Table 3 Research on mental health and stigma in LMICs, 2008–2017
Study (First author, year [ref.]) | Location | Sampling characteristics | Sample size | Study design | Type of stigma assessed | Description of stigma association (strength, significance) | Significant mediators/moderators |
|---|---|---|---|---|---|---|---|
Adewuya, 2009 [94] | Nigeria | Facility-based sample; any disorder | 342 | Cross-sectional | Internalized (ISMI) | Poor medication adherence for high relative to low stigma | None |
Assefa, 2012 [93] | Ethiopia | Facility-based sample; schizophrenia | 212 | Cross-sectional | Internalized (ISMI) | Discontinuation of psychotropic medication for high relative to low stigma Psychotic symptoms for high relative to low stigma Suicide attempt for those with high relative to low stigma | None |
Ethiopia | Facility-based sample; schizophrenia | 411 | Cross-sectional | Perceived (PDD), resistance (ISMI-SR) | Poor antipsychotic medication adherence for high perceived relative to low perceived stigma and for high relative to low stigma resistance Duration of illness less than 1 year for high relative to low perceived stigma (NS for stigma resistance) Poor follow-up care NS for perceived stigma or stigma resistance | None | |
Cai, 2017 [188] | China | Facility-based sample; schizophrenia | 172 | Cross-sectional | Internalized (ISMI) | Stigma not associated with quality of life | None |
Dardas, 2017 [106] | Jordan | School-based; depression | 2349 | Cross-sectional | Personal and perceived (DSS) | Stigma associated with care seeking | Mod: Significant interaction between stigma and depression for willingness to seek help |
Devi Thakoor, 2016 [189] | China, Mauritius | Facility-based sample; SMI | 300 | Cross-sectional | Internalized (ISMI) | Duration of psychosis of greater than 3 months relative to less than 3 months was associated only with the following ISMI items: increased perceived break up due to illness and increased perceived disinheritance due to illness by family (China); decreased patient awareness of illness and decreased family awareness of illness (Mauritius) | None |
Elkington, 2010 [92] | Brazil | Facility-based sample; SMI | 98 | Cross-sectional | Internalized, experienced discrimination, perceived (SPISEW) | Significantly higher mean personal experiences of stigma score for individuals in the mild to moderate vs. moderate to marked illness severity group Perceived attractiveness and relationship discrimination stigma scales were NS HIV risk and protective behaviors associated with relationship discrimination for sexual activity, unprotected sex, and fewer partners; perceived attractiveness – all NS; personal experiences – all NS | None |
Fawzi, 2016 [91] | Egypt | Facility-based sample; depression | 196 | Cohort | Internalized (ISMI) | Treatment acceptance: patients refusing treatment had a higher stigma score than those who accepted treatment Diabetes: increase in ISMI score was associated with change in fasting plasma glucose and standardized 8-week percentage change in HbA1c levels in multiple regression analyses | None |
Fresan, 2017 [190] | Mexico | Facility-based sample; schizophrenia | 217 | Cross-sectional | Perceived and experienced discrimination (KSS) | Length of hospitalization increase of 1 week associated with KSS score Duration of untreated psychosis was NS | None |
Grover, 2017 [99] | India | Facility-based sample; SMI | 1403 | Cross-sectional | Internalized (ISMI) | Shorter duration of illness was significantly correlated with higher overall internalized stigma among patients with schizophrenia, but the SE and DE subscales were NS; overall stigma and all subscales were NS among patients with recurrent depressive disorder Shorter duration of treatment was significantly correlated with higher overall internalized stigma among patients with schizophrenia, but the SE and SR subscales were NS Among patients with recurring depression, higher overall internalized stigma was significantly correlated, but the SE, DE, and SR subscales were NS Lesser symptom severity among patients with schizophrenia, as measured by the PANSS-P, was significantly correlated with overall stigma and only the SR subscale was NS; however, the PANSS-N and PANSS general psychopathology scales were NS with overall stigma For patients with depression as measured by the HDRS, overall stigma was significantly correlated, but not the SE, SW, or SR subscales Greater participation restriction was significantly correlated with overall stigma score and all subscales among patients with schizophrenia; for patients with depression, overall stigma score was significant, but the A and SE subscales were NS | None |
Koschorke, 2014 [101] | India | Schizophrenia | 282 | Cross-sectional | Anticipated and experienced discrimination (DISC) | Symptom severity, as measured by total PANSS score, was NS in association with discrimination; however, belonging in a higher PANSS-N quartile was associated with reduced odds of experiencing negative discrimination, while belonging in a higher PANSS-P quartile was associated with increased odds of experiencing negative discrimination | None |
Kulesza, 2014 [102] | India | Facility-based sample; majority exhibited depression | 60 | Cross-sectional | Anticipated and perceived (EMIC-SS) | Symptom severity for depression was positively correlated with stigma | None |
Lahariya, 2010 [97] | India | Facility-based sample; SMI | 295 | Cross-sectional | One question on fear of stigma related to care seeking | Delay in care seeking: 73% of patients had delayed seeking care at least in part due to a fear of stigma | None |
