Alan L. Hillman

Avoiding bias in the conduct and reporting of cost-effectiveness research sponsored by pharmaceutical companies.

Because of the growing focus on containing health care costs, pharmaceutical companies are trying to demonstrate the cost effectiveness of their products relative to alternatives. In Europe and Aus...

Economic analysis of health care technology: A report on principles

Preamble Although economic outcomes research is an evolving field in health services research, there are correct and incorrect ways to conduct and report on economic outcomes studies. Research practices that help to minimize real or perceived bias will increase the quality and usefulness of such studies for those who sponsor, publish, and use them. Because of public concerns about the potential for bias in the design, analysis, and reporting of economic analyses of health care technology, we formed a task force to develop principles to enhance the credibility of these studies. The Task Force on Principles for Economic Analysis of Health Care Technology included participants from academia, the pharmaceutical industry, the public sector, and private research organizations. As health care resources become increasingly constrained, the information used to make resource allocation decisions must be as reliable, valid, and free of bias as possible. Getting it right at the level at which economic results are produced will help to protect consumers and will advance the health of the public. Bias stems from two broad categories: lack of appropriate independence for researchers and lack of consensus about methods. We focused heavily on the first of these categories for two major reasons. First, few have yet considered the unique issues of researcher independence in economic outcomes research [1, 2]. Second, other investigators have begun to consider and define proper methods for economic outcomes work (an area of considerable controversy) [3-5]. We also looked closely at the requirements for the reporting of economic analyses, which are intended to ensure methodologic transparency and accountability. The Need for Voluntary Guidelines Widespread use of economic analysis as part of the development of pharmaceutical, biotechnologic, and medical devices is relatively new. To date, many economic analyses of health care have focused on pharmaceutical agents, and many such studies have been funded by pharmaceutical companies. Results of these economic outcomes studies are used by medical technology firms to support the pricing and marketing of new interventions and to influence national health care systems and third-party payers in their development of coverage and payment decisions. Managed care organizations, hospitals, and government-subsidized health care programs rely on economic analysis of medical technology to help make formulary purchasing and utilization decisions. Physicians may use the results of these analyses to help guide treatment and prescription decisions. Health care economic outcomes projects are sponsored and conducted by pharmaceutical, biotechnologic, and medical device companies; government agencies; nonprofit foundations; academic investigators; and private research and consulting firms. The standards and methods used to evaluate the safety and efficacy of pharmaceutical products in randomized, controlled trials have evolved over 50 years of collaboration among researchers in private, public, and academic settings. Compared with those established to ensure the safety and efficacy of clinical trials, the principles and methods for the conduct of economic studies of health care technology are far less developed. Problems of conduct, reporting, and bias exist in all types of research [6-14]. In response, codes of conduct, such as those developed by the American Federation for Clinical Research and the Institute of Medicine, have been developed for many scientific disciplines [15-24]. These are good models on which to base principles of conduct for economic outcomes analysis. Although many published principles apply to economic studies, others should be modified and new ones should be developed to guide the conduct and reporting of economic outcomes analyses. Economic outcomes research requires unique guidelines for the following reasons: 1) As a field, it continues to evolve and is often misunderstood by end users; 2) peer review of it requires special expertise that often exceeds the capabilities of reviewers and scientific journals; 3) it often uses secondary data and requires that many assumptions be made [for example, attribution of a dollar cost to a unit of resource use]; 4) it offers unique methodologic choices, such as which types of costs to include (direct, indirect, intangible, induced), which perspective to apply (that of society, payer, provider, patient), which design to adopt (cost-identification, costbenefit, cost-effectiveness, costutility), from where to obtain costs [indemnity database, managed care or capitated database, hospital cost systems, Medicare, Medicaid], and whether to collect resource consumption data prospectively or retrospectively through various modeling techniques; and 5) economic studies play an increasingly important role in health care decision making because of increasing financial constraints throughout the health care industry. The financial and medical implications of decision making done on the basis of these studies, coupled with the lack of widely accepted guidelines about the conduct and reporting of economic analyses, undermine the credibility of this research. A major issue is that the primary source of funding for this research is often the primary financial beneficiary of positive study results. Unfortunately, even valid studies done under the best of circumstances may be suspect [25-29]. This has led at least one major journal to conclude that these analyses should be viewed much like editorials or review articles are viewed in terms of potential for conflicts of interest [30]. We developed the guidelines reported here after extensive consultation with experts from the public, private, and academic sectors. We recommend that researchers and sponsors adhere to these guidelines, and we suggest they state publicly within their manuscripts that they have done so. End users, including journal editors and readers, consumers, and social decision makers, may then feel more secure in accepting the results of the research, while recognizing that intensive critique of the research will always be necessary. Operations of the Task Force The Task Force on Principles for Economic Analysis of Health Care Technology was initiated and organized by faculty from the Leonard Davis Institute (LDI) Center for Health Policy of the University of Pennsylvania. However, because not all LDI faculty were involved, this paper does not represent an official LDI position statement. The Task Force was funded by a coalition of pharmaceutical companies (Appendix A). Funding was also requested (but not obtained) from various government and private foundations. All funding was provided in the form of unrestricted research grants or gifts to the University of Pennsylvania. Guidelines derive their credibility in part from the composition of the panel that creates them and the process by which they are developed. Candidates for participation in the Task Force from the private sector and the academic research community were identified by the frequency with which they were cited in the health economics literature, which was obtained using a MEDLINE search of literature related to economic analyses of medical technology published between 1983 and 1992. Approximately 15 members attended each meeting. Minutes were distributed after each meeting and approved by all members present. Members from the sponsoring pharmaceutical companies and the academic organizers rotated so that they numbered three and one, respectively, at the table for each formal Task Force meeting. Several professional and governmental organizations, including the Institute of Medicine, the Agency for Health Care Policy and Research, the Food and Drug Administration (FDA), the Health Care Financing Administration, the Centers for Disease Control and Prevention, and a managed care organization, were asked to suggest persons who might participate. In addition, other organizations with a stake in economic analyses were each asked to suggest a person who, because of his or her professional background, had extensive knowledge of or experience with economic analyses. An academic pharmacist, an academic researcher, a private researcher, an ethicist and patient advocate, and an attorney specializing in medical ethics rounded out the Task Force (Appendix B). Two members withdrew from participation. A small audience consisting of industry sponsors, academic organizers, staff, and a few other interested parties were invited to each meeting and allowed to comment. Four formal meetings of the Task Force were held in Philadelphia during 1993 and 1994, and the most substantial work was done by various subcommittees between these meetings. The three main subcommittees were titled Ethical Conduct, Responsibility and Control (the findings of these two subcommittees were later merged into one report), and Reporting Requirements for Economic Evaluations (Appendix C). The Task Force was assisted by a professional facilitator. Although Task Force members sought consensus wherever possible on the key issues, consensus was not forced and recommendations were issued only when substantial agreement existed among the members. Task Force members were not asked to formally represent any organization. Formal endorsement of the final document was not sought. Copies of this report will be distributed to all organizations that were asked to suggest a participant and to other parties who responded to announcements placed in The New England Journal of Medicine, Scrip, and the Pink Sheet. The Task Force findings will also be reported at academic conferences, medical profession meetings, and appropriate trade conventions. Findings and Recommendations Valid approaches to economic analyses can be defined; acceptable methods can be differentiated from unacceptable ones. Bias in economic research stems from two major sources: lack of appropriate independence for researc

