Abstract
Purpose
There is scant research on depressive symptoms (DS), suicidal ideation (SI), and mental health care-seeking in Mozambique.
Methods
Generalized estimating equations were used to assess factors associated with DS, SI, and mental health care-seeking among 3080 individuals interviewed in a representative household survey in Sofala and Manica provinces, Mozambique.
Results
19% (CI 17–21%) of respondents reported DS in the past year and 17% (CI 15–18%) lifetime SI. Overall, only 10% (CI 8–11%) of respondents ever sought any care for mental illness, though 26% (CI 23–29%) of those reporting DS and/or SI sought care. 90% of those who sought care for DS received treatment; however, only 46% of those who sought care for SI received treatment. Factors associated with DS and SI include: female gender, divorced/separated, widowed, and > 55 years old. Respondents in the bottom wealth quintile reported lower DS, while those in upper wealth quintiles reported higher prevalence of SI. Individuals with DS or SI had significantly elevated measures of disability—especially in doing household chores, work/school activities, standing for long periods, and walking long distances. Factors associated with care-seeking include: female gender, rural residence, divorced/separated, and > 45 years old. Individuals in lower wealth quintiles and with no religious affiliation had lower odds of seeking care.
Conclusions
DS and SI are prevalent in central Mozambique and treatment gaps are high (68% and 89%, respectively). An urgent need exists for demand- and supply-side interventions to optimize the delivery of comprehensive community-based mental healthcare in Mozambique.
Keywords: Mozambique, Mental health, Depressive symptoms, Suicidal ideation, Care-seeking
Introduction
Factors associated with depression and suicide in low- and middle-income countries
Depression is ranked as the seventh most important cause of disease burden in low- and middle-income countries (LMICs) and 85% of suicides in the world occur in LMICs [1, 2]. Despite this high burden, epidemiologic data on depression and suicide are not readily available across many LMICs. A cross-national study conducted in 8 LMICs reported that the average lifetime and 12-month prevalence estimates of depression in LMICs were 11.9% and 5.9%, respectively [3]. The same study found that the average age of onset for depression in LMICs was 24 years, the female–male ratio was 2:1, and the strongest demographic correlate was being divorced or widowed. These findings are consistent with other studies, reporting that being female and unmarried are consistently correlated with major depressive episodes [4, 5]. Available evidence from LMICs suggests that there is either a nonmonotonic relationship between depression and age, or that the prevalence of depression increases with age [6–10].
With regard to suicide, a cross-national study conducted in seven LMICs found that the lifetime prevalence of suicidal ideation ranged from 3.1 to 12.4% [11]. A systematic review of suicide in African countries reported that risk factors for suicide include: interpersonal difficulties, mental and physical health problems, low socioeconomic standing, and drug and alcohol use/abuse [12]. Other commonly cited risk factors in LMICs include youth or old age, low levels of education, being unmarried, previous suicide attempts, family history of psychopathology, and stressful life events [11, 13–15]. It has also been reported in a number of studies in both LMICs and high-income countries (HICs) that females have higher rates of suicidal ideation and behavior than males, but males have higher mortality rates from suicide than females [16, 17].
In the Mozambican population, there is limited research on suicide and common mental disorders, such as depression. To our knowledge, there is currently only one peer-reviewed publication describing factors associated with suicidal attempts and deaths in Mozambique [17]. This paper focused at the clinic level and was limited in geographic scope. Two community-level mental health studies have been conducted previously in Mozambique [18, 19]. The former was focused solely on severe mental disorders, while the latter was specifically focused on depression among women.
The present paper seeks to describe the burden, risk factors, and care-seeking behavior for depressive symptoms and suicidal ideation in central Mozambique. This study aims to assess the burden of and care needs for common mental disorders and suicidal thoughts in Mozambique and other similar LMICs with high disease burden and limited resources.
Depression and suicide in Mozambique
In Mozambique, mental and substance use disorders are estimated to account for 21.7% of all years lived with disability for those aged 15–49 [20]. Mental, neurological, and substance use disorders are estimated to account for more years lived with disability in the country than HIV, TB, and malaria combined [21]. Additionally, modeling studies estimate that the suicide rate among Mozambican males (20.5/100,000) is nearly three times the Africa regional average (7.4/100,000) [22].
Mozambique has made great strides in reducing the treatment gap for mental health services. In 1994, the Ministry of Health implemented a task-shifting strategy through the development of a 30-month psychiatric technician training program, in which mid-level professionals are trained to provide mental health services in primary care settings [23]. This strategy has increased availability of mental health services, and this new cadre of health professionals currently provide the vast majority of psychiatric services nationwide [23, 24]. Despite this strategy, there is a significant shortage of mental health professionals and limited health system capacity to bring evidence-based mental health interventions to scale. For instance, essential psychotropic medications are routinely unavailable at primary care facilities, and psychiatric technicians are predominately located at district-level hospitals, leaving a majority of the population whose care is provided by primary care facilities without ready access to mental health services [24, 25]. This is particularly evident in central Mozambique, where it is estimated that 99% of formal healthcare is provided in the public sector, and primarily at the primary care level [26].
In addition to limited human and financial resources, research suggests that mood disorders are currently not well addressed by the Mozambique healthcare system. Less than 4% of yearly consultations in the primary care system are for mood disorders, which is far lower than would be expected given their estimated disease burden, as well as their link to suicide attempts and deaths [24]. Deeper community-level understanding of the prevalence and associated factors for depressive symptoms and suicidal ideation is needed to develop implementation strategies to address the treatment gap for community-based public-sector mental health services.
Methods
Study setting
Mozambique is a southeastern African country with among the lowest health and development indices globally. The country has a population of 28.9 million with over 50% of the population under the age of 18 [27–29]. Mozambique is classified as a low-income country, and a majority of the population (67%) resides in rural areas [28]. Mozambique has a high prevalence of HIV, with 13.2% of the adult population infected [30]. The child mortality rate has declined significantly in the past decade; however, children under five continue to die from preventable causes at a rate of 79/1000 live births [31].
This survey was carried out in Manica and Sofala provinces in central Mozambique. Sofala’s population is 2.2 million, while Manica’s is 1.9 million [29]. Both have relatively high population densities (Sofala 24.8/km2, Manica 23.3/km2) [32]. Health and socioeconomic status indicators are comparable across both provinces, with approximately 60% of the population at or below the middle wealth quintile [30].
Study design and sampling procedure
Data for this study are from a cross-sectional infant mortality impact evaluation, powered to detect a 15% decrease in infant mortality in Sofala. Additional survey questions were added to assess the prevalence and factors associated with mental illness and noncommunicable diseases. The survey was conducted between September, 2016 and February, 2017. The full sampling methods for this community survey have been previously published [33]. Briefly, remote satellite imagery was used to develop a provincial-level representative community survey sampling frame. The satellite imagery was integrated with the open-source OpenStreetMap platform (http://www.openstreetmap.org) to digitize all buildings, which were then used to represent population density. The study area was divided into sampling units of approximately 2 × 2 km, and their population densities used to generate probability proportional to size sampling. Twenty households were invited to participate for each time a sampling unit was selected. Due to civil conflict, 8/31 administrative units in Sofala and 15/39 in Manica were excluded from our sample. The final sample included 1549 households in Sofala and 1538 households in Manica.
Data collection
Data were collected on Samsung tablets using Open Data Kit (ODK) software, and transferred from ODK to a REDCap database through a cloud server [34]. There were a total of 23 data collectors (10 male and 13 female). Each participant interview lasted approximately 30 min and was conducted inside the participant’s home or outside close to it, according to their preference. Due to literacy barriers, the data collector asked each survey question verbally and recorded the participant’s response.
The household survey questions were adapted from the 2011 Mozambique Demographic and Health Survey with an additional module to address the burden of mental health conditions and general disability. If a participant reported current suicidal ideation, he or she was referred to the nearest health facility.
Outcome and explanatory variables
We analyzed the prevalence and associated factors for three outcomes: depressive symptoms (DS), suicidal ideation (SI), and mental health care-seeking. Outcomes were defined as follows: (1) DS: Have you ever had a period of sadness and/or loss of energy lasting more than 2 weeks in the past 12 months?; (2) SI: Have you had thoughts of suicide or self-harm (in your lifetime, in the past month, and/or currently)?; and (3) care-seeking behavior: Have you ever sought care (allopathic or non-allopathic) for a mental health problem, depressive symptoms, and/or suicidal ideation?
