Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 BMC Psychiatry https://doi.org/10.1186/s12888-023-04775-z R E S E A R C H Open Access Prevalence of probable mental, neurological and substance use conditions and case detection at primary healthcare facilities across three districts in Ghana: findings from a cross-sectional health facility survey Kenneth Ayuurebobi Ae-Ngibise1,2, Lionel Sakyi1, Lyla Adwan-Kamara1, Crick Lund3,4 and Benedict Weobong5* Abstract Background Few studies have examined the prevalence of mental, neurological and substance use (MNS) conditions, case detection and treatment in primary healthcare in rural settings in Africa. We assessed prevalence and case detection at primary healthcare facilities in low-resource rural settings in Ghana. Methods A cross-sectional survey was conducted at the health facility level in three demonstration districts situated in Bongo (Upper East Region), Asunafo North (Ahafo Region) and Anloga (Volta Region) in Ghana. The study participants were resident adult (> 17 years) out-patients seeking healthcare at primary care facilities in each of the three demonstration districts. Data were collected on five priority MNS conditions: depression, psychosis, suicidal behaviour, epilepsy and alcohol use disorders. Results Nine hundred and nine (909) people participated in the survey. The prevalence of probable depression was 15.6% (142/909), probable psychotic symptoms was 12% (109/909), probable suicidal behaviour was 11.8% (107/909), probable epilepsy was 13.1% (119/909) and probable alcohol use disorders was 7.8% (71/909). The proportion of missed detection for cases of depression, self-reported psychotic symptoms, epilepsy and alcohol use disorders (AUD) ranged from 94.4 to 99.2%, and was similar across study districts. Depression was associated with self-reported psychotic symptoms (RR: 1.68; 95% CI: 1.12–1.54). For self-reported psychotic symptoms, a reduced risk was noted for being married (RR: 0.62; 95% CI: 0.39–0.98) and having a tertiary level education (RR: 0.12; 95% CI: 0.02–0.84). Increased risk of suicidal behaviour was observed for those attending a health facility in Asunafo (RR: 2.31; 95% CI: 1.27–4.19) and Anloga districts (RR: 3.32; 95% CI: 1.93–5.71). Age group of 35 to 44 years (RR: 0.43; 95% CI: 0.20–0.90) was associated with reduced risk of epilepsy. Being female (RR: 0.19; 95% CI: 0.12–0.31) and having a tertiary education (RR: 0.27; 95% CI: 0.08–0.92) were associated with reduced risk of AUD. *Correspondence: Benedict Weobong bweobong@ug.edu.gh Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 2 of 12 Conclusions Our study found a relatively high prevalence of probable MNS conditions, and very low detection and treatment rates in rural primary care settings in Ghana. There is a need to improve the capacity of primary care health workers to detect and manage MNS conditions. Keywords Mental health conditions, Detection rates, Prevalence, Primary healthcare Background capacity for mental health services in LMICs through An estimated 10% of Ghanaians have common men- the introduction of the Mental Health Gap Action Pro- tal health conditions, while 1–3% have severe mental gramme (mhGAP) [6, 19]. The mhGAP provides key health conditions such as schizophrenia [1, 2]. Yet only strategies such as integrated management plans and 2% of such persons will receive treatment [3]. Epilepsy evidence-based guidelines for scaling up mental health for instance is a common neurological disorder especially services in LMICs for priority mental health conditions in poor areas of the world, the burden of it in low and- including depression, psychosis, alcohol use disorder, and middle-income countries LMICs is more than twice that epilepsy [20]. In the light of these global efforts at improv- found in high-income countries [4]. In Ghana epilepsy is ing mental health services in LMICs, Ghana Somubi the most common treated disorder among adolescents Dwumadie (Ghana Participation Programme), a four- receiving services in four health facilities in the two dis- year disability programme, is supporting the generation tricts in the Bono East Region [5]. of evidence to inform the scale-up of quality integrated As in many other low and middle-income countries mental healthcare in primary care in Ghana. Part of the (LMICs), few data are available in Ghana to inform the programme is to support selected districts to develop and scale up of mental health services and reduce the large implement mental healthcare plans including the capac- treatment gap [6, 7]. There is little epidemiological data ity development of primary healthcare workers through on persons living with mental health conditions who mhGAP training. The implementation and evaluation of attend primary care settings in Ghana, and to the best of mental health programmes in real-world primary care our knowledge, none in the northern and central parts of and community settings is critical to identify factors that the country [8]. In LMICs, lack of reliable data on mental contribute to the effective scale-up of mental health ser- health systems affects mental healthcare delivery efforts vices [21]. However, very few high-quality evaluations as more than one-quarter of these countries have no sys- have been carried out in LMICs [7, 22], where scaling up tem for reporting basic mental health information [9]. mental health services is of great importance. Evaluation In addition, empirical research is seldom used to guide studies done in LMICs have not assessed important base- policy development and implementation, especially in line factors such as case detection within routine primary LMICs [10, 11]. Out of 136 global studies of scaling up healthcare facilities [21]. mental health programmes according to a systematic Ghana Somubi Dwumadie, together with the Ghana review, only 15 were carried out in LMICs [12]. Health Service and the Mental Health Authority in It is important to conduct studies of prevalence and Ghana is mindful of these limitations and is support- needs for care in primary care settings in remote rural ing an approach that can inform the systematic scale-up areas of Ghana for two main reasons: first to assess the of integrated mental health services. This is being done need; and second to provide baseline data to assess the through demonstration sites in three districts of Ghana. impact of proposed programmes to integrate mental The goal of this paper is to report on prevalence and health into primary care and narrow the treatment gap. mental health case detection in three districts in three In Ghana, notwithstanding the existence of the Mental regions in order to provide a baseline estimate for assess- Health Act (Act 846) since its passing in 2012 [13, 14], ing the impact of subsequent district mental healthcare the infrastructure and public services have not been plans. properly developed, including mental healthcare services to align with the population growth [15, 16]. For example Methods whilst integration of mental health services in primary Study setting healthcare is a globally accepted approach to optimising The study was conducted in 15 primary healthcare facili- healthcare in view of its capacity to meet multiple health ties across three districts purposively selected (out of 216 and social needs from a single platform of care [17], men- districts in Ghana) as demonstration sites for implemen- tal health