Abstract
Objective:
We examined racial/ethnic disparities in the receipt of minimally adequate depression treatment among Medicaid-enrolled youth.
Method:
We used 2008–2011 Medicaid claims data to derive a cohort of youth (age 5 to 17) that was diagnosed with a new episode of major depression (N=45,816) across nine states. Dichotomous outcomes measured the receipt of: (1) minimally adequate psychotherapy (≥ four psychotherapy visits within 12 weeks of initiation); (2) minimally adequate medication (filled antidepressants for 84 of 144 days); (3) any minimally adequate treatment (psychotherapy or medication); and (4) no psychotherapy or medication. Racial/ethnic disparities were estimated using logistic regressions that controlled for predisposing, enabling, and need-related factors.
Results:
Less than four-tenths (38.3%) of the cohort received minimally adequate psychotherapy, 19.2% received minimally adequate pharmacotherapy, and 49.9% received any minimally adequate treatment; conversely, 16.4% received no treatment. The adjusted percentages of Black (42.3%, p<0.001) and Hispanic (48.2%, p<0.001) youth that received minimally adequate treatment were significantly lower than among non-Hispanic whites (54.7%), due to lower likelihoods of receiving minimally adequate psychotherapy and/or minimally adequate pharmacotherapy. Additionally, the adjusted percentages of Black (20.2%, p<0.001) and Hispanic (15.0%, p<0.01) youth that received no treatment were significantly greater than among non-Hispanic white youth (12.9%).
Conclusions:
The percentage of Medicaid-enrolled youth that receive minimally adequate treatment for depression is low overall, and even lower among racial/ethnic minorities than among whites. Future research is needed to identify strategies that improve the overall quality of depression treatment among Medicaid-enrolled youth, as well as reduce disparities in care.
Keywords: major depression, psychotherapy, pharmacotherapy, quality of care, disparities, race and ethnicity
Lay Summary
We used Medicaid data from nine states to identify youth with a new episode of major depression. Less than half (49.9%) of youth in the cohort received minimally adequate treatment with either psychotherapy or pharmacotherapy in the acute phase of care. Compared to non-Hispanic whites, Black and Hispanic youth were significantly less likely to receive minimally adequate treatment, respectively. Future research is needed to identify strategies that improve the overall quality of depression treatment among Medicaid-enrolled youth, as well as reduce disparities among racial and ethnic minority populations.
Introduction
Major depression is one of the most common mental health (MH) disorders among youth,1 and it is associated with a number of poor health and social outcomes.2, 3 In spite of these consequences, large racial and ethnic disparities have been documented in the receipt of treatment for major depression among youth in the general population.4 Medicaid covered 37 million U.S. youth in 2016,5 is the largest payer of MH services,6 and disproportionately serves racial/ethnic minority youth.7 Therefore is critical to understand the patterns of depression care among Medicaid-enrolled youth, and how these differ across racial/ethnic groups.
The American Academy of Child and Adolescent Psychiatry (AACAP) recommends that psychotherapy approaches should generally be used as the first line treatment for youth with depression, and that antidepressants should only be used when response to psychotherapeutic approaches have failed and/or among those with more severe depressive episodes.8 To date, however, most studies on depression care for Medicaid-enrolled youth have only examined pharmacotherapy.9–12 Thus, more research is needed to understand the overall treatment patterns for major depression among Medicaid-enrolled youth -- including treatment with psychotherapy and/or pharmacotherapy -- and how these differ by race and ethnicity.
To address these gaps in the literature, we used administrative claims data from multiple states to derive a cohort of Medicaid-insured youth with a new episode of major depression. We examined patterns of the receipt of minimally adequate depression care in the acute phase of treatment, including treatment with psychotherapy and/or antidepressant medication. We subsequently estimated racial/ethnic differences in these measures, before and after adjusting for confounding measures. We discuss study findings in light of recent policy deliberations concerning the future of the Medicaid program.
METHODS
Data
We used data from the 2008–2011 Medicaid Analytic eXtract (MAX) Files from nine states (Alabama, Georgia, Kentucky, Louisiana, Missouri, North Carolina, Tennessee, Texas, Virginia). These states have been identified as having sufficiently complete managed care claims reporting;13 thus, we included youth in both fee-for-service plans and managed care plans in the cohort. The MAX data were merged with contextual-level measures from the Area Health Resources File (AHRF)14 and the 2008 National Survey of Mental Health Treatment Facilities (NSMHTF).15
This study received approval from the Emory University Institutional Review Board.
Study Cohort
We identified a cohort of 52,657 youth between the ages of 5 to 17 that initiated treatment for a new episode of major depression (ICD-9 code 296.2, 296.3) between January 1, 2008 and August 8, 2011. The cohort included youth that had at least: (1) one inpatient health care encounters with a primary diagnosis of major depression; and/or (2) two outpatient encounters with a primary diagnosis of major depression, the second of which needed to occur within twelve weeks of the index diagnosis. We defined the index visit using the first outpatient visit or inpatient stay with a primary diagnosis of major depression following a 90-day exclusion period during which there was no health care encounter with a major depression diagnosis or antidepressant medication that was filled.16, 17 We excluded those who: (1) were not continuously enrolled in Medicaid during the 90-day exclusion and 144-day treatment periods (N=6,181), (2) had dual Medicare eligibility (N=23), (3) had inaccurate or multiple county codes (N=636), and/or (3) had a missing value on gender (N=1). Our final cohort comprised 45,816 Medicaid-enrolled youth diagnosed with a new episode of major depression across nine states.
Measures
Minimally adequate depression care in the acute phase of treatment
AACAP clinical guidelines recommend that the treatment of depressive disorders in children and adolescents should include an acute and continuation phase.8 The duration of the acute phase is at least 6 to 12 weeks, and the clinician’s goal is to achieve response to treatment and full symptomatic remission.8, 18
We created a dichotomous indicator for those that received minimally adequate psychotherapy, defined as receiving at least four individual, family, or group psychotherapy visits (Current Procedure Terminology Codes presented in Table S1) on distinct days outside of an inpatient setting in the 12 weeks following the index depression diagnosis (Figure 1). AACAP guidelines recommend that four to six weeks of supportive therapy (in uncomplicated or brief cases of depression) or six to eight weeks of other types of psychotherapy (e.g., cognitive behavioral therapy) are needed to assess responsiveness to treatment.8 Consistent with Stein and colleagues,17 we use a four-visit threshold for the main analyses to capture the lower bound of this recommendation (assuming one visit per week, over 4 weeks). We also created an alternative measure using an eight-visit threshold in supplemental analyses.
