Menu
Menu
Menu
Menu
Objectives: Mental disorders are vastly underdiagnosed in low-income countries that disproportionately affect women. We aimed to evaluate the prevalence of common mental disorders in newly postpartum women, and stigma associated with mental health reporting in an Ethiopian community using a validated World Health Organization survey. Methods: The Self Reporting Questionnaire (SRQ) for psychological distress was administered in Amharic by nurses to 118 women aged 18–37 years who had given birth in the prior 3 months in the Glenn C. Olsen Memorial Primary Hospital in Yetebon. Mental health stigma among the four nursing staff was assessed using Link and Phelan’s Components of Stigma. Results: Among 118 women surveyed, 18% had a probable common mental disorder using the SRQ 4/5 cutoff and 2% admitted to suicidal thoughts. Presence of stigma in the healthcare staff was verified, including labeling, stereotyping, separating, and status loss and discrimination. Conclusion: Postpartum mental health disorders as well as stigma against such diagnoses are common in the Yetebon community. There is an urgent need for increased availability of properly trained and supervised healthcare staff in the identification and referral of postpartum women with common mental disorders.
The study was conducted at Project Mercy, a holistic community development organization founded in 1977, which implements integrated programs in K-12 education, healthcare, food security, orphan care, adult skills and literacy enhancement, and infrastructure development. About 11,000 individuals are served in the 52-bed Glenn C. Olsen Memorial Primary Hospital each year. Data were collected over a 9-month period, from October 2017 to June 2018. All women coming in for postnatal vaccinations and check-ups within 3 months of giving birth and providing verbal informed consent to participate in the survey were included. This was a time-limited, mixed-method, observational cohort survey. Among the healthcare staff who were being observed for stigma, few to no nurses in the Hospital had received any formal mental health pre-service education or in-service training. At the beginning of each survey, basic sociodemographic information was collected from the respondents including marital status, age, and number of children. This data was then housed in a separate document that was linked to the survey results through a number that was only available to SM and was kept in a password protected folder to ensure the survey results could not be linked back to the individual. Common mental disorders were assessed using the Self Reporting Questionnaire (SRQ) created by the WHO for use in primary health care settings in low- and middle-income countries (LMICs). The SRQ is a 20-question survey that asks yes or no questions and is scored based on the number of yes answers collected, with a scoring scale from 0 to 20 and a cut-off score of 4/5. The WHO uses the SRQ to assess the possible existence of non-specific psychological distress, including depression and anxiety disorders, as well as suicidality, in women in low-income countries, with further follow-up psychiatric assessment recommended to determine needed treatment (Beusenberg et al., 1994). The SRQ has been validated in neighboring Butajira, Ethiopia (Hanlon et al., 2008). Hanlon et al. provided the Amharic version of the survey, and a Project Mercy staff member made slight translation modifications, with attention to the most culturally appropriate way to word the questions to account for the Guragina language spoken in the population. The Amharic version of the survey was used in the interviews, however SM followed along on the back-translated English version to ensure fidelity in meaning of the questions. The four nurses who rotated through the Maternal and Children’s Health ward, including the head of the ward, were trained by investigator SM on the meaning of each of the questions in the survey and how to administer it, in consultation with a psychiatrist in Addis Ababa (see Acknowledgments). SM was present during all of the interviews conducted by the nurses to ensure they followed the proper protocol for administering the survey according to the WHO guidelines (Beusenberg et al., 1994). In order to minimize the impact of the power differential, SM sat together with the nurses to discuss the research about identification of mental illness and emphasized that the nurses were under no obligation to administer the survey. The head nurse, Sister Meseret, led survey implementation. During postpartum visits, mothers were seated while their child received their immunization, and only the nurse, mother, and researcher were present. Following the child’s immunization, nurses asked women if they were willing to answer 20 yes or no questions about their emotional wellness in the 30 days prior to their clinic visit. After obtaining verbal, informed consent, nurses administered the survey to the women as part of their postpartum checkup, including basic sociodemographic information at the beginning of the survey, to learn if they had experienced any psychological distress or postpartum depressive symptoms, and whether they were willing to discuss their emotions. Given that this was the first time that mental health was addressed at this hospital, we focused data collection on the SRQ and minimized other data collection. To combat social desirability bias, the same neutral language was used for every survey conducted, and the women’s data was deidentified using participant codes that were linked to the women’s names on a password-protected document on SM’s Protected