DS were assessed as a single survey question and cannot be interpreted as clinical depression. While this question lacks specificity, we anticipate it is sensitive, allowing for a general understanding of DS in this study population. Our care-seeking variable included respondents who reported seeking any care for mental ill-health, DS, and/or SI. We included all three subcategories in our outcome given the common misconception that mental illness only refers to severe mental disorders.
Explanatory variables were selected from existing literature on factors associated with DS and SI, focusing on socio-demographic factors, but including additional survey variables hypothesized to increase risk for DS or SI. Factors assessed included age, gender, urban vs. rural household location, marital status, socioeconomic status (SES), education level, religious affiliation, alcohol consumption, history of injury in the past year, history of injury by assault in the past year, and overall disability (Table 1).
Table 1.
Demographic characteristics of 3080 individuals interviewed in the present study, 2016–2017
| Characteristic | Sofala N (%) | Manica N (%) | N (%) |
|---|---|---|---|
| Sex | |||
| Male | 757 (54%) | 638 (46%) | 1395 (46%) |
| Female | 767 (47%) | 879 (53%) | 1646 (54%) |
| Urban or rural | |||
| Urban | 480 (53%) | 434 (47%) | 914 (30%) |
| Rural | 1065 (49%) | 1088 (51%) | 2153 (70%) |
| Level of school | |||
| No School | 342 (60%) | 232 (40%) | 574 (19%) |
| Basic (literacy and primary) | 803 (50%) | 817 (50%) | 1620 (54%) |
| Higher (secondary and higher) | 340 (43%) | 450 (57%) | 790 (26%) |
| Age | |||
| 18–25 | 332 (47%) | 373 (53%) | 705 (23%) |
| 26–35 | 457 (47%) | 512 (53%) | 969 (31%) |
| 36–45 | 328 (52%) | 302 (48%) | 630 (20%) |
| 46–55 | 199 (59%) | 137 (41%) | 336 (11%) |
| > 55 | 226 (57%) | 168 (43%) | 394 (13%) |
| Marital status | |||
| Married | 1272 (50%) | 1284 (50%) | 2556 (83%) |
| Divorced/separated | 82 (48%) | 88 (52%) | 170 (6%) |
| Widowed | 130 (54%) | 110 (46%) | 240 (8%) |
| Single | 36 (50%) | 36 (50%) | 72 (2%) |
| Religion | |||
| Pentecostal | 678 (49) | 707 (51% | 1385 (46%) |
| Catholic | 231 (59%) | 158 (41%) | 389 (13%) |
| Muslim | 29 (74%) | 10 (26%) | 39 (1%) |
| Zion | 138 (35%) | 255 (65%) | 393 (13%) |
| Anglican | 22 (52%) | 20 (48%) | 42 (1%) |
| Johan Masowe/Johan Maranga | 12 (38%) | 20 (63%) | 32 (1%) |
| No religion | 287 (57%) | 217 (43%) | 504 (17%) |
| Other Christian | 71 (57%) | 53 (43%) | 124 (4%) |
| Not sure | 10 (83%) | 2 (17%) | 12 (0.4%) |
| Other | 51 (58%) | 37 (42%) | 88 (3%) |
| SES | |||
| 1st quintile (poorest) | 383 (62%) | 232 (38%) | 615 (20%) |
| 2nd quintile | 301 (49%) | 312 (51%) | 613 (20%) |
| 3rd quintile | 264 (43%) | 348 (57%) | 612 (20%) |
| 4th quintile | 256 (42%) | 358 (58%) | 614 (20%) |
| 5th quintile | 341 (56%) | 271 (44%) | 612 (20%) |
| Alcohol consumption | |||
| Never | 591 (36%) | 1065 (64%) | 1656 (54%) |
| Once a month or less | 140 (46%) | 166 (54%) | 306 (10%) |
| 2–4 × per month | 47 (33%) | 97 (67%) | 144 (5%) |
| 2–4 × per week | 33 (35%) | 60 (65%) | 93 (3%) |
| 4× or more per week | 14 (44%) | 18 (56%) | 32 (1%) |
| Missing | 729 (86%) | 120 (14%) | 849 (28%) |
| Overall disability | |||
| 0–12 | 498 (50%) | 485 (50%) | 993 (32%) |
| 13–24 | 991 (52%) | 913 (48%) | 1904 (62%) |
| 25–36 | 36 (31%) | 81 (69%) | 117 (4%) |
| 37–60 | 29 (44%) | 37 (56%) | 66 (2%) |
| Any injury (in the past year) | |||
| No injury | 1378 (50%) | 1376 (50%) | 2754 (90%) |
| Injury | 170 (54%) | 146 (46%) | 316 (10%) |
| Injury by assault (in the past year) | |||
| No assault injury | 1535 (50%) | 1506 (50%) | 3041 (99%) |
| Assault injury | 19 (49%) | 20 (51%) | 39 (1%) |
Missing data were < 5% except where indicated
The SES variable was generated using principal component analyses (PCA) of household characteristics and ownership of household items [35, 36], and then categorized into wealth quintiles. Overall disability was measured using the short version of the WHO Disability Assessment Schedule 2.0 (WHODAS), which is a 12-item questionnaire to assess health and disability [37]. The 12 questions relate to the functioning difficulties experienced by the respondent during the previous 30 days and scores can range from 0 to 60. History of injury in the past year included all injury types, including injury by assault.
Statistical analyses
To calculate sampling weights, we enumerated all buildings in each sampling unit. Out of 176 sampling units, the study team excluded 23 due to sudden conflict and/or because they were too difficult to reach. To adjust for non-response in these 23 sampling units, their sampling weights were redistributed to neighboring areas by province, consistent with the province-stratified sampling approach. Sampling weights were applied to survey responses to generate prevalence estimates (Table 2) and analytical models (Tables 3 and 4). The full sampling methods have been previously published [33].
Table 2.
Weighted prevalence of depressive symptoms and suicidal ideation among 3080 individuals in the present study, 2016–2017
| Description | Percentage (95% CI) |
|---|---|
| Proportion of respondents who reported period of sadness or loss of energy that lasted more than 2 weeks (yes/no) out of total number of respondents | 19.1% (17.1–21.0) |
| Proportion of respondents who reported thoughts of suicide or self-harm in their lifetime (yes/no) out of total number of respondents | 17.0% (14.8–18.2) |
| Proportion of respondents who reported thoughts of suicide or self-harm in the last month (yes/no) out of total number of respondents | 5.9% (4.9–6.8) |
| Proportion of respondents who reported thoughts of suicide or self-harm currently (yes/no) out of total number of respondents | 1.6% (1.1–2.2) |
| Proportion of respondents who reported thoughts of suicide or self-harm in the last month (yes/no) out of total number of respondents with lifetime suicidal ideation | 35.7% (31.4–40.0) |
| Proportion of respondents who reported thoughts of suicide or self-harm currently (yes/no) out of total number of respondents with lifetime suicidal ideation | 10.0% (7.0–12.9) |
95% CI 95% confidence interval
Table 3.
WHO-DAS disability measures: ordered logistic regression analyses of individuals in the present study, 2016–2017
| Depressive symptoms | Suicidal ideation | Mental health care-seeking | ||||
|---|---|---|---|---|---|---|
| AOR | P value | AOR | P value | AOR | P value | |
| Difficulty standing for long periods of time | 1.39 | 0.002 | 1.57 | 0.000 | 1.39 | 0.020 |
| Difficulty doing household chores | 1.92 | 0.000 | 1.57 | 0.005 | 1.93 | 0.000 |
| Difficulty learning new task | 1.45 | 0.001 | 1.19 | 0.171 | 1.26 | 0.081 |
| Difficulty participating in community activities | 1.04 | 0.794 | 1.32 | 0.032 | 1.12 | 0.514 |
| Health affects emotional state | 0.88 | 0.183 | 1.20 | 0.116 | 1.30 | 0.064 |
| Difficulty concentrating 10 min | 1.14 | 0.348 | 1.68 | 0.000 | 1.69 | 0.002 |
| Difficulty walking long distance | 1.53 | 0.002 | 1.48 | 0.011 | 1.86 | 0.000 |
| Difficulty taking a bath | 1.82 | 0.178 | 1.36 | 0.470 | 2.55 | 0.072 |
| Difficulty getting dressed | 2.01 | 0.123 | 1.95 | 0.136 | 2.16 | 0.159 |
| Difficulty dealing with strangers | 0.93 | 0.559 | 1.14 | 0.285 | 1.01 | 0.945 |
| Difficulty maintaining friendships | 0.78 | 0.060 | 1.15 | 0.357 | 1.11 | 0.534 |
| Difficulty at work or in school | 1.37 | 0.046 | 1.63 | 0.006 | 1.83 | 0.001 |
Table 4.