Figure 1:
Derivation of measures: Minimally adequate treatment for depression in the acute phase of care
We also created an indicator for the receipt of minimally adequate pharmacotherapy in the acute phase of treatment, defined as filling antidepressant medication prescription(s) for 84 days (i.e., 12 weeks) out of the 144 days following the index date (Table S2, Figure 1). This measure follows the strategy employed by Stein and colleagues,17 which modified the Healthcare Effectiveness Data and Information Set measure of acute antidepressant medication treatment (i.e., filled antidepressant medication for 84 out of the first 114 days of treatment).19 The 114 day treatment period in HEDIS allows for up to 30 days for gaps in medication refills or for washout periods when switching medications.19 The modification extends the treatment period by an extra 30 days (from 114 to 144 days), to account for a potential lag between the receipt of the initial depression diagnosis, referral to a prescribing physician, and filling a prescription for an antidepressant medication.
A third indicator for the acute phase of treatment identified those that received any minimally adequate treatment, defined as the receipt of either minimally adequate pharmacotherapy or minimally adequate psychotherapy.
Finally, we created an indicator for those that did not receive any psychotherapy and did not fill any antidepressant medication (i.e., no treatment) during the acute phase of treatment.
Race/ethnicity
Race/ethnicity was measured with five mutually exclusive categories: non-Hispanic White, non-Hispanic Black, Hispanic, non-Hispanic other racial/ethnic group, and unknown race/ethnicity. Other race/ethnicity included smaller groups and those with more than one racial/ethnic background.
Individual-level measures
We included a robust set of covariates to control for individual-level predisposing, enabling, and need-related characteristics of our cohort.20 Specifically, we included age at the index date (in years) and an indicator for female (versus male) gender. To adjust for enabling factors, we created a categorical measure of the type of health plan using the monthly information regarding the health plan in which the child was enrolled (Table 1). To adjust for individual-level need-related characteristics, we created a categorical measure to assess the basis of Medicaid eligibility using the eligibility code from the most recent month observed in the study period (blind disabled, foster care, and other basis of eligibility category [e.g., low family income]). We created indicators to measure comorbid MH disorders, including attention deficit hyperactivity disorder [ADHD], oppositional defiance disorder and/or conduct disorder, anxiety disorder, bipolar disorder, and any other MH or substance use disorder. We also created two indicators to assess physical health problems, including asthma and other chronic conditions.21 Each indicator was created for those who had at least two inpatient and/or outpatient claims with the respective ICD-9 diagnosis codes (in any position) during the study timeframe, including the exclusion period and 144-day treatment period combined (See Table S3 for details).
Table 1:
Cohort characteristics of Medicaid-enrolled youth (age 5 to 17 years old) diagnosed with a new episode of major depression, by race/ethnicity
| Total N=45,816 | Non-Hispanic White n=22,464 | Non-Hispanic Black n=12,836 | Hispanic n=7,811 | Non-Hispanic Other n=430 | Unknown n=2,275 | |
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Age, mean (SD) | 13.3 (2.9) | 13.3 (2.9) | 13.3 (2.9) | 13.2*** (2.9) | 13.4 (3.0) | 13.0*** (3.0) |
| Female, % | 56.3 | 56.8 | 55.7 | 59.2*** | 59.3 | 43.6*** |
| Health plan type, % | ||||||
| Fee-for-service or primary care case management (no carve-out) | 32.4 | 33.7 | 27.7*** | 35.4** | 35.6 | 35.2 |
| Comprehensive managed care plan or any behavioral health care carve-out plan | 52.9 | 54.2 | 59.8*** | 38.6*** | 48.1* | 51.0** |
| More than one type of plana | 14.8 | 12.2 | 12.6 | 26.0*** | 16.3* | 13.8* |
| Basis of Eligibility, % | ||||||
| Blind/disabled | 10.7 | 4.4 | 12.7*** | 10.2*** | 9.1*** | 63.9*** |
| Foster care | 14.7 | 14.3 | 18.2*** | 12.5*** | 14.7 | 5.5*** |
| Other eligibility typeb | 74.6 | 81.3 | 69.1*** | 77.3*** | 76.3* | 30.6*** |
| Physical health comorbidity, % | ||||||
| Asthma | 4.4 | 3.7 | 5.2*** | 4.5** | 3.0 | 6.4*** |
| Any other chronic conditionc | 2.4 | 2.0 | 2.4* | 2.6** | 1.9 | 5.5*** |
| Mental health comorbidity, % | ||||||
| Oppositional defiant disorder and/or conduct disorder | 17.3 | 14.8 | 24.3*** | 11.4*** | 22.1*** | 21.9*** |
| Attention deficit hyperactivity disorder | 21.7 | 20.5 | 24.3*** | 17.6*** | 18.8 | 34.0*** |
| Anxiety disorder | 8.6 | 9.5 | 7.2*** | 7.9*** | 7.7 | 9.9 |
| Bipolar disorder | 18.8 | 19.6 | 18.0*** | 15.5*** | 17.7 | 27.0*** |
| Other mental health disorder and/or substance use disorder | 29.1 | 29.9 | 28.9* | 26.5*** | 27.7 | 31.9 |
| County-level characteristics, mean (SD) | ||||||
| Percentage residents living in urban area | 67.3 (30.4) | 54.9 (30.4) | 77.8*** (26.1) | 84.8*** (19.8) | 61.2*** (31.4) | 71.5*** (29.0) |
| Percentage residents living in poverty | 16.6 (6.4) | 15.5 (5.1) | 16.0*** (5.3) | 21.1*** (9.0) | 17.5*** (7.8) | 16.4*** (5.9) |
| Outpatient mental health facilities per 100K residents | 1.3 (2.0) | 1.6 (2.3) | 1.1*** (1.6) | 0.6*** (1.5) | 1.3*** (1.8) | 1.2*** (2.0) |
| Community health centers per 100K residents | 4.4 (8.3) | 6.0 (9.9) | 2.5*** (5.4) | 3.3*** (6.7) | 5.6 (8.4) | 3.0*** (5.9) |
| Primary care physicians per 100K residents | 60.7 (28.3) | 57.1 (29.5) | 70.5*** (28.4) | 54.5*** (19.6) | 59.9 (29.2) | 62.4*** (25.9) |
| Psychologists per 100K residents | 17.9 (19.1) | 16.2 (18.6) | 23.9*** (21.1) | 12.6*** (13.3) | 17.4 (20.6) | 19.6*** (19.6) |
Notes:
p<0.05,
p<0.01,
p<0.001; SD: standard deviation.
