in the targeted health outcomes. Efforts should focus on developing theories and methods that can improve the design and evaluation of community-based interventions.

Abstract: OBJECTIVES: To evaluate the costs of implementing a church-based, telephone-counseling program for increasing mammography use, and to identify the components of costs and the likely cost-effectiveness in hypothetical communities with varying characteristics. DATA SOURCES/STUDY SETTING: An ethnically and socioeconomically diverse sample of 1,443 women recruited from 45 churches participating in the Los Angeles Mammography Promotion (LAMP) program were followed from 1995 to 1997. STUDY DESIGN: Churches were stratified into blocks and randomized into three intervention arms-telephone counseling, mail counseling, and control. We surveyed participants before and after the intervention to collect data on mammography use and demographic characteristics. DATA COLLECTION/EXTRACTION METHODS: We used call records, activity reports, and interviews to collect data on the time and materials needed to organize and carry out the intervention. We constructed a standard model of costs and cost-effectiveness based on these data and the Year One results of the LAMP program. PRINCIPAL FINDINGS: The cost in materials and overhead to the church site was $10.89 per participant and $188 per additional screening. However, when the estimated cost for church volunteers' time was included, the cost of the intervention increased substantially. CONCLUSIONS: A church-based program to promote the use of mammography would be feasible for many churches with the use of volunteer labor and resources.

Abstract: OBJECTIVE: To inform the debate about managed care by examining how different types of private insurance-indemnity insurance, PPOs, open model HMOs, and closed model HMOs-affect the use of health services and consumer assessments of care. DATA SOURCES/DATA COLLECTION: The 1996-1997 Community Tracking Study Household Survey, a nationally representative telephone survey of households, and the Community Tracking Study Insurance Followback Survey, a supplement to the Household Survey, which asks insurance organizations to match household respondents to specific insurance products. The analysis sample includes 27,257 nonelderly individuals covered by private insurance. STUDY DESIGN: Based on insurer reports, individuals are grouped into one of the four insurance product types. Measures of service use include ambulatory visits, preventive care use, hospital use, surgeries, specialist use, and whether there is a usual source of care. Consumer assessments of care include unmet or delayed care needs, satisfaction with health care, ratings of the last physician visit, and trust in physicians. Estimates are adjusted to control for differences in individual characteristics and location. PRINCIPAL FINDINGS: As one moves from indemnity insurance to PPOs to open model HMOs to closed model HMOs, use of primary care increases modestly but use of specialists is reduced. Few differences are observed in other areas of service use, such as preventive care, hospital use, and surgeries. The likelihood of having unmet or delayed care does not vary by insurance type, but the reasons that underlie such access problems do vary: enrollees in more managed products are less likely to cite financial barriers to care but are more likely to perceive problems in provider access, convenience, and organizational factors. Consumer assessments of care-including satisfaction with care, ratings of the last physician visit, and trust in physicians-are generally lower under more managed products, particularly closed model HMOs. CONCLUSIONS: The type of insurance that people have-not just whether it is managed care but the type of managed care-affects their use of services and their assessments of the care they receive. Consumers and policymakers should be reminded that managed care encompasses a variety of types of insurance products that have different effects and may require different policy responses.

Abstract: OBJECTIVE: To describe the efficiency of HMOs and to test the robustness of these findings across alternative models of efficiency. This study examines whether these models, when constructed in parallel to use the same information, provide researchers with the same insights and identify the same trends. DATA SOURCES: A data set containing 585 HMOs operating from 1985 through 1994. Variables include enrollment, utilization, and financial information compiled primarily from Health Care Investment Analysts, InterStudy HMO Census, and Group Health Association of America. STUDY DESIGN: We compute three estimates of efficiency for each HMO and compare the results in terms of individual performance and industry-wide trends. The estimates are then regressed against measures of case mix, quality, and other factors that may be related to the model estimates. PRINCIPAL FINDINGS: The three models identify similar trends for the HMO industry as a whole; however, they assess the relative technical efficiency of individual firms differently. Thus, these techniques are limited for either benchmarking or setting rates because the firms identified as efficient may be a consequence of model selection rather than actual performance. CONCLUSIONS: The estimation technique to evaluate efficient firms can affect the findings themselves. The implications are relevant not only for HMOs, but for efficiency analyses in general. Concurrence among techniques is no guarantee of accuracy, but it is reassuring; conversely, radically distinct inferences across models can be a warning to temper research conclusions.

Abstract: OBJECTIVE: To examine ways in which the management and organization of medical care is changing in response to the shifting incentives created by managed care. DATA SOURCES: Site visits conducted in 12 randomly selected communities in 1996/ 1997. STUDY DESIGN: Approximately 35-60 interviews were conducted per site with key informants in healthcare and community organizations; about half were with providers. DATA COLLECTION: A standardized interview protocol was implemented across all sites, enabling cross-site comparisons. Multiple respondents were interviewed on each issue. PRINCIPAL FINDINGS: A great deal of experimentation and apparent duplication exist in efforts to develop programs to influence physician practice patterns. Responsibility for managing care is being contested by health plans, medical groups and hospitals, as each seeks to accrue the savings that can result from the more efficient delivery of care. To manage the financial and clinical risk, providers are aggressively consolidating and reorganizing. Most significant was the rapid formation of intermediary organizations, such as independent practice arrangements (IPAs), physician-hospital organizations (PHOs), or management services organizations (MSOs), for contracting with managed care organizations. CONCLUSIONS: Managed care appears to have only a modest effect on how healthcare organizations deliver medical care, despite the profound effect that managed care has on how providers are organized. Rather than improving the efficiency of healthcare organizations, provider efforts to build large systems and become indispensable to health plans are exacerbating problems of excess capacity. It is not clear if new organizational arrangements will help providers manage the changing incentives they face, or if their intent is to blunt the effects of the incentives by forming larger organizations to improve their bargaining power and resist change.