services are not integrated in primary health- tation of district mental healthcare plans. These plans care in Ghana due to lack of resources and prioritisation include detailed implementation strategies to achieve [13, 18]. integrated mental healthcare in primary healthcare set- The WHO, in an effort to reduce this mental health tings, as part of Ghana Somubi Dwumadie [23]. These treatment gap, has made significant efforts to build districts included Anloga (population: 99,996) in the Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 3 of 12 Volta Region (Southern belt), Asunafo North (popula- Data were collected on five priority mental, neurologi- tion: 157,732) in the Ahafo Region (Middle belt) and cal and substance use (MNS) conditions of interest in Bongo (population: 105,206) in the Upper East Region the mhGAP-IG namely: depression, psychosis, suicidal (Savannah belt) (Fig. 1: Map of Ghana showing the three behaviour, epilepsy and alcohol use disorders (AUD) to demonstration districts). The selection of the ‘demonstra- assess the prevalence and detection of MNS conditions tion’ districts was guided by a set of criteria outlined in at the routine primary healthcare facilities. Each data col- Ghana Somubi Dwumadie’s framework for implement- lector was assigned to a health facility for the daily data ing district mental healthcare plans, ratified in 2020 in collection for the period used for the data collection. a high-level stakeholder engagement meeting in Accra, Eligible participants were resident adult (> 17 years) Ghana [24]. The criteria for selection included: geo- out-patients seeking healthcare at primary care facilities political equity (each of the districts with health facilities in each of the three demonstration districts. All consecu- should spread across the three zones of Ghana: South- tive out-patients exiting their clinical consultations were ern, Middle Belt, and Savannah); representativeness approached and those who consented to participate were (i.e. the districts should not be over-resourced, but also recruited. Nine trained and experienced data collectors not under-resourced, particularly in relation to human administered a two-part survey adapted from the Pro- resources so that lessons could be generalised to inform gramme for Improving Mental health care (PRIME) in scale up in other districts); an appropriate level of prevail- five LMICs [21]: first, a battery of five screening mea- ing mental health activity (i.e. there should be no existing sures to elicit responses on symptoms and help-seeking sites for mental health research or previous/ or on-going behaviour for depression, psychosis, suicidal behaviours, district mental health plans or national academic cen- epilepsy, and AUD; and second, a quantitative/qualita- tres for mental health should exist as this would create tive exit interview to collect data on the patient’s expe- an unrealistic environment which is not representative rience regarding the clinical consultation. Each interview of other districts); and willingness of district leadership session lasted between 30 and 40  min and we did not to engage. Two key stakeholders that were involved in encounter patients that refused consent to participate in selecting the districts included Ghana Health Service the study. This section of the form also included extrac- (GHS) and the Mental Health Authority (MHA). tion of data on clinical notes from individual patient The 15 health facilities are presented on Table 1. These folders. included 10 health centres, 2 hospitals, 2 Community Health Planning and Services and 1 clinic. Measures Following the selection of districts, primary healthcare The measures included items on participant socio- facilities were identified for the baseline facility survey. demographic identifiers and a battery of instruments Within each district five health facilities were randomly including: the Patient Health Questionnaire (PHQ-9), selected from a pool of 6–10 health facilities with high Psychosis screener, Suicidal Ideation screener, Epilepsy OPD attendance, including the availability of physician Screener, and Alcohol Use Disorders Identification Test. assistants, mental health nurses or midwives. On the Some of the measures (suicidal ideation screener, epi- average these health facilities had a minimum of 48 OPD lepsy screener) were from the PRIME study. The other daily attendance visits. measures have been previously used in Ghana [25, 26]. Training on the English version of the measures was Study design and research participants conducted by BW, AK, and LS. Experienced bi-lingual We employed a cross-sectional health facility detec- (English and the predominant local language in the study tion survey design. The baseline survey (the focus of this districts) data collectors were trained in a 3-day work- paper) was conducted from November 2021 to Decem- shop on the content and administration of all study ques- ber 2021, prior to the implementation of the district tionnaires, including the exit interview. This involved mental healthcare plans and mhGAP in each demonstra- forward translation to the local languages (Gurene, Twi/ tion district. We estimated a minimum sample size of 240 Bono, and Ewe), and back translation, particularly paying participants in each district to detect a 15% increase in attention to key constructs in the backtranslation. Con- case detection at 80% power, 5% alpha and a design effect sensus on the translation of each item was obtained. The of 2. This is based on the test of our primary hypothesis data collectors were trained on how to administer the that there will be no difference in case detection and ini- study tools. They were first introduced to mental health tiation of treatment (defined as proportion of patients and the various conditions especially the priority MNS with the target MNS conditions receiving diagnosis and disorders. The trainees systematically went through all initiation of treatment) between baseline and endline the questions in each study tool and clarified all concerns (after primary health care worker training). regarding how the questions should be asked and the expected responses. Interviewing skills was treated and a Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 4 of 12 Fig. 1 Map of Ghana showing the three demonstration districts Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 5 of 12 Table 1 List of Health Facilities in 3 demonstration districts, 2019–2021 District Facility name OPD Attendance Facility Type Ownership 2019 2020 2021 Anloga Tegbi 12,027 11,011 NA Health Centre Public Dziedzorve 3438 3140 NA CHPS Public Kotoka Memorial 2446 1392 NA Health Centre Public Tregui 1361 951 NA Health Centre Public Anloga 10,035 9955 NA Health Centre Public Bongo Namoo 9287 5504 4479 Health Centre Public Dua 5775 4219 2926 Health Centre Public Kodorogo 977 853 345 Health Centre Public Bongo 23,102 21,481 18,002 Hospital Public Wagliga 4133 2 814 2240 CHPS Public Asunafo North Akrodie NA 30,958 18,274 Health Centre Public Mim NA 19,512 11,194 Health Centre Public Kasapine NA 4486 3158 Clinic Public Asumura NA 1780 4325 Health Centre Public Goaso NA 55,446 44,010 Hospital Public *NA: information on OPD attendance not available *CHPS: Community Health Planning and Services session on the translation of key constructs from English sensitivity and specificity of 0.88 at a cut-off of 10 [29], to the local languages. Both