Multivariable analyses of depressive symptoms, suicidal ideation, and mental health care-seeking in the present study, 2016–2017
| Depressive symptoms | Suicidal ideation | Mental health care-seeking | ||||
|---|---|---|---|---|---|---|
| AOR (95% CI) | P value | AOR (95% CI) | P value | AOR (95% CI) | P value | |
| Sexa | ||||||
| Male | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Female | 1.40 (1.13–1.73) | 0.002 | 1.46 (1.16–1.84) | 0.001 | 1.37 (1.03–1.82) | 0.030 |
| Residenceb | ||||||
| Urban | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Rural | 1.47 (1.08–1.99) | 0.013 | 0.86 (0.62–1.18) | 0.348 | 2.32 (1.52–3.55) | 0.000 |
| Level of Schoolc | ||||||
| No school | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Basic | 0.96 (0.75–1.24) | 0.777 | 0.80 (0.62–1.03) | 0.081 | 1.18 (0.84–1.66) | 0.336 |
| Higher | 0.96 (0.70–1.32) | 0.796 | 0.64 (0.46–0.90) | 0.010 | 0.88 (0.57–1.36) | 0.565 |
| Aged | ||||||
| 26–35 | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| 18–25 | 0.86 (0.67–1.12) | 0.265 | 1.10 (0.81–1.43) | 0.633 | 1.11 (0.79–1.57) | 0.541 |
| 36–45 | 1.09 (0.84–1.42) | 0.502 | 1.60 (1.20–2.09) | 0.001 | 1.33 (0.94–1.88) | 0.109 |
| 46–55 | 1.14 (0.83–1.56) | 0.429d | 1.76 (1.27–2.44) | 0.001 | 1.73 (1.16–2.58) | 0.007 |
| > 55 | 1.83 (1.38–2.42) | 0.000 | 1.62 (1.18–2.22) | 0.003 | 1.57 (1.06–2.33) | 0.024 |
| Marital statusf | ||||||
| Married | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Divorced/separated | 1.80 (1.24–2.62) | 0.002 | 1.77 (1.20–2.60) | 0.004 | 1.75 (1.10–2.78) | 0.018 |
| Widowed | 2.00 (1.40–2.86) | 0.000 | 1.73 (1.20–2.48) | 0.003 | 1.00 (0.61–1.64) | 0.991 |
| Single | 1.23 (0.67–2.26) | 0.511 | 1.00 (0.49–2.08) | 0.990 | 0.98 (0.41–2.31) | 0.955 |
| Religionf | ||||||
| Pentecostal | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Catholic | 1.06 (0.79–1.42) | 0.715 | 0.78 (0.56–1.10) | 0.163 | 0.96 (0.64–1.42) | 0.826 |
| Muslim | 1.11 (0.49–2.48) | 0.806 | 0.73 (0.25–2.12) | 0.565 | 1.06 (0.36–3.12) | 0.923 |
| Zion | 1.09 (0.81–1.47) | 0.553 | 1.09 (0.80–1.49) | 0.568 | 1.27 (0.88–1.85) | 0.198 |
| Anglican | 1.30 (0.61–2.77) | 0.503 | 0.61 (0.21–1.76) | 0.363 | 0.75 (0.23–2.46) | 0.638 |
| Johan Masowe/Johan Maranga | 0.42 (0.13–1.40) | 0.159 | 1.10 (0.45–2.72) | 0.833 | 0.71 (0.19–2.68) | 0.617 |
| No religion | 0.72 (0.53–0.97) | 0.033 | 0.91 (0.67–1.22) | 0.528 | 0.64 (0.42–0.97) | 0.036 |
| Other Christian | 0.62 (0.35–1.09) | 0.098 | 1.09 (0.65–1.83) | 0.739 | 1.01 (0.53–1.92) | 0.971 |
| Not sure | 0.49 (0.06–3.82) | 0.499 | 0.67 (0.09–4.76) | 0.685 | 1.18 (0.16–8.87) | 0.870 |
| Other | 1.56 (0.94–2.58) | 0.086 | 1.08 (0.61–1.92) | 0.792 | 0.96 (0.46–2.00) | 0.917 |
| SESe | ||||||
| 5th quintile | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| 1st quintile (poorest) | 0.65 (0.43–0.98) | 0.041 | 1.44 (0.91–2.28) | 0.118 | 0.51 (0.29–0.88) | 0.015 |
| 2nd quintile | 0.70 (0.47–1.04) | 0.079 | 1.31 (0.83–2.05) | 0.242e | 0.47 (0.27–0.81) | 0.007 |
| 3rd quintile | 0.79 (0.54–1.15) | 0.216 | 1.39 (0.90–2.13) | 0.136 | 0.64 (0.39–1.07) | 0.087 |
| 4th quintile | 0.82 (0.60–1.14) | 0.236 | 1.45 (1.01–2.07) | 0.045 | 0.99 (0.64–1.52) | 0.951 |
| Alcoholf | ||||||
| Never | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Once a month or less | 1.31 (0.97–1.77) | 0.081 | 1.46 (1.06–2.01) | 0.019 | 1.06 (0.70–1.60) | 0.783 |
| 2–4× per month | 1.19 (0.77–1.84) | 0.432 | 1.42 (0.90–2.23) | 0.133 | 1.48 (0.88–2.51) | 0.143 |
| 2–4× per week | 1.58 (0.96–2.59) | 0.073 | 1.63 (0.96–2.77) | 0.073 | 0.95 (0.46–1.97) | 0.894 |
| 4 or more per week | 0.80 (0.31–2.11) | 0.658 | 2.56 (1.15–5.71) | 0.022 | 1.99 (0.76–5.21) | 0.162 |
| Overall disabilityf | ||||||
| ≤ 12 | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| 13–24 | 1.00 (0.81–1.23) | 0.994 | 1.16 (0.92–1.46) | 0.201 | 1.37 (1.03–1.82) | 0.033 |
| 25–36 | 1.02 (0.63–1.67) | 0.922 | 2.04 (1.29–3.23) | 0.002 | 1.50 (0.81–2.75) | 0.196 |
| > = 37 | 2.89 (1.64–5.09) | 0.000 | 1.64 (0.85–3.18) | 0.143 | 2.01 (0.93–4.32) | 0.074 |
| Any injuryf | ||||||
| No injury | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Injury | 1.62 (1.21–2.13) | 0.001 | 1.62 (1.20–2.18) | 0.002 | 1.55 (1.09–2.20) | 0.016 |
| Assault injuryf | ||||||
| No injury from assault | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Injury from assault | 1.49 (0.71–3.12) | 0.289 | 2.08 (1.01–4.32) | 0.048 | 1.75 (0.73–4.18) | 0.208 |
Adjusted for age, SES, education, residence
Adjusted for age, SES, education, sex
Adjusted for SES, sex, residence
Adjusted for SES, residence
Adjusted for age, sex, education, residence
Adjusted for age, sex, education, SES, residence
Factors associated with DS, SI, and mental health care-seeking were identified using generalized estimating equations with clustering by primary sampling unit, using the binomial family, a logit link function, and an exchangeable working correlation matrix. Selected explanatory factors were additionally analyzed stratified by gender to determine whether associations differed strongly. We conducted both univariable and multivariable complete-case analyses for each exposure–outcome relationship. Adjustment variables in the multivariable models were selected a priori based on existing literature, and were selected individually for each exposure of interest. We also analyzed the associations between the 12 WHODAS disability measures and DS, SI, and mental health care-seeking. These associations were identified using ordered logistic regressions with robust standard errors clustered by primary sampling unit.
Results
Prevalence estimates
The estimated weighted prevalence of DS in our study area is 19.1% (Table 2). Of those who reported DS (n = 591), only 36.3% sought care, of whom 89.9% reported receiving treatment (Fig. 1a).
Fig. 1.
Prevalence of depressive symptoms, suicidal ideation, and care-seeking behavior
The estimated weighted prevalence of lifetime SI in our study area is 16.5%. The estimated weighted prevalence of SI in the last month is 5.9% and current SI is 1.6%. Of those who reported SI in the last month (n = 181), only 23.1% sought care, of whom only 45.8% reported receiving treatment (Fig. 1b).
A small proportion (9.7%) of all respondents had ever sought any care for a mental health problem. Of these, the vast majority sought care for DS (94.8%). Among those who reported DS and/or SI (n = 896), roughly one quarter (26.1%) sought any care for mental ill-health and 22.9% received any treatment.