Issues in the Cross-National Assessment of Health Technology

With the growing international literature in economic evaluation and the rapid spread of new health technologies, there is a need to undertake, or at least interpret, economic evaluations on the international level. However, the ways in which cross-national differences affect the cost-effectiveness of health technologies or their evaluations have never been studied. This paper explores these issues by taking advantage of a unique situation in which the same economic evaluation of a new indication for a health technology was conducted simultaneously in four countries using an identical methodology. The study showed that if prior agreement on methods can be reached and local data applied, economic evaluations can be undertaken in a way that facilitates the extrapolation of results from country to country.

The Use of Physician Financial Incentives and Feedback to Improve Pediatric Preventive Care in Medicaid Managed Care

Objective. Immunizations and other cost-effective preventive services remain underused by many children, especially those living in poverty. Given the effectiveness of provider-based tracking systems and the widespread use by managed care organizations of financial incentives to influence physician practice patterns, we designed and tested an intervention combining these strategies. We studied whether a system of semiannual assessment and feedback, coupled with financial incentives, could improve pediatric preventive care in a Medicaid health maintenance organization (HMO). Methodology. We randomly assigned primary care sites serving children in a Medicaid HMO to one of three groups: a feedback group (where physicians received written feedback about compliance scores), a feedback and incentive group (where physicians received feedback and a financial bonus when compliance criteria were met), and a control group. We evaluated compliance with pediatric preventive care guidelines through semiannual chart audits during the years 1993 to 1995. Results. Compliance with pediatric preventive care improved dramatically in the study period. Repeated measures ANOVA demonstrated a significant increase in all three study groups throughout the time in total compliance scores (from 56%–73%), as well as scores for immunizations (from 62%–79%) and other preventive care (from 54%–71%). However, no significant differences were observed between either intervention group and the control group, nor were there any interaction (group-by-time) effects. Conclusions. Feedback to physicians, with or without financial incentives, did not improve pediatric preventive care in this Medicaid HMO during a time of rapid, secular improvements in care. Possible explanations include the context and timing of the intervention, the magnitude of the financial incentives, and lack of physician awareness of the intervention.

A Reappraisal of Hepatitis B Virus Vaccination Strategies Using Cost-Effectiveness Analysis

