What is the role of two-wave designs in clinical research? Comment on Hageman and Arrindell.

Traditional beliefs about regression to the mean, difference scores (d-scores) and corrections for regression in determining the significance of individual person change have been challenged by proponents of growth curve modeling for the study of change. These challenges have generally not been adequately addressed by those proposing modifications to Jacobson's Reliable Change Index (RC). It is proposed that (1) two-wave designs and RC or variations of it are not very good methods for the scientific study of change, (2) multi-wave designs are better, (3) d-scores are unbiased data for analyzing two-wave studies, (4) RC is relatively more justifiable than most corrected or adjusted forms of RC, (5) the more appropriate place for two-wave designs and RC are in program or outcome evaluation and (6) the most appropriate function for RC is to facilitate communication among evaluators, public decision makers, providers and the public. In the absence of a 'gold standard' or consensus, it is recommended the only RC be used in outcome evaluation, in the near future, in order to enhance comparability of change rate data among different studies.

Ethnic, demographic, and social differences among middle and older adults with HIV/AIDS.

Although the prevalence of AIDS among middle and older adults is increasing, little is known about them. The objective of this study was to obtain basic demographic and social information about people with HIV/AIDS (PWAs) over the age of 44 years. This was a descriptive, cross-sectional survey of convenience samples of 76 African-American (AA) and 80 White older PWAs. Participants were interviewed by trained peer interviewers using a structured and largely objective schedule. Participants and interviewers were each paid $35 per completed interview. The schedule consisted of wide-ranging demographic, HIV/AIDS, stressor, coping, social, and support questions. Although both ethnic groups had incomes significantly below national norms and poverty rates three times their national rates, AA PWAs were markedly disadvanted socioeconomically relative to White PWAs. Although over 60% of all PWAs reported that having enough money to live on was a problem, AAs also reported significantly more Stressors, many related to economics, then White PWAs. AA PWAs reported being predominantly heterosexual, while the largest group of Whites was homosexual or bisexual. Sources of infection paralleled sexual orientation. Both groups used relatively effective coping strategies and reported moderate levels of social support and activity. It was concluded that the course of illness, treatment, and quality of life of middle and older PWAs are likely to be complicated by economic factors, and this will be particularly serious among AA PWAs.

Five methods for computing significant individual client change and improvement rates: support for an individual growth curve approach.

Interest has been renewed in methods for determining individual client change. Currently, there are at least 4 pretreatment-posttreatment (pre-post) difference score methods. A 5th method, based on a random effects model and multiwave data, represents a growth curve approach and was hypothesized to be more sensitive to detecting significant (p < .05) change than the pre-post methods. The change rates produced by the 5 methods were compared in a sample of 73 older outpatients with 3 to 5 assessments per client on a measure of well-being (H. J. Dupuy, 1977). Results indicated that the growth curve approach improvement rate was the highest (68.5%). The growth curve and the Edwards-Nunnally (63.0%) methods produced significantly (p < .05) higher improvement rates than the other 3 methods, with 1 exception.

The relationship between hospice length of stay and caregiver adjustment.

This study investigated the hypothesis that the greater the predeath length of stay in hospice the better would be the primary caregivers' bereavement adjustment 6 months post death. One hundred sixty caregivers from a 30% pool of hospice survivors who voluntarily returned 6-month, follow-up questionnaires were studied. Five length of stay groups were compared. There were no statistically significant differences in bereavement adjustment between caregivers with varying predeath lengths of stay. It was concluded that bereavement resolution is a complex process, and that cumulative medical illness burden, impairment of ADLs, and caregiver burden need to be considered.

Can treatment research inform decision makers? Nonexperimental method issues and examples among older outpatients.

Outcome research does not seem to have informed policymakers, decision makers, or practitioners. It is proposed that reporting of improvement-deterioration rates and greater use of single-sample, pretreatment-posttreatment designs in community clinic settings will facilitate communication and, perhaps, mitigate the external validity problem of psychotherapy experiments. The Edwards-Nunnally method (D. C. Speer, 1992) was used to investigate change rates from a single cohort of older adult outpatients (N = 92). The cohort improvement rate was 51%, and the deterioration rate was 5%. However, improvement rates for clients with major depression and adjustment disorders were greater than 70%. Change rates of student and staff therapists and of clients treated in the clinic and in their homes were not significantly different. Deteriorated clients are briefly described.

Comorbid mental and substance disorders among older psychiatric patients.

