Suicidal behaviours among deaf adolescents in Ghana: a cross-sectional study

Abstract Background A growing global concern is that suicide research has paid little attention to young people with disabilities, particularly, in low- and middle-income countries (LAMICs). We aimed to estimate the 12-month prevalence of suicidal ideation and attempt and describe some associations among deaf adolescents in Ghana. Methods This is a cross-sectional anonymous self-report survey involving a nationally representative random sample of 450 school-going deaf adolescents. Data analysis included bivariate and multivariable approaches. Results The overall 12-month prevalence of suicidal ideation was 19·3% (95% confidence interval [CI] = 15·8–23·3) and suicidal attempt was 15·6% (95% CI = 12·3–19·2). Although alcohol use and parental divorce were strongly associated with increased odds of both suicidal ideation and attempt, high subjective mental well-being was associated with reduced odds of both suicidal ideation and attempt. Living with no parents and being a final year student were associated with suicidal ideation, while male gender was associated with suicidal attempt. Conclusions The prevalence of suicidal behaviours among school-going deaf adolescents in this study compares with estimates among in-school non-deaf adolescents in Ghana and other LAMICs in Africa, and also highlights the need for prevention efforts against the onset of suicidal ideation and possible transition to attempt and suicide among deaf adolescents.


Introduction
The World Health Organization (WHO) defines suicidal behaviour as 'a range of behaviours that include thinking about suicide (or ideation), planning for suicide, attempting suicide and suicide itself '. 1 Globally, suicide is the second leading cause of death among 15-19-year-olds, 1,2 with 79% of the annual world-wide suicide-related mortality reported in low-and middle-income countries (LAMICs). 1,3,4 Across the general population, suicidal ideation is a strong risk for attempting suicide; attempted suicide represents the single significant risk for suicide. 1,5,6 Given that disabilities are associated with negative physical and mental health outcomes, persons with functional disabilities-related to hearing and sight-have been found to be at an increased risk of attempted suicide, suicide and other emotional and behavioural problems. [7][8][9][10][11][12] However, there is a growing global concern that suicide research has paid little attention to young people with disabilities, especially, in LAMICs: in particular, much of our understanding of suicidal behaviours among deaf adolescents is based on evidence from high-income countries. 9,11,[13][14][15][16][17][18] We performed and updated a systematic search for literature (We searched Global Index Medicus, Index Medicus for South-East Asia Region, PsycINFO, African Journals OnLine, MEDLINE and African Index Medicus up to July 2020, using keywords [e.g. ('suicide ideation' OR 'suicidal ideation' OR 'self-harm ideation' OR suicid * OR self-injur * OR mental) AND (deaf children OR 'deaf adolescents' OR 'deaf young people' OR disability OR adolescen * OR child * OR students OR teen * OR 'young adults' OR youth OR pupils)]. We did not apply any date or language restrictions. Our geographic search filter to identify countries in within low-and middle-income context included names of the countries in both English and languages relevant to the countries. We supplemented the searches by reviewing references and forward citations of relevant articles.) from Africa to contextualize the current study. Apart from relevant studies involving school-going non-deaf adolescents, we found neither published primary studies nor systematic reviews or meta-analyses with included studies involving deaf adolescents from African countries. [19][20][21][22][23][24][25] Among in-school non-deaf adolescents, recent regional systematic reviews and meta-analyses suggest that relative to other LAMICs, the 12-month prevalence estimate of suicidal ideation (20 [20][21][22][23] The evidence also suggests that suicidal behaviours among adolescents are multicausal, with specific risk and protective factors existing at the personal level (e.g. female gender, alcohol and substance use, depression), family context (e.g. parental understanding, conflict with parents), school (e.g. schoolwork problems, bullying victimization) and other social relationship contextual factors (e.g. break-up, sexual abuse victimization, peer support). [21][22][23] In Ghana, only one study has reported social discrimination as a critical reason for suicidal ideation among adult mothers with physical disabilities. 26 Evidence also suggests that although deaf adolescents have access to health facilities in Ghana, professional health care is hampered by communication barriers such as the lack of sign language interpreters. 27 Taken together, considering that the reduction of suiciderelated deaths has been given special priority in the sustainable development goals, it is important to contribute sound and timely research evidence, particularly, among young vulnerable groups such as deaf adolescents in LAMICs to inform intervention and prevention efforts.

