Quality of Follow-Up After Hospitalization for Mental Illness Among Patients From Racial-Ethnic Minority Groups
Abstract
Objective
Outpatient follow-up after hospitalization for mental health reasons is an important indicator of quality of health systems. Differences among racial-ethnic minority groups in the quality of service use during this period are understudied. This study assessed the quality of outpatient treatment episodes following inpatient psychiatric treatment among blacks, whites, and Latinos in the United States.
Methods
The Medical Expenditure Panel Survey (2004–2010) was used to identify adults with any inpatient psychiatric treatment (N=339). Logistic regression models were used to estimate predictors of any outpatient follow-up or the beginning of adequate outpatient follow-up within seven or 30 days following discharge. Predicted disparities were calculated after adjustment for clinical need variables but not for socioeconomic characteristics, consistent with the Institute of Medicine definition of health care disparities as differences that are unrelated to clinical appropriateness, need, or patient preference.
Results
Rates of follow-up were generally low, particularly rates of adequate treatment (<26%). Outpatient treatment prior to inpatient care was a strong predictor of all measures of follow-up. After adjustment for need and socioeconomic status, the analyses showed that blacks were less likely than whites to receive any treatment or begin adequate follow-up within 30 days of discharge.
Conclusions
Poor integration of follow-up treatment in the continuum of psychiatric care leaves many individuals, particularly blacks, with poor-quality treatment. Culturally appropriate interventions that link individuals in inpatient settings to outpatient follow-up are needed to reduce racial-ethnic disparities in outpatient mental health treatment following acute treatment.
Hospitalization serves a crucial function in mental health systems by providing treatment for individuals with acute psychiatric needs. Timely follow-up after hospitalization can reduce the duration of disability and, for certain conditions, the likelihood of rehospitalization (1–3). For these reasons, the time between inpatient discharge and outpatient follow-up is considered an important indicator of health system quality. For example, the National Committee for Quality Assurance (NCQA) reports follow-up within seven or 30 days after psychiatric hospitalization in the Healthcare Effectiveness Data and Information Set (HEDIS), and these indicators are used nationally to assess quality and continuity of mental health care (4). These service use data are reported by health organizations to the NCQA and represent how these organizations are improving on important health outcomes over time. In 2011, the rate of follow-up care within 30 days of inpatient discharge ranged from 56% among patients enrolled in Medicare to 77% among patients with commercial insurance plans (5).
In the United States, individuals from racial-ethnic minority groups frequently use inpatient psychiatric services (6,7). Such individuals, particularly black patients, are at risk of poor follow-up, although few studies of service use among racial-ethnic groups have described patterns of aftercare beyond the initial follow-up visit, and most have focused on regional Medicaid data or the elderly (8,9). Studying the quality of treatment episodes after psychiatric hospitalization is especially relevant for racial-ethnic minorities because people in these groups suffer greater persistence and severity of illness (10,11). For example, this approach revealed that black individuals were less likely than whites to receive adequate depression treatment during treatment episodes after a psychiatric hospitalization (12). Predictors of racial-ethnic disparities in follow-up after hospitalization include public insurance, co-occurring substance use and mental disorders, lack of follow-up care, and psychiatric treatment prior to hospitalization (9,13–16).
Health care disparities have been measured by comparing differences in unadjusted means (17), by interpretation of race coefficients determined by regression modeling (18–21), and by model-based estimations of disparities (22–26). The latter approach, which was the one used in this study, uses the Institute of Medicine (IOM) definition of disparities as any difference in health care that is unrelated to clinical appropriateness, need, or patient preferences (27). Treatment differences that are due to these three factors are justifiably excluded from the estimate of disparity (28), although information about preferences and clinical need is rarely available in national data sets.
The IOM definition suggests that normative differences in illness severity should not contribute to the disparity in care. In the context of follow-up care after hospitalization, differences in severity among racial-ethnic groups beyond those that led to inpatient admission should not contribute to the disparity. For example, if one group has a higher burden of illness that reduces their ability to seek follow-up care, related treatment differences should not be considered part of the disparity.
In contrast, treatment differences that are due to the operation of health care systems or the legal or regulatory climate should be considered factors in the disparity, because they unfairly decrease access by racial-ethnic minorities. Differences that are due to discrimination are also considered part of the disparity, according to the IOM. In regression models of survey data, independent effects of race-ethnicity on service use can be considered a proxy for discrimination (29).
In this study, we implemented a conceptually based definition of health care disparities to assess differences in quality of follow-up after hospitalization among members of racial-ethnic groups. Quality of follow-up was measured by rates of any follow-up and of adequate treatment after discharge. On the basis of available literature, we hypothesized that black and Latino patients received less follow-up (both any and adequate) after inpatient discharge compared with non-Latino white patients.