Hepatitis B virus (HBV) is the worlds most common blood-borne viral infection, chronically infecting more than 200 million people worldwide [1]. It is responsible for substantial morbidity and mortality and is a leading cause of hepatocellular carcinoma. Approximately 300 000 Americans are infected with HBV each year, an annual incidence rate of 11.5/10 000 population [2, 3]. Serologic evidence of previous infection is found in 4.8% of the U.S. population [4]. Each year approximately 4000 persons in the United States die of HBV-related cirrhosis and about 800 people die of HBV-related hepatocellular carcinoma [3, 5, 6]. The goal of our study was to determine the clinical and economic impact of alternative HBV vaccination strategies for selected U.S. populations. This analysis updates the earlier work of Mulley and colleagues [7] by incorporating more recent data on incidence of HBV infection, cost of acute disease and chronic sequelae, and safety and effectiveness of HBV vaccine, by focusing on a broader range of populations, and by modeling a longer period. A safe and effective vaccine to prevent HBV infection was introduced in the United States in 1981. Nonetheless, the overall incidence of HBV infection increased 14.5% between 1981 and 1987 [2, 3], especially in selected high-risk groups [8] (77% increase between 1981 and 1988 among injection drug abusers). The clear relationship between HBV infection and primary hepatocellular carcinoma indicates that HBV vaccine is the first effective anti-cancer vaccine [6]. However, the vaccine is not administered to most high-risk people (fewer than 50% of high-risk health care workers have received a complete HBV vaccination series) [3, 9, 10]. Acute HBV infection manifests itself clinically along a spectrum from asymptomatic, subclinical infection through fatal fulminant hepatitis [1, 5, 11]. More than 50% of HBV infections in adolescents and adults in the United States are asymptomatic [2, 9, 12, 13]. Others exhibit a variety of nonspecific viral symptoms; one quarter become ill with jaundice [3]. More than 10 000 people with acute HBV infection are hospitalized annually [3]. About 250 die each year of fulminant infection [3]. Both asymptomatic and nonfulminant symptomatic HBV infection are similarly associated with chronic forms of infection, which, in turn, may lead to cirrhosis and hepatocellular carcinoma [1, 5, 6, 11, 14-16]. The epidemiology of HBV infection varies greatly by geography and sociodemographic factors [1-4, 17]. Populations at greatest risk include those with multiple sex partners, sexually active male homosexuals, injection drug abusers, family members of HBV carriers, newborn children of actively infected mothers, medical personnel who are exposed to blood and blood products, immigrants from HBV endemic areas, and sexual partners of members of high-risk groups [3]. Epidemiologic studies have documented serologic evidence of HBV infection in as many as 50% of members of high-risk groups [2, 3, 18]. This combination of increasing incidence, expanding high-risk populations, and failure to deliver the vaccine to those at highest risk warrants a re-evaluation of current HBV vaccine policy. The Centers for Disease Control (CDC) has recommended universal vaccination of newborns and of high-risk adolescents [19]. The Occupational Safety and Health Administration (OSHA) has issued guidelines requiring HBV vaccination for all at-risk health care workers [20]. Others have proposed more limited vaccination strategies [21]. Methods Structure of the Model: Basic Decision Tree We used decision analysis to model clinical and economic consequences of alternative approaches to prevention and management of HBV infection. We focused on number and costs of cases prevented, deaths prevented, and years of life saved [22]. The decision model examined three different HBV management strategies (no vaccination, universal vaccination, and screen and vaccinate) in four different U.S. cohorts: 1) newborns; 2) adolescents at age 10; 3) a high-risk adult population with HBV infection incidence of 5% per annum; and, 4) the general adult population (12 to 50 years old). In addition, we examined a mixed population strategy of screening all pregnant women before or during labor for evidence of active HBV infection (HBsAg), administering HBV vaccine and hepatitis B immune globulin (HBIG) to the newborns of mothers who test positive, and then vaccinating all children at age 10 and again 10 years later. Each analysis followed a cohort of 10 000 individuals for three consecutive 10-year periods, because probabilities of infection or sequelae change over time. Cyclical Markov modeling could have been used, but its complexity makes presentation more difficult to understand, and accuracy is not importantly improved. The decision model used in this analysis was developed after a critical review of the literature, with input from five HBV experts to ensure accuracy and validity. The basic