Abstract
Background and Purpose
Although post-stroke depression is common, racial-ethnic disparities in depression among stroke survivors remain under-explored. Thus, we investigated the relationship between race/ethnicity and depression in a multi-racial-ethnic stroke cohort.
Methods
Baseline survey data of validated scales of depression and functional status, demographics, comorbidities and socioeconomic status were used from a recurrent stroke prevention study among community-dwelling urban stroke/TIA survivors.
Results
The cohort included 556 participants with a mean age of 64 years. The majorities were Black (44%) or Latino (42%), female (60%), had their last stroke/TIA nearly two years prior study enrollment, and lived below the poverty level (58%). Nearly one in two Latinos, one in four Blacks and one in eight Whites were depressed. Multivariate logistic regression showed that survivors who were younger, female, had three or more comorbid conditions, were functionally disabled from stroke, lacked emotional or social support, and who took antidepressants prior to study entry had higher risk of depression. Time since last stroke/TIA did not affect the chance of depression. After adjusting for all above risk factors, Latinos had three times the odds of depression (95% CI: 1.18 – 6.35) than Whites; Blacks and Whites had similar odds of depression.
Conclusion
This study reveals that Latino stroke survivors have a significantly higher prevalence of depression compared to their non-Latino counterparts.
Keywords: Depression, Prevalence, Racial/Ethnic Disparity
Introduction
Post-stroke depression (PSD) is prevalent, underdiagnosed and one of the most commonly ignored complications of stroke.1 Racial-ethnic minorities have higher incidence and recurrence of stroke, poorer clinical outcomes, poorer functional scores and 1-year all-cause re-hospitalization after acute ischemic strokes.2, 3 Though data are limited, there appear to be racial-ethnic disparities in the prevalence of depression in the general population; Blacks (14.2%), and Hispanics (14.6%) have a higher prevalence of depression than Whites (10.9%).4 Yet, little is known about disparities in PSD. Most prior studies of racial-ethnic differences in PSD have relied heavily on hospital registry data alone.5, 6
To our knowledge, no previous studies have disaggregated PSD prevalence in racial-ethnic minorities. We conducted a comprehensive study of PSD using surveys of community-dwelling stroke/TIA survivors in order to examine racial-ethnic differences in PSD among a diverse, urban population.
Methods
For this secondary analysis, we used data from the Prevent Recurrent All-Inner city Stroke through Education (PRAISE) study conducted in New York City between 2009 and 2012, using a community-based participatory research approach, discussed in detail elsewhere.7 This study received Institutional Review Board approval.8 For the survey, the team recruited cognitively intact adults (screened with an adaptation of the Blessed Dementia Scale) who had a stroke/TIA within the past five years, from local hospitals, health centers, a visiting nurse service and community organizations such as churches and senior centers. We evaluated PSD using the Patient Health Questionnaire-8, a commonly used scale validated in English and Spanish.9, 10 In the current sample, the internal consistency of the eight PHQ questions was reliable with a Cronbach’s alpha of 0.78. We defined participants as having PSD if they had PHQ-8 sum score ≥ 10.11 We assessed race/ethnicity through self-report, categorized into three groups: non-Hispanic White (White), non-Hispanic Black (Black), and Latino. We excluded those not self-identified in one of the three groups due to their very small numbers. Other demographic covariates included age, gender, socioeconomic status (SES, including income and insurance status) and current smoking status.
The team used other validated instruments to assess stroke-related disability with the modified Rankin scale (mRS) (0–2 functional, 3–4 disabled),12 comorbidity with the Charlson Index,13 and emotional support with a single question, “How many people can you count on to give you emotional support, such as talking over problems to help you with a difficult decision?”.14 Direct prescription fill records of antidepressant medications obtained from participants’ pharmacies for the one year prior to the study were used as an indicator of antidepressant use before joining the study.
Statistical analysis
Variables were summarized as mean ± standard deviation or proportion. Bivariate analyses were conducted using student t-tests, and chi-square tests for continuous and categorical variables respectively. Multivariate logistic regression was performed to assess racial-ethnic differences in PSD, by adjusting for independent risk factors significantly associated with PSD in addition to potential confounders associated with both race/ethnicity and PSD in bivariate analyses. All analyses were conducted in SAS 9.3 (SAS Institute, Inc. Cary, NC). Statistical significance was assessed at the 0.05 level.
