Mental health care utilization in prepaid and fee-for-service plans among depressed patients in the Medical Outcomes Study

OBJECTIVE

We compare mental health utilization in prepaid and fee-for-service plans and analyze selection biases.

DATA SOURCE

Primary data were collected every six months over a two-year interval for a panel of depressed patients participating in the Medical Outcomes Study, an observational study of adults in competing systems of care in three urban areas (Boston, Chicago, and Los Angeles).

STUDY DESIGN

Patients visiting a participating clinician at baseline were screened for depression, followed by a telephone interview, which included the depression section of the NIMH Diagnostic Interview Schedule. Patients with current or past lifetime depressive disorder and those with depressed mood and three other lifetime symptoms were eligible for this analysis. We analyze mental health utilization based on periodic patient self-report. ANALYTIC METHODS: We use two-part models because of the presence of both nonuse and skewness of use. Standard errors are corrected nonparametrically for correlations across observations due to clustered sampling within participating physicians and repeated observations on the same individual.

PRINCIPAL FINDINGS

The average number of mental health visits was 35-40 percent lower in the prepaid system, adjusted and unadjusted for observed differences in patient characteristics, including health status. Utilization differences were concentrated among patients of psychiatrists, with only minor differences among patients of general medical providers. Analyzing the effect of switches that patients make between payment systems over time, we found some evidence of adverse selection into fee-for-service plans based on baseline utilization, but not based on utilization at the end of the study. In particular, after adjusting for observed patient characteristics and health status, patients switching out of prepaid plans had higher baseline use than predicted, whereas patients switching out of fee-for-service had lower use than predicted. Switching itself appears to be related to an immediate decline in utilization and was not followed by an increase or "catch-up" effect.

CONCLUSIONS

The absence of the commonly found "catch-up" effect following switching and the significant decrease in utilization during the switching period suggests an interruption in care that does not occur for patients staying within a payment system. This finding emphasizes the need for integrating new patients quickly into a system, an issue that should not be neglected in the current policy discussion.

Personal and psychosocial risk factors for physical and mental health outcomes and course of depression among depressed patients

This article focuses on personal and psychosocial factors to identify those that predict change in functioning and well-being and clinical course of depression in depressed outpatients over time. Data from 604 depressed patients in The Medical Outcomes Study showed improvements in measures of functioning and well-being associated with patients who were employed, drank less alcohol, and had active coping styles. Better clinical course of depression was associated with patients who had high levels of social support, who had more active and less avoidant coping styles, who were physically active, and who had fewer comorbid chronic conditions. Findings provide some guidance as to what can be done to improve depressed patients' levels of physical and mental health and affect the clinical course of depression.

Prevalence of behavior problems in US children with asthma

OBJECTIVES

To present national population-based estimates of the prevalence of parent-reported emotional and/or behavioral problems in children with asthma and the relationship of medical comorbidity and asthma severity with behavior problems.

DESIGN

Cross-sectional study of the 1988 National Health Interview on Child Health.

MAIN OUTCOME MEASURES

Parent responses to a checklist of child health conditions were used to assign school-age children (5 to 17 years old) into one of four groups: children without reported chronic conditions; children with asthma alone; children with asthma and additional reported chronic conditions; and children with the same chronic conditions, but without asthma. Parental responses to the Behavior Problem Index (BPI) were used for construction of an overall BPI score, as well as subscale scores. Cross-tabulation and linear and logistic regression were used to determine the relation of the different condition categories to emotional and/or behavioral problems expressed by relative values of the BPI.

RESULTS

Children with asthma and comorbid conditions had a mean BPI score of 7.3, compared with 5.4 for children without chronic conditions, and all subscale scores, except those for antisocial conduct and immature behavior, were significantly elevated. Using logistic regression to control for confounding variables, children with severe asthma alone had nearly three times the odds (odds ratio, 2.96; 95% confidence interval, 1.22 to 7.17) and children with asthma plus comorbid conditions nearly twice the odds (odds ratio, 1.86; 95% confidence interval, 1.20 to 2.90) of children without chronic conditions to have severe behavior problems.

CONCLUSIONS

Severe asthma and asthma with medical comorbidity represent significant risk factors for emotional and/or behavioral problems. Clinicians caring for children with asthma and their families should be aware of the relationship between asthma and emotional and/or behavioral problems and anticipate that a substantial number of their patients may have mental health services needs.

Cost containment and mental health outcomes: experiences from US studies

BACKGROUND

Cost containment mechanisms, such as prepayment, are being considered or implemented in the US and elsewhere, but there have been few studies of the effects of such mechanisms on quality or outcomes of care for individuals with serious psychiatric disorders.

METHOD

Key results from US studies on cost containment and their implications are reviewed.

RESULTS

Cost savings in out-patient mental health care can be achieved through increasing the share of costs paid by the covered individual or through prepayment, but individuals with the greatest psychological distress or poor people may achieve worse outcomes under greater cost containment. Quality of care may be poorer under some forms of prepayment than under fee-for-service care, yet a national prospective payment mechanism for depressed elderly in-patients was not associated with a marked drop in quality or outcomes of care among those admitted.

CONCLUSIONS

Prepayment, relative to fee-for-service is not always associated with lower outcomes or quality of care for affective disorders. Under cost containment, quality and outcomes of care, especially for the sick poor, should be monitored to identify adverse consequences.