Factors associated with depressive symptoms
Being female [adjusted odds ratio (aOR): 1.40, 95% Confidence Interval (CI) 1.13–1.73] and living in a rural area (aOR: 1.47, CI 1.08–1.99) was significantly associated with self-reported DS. There was a monotonic positive relationship between age and DS; however, the association was significant only for those aged > 55 years (aOR: 1.83, CI 1.38–2.42) compared to those aged 26–35 years. Of all socio-demographic factors, marital status had the strongest association with DS, with unmarried single, divorced or separated, and widowed individuals having greater odds of DS in comparison to married individuals (aOR: 1.23, CI 0.67–2.26; aOR: 1.80, CI 1.24–2.62; aOR: 2.00, CI 1.40–2.86 respectively). In stratified analyses, associations between marital status and DS were stronger among women than men (Appendix Table 7). In contrast to our hypothesis, the lowest wealth quintile had lower self-reported DS (aOR: 0.65, CI 0.43–0.98) in comparison to the highest wealth quintile, and individuals with higher levels of education did not differ significantly from individuals with no schooling (Table 3).
Factors associated with suicidal ideation
Factors associated with DS were similar to factors associated with SI. Females were more likely to report SI (aOR: 1.46, CI 1.16–1.84) in comparison to males. Divorced or separated individuals (aOR: 1.77, CI 1.20–2.60) and widowed individuals (aOR: 1.73, CI 1.20–2.48) had greater odds of SI in comparison to married persons. There was a monotonic positive association between age and SI, with individuals 36–45 (aOR: 1.60, CI 1.20–2.09), 46–55 (aOR: 1.76, CI 1.27–2.44), and > 55 years (aOR: 1.62, CI 1.18–2.22) being significantly more likely to report lifetime SI than individuals aged 26–35 years. A similar monotonic positive association was observed between age and SI in the last month as well.
Compared to highest wealth quintile SES, all other quintiles were more likely to report lifetime SI; however, the only statistically significant difference was among the 4th quintile (aOR: 1.45, CI 1.01–2.07). Individuals with higher levels of education were less likely to report SI than those with no schooling (aOR: 0.64, CI 0.46–0.90).
Factors associated with mental health care-seeking
Mental health care-seeking was estimated for multiple sub-groups, including: (1) all respondents; (2) respondents who reported DS and/or SI; (3) respondents who reported DS only; and (4) respondents who reported SI only. Generally speaking, the associations were similar across all subgroups (Appendix Table 6).
Among all respondents, females were significantly more likely to seek care for mental health than males (aOR: 1.42, CI 1.06–1.89). Older individuals were more likely to seek care compared to those aged 26–35 years (46–55 years aOR: 1.78, CI 1.18–2.68; > 55 years aOR: 1.51, CI 1.00–2.27). Compared to married individuals, those who were separated or divorced were significantly more likely to have sought mental health care (aOR: 1.80, CI 1.12–2.88). Individuals with lower levels of SES were significantly less likely to seek care for mental health compared to high SES individuals (first quintile SES aOR: 0.51, CI 0.29–0.89; second quintile SES aOR: 0.53, CI 0.30–0.92). Rural residents were more likely to seek care than urban residents (aOR: 2.28, CI 1.49–3.51).
Discussion
Prevalence of depressive symptoms, suicidal ideation, and care-seeking
One out of every five (19%) respondents self-reported DS in the past year and one out of every six (17%) reported lifetime SI. These prevalence estimates are comparable to a recently published study in central Mozambique which found that 14% of female heads-of-household screened positive for depression [19]. However, generally speaking, these results are higher than previous research in LMICs. For example, a study conducted in 18 countries found that the average 12-month prevalence of individuals who screened positive for depression in LMICs was 10.5% [3]. Similarly, a nationally representative household survey in South Africa found that 4.9% of respondents experienced a depressive episode in the past 12 months [38]. With regard to suicide, a cross-national study conducted in 17 countries found that the average lifetime prevalence of SI was 9.2% [11]. And a community-level study conducted in Nigeria found that the weighted prevalence of SI was 7.3% [39].
The higher prevalence estimates we obtained may partly be explained by our definition of DS, which is not based on a validated screening tool and may be an overestimate of the true prevalence of depression. However, our SI questions are unlikely to be misinterpreted and believed to produce valid SI prevalence results. An additional challenge to these comparisons is the scant evidence regarding prevalence of SI globally.
It is reasonable to infer that civil unrest is one key under-lying cause of these high prevalence estimates. There is a significant amount of research suggesting that war-affected and post-disaster regions have a higher prevalence of depressive symptomatology and suicidal ideation [40–45]. After gaining independence from Portugal in 1975, Mozambique endured a 16-year civil war that displaced millions of refugees and destroyed key infrastructure, disproportionately affecting central Mozambique. Furthermore, in the 1970s, central Mozambique served as a base for neighboring Zimbabwe’s guerrilla campaign against the minority government [46]. Recently, there has been a flare-up of tensions with the RENAMO opposition group resulting in ongoing conflict, violent attacks, and nighttime raids, especially in Sofala and Manica provinces.
Despite the high burden of self-reported mental health issues in our survey, only 9.7% of respondents had ever sought care for mental health, including 36% of respondents who reported DS and 23% of respondents who reported SI in the last month. Most of those who sought care for DS (89.9%) reported receiving treatment; however, only 45.8% of respondents who sought care for SI reported receiving treatment. As such, the treatment gap for SI is much higher than the treatment gap for DS (89.3% and 67.9%, respectively). Similar to a study conducted in Sofala, these findings suggest that a large proportion of individuals with common mood disorders do not receive treatment [24].
We found that the vast majority (94.8%) of mental health care-seeking was for DS; yet, previous research indicates that mood disorders represent a very low proportion (less than < 4%) of yearly psychiatric consultations at facilities across Sofala province [24]. The differences in these findings may be due to alternative care pathways for treatment of depression outside of formal psychiatric services. It would not be surprising for individuals to seek community resources over clinical care, especially in settings where the formal mental health system is nascent and focuses primarily on severe mental illness. Previous work in Haiti, Nigeria, and Ethiopia has found that it is common for individuals to seek mental health care from traditional or religious leaders prior to, or instead of, allopathic clinical care [47–49]. These findings highlight the need to expand mental health services beyond district-level facilities and to engage with informal health providers and community resources [47, 50].
Our findings that 89.3% of individuals with SI and 67.9% of individuals with DS did not receive treatment highlights the urgent need for multifaceted and multi-level implementation strategies to close the mental health treatment gap in low-resource settings. Such approaches include improving community-level screening for common mental disorders [51, 52], integrating mental healthcare within the primary care system [53–55], reducing community-level stigma [56, 57], creating links between community and health facility allopathic care pathways [50, 58], increasing use of mental health services [59], and optimizing the performance of the existing mental health system.
Factors associated with depressive symptoms and suicidal ideation
Socio-demographic factors (female gender, being unmarried, and older age) associated with DS and/or SI are generally consistent with previous literature [3–7]. Epidemiologic research indicates that being female is associated with a twofold increased risk of a lifetime diagnosis of major depressive disorder, as well as an increased risk of SI [3, 5, 17, 60]. We found similar patterns in our study, with women having 40% higher odds of DS and 50% higher odds of SI.
Our study is also in line with previous research showing that low levels of social support, due to being single, divorced, or widowed, can result in social isolation and increased potential for development of SI or depression [4, 15]. Widowhood and divorce are well recognized as stressful life events that precipitate DS or SI. Our associations with widowhood and divorce were predominately driven by women; in stratified analyses, associations between marital status and DS/SI were much stronger among women than men (Appendix Table 7). These findings suggest that the social consequences of divorce and widowhood may be greater for females than males in central Mozambique. Last, our findings are consistent with the limited research in LMICs that suggests an increasing prevalence of depression with age [6, 7, 9, 10], and that old age is a risk factor for suicide deaths [13].
Both self-reported DS and SI were strongly associated with a history of injury and injury by assault in the past year. In our study population, 10% of respondents experienced any injury in the past year, of which 14% were injuries by assault. There is limited research in LMICs on the relationship between injury and DS or SI. However, substantial evidence in HICs demonstrates a strong relationship between traumatic physical injury and subsequent DS and/or SI [61, 62]. Recovery from physical trauma is emotionally challenging, and can have many impacts on health and well-being, including employment and ability to carry out general physical activities [63]. The process of recovering from physical trauma may be particularly difficult in low-resource settings where a loss of employment or persistent disability could result in catastrophic difficulties for maintaining household income.