Results
As shown in Table 1, the 556 participants included in the study had an average age of 63.5 (SD=11.2) years, 60% were women, 14% White, 44% Black, and 42% Latino. Of the 231 Latinos, 61% spoke Spanish only, 53% reported their place of origin as Puerto Rico, 29% as the Dominican Republic, and the remainder were from other Caribbean, Central or South American nations. Participants had 1.6 stroke/TIAs on average, and the majority (80%) had had their last stroke/TIA more than a year ago. Nearly half were disabled from their stroke/TIA, as indicated by a mRS of 3 or 4, and nearly two thirds had three or more comorbid conditions. Only one third were married or living with a partner, but nearly all reported having emotional social support.
Table 1.
Patient Characteristics by PSD
| Total (N=556) |
Without PSD N=386 (69%) |
With PSD N=170 (31%) |
p | |
|---|---|---|---|---|
| Socio Demographics | ||||
| Age, mean ± SD (years) | 63.5 ± 11.2 | 64.6 ± 11.1 | 60.8 ± 10.9 | 0.0002 |
| Female | 333 (60%) | 214 (55%) | 119 (70%) | 0.0012 |
| Race/Ethnicity | <.0001 | |||
| White | 80 (14%) | 70 (18%) | 10 (6%) | |
| Black | 245 (44%) | 180 (47%) | 65 (38%) | |
| Hispanic | 231 (42%) | 136 (35%) | 95 (56%) | |
| Insurance | 0.0006 | |||
| Medicaid | 172 (31%) | 103 (27%) | 69 (41%) | |
| Medicare | 219 (39%) | 150 (39%) | 69 (41%) | |
| Commercial | 137 (25%) | 111 (29%) | 26 (15%) | |
| Other | 28 (5%) | 22 (6%) | 6 (4%) | |
| Low Income (≤ $15,000/year) | 319 (58%) | 201 (52%) | 118 (69%) | 0.0002 |
| Married/Live with partner | 186 (33%) | 140 (36%) | 46 (27%) | 0.0339 |
| Have emotional-social support | 504 (91%) | 359 (93%) | 145 (85%) | 0.0040 |
| Health Status & Behavior | ||||
| Disabled (mRS 3–4) | 263 (47%) | 151 (39%) | 112 (66%) | <.0001 |
| High Comorbidity (Charlson ≥ 3) | 347 (62%) | 224 (58%) | 123 (72%) | 0.0013 |
| Year since stroke, mean ± SD | 1.8 ± 1.5 | 1.8 ± 1.4 | 1.7 ± 1.5 | 0.5640 |
| # strokes/TIAs, mean ± SD | 1.6 ± 1.2 | 1.6 ± 1.1 | 1.8 ± 1.2 | 0.0608 |
| Pre-study anti-depressant usage | 152 (27%) | 87 (23%) | 65 (38%) | 0.0001 |
| Smoking | 95 (17%) | 56 (15%) | 39 (23%) | 0.0136 |
About one-third (31%) of the participants were currently experiencing PSD, and about one fourth were taking antidepressant medications prior to study enrollment. Participants who had a stroke/TIA within a year and those had more than a year ago had similar rates of PSD 33% vs. 30% (p=0.47). The rate of PSD differed significantly across racial-ethnic groups: Latino (41%), Black (26%), and White (13%), p<.0001. In comparison, rates of depression in the general adult population do not show such large disparity: Latino (15%), Black (14%) and White (11%).4 Individuals of Puerto Rican origin had a higher prevalence of PSD compared to other Latinos (49% vs. 32%, p=0.01). Subjects with PSD were about 4 years younger than non-PSD subjects (61 vs. 65 year, p=0.0002), more were female (70% vs. 55%, p=0.001), had greater than 3 comorbidities (72% vs. 58%, p=0.001) and were disabled (66% vs. 39%, p<.0001; Table 1).
