The study was approved by the Committee on Human Research, Publication and Ethics (CHRPE) of the Kwame Nkrumah University of Science and Technology (KNUST) with approval number CHRPE/AP/335/19. All participants provided written informed consent. All study procedures conformed with the principles guiding research in human subjects as set out in the Declaration of Helsinki .
Study sites and participants
This was a case control study to assess the mental distress associated with BUD among patients with active infection (any manifestation), past infection (i.e., scars and/or contractures), and their caregivers. For the purposes of this study, we have defined BUD as active and past infection. A patient with active infection was defined as one who had been recently confirmed (in the preceding 3 months) to have BUD and was on treatment but whose lesion had not yet healed. BUD patients with past infection were those diagnosed within the preceding three years, who had completed a course of antibiotic therapy for BUD and whose lesions had healed with or without disability. According to the WHO report on ageing and health, ‘a caregiver provides care and support to someone else; such support may include: helping with self-care, household tasks, mobility, social participation and meaningful activities; offering information, advice and emotional support, as well as engaging in advocacy, providing support for decision making and peer support, and helping with advance care planning; offering respite services; and engaging in activities to foster intrinsic capacity’ . In this study, a caregiver was defined as a person who provided such assistance to a BUD patient with either active or past infection. The study was conducted in three rural districts in Ghana namely; Asante Akim North, Ahafo Ano North (both districts in the Ashanti region) and Upper Denkyira East (Central region).
Participant recruitment and sampling
BUD patients with active infection and their caregivers visiting BUD clinics for medical care were identified and recruited using a convenience sampling technique. For the recruitment of BUD patients with past infection, the hospital records of these patients were retrieved and they were contacted and informed of the study and subsequently invited to participate in their community setting. In addition, age and sex matched healthy individuals residing in BUD endemic communities were also contacted and recruited as controls using convenience sampling. Participants were provided with information leaflets informing them of the study. This was read and explained in the local language of Twi to those who were unable to read. Caregivers and healthy controls were included in the study if they had been resident in a BUD endemic community for more than two years, and were aged 18 to 60 years. Participants were excluded if they had a known physical or psychiatric illness that could confound study results. In addition, persons aged < 18 years and those unable to respond to questions were excluded. All participants provided written informed consent.
The data collection tools (Additional file 1) used in this study were chosen to allow comparison with the results of previous psychological [19, 20] and quality of life  studies of BUD, along with consideration to BUD-specific instruments and those recommended in the NTD Toolkit .
WHO Quality of Life-BREF (WHOQOL-BREF), is an international cross-culturally comparable QoL assessment scale used to evaluate people’s perception of their quality of life in relation to their personal goals, concerns and culture. It consists of 26 items which measure across 4 domains: physical health, psychological health, social relationships and the environment. The score for each item ranges from 1 to 5 and the total score for a domain is from 20 to 100 with higher scores indicating greater QoL [24,25,26,27].
WHO Disability Assessment Schedule (WHODAS) 2.0 and the Buruli Ulcer Functional Limitation Score (BUFLS) were employed to assess the degree of functional limitation and participation in BUD patients. The WHODAS 2.0 generally assesses functioning in 6 domains; cognition, mobility, self-care, relationships, life activities and participation in community activities . Scores for each item on WHODAS 2.0 range from 0 to 4. The scores of items across each domain are computed with higher scores representing higher functional limitation.
BUFLS was designed and validated in Benin and Ghana  and is specifically used for assessing functional limitation in BUD. It consists of items related to the performance of 19 common daily activities. Each item is scored between 0 and 2 (0 = can perform activities easily, 1 = difficulties in performing activities and 2 = cannot perform activities at all). Functional limitation score is calculated by summing the individual item scores and dividing by the maximum possible score for an individual and finally multiplied by 100. Higher scores indicate more functional limitations with range between 0 and 100%. A score was not calculated if more than 6 items were not applicable.
Self-reporting questionnaire SRQ-20 is a 20-item scale used to screen for symptoms of mental distress. The score for each item ranges from 0 (symptom absent) to 1 (symptom present). Score items are summed to obtain the total score. Score above cut-off point indicates probable mental distress . A cut-off score of 8 is widely used, however optimal cut off ranges vary across languages and settings [31,32,33]. In this study, we cautiously used 5 as our cut-off point to detect distress that may be present in BUD patients with pre-ulcerative forms.
Hospital Anxiety and Depression Scale (HADS) was chosen to screen for anxiety and depression among the study population. The HADS tool assesses two subscales (anxiety and depression) with scoring for each item ranging from 0 to 3 (0 = lowest anxiety or depression level; 3 = highest anxiety or depression level). A total subscale of 0–7 indicates an absence of anxiety and/or depression; scores 8–10 indicate mild symptoms of either anxiety and/or depression; scores 11–14 indicate moderate symptoms of anxiety and/or depression; scores 15–21 indicate severe symptoms of anxiety and/or depression .
To our knowledge, the above data tools had not previously been used in the local Twi language. In keeping with WHO guidance on research tool use in different languages , all data tools were first translated into Twi by a professional linguistics tutor (MO) familiar with the mental health issues addressed by the various data tools. Next, the translated tools (Additional file 2) were reviewed and portions back-translated by the study coordinator (NA) and study research assistant (MNO). Minor changes were made to better reflect items in the original English language tools. All staff involved in the study then received training on questionnaire administration and conduction of interviews. The study team later discussed and tested the translated versions of the data tools on seven patients at the Agogo BUD clinic for comprehensibility, acceptability and relevance of the items. No major changes were considered necessary.
Data collection took place between 1st August and 30th November 2019 in the BUD clinics and communities within the selected districts in Ghana. Face-to-face interviews were conducted using the study data tools; interviews lasted between 45 and 60 min. For BUD patients with active and past infection, WHODAS, WHOQOL, SRQ-20, HADS, and BUFLS were administered, while WHOQOL, SRQ-20, and HADS were used for caregivers and controls. The HADS tool was only administered to participants who scored ≥ 5 on the SRQ-20 (screening) tool. Persons who scored < 5 were deemed unlikely to have a common mental disorder (anxiety or depression) and so were not administered the HADS tool. Interviews for patients with active infection and their caregivers were conducted in private in the BUD treatment clinic. Interviews for participants with past infection and controls were conducted in health facilities located within the communities. All interviews were conducted in private and confidentiality was maintained at all times. Scores were computed according to the scoring manual for each data tool. Results were entered in Microsoft Excel version 2013 (Microsoft Corporation, Redmond, WA, USA) before being exported to STATA (Stata Corp LP, College Station, Texas, USA) for further analysis.
Study data was analysed using STATA version 14.0. Descriptive statistics were used to summarize the characteristics of the study participants. In addition, the degree of association was evaluated using student t test and chi-square (χ2) or Fisher’s exact tests, where appropriate, with a P value of ≤ 0.05 deemed to be statistically significant. Mean quality of life scores are reported with their respective 95% confidence intervals. All items on a Likert scale were assessed using mean scores. Each item response was scored into 3 different groups: ‘can do easily/at baseline level’; ‘can do with difficulty’; and ‘cannot do at all’. Scores were summarized using frequencies and percentage scores in a bar chart.