Methods
Source of data
We analyzed data from adults age 18 and older in the Medical Expenditure Panel Survey (MEPS) (30), a nationally representative survey of health care utilization by the noninstitutionalized U.S. population. The survey features a panel design, in which respondents participate in several rounds of interviewing over the course of two full calendar years. We combined six two-year panels (2004–2010), in which respondents were interviewed in five rounds per panel about medication prescriptions, provider visits, and inpatient stays. Diagnoses for mental health care “events” were translated to ICD-9 codes 291, 292, and 295–314. Events included office-based visits, for example, psychotherapy visits; partial hospitalizations; or prescription fills associated with mental health or substance use disorder diagnoses (31). Dates for inpatient, office-based, and outpatient visits and initial prescription fills were identifiable in the data. For subsequent prescription fills, the round of interviewing, not the exact date of the fill, was indicated. We imputed dates of prescription fills by using an established method (32). The study received approval from the institutional review board at the Cambridge Health Alliance.
Measures
Treatment episodes.
Treatment episodes started with hospitalization and ended at the last outpatient visit before a gap of 12 weeks or more. The first two dependent variables were HEDIS measures (4): any follow-up within seven days and within 30 days of discharge. The second two were extensions of the HEDIS measures: minimally adequate care beginning within seven days and within 30 days of discharge. Adequate treatment was defined as a treatment episode consisting of at least eight visits to a mental health provider or at least four mental health care visits or events and at least one psychotropic medication fill (33–36). This definition is consistent with recommendations for treatment type and duration by the American Psychiatric Association (37). For those with psychotic disorders, only the latter criterion was applied (34).
Race-ethnicity.
Racial-ethnic categories were based on U.S. Census definitions and included non-Latino white, non-Latino black (includes those identifying as African American), and Latino (includes those identifying as Hispanic). Asian Americans and Native Americans were not included because of small samples.
Clinical need variables.
Clinical need was measured by the Patient Health Questionnaire–2 (PHQ-2) depression checklist; the K6 measure of psychological distress; the mental and physical component summary scales of the 12-Item Short-Form Health Survey (SF-12), version 2; self-reported mental health; and number of psychiatric inpatient nights. The PHQ-2 is sensitive (93%) and specific (75%) for any depressive disorder (38), whereas the K6 is predictive of severe mental illness (39). Because comorbid general medical illness is predictive of mental illness severity (40–43), we also defined clinical need as any limitation due to general medical or mental health and self-reported “fair” or “poor” mental health. Sex, age, and marital status were considered proxies for clinical need, given the large differences in mental illness prevalence within these categories (44). Other indicators of need were number of hospitalizations during the study period and outpatient treatment prior to hospitalization.
“Non-need” variables.
“Non-need” variables were income, education, health insurance, participation in a health maintenance organization (HMO), region, employment status, and metropolitan statistical area residence.
Statistical analysis
Racial-ethnic differences in clinical need and non-need variables were measured by using chi square tests for dichotomous variables, such as sex, and t tests for continuous variables, such as mental health measures. Correlates of follow-up were examined with multivariate logistic regression. Individuals with multiple hospitalizations during a MEPS panel could have multiple episodes of care. Therefore, the nonindependence of individuals with multiple episodes of care was accounted for by using the balanced repeated replication (BRR) method of computing standard errors in disparity predictions. The panel number was entered into regression models to adjust for secular trends.
Implementation of the IOM definition of health care disparities (27) involved an adjustment for differences in service use due to clinical appropriateness, need, and patient preferences. Differences due to system-level factors, which are considered to be unjustifiable, were counted in the disparity estimation (24,28).
Implementing the IOM definition required a four-step process, beginning with the estimation of a logistic regression model of quality of care, conditional on both need and system-level covariates. Next, we used a rank-and-replace method to adjust only the distributions of clinical need variables in order to balance them across racial-ethnic groups (24,45). Predictions for quality of care among blacks, and then Latinos, were made by using the original coefficients from the model in step 1, the transformed clinical need characteristics, and the original non-need characteristics. In step 4, predicted rates of quality of care among blacks and Latinos were compared with those of whites to compute disparities. Standard errors for disparity estimates were generated with BRR estimation (46) from BRR samples by the Agency for Healthcare Research and Quality (47).
Whereas the logistic regression model in step 1 adjusted for clinical need and non-need variables, the analysis in step 4 adjusted only for clinical need variables. A comparison of the results of these models, therefore, provides information about the role of socioeconomic status in determining racial-ethnic disparities.
We estimated disparities both with and without adjustment for outpatient treatment in the three months preceding hospitalization. Not adjusting for prior treatment assumes that hospitalized individuals needed such treatment and that lack of access reflects a failure of the health system. Differences in this variable between racial-ethnic groups would, therefore, enter the disparity estimation. Conversely, adjusting for prior treatment assumes it is a justifiable indicator of need, with more severely ill individuals more likely to have had prior treatment.
Results
Our sample included 263 treatment episodes involving 206 white patients, 104 episodes involving 78 black patients, and 65 episodes involving 55 Latino patients, with a number of individuals from each racial-ethnic group having multiple treatment episodes during the two-year data collection period.