Statistical tests were conducted using adjusted Wald tests to compare youth in each racial/ethnic minority group to White youth.
We used monthly information about plan enrollment to measure health plan type. This category includes those that were enrolled in more than one plan type during the study period.
“Other eligibility type” includes children eligible for Medicaid based on household income, classification as “medically needy”, and/or other criteria specified in each state’s Section 1115 waiver.
“Other chronic conditions” included cerebral palsy, cystic fibrosis, diabetes, spina bifida, seizure disorder, congenital heart disease, sickle cell disease, and malignant neoplasms.
County-level measures
County-level measures were assessed for the most recent year of data available preceding the study period or the earliest year during the study period. County-level socioeconomic status was measured by the percentage of residents living in poverty in 2008. Urbanicity was measured by the percentage of county residents living in an urban area (2000), as defined by the U.S. Census Bureau.22 Because prior research suggests that the supply of MH treatment resources varies across communities by racial/ethnic composition23, 24 and that the supply of MH providers is associated with MH service use among depressed adolescents,25 we created county-level measures of the per capita number of: (1) community health centers (i.e., rural health clinics and federally qualified health centers) in 2008, (2) outpatient MH facilities that serve youth and accept Medicaid in 2008, (3) psychologists in 2009, and (4) primary care physicians in 2010.
Analysis
We conducted unadjusted, bivariate analyses using Wald tests to compare the outcome measures and covariates across racial/ethnic groups. Next, we estimated logistic regression models to identify racial/ethnic differences in depression care after adjusting for individual- and county-level covariates. We also examined whether racial/ethnic differences in minimally adequate psychotherapy were due to differences in having at least one visit and/or differences in the receipt of at least four visits (conditional on any visit) (Table S4). We conducted similar analyses for minimally adequate pharmacotherapy (Table S5).
In all models, we included state indicators to control for unobserved state-level characteristics that may be associated with health services utilization and population demographics, and we clustered standard errors at the county-level. For the ease of interpretation, we reported marginal effects calculated for each racial/ethnic group at the observed values of all other covariates in the model. The marginal effects reflect the predicted percentage point difference in the likelihood that the outcome of interest occurs among a racial/ethnic minority group, compared with non-Hispanic whites (i.e., the intercept).
RESULTS
Minimally Adequate Depression Care
Cohort characteristics are presented in Table 1. Of Medicaid-enrolled youth diagnosed with a new episode of depression, 68.6% received at least one psychotherapy session during the 12 weeks after their index diagnosis; of these, just over half (55.9%) received at least four psychotherapy visits and only 22.9% received at least eight visits (Table 2). Thus, 38.3% of the cohort received minimally adequate psychotherapy using the 4-visit threshold and 15.7% received minimally adequate psychotherapy using the 8-visit threshold.
Table 2:
Receipt of adequate treatment in the acute phase of care among Medicaid-enrolled youth (age 5 to 17 years old) diagnosed with a new episode of major depression, by race/ethnicity
| Total N=45,816 | Non-Hispanic White n=22,464 | Non-Hispanic Black n=12,836 | Hispanic n=7,811 | Non-Hispanic Other n=430 | Unknown n=2,275 | |
|---|---|---|---|---|---|---|
| Psychotherapy | ||||||
| Any psychotherapy visit | 68.6 | 73.4 | 58.1*** | 73.0 | 56.3*** | 67.7*** |
| ≥ 4 visits (of those that had any, n=31,408) | 55.9 | 57.6 | 53.0*** | 55.3** | 54.1 | 54.3* |
| ≥ 8 visits (of those that had any, n=31,408) | 22.9 | 24.4 | 23.4 | 17.7*** | 24.4 | 22.9 |
| Minimally adequate psychotherapy (≥ 4 visits) | 38.3 | 42.2 | 30.7*** | 40.4** | 30.5*** | 36.8*** |
| Minimally adequate psychotherapy, version 2 (≥ 8 visits) | 15.7 | 17.9 | 13.6*** | 12.9*** | 13.7* | 15.5** |
| Pharmacotherapy | ||||||
| Any antidepressant medication prescription | 43.2 | 44.9 | 36.8*** | 47.7*** | 40.0* | 47.0 |
| Filled medication for ≥ 84 days (of those that had any, n=19,770) | 44.6 | 48.6 | 37.6*** | 42.0*** | 43.0 | 46.6 |
| Minimally adequate pharmacotherapy (≥ 84 days) | 19.2 | 21.8 | 13.8*** | 20.0*** | 17.2* | 21.9 |
| Any Adequate Treatment | ||||||
| Minimally adequate psychotherapy (≥ 4 visits) and/or pharmacotherapy | 49.9 | 55.1 | 39.4*** | 52.3*** | 41.6*** | 50.7*** |
| Minimally adequate psychotherapy (≥ 8 visits) and/or pharmacotherapy | 31.6 | 35.7 | 25.0*** | 30.1*** | 27.4*** | 34.3 |
| No treatment with either psychotherapy or pharmacotherapy | 16.4 | 12.3 | 27.0*** | 10.6*** | 27.9*** | 15.5*** |
N=45,816 unless otherwise noted
p<0.05,
p<0.01,
p<0.001
Statistical tests were conducted using adjusted Wald tests to compare youth in each racial/ethnic minority group to White youth.
When examining pharmacotherapy, 43.2% filled any antidepressant medication during the 144-day treatment period; of these, four-tenths (44.6%) filled antidepressant medication prescription(s) for at least 84 days (Table 2). Thus, less than one-fifth (19.2%) of the cohort received minimally adequate pharmacotherapy.
Using the more conservative (4-visit) threshold for psychotherapy visits, 49.9% received any minimally adequate treatment in the acute phase of care (Table 2). When using the 8-visit threshold, this figure drops to 31.6%. One in six (16.4%) did not receive any treatment at all (psychotherapy or pharmacotherapy) in the acute phase.