decision tree (Figure 1) incorporated the sensitivity and specificity of screening tests for previous or current HBV infection, estimates of compliance with vaccine administration, the probability of protection if vaccine is administered, the probability of exposure to HBV, and the probability of acute infection after exposure to HBV. All strategies assume that some individuals will be protected from HBV by virtue of previous infection. The terminal branches that end with no hepatitis B infection identify persons who remain uninfected with HBV at the end of the period. They enter the model again at the same starting point as everyone else (no previous HBV infection) for up to two additional consecutive 10-year periods. The terminal branches of the basic tree that end with hepatitis B infection lead to a subtree Figure 2 that reflects the distribution of types of acute disease and the probability of long-term sequelae of HBV infection. Figure 1. Structure of basic adult decision tree for first 10-year period. Figure 2. Structure of the decision tree for persons who contract hepatitis B virus. Data Sources We obtained the data used in the model from three sources. First, we critically reviewed the medical literature and found probability estimates of incidence and prevalence, clinical course and patient management of HBV infection and its sequelae in the community; safety and efficacy of HBV vaccine; and compliance with the administration of HBV vaccines. Second, using a modified Delphi technique, an expert panel reviewed these estimates, reconciled conflicting data, and provided consensus estimates for probabilities, events, patient management in the community, and outcomes for which reliable data were unavailable in the medical literature. Third, we obtained data on costs from actual private insurer payments, adjusted for expected patient out-of-pocket expenditures. Screening Strategies We used two screening strategies. First, the antibody to hepatitis B surface antigen (anti-HBsAg) was used in the adult populations to screen for previous infection to identify those individuals who were protected from recurrent infection and who, therefore, did not require vaccination. Second, HBV surface antigen (HBsAg) was used to screen mothers for active infection to identify newborns at high risk for HBV infection who would benefit from HBV vaccine and HBIG [22, 23]. Compliance and Vaccine Effectiveness Table 1 shows the input values for screening tests, vaccine compliance, and vaccine effectiveness used for the base case analysis. We assumed 100% compliance with maternal screening for HBV before or at the time of delivery. For other groups, we assumed compliance with screening to be 20% greater than the probability that they would comply with at least the first dose of vaccine. The literature and expert panel suggested a 50% compliance rate with the full series of three vaccinations (0, 1, and 6 months) among adults and adolescents, with lower compliance rates among high-risk populations and higher compliance rates among newborns because of accessibility in the hospital for the first dose of vaccine. People who did not comply with any of the initial series of vaccinations were assumed not to comply with any subsequent vaccine booster doses. Table 1. Base Case Input Values (Per Person)* Efficacy for full and partial vaccination series were estimated from randomized and historical clinical trials (Table 1) [7, 17, 23-34]. Antibody titers decrease over time after vaccination [26, 35-38]. Although epidemiologic studies do not indicate that such decreases in titers result in reduced protection against HBV infection, such studies have been limited to 7 or fewer years after vaccination [38]. The CDC currently does not recommend administration of vaccine boosters but is following populations of immunized patients to determine if such boosters will be required. The base case model we report here assumed that vaccine efficacy persisted for 10 years, after which a single booster dose was required to provide continuing protection against HBV infection for an additional 10 years, a conservative assumption that biases the model against the cost-effectiveness of all vaccination strategies. Among those who responded to the vaccine by developing protective antibodies, we made counterbalancing assumptions: no decrease in protection from HBV vaccine before 10 years but no protection after 10 years unless a booster is administered. We assumed that HBV vaccine did not incur any side effects that required medical care [26, 27] and that persons infected with HBV despite vaccination followed the same clinical course as those never vaccinated. Newborns of mothers who tested positive for active hepatitis B infection (HBsAg+) were assumed to be treated with HBIG plus HBV vaccine. These infants also underwent serologic testing at 1 year of age and those with low antibody titers (10% of neonates receiving a full primary series of HBV immuniza

