Alcoholism treatment offset effects: an insurance perspective.

This study investigates whether alcoholism treatment costs are offset by reductions in other medical treatment costs by comparing people treated for alcoholism with a matched comparison group. The alcoholism treatment group is defined by diagnoses of alcohol dependence, abuse, or psychoses from health insurance claims field between January 1980 and June 1987. A comparison sample was matched on age, gender, and insurance coverage. In this primarily methodological study, expected costs for nonalcoholism treatments were calculated from standardized regressions. Offset effects were measured from the insurer's perspective through differences in expected total nonalcoholism treatment costs in the periods preceding and following alcoholism treatment. Members of the alcoholism treatment group were more likely than the comparison group to be hospitalized and to need other (nonalcoholism) medical treatment, thus incurring higher total costs. Offset effects emerged for patients with alcohol abuse and without mental psychosis comorbidities.

Short term drug abuse treatment costs and utilization: a multi-employer analysis.

OBJECTIVES: This report investigates three aspects of drug abuse treatment costs, with special emphasis on systematic differences among employers: (1) predictors of drug abuse treatment costs; (2) differentials in drug abuse treatment costs across employers; and (3) differential impacts of patient and employer characteristics on drug abuse treatment costs. METHODS: The study used multiple regression analysis of behavioral cost functions. It decomposed cost differences into employer and variable effects using an algebraic method that accounted for differences in cost functions and in population characteristics. An insurance claims database was used from 10 large self-insured employers for a 3-year period starting January 1989. RESULTS: Marginal inpatient costs generally exceeded average costs, leading to slightly increasing costs per day as length of stay increased. Marginal outpatient costs were generally about the same as average costs, implying that outpatient drug treatment maintained constant unit costs as utilization increased. Decomposition of cost differences among employers suggested that observed differences among employers and/or their carriers (who administer the benefits for the self-insured employers) and providers appeared to be at least as important as differences among the characteristics or the utilization of the people that they cover. CONCLUSIONS: National health policies aimed at reducing costs are likely to have differing impacts on different employers. Employers with high costs relative to the characteristics of their covered population may be able to achieve significant cost savings. Employers serving populations with greater risk factors may find it difficult to cut costs further.

A discrete choice model of drug abuse treatment location.

OBJECTIVE: To identify short-term drug abuse treatment location risk factors for ten large, self-insured firms starting January 1, 1989 and ending December 31, 1991. DATA SOURCES/STUDY SETTING: Study population selected from a large database of health insurance claims for all treatment events starting January 1, 1989 and ending December 31, 1991. STUDY DESIGN: A nested binomial logit method is used to estimate firm-specific patterns of treatment location. The differences in treatment location patterns among firms are then decomposed into firm effects (holding explanatory variables constant among firms) and variable effects (holding firm-specific parameters constant). PRINCIPAL FINDINGS: Probability of inpatient drug treatment is directly related to the type of drug diagnosis. The most important factors are diagnoses of drug dependence (versus drug abuse) and/or a cocaine dependence. Firm-specific factors also make a substantive difference. Controlling for patient risk factors, firm-specific probabilities of inpatient treatment vary by as much as 87 percent. Controlling for practices of firms and their insurance carriers, differing patient risk profiles cause probabilities of inpatient treatment to vary by as much as 69 percent among firms. Use of the outpatient setting increased over the three-year period. CONCLUSIONS: There are two plausible explanations for the findings. First, people beginning treatment later in the three-year period had less severe conditions than earlier cases and therefore had less need of inpatient treatment. Second, drug abuse treatment experienced the same trend toward the increased use of outpatient care that characterized treatment for other illnesses in the 1980s and early 1990s.

Cost and usage impacts of treatment initiation: a comparison of alcoholism and drug abuse treatments.

An extensive literature on substance abuse and mental health treatments suggests that they often lead to decreased usage and/or spending on other medical treatments. We compare alcohol and drug abuse treatment costs with a model that decomposes total treatment costs into amount of treatment (outpatient visits or inpatient days) and costs per treatment. The analysis compares alcohol and drug abuse treatment costs regarding: (1) the incremental costs attributable to changed short-term substance abuse and nonsubstance abuse treatments; (2) the impacts of current substance abuse treatments on short-term nonsubstance abuse, long-term substance abuse, and long-term nonsubstance abuse treatments; and (3) the difference in inpatient and outpatient impacts. Our findings indicate that alcoholism and drug abuse treatment initiation have similar impacts on coincident and subsequent utilization and costs. For both treatments, the largest portions of the cost impacts occur for inpatient treatments, and for treatments that occur within 6 months of the initiation. The similarity of results suggests that it may often be reasonable to infer utilization and cost impacts for one type of care from studies that examine the other.

Cost analyses of prostate cancer screening: frameworks for discussion. Investigators of the American Cancer Society-National Prostate Cancer Detection Project.

Our recent cost analysis of prostate cancer early detection evaluated the economic performance of various prostate specific antigen (PSA) screening approaches, detected marginal cost variations with time and used a benefit-cost calculation as a framework for further discussion. Receiver operator characteristic analysis initially suggested an optimal test performance for PSA of 2 to 3 ng./ml. when used alone and at approximately 3 ng./ml. in combination with digital rectal examination. However, lower PSA decision levels require cost justifications. Marginal cost analysis demonstrated markedly decreased use of digital rectal examination by year 3 due to significantly lower sensitivity for incident cancer. The benefit-cost equation acknowledges that many parameters of cost and probability are not definitive to date yet illustrated major points for discussion. The cost parameters most sensitive to incremental change in decreasing order are the specificity of the screening test, benefits obtained from early therapy and prevalence of the disease. Discussions about improving the likelihood of overall benefit for the United States population should focus on these parameters, as well as social and ethical implications. If we assume minimized future expenditures for terminal cancer care via decreases in therapy choices or coverage, no economic benefit for screening exists. If we also assume that potential costs to society are not roughly approximated by any benefits, we may engender inappropriate attempts at cost reduction by effectively discouraging screening in the highest risk groups. Perhaps the greatest immediate cost control issue is the marked increase in prostate cancer detection in the oldest age groups who have the least likelihood of mortality or morbidity benefits. Current cost savings may be possible with improved public health education about the appropriateness of early detection in the oldest age groups or those with significant preexisting medical conditions.