theory and practical sessions and high positive predictive value [30]. The PHQ-9 has were employed. Trainees performed role plays for the been previously validated in Ghana and showed superior facilitators to assess their competencies and understating psychometric properties when compared with the Edin- of key constructs. There were also group work and mock burgh Postnatal Depression Scale [25, 26]; it recorded a exercises involving the translation and administering the sensitivity of 94% and specificity of 75% at a cut-off score questionnaires in the various local languages and docu- of 5. The measure has subsequently been used in other mentation of key words in the local languages. studies [26, 31]. Probable depression ‘diagnosis’ (major or During the data collection, all study participants pro- minor) is defined in this study as having occurred when a vided written consent before completing the question- total score of at least 5 is indicated, and at least two car- naire in the local language of their choice (Ewe, Gurene, dinal symptoms of depression are reported as present and Twi), mostly based on the geographical location of for at least most of the time in the last two weeks; these the health facility although some interviews were also must include depression or anhedonia (loss of interest or conducted in English. The trained data collectors admin- pleasure). In this study, the PHQ-9 recorded a Cronbach’s istered the structured questionnaires in the participant’s reliability coefficient of 0.83, indicating an acceptable primary language, using an electronic mobile device or level of reliability. tablet. All the questionnaires were administered by the data collectors. For the purposes of this study, these mea- Psychosis screener sures were used for diagnosis of the priority MNS con- The psychosis screener used in this study (adapted from ditions, and there was no further clinical assessment to the PRIME study) is a structured 10-item questionnaire validate the diagnosis of these screening questionnaires. that enquires after symptoms of psychosis, which require The screening tools used for the data collection are reli- a yes or no answer. Only items 1–6 are scored, and these able and valid instruments that have been used widely to include items such as strange feelings, having special screen for priority MNS conditions [27, 28]. Each mea- powers, felt other people are too interested in you, felt sure is described in detail below. thoughts were directly interfered with, ever heard voices or participant had ever been prescribed anti-psychotic Patient health questionnaire (PHQ-9) medicine. The cut-off point applied for recording a per- The PHQ-9 is a structured questionnaire that enquires son as screen positive for ‘probable psychotic symp- after the nine symptom-based criteria for a diagnosis of toms’ was endorsing at least 2 items on the screener. In DSM-IV depression, including their duration and sever- this study, this instrument recorded a Cronbach’s reli- ity [29]. Each item is scored on a scale of 0 to 3 and gen- ability coefficient of 0.7, which is an acceptable level of erates a continuously distributed total score ranging reliability. from 0 to 27. In its initial review the PHQ-9 recorded Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 6 of 12 Suicidal ideation and action measure study a probable ‘diagnosis’ of any AUD was defined as a The suicidal ideation and action measure is a struc- total score of 8 and above. tured a 9-item questionnaire (adapted from PRIME) that enquires after signs and symptoms of suicidal thought Consultation experience and folder data extraction and behaviour, including help-seeking. For this study we This section of the survey included both quantitative and defined probable suicidal behaviour broadly as compris- qualitative questions on details of the consultation, and ing of three domains (thoughts, plan, and attempt). Three was used to ascertain detection of the priority MNS. The items covering these domains that provide a ‘yes’ or ‘no’ first part which was fully structured, extracted data from answer were used; any ‘yes’ response was considered a clinical folders of patients. The information included type screen positive for probable suicidal behaviour. In this of clinician seen, symptoms presented, diagnosis and study, using only three items it recorded a Cronbach’s treatment prescribed including any referrals made by the reliability coefficient of 0.4, indicating a low reliability. clinician. The research team engaged the services of pharmacists Epilepsy screener at the district hospitals to support the data collectors The epilepsy screener is a structured 10-item measure with extraction of data on whether or not a diagnosis was (adapted from PRIME) with two sub-scales on symp- given for each of the priority MNS conditions (except sui- toms and severity. The epilepsy screener seeks to estab- cidal behaviour as this was assessed separately for those lish if the patient has ever had a fit, or epilepsy. It assesses patients at risk and referral made where necessary). The symptoms such as falling to the ground with loss of support from the pharmacists was instructive in ensuring consciousness without reason and experiencing twitch- the diagnosis on patients folders are correctly captured, ing, shaking of the arms of legs without control, wetting where the information was not clear to the data collec- yourself or biting of the tongue. It further seeks to con- tors. The second part was a self-report administered to firm if a clinician had ever confirmed an index person as the patients on whether or not the health worker made having epilepsy, last time index person experienced epi- a diagnosis or mention if the patient had a health prob- leptic attack and the number of seizures recorded in the lem (e.g. Did the [health worker] give you a diagnosis of last 30 days. The symptom sub-scale elicits ‘yes’ or ‘no’ depression?). Given this was a self-report, only the folder- responses. For this study, a probable ‘diagnosis’ of epi- extracted data was used for this analysis. Other areas lepsy entailed a ‘yes’ response to any of the 6 questions assessed included duration of clinical consultation, type on the symptom sub-scale. In this study, it recorded a of clinician seen, medication given and whether adequate Cronbach’s reliability coefficient of 0.5, indicating a low advice was provided on how to take the medications. reliability. Participants that needed referral based on the screening were referred to the mental health unit for help. Alcohol Use Disorders Identification Test (AUDIT) The AUDIT is a 10-item screening tool developed by the Data analysis WHO for the detection of hazardous and harmful alcohol Both descriptive and inferential analyses were con- use including alcohol dependence in primary healthcare ducted by the investigators. Descriptive analyses were settings [32]. Each item is scored on a scale of 0 to 4 and performed to estimate the prevalence of the probable generates a continuously distributed total score ranging diagnoses of the priority MNS conditions (assessed by from 0 to 40. In the initial validation involving 6 coun- the survey team) and the proportion detected by health tries, scores of 8–15, 16–19, and 20 or more, represented care workers at the health facilities. Prevalence estimates probable diagnosis of hazardous use, harmful use, and were computed and reported