Health Information-compliant computer. Additionally, the nurses were in a healthcare setting that protects patient privacy. SM followed along using the English version of the survey as well as the script that the nurses were using to introduce the survey to make sure they were all following it accurately. SM observed and interjected when the nurses strayed from their instructions and training on administration of the survey to ensure consistency and minimize bias. To ensure survey quality and adherence to the questionnaire, these observations were recorded and discussed with the nurses in a debriefing session following the administration of each survey. In debriefing sessions, SM invited observations from the nurses on the survey process from the nurses. A study in Butajira found that using a 4/5 cutoff score—meaning five or more affirmative answers—indicated presence of mental disturbance with 80% sensitivity and specificity (Hanlon et al., 2008). Using this study as reference, for all enrolled postpartum women we summed and reported positive (“Yes”) responses by question number and determined the prevalence of mental disturbance using a primary cutoff of 4/5, as those identified in the positive category using the 4/5 cutoff were noted in a previous study as in need of further psychiatric evaluation by a licensed professional (Rahman et al., 2013). The presence of stigma, defined as a sign of disgrace or discredit which sets a person apart from others, was also assessed in the health care staff giving the survey, based on observations of survey administration by SM (Byrne, 2000). The main model used for observational data on stigma was adapted from Link and Phelan’s Components of Stigma (2001) that describes converging components of: (1) labeling, (2) stereotyping, (3) separating, and (4) status loss and discrimination to identify stigma (Figure 1; Link and Phelan, 2001). During administration of the SRQ, we observed for these four signs from the health care staff. The nurses were not informed of the stigma assessment in order to reduce potential bias; however, as part of the informed consent process they were told that SM was conducting a research project on mental health. Observations on the presence of stigma on the part of the nurses were not discussed during survey administration debriefing sessions to enable ongoing assessment of its presence. Moreover, the rigorous nature of supervision to survey administration and regular discussions of observations of the researcher were designed to ensure that stigma did not significantly affect the survey results. Conceptual framework for identification of stigma. Adapted from Link and Phelan (2001). SM used a pre-determined framework adapted from Link and Phelan to record qualitative observations on the components of stigma, without use of prompts, using a printed framework guide for each postpartum woman. This involved four categories that were previously defined and behaviors that fell under one of those clearly defined categories were recorded (Figure 1). Labeling occurs when people distinguish and label human differences, so it was noted every time one of the health care staff denoted one of the women as different, or used a particular label, because of how she was acting or answering the survey. For example, labeling was observed when health care staff expressed the belief that their community and race (of females in particular) were mentally superior to those of the first world because they were “strong” and “never cry.” The second component of stereotyping occurs when “dominant cultural beliefs link labeled persons to undesirable characteristics—to negative stereotypes.” (Link and Phelan, 2001). This was recorded when the health care staff would assign a certain characteristic to women they had previously labeled as a result of their definition of Ethiopian culture. The third component of separating occurs when those who have been labeled are assigned to a specific, distinct category that achieves a separation of them from the person who assigns it to them, i.e., “us” vs. “them.” When health care staff made comments on distinct categorization of the women that had been labeled, this was also recorded. The fourth component of stigma relates to status loss and discrimination, a measure harder to achieve in the one-on-one interactions in the clinic; however, further research into the practices of the Ethiopian culture highlighted the link between the first three components and resulting status loss or discrimination (Reta et al., 2016). When each of these elements come together so that identifying those as different and constructing stereotypes to place them into categories leads to their exclusion from aspects of society as well as discrimination against them that adversely affects their access to social, economic, and political power, this is identified as stigmatization (Link and Phelan, 2001). SPSS was used to quantify SRQ scores for all 20 questions across all respondents and to determine the proportion of respondents meeting the 4/5 cutoff for screen positivity for the presence of common mental disorders. For qualitative analysis of stigma, Link, and Phelan’s framework was followed in order to interpret the observations collected. Ethical approval was received from the governing non-governmental organization of the hospital, Project Mercy, and the protocol was deemed exempt from human subjects review by the Stanford Institutional Review Board. All interviews were conducted during routine postpartum appointments after obtaining verbal informed consent; confidentiality and a comfortable environment were assured.
N/A