Severity of depression in prepaid and fee-for-service general medical and mental health specialty practices

This study compares severity of depression for patients of general medical clinicians, psychiatrists, and nonphysician therapists receiving prepaid or fee-for-service care. Cross-sectional severity comparisons were conducted among 715 outpatients with current major depression or dysthymia, by independent assessment. Severity was assessed by counts of current and lifetime depressive symptoms, prognostic and treatment response indicators, and global measures of psychological and physical sickness. Patients of psychiatrists were the most psychologically ill, patients of nonphysician therapists were intermediate, and general medical patients were least ill; but even in the general medical sector, depression severity was at least ill; but even in the general medical sector, depression severity was at least moderate. No differences in global physical sickness by specialty remained after demographic adjustment. General medical patients whose depression had been detected were only slightly sicker than undetected cases. Type of payment was not consistently related to either psychological or physical aspects of sickness, and payment did not interact with specialty. Mental health specialists, especially psychiatrists, encountered more severely depressed patients, but patients in all sectors were sick enough to warrant treatment. Even undetected patients in the general medical sector were relatively sick, raising questions about gatekeeper policies. There was no evidence of a greater severity gradient by specialty in prepaid care. Because payment was unrelated to severity, treatment implications are similar under prepaid and fee-for-service care. Implications for clinical practice, public policy, and outcomes research design are discussed.

How can care for depression become more cost-effective?

OBJECTIVE

To determine the cost and health effects of changes in the content and quality of care for depressed patients treated in prepaid general medical practices (internal and family medicine) and mental health specialty practices and shifts in the proportion of patients treated in general medicine.

METHODS

Cost-effectiveness analysis and simulations, which are empirically based on data from the Medical Outcomes Study.

OUTCOME MEASURES

Change in serious functional limitations, annual treatment costs per patient, and costs per reduction in one functional limitation.

RESULTS

More appropriate care for depression (increased counseling, use of appropriate antidepressant medications, or avoidance of regular minor tranquilizer use) improves functioning outcomes. Although this approach increases total costs of care, it also improves the value of care because each dollar spent on care now provides more benefits in terms of health improvements. In contrast with the effects of more appropriate care for depression, the trend away from mental health specialty care and toward general medical provider care under current treatment patterns reduces costs, worsens outcomes, and does not increase the value of health care spending in terms of health improvement per dollar.

CONCLUSION

Quality improvement measures that roughly follow practice guidelines for depression can improve outcomes and the value or cost-effectiveness of care, but at increased treatment costs; shifting patients away from mental health specialists decreases costs but worsens functioning outcomes. The best strategy for making care for depression more cost-effective is through quality improvement, not through changing specialty mix. Yet combining these strategies may achieve better outcomes, lower treatment costs, and better value of care compared with current practice patterns. To realize this potential, however, substantial quality improvement of care for depression is necessary in general medical practice.

Functioning and well-being outcomes of patients with depression compared with chronic general medical illnesses

BACKGROUND

Cross-sectional studies have found that depression is uniquely associated with limitations in well-being and functioning that were equal to or greater than those of chronic general medical conditions such as diabetes and arthritis. However, whether these relative limitations persist over time is not known.

METHODS

We conducted a 2-year observational study of 1790 adult outpatients with depression, diabetes, hypertension, recent myocardial infarction, and/or congestive heart failure. Change in functional status and well-being was compared for depressed patients vs patients with chronic general medical illnesses, controlling statistically for medical comorbidity, sociodemographics, system, and specialty of care.

RESULTS

Over 2 years of follow-up, limitations in functioning and well-being improved somewhat for depressed patients; even so, at the end of 2 years, these limitations were similar to or worse than those attributed to chronic medical illnesses. Similar patterns were observed for depressed patients in the mental health specialty sector and those in the general medical sector, but the patients in the mental health specialty sector improved more. More severely depressed patients improved more in functioning, but even initially depressed patients without depressive disorder had substantial persistent limitations.

CONCLUSION

Depressed patients have substantial and long-lasting decrements in multiple domains of functioning and well-being that equal or exceed those of patients with chronic medical illnesses.

Care for depression in a changing environment

The U.S. health care system is quickly changing, but is it moving in the right direction? Focusing on care for clinical depression as a test case, this paper summarizes our previously published findings on the effects of various payment strategies, managed care, and primary care gatekeepers on the outcomes and costs for the treatment of mental health conditions. We then synthesize the policy implications of these findings for achieving value of care, lower costs, and good health outcomes.

Subthreshold depression and depressive disorder: clinical characteristics of general medical and mental health specialty outpatients

OBJECTIVE

The authors examined the clinical significance of depressive symptoms below the threshold for depressive disorder in outpatient samples.

METHOD

The subjects were 775 adult patients with current depressive disorder, 1,420 patients with subthreshold depression, and 1,767 hypertensive patients with and without depression, all of whom were visiting the offices of mental health specialists and general medical care providers in three U.S. cities. Data on demographic characteristics, severity of depression, extent of psychiatric and medical comorbidity, family psychiatric history, and treatment history for the patients with depressive disorder and those with subthreshold depression were compared.

RESULTS

The percentage of patients with subthreshold depression who had a family history of depression (41%) was nearly as high as that of the patients with depressive disorder (59%). The two groups of patients had similar levels of medical and psychiatric comorbidity except for anxiety disorders, which were greater among the patients with depressive disorder. Among the hypertensive patients in the general medical sector, those with subthreshold depression were more similar to those with depressive disorder than to the nondepressed hypertensive patients. Treatment rates were considerably lower for patients with subthreshold depression than for patients with depressive disorder in the general medical sector, but they were similar in the mental health specialty sector.

CONCLUSIONS

In these outpatients, subthreshold depression appeared to be a variant of affective disorder and was treated as such in the mental health specialty sector but not in the general medical sector. The findings emphasize the importance of treatment outcome studies of patients with subthreshold depression.

Clinician specialty and treatment style for depressed outpatients with and without medical comorbidities

BACKGROUND

The advent of clinical practice guidelines for the management of depression increases the importance of understanding variation across clinician specialty groups in treatment styles for depression and the role of medical comorbidities.

METHODS

Data are reported by clinicians (N = 470) and patients (N = 2545). Multiple regression was used to compare the treatment styles (counseling and prescribing antidepressants) of family physicians with those of psychiatrists, medical subspecialists, internists, psychologists, and other therapists for depressed patients with different medical comorbidities.

