Mental Health Treatment and Work among African American and Caribbean Black Welfare Recipients

Abstract

Objective:

Untreated depression among Temporary Assistance for Needy Families (TANF) participants greatly reduces chances of securing and holding gainful employment.

Methods:

Logistic regression models were estimated on data describing 1,000 African American and Caribbean black TANF recipients and 2,123 African American and Caribbean Black non-TANF recipients obtained from the National Survey of American Life (NSAL).

Results:

Black TANF participants were more likely than Black non-TANF participants to be depressed and treated. Treatment odds were lower for Caribbean Black than for US born black TANF participants.

Conclusion:

Results indicated that mental health treatment was likely among black TANF participants if depression was identified. TANF participants working less than full-time did not receive as much treatment.

Poverty is strongly associated with mental illness and the patchwork of safety net poverty programs targeting the poor disproportionately serve mentally ill persons. Poverty programs can facilitate receiving mental health treatment by detecting mental illness through disability certification, implementing other screening procedures, and by facilitating acquisition of health care coverage. Because African Americans have higher poverty rates than whites (Semega, Fontenot, & Kollar, 2017), poverty programs serve as a mental health treatment portal for African Americans especially.

Temporary Assistance for Needy Families (TANF), the primary cash assistance program for families in financial distress is an often-overlooked but important safety net program. TANF is a federal-state partnership permitting states considerable discretion in state-designed programs within a framework of federal guidelines and requirements. TANF’s goal is to prepare participants for entry into the world of work (Committee on Ways and Means, 2016) and a key TANF provision is that support cannot be open-ended; participants must leave the program after 5 years and they cannot subsequently return. Mental illness is prevalent among TANF recipients (Danziger et al., 2000; Lennon, Blome, & English, 2001; Pavetti, Derr, Kauff, & Barrett, 2010; Siefert, Bowman, Heflin, Danziger, & Williams, 2000), and mental illness undermines TANF participants’ ability to seek and hold gainful employment.

TANF is considered successful because it greatly reduced AFDC enrollment (Committee on Ways and Means, 2016), and because it demonstrated that program requirements and services design can be decentralized, that is, placed largely under state control and financed from a fixed, block grant federal contribution. This approach to “welfare reform”, as conversion to TANF from AFDC is widely known, inspired some Medicaid “repeal and replace” proposals (Holahan, Buettgens, Pan, & Blumberg, 2017), as well as proposals to transform remaining safety net programs, including, Supplemental Assistance for Needy Families (SNAP, known as “Food Stamps”) and low-income housing assistance (Greenstein, 2016). However, “The combination of state flexibility and a capped federal block grant creates incentives for states to provide cash assistance to fewer families and to not create employment programs, which can be expensive if they target families with multiple employment challenges” (Hahn, Aron, Lou, Pratt, & Okoli, 2017). Thus, in some states, TANF participants receive few programmatic supports for job training and employment sustainability.

As officials set priorities for TANF programming, allocating scarce resources for meeting the needs of people often suffering from significant physical and mental health-related problems (Office of Family Assistance, 2009), it is important that they recognize untreated depression symptoms [e.g., lingering sadness; feeling fatigue or worthlessness; difficulty thinking or concentrating. See DSM-5 for complete definition (American Psychiatric Association, 2013)] as a key hindrance to finding and keeping work. Many mental health treatments, including pharmacotherapy and psychotherapies (Frank & Glied, 2007), can improve prospects for symptom management and recovery. Opportunities to finance mental health treatment for the poor have increased under the Affordable Care Act (ACA) both from expansion of the Medicaid program and from arrangements for subsidized purchase of coverage on state or federal insurance exchanges. Furthermore, mental health coverage at “parity” with general health coverage must be provided as one of ten essential insurance benefits (Beronio, Glied, & Frank, 2014).