Table 1 reports the rates of follow-up, severity of illness, and sociodemographic information among white, black, and Latino patients. Across all patient groups, rates of follow-up ranged from 16% to 22% for any outpatient visit within seven days, 11% to 14% for adequate treatment beginning within seven days, 29% to 51% for any outpatient visit within 30 days, and 17% to 26% for adequate treatment beginning within 30 days. Differences in these variables by racial-ethnic group were not statistically significant.
| White (N=206) | Black (N=78) | Latino (N=55) | |||
|---|---|---|---|---|---|
| Variable | % | % | pa | % | pa |
| Dependent | |||||
| Follow-up treatment within 7 days of inpatient discharge | |||||
| Outpatient visit | 21.3 | 15.8 | .48 | 22.2 | .92 |
| Adequate treatment begun | 14.2 | 11.7 | .73 | 11.2 | .62 |
| Follow-up treatment within 30 days of inpatient discharge | |||||
| Outpatient visit | 39.7 | 28.9 | .19 | 51.0 | .19 |
| Adequate treatment begun | 25.6 | 16.7 | .31 | 25.5 | .99 |
| Need related | |||||
| Mental health | |||||
| K6 score (M±SD)b | 9.2±.5 | 9.2±.9 | .96 | 9.8±1.1 | .61 |
| PHQ-2 score (M±SD)c | 2.4±.2 | 2.1±.2 | .25 | 2.6±.3 | .58 |
| SF-12 mental component summary scale score (M±SD)d | 38.0±1.2 | 38.1±2.2 | .95 | 41.0±1.9 | .19 |
| Fair or poor mental health (self-rated) | 46.7 | 46.7 | 1.00 | 34.4 | .17 |
| Psychotic disorder diagnosis during inpatient admission | 6.7 | 24.8 | .003 | 4.0 | .52 |
| Physical health | |||||
| ≥2 comorbid general medical illnesses | 43.5 | 44.5 | .89 | 42.7 | .93 |
| SF-12 physical component summary scale score (M±SD)d | 44.7±1.0 | 42.8±2.0 | .38 | 43.3±2.0 | .49 |
| Any limitation in work, housework, or school | 58.5 | 63.6 | .51 | 52.5 | .53 |
| Other | |||||
| Inpatient nights (M±SD) | 7.3±.7 | 8.1±1.3 | .61 | 6.8±1.3 | .73 |
| Female | 54.7 | 58.9 | .61 | 49.1 | .57 |
| Age | |||||
| 18–34 | 22.9 | 24.5 | .84 | 33.0 | .26 |
| 35–64 | 59.2 | 66.6 | .36 | 56.7 | .81 |
| ≥65 | 17.9 | 9.0 | .05 | 10.3 | .25 |
| Married | 39.4 | 17.2 | .002 | 43.6 | .65 |
| Office-based visit for mental health care before hospitalization | 27.0 | 27.2 | .98 | 29.3 | .79 |
| ≥2 inpatient episodes during the study period | 20.5 | 22.0 | .82 | 10.3 | .14 |
| Non-need related | |||||
| Socioeconomic status | |||||
| Income ≥200% of the federal poverty level | 52.3 | 30.4 | .01 | 30.8 | .03 |
| Education less than high school graduate | 20.3 | 21.9 | .80 | 46.7 | .002 |
| Other | |||||
| Health insurance | |||||
| Private | 45.6 | 29.4 | .05 | 25.7 | .05 |
| Uninsured | 16.6 | 14.9 | .76 | 20.3 | .61 |
| Any public | 37.8 | 55.8 | .03 | 54.0 | .09 |
| HMO | 25.7 | 32.3 | .35 | 39.5 | .11 |
| Employed | 38.7 | 37.4 | .87 | 31.8 | .47 |
| Resides in Northeast | 16.9 | 17.5 | .92 | 29.9 | .10 |
| Urbanicity | 74.5 | 86.6 | .11 | 95.7 | <.001 |
a
For comparisons with white patients
b
Possible scores range from 0 to 24, with higher scores indicating greater psychological distress.
c
Possible scores on the Patient Health Questionnaire–2 range from 0 to 6, with a score above 3 indicating probable depressive disorder.
d
From the 12-Item Short Form of the Medical Outcomes Study. Possible scores range from 0 to 100, with higher scores indicating higher levels of mental health or physical health relative to individuals of similar age.
Approximately one-third to one-half reported fair or poor mental health, and greater than half reported a limitation in work, housework, or school. Scores on the K6, PHQ-2, and SF-12 were suggestive of moderate psychological distress (38,48,49). Relative to whites, blacks were significantly more likely to receive a psychotic disorder diagnosis while hospitalized, to have low family income, and to have public insurance. Latinos were significantly more likely than whites to have lower income and to live in urban areas and significantly less likely to be high school graduates. No racial-ethnic differences in outpatient treatment preceding inpatient care or in percentage of multiple inpatient episodes were identified.
Compared with whites, blacks were half as likely to receive any follow-up within 30 days of inpatient discharge (odds ratio [OR]=.45), and were one-third as likely to receive adequate care beginning within 30 days of discharge (OR=.36) (Table 2). Several clinical need variables were also predictive of follow-up, according to the regression models. Having two or more comorbid general medical illnesses and being age 35–64 years old (as compared with younger adults) were significantly positively predictive of some measures of follow-up care. For non-need variables, greater income, higher education, and employment were significant positive predictors of some measures of follow-up care. Outpatient mental health service use prior to inpatient treatment was a consistently strong and significant predictor of any and adequate treatment across both time periods, with ORs ranging from 4.29 to 8.45. Finally, a one-unit increase in MEPS panel enrollment was associated with decreased likelihood of receiving any treatment (OR=.79) or beginning adequate treatment (OR=.65) within seven days of discharge.