RESULTS

Relative to other primary care specialists, family physicians had the strongest preferences for both counseling and prescribing antidepressants for depressed patients. Family physicians reported preferences for treating with antidepressants that were similar to those of psychiatrists. However, in actual practice, medication use was higher among the patients of psychiatrists than those of family physicians. Mental health care specialists reported the strongest counseling preferences and provided the most counseling in actual practice, compared with general medicine physicians. Internists and subspecialists had similar preferences for prescribing antidepressants, but, compared with internists, subspecialists had lower preferences for counseling. Clinician preferences for counseling were similar for depressed patients with or without medical comorbidities, but preferences for prescribing antidepressants were lowest for patients with depression and myocardial infarction, compared with other patient groups.

CONCLUSIONS

Measures of clinician treatment style for depression are good proxies for counseling but not for prescribing practices. Among general medical physicians, family physicians have the strongest reported preferences in treating depression but, especially in terms of medication therapy, do not always follow those preferences. Preferred treatments of patients with and without medical comorbidities were similar. Findings have implications for improving the quality of treatment of depressed patients.

Switches between prepaid and fee-for-service health systems among depressed outpatients: results from the Medical Outcomes Study

We analyzed switches between prepaid and fee-for-service health care plans among depressed outpatients in the longitudinal part of the Medical Outcomes Study. Patients of mental health specialists in fee-for-service plans had the lowest adjusted rate of plan switching (8.1%), compared to fee-for-service general medical patients (13.5%) and prepaid patients of both types of providers (10.1% to 11.7%). Although there were no substantial differences in initial sickness by payment system among enrolled patients, differing switching rates by provider specialty and payment system indicated biased selection over time. In addition, we found that married, nonwhite, and wealthier individuals were significantly more likely to leave fee-for-service than prepaid care plans. We analyzed whether system switching had an effect on patient satisfaction and outcomes. None of the results were highly significant, but the power of the data to analyze this issue was limited. Nevertheless, it appears that patients switching from prepaid to fee-for-service may be at risk for poorer functioning outcomes, although there was no similar effect on mental health status.

Long-term functioning and well-being outcomes associated with physical activity and exercise in patients with chronic conditions in the Medical Outcomes Study

This study was carried out to determine whether levels of physical activity of patients with various chronic diseases are associated with subsequent functioning and well-being. It was an observational 2-year longitudinal design. The setting was offices of medical and mental health practices within health maintenance organizations, large multispecialty groups, and solo practices or small single-specialty group practices in three U.S. cities. Included in the study were 1758 adult patients with one or more of the following: diabetes, hypertension, congestive heart failure, recent myocardial infarction, depressive symptoms, or current depressive disorder. Outcome measures included physical, role, and functioning; energy/fatigue; pain intensity; sleep problems; depressed affect, anxiety, positive affect, and overall psychological distress/well-being; health distress; and current health perceptions. Cross-sectional (base-line), 2-year endpoint, and change score relationships were evaluated between baseline levels of physical activity and each outcome, controlling for chronic conditions, comorbidity, smoking, alcohol use, overweight, self-reported adherence, and other patient and study characteristics. Higher baseline levels of exercise were uniquely associated with better functioning and well-being at baseline and 2 years later for some measures. The magnitude of the differences varied by disease group, but tended to be between 0.17 and 0.39 of the baseline SD. Greater levels of exercise are associated with feeling and functioning better for patients with chronic conditions over a 2-year period, suggesting that this is a fruitful area for further study using controlled interventions.

Depression in general medical settings. Implications of three health policy studies for consultation-liaison psychiatry

Prepaid or prospective reimbursement has implications for the consultation-liaison (C-L) psychiatrist. The author reviews results from three health policy studies that indicated 1) degree of reliance on general medical providers for mental health care is not affected by generosity of fee-for-service (FFS) coverage, but is greater in some prepaid health care systems; 2) psychological sickness of depressed outpatients visiting general medical providers is similar across prepaid and FFS systems of care; 3) prepaid care is associated with lower rates of detection of depression and counseling in the general medical sector; 4) depression outcomes in the general medical sector are similar under prepaid or FFS care; 5) quality of care for depressed patients is moderate to low in the general medical sector; and 6) depressed elderly inpatients receive higher quality of psychological care in psychiatric units, but they receive higher quality of physical care in general medical wards. The discussion emphasizes the C-L psychiatrist's role in educating general medical providers, improving outcomes for the sickest patients, and improving psychosocial care in prepaid practices.

Use of minor tranquilizers and antidepressant medications by depressed outpatients: results from the medical outcomes study

OBJECTIVE

The purpose of this study was to compare use of minor tranquilizers and antidepressant medications by depressed outpatients across different treatment settings.

METHOD

The study subjects were 634 patients with current depressive disorder or depressive symptoms who visited general medical clinicians, psychiatrists, psychologists, or other therapists. Data on use of medication in different types of clinical practices with different types of payment plans were gathered from structured interviews by study clinicians and from surveys of patients.

RESULTS

Of the depressed patients, 23% had recently used an antidepressant medication and 30% had used a minor tranquilizer. The level of use was similar for different types of depression. Patients of psychiatrists were the most likely to use medications. In the practices of physicians, but not nonphysicians, the more severely distressed patients were more likely to use antidepressant medications. Of the patients taking an antidepressant medication, 39% used an inappropriately low dose. Patients in prepaid health care plans were twice as likely as those in fee-for-service care to use minor tranquilizers.

CONCLUSIONS

Less than one-third of the depressed outpatients used antidepressant medications, and the probability of use was similar for major depression and other types of depression for which efficacy is less well established. Use of antidepressant medications among patients of nonphysicians was unrelated to the level of psychological sickness, suggesting the need for more cooperation among provider groups. Minor tranquilizers were used more often than antidepressants, particularly among patients in prepaid plans, despite controversy over their efficacy.