Depression interferes with successful social and economic functioning, including finding and remaining employed (Danziger, Kalil, & Anderson, 2002; Derr, Douglas, & Pavetti, 2001; Jayakody, Danziger, & Pollack, 2000; Seefeldt & Orzol, 2005; Sweeney, 2000) and 59% of persons diagnosed with depression report severe or very severe impairment in meeting social role obligations (Kessler et al., 2003; Kessler & Bromet, 2013). Subsequent research continues to document high levels of functional impairment from depression (Hastings, Jones, & Martin, 2015; Kessler & Bromet, 2013; Rusch, Kanter, Manos, & Weeks, 2008; Segrin & Flora, 2000; Siefert et al., 2000). Worldwide, depression’s functional impairment is so great that it is a leading source of “Disease Burden” (Kessler & Bromet, 2013; Prince et al., 2007).

Depression rates are higher among poor and unemployed women than among other women (Danziger et al., 2002). For TANF participants with major depression, Danziger et al. (2000) found that they were significantly less likely than non-depressed recipients to be working (Danziger et al., 2000). When employed, depressed and other mentally ill TANF recipients held jobs for shorter periods than non-mentally ill TANF recipients. (Corcoran, Danziger, & Tolman, 2004). Arranging for African American TANF participant’s mental health treatment is crucial because states with greater African American proportional representation tend to design less generous TANF programs, placing participants under greatest pressure to find work assisted by the fewest support services (Hahn, Aron, Lou, Pratt, & Okoli, 2017).

Because 96% of TANF recipients are women, and 38.6% are African American — more than three times African American’s population representation -- considering TANF, depression, and treatment are especially important for understanding African American women’s mental health. By one calculation, TANF recipients received about 10% of all specialty mental health treatment provided to African American women (Acs & Loprest, 2007; Derr et al., 2001). Yet, reported national estimates suggesting the scope of depression and depression treatment among African American TANF recipients remain absent from the literature.

Using data from a representative national sample and focusing on black women solely for the present study, the current study sought to determine whether depression rates and treatment rates are greater among black TANF participants than among comparable others. Also examined in the study was whether black TANF recipients, who were failing to meet program expectations by not working full-time and who were depressed, were especially likely to be treated. Depression for the study was defined as a diagnosis according to American Psychiatric Association Diagnostic and Statistical manual criteria (American Psychiatric Association, 2013).

African American from Caribbean black TANF recipients were separated for several reasons. At least 20 percent of black population growth between 2000 and 2016 (U.S. Bureau of the Census, June 2017; U.S. Census Bureau, 2007) came from Caribbean immigrants who self-identify as having Caribbean Black origins. Capturing the racial differences in the research is important because it shows important heterogeneity differences in the black population. The literature reports that persons identifying racially as Caribbean report lower levels of mental health treatment (Escobar, Hoyos-Nervi, & Gara, 2000; Williams et al., 2007). We investigated whether lower treatment rates among Caribbean Blacks would characterize TANF participants, despite TANF’s, mechanisms for mental health screening, treatment referral, and financing -- presumably equally available to all.

Methods

Participants and Design

The analytic TANF sample was selected by choosing respondents with one or more children and answering “yes” to the following question: “Are you (or your family) currently receiving public assistance? (Aid to Families with Dependent Children [AFDC], TANF, or General Assistance [GA])?” AFDC is now TANF and any remaining former AFDC participants at the time of the study had become TANF participants. GA is a program for single, childless and adults and, by requiring that respondents have at least one child, GA participation was ruled out. The non-TANF sample comprised women not meeting TANF sample’s inclusion criteria. The study sample included 3,123 adult women.

African Americans were persons who self-identified as black but who did not report Caribbean ancestry. Caribbean blacks were persons who self-identified as black and endorsed any of these inclusion criteria: (1) they were of West Indian or Caribbean- descent, (2) they were of from a Caribbean-area country, or (3) they had parents or grandparents who were born in a Caribbean-area country. The Caribbean black sample was selected from residential areas that were sampled to reflect the distribution of the African American population and from additional metropolitan areas where Caribbean blacks composed more than 10% of the population.

Data

The National Survey of American Life (NSAL) was part of the National Institutes of Mental Health Collaborative Psychiatric Epidemiology Surveys initiative that included three nationally representative surveys: the NSAL, the National Comorbidity Survey Replication (NCS-R), and the National Latino and Asian American Study (NLAAS). The data for NSAL were collected through face-to-face interviews conducted in English using a computer-assisted personal interview system after participants provided written informed consent. Interviews lasted an average of 2 hours and 20 minutes (Jackson et al., 2004). The NSAL included a household probability sample of 3,570 African Americans, 1,621 blacks of Caribbean descent (hereafter referred to as Caribbean blacks), and 891 non-Hispanic whites aged 18 years and older. Because of TANF’s special importance for African Americans due to African American’s overrepresentation on TANF, the present study’s focus was on African Americans and it used the black American samples only for purposes of analysis.