with 95% CI. To estimate alcohol dependence respectively [33–35]. The initial vali- detection proportions, positive cases (i.e. met criteria dation generated sensitivities around 0.90 and specifici- for depression, psychosis, suicidal behaviour, epilepsy or ties averaging 0.80 [36]. In a recent review of the AUDIT AUD) screened by the survey team were compared with in LMICs, lower cut-off scores were observed [37]: haz- the health worker diagnosis. A priority MNS was classi- ardous drinking cut-off scores of > 3 or > 5 yielded sen- fied as ‘missed detection’ if a solicited diagnosis was not sitivities ranging from 93.5 to 96.2%, and specificities of provided in the patient folder by the health worker but 63.3–91.5%; harmful drinking cut-off score of > 7 or > 8 was screened positive for any of the priority MNS con- yielded 90.0% sensitivity, 86.2% specificity; dependent ditions by the independent survey team using the study drinking cut-off scores of > 7 to > 24, yielded sensitivity screening tools. Following this, inferential analyses were 63.6% and specificity of 75%. In this study, it recorded a conducted to ascertain patient-level correlates of ‘missed Cronbach’s reliability coefficient of 0.86, which confirms detection’ and prevalence of the priority MNS condi- the screening tool to have acceptable reliability. For this tions. For the analysis on correlates of ‘missed detec- tion’, Fisher’s exact chi-square test (opted for because of Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 7 of 12 insufficient cell observations (< 5)) was used to deter- Association between patient-level factors and missed mine if there was a significant association between case detection of MNS disorders was assessed. The Fisher’s detection (except suicidal behaviour as health worker exact test shows no statistically significant association data was not collected) and each measured patient level between patient level factors and case detection. factor (age, sex, marital status, education, employment and district of residence); p-values were reported with Correlates of MNS conditions in the 3 demonstration statistically significant level determined at 0.05. For the districts analysis of correlates of the prevalence outcomes, logis- Table 2 shows patient-level correlates of the priority MNS tic regression was used to examine associations first conditions. For probable depression, independent asso- between potential determinants (patient level factors) ciations of increased risk were noted only with attending and the prevalence of each of the priority MNS condi- a health facility in the Anloga district. We noted that the tions. Second, associations between other priority MNS data collectors from the Anloga and Asunafo North dis- conditions and prevalence of depression (as the outcome) tricts included trained mental health professionals and were examined. Effect sizes are reported as crude and may have accounted for the higher case detection in that adjusted relative risks (aRR) estimated using the marginal district. For probable psychotic symptoms, independent standardisation technique with 95% confidence intervals associations of reduced risk were noted for being married estimated via the delta method [38]. Analyses were con- and having a tertiary level education; all other associa- ducted using Stata 14 [39], charts were generated using tions were not statistically significant. Independent asso- Microsoft Excel. ciations of increased risk of probable suicidal behaviour were noted for those attending a health facility in Asu- Results nafo and Anloga districts (compared to the Bongo dis- Socio-demographic characteristics of study population trict). For probable epilepsy, being in the age group of 35 A total of 909 people (Bongo district: 301, Asunafo North to 44 years was associated with reduced risk of epilepsy; district: 308, and Anloga district: 300) attending primary all other associations were not deemed statistically signif- health facilities within the period (November – Decem- icant. Many more factors were associated with probable ber 2021) participated in the study of which approxi- alcohol use disorder, for instance, being more than 35 mately 82% (741/909) were women, 30% (273/909) aged years and those attending a health facility in the Anloga 18–24 (Table  2). Approximately 65% (587/909) of the district had an increased risk of alcohol use disorder. participants were married, 29% (265/909) had no formal Additionally, being a woman and having a tertiary educa- education, while 64% (583/909) had ever been employed. tion were independently associated with reduced risk of alcohol use disorder. All other associations were not sta- Prevalence and detection of MNS conditions in the 3 tistically significant. demonstration districts The study also assessed possible associations between This study assessed active probable cases of MNS. The the other MNS conditions and depression. Independent overall prevalence of any probable depression and psy- associations of increased risk were observed for having chotic symptoms among the study participants was self-reported psychotic symptoms and an alcohol use dis- 15.6% (95% CI: 13.3 – 18.1%) and 12.0% (95% CI: 10.0 order (RR: 1.68; 95% CI: 1.12–2.54). − 14.3%) respectively. Prevalence of depression in men is marginally higher compared with women in the whole Discussion sample. The prevalence of probable suicidal behaviour This study set out to assess prevalence and case detec- and epilepsy was 11.8% (95% CI: 9.7 − 14.0%) and 13.1% tion of priority MNS conditions at primary healthcare (95% 10.9 − 15.5%) respectively. Probable AUD among facilities in three rural settings in Ghana. Specifically we the study participants was 7.8% (95% CI: 6.2 − 9.8%). sought to estimate primary healthcare workers’ ability to Fig. 2 shows the proportion of patients with the prior- detect depression, psychotic symptoms, suicidal behav- ity MNS conditions (except suicide behaviour) that were iour, epilepsy, and alcohol use disorder, in comparison to missed by the health care workers across the three dis- detection by an independent team of researchers using tricts. Almost all probable cases identified by the survey screening tools for these conditions. team were missed (missed detection) by the health care Prevalence of priority MNS. workers with missed detection proportions ranging from An overall prevalence of probable depression among 94.4–99.2% across each priority MNS condition. Missed the study participants of 15.6% is lower than previous detection for people with AUD even though very high studies in Ghana that reported depression prevalence in (94.4%), was better compare with depression, psychosis the range of 25–62% [40, 41]. We observed a marginally and epilepsy. higher prevalence of depression in men than in women in the whole sample. Unlike previous studies, which show Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 8 of 12 Table 2 Baseline socio-demographic correlates of priority MNS among study participants across the demonstration sites Correlates Participants Depression Psychosis Suicidal Behaviour Epilepsy AUD n (%) of n (%) with RR1 (95% CI) n (%) with RR1 (95% CI) n (%) with RR1 (95% CI) n (%) RR1 (95% CI) n (%) RR1 (95% CI) participants depression psychosis suicidal with with AUD (N = 909) (N = 142) (N = 109) behaviour epilepsy (N = 71) (N = 107) (N = 119) Age Group 18–24 273(30.03) 37 (13.55) 1 36(13.19) 1 