Coinciding with the downstream impact of injuries on mental health, individuals with DS and/or lifetime SI had significantly elevated measures of disability—especially expressing difficulty standing for long periods of time, doing household chores, walking long distances, and completing work or school activities. These results concur with the published literature that adults with disability have a significantly higher incidence of DS than the general population [64, 65]. Our observed strong association between DS and disability also suggests face validity of our question to represent clinically important depressive symptomatology, and suggests there is a large number of individuals with depression-related function impairment who currently are not receiving treatment and likely would benefit from clinical treatment interventions.
Our findings that lower SES was associated with lower levels of self-reported DS and that higher quintiles of SES were significantly associated with SI bears additional discussion. Given that low SES populations experience more adverse living circumstances than their counterparts, this subpopulation tends to have higher rates of DS and SI. Our observation is contrary to most of the published literature, which predominantly finds an inverse relationship between increasing levels of SES and common mental disorders [66, 67]. Given the high rates of poverty in central Mozambique, it is likely that the majority of our study sample is poor. Therefore, individuals in the higher wealth quintile may represent the ‘working poor’ who feasibly experience heightened distress, as they are busy maintaining regular employment, but remain in relative poverty, potentially unable to save money or support their families.
In comparison to abstainers, alcohol consumption was associated with elevated levels of DS and SI. These associations were especially strong for SI. We found that 46% increased odds of SI associated with minimal alcohol consumption compared to no alcohol consumption, and 156% increased odds of SI for significant alcohol consumption compared to no alcohol consumption. These findings concur with previous research, which indicates that excessive alcohol consumption and DS commonly co-occur and that alcohol misuse predisposes to suicidal behavior [68–70], though we are not able to determine from our data whether alcohol consumption predated mental health symptoms. When stratified by gender, females generally had stronger associations between alcohol consumption and DS or SI than men (Appendix Table 7).
Factors associated with mental health care-seeking
In the present study, we found that being female was significantly associated with mental health care-seeking. This appears consistent with published literature, where women are generally more likely to seek healthcare than men [71–73]. We also found that older age groups (46–55 years and > 55 years) had increased odds of care-seeking. These findings also coincide with the literature, which has found that middle-aged persons more likely to seek help for DS than other age groups [71]. However, this research is primarily from HICs and may not be representative of LMICs.
Contrary to initial hypotheses, rural residents were significantly more likely to seek care for mental ill-health than urban residents. Given the shortage of mental health professionals in Mozambique and that psychiatric technicians are predominately located at district-level health facilities, it is counterintuitive that rural residents were more likely to seek mental health care. One explanation for this finding may be that our care-seeking question did not specify if care-seeking was allopathic or non-allopathic, or from the formal or informal care delivery sectors. It is possible that many individuals reporting mental health care-seeking utilized non-allopathic community providers, given previous research suggesting nearly half of individuals seeking care for mental health in Africa choose traditional and religious healers over allopathic clinical care [74].
Lastly, respondents with lower SES (1st and 2nd quintiles) were significantly less likely to seek care for mental ill-health. These findings are consistent with the literature, where SES is one of the most significant determinants of health-seeking behavior [75, 76]. Barriers to care-seeking for low SES populations, including medical expenses, transport to health facilities, and loss of income, inhibit low SES populations from accessing health care.
Study limitations
This study has a number of important limitations. First, as a cross-sectional community household survey, only associations can be inferred between explanatory and outcome variables; we cannot infer the temporal direction. Second, as mentioned previously, the DS outcome variable was a single non-validated survey question that cannot be interpreted as a clinical diagnosis of depression. We recognize this question lacks specificity, but believe it is sensitive and able to capture a general understanding of DS in the study population. Third, several administrative units of Sofala and Manica provinces were excluded from data collection due to violent civil conflict ongoing at the time of study. Unfortunately, the exclusion of these administrative units negatively impacts the representativeness of our sampling frame. It is important to highlight notable strengths of this study, namely, this study is the first community-level assessment of SI and mental health care-seeking in Mozambique, and the second community-level assessment of DS. It includes a relatively large sample and relied on an up-to-date sampling frame that used satellite imagery to enumerate households.
Conclusions
There is a high prevalence of DS and lifetime SI in central Mozambique; yet, the majority of individuals suffering from these common mental health conditions do not receive treatment (68% and 89%, respectively). Urgent investments are needed to develop approaches to scale up access to care and treatment for common mental disorders, such as community-level screening, decreased stigma, explicit linkages between allopathic and non-allopathic community providers, programs to increase demand for mental healthcare, the integration of mental health services within the primary care system, and systems analysis and improvement approaches to optimize the delivery of existing of community-based public-sector mental healthcare. There were substantial similarities between our findings and research in LMICs regarding factors associated with DS and SI, with particularly high odds among women, unmarried, older, rural residents, and those with low education. Policymakers and health systems managers should consider targeting limited resources towards these populations.
Acknowledgements
This work was supported by the African Health Initiative of the Doris Duke Charitable Foundation. INCOMAS Study Team includes: Falume Chale; Alfredo Covele; Fatima Cuembelo; Stephen Gloyd; Catherine Henley; Leecreesha Hicks; Joaquim Lequechane; Arlete Mahumane; Nelia Manaca; Cathy Michel; Alberto Muanido; Miguel Nhumba; James Pfeiffer; and Lucia Vieira.
Funding This study was supported by the Doris Duke Charitable Foundation’s African Health Initiative. Bradley H. Wagenaar was supported by Grant number K01MH110599 from the US National Institute of Mental Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Appendix
See Tables 5, 6, 7.
Table 5.
Prevalence of Mental Health Care-seeking among 3080 individuals in the present study, 2016
| Description | Percentage (95% CI) |
|---|---|
| Proportion of respondents who sought care for mental health (yes/no) out of total number of respondents | 9.7% (8.2–11.2) |
| Proportion of respondents who sought care for mental health (yes/no) out of total number of respondents who identified a mental health problem | 31.0% (27.7–34.4) |
| Proportion of respondents who sought care for mental health (yes/no) out of total number of respondents who identified depressive symptoms and/or suicidal ideation | 26.0% (22.8–29.3) |
| Proportion of respondents who sought care for depressive symptoms out of total number of respondents who sought care | 94.8% (91.8–97.8) |
| Proportion of respondents who were treated for mental health (yes/no) out of total number of respondents | 8.4% (7.1–9.7) |
| Proportion of respondents who were treated for depression (yes/no) out of total number of respondents who identified period of sadness or loss of energy that lasted more than 2 weeks | 33.0% (28.8–37.3) |
| Proportion of respondents who were treated for suicidal ideation (yes/no) out of total number of respondents who identified thoughts of suicide or self-harm in the last month | 10.7% (5.2–16.2) |
Table 6.