Data were collected for the study between February 2001 and June 2003. The study’s overall response rate was 72% for whites, 71% for African Americans, and 78% for Caribbean blacks. The University Albany, Office for Pre-Award and Compliance Services granted approval for this study on March 28, 2017.

Measures

Dependent Variable

Mental health treatment was measured dichotomously (yes/no) in response to “Did you ever in your life have a session of psychological counseling or therapy that lasted 30 minutes or longer with any type of professional?”

Demographic and Control Variables

Age was measured as the number of years attained by the date of the interview; education was measured as the number of school years finished; household income was measured as total monetary resources (dollar amount) from all sources; ethnicity was coded African American = 1 and Caribbean black = 0; marital status categories were represented as a series of dummy-coded variables where married served as the primary reference group. Dummy coded variables were never married (1) versus all other marital statuses (0); partnered (1) versus all other marital statuses (0) and separated, divorced, and widowed (1) versus all other marital statuses coded as (0). Children were measured as the number of children between newborn and 12 years living in the household; Adolescents (13-17 years) were measured as the number living in the household; Household size was measured as the total number of persons living in the household; nativity was measured as the country respondents were born and coded as US-born (1) and Foreign-born (0). Employment was coded dichotomously; 1 represented less than full-time employment and 0 represented full-time employment. Less than full-time employment was based on 40 hours per work-week in the calendar year.

Depression was measured using the World Mental Health CIDI, a fully structured diagnostic interview, was used to evaluate a wide range of DSM-IV mental disorders (American Psychiatric Association, 1994). The psychiatric disorders assessed in the NSAL, which included major depressive disorder, are slightly modified versions of those developed for the World Mental Health project (WMH-CIDI) and was used in the National Comorbidity Survey Replication and the National Latino and Asian American Study (Kessler et al., 2004; Pennell et al., 2004). The algorithm for major depressive disorder (MDD) is the same as that for major depressive episode (MDE): criteria C, the presence or absence of a manic episode, is not considered. Major depressive disorder represented lifetime prevalence of depression and was calculated as the proportion of the entire sample reporting the outcome. Major depressive disorder was coded 1 if respondents reported the presence of depression and 0 otherwise.

Analysis Strategy

TANF and non-TANF subsamples for African American and Caribbean black women were first compared separately to determine whether depression rates and treatment rates were higher among TANF participants. Chi-square tests were employed to assess unadjusted depression rate and treatment rate differences, followed by logistic regression for controlled evaluation of these differences. Logistic regression models were estimated to present odds ratios after adjustment to establish correlates of depression and treatment (Hilbe, 2009) which also include age, education, annual household income, marital status, children, adolescents, African American vs. Caribbean Black status. The models were adjusted for depression where treatment served as the dependent variable.

Next, the logistic regression analysis evaluated whether TANF participants who were considered depressed and employed less than full-time could predict receipt of treatment. The model included a depression x employment interaction term to estimate whether women who were both depressed and employed less than full-time were especially likely to be treated. In this phase, the principal concern resided with TANF participants’ mental illness and treatment experiences because of the policy relevance given toward mental illnesses and treatment’s importance for meeting TANF policy goals. Logistic regression models were also performed to estimate an an equation for non-TANF participants as well as TANF participants in exploratory mode, and to provide a descriptive frame of reference for understanding TANF-based findings.

To further investigate findings revealed in earlier analysis, additional logistic regression models were estimated to determine whether length of exposure to US communities and cultures helps to explain ethnicity-related differences. The regression equations compared African American and Caribbean black women as well as nativity (US born vs. foreign born) which can be considered an important ethnicity-related, and possibly explanatory variable.

For computing regression equations, the data were weighted to adjust for differential probabilities of selection, non-response, and post-stratified to represent the black population accordingly. Stata 14 represented the software used to adjust the statistical tests for the survey design (StataCorp., 2016). Statistical significance was evaluated with 2-sided design-based tests at the α = .05 or smaller level of significance.