35(12.82) 1 43(15.75) 1 9(3.30) 1 25–34 250(27.50) 35(14.0) 1.010.62–1.63) 27(10.80) 0.67(0.41–1.11) 24(9.60) 0.77(0.45–1.32) 26(10.4) 0.68(0.41–1.12) 11(4.40) 1.17(0.46–2.99) 35–44 136(14.96) 28(20.59) 1.62(0.95–2.78) 17(12.50) 0.88(0.49–1.61) 18(13.24) 1.01(0.53–1.93) 10(7.35) 0.43(0.20–0.90) 16(11.76) 3.06(1.14–8.24) 45–54 129(14.19) 20(15.50) 1.36 (0.74–2.50) 11(8.50) 0.62(0.30–1.28) 16(12.4) 0.87(0.43–1.76) 16(12.4) 0.63(0.32–1.24) 16(12.4) 3.10(1.13–8.52) >/=55 121(13.31) 22(18.18) 1.54(0.83–2.86) 18(14.80) 0.82(0.42–1.59) 14(11.57) 0.71(0.33–1.52) 24(19.83) 1.08(0.58–2.02) 19(15.7) 3.00(1.07– 8.43) Sex Male 168(18.48) 27 (16.07) 1 18 (10.71) 1 16(9.52) 1 27(16.07) 1 36(21.4) 1 Female 741(81.52) 115 (15.52) 1.01 (0.67–1.53) 91 (12.28) 0.98(0.62–1.57) 91(12.28) 1.15(0.68–1.93) 92(12.42) 0.82(0.54–1.24) 35(4.70) 0.19(0.12– 0.31) Marital Status Single 164(18.04) 24 (14.63) 1 28(17.07) 1 23(14.02) 1 27(16.46) 1 9(5.49) 1 Married 587(64.58) 94(16.01) 0.94(0.59–1.51) 61(10.39) 0.62(0.39–0.98) 62(10.56) 0.79(0.46–1.36) 68(11.58) 0.90(0.54–1.50) 43(7.33) 1.08(0.48–2.42) Living together 49(5.39) 7 (14.29) 1.27(0.59–2.71) 3(6.12) 0.60(0.21–1.72) 5(10.20) 0.68(0.26–1.76) 7(14.29) 0.61(0.26–1.41) 4(8.16) 3.59(1.42–9.07) Widow/separated 109(11.99) 17 (15.60) 0.78(0.38-1. 60) 17(15.6) 0.85(0.43–1.69) 17(15.6) 1.15(0.54–2.45) 17(15.6) 1.10(0.52–2.31) 15(13.76) 1.43(0.54–3.79) Education Status None 265(29.15) 45(16.98) 1 31(11.7) 1 32(12.08) 1 30(11.32) 1 28(10.57) 1 Primary 169(18.59) 23(13.61) 0.84(0.52–1.36) 21(12.4) 0.81(0.49–1.33) 18(10.65) 0.75(0.43–1.31) 18(10.65) 0.91(0.52–1.58) 14(8.28) 0.75(0.42–1.35) Middle/JHS 262(28.82) 42(16.03) 1.05(0.69–1.61) 29(11.07) 0.78(0.48–1.27) 35(13.36) 0.96(0.59–1.57) 40(15.27) 1.23(0.77–1.98) 14(5.34) 0.62(0.34–1.16) Technical/SHS 154(16.94) 21(13.64) 0.95(0.55–1.63) 27(17.53) 1.10(0.65–1.86) 20(12.99) 0.87(0.47–1.60) 25(16.23) 1.20(0.67–2.16) 12(7.79) 0.95(0.47–1.91) Tertiary 59(6.49) 11(18.64) 1.17(0.62–2.22) 1 (1.69) 0.12(0.02–0.84) 2(3.39) 0.29(0.07–1.18) 6(10.17) 0.95(0.41–2.22) 3 (5.08) 0.27(0.08– 0.92) Employment Ever employed 583(64.14) 48(14.72) 1 76(13.04) 1 69(11.84) 1 76(13.04) 1 54(9.26) 1 Never employed 326(35.86) 94 (16.12) 0.99(0.68–1.45) 33(10.12) 0.67(0.43–1.06) 38(11.66) 1.05(0.66–1.65) 43(13.19) 0.95 (0.61–1.46) 17(5.21) 1.03(0.58–1.84) District Bongo 301 (33.11) 45(14.95) 1 10(3.32) 1 17(5.65) 1 29(9.63) 1 15(4.98) 1 Asunafo North 300(33.00) 36(12.00) 0.72(0.46–1.13) 16(5.33) 1.48(0.66–3.32) 36(12.00) 2.31(1.27–4.19) 62(20.67) 2.26(1.44–3.56) 18(6.00) 0.91(0.44–1.88) Anloga 308 (33.88) 61(19.81) 1.29(0.89–1.87) 83(26.95) 7.60(3.96– 54(17.53) 3.32(1.93–5.71) 28(9.09) 0.96(0.57–1.60) 38(12.34) 2.69(1.49– 14.58) 4.88) R1: adjusted for each socio-demographic variable in univariable analysis. Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 9 of 12 rate of people with MNS conditions in the routine pri- mary healthcare facilities. The findings align with the PRIME study which reported low health worker detec- tion of priority MNS ranging from 0 to 11.7% in five LMICs [52]. The low detection rate could be a func- tion of the cadre of healthcare workers who attended to the routine primary healthcare attendees and who had received minimal training in mental health. In our study, participants would have been seen by either nurses/mid- wives who comprised of 51% of healthcare workers and/ or physician assistants (48%). Participants were not seen by doctors or mental health professionals such as psy- chiatrists or the four main types of mental health profes- sionals (community mental health officers, community Fig. 2 Proportion of cases of MNSconditions not detected at primary psychiatric nurses, registered mental health nurses, and health care facilities across districts clinical psychiatric officers) within the primary health- care system in Ghana [53]. Our findings highlight missed a higher prevalence of depression in females, there was opportunities for mental healthcare in these primary care no significant difference in this study. The prevalence settings, and the need to provide training and strengthen of epilepsy in this study is similar to a previous study health systems to improve detection and care for people involving five sub-Saharan Africa countries including living with MNS conditions. Ghana, Kenya, South Africa, Tanzania and Uganda which Correlates of MNS conditions. reported prevalence of active epilepsy in the range of Our study found some associations between patient- 7.8–14.8% [42, 43]. The prevalence of self-reported psy- level factors and the MNS conditions examined. Being chotic symptoms in the current study is similar to pre- married and having a tertiary level education appear to vious studies conducted in Ghana that reported that the be protective against developing psychosis symptoms. prevalence of positive psychotic like experiences range These findings are similar to what has been reported from 3.8 to 41.2% [44, 45]. In other sub-Saharan African from a systematic review of common mental disorders countries, a prevalence of self-reported psychotic symp- and poverty (including education and employment sta- toms of 3.9% and 13.9% have been reported in Tanzania tus) in LMICs which indicated that lower education sta- [46] and Kenya [47] respectively. We note however that tus was positively associated with increased prevalence there are substantial methodological challenges with of common mental disorders [54]. Although some stud- screening for psychosis using self-report screeners [48]. ies have reported that increasing age is associated with It is also possible that the self-report of psychotic-like depression [55], Deribew et al. did not find significant experiences may be transient and benign, and further association between age and depression [56]. research would be required to ascertain specific needs for Being a woman and having a tertiary level education care. were associated with reduced risk of alcohol use disor- The prevalence of probable alcohol use disorder of der. These findings compare with a previous study in 7.8% in the current study is comparable with a previous Ghana among adults aged 15–65 years which reported study in Ghana among students that found alcohol use that women and those with tertiary education were less to be 6.8%, 11.1% and 12.6% for alcohol problem, lifetime likely to engage in dependent drinking [57]. Our study drunkenness, and current alcohol use respectively [49], found an association between alcohol use disorder and although a higher prevalence of 43% among the youth has depression, in keeping with previous studies which have been reported with more men using alcohol than women reported a positive association between alcohol use dis- [50]. We surmise that the low prevalence of probable order and mental illness [58]. Our study also found asso- AUD reported in the current study is partially explained ciations between self-reported psychotic symptoms and by the small number of men in this sample. Also, the probable depression, which may be reflective of comor- prevalence of suicidal behaviour (11.8%) in the current bidity between these conditions. This finding is consis- study is comparable with a prevalence of between 5.0 tent with previous studies that have reported psychiatric and 14.8% [51] reported in a study involving five LMICs comorbidities among patients with schizophrenia includ- (Ethiopia, Uganda, South Africa, India and