Mental Health Care-seeking: multivariable analyses of individuals in the present study, 2016–2017
| Care-seeking among depressive symptoms and suicidal ideation | Care-seeking among depressive symptoms only | Care-seeking among suicidal ideation only | ||||
|---|---|---|---|---|---|---|
| AOR | P value | AOR | P value | AOR | P value | |
| Sexa | ||||||
| Male | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Female | 1.07 (0.75–1.52) | 0.710 | 1.12 (0.74–1.69) | 0.586 | 1.07 (0.63–1.83) | 0.798 |
| Urban or ruralb | ||||||
| Urban | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Rural | 1.78 (1.13–2.79) | 0.012 | 1.38 (0.84–2.27) | 0.204 | 2.01 (0.99–4.07) | 0.054 |
| Level of schoolc | ||||||
| No school | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Basic | 0.96 (0.64–1.46) | 0.858 | 1.00 (0.62–1.61) | 0.99 | 0.72 (0.41–1.27) | 0.256 |
| Higher | 0.75 (0.44–1.28) | 0.288 | 0.71 (0.39–1.31) | 0.271 | 0.65 (0.29–1.42) | 0.279 |
| Aged | ||||||
| 26–35 | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| 18–25 | 1.04 (0.67–1.63) | 0.856 | 1.10 (0.66–1.83) | 0.717 | 1.21 (0.61–2.41) | 0.583 |
| 36–45 | 1.21 (0.78–1.87) | 0.396 | 1.36 (0.82–2.24) | 0.229 | 1.35 (0.72–2.54) | 0.356 |
| 46–55 | 1.31 (0.78–2.22) | 0.313 | 1.66 (0.91–3.00) | 0.097 | 1.13 (0.52–2.42) | 0.760 |
| > 55 | 1.18 (0.74–1.89) | 0.489 | 1.22 (0.72–2.05) | 0.461 | 0.94 (0.42–2.07) | 0.872 |
| Marital statusf | ||||||
| Married | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Divorced/separated | 1.51 (0.87–2.63) | 0.145 | 1.27 (0.65–2.47) | 0.482 | 2.37 (1.11–5.06) | 0.025 |
| Widowed | 0.72 (0.41–1.28) | 0.265 | 0.64 (0.34–1.20) | 0.162 | 1.20 (0.53–2.72) | 0.654 |
| Single | 0.83 (0.26–2.62) | 0.754 | 0.95 (0.27–3.29) | 0.932 | 1.28 (0.24–6.86) | 0.771 |
| Religionf | ||||||
| Pentecostal | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Catholic | 0.92 (0.55–1.52) | 0.732 | 0.59 (0.33–1.04) | 0.067 | 2.06 (0.95–4.47) | 0.068 |
| Muslim | 0.63 (0.13–3.04) | 0.564 | 0.57 (0.11–2.99) | 0.503 | 2.18 (0.20–23.84) | 0.523 |
| Zione | 1.21 (0.76–1.94) | 0.426 | 1.00 (0.58–1.72) | 0.998 | 1.75 (0.90–3.40) | 0.102 |
| Anglican | 0.55 (0.12–2.63) | 0.456 | 0.48 (0.10–2.39) | 0.37 | 1.31 (0.12–14.63) | 0.827 |
| Johan Masowe/Johan Maranga | 0.72 (0.14–3.71) | 0.698 | N/A | N/A | 2.33 (0.40–13.69) | 0.348 |
| No religion | 0.66 (0.39–1.12) | 0.128 | 0.75 (0.41–1.38) | 0.355 | 0.85 (0.40–1.82) | 0.682 |
| Other Christian | 1.12 (0.46–2.72) | 0.801 | 1.38 (0.47–4.06) | 0.558 | 1.65 (0.53–5.14) | 0.39 |
| Not sure | N/A | N/A | N/A | N/A | N/A | N/A |
| Other | 0.76 (0.33–1.77) | 0.522 | 0.70 (0.28–1.76) | 0.446 | 0.65 (0.13–3.13) | 0.59 |
| SESe | ||||||
| 5th quintile | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| 1st quintile | 0.56 (0.29–1.08) | 0.083 | 0.70 (0.34–1.44) | 0.328 | 0.53 (0.20–1.42) | 0.209 |
| 2nd quintile | 0.55 (0.29–1.04) | 0.067 | 0.57 (0.28–1.17) | 0.125 | 0.58 (0.22–1.53) | 0.272 |
| 3rd quintile | 0.76 (0.42–1.36) | 0.349 | 0.87 (0.45–1.68) | 0.682 | 0.42 (0.16–1.08) | 0.071 |
| 4th quintile | 1.03 (0.61–1.72) | 0.923 | 1.13 (0.63–2.03) | 0.682 | 0.93 (0.42–2.04) | 0.847 |
| Alcoholf | ||||||
| Never | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Once a month or less | 0.62 (0.37–1.05) | 0.078 | 0.66 (0.37–1.20) | 0.174 | 0.48 (0.21–1.07) | 0.074 |
| 2–4× per month | 1.08 (0.55–2.12) | 0.822 | 1.50 (0.68–3.31) | 0.318 | 1.03 (0.38–2.82) | 0.953 |
| 2–4× per week | 0.78 (0.32–1.92) | 0.595 | 0.79 (0.31–2.03) | 0.625 | 1.46 (0.56–4.70) | 0.522 |
| 4 or more per week | 1.31 (0.38–4.57) | 0.671 | 2.83 (0.45–17.64) | 0.265 | 1.43 (0.26–7.71) | 0.678 |
| Overall disabilityf | ||||||
| ≤ 12 | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| 13–24 | 1.23 (0.86–1.75) | 0.264 | 1.29 (0.86–1.92) | 0.216 | 1.57 (0.89–2.77) | 0.121 |
| 25–36 | 1.19 (0.58–2.44) | 0.632 | 1.53 (0.63–3.70) | 0.346 | 1.18 (0.42–3.34) | 0.752 |
| > = 37 | 1.02 (0.41–2.54) | 0.968 | 0.73 (0.27–2.03) | 0.551 | 0.57 (0.07–4.27) | 0.581 |
| Any injury (in the past year)f | ||||||
| No injury | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Injury | 1.17 (0.75–1.83) | 0.476 | 1.35 (0.82–2.22) | 0.245 | 0.92 (0.48–1.78) | 0.805 |
| Injury by assault (in the past year)f | ||||||
| No assault injury | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Assault injury | 0.76 (0.24–2.39) | 0.644 | 0.76 (0.19–3.03) | 0.696 | 0.89 (0.17–4.58) | 0.891 |
Adjusted for age, SES, education, residence
Adjusted for age, SES, education, sex
Adjusted for SES, sex, residence
Adjusted for SES, residence
Adjusted for age, sex, education, residence
Adjusted for age, sex, education, SES, residence
Table 7.
Multivariable analyses of individuals in the present study stratified by gender, 2016–2017
| Depressive symptoms | Suicidal ideation | Mental health care-seeking | ||||
|---|---|---|---|---|---|---|
| AOR (95% CI) | P value | AOR (95% CI) | P value | AOR (95% CI) | P value | |
| Marital statusa | ||||||
| Female | ||||||
| Married | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Divorced/separated | 1.96 (1.32–2.92) | 0.001 | 1.74 (1.15–2.62) | 0.008 | 1.93 (1.17–3.18) | 0.010 |
| Widowed | 2.33 (1.55–3.50) | 0.000 | 1.54 (1.02–2.33) | 0.040 | 1.17 (0.67–2.06) | 0.576 |
| Single | 1.85 (0.88–3.88) | 0.106 | 1.30 (0.55–3.10) | 0.549 | 0.77 (0.23–2.62) | 0.680 |
| Male | ||||||
| Married | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Divorced/separated | 1.12 (0.31–3.94) | 0.876 | 1.82 (0.52–6.41) | 0.353 | 1.58 (0.35–7.20) | 0.551 |
| Widowed | 0.42 (0.05–3.67) | 0.435 | 0.72 (0.09–5.67) | 0.759 | 0.95 (0.12–7.73) | 0.962 |
| Single | 0.87 (0.29–2.63) | 0.806 | 0.54 (0.13–2.37) | 0.418 | 2.03 (0.57–7.22) | 0.272 |
| Alcohola | ||||||
| Female | ||||||
| Never | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Once a month or less | 1.74 (1.15–2.64) | 0.009 | 1.63 (1.06–2.51) | 0.025 | 0.90 (0.50–1.62) | 0.733 |
| 2–4× per month | 2.35 (1.15–4.80) | 0.019 | 2.07 (0.99–4.32) | 0.053 | 1.14 (0.42–3.12) | 0.792 |
| 2–4× per week | 2.50 (1.04–6.02) | 0.040 | 0.57 (0.16–1.99) | 0.377 | 0.66 (0.15–3.00) | 0.590 |
| 4 or more per week | 0.68 (0.15–3.16) | 0.627 | 3.34 (1.00–11.17) | 0.050 | 1.31 (0.27–6.32) | 0.737 |
| Male | ||||||
| Never | 1 (ref) | (ref) | 1 (ref) | (ref) | 1 (ref) | (ref) |
| Once a month or less | 1.05 (0.67–1.65) | 0.820 | 1.40 (0.86–2.29) | 0.175 | 1.42 (0.78–2.56) | 0.248 |
| 2–4× per month | 0.90 (0.52–1.56) | 0.705 | 1.27 (0.69–2.31) | 0.442 | 2.05 (1.09–3.88) | 0.027 |
| 2–4× per week | 1.62 (0.89–2.96) | 0.117 | 2.55 (1.37–4.75) | 0.003 | 1.25 (0.53–2.94) | 0.612 |
| 4 or more per week | 0.97 (0.29–3.25) | 0.965 | 2.03 (0.65–6.35) | 0.223 | 2.61 (0.74–9.20) | 0.137 |
Adjusted for age, sex, education, SES, residence
Footnotes
Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
Ethical approval and consent to participate The study was approved by the Institutional Bio-ethics Committee of the National Institute of Health in Mozambique. All survey respondents provided written informed consent. In cases where participant could not read or write, s/he provided a thumbprint.
Statement of Originality The authors attest that all work is original and that it has not been published or submitted anywhere other than to Social Psychiatry and Psychiatric Epidemiology.