OBJECTIVE: To investigate the rate and configurations of current comorbid mental and substance disorders among older psychiatric patients. DESIGN: A descriptive, retrospective study. SETTING: A non-acute, public residential psychiatric treatment facility for adults 55 years of age and older (mean length of stay: 3 months). PATIENTS: 128 patients discharged during a 2-year period. MEASURES: DSM-III-R diagnoses, demographic and history data. RESULTS: The overall prevalence rate of concurrent mental and substance disorders during the present treatment episode was 21%. This is comparable to the 6-month rate of 19.8% found in the Epidemiological Catchment Area Study among adults in treatment (all ages). Nearly 50% of the dually diagnosed subsample received more than two diagnoses, with 60% of the subsample receiving a personality disorder diagnosis. Concurrent affective, alcohol, and personality disorder diagnoses were common. CONCLUSION: Older patients with comorbid substance-use and mental disorders may differ substantively from comorbid younger patients. Among older patients in an intermediate-term psychiatric facility, the triple occurrence of alcoholism personality disorder and depression was common.

Clinically significant change: Jacobson and Truax (1991) revisited.

The relationship between statistically and clinically significant change has been enigmatic. Jacobson and Truax (1991) have proposed an important step toward rapprochement. However, their suggested index of clinically significant change neglects possible confounding of improvement rate estimates by regression to the mean. An alternative method is described that incorporates an adjustment that minimizes this confounding when statistical regression has been shown to be present. If regression is not present, the Jacobson and Truax method is more appropriate; if regression is present, the Edwards-Nunnally method (Edwards, Yarvis, Mueller, Zingale, & Wagman, 1978) is more appropriate. The two methods are compared, and the effects of instrument reliability and sample deviance on estimated improvement rates are demonstrated using general well-being test-retest data from a sample of older adult mental health outpatients.

Differences in social resources and treatment history among diagnostic groups of older adults.

Differences in demographic and social characteristics and treatment history of 407 patients age 55 or older were examined in relation to diagnosis. Seven diagnostic groups in three mental health service programs were studied, all of which were characterized by surprisingly high rates of previous treatment (75 percent overall). Patients in the dysthymia, adjustment disorder, and V code groups appeared to be the most similar and to have the greatest psychosocial resources and fewest indicators of chronicity. Schizophrenic and bipolar patients had the least resources and most indicators of chronicity. Substance abuse patients tended to live alone, while those with major depression seemed to have more resources and later onset of their illness than schizophrenic and bipolar patients. About half of those with milder diagnoses had received prior outpatient treatment, suggesting that perhaps past treatment success leads older adults to revisit mental health services.

Older adult users of outpatient mental health services.

Descriptive demographic, clinical and interview data from clients of an outpatient mental health clinic for older adults are presented. These clients are better educated and have a more frequent history of marital difficulties than older adults in general. Although a significant minority received V code and Adjustment Disorder diagnoses, the majority were depressed and had histories of prior mental health services. Economic factors are a major issue in seeking services.

Dual diagnosis among older adults: a new array of policy and planning problems.

Case finding and successfully engaging and treating older adults with substance abuse disorders continue to be enigmatic processes. Because of the adverse consequences on health of substance abuse among the elderly, substance abuse has serious implications not only for the well-being of the elderly but also for the future costs and resources of the mental health and health care systems for older adults. However, older adults with concurrent mental and substance abuse disorders are appearing in the public gerontological mental health system. This system is both unprepared and ill-equipped to manage and treat these often difficult clients. Further complicating the situation is the seeming lack of knowledge about "dual diagnosis" among the elderly. Policy and planning issues necessary to address this newly discovered array of problems are discussed.

The "monitoring model" and the mortality X treatment interaction threat to validity in mental health outcome evaluation.

Symptom improvement rates among three cohorts of adult outpatients generated by a base-line monitoring model approach to outcome evaluation are presented. Follow-up questionnaire data were collected by mail. The cohorts represented approximately 40% self-selected samples from the parent populations of clients admitted for service. Efforts to interpret significant cohort improvement rate differences led to the conclusion that differences may have been the result of severely symptomatic clients differentially selecting themselves in/out of follow-up data collection. This Mortality X Treatment interaction threat to the validity of conclusions about service effectiveness is discussed in the context of Campbell and Stanley's recurrent institutional cycle design, with particular emphasis placed on reducing posttreatment data loss.

Evaluation of mental health service effectiveness: a "start up" model for established programs.

The practical difficulties of using the experimental model for program evaluation in direct service settings are discussed, and it is suggested that use of this model seems premature. These difficulties, however, can no longer be used as a rationale for delaying program evaluation. A practical alternative model is outlined.