Nepal). ing depression and anxiety disorders [40, 59], with Detection of priority MNS conditions. depression co-occurring in persons within first episode Our study reports a high rate of missed detection for psychosis [59]. MNS conditions of about 98%, indicating a low detection Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 10 of 12 Our study found an increased risk of the MNS con- all at all ages [69]. It is imperative therefore, for central ditions in some of the study districts (Asunafo North, government and other donor organisations to provide the Anloga) but not others (Bongo). This is an interesting needed resources to support the integration and expan- finding with theoretical backing in terms of the exposure sion of mental health services in the districts. to place of residence and mental health outcomes. For Strengths and limitations of the study. example, the neighbourhood disorder model, the envi- A strength of this study was the adaptation of meth- ronmental stress model [60] and the social stress model ods from previous research conducted in LMIC in the [61], could explain our findings but this requires further PRIME consortium [7, 70]. Nevertheless, there are some research to explore area level factors that may explain limitations. First, as this is a cross-sectional study it is not these differences. possible to infer temporal or causal relationships. Second, Implications of the survey. given not many factors were found to be associated with The findings of the current study underscore findings the prevalence of the MNS conditions examined in this from other studies in Ghana and sub-Saharan Africa study, other factors that our study did not measure could [62] indicating a large treatment gap for MNS condi- account for this. Third, the data collection was limited to tions in primary healthcare settings. The findings also fifteen health facilities within three districts across three demonstrate the need to strengthen the capacity of pri- regions of Ghana, we cannot claim that the demographic mary healthcare facilities to routinely detect and provide characteristics of these districts represents Ghana care for patients with MNS conditions. The prevalence nationally. Therefore the prevalence reported may not reported in the current study may be associated with be representative of the general population and should other factors that our study did not measure, such as the be interpreted in the context of these primary health- presence of other chronic diseases [63]. care settings. Fourth, the limitations of using screening Our findings compare with previous studies that report instruments for the survey need to be acknowledged. In high rates of common mental health conditions in many particular, we did not conduct any clinical diagnosis by a resource-constrained settings, indicating the need for mental health professional, limiting the conclusions that improved detection of patients attending primary health- can be drawn regarding the prevalence of the MNS disor- care facilities [62, 64, 65]. Studies have demonstrated that ders that are reported in this study. Fifth, some of the data developing the capacity of primary healthcare workers is collectors were non-mental health professionals, even a promising strategy to increasing access to mental health though they were trained on how to screen for probable services and thereby addressing the missed detection rate mental health conditions. Finally, one of the data collec- of MNS conditions [66]. tors from the Anloga district is a professionally trained Due to the absence of specialist mental health work- mental health nurse, and this may have accounted for the ers and the limited mental health service provision in higher prevalence reported in that district. Ghana [67], task-sharing may be a good strategy to bol- ster the insufficient health workforce in an attempt to Conclusions address the treatment gap and improve access to mental This study found a substantial unmet need for care health service provision [68]. Nevertheless, Agyapong et among people living with MNS conditions, attending pri- al. contended that task-sharing may not be practicable in mary healthcare facilities in the Bongo, Asunafo North the long run given that primary healthcare workers are and Anloga districts in Ghana. Despite a relatively high already over-burdened and additional responsibilities prevalence of probable MNS conditions, about 98% of may compromise on quality health service provision [53]. participants who screened positive for probable depres- Strategies to overcome this barrier may include provid- sion, psychotic symptoms, epilepsy, suicidal behaviour ing ongoing training, supervision, and support; strength- or alcohol use disorder, were not detected by healthcare ening of referral pathways; improving supply of essential workers at the routine primary healthcare facilities. The psychotropic medications; and inclusion of mental health findings demonstrate the need for improved detection indicators in routine District Health Information Man- and care for people living with MNS conditions in these agement Systems (DHIMS). Another strategy to increase districts. A proposed future study will evaluate changes access to mental health services is the WHO-recom- in detection rates following implementation of district mended interventions such as mhGAP-IG [6, 19] to mental healthcare plans in these districts. develop the capacity of primary healthcare workers to detect mental conditions in routine primary health- AbbreviationsAUD Alcohol used Disorders care facilities. This will help to bridge the treatment gap AUDIT Alcohol Use Disorders Identification Test to achieve the universal health coverage agenda by the GHS Ghana Health Service United Nations Sustainable Development Goal (SDG) 3 GHS-ERC Ghana Health Service Ethics Review CommitteeMHA Mental Health Authority of ensuring healthy lives and promoting well-being for GMHA Ghana Mental Health Act 846 Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 11 of 12 MNS Mental, Neurological and Substance used Disorders References PHQ-9 Patient Health Questionnaire version 9 1. Read UM, Doku V. Mental health research in Ghana: a literature review. Ghana SDGs Sustainable Development Goals Med J. 2012;46(2):29–38. WHO World Health Organisation 2. Addo R, Nonvignon J, Aikins M. Household costs of mental health care in Ghana. J Ment Health Policy Econ. 2013;16(4):151–9. Acknowledgements 3. Eaton J, Ohene S, editors., editors. Providing Sustainable Mental Health Care The current study is a product of Ghana Somubi Dwumadie (Ghana in Ghana: A Demonstration Project2016. Participation Programme. The authors thank Alice Tilton (Options) and Adam 4. Newton CR, Garcia HH. Epilepsy in poor regions of the world. The Lancet. Dokurugu Yahaya (Basic Needs Ghana) for reviewing and providing comments 2012;380(9848):1193–201. used for finalising this paper. The authors appreciate all participants for their 5. Formentos A, Ae-Ngibise KA, Nyame S, Asante KP. Situational analysis of role in contributing data for this study. service provision for adolescents with mental and neurological disorders in in two districts of Ghana. Int J Mental Health Syst. 2021;15(1):1–21. Authors’ contributions 6. Keynejad RC, Dua T, Barbui C, Thornicroft G. WHO Mental Health Gap Conceptualisation of study: KA, BW, CL. Data acquisition: KA, BW, LS. Data Action Programme (mhGAP) intervention guide: a systematic review of analysis: BW, KA. Data interpretation: KA, BW, CL. Drafting: KA. Critical revision: evidence from low and middle-income countries. Evid Based Ment Health. KA, BW, LS, LAK, CL. All authors read and approved the final manuscript. 