Li, 2017 [88] | China | Facility-based sample; schizophrenia | 384 | Cross-sectional | Internalized (ISMI) | Psychiatric symptoms: Stigma significantly increased with an increase in general symptoms measured via the BPRS in multiple regression analyses; PANSS-N NS Functioning: Stigma score significantly decreased with an increase in GAF in multiple regression analyses Quality of life: Stigma score significantly increased b = 0.01 (0.01–0.02) with an increase in SQLS score in multiple regression analyses | None |
Loch, 2012 [191] | Brazil | Facility-based sample; mostly SMI | 169 | Cohort | Question on dangerousness stereotyping | Re-hospitalization: Individuals who were readmitted over the year were significantly more likely to be stereotyped as dangerous by family members that those who were not readmitted | None |
Lu, 2012 [192] | China | Facility-based sample; schizophrenia | 92 | Cross-sectional | Internalized (ISMI); experienced discrimination (MCESQ) | Insight: MCESQ and ISMI total score was NS in multiple regression with insight as the outcome | None |
Lv, 2013 [100] | China | Facility-based sample; schizophrenia | 95 | Cross-sectional | Internalized (ISMI) | Symptom severity: Positive and negative symptoms of psychosis both NS Greater duration of illness was associated with a change in stigma score; greater number of hospitalizations was NS; greater quality of life was associated with a change in stigma score | None |
Mosanya, 2014 [98] | Nigeria | Facility-based sample; schizophrenia | 256 | Cross-sectional | Internalized (ISMI) | Medication side effects, comorbid medical problem, duration of illness, and number of episodes all NS Increase in BPRS score increased the odds of having high vs. low stigma Individual in the high stigma group had significantly lower mean quality of life as measured by all WHOQOL-Brief subscales (physical, psychological, social, and environment) as well as the overall quality of life and general health | None |
Rayan, 2017 [103] | Jordan | Facility-based sample; depression | 160 | Cross-sectional | Perceived (PDD) | Pain was NS An increase in number of relapses was associated with a significant change in stigma score Symptom severity of depression was associated with a significant change in stigma score | None |
Rayan, 2017 [104] | Jordan | Facility-based sample; schizophrenia | 161 | Cross-sectional | Perceived (PDD) | In a multivariate regression, increase in stigma was associated with a significant reduction in quality of life Symptom severity for depression was significantly correlated with stigma | None |
Roberts, 2017 [96] | Ukraine | Community-based time-location sampling; depression, anxiety or PTSD | 2203 | Cross-sectional | One question on stigma related to care seeking | Out of the 703 people with a mental health problem, only 180 (25.6%) had sought care from any medical source (including pharmacists, or NGO counselling center); of the 520 who did not seek care, 41 attributed this to stigma or embarrassment (8%) | None |
Sharaf, 2012 [107] | Egypt | Facility-based sample; schizophrenia | 200 | Cross-sectional | Internalized (ISMI) | In multivariate regression, increase in stigma was associated with increase in suicide risk Insight was correlated positively with stigma | Mod: Insight was measured but not a significant moderator of stigma–suicide relationship |
Shi-Jie, 2017 [90] | China | Facility-based sample; depression | 158 | Cross-sectional | Anticipated and perceived (EMIC) | The depression subscale of the SCL-90 was associated with a significant increase in stigma in multivariate regression MADRS, somatization, and the SCL-90 total and anxiety subscale score were all NS; fatigue was associated with a significant increase in stigma in multivariate regression; disability NS in multivariate regression; duration of illness NS in multivariate regression | None |
Singh, 2016 [89] | India | Facility-based sample; schizophrenia | 100 | Cross-sectional | Internalized (ISMI); anticipated and perceived (EMIC) | Functioning was significantly associated with decrease in all ISMI subscales in regression analyses except ISMI-A and ISMI-SR Increase in GAF score was associated with reduced odds of having high vs. low overall ISMI score Functioning was negatively correlated with EMIC score Duration of illness was NS in regression analyses, except an increase in duration was associated with increased odds of having high vs. low ISMI-SR score; treatment duration was NS Symptom severity was NS in regression analyses, except an increase in the general PANSS subscale was associated with increased odds of having high vs. low ISMI-A score | None |
Vidojevic, 2015 [193] | Serbia | Facility-based sample; depression | 52 | Cross-sectional | Anticipated and experienced discrimination (DISC) | Hospitalization history was associated with higher discrimination and lower ability to overcome stigma | None |
Wang, 2017 [194] | China | Facility-based sample; schizophrenia | 146 | Cross-sectional | Perceived and internalized (LSS) | Quality of life positively correlated with perceived stigma and a coping orientation of withdrawal, but NS with secrecy, educating challenging and distancing coping strategies; positively correlated with both stigma-related feelings subscales (misunderstood and different/ashamed) Medication adherence negatively correlated with perceived discrimination and a coping orientation of secrecy, but NS with withdrawal, educating, challenging, and distancing; negatively correlated with feeling different/ashamed but feeling misunderstood NS | None |
Xu, 2013 [105] | China | Facility-based sample; schizophrenia | 133 | Cross-sectional | Self-blame (CSQ-SB) | Symptom severity for depression was predicted by self-blame | None |