| Within 7 days of discharge | Within 30 days of discharge | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Any follow-upa | Adequate treatment begunb | Any follow-upa | Adequate treatment begunb | |||||||||
| Variable | OR | CI | p | OR | CI | p | OR | CI | p | OR | CI | p |
| Race-ethnicity (reference: white) | ||||||||||||
| Black | .46 | .17–1.23 | .12 | .49 | .14–1.71 | .26 | .45 | .22–.91 | .03 | .36 | .13–.97 | .04 |
| Latino | 1.06 | .26–4.41 | .93 | .63 | .13–3.20 | .58 | 1.75 | .66–4.67 | .26 | .90 | .27–2.98 | .86 |
| Need related | ||||||||||||
| Mental health | ||||||||||||
| Severe psychological distress (reference: no distress) | 1.07 | .96–1.19 | .23 | 1.06 | .93–1.21 | .38 | 1.02 | .94–1.11 | .64 | 1.03 | .93–1.14 | .52 |
| Probable depressive disorder (reference: no disorder) | .76 | .56–1.04 | .09 | .81 | .53–1.25 | .34 | .99 | .76–1.29 | .96 | 1.02 | .74–1.41 | .89 |
| SF-12 mental component summary scale score | .98 | .94–1.03 | .37 | .97 | .92–1.03 | .30 | .98 | .94–1.02 | .25 | .97 | .92–1.02 | .30 |
| Fair or poor mental health (self-rated) (reference: good, very good, or excellent) | .86 | .35–2.06 | .73 | 1.03 | .39–2.74 | .95 | .77 | .40–1.48 | .43 | .71 | .31–1.66 | .42 |
| Psychotic disorder diagnosis during inpatient admission (reference: no psychotic disorder) | .45 | .12–1.73 | .24 | .53 | .13–2.21 | .38 | 1.10 | .48–2.53 | .82 | 1.11 | .40–3.06 | .84 |
| Physical health | ||||||||||||
| ≥2 comorbid medical illnesses (reference: <2) | 2.96 | 1.31–6.69 | .01 | 3.19 | 1.20–8.49 | .02 | 1.90 | 1.02–3.53 | .04 | 1.45 | .74–2.87 | .28 |
| SF-12 physical component summary scale score | .98 | .95–1.01 | .14 | 1.00 | .96–1.04 | .92 | 1.01 | .98–1.03 | .68 | 1.01 | .98–1.05 | .54 |
| Any limitation in work, housework, or school (reference: none) | .68 | .28–1.60 | .37 | 1.02 | .35–3.00 | .97 | 1.19 | .62–2.30 | .59 | 2.25 | 1.01–5.04 | .048 |
| Other | ||||||||||||
| Number of inpatient nights | .97 | .93–1.01 | .11 | .97 | .92–1.02 | .27 | .98 | .95–1.01 | .13 | .98 | .95–1.02 | .36 |
| ≥2 inpatient episodes (reference: <2) | 1.50 | .74–3.03 | .26 | 1.16 | .43–3.17 | .76 | .83 | .47–1.46 | .51 | .60 | .29–1.24 | .16 |
| Office-based visit for mental health care before hospitalization (reference: none) | 6.18 | 3.18–12.02 | <.001 | 8.45 | 3.87–18.44 | <.001 | 4.29 | 2.26–8.13 | <.001 | 6.27 | 2.93–13.42 | <.001 |
| Female (reference: male) | 1.63 | .72–3.69 | .24 | 1.40 | .50–3.96 | .52 | 1.51 | .87–2.63 | .14 | 1.54 | .78–3.04 | .20 |
| Age (reference: 18–34) | ||||||||||||
| 35–64 | 1.09 | .41–2.94 | .86 | 3.38 | 1.10–10.35 | .03 | 1.31 | .68–2.52 | .41 | 2.20 | 1.02–4.75 | .05 |
| ≥65 | .28 | .05–1.52 | .14 | .35 | .02–5.23 | .44 | .65 | .22–1.88 | .42 | .43 | .05–3.73 | .43 |
| Married (reference: not married) | .68 | .30–1.53 | .35 | .92 | .33–2.52 | .86 | .51 | .28–.93 | .03 | 1.05 | .51–2.15 | .90 |
| Survey panel | .79 | .63–.99 | .04 | .65 | .47–.89 | .01 | .86 | .74–1.01 | .07 | .82 | .67–1.00 | .05 |
| Non-need related | ||||||||||||
| Socioeconomic status | ||||||||||||
| Income ≥200% of the federal poverty level (reference: <200%) | 1.97 | .78–4.95 | .15 | 2.92 | .91–9.35 | .07 | 2.72 | 1.33–5.59 | .01 | 1.81 | .78–4.19 | .17 |
| Less than high school graduate (reference: high school or college graduate) | .68 | .28–1.64 | .39 | .21 | .05–.82 | .03 | 1.03 | .53–2.03 | .92 | .24 | .08–.67 | .01 |
| Other | ||||||||||||
| Health insurance (reference: private) | ||||||||||||
| Any uninsured | .40 | .12–1.30 | .13 | .38 | .10–1.41 | .15 | .70 | .32–1.54 | .37 | .86 | .33–2.29 | .76 |
| Any public, never uninsured | 1.56 | .54–4.45 | .41 | 2.74 | .62–12.18 | .18 | 1.56 | .75–3.25 | .23 | 2.04 | .82–5.11 | .12 |
| HMO | .52 | .21–1.28 | .15 | .42 | .13–1.37 | .15 | .85 | .48–1.53 | .59 | 1.05 | .52–2.10 | .90 |
| Employed (reference: unemployed) | 2.05 | .89–4.72 | .09 | 3.93 | 1.21–12.78 | .02 | 1.06 | .51–2.18 | .88 | 1.30 | .49–3.48 | .59 |
| Resides in Northeast (reference: residence elsewhere) | 1.42 | .62–3.26 | .40 | 1.81 | .68–4.78 | .23 | 1.10 | .58–2.06 | .77 | 1.46 | .65–3.25 | .35 |
| Urbanicity (reference: rural) | 1.50 | .55–4.10 | .43 | 2.63 | .68–10.13 | .16 | 1.93 | .88–4.26 | .10 | 2.45 | 1.01–5.99 | .049 |
| Constant | 5.16 | .06–419.44 | .46 | 1.08 | .01–196.98 | .98 | .96 | .04–24.92 | .98 | .22 | .00–15.97 | .49 |
a
Any outpatient visit
b
Adequate treatment was defined as a treatment episode consisting of at least eight visits to a mental health provider or at least four mental health care visits or events and at least one psychotropic medication fill.