Racial/Ethnic Differences in Minimally Adequate Depression Care
The percentages of youth that received minimally adequate psychotherapy and minimally adequate pharmacotherapy were significantly lower among racial/ethnic minorities compared to their white peers. In the unadjusted comparisons (Table 2), 42.2% of white youth received minimally adequate psychotherapy, compared to 30.7% (p<0.001) and 40.4% (p=0.004) among Black and Hispanic youth, respectively. Moreover, 21.8% of white youth received minimally adequate pharmacotherapy compared to 13.8% (p<0.001) and 20.0% (p<0.001) among Black and Hispanic youth, respectively.
After controlling for confounders, the predicted percentage that received minimally adequate psychotherapy was 7.5 percentage points (Marginal effect [M.E.] = −7.5, 95% CI = −9.4, −5.7) lower among Black youth and 2.5 percentage points (M.E. = −2.5, 95% CI = −4.6, −0.3) lower among Hispanic youth than among non-Hispanic white youth (intercept=40.9%), respectively (Table 3, Figure 2). Black youth were less likely than non-Hispanic whites to receive any psychotherapy visit (M.E.= −7.9, 95% CI=−9.5,−6.2), or receive at least 4 visits conditional on having any visit (M.E.= −4.6, 95% CI=−6.5,−2.7) (Table S4).
Table 3:
Logistic regression results examining racial/ethnic differences in minimally adequate treatment among Medicaid-enrolled youth with a new episode of major depression
| Adequate psychotherapy (≥ 4 visits) | Adequate pharmacotherapy | Adequate psychotherapy (≥ 4 visits) or pharmacotherapy | No psychotherapy and no pharmacotherapy | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ME(%) | 95% CI | ME(%) | 95% CI | ME(%) | 95% CI | ME(%) | 95% CI | |||||||||
| Race/ethnicity (ref: non-Hispanic white) | ||||||||||||||||
| Non-Hispanic black | −7.5 | −9.4 | to | −5.7 | −10.4 | −11.8 | to | −8.9 | −12.4 | −14.1 | to | −10.7 | 7.3 | 6.5 | to | 8.2 |
| Hispanic | −2.5 | −4.6 | to | −0.3 | −6.1 | −8.0 | to | −4.1 | −6.5 | −8.6 | to | −4.3 | 2.1 | 0.7 | to | 3.5 |
| Non-Hispanic other | −6.6 | −13.7 | to | 0.6 | −3.5 | −8.3 | to | 1.4 | −7.7 | −14.4 | to | −1.0 | 6.0 | 2.0 | to | 10.1 |
| Unknown | −3.0 | −5.9 | to | −0.1 | −0.7 | −2.8 | to | 1.4 | −3.5 | −5.9 | to | −1.0 | 3.0 | 1.5 | to | 4.4 |
| Demographics | ||||||||||||||||
| Age | −0.4 | −0.6 | to | −0.2 | 1.4 | 1.1 | to | 1.6 | 0.1 | −0.1 | to | 0.3 | −0.4 | −0.5 | to | −0.3 |
| Female | 1.6 | 0.5 | to | 2.7 | 4.4 | 3.6 | to | 5.3 | 3.4 | 2.3 | to | 4.4 | −3.0 | −3.6 | to | −2.4 |
| Health plan type (ref: Fee-for-service or primary care case management) | ||||||||||||||||
| Comprehensive managed care plan or any behavioral health care carve-out plan | −8.1 | −10.6 | to | −5.7 | 2.2 | 0.7 | to | 3.7 | −5.3 | −7.4 | to | −3.3 | 2.6 | 1.1 | to | 4.0 |
| More than one type of plan | −7.4 | −9.1 | to | −5.6 | −1.3 | −2.7 | to | 0.2 | −6.0 | −7.7 | to | −4.4 | 1.4 | 0.4 | to | 2.5 |
| Basis of Eligibility (ref: blind/disabled) | ||||||||||||||||
| Foster care | 21.7 | 19.1 | to | 24.3 | 3.4 | 1.4 | to | 5.3 | 20.0 | 17.4 | to | 22.5 | −6.9 | −8.7 | to | −5.2 |
| Other eligibility type | 6.7 | 4.8 | to | 8.6 | 1.0 | −0.4 | to | 2.4 | 5.5 | 3.9 | to | 7.2 | −3.2 | −4.1 | to | −2.2 |
| Physical health comorbidity | ||||||||||||||||
| Asthma | 2.0 | 0.02 | to | 4.0 | 4.2 | 2.4 | to | 6.1 | 4.0 | 2.0 | to | 6.0 | −1.8 | −3.3 | to | −0.4 |
| Any other chronic condition | −2.8 | −5.8 | to | 0.1 | 1.0 | −1.5 | to | 3.4 | −1.4 | −4.3 | to | 1.5 | −0.1 | −1.9 | to | 1.7 |
| Mental health comorbidity | ||||||||||||||||
| Oppositional defiant disorder or conduct disorder | 9.2 | 7.6 | to | 10.7 | 2.0 | 0.9 | to | 3.2 | 8.7 | 7.0 | to | 10.4 | −1.8 | −2.7 | to | −0.8 |
| Attention deficit hyperactivity disorder | 3.8 | 2.5 | to | 5.2 | 2.5 | 1.1 | to | 3.9 | 4.6 | 3.4 | to | 5.9 | −1.9 | −2.7 | to | −1.1 |
| Anxiety disorder | 4.3 | 2.3 | to | 6.3 | 10.7 | 9.3 | to | 12.1 | 10.4 | 8.3 | to | 12.4 | −5.8 | −7.5 | to | −4.2 |
| Bipolar disorder | 5.6 | 3.8 | to | 7.4 | 4.7 | 3.3 | to | 6.1 | 6.9 | 5.2 | to | 8.5 | −2.6 | −3.7 | to | −1.4 |
| Other mental health disorder or substance use disorder | 9.7 | 8.2 | to | 11.1 | 6.7 | 5.8 | to | 7.5 | 11.0 | 9.7 | to | 12.3 | −4.4 | −5.1 | to | −3.6 |
| County-level characteristics | ||||||||||||||||
| Percentage residents living in urban areaa | −0.6 | −2.1 | to | 1.0 | −0.5 | −1.4 | to | 0.4 | −1.0 | −2.3 | to | 0.4 | 1.0 | −0.01 | to | 1.9 |
| Percentage residents living in povertya | −0.8 | −1.9 | to | 0.3 | −1.0 | −1.9 | to | −0.1 | −1.1 | −2.0 | to | −0.1 | −0.5 | −1.4 | to | 0.4 |
| Outpatient mental health facilities per 100K residentsa | −0.3 | −1.1 | to | 0.5 | −0.03 | −0.6 | to | 0.5 | −0.4 | −1.1 | to | 0.3 | 0.2 | −0.3 | to | 0.7 |
| Community health centers per 100K residentsa | −0.5 | −1.7 | to | 0.6 | −0.5 | −1.2 | to | 0.3 | −0.8 | −1.9 | to | 0.3 | 1.3 | 0.4 | to | 2.2 |
| Primary care physicians per 100K residentsa | 0.5 | −0.9 | to | 1.9 | −0.2 | −1.2 | to | 0.8 | 0.1 | −1.1 | to | 1.4 | −0.3 | −1.2 | to | 0.7 |
| Psychologists per 100K residentsa | 1.0 | −1.2 | to | 3.1 | 0.1 | −0.9 | to | 1.1 | 0.9 | −1.3 | to | 3.0 | 0.3 | −1.2 | to | 1.8 |
| Intercept | 40.9 | 22.8 | 54.7 | 12.9 | ||||||||||||
Notes: ME = marginal effect; Ref = reference; CI = confidence interval. N = 45,816
Marginal effects and intercept were estimated at the reference of race/ethnicity (i.e., whites) and the observed value of other covariates.