Quality of care for depressed elderly pre-post prospective payment system: differences in response across treatment settings

We evaluated the quality of care for depressed elderly patients (n = 2,746) hospitalized in general medical hospitals (n = 297) before or after implementation of Medicare's Prospective Payment System, focusing on whether the response to time period differed for hospitals that in the post-PPS period had no psychiatric unit, an exempt psychiatric unit, or a nonexempt unit, and by ward placement within hospitals with psychiatric units. Quality of care increased over time, and for most measures of quality of care the level of improvement did not differ significantly across different types of hospitals or by ward placement. The intensity of use of therapeutic services, such as rehabilitation, occupation, or recreation therapy, increased over time, particularly in nonexempt psychiatric units and hospitals without psychiatric units, such that these locations caught up some over time in the level of use of these services to the level for exempt psychiatric units. Several outcomes of care improved over time, and the degree of improvement in the rate of inpatient medical and psychiatric complications and other outcomes was significantly greater for psychiatric units that were exempt post-PPS than for nonexempt treatment locations.

Emotional and behavioral problems and severe academic delays among sheltered homeless children in Los Angeles County

OBJECTIVES

Few studies have estimated the extent of specific emotional, behavioral, and academic problems among sheltered homeless children. The objectives of this study were to describe such problems, identify those children with the problems, and evaluate the relationship between child problems and use of physical and mental health services.

METHODS

From February through May 1991, 169 school-age children and their parents living in 18 emergency homeless family shelters in Los Angeles County were interviewed. To evaluate the answers, interviewers used standardized measures of depression, behavioral problems, receptive vocabulary, and reading.

RESULTS

The vast majority (78%) of homeless children suffered from either depression, a behavioral problem, or severe academic delay. Among children having a problem, only one third of the parents were aware of any problem, and few of those children (15%) had ever received mental health care or special education.

CONCLUSIONS

Almost all school-age sheltered homeless children in Los Angeles County have symptoms of depression, a behavioral problem, or academic delay severe enough to merit a clinical evaluation, yet few receive specific care. Programs targeted at sheltered homeless school-age children are needed to close this gap.

Do depressed patients in different treatment settings have different levels of well-being and functioning?

Differences in the functioning and well-being of adult patients with current or past depressive disorder who visited clinicians of different specialties in health maintenance organizations, solo practices, or large multispecialty group practices were examined. For patients in different systems, there were no significant differences in functioning and well-being across 12 domains tested. Patients of mental health specialists had worse mental health and more limitations in social activities, whereas patients of medical clinicians had worse physical functioning, more pain, more physical/psychophysiologic symptoms, and worse health perceptions. Thus, each system of care had depressed patients with a similar functioning and well-being "burden" but specialty sectors had patients with slightly different functioning and well-being profiles, probably reflecting patient selection of type of provider.

Quality of antidepressant medications prescribed at discharge to depressed elderly patients in general medical hospitals before and after prospective payment system

This study describes the quality of antidepressant medication use at hospital discharge for depressed elderly inpatients and compares quality of care before and after implementation of Medicare's Prospective Payment System (PPS). The study reviewed data from medical records of 2746 depressed, elderly, hospitalized patients in acute-care general medical hospitals in five U.S. states (pre-PPS period 1981-82; post-PPS period 1985-86). The majority were discharged on antidepressant medication both pre-PPS and post-PPS. After PPS' implementation, sedating medications were used less often in all treatment settings. In general medical wards, a higher percentage post-PPS (24%) than pre-PPS (14%) were discharged 48 hours or less after first starting an antidepressant medication. In both time periods, one-third of patients receiving antidepressant medications were prescribed daily dosages at discharge below recommended, minimum, therapeutic levels, whether treated in general medical wards or psychiatric units. Otherwise, patients previously treated in psychiatric units received higher quality of medication management than those treated in general medical wards. Over time, patients discharged on antidepressant medication were less likely to use sedating medication, suggesting improved quality of care. In general medical wards, however, patients were discharged more rapidly after starting medication, possibly suggesting lower quality of care. A substantial percentage of patients received subtherapeutic dosages of medication or sedating medications, suggesting that improved management of discharge antidepressant medication in the elderly is needed in general medical hospitals.

Quality of care for hospitalized depressed elderly patients before and after implementation of the Medicare Prospective Payment System

OBJECTIVE

The authors evaluated the impact of Medicare's Prospective Payment System on aspects of quality of care and outcomes for depressed elderly inpatients in acute-care general medical hospitals.

METHOD

The depressed elderly inpatients (N = 2,746) were hospitalized in 297 acute-care general medical hospitals. The authors used a retrospective before-and-after design, controlling for differences over time in sickness at admission. Quality of care and outcomes were assessed through clinical review of explicit and implicit information in the medical records; secondary data sources provided information on postdischarge outcomes.

RESULTS

After implementation of the prospective payment system 1) a higher percentage of patients had clinically appropriate acute-care admissions; 2) the initial assessment of psychological status by the treating provider was more complete; 3) the quality of psychotropic medication management, as rated by the study psychiatrists, improved; 4) the rates of any inpatient medical or psychiatric complication, of discharge to another hospital or a nursing home, and of inpatient readmission declined; and 5) there was no marked change in the percentage of patients rated by study clinicians as having acceptable overall clinical status at discharge or the rate of mortality 1 year after admission.

CONCLUSIONS

After the implementation of the Medicare Prospective Payment System, the quality of care for depressed elderly inpatients improved and there was no marked increase in adverse clinical outcomes. Despite these gains, after implementation the quality of care was moderate at best and over one-third of the patients had unacceptable clinical status at discharge.

Prevalence of comorbid alcohol disorder and consumption in medically ill and depressed patients

OBJECTIVE

To estimate the extent to which alcohol disorder co-occurs in patients with major medical and psychiatric conditions.

DESIGN

Observational study.

SETTING

Offices of general medical providers and mental health specialists in three US cities.

PATIENTS

Adult patients (N = 2296) with hypertension, diabetes, heart disease (congestive heart failure or myocardial infarction), and/or current depressive disorder or subthreshold depressive symptoms.