Black and Latino participants were about eight years younger, on average, than Whites (p<.0001). A higher proportion of Latino and Black participants had 3 or more comorbid conditions than Whites (70% vs. 61% vs. 45% respectively, p=0.0004); were disabled by stroke (54% vs. 49% vs. 23%, p<.0001); Latinos also received antidepressant medications prior to joining the study at a higher rate than Whites and Blacks, 32% vs. 28% and 22% respectively (p=0.0496). Other demographic and clinical comparisons by racial-ethnic groups can be found in Table 2.
Tables 2.
Demographics and Biomarkers by Race/Ethnicity
| Black N=245 (44%) |
Latino N=231 (42%) |
White N=80 (14%) |
p | |
|---|---|---|---|---|
| Age, mean ± SD (years) | 62 ± 11 | 63 ± 11 | 71 ± 10 | <.0001 |
| Female | 160 (65%) | 131 (57%) | 42 (53%) | 0.0555 |
| Insurance | <.0001 | |||
| Medicaid | 84 (34%) | 83 (36%) | 5 (6%) | |
| Medicare | 88 (36%) | 111 (48%) | 20 (25%) | |
| Commercial | 62 (25%) | 25 (11%) | 50 (63%) | |
| Other | 11 (4%) | 12 (5%) | 5 (6%) | |
| Low Income (≤ $15,000/year) | 142 (58%) | 167 (73%) | 10 (13%) | <.0001 |
| Have emotional-social support | 224 (91%) | 207 (90%) | 73 (91%) | 0.7774 |
| Disabled (mRS 3–4) | 120 (49%) | 125 (54%) | 18 (23%) | <.0001 |
| High Comorbidity (Charlson ≥ 3) | 150 (61%) | 161 (70%) | 36 (45%) | 0.0004 |
| Years since stroke, mean ± SD | 1.8 ± 1.5 | 1.7 ± 1.4 | 2.2 ± 1.5 | 0.0137 |
| # strokes/TIAs, mean ± SD | 1.6 ± 1.0 | 1.7 ± 1.3 | 1.5 ± 1.1 | 0.3300 |
| Pre-study anti-depressant usage | 55 (22%) | 75 (32%) | 22 (28%) | 0.0496 |
| Smoking | 54 (22%) | 32 (14%) | 9 (11%) | 0.0183 |
Multivariate logistic regression showed that survivors who were female (OR=1.64; 95%CI: 1.07–2.53), functionally disabled from stroke (OR=2.43; 95%CI: 1.59–3.70), had 3 or more comorbid conditions (OR=1.55; 95%CI: 1.00–2.40), and taking antidepressant prior study entry (OR=1.85; 95%CI: 1.20–2.85) had higher risk of PSD; those of older age (OR=0.97; 95%CI: 0.94–0.99) and had emotional social support (OR=0.38; 95%CI: 0.20–0.73) were less likely to have PSD; time since last stroke/TIA did not affect the chance of depression (OR=1.00; 95%CI: 0.87–1.15). After adjusting for all above protective/risk factors, the odds of PSD for Latinos were 2.75 (95% CI: 1.18–6.35) times higher than that of Whites; the odds of PSD were similar for Blacks compared to Whites (OR=1.33; 95% CI: 0.58–3.05). Detailed unadjusted and adjusted results are shown in Table 3. Additional multivariate logistic regression also confirmed that Puerto Ricans had 4.5 times higher odds (95%CI: 1.8–11.2) of PSD compared to Whites, while Other Hispanics had 3 times higher odds (95%CI: 1.1–8.2) of PSD compared to Whites; participants spoke Spanish only had similar odds of having PSD compared to those who spoke English.
Table 3.