Table 3 presents results for disparities in predicted probability of treatment after the analyses were adjusted for clinical need. Rates of follow-up within seven days of discharge were low for all groups (21%, 11%, and 23% for any outpatient visits and 14%, 8%, and 13% for adequate treatment among whites, blacks, and Latinos, respectively). These differences, although large in magnitude, were not statistically significant. However, blacks were significantly less likely than whites to receive a follow-up visit within 30 days of discharge (23% versus 40%, p=.02). This disparity persisted after the analysis was adjusted for prior outpatient treatment (p=.03). Whites were more likely than blacks to receive adequate care beginning within 30 days of discharge, but the difference was only marginally significant (26% versus 14%, p=.10). Latinos and whites received follow-up care at similar rates, and there were no significant differences between the groups.
| Follow-up measure | Before adjustment | After adjustment | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| White (N=206) | Black (N=78) | p | Latino (N=55) | p | White (N=206) | Black (N=78) | p | Latino (N=55) | p | |
| Within 7 days | ||||||||||
| Outpatient visit | 21.3 | 11.4 | .12 | 22.6 | .88 | 21.3 | 11.9 | .11 | 21.8 | .96 |
| Adequate treatment begunb | 14.2 | 8.0 | .34 | 12.5 | .82 | 14.2 | 8.3 | .34 | 10.9 | .65 |
| Within 30 days | ||||||||||
| Outpatient visit | 39.7 | 23.1 | .02 | 53.1 | .12 | 39.7 | 24.1 | .03 | 53.0 | .16 |
| Adequate treatment begunb | 25.6 | 13.9 | .10 | 26.3 | .93 | 25.6 | 14.1 | .10 | 25.0 | .94 |
a
Both analyses adjusted for clinical need variables.
b
Adequate treatment was defined as a treatment episode consisting of at least eight visits to a mental health provider or at least four mental health care visits or events and at least one psychotropic medication fill.
Discussion
This study found low rates of follow-up, and even lower rates of adequate treatment, following discharge, across all racial-ethnic groups. Ensuring continuity of care from acute to outpatient mental health settings remains a difficult challenge. Using the IOM definition of disparities, we found that blacks were significantly less likely than whites to receive any follow-up within 30 days of discharge. They were also less likely to receive adequate care beginning within 30 days of discharge, but this finding was only marginally significant. Even when the analyses controlled for prior outpatient treatment, a potent predictor of outpatient follow-up (50), these disparities persisted.
The sample had a high burden of both mental and general medical problems, in contrast to the overall MEPS sample, which had low rates of such problems (22). However, the rates of adequate care for the sample were low, suggesting that overall illness severity in itself does not ensure follow-up. In fact, comorbid general medical illnesses were a strong predictor of follow-up, suggesting that coordination of mental and general medical services is important to ensure continuity of mental health treatment (51).
The low rates of follow-up among black patients suggest that the high burden of illness for this group is being managed, in most cases, by just a small number of outpatient visits, if any. There are several findings in the literature relevant to improving disparities in follow-up. Among black inpatients, the intent to follow up after inpatient treatment is influenced by the level of empathic understanding they receive from mental health providers during the hospitalization (52). When patients from racial-ethnic minority groups perceive provider discrimination or bias during inpatient treatment, the negative impacts on mental health and follow-up can be profound and potentially debilitating (53). Culturally competent inpatient care of black patients might include efforts by inpatient staff to use brief psychotherapeutic approaches to convey empathy and act as ambassadors for outpatient colleagues. Collaborative-care approaches—in primary care or by telephone—also are associated with strong evidence for engaging individuals from racial-ethnic minority groups in outpatient mental health treatment (54).
