Methods
The study was conducted at the Psychiatric Unit of Muhimbili National Hospital (MNH) in Dar es Salaam, the economic centre and fastest growing city in Tanzania with a population of more than 3 million people. It has three districts namely Ilala, Kinondoni and Temeke. MNH is located in Ilala and is the nation government referral hospital with the highest specialized health services. It is a place where patients with complicated health problems from all over the country can be found. The psychiatric unit is one of the departments in the directorate of clinical services at MNH. Unlike other departments which receive referrals from all over the country, this unit only caters for patients who are referred from the three districts. Psychiatric services at this unit are provided in two major forms depending on age. Firstly, it is provided for adults who receive either inpatient or outpatient care, and more serious patients who require long term hospitalization are referred to Mirembe hospital located in Dodoma region which is a specialized national psychiatric hospital. Secondly, there is child and adolescent care which is usually provided on the outpatient basis. A special day has been allocated every week (Thursday) for children and adolescents to be seen by health care providers. According to the records obtained at the unit, about 30 children and adolescents were seen every week.
Respondents were parents or guardians who brought their children to attend the weekly outpatient psychiatric clinic at the unit. Convenience and purposive sampling methods was used. Respondents were recruited through the identified child and adolescent nurse counselor working at the psychiatric unit. The first author worked together with the counselor in recruiting respondents while waiting for their children to be seen by the health care providers. The inclusion criterion was a parent or guardian who had lived for at least six months with the mentally ill child. This was considered an adequate period for having reasonable experience. Parents/guardians who met this criterion and consented were included in the study.
Focus group discussions (FGDs) and in-depth interviews were used to gather data. A semi-structured interview guide was used to interview respondents. All interviews were audio-recorded and a note book was used to take field notes. During FGDs the moderator (author) led the discussion and kept the conversation flowing while the research assistant was recording the interviews and taking field notes. Basic demographic data was also collected from respondents after conducting the interviews. All interviews were transcribed verbatim. Two FGDs and 8 in-depth interviews were conducted. The first FGD was attended by 5 respondents (all were mothers) and the second one by 6 respondents (3 fathers and 3 mothers). Eight in-depth interviews were conducted with 7 biological mothers and one grandmother. Conducting in-depth interviews was stopped when no more information could be obtained from caregivers. The purpose of using FGDs was to help get general information and ideas that were further explored during in-depth interviews.
Analysis of data was done by using content analysis which consisted of reading and re-reading the text, manual coding in the margins, and through memos, synthesizing and grouping of data in relatively exhaustive categories [7]. Data was analyzed in the original language (Swahili) in order to minimize the possibility of losing the original meaning of concepts. Analysis of data was done by the two authors and whenever there was a discrepancy in forming codes, categories or themes discussions were done to reach a consensus. Additionally, the analysis process was audited by a third person who was not part of the study but who is conversant with qualitative methods. Translation into English was done for what was included in the report.
The ethical approval of the study was obtained from the Research and Publications Committee of Muhimbili University of Health and Allied Sciences and permission to conduct the study was obtained from Muhimbili National Hospital. Written informed consent was sought from all participants prior to interview sessions.