CONTRACTUAL ARRANGEMENTS BETWEEN HMOS AND PRIMARY CARE PHYSICIANS : THREE-TIERED HMOS AND RISK POOLS

Concern about certain contractual arrangements between health maintenance organizations (HMOs) and primary care physicians has led policymakers to consider curbing these arrangements; one law has already been passed. However, these arrangements are complex and their impact is neither obvious nor well understood. This article first presents a conceptual approach to understanding the relationship between HMOs and primary care physicians and discusses how they influence the locus of financial risk and managerial control. It then refines understanding of two critical dimensions (three-tiered HMOs and risk pools) by examining survey responses of 260 HMOs (representing over 50% of total HMO enrollment.) Results of the evaluation led to the conclusion that primary care physicians in three-tiered HMOs are sheltered from some of the financial incentives and contractual arrangements enacted by the HMO and that the reason for using risk pools may be due more to peer group effects or interaction with other incentives, rather than the direct financial implications of the risk pool on individual physicians. These concepts and observations have relevance for other types of health care systems in this country and elsewhere. Policymakers risk enacting misguided policies unless they understand the details of these arrangements.

A comparison of international health outcomes and health care spending

Does increased spending improve health outcomes? We analyzed 1988 data from OECD countries to determine how key health care indexes correlate with health care outcomes. Total health care spending per capita and outpatient and inpatient utilization are not related to health outcomes. How our resources are allocated seems to be more important than how much money is actually spent.

The Adoption and Diffusion of Ct and Mri in the United States: A Comparative Analysis