Costs and benefits of prostate cancer screening. Investigators of the American Cancer Society--National Prostate Cancer Detection Project.

Optimal combinations of digital rectal examination (DRE), transrectal ultrasound (TRUS) and prostate specific antigen (PSA) may better detect patients at high risks, as well as those in whom continued screening may not be cost effective. Our recent cost analysis of prostate cancer early detection used current data from three consecutive years of the American Cancer Society's National Prostate Cancer Detection Project. Marginal cost analysis showed marked increased costs for the DRE by year three due to significantly reduced sensitivity for incident cancers. The benefit-cost equation acknowledges that many parameters of both cost and probability are not definitive at this time, yet illustrated major points for discussion. The cost parameters most sensitive to incremental change in decreasing order are: the specificity of the screening test > benefits obtained from early therapy > prevalence of the disease. Benefit-cost calculations demonstrated that DRE, when performed by highly skilled examiners, had the lowest cost. However, DRE became one of the most costly detection scenarios when a minor decrease in DRE performance was assumed for more general examiners. If slightly more specific PSA usage (or assay) is developed, the higher prevalence of clinically detectable prostate cancer could make screening less costly than breast cancer screening. If we assume minimized future expenditures for terminal cancer care via reductions in therapy choices or coverage, no economic benefit for screening exists. If we also assume that potential costs to society are not roughly approximated by any benefits, we may engender inappropriate attempts at cost reduction by effectively discouraging screening in the highest risk groups.(ABSTRACT TRUNCATED AT 250 WORDS)

The benefit and cost of prostate cancer early detection. The Investigators of the American Cancer Society-National Prostate Cancer Detection Project.

Cost-effectiveness calculations of prostate cancer early detection have not been possible due to the lack of any data demonstrating reduction in mortality from any test or procedure. Prior analyses focused only on cost assessments without consideration of any possible benefits. We used current data from three consecutive years of the American Cancer Society-National Prostate Cancer Detection Project to assess different economic perspectives of test performance, marginal costs, and benefit-cost analysis. The marginal cost, or cost per cancer, of digital rectal examination (DRE) markedly increased by the third year relative to several proposed prostate-specific antigen (PSA) scenarios. Sensitivity analysis for average cost showed that at 4 ng/ml, pricing PSA below $30 would be the most potent factor in potentially lowering costs. Analysis of receiver operator characteristic curves suggested that optimal performance for PSA may be at 3 ng/ml when combined with DRE or between 2 to 3 ng/ml when used alone. Benefit-cost calculations demonstrated that DRE when performed by highly skilled examiners had the lowest cost. However, DRE became one of the most costly detection scenarios when a minor decrease in performance was assumed. Sensitivity analysis demonstrated that the three most determinant parameters of net benefit, in decreasing order, are: specificity, benefits from earlier therapy, and prevalence. If a slightly more specific PSA assay is developed, the higher prevalence of clinically detectable prostate cancer could also make screening less costly than breast cancer screening. Under the assumptions of these analyses, the combination of PSA and DRE appears to represent an ethical and economical detection choice for individual patients in consultation with their physicians. Additional research is needed to quantify the significance of differences between different screening strategies.

A discrete choice model of alcoholism treatment location.

In this study, a discrete choice model of alcoholism treatment location, with special emphasis on the roles of comorbidities is considered. Three specific questions are addressed: 1) what demographic and health factors have significant impacts on treatment location for both short- and long-term alcoholism and nonalcoholism treatments?; 2) how does the impact of alcohol dependence differ from the impact of alcohol abuse, on probabilities of short-term or long-term inpatient treatment?; and 3) what are the impacts of health comorbidities on probabilities of inpatient treatment in the short or long term? A binomial logit model is estimated for short- and long-term alcoholism treatment, as well as for short- and long-term nonalcoholism treatment (which occurs at the same time). The results indicate the importance of comorbidities in predicting treatment location. They also indicate a trend during the 1980s toward increased use of outpatient rather than inpatient treatment.

An analysis of short-term alcoholism treatment cost functions.

A number of alcohol treatment studies have documented variations in the average cost of treating alcoholics. However, these studies have provided little explanation for these variations. In this study, three major issues in the measurement of alcoholism treatment costs are investigated: 1) choice of treatment location, i.e., inpatient versus outpatient; 2) interaction of treatment locations in the estimation of costs; 3) impact of type of alcohol problem and comorbidities on treatment costs. The study includes an integrated framework that jointly estimates treatment location and treatment costs conditional on treatment location, concentrating on short-term alcoholism treatment and using insurance claims data to specify a 6-month period beginning with each individual's first treatment for alcoholism. The different treatment types subsumed in the categories alcohol abuse and alcohol dependence are also addressed. Results indicate that comorbidities are crucial in determining treatment location. Once treatment location is determined, however, their effects on treatment costs, while measurable, are statistically insignificant. Partial treatment effects, conditional on treatment location, differ substantially from full treatment effects, which are determined jointly with treatment location.