Results

Sample Demographics

The samples comprised 1,000 TANF participants and 2,123 African American and Caribbean black female non-TANF participants. Weighted demographic characteristics appear in Tables 1 and 2.

Table 1:

Demographic and socioeconomic characteristics by Race for TANF Recipients^* (n = 1, 000)

Characteristic	African American (n= 827)		Caribbean blacks (n= 173)		F	p-value
	$\stackrel{‒}{X}$	SE	$\stackrel{‒}{X}$	SE
Age (mean years)	40	(0.7)	40	(2.8)	0.00	0.95
Education (years)	12.1	(0.1)	12.2	(0.18)	0.19	0.67
Annual Household Income (mean $)	24,900	(1,724)	27,100	(2,640)	0.45	0.50
Number of Children	1.0	(0.6)	1.0	(0.1)	0.02	0.88
Number of Adolescents	0.5	(0.3)	0.4	(0.1)	1.17	0.28
Household Size	3.3	(0.7)	3.4	(0.2)	0.15	0.70
	%	SE	%	SE	χ2	p-value
Mental Health Service Use					4.83**	0.03
Yes	24.3	(2.3)	14.1	(3.4)
No	75.7	(2.3)	85.9	(3.4)
Depressed					0.10	0.75
Yes	18.9	(1.5)	20.5	(4.8)
No	81.1	(1.5)	79.5	(4.8)
Marital Status (%)					0.72	0.50
Married	22.0	(1.9)	16.6	(4.3)
Partner	11.9	(1.5)	18.6	(8.0)
Sep., Wid., Div.	29.9	(1.4)	36.1	(11.4)
Never Married	36.2	(2.1)	28.6	(8.0)
Employment Status					0.11	0.74
Full-Time Work	50.9	(1.7)	53.5	(7.7)
Less than Full-Time Work	49.1	(1.7)	46.5	(7.6)
nativity					209.2***	0.00
US-born	98.7	(0.5)	50.5	(8.3)
Foreign-born	1.3	(0.5)	49.5	(8.3)

^*Data source: National Survey of American Life (NSAL).

All calculations are weighted.

Table 2:

Demographic and socioeconomic characteristics by race for general population, non-TANF Recipients^* (n = 2,123)

Characteristic	African American (n= 1,353)		Caribbean blacks (n= 770)
	$\stackrel{‒}{X}$	SE	$\stackrel{‒}{X}$	SE	F	p-value
Age (mean years)	44	(0.7)	41	(1.3)	3.67	0.06
Education (years)	12.6	(0.1)	13.0	(0.2)	1.89	0.17
Annual Household Income (mean $)	36,900	(1,360)	44,800	(4,600)	2.71	0.11
Number of Children	1.0	(0.0)	1.0	(0.1)	0.15	0.70
Number of Adolescents	0.2	(0.0)	0.3	(0.0)	7.08*	0.01
Household Size	2.8	(0.1)	3.0	(0.1)	3.66	0.06
	%	SE	%	SE	χ2	p-value
Mental Health Service Use					0.50	0.48
Yes	14.4	(1.1)	12.7	(2.0)
No	85.6	(1.1)	87.3	(2.0)
Depressed					0.44	0.51
Yes	11.9	(10)	13.9	(2.9)
No	88.1	(1.0)	86.1	(2.9)
Marital Status (%)					2.07	0.12
Married	30.5	(1.5)	32.3	(3.2)
Partner	6.2	(0.8)	9.8	(2.8)
Sep., Wid., Div.	33.0	(1.4)	24.8	(3.4)
Never Married	30.3	(1.8)	33.1	(2.4)
Employment Status					4.63*	0.04
Full-Time Work	54.9	(1.6)	62.4	(3.0)
Less than Full-Time Work	45.1	(1.6)	37.6	(3.0)
nativity					1306.56***	0.00
US-born	98.3	(0.3)	67.3	(2.6)
Foreign-born	1.7	(0.3)	32.7	(2.6)

^*Data source: National Survey of American Life (NSAL).

All calculations are weighted.