This study examines and compares the rates and patterns of diffusion of computerized tomography (CT) and magnetic resonance imaging (MRI) over the first 4 years of their availability. Although early diffusion of CT was more rapid than that of MRI, adoption of MRI in nonhospital settings equaled that of CT. Analysis of attributes of the technologies and attributes of the regulatory, reimbursement, and market environments surrounding the early diffusion of these technologies provides insight into their different diffusion patterns. In particular, the technical and financial uncertainties surrounding MRI have inhibited its diffusion compared with that of CT. Medicares DRG-based prospective reimbursement system and certificate-of-need (CON) regulation by states have reduced overall MRI diffusion and stimulated purchases of MRI by nonhospital organizations. The FDAs premarket approval (PMA) program has changed marketing strategies and influenced the diffusion of MRI to a lesser degree. This analysis identifies problems in how the present health care system evaluates and adopts new, expensive, diagnostic technologies and suggests changes to make the system more responsive to present needs.This study examines and compares the rates and patterns of diffusion of computerized tomography (CT) and magnetic resonance imaging (MRI) over the first 4 years of their availability. Although early diffusion of CT was more rapid than that of MRI, adoption of MRI in nonhospital settings equaled that of CT. Analysis of attributes of the technologies and attributes of the regulatory, reimbursement, and market environments surrounding the early diffusion of these technologies provides insight into their different diffusion patterns. In particular, the technical and financial uncertainties surrounding MRI have inhibited its diffusion compared with that of CT. Medicares DRG-based prospective reimbursement system and certificate-of-need (CON) regulation by states have reduced overall MRI diffusion and stimulated purchases of MRI by nonhospital organizations. The FDAs premarket approval (PMA) program has changed marketing strategies and influenced the diffusion of MRI to a lesser degree. This analysis identifies problems in how the present health care system evaluates and adopts new, expensive, diagnostic technologies and suggests changes to make the system more responsive to present needs.

Competing Practice Guidelines: Using Cost-Effectiveness Analysis To Make Optimal Decisions