As time passes after discharge, hospitals have decreasing control over a patient’s disposition, and the influence of low community resources and other social determinants grows (55). The emergence of significant disparities 30 days after discharge suggests that outpatient treatment among black patients is influenced by community factors, such as a lack of available treatment or stigma against pursuing further care. Because the odds of receiving follow-up decreased significantly during successive MEPS panels (Table 2), it also seems possible that the economic downturn may have had a negative impact on both hospital resources to ensure follow-up, such as social work and case management, and community availability of mental health treatment (56). African Americans were particularly affected by the recession, as indicated by more unemployment, lower rates of insurance, and decreases in health visits compared with whites (57). Future research should examine ways to modify community factors in order to improve follow-up. Such work would also be useful for understanding contributors to 30-day rehospitalization, another important health system quality indicator (58,59).
We chose to treat duration of inpatient treatment as a measure of clinical need, and, therefore, we adjusted for this variable. However, duration of inpatient stay may also represent availability of follow-up in the health system or the management of behavioral health care by payors. In post hoc sensitivity analyses, we treated this outcome as a non-need variable; when we compared the racial-ethnic groups on this variable, we found no differences of significance nor any changes in direction of our predicted outcomes. Further research using data sets that make it possible to describe management of care, variation in inpatient duration policies, and associations with disparities in follow-up is warranted.
Our results reassert the need for interventions to improve continuity of care for all acuity levels of mental health services, especially for black patients. Strategies supported by evidence include conducting discharge planning with outpatient providers, initiating outpatient programs before discharge, and engaging family during inpatient treatment (60). Use of care coordinators is supported by good evidence (61–63), although research on their use with racial-ethnic minorities is needed. In the United States, black patients are served by a concentrated number of health systems (64), and these systems merit increased support to improve treatment both before and after psychiatric hospitalization.
Our findings should be assessed in the context of several limitations. Because of the limited sample size, statistically insignificant results for certain comparisons may be attributable to low statistical power, as opposed to a true absence of effects. For example, the disparity between blacks and whites in rates of starting adequate treatment within 30 days was clinically significant but not statistically significant. For the comparison to achieve 80% power, an odds ratio of .38 would have been required, which is likely too small to have been reasonably observed in practice. A second limitation was that the MEPS excludes individuals who are institutionalized, live in congregate housing, or are homeless and, therefore, is not representative of the national population of persons with severe psychiatric disorder. In defining minimally adequate treatment, we allowed for any psychotropic medication fill during an episode, regardless of indication or diagnosis. This approach allowed broad comparisons of adequate care across psychiatric diagnoses but could not determine adequate treatment for specific diagnoses. Nor was it possible to assess the precise duration of medication use, given that dates of medication fills were imputed within MEPS rounds. As noted by other authors using similar methods (65), the MEPS cannot account for the local hospital conditions that might influence rates of follow-up mental health treatment, such as rates of referral to mental health care or availability of mental health professionals in the community (66).
Conclusions
This nationally representative analysis of follow-up after hospitalization showed low rates of adequate care and disparities between whites and blacks in rates of starting adequate follow-up within 30 days of discharge. Further investigation is needed to explore the impact of adequate care after discharge from an inpatient unit on clinical outcomes and other indicators of treatment quality, such as rehospitalization within 30 days.
Acknowledgments and disclosures
Funding for this project was provided by grant 1R01MH091042-01 from the National Institute of Mental Health.
The authors report no competing interests.
References
1.
Ilgen PDM, Hu KU, Moos PDR, et al.: Continuing care after inpatient psychiatric treatment for patients with psychiatric and substance use disorders. Psychiatric Services 59:982–988, 2008
2.
Kim HM, Pfeiffer P, Ganoczy D, et al.: Intensity of outpatient monitoring after discharge and psychiatric rehospitalization of veterans with depression. Psychiatric Services 62:1346–1352, 2011
3.
Nelson EA, Maruish ME, Axler JL: Effects of discharge planning and compliance with outpatient appointments on readmission rates. Psychiatric Services 51:885–889, 2000
4.
HEDIS 2014. Washington, DC, National Committee for Quality Assurance, 2014. Available at www.ncqa.org/HEDISQualityMeasurement/HEDISMeasures/HEDIS2014.aspx. Accessed March 6, 2014
5.
The State of Health Care Quality 2012: Focus on Obesity and on Medicare Plan Improvement. Washington, DC, National Committee on Quality Assurance, 2012
6.
Samnaliev M, McGovern MP, Clark RE: Racial/ethnic disparities in mental health treatment in six Medicaid programs. Journal of Health Care for the Poor and Underserved 20:165–176, 2009
7.
Collins KS, Hall A, Neuhaus CUS: Minority Health: A Chartbook. Washington, DC, Commonwealth Fund, 1999
8.
Schneider EC, Zaslavsky AM, Epstein AM: Racial disparities in the quality of care for enrollees in Medicare managed care. JAMA 287:1288–1294, 2002
9.
Virnig B, Huang Z, Lurie N, et al.: Does Medicare managed care provide equal treatment for mental illness across races? Archives of General Psychiatry 61:201–205, 2004
10.
Breslau J, Kendler KS, Su M, et al.: Lifetime risk and persistence of psychiatric disorders across ethnic groups in the United States. Psychological Medicine 35:317–327, 2005
11.