Depression: Comparing African American and Caribbean Black TANF Participants With Other African American and Caribbean Black Women

African American TANF recipients (20.5%) were significantly more likely than other African American women (10.7%) to be depressed (χ² = 29.42, df = 1, p = 0.00). Similarly, Caribbean Black TANF recipients (20.9%) were significantly more likely than Non-TANF Caribbean Black women (9.2%) to be depressed (χ² = 15.49, df = 1, p = 0.00). Among TANF participants, the difference between Caribbean Black (20.9%) and African American TANF participants in depression rates (20.5%) was not statistically significant (χ² = 0.01, df = 1, p = 0.92).

Multivariate analysis confirmed these findings. In models predicting whether, TANF participants were more likely to be depressed than others and controlling for covariates as well as treatment rate differences, the depression rate differences proved statistically significant (OR= 4.53, 95% CI: 3.04 – 6.74). After entering interaction terms capturing whether the treatment rate difference between TANF and other black women varied according to whether women were African American or Caribbean black, results revealed no statistically significant interaction (OR = 0.76, 95%CI: .42 – 1.37).

Treatment: Comparing African American and Caribbean Black TANF Participants With Other African American and Caribbean Black Women

African American TANF participants (23.9%) were significantly more likely to be treated than non-TANF-participating African American women (12.3%) (χ² = 37.35, df = 1, p = 0.00). Caribbean Black TANF participants, too, were significantly more likely than Caribbean Black non-TANF participants to be treated (18.8% vs 10.2%), (χ² = 7.72, df = 1, p = 0.01). Among TANF participants, the difference between African American TANF participants (23.9%) and Caribbean Black TANF participants (18.8 %) in treatment rates was noticeable but did not reach statistical significance (χ² = 1.58, df = 1 p = 0.21).

In multivariate analysis (not shown) controlling for covariates as well as depression rate differences, TANF participants again proved more likely to be treated than others (OR = 4.72, 95%CI: 3.59 – 6.19). After entering interaction terms, to capture whether the treatment rate difference between TANF and other black women varied according to whether the women were African American or Caribbean black, results indicated absence of statistically significant interaction (OR = 1.25, 95%CI: .68 – 2.29).

Presented in Table 3 are results from logistic regression analyses for the TANF subsample and the larger subsample of African American women who did not participate in TANF. After controlling for covariates, the odds of receiving treatment were greater for depressed TANF participants (OR=4.53; 95%-CI: 3.05, 6.74) and for TANF participants working less than full time (OR=1.70; 95%-CI: 1.15, 2.50). Although Caribbean Black TANF participants were no longer less likely to receive treatment, foreign born TANF participants, who overwhelmingly Caribbean Black, were less likely than their US born counterparts to receive treatment. The depression x work interaction term, for evaluating whether TANF participants working less than full time were especially likely to receive treatment, was not statistically significant (OR =0.65; CI = 0.31, 1.35).

Table 3:

Odds of Mental Health Treatment for TANF and Non-TANF Recipients Controlling for Demographic Characteristics

	Entire TANF Sample (N=1,000)		Entire Non-TANF Sample (N = 2,123)
Independent Variables	Odds Ratio (SE)	95% CI	Odds Ratio (SE)	95% CI
Work
Less Than Full-Time (W)	1.70 (0.33)	[1.16 - 2.50]**	1.28 (0.22)	[0.91 - 1.81]
Full Time (Omit)
Depression (D)	4.53 (0.92)	[3.04 - 6.74]**	4.69 (0.91)	[3.20 - 6.85]***
Ethnic Background	1.19 (0.36)	[0.65 - 2.17]	0.66 (0.16)	[0.42 - 1.05]
African American
Caribbean black (Omit)	0.99 (0.01)	[0.97 - 1.01]	0.99 (0.01)	[0.97 - 0.99]**
Age (mean)
Education (years)	1.15 (0.05)	[1.05 - 1.26]**	1.12 (0.05)	[1.03 - 1.22]**
Annual Household Income (mean $)	1.00 (0.00)	[1.00 - 1.00]	1.00 (0.00)	[1.00 - 1.00]
Marital Status
Married (Omit)
Partner	0.64 (0.24)	[0.31 - 1.34]	1.79 (0.67]	[0.86 - 3.72]
Separated, Widowed, Divorced	0.95 (0.25)	[0.56 - 1.60]	1.70 (0.40)	[1.08 - 2.69]*
Never Married	0.72 (0.19)	[0.43 - 1.22]	1.08 (0.25)	[0.69 - 1.70]
Children	0.77 (0.07)	[0.65 - 0.92]**	0.83 (0.09)	[0.67 - 1.04]
Adolescents	0.84 (0.12)	[0.64 - 1.10]	1.06 (0.16)	[0.79 - 1.41]
Nativity
US-Born	2.05 (0.91)	[0.85 - 4.90]*	3.16 (0.89)	[1.82 - 5.50]***
Foreign-Born (ref.)