2018;21(1):30–4. 7. Lund C, Tomlinson M, De Silva M, Fekadu A, Shidhaye R, Jordans M, et al. Funding PRIME: a programme to reduce the treatment gap for mental disorders in five Funding for the study was provided by the UK Aid (Foreign and low-and middle-income countries. PLoS Med. 2012;9(12):e1001359. Commonwealth Development Office) from the UK government through 8. Weobong B, Ae-Ngibise K, Mwangi G, Sakyi L, Lund C. Mental health and Ghana Somubi Dwumadie (Ghana Participation Programme). However, the disability research priorities and capacity needs in Ghana: findings from views conveyed in this paper do not necessarily reflect the UK Government’s a rapid review and research priority ranking survey. Global Health Action. official policy position. The funder had no role in the study design, data 2022;15(1):2112404. collection and analysis, decision to publish, or preparation of the manuscript. 9. Upadhaya N, Jordans MJ, Abdulmalik J, Ahuja S, Alem A, Hanlon C, et al. Infor- mation systems for mental health in six low and middle income countries: Data Availability cross country situation analysis. Int J Mental Health Syst. 2016;10:1–11. The dataset used for the current study is available from the corresponding 10. Benbow S, Forchuk C, Gorlick C, Berman H, Ward-Griffin C. Social exclusion author on reasonable request. and health: The development of nursing knowledge.Canadian Journal of Nursing Research Archive. 2015:56–72. 11. Wainberg ML, Scorza P, Shultz JM, Helpman L, Mootz JJ, Johnson KA, et al. Declarations Challenges and opportunities in global mental health: a research-to-practice perspective. Curr psychiatry Rep. 2017;19(5):1–10. Ethics approval and consent to participate 12. De Silva MJ, Lee L, Fuhr DC, Rathod S, Chisholm D, Schellenberg J, et al. Ethical approval was granted by the following ethics committees: King’s Estimating the coverage of mental health programmes: a systematic review. College London Research Ethics Committee (approval number; HR/DP-21/22- Int J Epidemiol. 2014;43(2):341–53. 26013), and the Ghana Health Service Ethics Review Committee (approval 13. Doku V, Wusu-Takyi A, Awakame J. Implementing the Mental Health Act in number; GHS-ERC:006/08/21). The team also followed the Ghana Health Ghana: any challenges ahead? Ghana Med J. 2012;46(4):241. Service-provided guidelines to investigators which must be complied with to 14. Walker GH. Ghana mental health act 846 2012: a qualitative study of the chal- mitigate the effect of COVID-19 [71]. Written informed consent was obtained lenges and priorities for implementation. Ghana Med J. 2015;49(4):266–74. from all the participants and their legal guardians. Literate participants and 15. Mfoafo-M’Carthy M, Sossou M-A. Stigma, discrimination, and social exclu- non-literate participants consented by signed or by thumbprinting the sion of the mentally ill: the case of Ghana. J Hum Rights Social Work. consent form. All the methods are followed by relevant regulations and 2017;2(4):128–33. guidelines. 16. Adu-Gyamfi S. Mental health service in Ghana: a review of the case. Int J Public Health Sci. 2017;6(4):299–313. Consent for publication 17. Porter ME, Lee TH. Why health care is stuck—and how to fix it. Harvard Busi- Not applicable. ness Rev. 2013;91:50–70. 18. Walker GH, Osei A. Mental health law in Ghana. BJPsych Int. 2017;14(2):38–9. Competing interests 19. World Health Organization. mhGAP intervention guide for mental, neurologi- The authors declared to have no conflicts of interest. cal and substance use disorders in non-specialized health settings: mental health gap action Programme (mhGAP). World Health Organization; 2016. Author details 20. Dua T, Barbui C, Clark N, Fleischmann A, Poznyak V, van Ommeren M, et al. 1Ghana Somubi Dwumadie (Ghana Participation Programme), PMB 6 Evidence-based guidelines for mental, neurological, and substance use disor- Asoyi Crescent, East Legon, Accra, Ghana ders in low-and middle-income countries: summary of WHO recommenda- 2Kintampo Health Research Centre, Research and Development Division, tions. PLoS Med. 2011;8(11):e1001122. Ghana Health Service, Kintampo North Municipality, Bono East Region, 21. De Silva MJ, Rathod SD, Hanlon C, Breuer E, Chisholm D, Fekadu A, et al. Ghana Evaluation of district mental healthcare plans: the PRIME consortium meth- 3Centre for Global Mental Health, Health Service and Population Research odology. Br J Psychiatry. 2016;208(s56):63–s70. Department, Institute of Psychiatry, Psychology and Neuroscience, King’s 22. Eaton J, McCay L, Semrau M, Chatterjee S, Baingana F, Araya R, et al. Scale up Global Health Institute, King’s College London, London, UK of services for mental health in low-income and middle-income countries. 4Department of Psychiatry and Mental Health, Alan J Flisher Centre for The Lancet. 2011;378(9802):1592–603. Public Mental Health, University of Cape Town, Cape Town, South Africa 23. Ghana Somubi Dwumadie. Theory of Change for district mental healthcare 5Department of Social and Behavioural Sciences, College of Health plans in three demonstration districts: (https://www.ghanasomubi.com). Sciences, University of Ghana, Accra, Ghana 2021 July 2021. 24. Weobong B, Sakyi L, Lund C. Towards Implementation of Context-specific Received: 12 October 2022 / Accepted: 12 April 2023 Integrated District Mental Health Care Plans: A Situation Analysis of Mental Health Services in Five Districts in Ghana. 2021. 25. Weobong B, Akpalu B, Doku V, Owusu-Agyei S, Hurt L, Kirkwood B, et al. The comparative validity of screening scales for postnatal common mental disorder in Kintampo, Ghana. J Affect Disord. 2009;113(1–2):109–17. 26. Weobong B, Soremekun S, Ten Asbroek AH, Amenga-Etego S, Danso S, Owusu-Agyei S, et al. Prevalence and determinants of antenatal depression Ae-Ngibise et al. BMC Psychiatry (2023) 23:280 Page 12 of 12 among pregnant women in a predominantly rural population in Ghana: the 51. Jordans M, Rathod S, Fekadu A, Medhin G, Kigozi F, Kohrt B, et al. Suicidal ide- DON population-based study. J Affect Disord. 2014;165:1–7. ation and behaviour among community and health care seeking populations 27. Carroll HA, Hook K, Perez OFR, Denckla C, Vince CC, Ghebrehiwet S, et al. in five low- and middle-income countries: a cross-sectional study. Epidemiol Establishing reliability and validity for mental health screening instruments Psychiatric Sci. 2018;27(4):393–402. in resource-constrained settings: systematic review of the PHQ-9 and key 52. Rathod SD, Roberts T, Medhin G, Murhar V, Samudre S, Luitel NP, et al. recommendations. Psychiatry Res. 2020;291:113236. Detection and treatment initiation for depression and alcohol use disorders: 28. Gelaye B, Williams MA, Lemma S, Deyessa N, Bahretibeb Y, Shibre T, et al. facility-based cross-sectional studies in five low-income and middle-income Validity of the patient health questionnaire-9 for depression screening and country districts. BMJ open. 2018;8(10):e023421. diagnosis in East Africa. Psychiatry Res. 2013;210(2):653–61. 53. Agyapong VI, Osei A, Farren CK, McAuliffe E. Task shifting–Ghana’s community 29. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression mental health workers’ experiences and perceptions of their roles and scope severity measure. J Gen Intern Med. 2001;16(9):606–13. of practice. Global health action. 2015;8(1):28955. 30. Kroenke K, Spitzer RL. The PHQ-9: a new depression diagnostic and severity 54. Lund C, Breen A, Flisher AJ, Kakuma R, Corrigall J, Joska JA, et al. Poverty and measure. SLACK Incorporated Thorofare, NJ; 2002. pp. 509–15. common mental disorders in low and middle income countries: a systematic 31. Weobong B, Ten Asbroek AH, Soremekun S, Manu AA, Owusu-Agyei S, Prince review. Soc Sci Med. 2010;71(3):517–28. M, et al. Association of antenatal depression with adverse consequences for 55. Hailemariam S, Tessema F, Asefa M, Tadesse H, Tenkolu G. The prevalence of the mother and newborn in rural Ghana: findings from the DON population- depression and associated factors in Ethiopia: findings from the National based cohort study. PLoS ONE. 2014;9(12):e116333. Health Survey. Int J mental health Syst. 2012;6(1):1–11. 32. Üstün TB, Kostanjsek N, Chatterji S, Rehm J. Measuring health and disability: 56. Deribew A, Tesfaye M, Hailmichael Y, Apers L, Abebe G, Duchateau L, et al. Manual for WHO disability assessment schedule WHODAS 2.0. World Health Common mental disorders in TB/HIV co-infected patients in Ethiopia. BMC Organization; 2010. Infect Dis. 2010;10(1):1–8. 33. Endsley P, Weobong B, Nadkarni A. Psychometric properties of the AUDIT 57. Anyinzaam-Adolipore JN, Alhassan AR. Sociodemographic Correlates of among men in Goa, India. Asian J Psychiatry. 2017;29:54–8. Alcohol Abuse in Kassena-Nankana Municipality, Ghana. Journal of Addiction. 34. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The alcohol use 2020;2020. disorders identification test. World Health Organization Geneva; 2001. 58. Michael A, Emmanuel A. A study of the diagnostic practices for mental disor- 35. Organization WH. AUDIT: the alcohol use disorders identification test: guide- ders in Ghana. J Clin Rev Case Rep. 2020;5(4):189. lines for use in primary health care. World Health Organization; 2001. 59. Wilson RS, Yung AR, Morrison AP. Comorbidity rates of depression and anxiety 36. Saunders JB, Aasland OG, Babor TF, De La Fuente JR, Grant M. Development in first episode psychosis: a systematic review and meta-analysis. Schizophr of the alcohol use disorders identification test (AUDIT): WHO collaborative Res. 2020;216:322–9. project on early detection of persons with harmful alcohol consumption-II. 60. Wandersman A, Nation M. Urban neighborhoods and mental health: psycho- Addiction. 1993;88(6):791–804. logical contributions to understanding toxicity, resilience, and interventions. 37. Nadkarni A, Garber A, Costa S, Wood S, Kumar S, MacKinnon N, et al. Auditing Am Psychol. 1998;53(6):647. the AUDIT: a systematic review of cut-off scores for the Alcohol Use Disorders 61. Pearlin LI, Menaghan EG, Lieberman MA, Mullan JT. The stress process.Journal Identification Test (AUDIT) in low-and middle-income countries. Drug Alco- of Health and Social behavior. 1981:337–56. hol Depend. 2019;202:123–33. 62. Badu E, O’Brien AP, Mitchell R. An integrative review of potential enablers and 38. Localio AR, Margolis DJ, Berlin JA. Relative risks and confidence intervals barriers to accessing mental health services in Ghana. Health Res Policy Syst. were easily computed indirectly from multivariable logistic regression. J Clin 2018;16(1):110. Epidemiol. 2007;60(9):874–82. 63. Schaakxs R, Comijs HC, van der Mast RC, Schoevers RA, Beekman ATF, Penninx 39. Stata A. Stata base reference manual release 14. 2015. BWJH. Risk factors for Depression: Differential Across Age? Am J Geriatric 40. Amu H, Osei E, Kofie P, Owusu R, Bosoka SA, Konlan KD, et al. Prevalence Psychiatry. 2017;25(9):966–77. and predictors of depression, anxiety, and stress among adults in Ghana: a 64. Kagee A, Tsai AC, Lund C, Tomlinson M. Screening for common mental community-based cross-sectional study. PLoS ONE. 2021;16(10):e0258105. disorders in low resource settings: reasons for caution and a way forward. Int 41. Asante KO, Andoh-Arthur J. Prevalence and determinants of depres- Health. 2012;5(1):11–4. sive symptoms among university students in Ghana. J Affect Disord. 65. Thornicroft G, Ahuja S, Barber S, Chisholm D, Collins PY, Docrat S, et al. 2015;171:161–6. Integrated care for people with long-term mental and physical health condi- 42. Ngugi AK, Bottomley C, Kleinschmidt I, Wagner RG, Kakooza-Mwesige A, Ae- tions in low-income and middle-income countries. The Lancet Psychiatry. Ngibise K, et al. Prevalence of active convulsive epilepsy in sub-saharan Africa 2019;6(2):174–86. and associated risk factors: cross-sectional and case-control studies. Lancet 66. Jordans MJ, Luitel NP, Kohrt BA, Rathod SD, Garman EC, De Silva M, et al. Neurol. 2013;12(3):253–63. Community-, facility-, and individual-level outcomes of a district mental 43. Ae-Ngibise KA, Akpalu B, Ngugi AK, Akpalu A, Agbokey F, Adjei P et al. Preva- healthcare plan in a low-resource setting in Nepal: a population-based evalu- lence and risk factors for Active Convulsive Epilepsy in Kintampo, Ghana.The ation. PLoS Med. 2019;16(2):e1002748. Pan African Medical Journal. 2015;21(29). 67. Agyapong V, McAuliffe E, Farren C. Improving Ghana’s mental health care 44. Adjorlolo S, Anum A. Positive and negative psychosis risk symptoms among through task shifting–psychiatrists and health policy directors views. Eur adolescents in Ghana. Int J Adolescence Youth. 2021;26(1):307–20. Psychiatry. 2016;33(S1):488–S. 45. Adjorlolo S, Mensah G, Badzi CD. Psychosis risk among pregnant women in 68. Okyere E, Mwanri L, Ward P. Is task-shifting a solution to the health workers’ Ghana. PLoS ONE. 2022;17(2):e0263011. shortage in Northern Ghana? PLoS ONE. 2017;12(3):e0174631. 46. Jenkins R, Mbatia J, Singleton N, White B. Prevalence of psychotic symptoms 69. Costanza R, Daly L, Fioramonti L, Giovannini E, Kubiszewski I, Mortensen LF et and their risk factors in urban Tanzania. Int J Environ Res Public Health. al. The UN Sustainable Development Goals and the dynamics of well-being. 2010;7(6):2514–25. 2016. 47. Jenkins R, Othieno C, Ongeri L, Ogutu B, Sifuna P, Kingora J, et al. Adult psy- 70. Lund C, Tomlinson M, Patel V. Integration of mental health into primary care chotic symptoms, their associated risk factors and changes in prevalence in in low- and middle-income countries: the PRIME mental healthcare plans. Br men and women over a decade in a poor Rural District of Kenya. Int J Environ J Psychiatry. 2016;208(s56):1–s3. Res Public Health. 2015;12(5):5310–28. 71. Quakyi NK, Asante NAA, Nartey YA, Bediako Y, Sam-Agudu NA. Ghana’s 48. Moreno-Küstner B, Martin C, Pastor L. Prevalence of psychotic disorders and COVID-19 response: the Black Star can do even better. BMJ global health. its association with methodological issues. A systematic review and meta- 2021;6(3):e005569. analyses. PLoS ONE. 2018;13(4):e0195687. 49. Oppong Asante K, Kugbey N. Alcohol use by school-going adoles- cents in Ghana: prevalence and correlates. Mental Health & Prevention. Publisher’s Note 2019;13:75–81. Springer Nature remains neutral with regard to jurisdictional claims in 50. Osei-Bonsu E, Appiah PK, Norman ID, Asalu GA, Kweku M, Ahiabor SY, et published maps and institutional affiliations. al. Prevalence of alcohol consumption and factors influencing alcohol use among the youth in Tokorni-Hohoe, Volta region of Ghana. Sci J Public Health. 2017;5(3):205–14.