Live long, live well: quantifying the health of heterogeneous populations.

Various health, quality, utility and disability adjusted life years or life expectancy (HALY, QALY, DALY; HALE, QALE, DALE) measures have become gold standards for defining outcomes in technology evaluation, population health monitoring, and other evaluative efforts. As such, the analytical framework within which these measures are used for descriptive and evaluative purposes should be theoretically consistent and statistically rigorous. For instance, widely accepted definitions of cost-effectiveness ratios and other technology evaluation criteria that are based on expectations of the respective cost and outcome measures must, as such, be defined in terms of expected HALYs or QALYs. Similarly, measures like HALEs or QALEs used for population health monitoring are typically concerned with population expectations of such measures (or their corresponding totals). This paper demonstrates that estimation of such expectations necessitates consideration of the population variation in, and covariation between, quality and longevity. From the perspective of several different environments characterizing such heterogeneity, quantification or estimation of measures like QALYs is reconsidered. An empirical example of the central issues is provided by means of an analysis of the years of healthy life (YHL) measure drawn from the US National Health Interview Survey.

Estimating log models: to transform or not to transform?

Health economists often use log models to deal with skewed outcomes, such as health utilization or health expenditures. The literature provides a number of alternative estimation approaches for log models, including ordinary least-squares on ln(y) and generalized linear models. This study examines how well the alternative estimators behave econometrically in terms of bias and precision when the data are skewed or have other common data problems (heteroscedasticity, heavy tails, etc.). No single alternative is best under all conditions examined. The paper provides a straightforward algorithm for choosing among the alternative estimators. Even if the estimators considered are consistent, there can be major losses in precision from selecting a less appropriate estimator.

It'll only hurt a second? Microeconomic determinants of who gets flu shots.

Appreciating how the propensity to be immunized against the flu depends on individual characteristics and environments is essential for policies regarding influenza control to be formulated sensibly. To this point, the literature has offered little documentation on the determinants of influenza immunization. Beyond epidemiology, there are important economic issues that must be addressed to understand this form of preventive care. One concerns the relationship between labour supply and immunization propensity: While it is relatively costly (in terms of time costs) for workers to obtain immunizations, workers also have relatively more to lose from being ill with the flu. Another concern not generally appreciated is the extent to which individuals' perceived risks of infection may affect their propensities to be immunized. The paper also attempts to shed light on these issues. The analysis uses data from the 1991 National Health Interview Survey. Immunization propensity displays expected patterns by age and health status, while the results with respect to race, household structure, income and insurance are somewhat more surprising and/or novel. The estimated labour supply and perceived risk effects suggest that some aspects of the economics of preventive care generally not considered in empirical work are important and merit further consideration.

Drinking, problem drinking, and productivity.

This chapter surveys and critiques the recent economic literature dealing with the relationships between labor market productivity and alcohol use and misuse. The focus here is twofold. First is to present and discuss the relevant conceptual issues that must be appreciated in assessing such relationships. Second is to summarize and assess the empirical findings that have been offered in the literature.

Much ado about two: reconsidering retransformation and the two-part model in health econometrics.

In health economics applications involving outcomes (y) and covariates (x), it is often the case that the central inferential problems of interest involve E[y/x] and its associated partial effects or elasticities. Many such outcomes have two fundamental statistical properties: y > or = 0; and the outcome y = 0 is observed with sufficient frequency that the zeros cannot be ignored econometrically. This paper (1) describes circumstances where the standard two-part model with homoskedastic retransformation will fail to provide consistent inferences about important policy parameters; and (2) demonstrates some alternative approaches that are likely to prove helpful in applications.

Net health benefits: a new framework for the analysis of uncertainty in cost-effectiveness analysis.

In recent years, considerable attention has been devoted to the development of statistical methods for the analysis of uncertainty in cost-effectiveness (CE) analysis, with a focus on situations in which the analyst has patient-level data on the costs and health effects of alternative interventions. To date, discussions have focused almost exclusively on addressing the practical challenges involved in estimating confidence intervals for CE ratios. However, the general approach of using confidence intervals to convey information about uncertainty around CE ratio estimates suffers from theoretical limitations that render it inappropriate in many situations. The authors present an alternative framework for analyzing uncertainty in the economic evaluation of health interventions (the "net health benefits" approach) that is more broadly applicable and that avoids some problems of prior methods. This approach offers several practical and theoretical advantages over the analysis of CE ratios, is straightforward to apply, and highlights some important principles in the theoretical underpinnings of CE analysis.

Linking the health utilities index to National Medical Expenditure Survey data.