Wide variation in physician care patterns [1-5] in the setting of rapidly increasing health care spending [6, 7] has led to efforts to foster greater consistency and value. For example, numerous clinical practice guidelines, algorithms, critical pathways, and standards (hereafter referred to collectively as clinical guidelines) have been developed in attempts to enhance quality of care while reducing avoidable variation in the costs of providing that care [8, 9]. Making the best clinical decision for a given patient requires knowing the potential costs and outcomes of different choices about treatment [10, 11]. This allows a decision maker to prioritize options according to their value, or cost-effectiveness [12, 13]. Applying more valuable clinical strategies first and following them with strategies of successively decreasing value should achieve optimal allocation of limited clinical and financial resources at the level of the individual patient. At the level of a population of patients with multiple clinical conditions, how does one decide among numerous different, clinically acceptable, and ethically valid treatment options, all of which differ in effectiveness and cost? To efficiently manage scarce resources, planners in the industrial sector have constructed complex mathematical models to capture key relations between resource and output variables, such as the availability of financial resources, suppliers, raw materials, producers, and distribution channels and expected demand. After describing the values that certain variables are allowed to take, one can use the set of mathematical techniques collectively known as optimization (using linear or nonlinear programming) [14] to maximize or minimize one key variable (such as benefit or risk). For example, an airline carrier with routes connecting several dozen cities and a limited number of aircraft and crew members generally wishes to minimize total cost. Optimization enables efficient routing adjustments and aircraft and crew member deployment by taking into account such constraints as local costs of jet fuel and required rest time for crew members. The result is the best possible arrangement for delivering the best possible outcomes with limited resources. In health care, the use of optimization is still new and is limited to well-defined areas in which one can easily summarize pathophysiology with mathematical equations. These areas include ventilator management in critically ill patients, adjustment of oral anticoagulation, treatment planning in radiation therapy, and maintenance of proper dialysate content in hemodialysis [15-18]. In health care, an important implication of this industrial strategy is that choosing a slightly less costly and less clinically effective treatment for a prevalent condition may conserve enough resources to permit the purchase of more valuable treatments for other, less prevalent conditions. Whereas each choice may not be the most cost-effective option for an individual patient, the constellation of interventions could best improve overall public health. In this article, we use optimization, supported by existing clinical guidelines, to show 1) which group of clinical options maximizes overall benefit for a population of patients and 2) how this group of options differs from options that maximize benefit for individual patients. We also show how the group of selected options changes according to the extent of resource constraints. Finally, we suggest ways in which cost-effectiveness analysis should be used to allocate resources. Methods In this study, we used optimization to select the best clinical options that, taken together, maximized the number of years of life added to a hypothetical population of 100 000 persons with an age and sex distribution similar to that of the United States in 1991 [19, 20]. For clarity, we considered only a limited number of diseases in the model. (More current clinical guidelines and epidemiologic inputs could easily be used to update our study.) Selection of Interventions and Clinical Situations Using MEDLINE to search the clinical literature from 1986 to the present, we sought clinical practice guidelines that 1) addressed clinical situations in which guidelines have actually been used, 2) evaluated differences in outcomes and direct medical costs between or among two or more ways of providing care, 3) used added years of life per patient (unadjusted for quality of life) to measure outcomes of care [21, 22], 4) made recommendations on the basis of cost per unit of outcome [for example, per added year of life], and 5) discounted both costs and outcomes at 5% per year (the standard approach to discounting) [23, 24]. Table 1 lists the six interventions that we selected for the model [25-30]. We included examples of major categories of health activities: prevention (prevention of hepatitis B), screening (screening for colorectal cancer), diagnosis (diagnosis of stable angina), risk factor reduction (risk factor reduction for hypercholesterolemia and smoking), and treatment (treatment of recurrent ventricular arrhythmia). Table 1. Interventions, Clinical Options, and Clinical Subgroups for Decision Making* For each clinical intervention, we summarized the key, mutually exclusive directions that could be followed, listing them as clinical options (Table 1, column 2). For the sake of clarity, Table 1 shows only the relevant characteristics of each option; the original [25-30] may be consulted for specifics on such variables as age ranges and doses. Each clinical option is considered to be of some benefit and is part of the standard repertoire of options that competent physicians might have offered their patients in 1991. Our task was to choose, for a population of patients, a single best option for each clinical intervention. We based our selection on total population benefit rather than on benefit for the individual groups of patients for whom guidelines were developed. Finally, in addition to assuming the existence of a standard U.S. population in terms of age and sex, we accounted for the fact that several of the cost-effectiveness studies reported results for multiple, more specific types of patients. Columns three and four in Table 1 highlight the instances in which more than one selection per clinical intervention was needed. For example, where guideline data existed for several types of patients, the model was programmed to select the best option for each type of patient. Selecting a clinical option for colorectal cancer screening, for instance, required only one decision (among the five clinical options) because the reference that we used discussed the group of persons at 65 years of age. However, selecting an option for the diagnosis of stable angina required four decisions, one for each of the clinical subgroups, beginning with those 35 to 44 years of age and ending with those 65 to 74 years of age. We sought the choice of one specific clinical option for each of the 22 different clinical subgroups of patients. Derivation of Population Cost and Effectiveness Data For each of the 22 clinical subgroups that required a selection decision, we assigned (from a cluster of allowable options) one baseline clinical option to serve primarily as the standard against which the cost and effectiveness of each competing option could be compared (Appendix Figure 1). Each baseline strategy was considered to be the option most widely currently practiced. If the optimization program selected the baseline strategy option for the population-wide solution, no additional years of life were added to the population at no additional cost because the incremental cost and effectiveness of a clinical option compared with itself are zero. Appendix Figure 1. For each selection decision, we calculated the incremental cost and incremental effectiveness (in years of life) of each of the allowable alternative clinical options. In other words, we obtained cost and effectiveness data directly from each study in the literature but recalibrated the data, when necessary, to express cost and effectiveness in terms of the baseline option. To determine the incremental cost and effectiveness of each alternative option, we adjusted the data, expressed per patient as obtained from the literature, for the approximate demand for that option over the next 12 months in our 100 000-member population. For example, a paper might report that a specific alternative clinical option (compared with the baseline strategy) resulted in a gain of 1.5 more years of life per patient at an additional cost of

Impact of a Mandatory Medicaid Case Management Program on Prenatal Care and Birth Outcomes: A Retrospective Analysis

20 000 per patient. If the estimated incidence of that clinical condition in our population over the next year was 30 patients, the population-wide incremental cost and effectiveness entries would be