Odds Ratios With 95% Confidence Intervals in Predicting PSD
| Unadjusted model |
Adjusting for social demographics |
Adjusting for clinical risks |
Adjusting for social- demographic and clinical risks |
|
|---|---|---|---|---|
| White | Ref | Ref | Ref | Ref |
| Black | 2.53 (1.23–5.20) | 1.47 (0.67–3.25) | 1.90 (0.89–4.06) | 1.33 (0.58–3.05) |
| Latino | 4.89 (2.40–9.97) | 3.02 (1.36–6.70) | 3.60 (1.70–7.60) | 2.75 (1.18–6.35) |
| Age | 0.97 (0.95–0.99) | 0.97 (0.94–0.99) | ||
| Male | Ref | Ref | ||
| Female | 1.92 (1.28–2.89) | 1.64 (1.07–2.53) | ||
| Income > $15,000/year | Ref | Ref | ||
| Low income | 1.26 (0.78–2.01) | 1.14 (0.70–1.86) | ||
| Commercial insurance | Ref | Ref | ||
| Medicaid | 1.38 (0.72–2.64) | 0.93 (0.47–1.85) | ||
| Medicare | 1.34 (0.73–2.47) | 1.02 (0.54–1.95) | ||
| Other | 0.81 (0.28–2.38) | 0.86 (0.28–2.62) | ||
| No emotional social support | Ref | Ref | ||
| Have support | 0.43 (0.23–0.79) | 0.38 (0.20–0.73) | ||
| Year since stroke/TIA | 0.99 (0.86–1.13) | 1.00 (0.87–1.15) | ||
| Charlson comorbidity < 3 | Ref | Ref | ||
| ≥ 3 | 1.43 (0.94–2.17) | 1.55 (1.00–2.40) | ||
| Functional (mRS 0–2) | Ref | Ref | ||
| Disabled (mRS 3–4) | 2.48 (1.67–3.68) | 2.43 (1.59–3.70) | ||
| No antidepressant use | Ref | Ref | ||
| Used prior to study | 1.85 (1.22–2.81) | 1.85 (1.20–2.85) | ||
| Non-smoking | Ref | Ref | ||
| Smoking | 1.73 (1.06–2.83) | 1.44 (0.85–2.44) |
Discussion
We found that disproportionately more Latino stroke survivors have PSD than Whites and Blacks, and this racial-ethnic difference is widened among stroke/TIA survivors than in the general population. Puerto Ricans have strikingly higher rate of PSD as compared to other Latinos, this seems to exacerbate their increased general prevalence of depression.15
The lack of Black-White difference is consistent with the findings of Robinson and Price.5 Other studies have found that Blacks were less likely to be diagnosed with PSD, and to receive treatment for depression if diagnosed.6 However, our definition of depression was based on patient self-report, not clinical reporting, allowing us to include undiagnosed depression.
Our study had several strengths. We screened stroke survivors from a community setting with a high proportion of minority patients, enabling us to have a better understanding of racial-ethnic disparities of PSD. The in-depth information about place of origin suggested within-group heterogeneity in PSD, and could help to target specific vulnerable Latino subgroups. Our study included pharmacy records for the year prior to study enrollment, which allowed us to control for pre-enrollment depression, since depression itself is an important and significant predictor of PSD.
There were several limitations to this study. We had no data on lesion location or information from neuroimaging, and some literatures show that location may be associated with PSD. There may be a selection bias of who participated. Patients who declined or were unable to participate due to their physical or mental state, may have given us an underestimate of the prevalence of PSD. These are estimates of PSD prevalence about two years post stroke on average, so our findings do not capture PSD immediately post stroke, and PSD prevalence varies overtime. Lastly, as is the case for most non-randomized studies, there is always the potential for bias as a result of unmeasured confounding.
Conclusions
Our study reveals a significantly higher rate of PSD among Latinos, as compared with Whites and Blacks, after adjusting for demographic and clinical risk factors. Further elucidation of possible modifiable factors to reduce PSD burden is needed given the significant burden of recurrent stroke in this population. Researchers and clinicians could consider studying and providing targeted screening and treatment of PSD, especially among Latino stroke survivors.
Acknowledgements
We would like to thank the patients who participated in the intervention, and the East and Central Harlem Health Outcomes Community Advisory Board, for their essential roles in this study.
Funding Sources:
NIH/National Institute on Minority Health and Health Disparities: P60 MD000270; NIH/National Center for Advancing Translational Sciences: UL1 TR000067; New York State Department of Health Empire State Clinician Research Program.
Footnotes
Disclosures: None.