Logistic regressions were estimated with state indicators and standard errors were clustered at the county level.
Continuous measures were standardized such that a one-unit increase corresponds to a one standard deviation increase in the measure above its mean value.
Figure 2:
Model-adjusted predicted percentages of minimally adequate treatment for depression among Medicaid-enrolled youth, by race/ethnicity§±
Notes: * p < 0.05, **p < 0.01, ***p < 0.001; N = 45,816
§ Predicted percentages were estimated at the reference of race/ethnicity (i.e., Whites) using logistic regression models.
± Results from multivariate logit models were adjusted for state indicators; the county-level measures of socio-demographic characteristics and mental healthcare resources; and the individual-level measures of socio-demographic and need-related characteristics.
Turning to minimally adequate pharmacotherapy, the model-adjusted value of this outcome measure was 10.4 percentage points lower among Black youth (95% CI= −11.8,−8.9) and 6.1 percentage points lower among Hispanic youth (95% CI= −8.0,−4.1) than among non-Hispanic white youth (intercept=22.8%), respectively (Table 4, Figure 2). Black and Hispanic youth were less likely than non-Hispanic white youth to receive any pharmacotherapy (p<0.010), or fill an antidepressant for at least 84 days (12 weeks) conditional on receiving any antidepressant (p<0.001) (Table S5).
The significantly lower percentages of minimally adequate psychotherapy and pharmacotherapy among Black and Hispanic youth translated into significantly lower percentages of any minimally adequate depression treatment in the bivariate (Table 2) and multivariate comparisons (Table 3). The model-adjusted predicted percentage of those that received any minimally adequate treatment was 12.4 percentage points (95% CI = −14.1, −10.7) lower among Black youth and 6.5 percentage points (95% CI = −8.6, −4.3) lower among Hispanic youth than among non-Hispanic white youth (intercept=54.7%), respectively. These findings are depicted graphically in Figure 2, in which the model-adjusted percentages of Black (42.3%) and Hispanic (48.2%) youth that received minimally adequate depression treatment were significantly lower than among white youth (54.7%).
When using the 8-visit threshold for psychotherapy, , the adjusted percentages of Black (M.E.= − 10.7 percentage points; 95% CI= −12.3, −9.2) and Hispanic (M.E.= −7.0 percentage points; 95% CI= −8.9, −5.0) youth that received minimally adequate treatment were also significantly lower than among non-Hispanic whites (intercept=35.5%) (Table S6).
Racial/Ethnic Differences in the Receipt of No Treatment
In multivariate comparisons (Table 3, Figure 2), Black youth (M.E. = 7.3 percentage points, 95% CI = 6.5, 8.2) and Hispanic youth (M.E. = 2.1 percentage points, 95% CI = 0.7, 3.5) were more likely than non-Hispanic whites to receive no treatment with either psychotherapy or pharmacotherapy. Thus, Figure 2 illustrates that the adjusted percentages of Black (20.2%) and Hispanic (15.0%) youth that received no treatment were significantly greater than among non-Hispanic white youth (12.9%).
Supplemental Analyses
We compared models with and without county-level measures of the health and MH care infrastructure to assess whether community-level differences in these resources explained racial/ethnic disparities in depression care. The estimates for race/ethnicity were qualitatively similar in magnitude and significance in both sets of models.
We also conducted analyses to further elucidate the Black-White disparity in the likelihood of receiving no treatment with psychotherapy or antidepressants. In the subgroup that received no treatment, we compared the diagnosed prevalence of other MH problems across racial groups (Table S7) and found that Black youth were significantly more likely than non-Hispanic whites to have a diagnosis of oppositional defiant disorder and/or conduct disorder (22.2% versus 13.0%, p<0.001). However, the diagnosed prevalence of all types of MH disorders combined (other than depression) did not differ between Black and non-Hispanic white youth (54.0% versus 51.8%, p=0.08). We also examined racial/ethnic differences in the likelihood that youth in this subgroup received psychotropic medications other than an antidepressant including: stimulants, other (non-stimulant) medication for ADHD, antianxiety medications (including benzodiazepines), mood stabilizers, and antipsychotics (Table S8). However, Black youth in this subgroup were significantly less likely to receive any other psychotropic medication than non-Hispanic whites (28.8% versus 38.4%, p<0.001).
Lastly, we conducted supplemental analyses for a subgroup less than 12 years of age (n=12,939) because the selection of treatment modality (i.e., psychotherapy versus pharmacotherapy) may differ for younger (versus older) youth with major depression. The overall percentage of younger youth that received any minimally adequate treatment did not differ statistically compared to older youth between the ages of 12 and 17 (50.6% versus 49.6%; p=0.051). However, younger youth were more likely to receive any adequate psychotherapy (41.8% versus 36.9%; p<0.001), and less likely to receive adequate pharmacotherapy (14.2% versus 21.2%; p<0.001) than older youth. When estimating the regression models in the younger subgroup (Table S9), the overall pattern of findings for race/ethnicity in the receipt of minimally adequate psychotherapy and any minimally adequate treatment were qualitatively similar to our findings from the main analysis in direction and significance. The relative sizes of the racial/ethnic disparities estimated for minimally adequate pharmacotherapy, however, were less pronounced in the younger subgroup; nevertheless, the marginal effect remained negative and significant for Black youth (M.E. = −2.9 percentage points, 95% CI= −4.8, −0.9) and approached significance for Hispanic youth (M.E.= −2.1 percentage points, 95% CI= −4.6, 0.4, p<0.10).
