Faculty Bibliography

BACKGROUND: The September 11th disaster in New York City resulted in an increase in mental health service delivery as a vast network of providers responded to the urgent needs of those impacted by the tragedy. Estimates of current capacity, potential additional capacity to deliver services and of potential shortfall within the mental health sector are needed pieces of information for planning the responses to future disasters. AIMS OF THE STUDY: Using New York State data, to determine the distribution of clinical service delivery rates among programs and to examine an explanatory model of observed variation; to estimate potential additional capacity in the mental health sector; and to estimate shortfall based on this capacity and data from studies on the need and use of services post September 11th METHODS: Empirical distributions of weekly clinical service delivery rates in programs likely to be used by persons with post disaster mental health problems were obtained from available data. Three regression models were fit to explain rate variation in terms of unmodifiable program characteristics likely to impact the rates. We argue that rates could not be easily increased if any of the models had good explanatory power, and could be increased if it did not. All models had poor fit. We then assumed that the median and 75th percentile of the clinical service delivery rates were candidates for the minimum production capability of a clinician. The service rates of those clinicians whose rates fell below these quartiles were increased to the quartile value to yield estimates of potential additional capacity. These were used along with data on clinical need to estimate shortfall. RESULTS: There is substantial variation in clinical service delivery rates within impact regions and among programs serving different age populations. The estimate of the percent increase in services overall based on the median is 12% and based on the 75th percentile is 27%. Using an estimate of need of.03 suggested by available data, and a range of services (1-10) that might be required in a six month period, shortfall estimates based on the median ranged between 22-92% and for the 75th percentile from no shortfall to 86%. A less conservative estimate of need of.05 produces median shortfall ranging between 59-96% and for the 75th percentile between 10-91%. LIMITATIONS: While the program descriptor variables used in the explanatory model of rates were those most likely to impact rates, explanatory power of the model might have increased if other characteristics that are not modifiable had been included. In this case, the assumption that service production can be increased is called into question. IMPLICATIONS FOR HEALTH CARE PROVISION AND USE: In the first six months post September 11th, in New York State (NYS) 250,000 persons received crisis counseling through Project Liberty. In 1999, NYS served approximately that same number in mental health clinic programs during the entire year. The estimates of this study suggest that additional funding and personnel are needed to provide mental health services in the event of a major disaster. IMPLICATIONS FOR HEALTH POLICIES: A disaster plan is needed to coordinate the use of current and additional personnel including mental health resources from other sources and sectors

The capture-recapture approach to estimating the size of a population is a well-studied area of statistics. The number of distinct individuals, N(A) and N(B), on each of two lists, A and B, and the number common to both lists, N(AB), are used to form an estimate of the binomial probability of being on one of the lists, which then allows an estimate to be made of the size of the population. Critical to the method is an accurate count of N(AB). We consider situations in which this count is not available. Such problems arise in a variety of behavioural health contexts in which the need for protection of privacy may prevent sharing identifying information, so it is not possible to specifically match an individual who appears on one list with an individual on the other. Suppose that the birth dates and/or other demographics of individuals on each list are known. We introduce two methods for estimating the duplication rates and the size of the population. Conditioning on the set beta of birth dates of those on list B, N(A) and N(B), the maximum likelihood estimators (MLEs) and their variance are derived. The MLEs are based on the proportion of individuals on list A whose birth dates fall in beta. This approach is particularly useful if list B itself contains duplicates. The second model utilizes the full sample distribution of the birth dates. We generalize this approach to accommodate multiple demographic characteristics. The approaches are applied to the problem of estimating duplication rates and the population size of veterans who have mental illness in Kings County, NY. The data are lists of those receiving service from the Veterans Administration system and from providers funded or certified by the New York State Office of Mental Health

A project is described in which performance measures of cultural competency in behavioral health care were selected and benchmarked. Input from an Expert Panel representing the four major ethnic and racial groups in the U.S. and persons with extensive experience in implementing cultural competency in health care, along with survey data from 21 sites were used in the process. Measures and benchmarks are made specific to organizations that administrate care networks, and to service entities that deliver care. Measures were selected to parallel an implementation process, and benchmarks were set at 'gold standard' levels

Interest in the use of net health benefit in cost-effectiveness analysis derives from its optimality property for decision-making. A description of the results of an economic evaluation of health care interventions is incomplete if it does not include point and interval estimates of this outcome measure. A simultaneous confidence band for the incremental net health benefit, INHB(lambda), for all lambda may be obtained by forming a confidence interval based on student's t statistic, and letting the willingness-to-pay value, lambda, run over all values. The familywise error rate (FWER) of the simultaneous confidence band is the probability that the confidence interval does not cover the true INHB(lambda) for some value of lambda. We show that the FWER equals P(T(2)>t(2)), where T(2) follows Hotelling's central distribution and that the simultaneous confidence band does not cover the true INHB(lambda) if and only if a T(2) based confidence ellipsoid does not cover the true mean c-e vector

Statistical methods are given for producing a cost-effectiveness frontier for an arbitrary number of programs. In the deterministic case, the net health benefit (NHB) decision rule is optimal; the rule funds the program with the largest positive NHB at each lambda, the amount a decision-maker is willing to pay for an additional unit of effectiveness. For bivariate normally distributed cost and effectiveness variables and a specified lambda, a statistical procedure is presented, based on the method of constrained multiple comparisons with the best (CMCB), for determining the program with the largest NHB. A one-tailed t test is used to determine if the NHB is positive. To obtain a statistical frontier in the lambda-NHB plane, we develop a method to produce the region in which each program has the largest NHB, by pivoting a CMCB confidence interval. A one-sided version of Fieller's theorem is used to determine the region where the NHB of each program is positive. At each lambda, the pointwise error rate is bounded by a prespecified alpha. Upper bounds on the familywise error rate, the probability of an error at any value of lambda, are given. The methods are applied to a hypothetical clinical trial of antipsychotic agents