^*p <.05;

^**p <.01;

^***p <.000;

Source: National Survey of American Life (NSAL). All calculations are weighted.

For non-TANF African American and Caribbean Black women, the odds of receiving treatment were greater for depressed than for non-depressed women (OR=4.69; CI: 3.20, 6.85). However, there was no significant difference in treatment odds between less than full-time employed and other women. For non-TANF women, the depression x less than full-time work interaction was significant (OR = 2.85; CI: 1.36, 5.98), indicating that for depressed women, working less than full time increased odds of receiving treatment.

Sensitivity Analysis

To evaluate whether our findings generalized beyond depression, we repeated our multivariate analyses for a diagnostic category composed of all non-depression mental illnesses. Results exactly paralleled those found for depression. For TANF participants, having a non-depression mental illness significantly increased treatment odds (OR=3.77; CI: 2.51, 5.65), working less than full time increased treatment odds (OR=1.76; CI: 1.07, 2.89), and the depression x full time work interaction was not statistically sig (OR=0.65; CI: 0.31, 1.35). For non-TANF participants, having a non-depression mental illness increased treatment odds (OR = 3.34, CI: 2.01, 5.55). However, working versus not working full-time was not significantly related to receiving treatment (OR = 0.97, CI: 0.6, 1.44), and the mental illness x less than full-time work interaction was statistically significant (OR = 2.57; CI: 1.22, 5.40).

Discussion

In a representative national sample of African American women and employing DSM IV-TR criteria, we find that black TANF participants, whether African American or Caribbean Black, are more likely to be depressed than are other black women. This situation may come about for several reasons. TANF may attract women who cannot work due to depression or associated general health conditions. Also, TANF participants may be exposed more to depression-inducing stressors than others, including stress from meeting TANF requirements and deadlines and stigma from TANF participation. It also may be true that women who receive TANF have less non-financial resources from their family, friends, and or community to help mitigate some of the effects of poverty. Thus, TANF participation may be a marker of depression risk and population-based studies of depression, especially those studying African American populations, should evaluate TANF participation as a risk factor.

In keeping with greater treatment need, both African American and Caribbean black TANF participants but native-born women especially, were more likely than other African American and Caribbean black women to be treated. TANF programs’ arrangements to facilitate mental health treatment entry appear to bring more TANF participants into care than to comparable women who do not participate in TANF. In view of their high levels of depression treatment need (as reflected in high depression rates), TANF participants tend to receive more treatment than women who are not in TANF. This is important because TANF participants experience TANF-related pressure to overcome incapacity from depression and other disabling conditions to secure full-time employment and thus, needs appropriate treatment. Some TANF programs focused on providing mental health treatment have been documented (Derr et al., 2001) and the present findings suggest that enough TANF programs are responding to mental health needs of TANF participants for treatment to occur at levels exceeding those for non-TANF black women.

Although proportionally more depressed black TANF recipients were treated, many depressed TANF recipients remained untreated. Because of the program’s screening mechanisms and opportunities for Medicaid and subsidized insurance coverage purchase, TANF provides an important opportunity to detect and treat depression among a vulnerable population segment, poor black women. It is significant that most diagnosed depression remained untreated. That so many women needing treatment do not receive it suggests that this opportunity to provide help remains largely unexploited. TANF participant’s depression treatment must be assigned a high priority because states with proportionally more African Americans tend to design TANF programs that allow for shorter-term TANF participation (Hahn, Aron, Lou, Pratt, & Okoli, 2017). TANF resources in these states are especially meager, but the disability from untreated depression is so great that treatment is essential if depressed TANF participants are to become employed.

