Introduction:

3. Method

3.1. Study Area and Period

Woliso is situated inside the regional state of Oromia.South West Shoa Zone, 96 kilometers from Addis Abeba. Woliso town is bounded on the west by Woliso Rural subdistricts. Woliso jail was established in 1956 and serves regions in the South West Shoa Zone, according to an administrative report. Woliso jail has 1644 inmates, 165 staff, and a total of 1805 inmates. The facility includes a gaming zone, a protestant church, an orthodox church, a mosque, a medium clinic with six health workers and one psychology professional, a KG to Grade 10 school, two technique training centers, and a library.  The research was conducted in Woliso Prison between March ,8-25, 2023.

3.2.   Study design

A cross-sectional study was carried out in an institution, and it was complemented by a qualitative investigation.

3.3. Population

3.3.1.   Source population

All prisoners in Woliso Prison

3.3.2. Study population

Randomly selected prisoner in Woliso Prison  

 3.3.4. Inclusion and exclusion criteria

 3.3.4.1 Inclusion criteria

All prisoners stay for at least 15 days in prison

3.3.4.2. Exclusion criteria

Prisoner on trial process

Prisoners stay <15 days were excluded from both qualitative and quantitative study.

 3.4. Sample Size and Sampling Technique

3.4.1. Sample size determination

3.4.1.1. Sample size determination for Quantitative

Based on a prior study conducted in a comparable context, the single population proportion method was utilized to calculate the sample size. The study conducted at Jimma Prison (40) found that the prevalence of depression was 41.9%. The formula was based on the assumption of 95% CI and 5% marginal error.

                                                =   (1.96)² ((0.419) (0.581)      

                                                                (0.05)²

                                                                                  ^{= 374}

Where,

n = the desired minimum sample size

                   Zα/₂ = the value under standard normal table for the given value of confidence level (z value at α/₂ =0.05)

d = 5% Margin of error

p = prevalence of depression, which is 41.9%.

Sample size for the second objective was calculated using Epi info 7 software with a 95% confidence interval and an 80% power. Three important factors were picked from a prior study: social support, joblessness in jail, and alcohol usage, and sample sizes were determined for each of them.

Table 1: Sample size determination for second objective, 2023

Given that the larger sample size was chosen as the final sample size, the computed sample sizes for the first and second objectives were compared to one another. As a result, 374-sample size was chosen as the initial target for determining sample size. Since there are 1640 inmates overall—less than 10,000—we must apply the correction formula to get the final sample size.

                                    n_f   =   n1+nN  

=   3741+3741640

                                                 = 304 by considering 10% non-response rate

                                                 = 334

Therefore, the final sample size that was included in the study was 334 prisoners.

3.4.1.2 Sample size determination for Qualitative

Two focus group discussions comprise eight prisoners selected from prisoners committee and representatives was held. The participants selected for focus group discussion was excluded from sample frame for qualitative study.

4.4.2. Sampling Technique

A simple random sampling was used to select study participants by using the prisoner’s registration number. The study individual was selected by using table of random number ranges from 1 to 1624.

Figure 2: Schematic diagram of sampling procedure of prisoners in Woliso, Oromia, Ethiopia 2023

3.5. Study Variables

3.5.1. Dependent variable

Depression

3.5.2. Independent variables

Socio-demographic and socio-economic factors: - Age, sex, religion, marital status, residence, income, occupation before incarceration, educational status perceived of social support.

Prison environment related factors: - Previous history of imprisonment, duration of sentence, duration of stay, sentenced for no crime, type of crime, religion practicing place, work in prison, physical exercise.

Medical and mental health related factors: - History of chronic disease, past mental illness, family history of mental illness.

 Behavioral factors (Lifetime Substance use): - History of Smoking status, alcohol consumption and khat chewing.

3.6. Operational Definition

Depression- Depression is a mental health condition marked by persistent sadness, discouragement, low self-esteem, and a lack of interest in routine tasks. Inmates' levels of depression throughout the past 15 days will be assessed using the PHQ9 (46), a patient health questionnaire. If there is depression, it will be noted as 1, and if not, it will be marked as 0. A score of 0–4 means that there are no symptoms of depression, 5–9 means that there are mild symptoms, 10–14 means that there are moderate symptoms, 15–19 means that there are moderately severe symptoms, and >20 means that there are severe symptoms., and >20 indicate severe depressive symptoms.