Design, setting and participants
The community-agreed recommendations of Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guided the design and reporting of this study. 28 We conducted a cross-sectional survey, using a self-report anonymous questionnaire, among deaf students attending Junior High Schools (JHSs) (In Ghana, Junior High Schools have a 3-year duration and are targeted at young people aged 12-14 years, but due to significant delayed school enrolment of young people with disabilities, typically, late and older adolescents aged 15-24 years are also predominantly found at this level of basic education in special schools in Ghana.) in Ghana. At the time of designing this study (August 2019) there were 1030 students across all the 13 JHSs for the deaf in the country. We predetermined a minimum sample size of 288, using Yamane's (1967) formula for proportions, with 0·05 level of precision. 29 We used a two-stage cluster sampling approach to firstly select seven JHSs for the deaf with probability proportional to enrolment size across the 16 administrative regions of Ghana. In stage 2, classes were randomly selected with all students in each class eligible to participate in the survey. Figure 1 summarizes the school selection and sampling of participants for this study. We approached and invited 468 participants, but 450 provided complete data included in the final analysis, representing a response rate of 96.1% (171 females and 221 males; aged 13-24 years; mean = 18.4; standard deviation = 2.60; modal age = 18).

Outcome measures
Two outcome variables were included in this study: suicidal ideation and attempt. In order to facilitate comparison, we adopted two single-items from the 2012 Ghana WHO-Global School-based Health Survey (WHO-GSHS), 30 one each to measure suicidal ideation and attempt. Specifically, the item, 'during the past 12 months, did you ever seriously consider attempting suicide?' was used to measure suicidal ideation; the responses were dichotomized as No (0) or Yes (1). Suicidal attempt was measured with the question 'during the past 12 months, how many times did you actually attempt suicide?' The responses were '0', '1', '2 or 3', '4 or 5' and '6 or more times'. Consistent with most published analyses of the WHO-GSHS data from sub-Saharan Africa, 20-24 we applied a binary recoding to this item for analysis, No (no attempt = 0) and Yes (one or more attempts = 1).

Exposure variables
This study involved 12 items assessing personal, family, school and interpersonal factors found in the 2012 Ghana WHO-GSHS to be associated with suicidal behaviours as risk and protective factors (see in Supplementary Table 1). For example, alcohol use (never drink alcohol or ≥1 drinks), parental divorce (no/yes), schoolwork problems (no/yes), bullying victimization (no/yes) and break-up (no/yes). Also, this study adopted one item from the 5-item Duke University Religion Index (DUREL) to measure religious participation: 'How often do you attend church or other religious meetings?' with response rating options ranging from (1) 'never' to (6) 'more than once per week'. 31 Considering the strong association between suicidal behaviours, and psychological functioning, life satisfaction and one's ability to develop and maintain mutually beneficial social relationships, 32 we adopted the 14-item Warwick-Edinburgh Mental Well-being Scale (WEMWBS) to assess subjective mental well-being. 13,33 It is a 5-score scale, from 'none of the time' [1] to 'all of the time' [5] (e.g. 'I've been feeling optimistic about the future'), with the possible total score ranging from 14 to 70 points. Higher scores indicate better mental well-being.
Internationally, the WEMWBS has demonstrated strong internal consistency and a satisfactory Cronbach's alpha; 34 the Cronbach's alpha in the current study was 0·83. Also, as shown in Supplementary Table 1, eight socio-demographic variables were included (e.g. gender, age, family structure). Prior to administration, the questionnaire was expert reviewed by two basic school Ghanaian Sign Language teachers and an adolescent suicide researcher.

Procedure
The survey took place from October 2019 to January 2020. The survey was administered to participants in their school's assembly hall or a larger classroom, with sitting arrangement spaced by reasonable distance to ensure privacy. We clarified participants' queries in the Ghanaian Sign Language-the sign language that is used for the instruction and examination of deaf students at all levels of education in Ghana. Averagely, the completion of the questionnaire lasted between 55 and 75 min. After completing the survey, each student placed their answered questionnaire in an opaque box before exiting the hall.

Data analysis
All data analyses were performed using the Statistical Package for Social Sciences (SPSS version 26.0 for Windows).
As the loss of cases due to missing data was <5%, the list-wise deletion of missing data strategy was applied. 35 The coding of the variables included in our statistical analysis and proportion of missing data are shown in Supplementary Table 1. The data analysis proceeded in two stages. Firstly, we performed descriptive analysis by applying frequencies, proportions and the Pearson's Chi-square tests (χ 2 and point-biserial correlation (r pb ) tests to examine the bivariate relationships between suicidal behaviours and the exposure variables and socio-demographic factors. Secondly, adjusted multivariable logistic regression analyses examining the possible associations between suicidal behaviours and the specified correlates and socio-demographic variables were performed. As recommended by leading logistic regression modelling experts, the candidate correlates were entered in the multivariable logistic regression models regardless of their statistically significant bivariate relationship with the outcome variables. [36][37][38] The socio-demographic variables were included as covariates. We reported the results of the logistic regression as odds ratios with 95% confidence intervals (CI) and P-values. Statistically significant results were also determined using P < 0·05.