MAIN OUTCOME MEASURES

Current and lifetime alcohol disorder, alcohol consumption, current problem drinking, perceived need for help for alcohol or other drug problems, and unmet need.

METHODS

Comparisons of the prevalence of alcohol comorbidity in medically ill nondepressed patients of general medical providers and in depressed patients of both provider types.

RESULTS

Patients with chronic medical problems or depression had similar levels of lifetime alcohol disorder (14% to 19%) and current alcohol problems (18% to 29%), but depressed patients were more likely to report needing help for problems with alcohol or drugs. Current alcohol disorder was more prevalent among depressed patients in mental health specialty practices than in general medical practices. Many patients who perceived a need for care for alcohol and other drug problems reported that this need was unmet (37% to 84%).

CONCLUSIONS

Clinicians who treat patients with major medical and psychiatric conditions need to be prepared to identify and treat comorbid alcohol disorder.

Do depressed patients in different treatment settings have different levels of well-being and functioning?

Differences in the functioning and well-being of adult patients with current or past depressive disorder who visited clinicians of different specialties in health maintenance organizations, solo practices, or large multispecialty group practices were examined. For patients in different systems, there were no significant differences in functioning and well-being across 12 domains tested. Patients of mental health specialists had worse mental health and more limitations in social activities, whereas patients of medical clinicians had worse physical functioning, more pain, more physical/psychophysiologic symptoms, and worse health perceptions. Thus, each system of care had depressed patients with a similar functioning and well-being "burden" but specialty sectors had patients with slightly different functioning and well-being profiles, probably reflecting patient selection of type of provider.

Outcomes for adult outpatients with depression under prepaid or fee-for-service financing

OBJECTIVE

To compare change over time in symptoms of depression and limitations in role and physical functioning of patients receiving prepaid or fee-for-service care within and across clinician specialties.

METHOD

Observational study of change in outcomes over 2 years for 617 depressed patients of psychiatrists, psychologists, other therapists, and general medical clinicians in three urban sites in the United States.

RESULTS

Psychiatrists treated psychologically sicker patients than other clinicians in all payment types. Among psychiatrists' patients, those initially receiving prepaid care acquired new limitations in role/physical functioning over time, while those receiving fee-for-service care did not. This finding was most striking in independent practice associations but varied by site and organization. Patients of psychiatrists were more likely to use antidepressant medication than were patients of other clinicians, but among psychiatrists' patients, there was a sharp decline over time in the use of such medication in prepaid compared with fee-for-service care. Outcomes did not differ by payment type for depressed patients of other specialty groups, or overall.

CONCLUSION

Depressed patients of psychiatrists merit policy interest owing to their high levels of psychological sickness. For these patients, functioning outcomes were poorer in some prepaid organizations. The nonexperimental evidence favors (but cannot prove) an explanation based on care received, such as a reduction in medications, rather than on preexisting sickness differences.

Alcohol use, depressive symptoms, and cultural characteristics in two Mexican-American samples

We modeled associations of quantity and frequency of alcohol use with depressive symptoms (negative affect, lack of positive affect, somatic disturbance, interpersonal problems) in two household surveys of Mexican-Americans (Ns = 1,313 and 3,577). Multivariate analyses controlled cultural (immigration, acculturation) and demographic (age, income, household size, marital status, employment status) characteristics, and assessed interactions of these two classes of predictors. Alcohol use was inconsistently related to depression. In some analyses, cultural characteristics accounted for associations of alcohol use with depression. In others, associations of alcohol use with depression depended on cultural characteristics. Associations of drinking with depression tended not to be robust across samples.

Course of depression in patients with hypertension, myocardial infarction, or insulin-dependent diabetes

OBJECTIVE

The authors examined the course of depression over 2 years for outpatients with and without a history of hypertension, a history of myocardial infarction, or current insulin-dependent diabetes.

METHOD

Among outpatient visitors to the practices of 523 general medical clinicians and mental health specialists, 626 depressed patients were followed for 1 or 2 years with a telephone-administered interview based on the format of the National Institute of Mental Health Diagnostic Interview Schedule.

RESULTS

Depressed patients with and without medical illness had high rates of persistent depressive symptoms and spells over 2 years. Patients with a lifetime history of myocardial infarction had significantly more spells of depression over the first follow-up year, more total symptoms of depression in the second follow-up year, and more depressive symptoms at the end of each follow-up year than depressed patients without myocardial infarction. The course of depression did not differ significantly for depressed patients with and without a lifetime history of hypertension or current insulin-dependent diabetes.

CONCLUSIONS

Among depressed adult outpatients, a history of myocardial infarction is associated with a particularly poor clinical prognosis. A relatively high percentage of all depressed patients in this study had persistent depression regardless of the extent of medical comorbidity.

Risk factors for secondary depression among Mexican Americans and non-Hispanic whites. Alcohol use, alcohol dependence, and reasons for drinking

We evaluated demographic (age, gender, income), cultural (ethnicity, acculturation), clinical (alcohol use, alcohol dependence), and motivational (subjective reasons for drinking) potential risk factors for secondary depression in 372 persons with lifetime alcohol abuse or dependence (from a randomly selected community sample of 2393). Lifetime alcohol abuse or dependence increased the risk of major depression by a factor of approximately 2 to 7. Recent secondary depression was present in 2% to 18% of persons meeting criteria for alcohol use disorders. Low income, female gender, and among Mexican Americans, low acculturation were associated with increased risk of secondary depression. Persons with lifetime alcohol diagnoses who currently drank, but did not name relaxation as a reason for drinking, were also at higher risk for secondary depression. Drinking to forget was associated with increased risk of secondary depression among Mexican American alcoholics, but not among non-Hispanic white alcoholics. Current abstinence was associated with greater risk among lifetime alcoholics born in Mexico, but not among those born in the United States.

Partial hospitalization (day treatment) for psychiatrically ill elderly patients

OBJECTIVE

The purpose of this paper is to present initial findings from a retrospective chart review of geriatric day treatment patients in order to focus attention on this potentially important area, add to the limited database in this area, and generate hypotheses for future investigations.