Williams DR, Gonzalez HM, Neighbors H, et al.: Prevalence and distribution of major depressive disorder in African Americans, Caribbean blacks, and non-Hispanic whites: results from the National Survey of American Life. Archives of General Psychiatry 64:305–315, 2007
12.
Teh CF, Sorbero MJ, Mihalyo MJ, et al.: Predictors of adequate depression treatment among Medicaid-enrolled adults. Health Services Research 45:302–315, 2010
13.
Compton MT, Rudisch BE, Craw J, et al.: Predictors of missed first appointments at community mental health centers after psychiatric hospitalization. Psychiatric Services 57:531–537, 2006
14.
Klinkenberg W, Calsyn R: Predictors of receipt of aftercare and recidivism among persons with severe mental illness: a review. Psychiatric Services 47:487–496, 1996
15.
Olfson M, Mechanic D, Boyer CA, et al.: Linking inpatients with schizophrenia to outpatient care. Psychiatric Services 49:911–917, 1998
16.
Stein B, Kogan J, Sorbero M, et al.: Predictors of timely follow-up care among Medicaid-enrolled adults after psychiatric hospitalization. Psychiatric Services 58:1563–1569, 2007
17.
National Healthcare Quality Report 2008. Rockville, Md, Agency for Healthcare Research and Quality, 2008
18.
Guevara JP, Mandell DS, Rostain AL, et al.: Disparities in the reporting and treatment of health conditions in children: an analysis of the Medical Expenditure Panel Survey. Health Services Research 41:532–549, 2006
19.
Fiscella K, Franks P, Doescher MP, et al.: Disparities in health care by race, ethnicity, and language among the insured: findings from a national sample. Medical Care 40:52–59, 2002
20.
Trivedi AN, Zaslavsky AM, Schneider EC, et al.: Trends in the quality of care and racial disparities in Medicare managed care. New England Journal of Medicine 353:692–700, 2005
21.
Vaccarino V, Rathore SS, Wenger NK, et al.: Sex and racial differences in the management of acute myocardial infarction, 1994 through 2002. New England Journal of Medicine 353:671–682, 2005
22.
Cook BL, McGuire T, Lock K, et al.: Comparing methods of racial and ethnic disparities measurement across different settings of mental health care. Health Services Research 45:825–847, 2010
23.
Cook B, McGuire T, Miranda J: Measuring trends in mental health care disparities, 2000–2004. Psychiatric Services 58:1533–1539, 2007
24.
McGuire TG, Alegria M, Cook BL, et al.: Implementing the Institute of Medicine definition of disparities: an application to mental health care. Health Services Research 41:1979–2005, 2006
25.
Cook BL, Alegría M: Racial-ethnic disparities in substance abuse treatment: the role of criminal history and socioeconomic status. Psychiatric Services 62:1273–1281, 2011
26.
Depetris AE, Cook BL: Differences in diffusion of FDA antidepressant risk warnings across racial-ethnic groups. Psychiatric Services 64:466–471, 2013
27.
Institute of Medicine: Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC, National Academies Press, 2003
28.
Cook B, McGuire TG, Zaslavsky AM: Measuring racial/ethnic disparities in health care: methods and practical issues. Health Services Research 47:1232–1254, 2012
29.
Measuring Racial Discrimination. Edited by Blank RM, Dabady M, Citro CF. Washington, DC, National Academies Press, 2004
30.
Cohen JW, Monheit AC, Beauregard KM, et al.: The Medical Expenditure Panel Survey: a national health information resource. Inquiry 33:373, 1997
31.
Zuvekas SH: Trends in mental health services use and spending, 1987–1996. Health Affairs 20(2):214–224, 2001
32.
Selden TM: The within-year concentration of medical care: implications for family out-of-pocket expenditure burdens. Health Services Research 44:1029–1051, 2009
33.
Alegria M, Chatterji P, Wells K, et al.: Disparity in depression treatment among racial and ethnic minority populations in the United States. Psychiatric Services 59:1264–1272, 2008
34.
Wang PS, Demler O, Kessler RC: Adequacy of treatment for serious mental illness in the United States. American Journal of Public Health 92:92–98, 2002
35.
Wang P, Berglund P, Kessler R: Recent care of common mental disorders in the United States: prevalence and conformance with evidence-based recommendations. Journal of General Internal Medicine 15:284–292, 2000
36.
Wang P, Lane M, Olfson M, et al.: Twelve-month use of mental health services in the United States. Archives of General Psychiatry 62:629–640, 2005
37.
American Psychiatric Association: Practice guideline for the treatment of patients with major depressive disorder (revision). American Journal of Psychiatry 157(Apr suppl):1–45, 2000
38.
Kroenke K, Spitzer RL, Williams JB: The Patient Health Questionnaire–2: validity of a two-item depression screener. Medical Care 41:1284–1292, 2003
39.
Kessler RC, Barker PR, Colpe LJ, et al.: Screening for serious mental illness in the general population. Archives of General Psychiatry 60:184–189, 2003
40.
Afari N, Schmaling KB, Barnhart S, et al.: Psychiatric comorbidity and functional status in adult patients with asthma. Journal of Clinical Psychology in Medical Settings 8:245–252, 2001
41.