Ethics
This study was approved by the Department of Psychology Research Ethics Committee, University of Ghana, Accra. We followed the ethical procedures of the Ghana Education Service (GES) for conducting research involving students in special schools in Ghana. The Special Education Unit of GES and heads of all the participating schools permitted this study. Each participant signed a written consent form prior to responding to the survey; parents/guardians of underage participants provided consent and we obtained the assent of participants younger than 18 years, prior to responding to the survey.

Main findings of this study
Across the total sample, nearly two in 10 participants reported suicidal ideation, while one out of 10 participants reported suicidal attempt in the previous 12 months. The factors associated with suicidal ideation and attempt were multi-contextual: personal, family and school related. Although alcohol use and parental divorce were strongly associated with increased odds of both suicidal ideation and attempt, high subjective Grade: Living arrangement: Live with both parents Reference Reference Personal and lifestyle variables: Family factors: Note: β = beta value; AOR = adjusted odds ratios; CI = Confidence Interval; GOF = goodness of fit mental well-being was associated with reduced odds of both suicidal ideation and attempt. Living with no parents, being a final year student and being deaf postnatally were strongly associated with increased odds of suicidal ideation, while male gender was associated with increased odds of suicidal attempt. Although age showed positive associations with sui-cidal ideation and attempt, these associations did not reach the desired statistical significance.

What is already known on this topic
Most of what is already known about this topic is from highincome context, with earlier systematic reviews identifying no studies from LAMICs. 8,13 Earlier evidence from highincome countries suggests that lifetime estimates of suicidal ideation among deaf adolescents ranges widely between 4.6 and 44%, while 12-month attempted suicide ranges between 1.7 and 18%. 13 Recently, a county-wide cross-sectional study from the USA has reported a 12-month prevalence estimates of attempted suicide (14.9%) among deaf adolescents. 11 Generally, major depression, family dysfunction, low academic performance, alcohol use, peer victimization and peer relationship difficulties and sexual abuse victimization have been reported as significant associations of suicidal ideation and attempt among adolescents with functional disabilities in high-income countries. 11,13,14,39 The only available study reporting evidence from a low-income country is a recent school-based cross-sectional survey from Pakistan that reports lifetime suicidal attempt (19%) and 12-month suicidal ideation (17%) among school-going deaf adolescents. 15

What this study adds
This study represents a pioneering attempt at providing evidence on suicidal behaviours among deaf adolescents from Ghana; the evidence could be potentially informative for more expansive epidemiological studies from Ghana and other sub-Saharan African countries. Although the prevalence estimates and associated factors of suicidal behaviours in the current study are comparable to those reported among inschool non-deaf adolescents in Ghana and other countries in sub-Saharan Africa, 19-24 the adjusted multivariate logistic models of the current study have shown two new specific evidence: (1) there is a significant negative association between subjective mental well-being and both suicidal ideation and attempt, and (2) adolescents with congenital deafness are less likely to report suicidal ideation, compared to adolescents of postnatal deafness. These underscore the need for universal and targeted prevention efforts aimed at improving the mental well-being of school-going deaf adolescents in Ghana. Put together, although further studies (including qualitative research) are needed to explore the lived experiences and meanings of suicidal behaviours among school-going deaf adolescents in Ghana, the evidence of the current study could be pointing to the possibility that both adolescents without disabilities and those with disabilities may be faced with common multiple distress and challenges of health, education, growth and development, and family and social adversities in sub-Saharan Africa. [40][41][42][43][44] Limitations of this study Considering that in cross-sectional surveys both the outcome and exposure variables are measured at the same point in time, it stands to suggests that the evidence reported in this study cannot support causal inferences. 45 Although the use of single-item measures affords the opportunity of screening a larger sample at one point in time for suicidal behaviours, single-item measures have the potential of leading to misclassification of suicidal behaviour. 46,47 Relatedly, the self-report approach used (instead of a diagnostic approach) might have yielded less reliable responses in respect of suicidal ideation and attempt. Furthermore, besides the high tendency of nondisclosure among young people in suicide research, 48 some participants in this study might have also provided socially desirable responses due to the highly stigmatized and criminalized nature of attempted suicide in Ghana. 49,50 While these limitations provide a critical basis for cautious interpretations of the findings, the inclusion of a nationally representative sample of deaf students in JHSs in Ghana could make the findings a good reflection of the national prevalence and associated factors of suicidal ideation and attempt among deaf students attending JHSs.

Conclusions
The prevalence of suicidal ideation and attempt among deaf adolescents in this study compares with estimates among nondeaf adolescents in Ghana and other LAMICs in Africa, but also highlights the need for targeted and universal prevention efforts against the onset of suicidal ideation and possible transition to attempt and suicide among deaf adolescents.

Data availability statement
The data underlying this article will be shared on reasonable request to the corresponding author.

Supplementary data
Supplementary data are available at the Journal of Public Health online.