METHOD

Data were abstracted from the charts of 100 geriatric day treatment patients over a period of approximately 5 years (1985-1989). Descriptive, univariate, and multiple regression techniques were used to describe the patients and identify variables associated with their outcomes.

RESULTS

The typical patient in this program was a widowed white woman in her 70s who suffered from a depressive disorder. During the initial treatment period (usually approximately 3 months), 57% of the patients experienced some clinical improvement. Variables associated with a favorable outcome included diagnosis of a mood disorder rather than a psychotic disorder, better initial functional status, greater initial social support, fewer stressful events during treatment, and longer duration of treatment.

CONCLUSIONS

Geriatric day treatment can be effective and merits further study as a mode of treatment for psychiatrically ill elderly patients.

The course of depression in adult outpatients. Results from the Medical Outcomes Study

OBJECTIVE

To compare the course of depression during a 2-year period in adult outpatients (n = 626) with current major depression, dysthymia, and either both current disorders ("double depression") or depressive symptoms with no current depressive disorder.

METHODS

Depressed patients visiting 523 clinicians (mental health specialists and general medical providers) were identified using a two-stage screening procedure including the Diagnostic Interview Schedule. The course of depression was assessed in 2 follow-up years with a structured telephone interview based on the format of the Diagnostic Interview Schedule.

RESULTS

Baseline severity of depressive symptoms was greatest in patients with double depression, but initial functional status was poor in those with dysthymia with or without concurrent major depression. Patients with dysthymia had the worst outcomes, those with current major depression alone had intermediate outcomes, and those with subthreshold depressive symptoms had the best outcomes. Even the latter group, however, had a high incidence (25%) of major depressive episode over 2 years. Initial depression severity and level of functional status accounted for more explained variance in outcomes than did type of depressive disorder.

CONCLUSIONS

The findings emphasize the poor prognosis associated with dysthymia even in the absence of major depression; the prognostic significance of subthreshold depressive symptoms; and the clinical significance of assessing level of severity of symptoms as well as functional status and well-being, regardless of type of depressive disorder.

Clinical factors associated with better quality of life in a seriously mentally ill population

Improving the quality of life of persons with chronic mental illness is becoming an important treatment goal. In this study, 101 former acute care psychiatric inpatients with serious mental illness who were living in Mississippi communities were interviewed using portions of Lehman's Quality of Life Interview. A particular focus was whether clinical characteristics, such as medication compliance and social skills, that could be changed by interventions were associated with patients' ratings of their quality of life. Self-reports of better quality of life were associated with fewer depressive symptoms, fewer medication side effects, and better family interactions. Results indicate that clinical interventions to improve quality of life in this population should include family psychoeducational programs and better detection, evaluation, and treatment of both depressive symptoms and side effects of medication.

The effects of psychological distress and psychological well-being on use of medical services

In this study, the authors determined whether mental health status affects the use of general medical services, with and without adjustment for the correlated effects of general health perceptions and physical health status on such use. Data were used from the RAND Health Insurance Experiment, which has information on up to 5 years of use of medical services by a nonelderly, civilian, general population. Health status and other covariates were assessed by self-administered questionnaires at enrollment. In the absence of statistical control for general and physical health status, worse mental health status-whether assessed by a global self-report measure or its two component parts, psychological well-being and psychological distress-significantly increased the use of both inpatient and outpatient general medical services. After controlling for general health perceptions, physical health status, demographic factors, and insurance plan coverage, the effects of mental health status on use are reduced, but not eliminated. Psychological distress and psychological well-being retained independent effects on total medical expenses.

The effects of preferred provider options in fee-for-service plans on use of outpatient mental health services by three employee groups

Descriptions of how preferred provider organizations (PPOs), offered as options to employees enrolled in fee-for-service plans, affected use of outpatient mental health services are provided. Data are from the RAND Preferred Provider Organization Study, which has a sample of employees who enrolled in fee-for-service plans 1 year before and 2 years after a PPO option was offered by three employers in two U.S. sites. To study effects of the optional PPOs on access to mental health care, usage patterns among those who initially stated that they did or did not intend to use PPO providers were examined. By the end of the second post-PPO year, employees had a similar annual probability of having an outpatient mental health visit whether or not they initially intended to use PPO providers. However, during the first post-PPO year, there was a decrease in the probability of use for those initially intending to use PPO providers, relative to those who did not intend to do so, among employees who had no regular medical provider. To study effects of the PPO option on usage levels of mental health care services, users of mental health services who primarily visited PPO were compared with those who primarily visited non-PPO providers. Users who visited PPO providers had significantly lower levels of use, controlling for other factors, than those who primarily visited non-PPO providers. Therefore, despite lower cost sharing for services received from PPO providers, the PPO option appeared to lower outpatient mental health care costs while having no more than a transient effect on access. This study did not evaluate mental health outcomes.

How the medical comorbidity of depressed patients differs across health care settings: results from the Medical Outcomes Study

OBJECTIVE

Although depression is one of the most common problems of medical and psychiatric outpatients, it has not been clear whether the extent of medical comorbidity among depressed patients varies across major types of clinical settings in which depressed patients receive care--especially by type of treating clinician (general medical versus mental health specialty) or type of payment for services (prepaid versus fee-for-service).

METHODS

The authors examined these issues using data on 1,152 adult outpatients with current depressive symptoms and a lifetime history of unipolar depressive disorder who received care in one of three health care delivery systems in three U.S. sites.

RESULTS

Depressed patients had a similarly high prevalence (64.9%-71.0%) of any of eight common chronic medical conditions whether they were seen in the general medical or specialty mental health sector; however, those visiting medical clinicians had a significantly higher prevalence of the two most common chronic medical conditions, hypertension and arthritis. Among depressed patients with hypertension, those visiting the general medical sector were more likely to be taking antihypertensive medication than were those visiting the mental health specialty sector. Type of payment (prepaid versus fee-for-service) was unrelated to either prevalence or severity of comorbid medical conditions, suggesting that the typical depressed patient in all types of practices studied had medical comorbidity.