Alexopoulos GS, Meyers BS, Young RC, et al.: “Vascular depression” hypothesis. Archives of General Psychiatry 54:915–922, 1997
42.
Clarke JL, Meiris DC: Building bridges: integrative solutions for managing complex comorbid conditions. American Journal of Medical Quality 22:5S, 2007
43.
de Groot M, Anderson R, Freedland KE, et al.: Association of depression and diabetes complications: a meta-analysis. Psychosomatic Medicine 63:619–630, 2001
44.
Wang PS, Lane M, Olfson M, et al.: Twelve-month use of mental health services in the United States: results from the National Comorbidity Survey Replication. Archives of General Psychiatry 62: 629–640, 2005
45.
Cook BL, McGuire TG, Zaslavsky AM, et al.: Adjusting for health status in non-linear models of health care disparities. Health Services and Outcomes Research Methodology 9:1–21, 2009
46.
Wolter K: Introduction to Variance Estimation. New York, Springer-Verlag, 1985
47.
MEPS H36BRR09 Codebook: 1996–2009 Replicates for Calculating Variances File. Rockville, Md, Agency for Healthcare Research and Quality, Dec 9, 2011. Available at meps.s-3.com/data_stats/download_data/pufs/h36brr/h36b09cb.pdf
48.
Prochaska JJ, Sung H-Y, Max W, et al.: Validity study of the K6 scale as a measure of moderate mental distress based on mental health treatment need and utilization. International Journal of Methods in Psychiatric Research 21:88–97, 2012
49.
Salyers MP, Bosworth HB, Swanson JW, et al.: Reliability and validity of the SF-12 health survey among people with severe mental illness. Medical Care 38:1141–1150, 2000
50.
Dausey DJ, Rosenheck RA, Lehman AF: Preadmission care as a new mental health performance indicator. Psychiatric Services 53:1451–1455, 2002
51.
Cook BL, McGuire TG, Alegría M, et al.: Crowd-out and exposure effects of physical comorbidities on mental health care use: implications for racial–ethnic disparities in access. Health Services Research 46:1259–1280, 2011
52.
Gillispie R, Williams E, Gillispie C: Hospitalized African American mental health consumers: some antecedents to service satisfaction and intent to comply with aftercare. American Journal of Orthopsychiatry 75:254–261, 2005
53.
Shavers VL, Fagan P, Jones D, et al.: The state of research on racial/ethnic discrimination in the receipt of health care. American Journal of Public Health 102:953–966, 2012
54.
Interian A, Lewis-Fernández R, Dixon LB: Improving treatment engagement of underserved US racial-ethnic groups: a review of recent interventions. Psychiatric Services 64:212–222, 2013
55.
Joynt KE, Jha AK: Thirty-day readmissions—truth and consequences. New England Journal of Medicine 366:1366–1369, 2012
56.
Levit KR, Mark TL, Coffey RM, et al.: Federal spending on behavioral health accelerated during recession as individuals lost employer insurance. Health Affairs 32:952–962, 2013
57.
Mortensen K, Chen J: The great recession and racial and ethnic disparities in health services use. JAMA Internal Medicine 173:315–317, 2013
58.
Bhalla R, Kalkut G: Could Medicare readmission policy exacerbate health care system inequity? Annals of Internal Medicine 152:114–117, 2010
59.
Hernandez AF, Curtis LH: Minding the gap between efforts to reduce readmissions and disparities. JAMA 305:715–716, 2011
60.
Boyer CA, McAlpine DD, Pottick KJ, et al.: Identifying risk factors and key strategies in linkage to outpatient psychiatric care. American Journal of Psychiatry 157:1592–1598, 2000
61.
Orlosky MJ, Caiati D, Hadad J, et al.: Improvement of psychiatric ambulatory follow-up care by use of care coordinators. American Journal of Medical Quality 22:95–97, 2007
62.
Batscha C, McDevitt J, Weiden P, et al.: The effect of an inpatient transition intervention on attendance at the first appointment postdischarge from a psychiatric hospitalization. Journal of the American Psychiatric Nurses Association 17:330–338, 2011
63.
Dixon L, Goldberg R, Iannone V, et al.: Use of a critical time intervention to promote continuity of care after psychiatric inpatient hospitalization. Psychiatric Services 60:451, 2009
64.
Jha AK, Orav EJ, Li Z, et al.: Concentration and quality of hospitals that care for elderly black patients. Archives of Internal Medicine 167:1177–1182, 2007
65.
Olfson M, Marcus SC, Bridge JA: Emergency treatment of deliberate self-harm. Archives of General Psychiatry 69:80–88, 2012
66.
Cook BL, Doksum T, Chen CN, et al.: The role of provider supply and organization in reducing racial/ethnic disparities in mental health care in the US. Social Science and Medicine 84:102–109, 2013
Information & Authors
Information
Published In
Cover: The Nation Makers, by Howard Pyle 1903. Oil on canvas. Collection Brandywine River Museum of Art, purchased through a grant from the Mabel Pew Myrin Trust, 1984.
History
Published in print: July 2014
Published online: 27 October 2014
Authors
Funding Information
National Institute of Mental Health10.13039/100000025: R01MH091042-01
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
There are no citations for this item
View Options
View options
PDF/ePub
View PDF/ePubGet Access
Login options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).