References
- 1.Khan MM (2005) Suicide prevention and developing countries. J R Soc Med 98:459–463 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Patel V, Araya R, Chatterjee S, Chisholm D, Cohen A, De Silva M et al. (2007) Treatment and prevention of mental disorders in low-income and middle-income countries. Lancet 370:991–1005 [DOI] [PubMed] [Google Scholar]
- 3.Bromet E, Andrade LH, Hwang I, Sampson NA, Alonso J, de Girolamo G et al. (2011) Cross-national epidemiology of DSM-IV major depressive episode. BMC Med 9:90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Andrade L, Caraveo-Anduaga JJ, Berglund P, Bijl RV, De Graaf R, Vollebergh W et al. (2003) The epidemiology of major depressive episodes: results from the International Consortium of Psychiatric Epidemiology (ICPE) Surveys. Int J Methods Psychiatr Res 12(1):3–21 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Van de Velde S, Bracke P, Levecque K (2010) Gender differences in depression in 23 European countries. Cross-national variation in the gender gap in depression. Soc Sci Med 71(2):305–313 [DOI] [PubMed] [Google Scholar]
- 6.Kessler RC, Birnbaum HG, Shahly V, Bromet E, Hwang I, McLaughlin KA, et al. Age differences in the prevalence and comorbidity of DSM-IV major depressive episodes: results from the WHO World Mental Health Survey Initiative. Depress Anxiety [Internet]. 2010;27(4):351–64. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20037917%5Cnonlinelibrary.wiley.com/store/10.1002/da.20634/asset/20634_ftp.pdf?v=1&t=hiebojc9&s=73d7e920588e5c5a1a67d9826a671a02d49df746 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Guerra M, Prina AM, Ferri CP, Acosta D, Gallardo S, Huang Y et al. (2016) A comparative cross-cultural study of the prevalence of late life depression in low and middle income countries. J Affect 190:362–368 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kessler RC, Bromet EJ (2013) The epidemiology of depression across cultures. Annu Rev Public Health 34(1):119–138. 10.1146/annurev-publhealth-031912-114409 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.World Health Organization. Depression and other common mental disorders: global health estimates. 2017
- 10.Bromet EJ, Gluzman SF, Paniotto VI, Webb CPM, Tintle NL, Zakhozha V et al. (2005) Epidemiology of psychiatric and alcohol disorders in Ukraine: findings from the Ukraine World Mental Health Survey. Soc Psychiatry Psychiatr Epidemiol 40:681–690 [DOI] [PubMed] [Google Scholar]
- 11.Nock MK, Borges G, Bromet EJ, Alonso J, Angermeyer M, Beautrais A et al. (2008) Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. Br J Psychiatry 192(2):98–105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mars B, Burrows S, Hjelmeland H, Gunnell D (2014) Suicidal behaviour across the African continent: a review of the literature. BMC Public Health 14:606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Vijayakumar L, John S, Pirkis J, Whiteford H. Suicide in developing countries (2): risk factors. Crisis. 2005;26(3):112–9. Available from: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=16276753 [DOI] [PubMed] [Google Scholar]
- 14.Vijayakumar L, Rajkumar S (1999) Are risk factors for suicide universal? A case–control study in India. Acta Psychiatr Scand 99(6):407–411. 10.1111/j.1600-0447.1999.tb00985.x [DOI] [PubMed] [Google Scholar]
- 15.Yoshimasu K, Kiyohara C, Miyashita K (2008) Suicidal risk factors and completed suicide: meta-analyses based on psychological autopsy studies. Environ Health Prev Med 13:243–256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Schrijvers DL, Bollen J, Sabbe BGC (2012) The gender paradox in suicidal behavior and its impact on the suicidal process. J Affect Disord 138:19–26 [DOI] [PubMed] [Google Scholar]
- 17.Wagenaar BH, Raunig-Berhó M, Cumbe V, Rao D, Napúa M, Sherr K (2016) Suicide attempts and deaths in Sofala, Mozambique, from 2011 to 2014: Who, Where, and from What. Crisis 37(6):445–453 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Patel V, Simbine APF, Soares IC, Weiss HA, Wheeler E (2007) Prevalence of severe mental and neurological disorders in Mozambique: a population-based survey. Lancet 370(9592):1055–1060 [DOI] [PubMed] [Google Scholar]
- 19.Audet C, Wainberg M, Oquendo M, Yu Q, Blevins Peratikos M, Duarte C et al. (2018) Depression among female heads-of-house-hold in rural Mozambique: a cross-sectional population-based survey. J Affect Disord 227:48–55 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Institute for Health Metrics and Evaluation (IHME) (2016) GBD compare data visualization. IHME, University of Washington, Seattle [Google Scholar]
- 21.Abajobir AA, Abate KH, Abbafati C, Abbas KM, Abd-Allah F, Abdulkader RS et al. (2017) Global, regional, and national disability-adjusted life-years (DALYs) for 333 diseases and injuries and healthy life expectancy (HALE) for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 390(10100):1260–1344 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.World Health Organization (WHO) (2018) Global health observatory data: suicide rates per 100,000 population
- 23.Santos PF, Wainberg ML, Caldas-de-Almeida JM, Saraceno B, Mari J (2016) Overview of the mental health system in Mozambique: addressing the treatment gap with a task-shifting strategy in primary care. Int J Ment Health Syst 10:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wagenaar BH, Cumbe V, Raunig-Berhó M, Rao D, Napúa M, Hughes JP et al. (2015) Health facility determinants and trends of ICD-10 outpatient psychiatric consultations across Sofala, Mozambique: time-series analyses from 2012 to 2014. BMC Psychiatry 15:227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wagenaar BH, Stergachis A, Rao D, Hoek R, Cumbe V, Napúa M et al. (2015) The availability of essential medicines for mental healthcare in Sofala, Mozambique. Glob Health Action; 10.3402/gha.v8.27942 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sherr K, Cuembelo F, Michel C, Gimbel S, Micek M, Kariaganis M et al. (2013) Strengthening integrated primary health care in Sofala, Mozambique . BMC Health Serv Res 13(2):4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.The World Bank. The world bank: Mozambique [Internet]. Available from: https://data.worldbank.org/country/mozambique
- 28.United States Central Intelligence Agency. CIA world factbook: Mozambique [Internet]. Available from: https://www.cia.gov/library/publications/the-world-factbook/geos/print_mz.html
- 29.National Statistics Institute (2017) Mozambique 2017 census. Maputo, Mozambique [Google Scholar]
- 30.Mozambican Ministry of Health. Inquérito de Indicadores de Imunização, Malária e HIV/SIDA em Moçambique (IMASIDA). 2016
- 31.UNICEF. Monitoring the Situation of Children and Women: Mozambique [Internet]. Available from: https://data.unicef.org/country/moz/
- 32.Moçambique demographic and health survey. 2011
- 33.Wagenaar BH, Augusto A, Asbjornsdottir K, Akullian A, Manaca N, Chale F, Muanido A, Covele A, Michel C, Gimbel S, Radford T, Girardot B, Sherr K (2018) Developing a representative community survey sampling frame using satellite imagery in Mozambique. Int J Health Geogr 17:37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hartung C, Anokwa Y, Brunette W, Lerer A, Tseng C, Borriello G (2010) Open data kit: tools to build information services for developing regions. Proc Int Conf Inf Commun Technol Dev [Google Scholar]
- 35.O’Donnell O, van Doorslaer E, Wagstaff A, Lindelow M (2008) Analyzing health equity using household survey data: a guide to techniques and their implementation. Health Equity using, Pp. 69–82 [Google Scholar]
- 36.Filmer D, Pritchett L (2001) Estimating wealth effects without expenditure data–or tears: an application to educational enrollments in states of India. Demography 38(1):115–132. 10.1353/dem.2001.0003 [DOI] [PubMed] [Google Scholar]
- 37.World Health Organization (WHO) (2010) Measuring health and disability: manual for WHO disability assessment schedule WHODAS 2.0
- 38.Tomlinson M, Grimsrud AT, Stein DJ, Williams DR, Myer L (2009) The epidemiology of major depression in South Africa: results from the South African Stress and Health study. South African Med J 99(5):368–373 [PMC free article] [PubMed] [Google Scholar]