CONCLUSIONS

These data suggest that clinicians in all health care settings must be prepared to encounter chronic medical conditions and complaints in the depressed patients who visit them.

Mental health and selection of preferred providers. Experience in three employee groups

While Preferred Provider Organizations (PPOs) are designed to contain the costs of health care, they may not be able to do so if sicker individuals opt not to use PPO providers. This study examined how level of mental health status and prior use of mental health services affected the decision to use or not use PPO providers for mental health care for employees enrolled in fee-for-service plans with a PPO option. Data were obtained from an employee survey and claims data on three large employee groups. It was not possible to examine effects of sickliness on the intent to select PPO providers for mental health care directly because about one half of employees could not identify who they would visit for mental health care or even how they would select a provider for such care. The intent to use PPO or non-PPO providers for general medical care, however, was not significantly associated with mental health status when other factors were controlled. Furthermore, among persons who used mental health services after implementation of the PPO option, those who had previously visited providers who were to become part of the PPO panel tended to stay with PPO providers, while those who previously visited providers who were not to enter the PPO panel subsequently selected away from PPO providers for mental health care. This pattern of results suggests that established individual patient-provider relationships, rather than sickliness, determined the selection of PPO versus non-PPO providers for mental health care for employees enrolled in these optional PPO fee-for-service plans.

How do HMOs reduce outpatient mental health care costs?

OBJECTIVE

Because previous studies of differences in utilization of mental health care services have had important limitations, it is not clear if their findings that health maintenance organization (HMO) outpatient mental health care costs less than fee-for-service care are due to less access, less intensive care per user, or selective enrollment of healthier people by HMOs. Therefore, the authors used data from the National Institute of Mental Health Epidemiologic Catchment Area (ECA) study to examine differences in the prevalence of psychiatric disorder and differences in the use of outpatient mental health services for adults enrolled in HMO or fee-for-service health insurance plans.

METHOD

The subjects were an ECA community sample obtained from East Los Angeles and West Los Angeles. This sample included a large number of Hispanic subjects. The subjects were categorized according to their responses to a 5-item battery on insurance as Medicare enrolles, members of private fee-for-service plans, Medicaid enrollees, members of an HMO, and uninsured. The presence or absence of psychiatric disorders was determined by using the NIMH Diagnostic Interview Schedule. Both users and nonusers of mental health services were studied.

RESULTS

The HMO and fee-for-service plans had similar prevalence of psychiatric disorder and similar access to specialty mental health care. However, HMO enrollees had significantly fewer visits per user to providers of specialty care.

CONCLUSIONS

The most likely explanation for lower mental health care costs in HMOs is a less intensive style of care for a comparably sick population.

The effects of a prepaid group practice on mental health outcomes

Does a prepaid group practice relative to comparable fee-for-service plans lead to different mental health outcomes for its beneficiaries? To answer this question, we used data from the RAND Health Insurance Experiment. We observed no statistically significant or clinically meaningful differences in mental health outcomes for families randomly assigned to Group Health Cooperative of Puget Sound or to comparable fee-for-service insurance plans in the Seattle area. We found the same null result for overall mental health status as well as for psychological distress (e.g., anxiety and depression) and psychological well-being, and for the full population as well as the initially sick and poor, although our precision was low for the latter comparisons. Thus, the less intensive style of treatment in the prepaid group practice was not associated with noticeably worse mental health outcomes.

Patterns of outpatient mental health care over time: some implications for estimates of demand and for benefit design

The article examines patterns of starting and continuing outpatient mental health care as a function of time, and the implications of these patterns for estimates of the response of demand to generosity of fee-for-service insurance coverage. The data are from the RAND Health Insurance Experiment (HIE), which acquired a random sample of the nonelderly general population in six U.S. sites. People rarely had more than one episode of use of outpatient mental health services in a year. Persons who used in the prior year had high rates of continuing in treatment, while those without prior use entered treatment at a low, steady rate. Similar patterns of use by former users and nonusers were observed across insurance plans that varied widely in generosity, but the absolute probabilities of use were significantly lower in less generous plans. The probability of use of mental health services expanded significantly over time in the HIE; thus, estimates of demand in a steady state would be higher than those based on the HIE study years.

A structured interview version of the Hamilton Depression Rating Scale: evidence of reliability and versatility of administration

A structured interview version of the Hamilton Depression Rating Scale (SI-HDRS) is described. Data are presented in support of its inter-rater and internal consistency reliability. SI-HDRS scores were reproducible by trained interviewers who lacked psychiatric backgrounds. Test-retest scores of a subset of patients who were interviewed twice (once in person and once by telephone) were highly correlated. Scores on the SI-HDRS did not differ between face-to-face and telephone administration groups, controlling for demographic factors, depression-specific indicators, and social and physical functioning. Since the SI-HDRS does not require a face-to-face interview by an experienced clinician, this instrument can be used economically in large-scale, community-based research projects.

Alcohol use and limitations in physical functioning in a sample of the Los Angeles general population

The authors examined the relationships between drinking and perceived current health and physical functioning for a general household sample of Mexican Americans and non-Hispanic whites. These relationships differed by sex and by the presence or absence of medical and psychiatric comorbidity, but not by ethnicity. Among men with a chronic medical illness, current abstinence was uniquely associated with poor current health and physical functioning, especially when current abstinence was combined with a past history of alcohol disorder. Among men without a chronic medical illness, a history of alcohol disorder (irrespective of current drinking) was uniquely associated with poorer functioning. For women, among the medically or psychiatrically ill, drinking was not strongly associated with physical functioning; while among women without chronic medical or psychiatric illness, a history of drinking was uniquely associated with poor physical functioning. The authors interpret the findings in terms of adverse effects of drinking and chronic medical conditions on functioning and the tendency of physically limited and chronically medically ill persons to stop drinking.