DISCUSSION
Using administrative claims data from nine states, we conducted the most comprehensive examination of depression care quality to date among Medicaid-enrolled youth. The study cohort included youth enrolled in traditional fee-for-service plans as well as managed care plans. Results indicated less than half of Medicaid-enrolled youth diagnosed with a new episode of major depression received minimally adequate depression care. Moreover, one-sixth of these youth did not receive any psychotherapy or any antidepressant medication. Compared to non-Hispanic whites, Black and Hispanic youth were significantly less likely to receive minimally adequate treatment and significantly more likely to receive no treatment at all.
AACAP clinical guidelines recommend that the minimum standard of care should include an acute phase (to obtain treatment response and symptom remission) as well as a continuation phase of treatment that lasts for at least 6 months.8 Our results, however, reveal high percentages of youth that drop out of treatment in the acute phase of care across all racial/ethnic groups. Three-fifths received any psychotherapy visit, but only 38.3% received at least four visits and 15.7% received at least eight visits in the acute phase. Similarly, although two-fifths of the cohort initiated antidepressant medication, only one-fifth received minimally adequate pharmacotherapy in the acute phase. Thus, the vast majority of those that initiate treatment do not make it to the continuation phase. The high likelihood of treatment dropout may reflect, in part, logistical barriers to care (including family schedules, lack of reliable transportation, and distance to the nearest provider)26–28 as well as attitudinal barriers (e.g., stigma) to treatment for low-income families.26
Compared to non-Hispanic whites, Black and Hispanic youth were less likely to receive minimally adequate treatment with psychotherapy or pharmacotherapy before and after adjusting for a robust set of child- and community-level covariates. Notably, the relative size of the disparity was larger for minimally adequate pharmacotherapy than for minimally adequate psychotherapy. This pattern of findings is similar to prior research,17 which compared adequate treatment between non-white and white Medicaid-enrolled youth in an urban county. The residual disparities may be explained by unmeasured racial/ethnic differences in attitudinal barriers such as stigma concerning depression and its treatment or unmeasured systems-level barriers, such as patient-provider language concordance or provider cultural competency.26, 29 Cultural health beliefs concerning psychotropic medication among Black and Hispanic individuals30, 31 may be an especially important factor that contributes to a lower likelihood of receiving minimally adequate psychotropic medication among minority youth compared to their non-Hispanic white peers.
The overall likelihood and the Black-white disparity in the percentage of youth diagnosed with a new episode of major depression that did not receive any psychotherapy or any antidepressant medication are also concerning. As one explanation for this disparity, families and health care professionals may be more focused on treating other co-morbid MH problems among Black youth compared to whites. Prior research reported that Black youth are more likely to enter MH treatment through social agencies compared to their non-Hispanic white peers,32 and Black youth with major depression are more likely than their white peers to report seeking MH treatment due to problems at school.33 However, our analyses did not lend support to this explanation. In the regression analysis, we controlled for co-morbid MH disorders. In addition, the measure of psychotherapy use was conservative and counted all psychotherapy visits -- not just those with a diagnosis code for major depression. In the subgroup that received no treatment, Black youth had a similar likelihood of having a diagnosis for another co-morbid MH disorder and a lower likelihood of receiving treatment with other types of psychotropic medication (indicative of treatment for different types of disorders) compared to non-Hispanic whites.
Future research is needed to identify the contributing mechanisms to these disparities in depression care, as well as potential strategies to mitigate them. One area of inquiry may focus on whether expanded behavioral health services in other healthcare settings – such as community health centers and schools – have the potential to address these disparities in care. As one example, comprehensive school-based health centers (SBHC) are primary medical centers that typically offer integrated preventive, physical, and MH services on school grounds.34 Prior studies have reported that racial/ethnic minorities are more likely to receive any MH care,35, 36 and receive more MH services at SBHCs compared to their non-Hispanic white peers.37 Yet, these studies have generally been descriptive in nature and more rigorous research is needed on the potential of SBHCs to improve depression care and address racial/ethnic disparities in care among Medicaid-enrolled youth.
In the past year, policies proposed at the federal level have included dramatic reductions in federal funding for the Medicaid program (up to $1 trillion in reduced Medicaid spending over the next decade).38 If federal funding for the Medicaid program were reduced at this level, state policymakers would likely need to implement strategies to reduce Medicaid program expenditures such as scaling back covered services (including MH services that are not mandated as part of coverage), reducing reimbursement rates for providers, and/or reducing the number of those eligible for coverage by scaling back program eligibility criteria.39, 40 The first two strategies could exacerbate the state of depression care for Medicaid-enrolled youth if coverage for psychotherapy services becomes less generous or if fewer MH providers were to participate in Medicaid due to lower reimbursement rates. Furthermore, changes to the Medicaid program that disproportionately reduce the coverage for and/or availability of psychotherapy services (versus pharmacotherapy) could exacerbate racial/ethnic disparities in depression care among youth, given the lower absolute and relative sizes of the racial/ethnic disparities for these services compared to pharmacotherapy.
There are several limitations to acknowledge. First, these findings may not necessarily generalize to other regions of the country. Second, the data are several years old. A recent study using Medicaid data from 2002 to 2009 reported that the Food and Drug Administration’s 2004 antidepressant warning led to greater declines in antidepressant prescribing for White youth compared to Black and Latino youth.41 If this trend continued beyond our study period, the disparities we estimated in minimally adequate antidepressant use (and any minimally adequate treatment) may have narrowed since the end of our study period; this reduction may not have occurred because minority youth are receiving more treatment, but because white youth are receiving less. As more recent data become available, future research will be needed to assess changes in overall likelihood that youth receive minimally adequate care as well as racial/ethnic disparities in these measures.