- 39.Adewuya AO, Ola BA, Coker OA, Atilola O, Zachariah MP, Olug-bile O et al. (2016) Prevalence and associated factors for suicidal ideation in the Lagos state mental health survey, Nigeria. Br J Psychiatry 2(6):385–389 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Johnson K, Asher J, Rosborough S, Raja A, Panjabi R, Beadling C et al. (2008) Association of combatant status and sexual violence with health and mental health outcomes in postconflict liberia. JAMA, J Am Med Assoc 300(6):676–690 [DOI] [PubMed] [Google Scholar]
- 41.Wagenaar BH, Hagaman AK, Kaiser BN, McLean KE, Kohrt BA (2012) Depression, suicidal ideation, and associated factors: a cross-sectional study in rural Haiti. BMC Psychiatry 12:149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Kohrt B, Hruschka D, Worthman C, Kunz R, Baldwin J, Upad-haya N et al. (2012) Political violence and mental health in Nepal: prospective study. Br J Psychiatry 201(4):268–275 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Uwakwe R, Oladeji BD, Gureje O (2012) Traumatic events and suicidal behaviour in the Nigerian survey of mental health and well-being. Acta Psychiatr Scand 126:458–466 [DOI] [PubMed] [Google Scholar]
- 44.Kinyanda E, Musisi S, Biryabarema C, Ezati I, Oboke H, Oji-ambo-Ochieng R et al. (2010) War related sexual violence and it’s medical and psychological consequences as seen in Kitgum, Northern Uganda: a cross-sectional study. BMC Int Health Hum Rights 10:28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Kizza D, Knizek BL, Kinyanda E, Hjelmeland H (2012) Men in despair: a qualitative psychological autopsy study of suicide in Northern Uganda. Transcult Psychiatry 49:696–717 [DOI] [PubMed] [Google Scholar]
- 46.Newitt M A short history of Mozambique. 2017
- 47.Wagenaar BH, Kohrt BA, Hagaman AK, McLean KE, Kaiser BN (2013) Determinants of care seeking for mental health problems in rural Haiti: culture, cost, or competency. Psychiatr Serv 64(4):366–372. 10.1176/appi.ps.201200272 [DOI] [PubMed] [Google Scholar]
- 48.Kovess-Masfety V, Evans-Lacko S, Williams D, Andrade LH, Benjet C, Ten Have M et al. (2017) The role of religious advisors in mental health care in the World Mental Health surveys. Soc Psychiatry Psychiatr Epidemiol 52:353–367 [DOI] [PubMed] [Google Scholar]
- 49.Aghukwa CN (2012) Care seeking and beliefs about the cause of mental illness among Nigerian psychiatric patients and their families. Psychiatr Serv 63:616–618 [DOI] [PubMed] [Google Scholar]
- 50.Khoury NM, Kaiser BN, Keys HM, Brewster ART, Kohrt BA (2012) Explanatory models and mental health treatment: is vodou an obstacle to psychiatric treatment in rural Haiti? Cult Med Psychiatry 36:514–534 [DOI] [PubMed] [Google Scholar]
- 51.Monteiro NM, Ndiaye Y, Blanas D, Ba I (2014) Policy perspectives and attitudes towards mental health treatment in rural Senegal. Int J Ment Health Syst. 8:9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Abeyasinghe DRR, Tennakoon S, Rajapakse TN (2012) The development and validation of the peradeniya depression scale (PDS)—a culturally relevant tool for screening of depression in Sri Lanka. J Affect Disord 142:143–149 [DOI] [PubMed] [Google Scholar]
- 53.Petersen I, Lund C, Bhana A, Flisher AJ (2012) A task shifting approach to primary mental health care for adults in South Africa: human resource requirements and costs for rural settings. Health Policy Plan 27:42–51 [DOI] [PubMed] [Google Scholar]
- 54.Mendenhall E, De Silva MJ, Hanlon C, Petersen I, Shidhaye R,Jordans M et al. (2014) Acceptability and feasibility of using nonspecialist health workers to deliver mental health care: stakeholder perceptions from the PRIME district sites in Ethiopia, India, Nepal, South Africa, and Uganda. Soc Sci Med 118:33–42 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.World Health Organization (2010) mhGAP intervention guide for mental, neurological and substance use disorders in nonspecialized health settings: mental health gap action programme (mhGAP). pp 1–121 [PubMed] [Google Scholar]
- 56.Henderson C, Noblett J, Parke H, Clement S, Caffrey A, Gale-Grant O et al. (2014) Mental health-related stigma in health care and mental health-care settings. Lancet Psychiatry. 1:467–482 [DOI] [PubMed] [Google Scholar]
- 57.Ungar T, Knaak S, Szeto ACH (2016) Theoretical and practical considerations for combating mental illness stigma in health care. Community Ment Health J 52:262–271 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Thornicroft G, Alem A, Dos Santos RA, Barley E, Drake RE, Gregorio G et al. (2010) WPA guidance on steps, obstacles and mistakes to avoid in the implementation of community mental health care. World Psychiatry. 9:67–77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Eaton J, Nwefoh E, Okafor G, Onyeonoro U, Nwaubani K, Henderson C (2017) Interventions to increase use of services; mental health awareness in Nigeria. Int J Ment Health Syst 11:66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Vijayakumar L (2015) Suicide in women. Indian J Psychiatry 57(Suppl 2):233–238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Kuo CY, Liao SC, Lin KH, Wu CL, Lee MB, Guo NW et al. (2012) Predictors for suicidal ideation after occupational injury. Psychiatry Res 198(3):430–435 [DOI] [PubMed] [Google Scholar]
- 62.Asfaw A, Souza K (2012) Incidence and cost of depression after occupational injury. J Occup Environ Med 54(9):1086–1091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Wiseman T, Foster K, Curtis K (2013) Mental health following traumatic physical injury: an integrative literature review. Injury 44:1383–1390 [DOI] [PubMed] [Google Scholar]
- 64.Meltzer H, Bebbington P, Brugha T, McManus S, Rai D, Dennis MS et al. (2012) Physical ill health, disability, dependence and depression: results from the 2007 national survey of psychiatric morbidity among adults in England. Disabil Health J 5:102–110 [DOI] [PubMed] [Google Scholar]
- 65.Shen S- C, Huang K- H, Kung P- T, Chiu L- T, Tsai W- C (2017) Incidence, risk, and associated factors of depression in adults with physical and sensory disabilities: a nationwide population-based study. PLoS ONE 12(3):e0175141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Lund C, Breen A, Flisher AJ, Kakuma R, Corrigall J, Joska JA et al. (2010) Poverty and common mental disorders in low and middle income countries: a systematic review. Soc Sci Med 71(3):517–528 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Lund C, De Silva M, Plagerson S, Cooper S, Chisholm D, Das J et al. (2011) Poverty and mental disorders: breaking the cycle in low-income and middle-income countries. Lancet 378:1502–1514 [DOI] [PubMed] [Google Scholar]
- 68.Sullivan LE, Fiellin DA, O’Connor PG (2005) The prevalence and impact of alcohol problems in major depression: a systematic review. Am J Med 118(4):330–341 [DOI] [PubMed] [Google Scholar]
- 69.Agabio R (2017) Non-specialist health workers to treat excessive alcohol consumption and depression. Lancet 389:133–135 [DOI] [PubMed] [Google Scholar]
- 70.Brady J (2006) The association between alcohol misuse and suicidal behaviour. Alcohol Alcohol 41:473–478 [DOI] [PubMed] [Google Scholar]
- 71.Magaard JL, Seeralan T, Schulz H, Levke A, Tt B (2017) Factors associated with help-seeking behaviour among individuals with major depression: a systematic review. PLoS ONE 12(5):1–17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 72.Abaerei AA, Ncayiyana J, Levin J (2017) Health-care utilization and associated factors in Gauteng province, South Africa. Glob Health Action 10:1305765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 73.Muula AS, Ngulube TJ, Siziya S, Makupe CM, Umar E, Prozesky HW et al. (2007) Gender distribution of adult patients on highly active antiretroviral therapy (HAART) in Southern Africa: a systematic review. BMC Public Health 7:63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Burns JK, Tomita A (2015) Traditional and religious healers in the pathway to care for people with mental disorders in Africa: a systematic review and meta-analysis. Soc Psychiatry Psychiatr Epidemiol 50:867–877 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 75.Ahmed SM, Tomson G, Petzold M, Kabir ZN (2005) Socioeconomic status overrides age and gender in determining health-seeking behavior in rural Bangladesh. Bull World Health Organ 83(2):109–117 [PMC free article] [PubMed] [Google Scholar]
- 76.Van Der Hoeven M, Kruger A, Greeff M (2012) Differences in health care seeking behaviour between rural and urban communities in South Africa. Int J Equity Health 11(1):31. [DOI] [PMC free article] [PubMed] [Google Scholar]