Unemployed African American and Caribbean black TANF participants were more likely than those who were employed to receive treatment, whether or not their distress met DSM criteria to be officially counted as clinical depression. It appears that some TANF programs may conceptualize psychological distress as a basis for referral more broadly than is described in the DSM. A more flexible approach might be appropriate in view of the urgent need to support TANF recipient’s job-finding and job-holding efforts. As such, TANF programs may prefer to err in favor of providing treatment rather than erring by withholding needed treatment.

Conceivably, TANF personnel either do not recognize or do not refer suspected depression as much among foreign-born as among US born depressed black women. This may occur because of non-US born Caribbean black’s reluctance to disclose mental illness-related suffering or because of screening or referral agent’s inability to recognize their symptoms. This also may occur because of non-US born Caribbean black’s mental illness-related mistrust and stigma, or because of a preference for culturally sanctioned styles of coping and traditional healing practices. Whatever the explanation, despite being under equal pressure to secure employment and depression’s interference with achieving that goal, immigrant Caribbean black TANF participants are more likely than other black women not to be treated. Further research is needed to ascertain why disproportionately many immigrant black TANF participants remain untreated.

Several limitations of the study should be kept in mind when considering the present findings. To qualify as receiving TANF, respondents reported that they received either TANF (or AFDC before TANF) or General Assistance (GA) as well as meeting a qualification for TANF only, having children who were living in the household. Under this procedure, any women with children living in the household and receiving GA would have been counted as TANF participants.

However, GA’s availability is limited: only 30 states offered GA programs in 2011 and only 12 offered programs statewide, providing very limited support largely to childless and very poor disabled adults lacking other personal or program support options (Schott & Cho, 2011). We cannot evaluate classification errors but we do not expect them to be large.

Another limitation stems from single item measures. Although common in epidemiologic and health services research, such questions are open to challenge on the basis of being lower in reliability than multi-item questions.

We were unable to document the mix of treatments provided and cannot determine which treatments, or which aspects, account for our results. In some cases, “treatment” may have consisted of mental health counseling not provided by mental health treatment specialists or merely addressing job-related concerns. Crucially, we cannot ascertain how much evidence-based psychotherapy and pharmacotherapy treatments most likely to be effective, were provided. Further research is needed to identify treatments targeted toward TANF participants. And it is especially important to determine which treatments increase employment prospects, as well as otherwise improving quality of life.

The present study finding also is unable to suggest whether the results apply to white as well as non-African American ethnic minority women. To acknowledge that African American TANF recipients are a high priority for research, and for other mental health research, is not to deny that white TANF recipients and recipients from other ethnic groups also are important to study. As an initial step, efforts are needed to replicate our findings on national data for whites and other ethnic groups. Future research efforts may also wish to explore how TANF recipients incorporate non-financial resources to lessen the effects of poverty. The results from the present study only begin a process of seeking to understand the role of mental health treatment in helping TANF recipients to become employed. Despite these limitations, the present study advances knowledge in this understudied area of inquiry.

African American women carry great family and community responsibility in African American communities, often with limited significant other support (Brown, Brody, & Stoneman, 2000; Ennis, Hobfoll, & Schroder, 2000; Schulz, Israel, et al., 2006; Todd & Worell, 2000). There are fewer employment opportunities in many African American communities than elsewhere, and expiring time limits are especially a burden where opportunities for gainful employment are limited (Cutrona et al., 2005; Schulz, Gravlee, et al., 2006; Wilson, 1996). Another factor that may influence engagement in the workforce is the women’s ability to recognize the need for depression treatment. It is not well understood in the literature that Black women receiving TANF see themselves as depressed or (stressed because they are poor) and as such, their opinions and behaviors toward engaging in mental health treatment might be influenced. More research needs to be conducted to uncover additional influences. Thus, as freedom from depression does not bring health benefits for blacks equal to those for whites (Assari, 2018), it is unlikely to bring equal employment prospects. However, depression treatment works equally well for blacks and whites (Lesser et al., 2011) and there is every reason to believe that freedom from depression improves African American women’s employment prospects beyond what they are when the women are actively depressed, and otherwise improves their quality of life. For African American TANF recipients, stakes are high and access to effective treatment is important.