References
- 1.Paolucci S. Epidemiology and treatment of post-stroke depression. Neuropsychiatr Dis Treat. 2008;4:145–154. doi: 10.2147/ndt.s2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ottenbacher KJ, Campbell J, Kuo YF, Deutsch A, Ostir GV, Granger CV. Racial and ethnic differences in postacute rehabilitation outcomes after stroke in the united states. Stroke. 2008;39:1514–1519. doi: 10.1161/STROKEAHA.107.501254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Qian F, Fonarow GC, Smith EE, Xian Y, Pan W, Hannan EL, et al. Racial and ethnic differences in outcomes in older patients with acute ischemic stroke. Circ Cardiovasc Qual Outcomes. 2013;6:284–292. doi: 10.1161/CIRCOUTCOMES.113.000211. [DOI] [PubMed] [Google Scholar]
- 4.Yang F-D. Racial differences in the prevalence of depressive disorders among U.S. Adult population. [Accessed October 20, 2015];Georgia State University. 2012 http://scholarworks.gsu.edu/math_theses/125/. [Google Scholar]
- 5.Robinson RG, Price TR. Post-stroke depressive disorders: A follow-up study of 103 patients. Stroke. 1982;13:635–641. doi: 10.1161/01.str.13.5.635. [DOI] [PubMed] [Google Scholar]
- 6.Jia H, Chumbler NR, Wang X, Chuang HC, Damush TM, Cameon R, et al. Racial and ethnic disparities in post-stroke depression detection. Int J Geriatr Psychiatry. 2010;25:298–304. doi: 10.1002/gps.2339. [DOI] [PubMed] [Google Scholar]
- 7.Goldfinger JZ, Kronish IM, Fei K, Graciani A, Rosenfeld P, Lorig K, et al. Peer education for secondary stroke prevention in inner-city minorities: Design and methods of the prevent recurrence of all inner-city strokes through education randomized controlled trial. Contemp Clin Trials. 2012;33:1065–1073. doi: 10.1016/j.cct.2012.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kronish IM, Goldfinger JZ, Negron R, Fei K, Tuhrim S, Arniella G, et al. Effect of peer education on stroke prevention: The prevent recurrence of all inner-city strokes through education randomized controlled trial. Stroke. 2014;45:3330–3336. doi: 10.1161/STROKEAHA.114.006623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of prime-md: The phq primary care study. Primary care evaluation of mental disorders. Patient health questionnaire. JAMA. 1999;282:1737–1744. doi: 10.1001/jama.282.18.1737. [DOI] [PubMed] [Google Scholar]
- 10.Diez-Quevedo C, Rangil T, Sanchez-Planell L, Kroenke K, Spitzer RL. Validation and utility of the patient health questionnaire in diagnosing mental disorders in 1003 general hospital spanish inpatients. Psychosom Med. 2001;63:679–686. doi: 10.1097/00006842-200107000-00021. [DOI] [PubMed] [Google Scholar]
- 11.Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The phq-8 as a measure of current depression in the general population. J Affect Disord. 2009;114:163–173. doi: 10.1016/j.jad.2008.06.026. [DOI] [PubMed] [Google Scholar]
- 12.Banks JL, Marotta CA. Outcomes validity and reliability of the modified rankin scale: Implications for stroke clinical trials: A literature review and synthesis. Stroke. 2007;38:1091–1096. doi: 10.1161/01.STR.0000258355.23810.c6. [DOI] [PubMed] [Google Scholar]
- 13.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis. 1987;40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 14.Berkman LF, Glass T. Social integration, social networks, social support, and health. In: Berkman LF, Kawachi I, editors. Social epidemiology. New York, NY: Oxford University Press; 2000. pp. 137–173. [Google Scholar]
- 15.Moscicki EK, Rae D, Regier DA, Locker B. The Hispanic health and nutrition examination survey: Depression among Mexican Americans, Cuban Americans, and Puerto Ricans. In: Gaviria M, Arana JD, editors. Health and behavior: Research agenda for Hispanics (simón bolivar research monograph series i) Chicago, IL: University of Illinois at Chicago; 1987. pp. 145–159. [Google Scholar]