Measures of health-related quality of life (HR-QOL) are becoming increasingly important in assessing the effects of chronic illness and healthcare interventions designed to treat them. Obtaining measures of HR-QOL for a nationally representative sample of individuals would enhance understanding of health status in the US, and promote further study of the economic causes and effects of health status. This study reports on our efforts to link a prominent HR-QOL scale, the Health Utilities Index Mark I (HUI), to the National Medical Expenditure Survey (NMES). Six distinct algorithms were constructed for linking the HUI to NMES. These alternative linkage algorithms yielded HUI measures that were highly intercorrelated (p = 93 to 99%). Multivariate regression analyses performed to predict variations in HR-QOL revealed that the HUI exhibited good predictive validity--the HUI demonstrated lower quality of life for a variety of chronic illnesses, and wealthier individuals and better educated individuals had a higher quality of life. In contrast to some previous HR-QOL research, the present analysis demonstrates that: (i) cancer is negatively and significantly related to quality of life; and (ii) smoking is negatively and significantly related to quality of life. Overall, the results suggest that the HUI linkages to NMES provide reliable and valid measures of quality of life. As such, items from the NMES can be grouped and linked in such a way as to obtain health state utility values. These values should be of use to those who wish to understand the global health of the US population for policy-making efforts.

Women and work: tipplers and teetotalers.

We seek to understand better the puzzling finding that, for women, alcoholism appears to be positively associated with the probability of being employed. Using the 1988 Alcohol Survey of the National Health Interview Survey, we find that this association holds for white women only. For white women, alcoholism and early drinking are associated with higher educational attainment, a smaller family size and a lower probability of being married. In turn, these human capital indicators are associated with greater labour supply, thus helping to explain the curious positive relationship between alcoholism and employment for women. An advance in this paper over our previous work is to examine life-time abstention from alcohol and its association with employment and human capital variables. We find that lifetime abstention is associated with lower employment, unemployment and education and greater propensity to be married for both white and non-white women.

Alcohol screening and brief intervention: where research meets practice.

Numerous studies have consistently shown that quick screening instruments can identify people whose drinking is likely to present health risks and that low-cost, brief interventions are effective in reducing drinking among many such at-risk drinkers. This article describes the results of a one-year policy analysis that explored how alcohol screening and brief intervention (SBI) can be moved to widespread clinical applications in the United States. It introduces the concept of risky drinking and considers the potential of this new technology to reduce it. The research evidence behind this approach is reviewed, and a description of current programs in this and other countries beginning to apply SBI is provided. Economic issues attendant to applications are identified and discussed. The potential for applications in health care is analyzed and summary conclusions from market research are set forth. Recommendations are offered for immediate action.

Employment, unemployment, and problem drinking.

The misuse of alcohol is estimated to result in enormous economic costs, composed largely of reduced labor market productivity. However, there has been debate on this issue. The purpose of this paper is to help to resolve this debate by presenting sound structural estimates of the relationship between measures of problem drinking and of employment and unemployment. The analysis is based on the 1988 Alcohol Supplement of the National Health Interview Survey. We find that for both men and women, problem drinking results in reduced employment and increased unemployment.

Alcoholism and income: the role of indirect effects.

The potential pathways by which alcoholism might affect income are examined using data on males from the New Haven site of the Epidemiologic Catchment Area (ECA) data set to illustrate important indirect, and direct, effects of alcoholism on income. The detrimental indirect effects occur through reduction of educational attainment and increased probability of divorce. Because both educational attainment and marriage positively influence income, the adverse impact of alcoholism on both of these variables translates into lower income for alcoholics. The implications should lead researchers to distinguish more carefully between the direct and indirect costs of this condition. Policy analysts also should explore the financial impact of early onset when they look at overall costs. Because this has often been underestimated, fewer resources are devoted to prevention and treatment of alcoholism than a considered awareness of the problem would indicate is warranted.

Air pollution, cigarette smoking, and the production of respiratory health.

Previous studies of the determinants of respiratory health have treated both smoking and air pollution as being exogenous. Using an instrumental variables approach, we estimate a simple production technology in which smoking is treated as being endogenously determined. Doing so, we find, increases the predicted absolute effects of smoking on respiratory health; relative to air pollution, smoking becomes a more important determinant when it is treated as an endogenous variable.

Life-cycle effects of alcoholism on education, earnings, and occupation.

In this article we examine the life-cycle effects of alcoholism on human capital formation and labor market outcomes. We attempt empirically to unravel some of the complicated structural relationships among family background, educational attainment, labor force participation, earnings, and current as well as past alcoholism. Access to a new data set allows us to study some of these relationships. The Epidemiological Catchment Area data set contains not only the standard socioeconomic and demographic variables but also information on family background and age of onset of symptoms and current alcoholism. Thus, when estimating models of occupational choice and earnings, we can assess both the indirect and the direct effects of alcoholism. We find that early onset of alcoholism significantly retards educational attainment.