Third, the outcome measures are imperfect proxies for minimally adequate care in the acute phase of treatment. They are, however, the best available measures and were derived based on a combination of AACAP guidelines,8 HEDIS performance measures,19 and prior research.17 For the measures of minimally adequate psychotherapy, we chose conservative thresholds of 4 and 8 visits that are suggested by clinical guidelines to assess if the child has responded to treatment; thus, these thresholds should not be interpreted as the number of visits needed to achieve an adequate course of care and full symptomatic remission.
Another possible limitation pertains to the measurement of minimally adequate psychotherapy. When deriving this measure, we used national CPT codes for individual, family, or group psychotherapy. In some states, however, it is possible that psychotherapy services could be provided as a component of other services (e.g., wraparound services) that are billed using different codes. To the extent that this occurs, our measure of minimally adequate psychotherapy would provide a conservative, lower bound estimate of this outcome measure, particularly among youth with more severe depression and/or MH comorbidities. In addition, coding errors in administrative claims databases may also result in measurement error.42
In spite of these limitations, this study sheds new light on the state of depression care quality in a large, diverse cohort of Medicaid-enrolled youth. Our findings indicate that most Medicaid-enrolled youth identified with a new episode of major depression do not receive minimally adequate treatment in the acute phase, and that racial and ethnic minorities are even less likely to receive minimally adequate treatment compared to non-Hispanic whites. Future research is needed to elucidate the contributing mechanisms to these disparities, and to identify the impact of poor care quality across populations on functional outcomes such as academic performance and social interactions. More work is also needed to examine the potential of integrated delivery models of care to improve access to and quality of MH services for diverse, low-income populations.
Supplementary Material
Clinical Guidance.
When Medicaid-enrolled youth initiate depression care, clinicians should consider helping families proactively identify and address barriers to engaging in treatment (both pharmacotherapy and psychotherapy).
Culturally tailored efforts to improve treatment engagement and retention would be especially important for Black and Hispanic youth, since these groups are even less likely than non-Hispanic whites to receive minimally adequate psychotherapy or pharmacotherapy.
Strategies to improve treatment engagement may need to address logistical barriers to care, cultural health beliefs, and/or stigma concerning depression treatment.
Acknowledgement & Disclosure:
This work was supported by the National Institute of Mental Health (K01MH095823). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
The authors have no conflicts of interest to disclose.
Presentation: Panel Presentation at AACAP’s 63rd Annual Meeting
References
- 1.Avenevoli S, Swendsen J, He J-P, Burstein M, Merikangas KR. Major depression in the National Comorbidity Survey–Adolescent Supplement: prevalence, correlates, and treatment. J Am Acad Child Adolesc Psychiatry. 2015;54(1):37–44. e32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Weissman MM, Wolk S, Wickramaratne P, et al. Children with prepubertal-onset major depressive disorder and anxiety grown up. Archives of general psychiatry. 1999;56(9):794–801. [DOI] [PubMed] [Google Scholar]
- 3.Weissman MM, Wolk S, Goldstein RB, et al. Depressed adolescents grown up. Journal of the American Medical Association. May 12 1999;281(18):1707–1713. [DOI] [PubMed] [Google Scholar]
- 4.Cummings JR, Druss BG. Racial/ethnic differences in mental health service use among adolescents with major depression. J Am Acad Child Adolesc Psychiatry. February 2011;50(2):160–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Centers for Medicare & Medicaid Services (CMS). FY 2016 unduplicated number of children ever enrolled in Medicaid as reported by the Statistical Enrollment Data System (SEDS) February 2017. Available at: https://www.medicaid.gov/chip/downloads/fy-2016-childrens-enrollment-report.pdf. Accessed August 14, 2017.
- 6.Centers for Medicare and Medicaid Services. Behavioral Health Services. 2016;https://www.medicaid.gov/medicaid/benefits/bhs/index.html. Accessed January 1, 2018.
- 7.Brooks Tricia, Wagnerman Karina. Snapshot of Children’s Coverage by Race and Ethnicity. Washington, DC: Georgetown University Health Policy Institute, Center for Children & Families (CCF). April 2017. Available at: https://ccf.georgetown.edu/2017/04/26/snapshot-of-childrens-coverage-by-race-and-ethnicity/. Accessed February 1, 2018. [Google Scholar]
- 8.Birmaher Boris, Brent David, AACAP Work Group on Quality Issues. Practice parameter for the assessment and treatment of children and adolescents with depressive disorders. J Am Acad Child Adolesc Psychiatry. 2007;46(11):1503–1526. [DOI] [PubMed] [Google Scholar]
- 9.Zito JM, Safer DJ, Gardner JF, Soeken K, Boles M, Lynch F. Rising prevalence of antidepressants among US youths. Pediatrics. 2002;109(5):721–727. [DOI] [PubMed] [Google Scholar]
- 10.Richardson LP, DiGiuseppe D, Christakis DA, McCauley E, Katon W. Quality of Care for Medicaid-Covered Youth Treated With Antidepressant Therapy. Archives of General Psychiatry. 2004;61(5):475–480. [DOI] [PubMed] [Google Scholar]
- 11.Richardson LP, DiGiuseppe D, Garrison M, Christakis DA. Depression in Medicaid-covered youth: differences by race and ethnicity. Archives of Pediatrics and Adolescent Medicine. 2003;157(10):984–989. [DOI] [PubMed] [Google Scholar]
- 12.Fontanella CA, Bridge JA, Marcus SC, Campo JV. Factors associated with antidepressant adherence for Medicaid-enrolled children and adolescents. Annals of Pharmacotherapy. 2011;45(7–8):898–909. [DOI] [PubMed] [Google Scholar]
- 13.LewinGroup. Evaluating Encounter Data Completeness-For Researchers using the Centers for Medicare & Medicaid Services Chronic Condition Data Warehouse (CCW). Falls Church, VA. May 2012. https://www.ccwdata.org/web/guest/ccw-medicaid-data-white-papers. Accessed August 22, 2017. [Google Scholar]
- 14.Area Health Resources Files. US Department of Health and Human Services, Health Resources and Services Administration, Bureau of Health Professions; Rockville, MD: 2011–2012. [Google Scholar]
- 15.National Survey of Mental Health Treatment Facilities. Center for Mental Health Services, Substance Abuse and Mental Health Services Administration. Rockville, MD; 2008. [Google Scholar]
- 16.Teh CF, Sorbero MJ, Mihalyo MJ, et al. Predictors of adequate depression treatment among Medicaid-enrolled adults. Health Services Research. February 2010;45(1):302–315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Stein BD, Sorbero MJ, Dalton E, et al. Predictors of adequate depression treatment among Medicaid-enrolled youth. Soc Psychiatry Psychiatr Epidemiol. May 2013;48(5):757–765. [DOI] [PubMed] [Google Scholar]
- 18.Work Group on Major Depressive Disorder, Gelenberg AJ, Freeman MP, et al. Practice Guideline for the Treatment of Patients with Major Depressive Disorder, 3rd Edition. Washington D.C.: American Psychiatric Association; 2010. [Google Scholar]
- 19.National Quality Measures Clearinghouse. Antidepressant Medication Management. In: Agency for Healthcare Research and Quality, ed. Vol 2017. Rockville, MD; 2013. [DOI] [PubMed] [Google Scholar]
- 20.Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? Journal of Health and Social Behavior. 1995:1–10. [PubMed] [Google Scholar]
- 21.Cummings JR, Ji X, Allen L, Lally C, Druss BG. Racial and Ethnic Differences in ADHD Treatment Quality Among Medicaid-Enrolled Youth. Pediatrics. June 2017;139 (6). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.US Census Bureau. Census 2000 Geographic Terms and Concepts. https://www2.census.gov/geo/pdfs/reference/glossry2.pdf. Accessed November 1, 2017.