BACKGROUND: Poorly defined cohorts and weak study designs have hampered cross-cultural comparisons of course and outcome in schizophrenia. AIMS: To describe long-term outcome in 18 diverse treated incidence and prevalence cohorts. To compare mortality, 15- and 25-year illness trajectory and the predictive strength of selected baseline and short-term course variables. METHODS: Historic prospective study. Standardised assessments of course and outcome. RESULTS: About 75% traced. About 50% of surviving cases had favourable outcomes, but there was marked heterogeneity across geographic centres. In regression models, early (2-year) course patterns were the strongest predictor of 15-year outcome, but recovery varied by location; 16% of early unremitting cases achieved late-phase recovery. CONCLUSIONS: A significant proportion of treated incident cases of schizophrenia achieve favourable long-term outcome. Sociocultural conditions appear to modify long-term course. Early intervention programmes focused on social as well as pharmacological treatments may realise longer-term gains

Statistical methods for cost-effectiveness analysis (CEA) for two treatments that mimic the deterministic optimal rules of CEA are presented. In these rules the objective is to determine the treatment with the maximal effectiveness whose unit cost is less than an amount, lambda, that a decision-maker is willing to pay (WTP). This is accomplished by identifying the treatment with the largest positive net health benefit (NHB), which is a function of lambda, while controlling the familywise error rate both when the WTP value is given and when it is unspecified. Fieller's theorem is used to determine a region of WTP values where the NHBs of the treatments are not distinguishable. For each lambda outside of the confidence region, the larger treatment is identified. A newly developed one-tailed analogue of Fieller's theorem is used to determine the WTP values where a treatment's NHB is positive. The situation in which both treatments are experimental is distinguished from the case where one of the treatments is usual care. The one-tailed confidence region is used in the latter case to obtain the lambda values where the NHBs are not different, and determining the region of positivity of the NHBs may be unnecessary. An example is presented in which the cost-effectiveness of two antipsychotic treatments is evaluated

BACKGROUND: Analyses that have been conducted previously on the implications of parity have focused on the concern that mental health costs of private payers will substantially increase. A complete analysis of the cost implications of parity, however, also needs to consider whether the mental health costs of public payers may increase particularly if employers or private insurers attempt to extrude enrollees with severe mental illness. This study examines the extent of mental health cost shifting from private to public payers during two separate two-year periods prior to the implementation of parity legislation. The results of the analyses can serve as a necessary baseline against which the consequences of parity legislation on this direction of cost-shifting can be examined. METHODS: The study utilizes an all payer data set that contains information on the use of specialty mental health services (excluding private practitioners) by adults in an urban and a rural county in New York State. For each year of two time periods -1991/1992 and 1995/ 1996 - consumers were classified into payer groups based on whether their services were paid for by 'Private Only', 'Public Only', 'Private/Public', 'Self Pay' or 'Other' payers. The proportion of individuals who moved from one payer group to another from one year to the following year of each time period and the average yearly costs under these payers were examined. Logistic regression models were used to identify the characteristics of persons most likely to remain with Private Only Payers in contrast to those likely to shift to Private/Public Only payers or to Public Only Payers. RESULTS: In both two-year time periods, the percent of persons who shifted in one year from Private Only to either Private/Public or Public Only payers was small. In contrast, a person in the Private/Public group has more than a 12 percent likelihood of shifting to a Public Only payer in the subsequent year. The average annual costs of the Private/Public group were higher than that of any other payer group. The average annual costs of persons who shifted into the Private/Public group from any other payer group or remained there from the previous year were even higher. The logistic regression analyses for both time periods showed that persons who shifted from Private Only to Private/Public or Public Only payers in contrast to those who remained with Private Only payers were more likely to have subsidized incomes, be younger and have a mental health disability. In 1995, the likelihood of the shift was also increased for those who were nonwhite and/or had a substance abuse disability. IMPLICATIONS: This study has found that individuals rarely shift directly from private payers to public payers. Rather, they first shift to having services reimbursed by both private and public payers, and during this period their average total service costs are extremely high. Persons who shift from private payers to having at least some of their services paid by public payers in subsequent years appear to be either young employees or young dependents who have severe mental illness or mental illness disabilities. Abusing substances and/or being nonwhite also increase the likelihood of a shift to public payers. Along with parity mandates, there has been an increase in managed care controls. The extent to which these controls will be used to accelerate the movement of these high cost persons from private to public payers needs close watch

The authors utilized numerous documents created by advisory groups, expert panels and multicultural focus groups to develop performance measures for assessing the cultural competency of mental health systems. Competency was measured within three levels of organizational structure: administrative, provider network, and individual caregiver. Indicators, measures and data sources for needs assessment, information exchange, services, human resources, plans and policies, and outcomes were identified. Procedures for selection and implementation of the most critical measures are suggested. The products of this project are broadly applicable to the concerns of all cultural groups

It is common for clinical trials designed to compare treatments for migraine to incorporate a component for estimating onset. Our objective is to describe a stopwatch method for collecting data on time to meaningful relief and a conceptual framework for describing and analysing the results. The survival distribution of onset is modelled in two parts: the probability that onset does not occur, and the survival distribution conditional on its occurrence. Using data from a clinical trial comparing an active treatment and placebo, we illustrate the method and find that the distributions of onset among those with onset do not differ, but the probabilities that onset occurs are substantially different. We illustrate how the model can be used to help determine how long patients without onset should wait before further intervention, how patients interpret the phrase meaningful relief, and how baseline clinical characteristics affect the onset