Detection of depressive disorder for patients receiving prepaid or fee-for-service care. Results from the Medical Outcomes Study

We estimated clinicians' awareness of depression for patients with current depressive disorder (N = 650) who received care in either a single-specialty solo or small group practice, a large multispecialty group practice, or a health maintenance organization in three US sites. Depressive disorder was determined by independent diagnostic assessment shortly after an office visit. Detection and treatment of depression were determined from visit-report forms completed by the treating clinician. Depending on the setting, from 78.2% to 86.9% of depressed patients who visited mental health specialists had their depression detected at the time of the visit, compared with 45.9% to 51.2% of depressed patients who visited medical clinicians, after adjusting for case-mix differences. Among patients of mental health specialists, there were no significant differences by type of payment in the likelihood of depressive disorder being detected or treated. Among patients of medical clinicians, however, those receiving care financed by prepayment were significantly less likely to have their depression detected or treated during the visit than were similar patients receiving fee-for-service care.

Chronic medical conditions in a sample of the general population with anxiety, affective, and substance use disorders

The authors studied the prevalence of eight chronic medical conditions in an adult population sample (N = 2,554) with and without psychiatric disorders. Adjusted for age and sex, the prevalence of any lifetime chronic medical condition for persons with any lifetime affective, anxiety, or substance use disorder was 61.4%, 57.1%, and 57.7%, respectively. Each of these percentages was significantly higher than that for persons with no lifetime psychiatric disorder (53.4%). Both lifetime affective and anxiety disorders were uniquely associated with a greater prevalence of any lifetime chronic medical condition, but the only psychiatric disorders uniquely associated with current (i.e., active) chronic medical conditions were anxiety disorders, suggesting that the association between anxiety disorders and chronic medical conditions develops more quickly than associations between medical conditions and other psychiatric disorders.

Affective, substance use, and anxiety disorders in persons with arthritis, diabetes, heart disease, high blood pressure, or chronic lung conditions

The authors estimated the sex- and age-adjusted prevalence of affective, substance use, and anxiety disorders in persons in a general population sample who identified themselves as having arthritis, diabetes, heart disease, high blood pressure, chronic lung disease, or no chronic medical conditions. Persons who reported ever having arthritis, heart disease, chronic lung disease, or high blood pressure had a significantly increased adjusted prevalence of each of the three groups of lifetime psychiatric disorders, relative to a no-chronic conditions comparison group (each p less than 0.05). Persons who ever had diabetes had an increased adjusted prevalence of lifetime affective and anxiety but not substance use disorder. Persons with current (i.e., active) arthritis, heart disease, or high blood pressure had a significantly increased adjusted prevalence of recent (6-month) anxiety disorder, whereas those with current chronic lung disease had an increased adjusted prevalence of recent affective and substance use but not anxiety disorder.

Acculturation and the probability of use of health services by Mexican Americans

How does level of acculturation affect the probability that Mexican Americans use general health, mental health, and human social services? We studied this question using data from a general population sample of Mexican Americans (N = 1,055). Data were elicited in face-to-face interviews. After controlling for sociodemographic and economic factors, health status, and insurance coverage, Mexican Americans who were less acculturated had significantly lower probabilities of an outpatient medical visit for physical health problems and of a visit to a mental health specialist or human service provider for emotional problems. The less acculturated with good perceived general health were especially unlikely to receive outpatient medical care. Having Medicaid coverage was associated with a larger increase in the probability of an inpatient medical admission for the more acculturated than for the less acculturated. Other individual characteristics had generally similar effects on use of medical and mental health services for both the more and the less acculturated Mexican Americans.

The effects of insurance generosity on the psychological distress and psychological well-being of a general population

Reductions in the generosity of fee-for-service insurance lower the use of general medical and mental health services, but do they lead to lower mental health status for the covered population? We addressed this question using data from the RAND Corporation Health Insurance Experiment. Families in six sites in the United States were randomly assigned to one of 14 insurance plans for three- or five-year periods. On average, there were no significant adverse effects of cost-sharing plans, relative to a free-care plan, on either psychological well-being or psychological distress, when the cost-sharing plans included full catastrophic coverage. Those with high mental health status but low income at baseline had significantly more favorable mental health outcomes on the cost-sharing plans than on the free-care plan. We cannot definitively comment on the effects of insurance generosity for the sick poor. Our findings apply in the context of mandated comprehensive mental and general health coverage for a general nonelderly, nondisabled household population.

The demand for episodes of mental health services

Observational studies of demand for mental health services showed much greater use by those with more generous insurance, but this difference may have been due to adverse selection, rather than in response to price. This paper avoids the adverse selection problem by using data from a randomized trial, the RAND Health Insurance Experiment (HIE). Participating families were randomly assigned to insurance plans that either provided free care or were a mixture of first dollar coinsurance and free care after a cap on out-of-pocket spending was reached. We estimate that separate effects of coinsurance and the cap on the demand for episodes of outpatient mental health services. We find that outpatient mental health use is more responsive to price than is outpatient medical use, but not as responsive as most observational studies have indicated. Those with no insurance coverage would spend about one-quarter as much on mental health care as they would with free care. Coinsurance reduces the number of episodes of treatment, but has only a small effect on the duration and intensity of use within episodes. Users appear to anticipate exceeding the cap, and spend at more than the free rate after they do so.

Development of a brief screening instrument for detecting depressive disorders

A very short (8-item), self-report measure was developed to screen for depressive disorders (major depression and dysthymia). The screener departs from traditional depressive symptom scales in that 1) individual items are differentially weighted and 2) two of the eight items concern diagnostically-relevant durations of depressed mood. Analyses of data from a general population and from primary care and mental health patients showed that the screener had high sensitivity and good positive predictive value for detecting depressive disorder, especially for recent disorders and those that met full DSM-III criteria. The high predictive utility of the screener, in combination with its brevity, suggests that it may be a useful tool for screening for depression in health care settings.