- 23.Cummings JR, Wen H, Ko M, Druss BG. Geography and the Medicaid mental health care infrastructure: implications for health care reform. JAMA Psychiatry. 2013;70(10):1084–1090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cummings JR, Case BG, Ji X, Marcus SC. Availability of Youth Services in US Mental Health Treatment Facilities. Administration and Policy in Mental Health and Mental Health Services Research. 2016;43(5):717–727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Cummings JR. Contextual socioeconomic status and mental health counseling use among US adolescents with depression. Journal of Youth and Adolescence. 2014;43(7):1151–1162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Santiago CD, Kaltman S, Miranda J Poverty and Mental Health: How Do Low-Income Adults and Children Fare in Psychotherapy? Journal of Clinical Psychology. 2013;69(2):115–126. [DOI] [PubMed] [Google Scholar]
- 27.Syed ST, Gerber BS, Sharp LK. Traveling Towards Disease: Transportation Barriers to Health Care Access. Journal of Community Health. 2013;38(5):976–993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Cummings JR, Allen L, Clennon J, Ji X, Druss BG. Geographic Access to Specialty Mental Health Care Across High- and Low-Income US Communities. JAMA Psychiatry. May 01 2017;74(5):476–484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Cauce AM, Domenech-Rodriguez M, Paradise M, et al. Cultural and contextual influences in mental health help seeking: a focus on ethnic minority youth. Journal of Consulting and Clinical Psychology. February 2002;70(1):44–55. [DOI] [PubMed] [Google Scholar]
- 30.Schnittker J. Misgivings of medicine?: African Americans’ skepticism of psychiatric medication. Journal of Health and Social Behavior. 2003:506–524. [PubMed] [Google Scholar]
- 31.Lewis-Fernandez R, Das AK, Alfonso C, Weissman MM, Olfson M. Depression in US Hispanics: diagnostic and management considerations in family practice. J Am Board Fam Pract. Jul-Aug 2005;18(4):282–296. [DOI] [PubMed] [Google Scholar]
- 32.Takeuchi DT, Bui KV, Kim L. The referral of minority adolescents to community mental health centers. Journal of Health and Social Behavior. June 1993;34(2):153–164. [PubMed] [Google Scholar]
- 33.Cummings JR, Case BG, Ji X, Chae DH, Druss BG. Racial/ethnic differences in perceived reasons for mental health treatment in US adolescents with major depression. J Am Acad Child Adolesc Psychiatry. September 2014;53(9):980–990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.School-Based Health Alliance. 2013–2014 Digital Census Report. 2015; http://censusreport.sbh4all.org/#comprehensivecare. Accessed July 24, 2017.
- 35.Bains RM, Diallo AF. Mental Health Services in School-Based Health Centers: Systematic Review. J Sch Nurs. February 2016;32(1):8–19. [DOI] [PubMed] [Google Scholar]
- 36.Anglin TM, Naylor KE, Kaplan DW. Comprehensive school-based health care: high school students’ use of medical, mental health, and substance abuse services. Pediatrics. March 1996;97(3):318–330. [PubMed] [Google Scholar]
- 37.Juszczak L, Melinkovich P, Kaplan D. Use of health and mental health services by adolescents across multiple delivery sites. The Journal of Adolescent Health. June 2003;32(6 Suppl):108–118. [DOI] [PubMed] [Google Scholar]
- 38.Morse S. Billions in proposed budget cuts to Medicare, Medicaid threaten provider compensation. Healthcare Finance October 23, 2017. http://www.healthcarefinancenews.com/news/billions-proposed-budget-cuts-medicare-medicaid-threaten-provider-compensation. Accessed October 23, 2017. [Google Scholar]
- 39.Garfield R, Zur J. Medicaid Restructuring Under the American Health Care Act and Implications for Behavioral Health Care in the US. Menlo Park, CA: The Henry J. Kaiser Family Foundation. March 2017. Available at: http://files.kff.org/attachment/Issue-Brief-Medicaid-Restructuring-Under-the-American-Health-Care-Act-and-Implications-for-Behavioral-Health-Care-in-the-US. Accessed August 15, 2017. [Google Scholar]
- 40.Zur J, Musumeci M, Garfield R. Medicaid’s Role in Financing Behavioral Health Services for Low-Income Individuals. Menlo Park, CA: The Henry J. Kaiser Family Foundation. June 2017. Available at: http://www.kff.org/medicaid/issue-brief/medicaids-role-in-financing-behavioral-health-services-for-low-income-individuals/. Accessed August 15, 2017. [Google Scholar]
- 41.Carson NJ, Progovac AM, Wang Y, Cook BL. A decline in depression treatment following FDA antidepressant warnings largely explains racial/ethnic disparities in prescription fills. Depression and Anxiety. December 2017;34(12):1147–1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.O’Malley KJ, Cook KF, Price MD, Wildes KR, Hurdle JF, Ashton CM. Measuring diagnoses: ICD code accuracy. Health Services Research. 2005;40(5p2):1620–1639. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.