Social support- Social support is the provision of aid during tough and crucial conditions such as financial, social, and psychosocial in jail, as measured by the Oslo-3 scale, which has 14 scores and is divided into three major areas. A score of 3-8 indicates weak social support, a score of 9-11 indicates moderate social support, and a score of 12-14 indicates good social support(47). 

Substance use: Using at least one of specific substance like alcohol, chat and cigarettes which assessed by Yes/No response.

3.7. Data collection Tool and Technique

The depressive status was evaluated using the Patient Health Questionnaire (PHQ-9). The PHQ-9 employs a quick self-report method to evaluate depressed symptoms based on the DSM IV depression criteria. Zero to three is the range of the PHQ-9 scale. The 14-item Oslo-3 social support scale is used to measure social support. The questionnaire includes socio-demographic and economic indicators, health-related factors, prison-environmental elements, and a history of drug use. The data was gathered using structured questionnaires. Closed-ended questionnaires with a list of choices from several research was created. Four Health Officers and one supervisor collected data. Participants in the research were interviewed face to face. The FGD was guided by a semi-structured questionnaire. One prisoner's representative, twelve prisoner's homeroom representatives, one prisoner's sanitation committee leader, one prisoner's nutrition committee leader, and one prisoner's daily labor committee leader were purposefully chosen, as were sixteen prisoners, fourteen males and two females. The talk was held in the male prisoner's family visit room on the close day of family visits in the male prisoner's compound. The investigator mediated the FGD, which was helped by supervisors as a note taker. The conversations were recorded using a mobile audio recorder. The frequency of recurrence of themes, words, and expressions used by discussants to communicate their perspectives in relation to the study issue was carefully monitored. The FGD data were collected until saturation or information was acquired.

3.8. Data Quality Control and Management

A standardized questionnaire created from earlier studies was employed for this investigation. After being developed in English and translated into Afan Oromo, the questionnaire was reviewed twice to ensure uniformity. Before the final actual data collection, the questionnaire that was confusing or incorrect was amended. Supervisors and data collectors received one day of training on data gathering procedures and tools. The investigator and supervisors checked the collected data for consistency and completeness. As a result, incomplete values were instantly fixed.SPSS was used to evaluate data completeness, missing values, and outliers, as well as the accuracy of the data entered. The audio was regularly listened to after data collection and before transcription for qualitative data.

3.9. Data Analysis Procedures

Following data collection, each questionnaire was reviewed for completion. Each questionnaire was coded individually. To characterize the research population in respect to key factors, summary statistics were used. For the study population, frequencies and measures of central tendency, as well as measures of variance, were characterized in respect to pertinent factors. Following the descriptive analysis, binary logistic regression analysis was utilized to select candidate variables with a p- value threshold of 0.25. To fit the final model, backward LR approaches were utilized. The final model is then fitted using multivariable regression analysis. Risk factors for depression in prisoners were identified using the binary logistic regression approach. The variance inflation factor (VIF), which was used in this study to test for multi-collinearity, produced a high of 1.48VIF, which is well below the allowed cutoff limit. The Hosmer-Lemeshow goodness of fit (GOF) test was used to evaluate the model. The model is well-fitting, as indicated by the Hosmer-Lemeshow GOF test's p-value of 0.761.An adjusted odds ratio (AORs) with 95% confidence intervals was calculated, and 5% p-value significance was evaluated and reported. The results were presented using a table and narrative. The FGD data was transcribed and then translated from Afan Oromo to English using the theme analysis approach.

4. Results

4.1. Socio-demographic characteristics of study participants

A 100% response rate was achieved by the 334 inmates that took part. The respondents' ages vary from 18 to 70, with a mean age of 25.7 years (25.7+ 8.24), and 134 (40.1%) of the participants had completed primary education. About 344 (43.1%) of the participants were Orthodox, while 108 (32.3%) were Protestant, and the rest were Muslims, Wakefatta, Catholics, and Adventists. With regard to the marital status of inmates, 136 (40.7%) were married, 49.6% were married, and the remaining 8.7% were widowed, divorced, or separated. A total of 111 (33.2%) of the inmates received sentences of less than a year in jail, 146 (46.7%) received sentences of one to five years in prison, and 77 (23.1%) received sentences of more than five years in prison.

Table 2: Socio-demographic characteristics of Woliso prisoners, Woliso Prison, Ethiopia, 2023 (n =334)

4.2. Prison Environment Related Factors

Of the participants, 24 (7.2%) had a history of prior sentences, according to the report. About 32% of them were incarcerated for less than a year, while the median sentence was two years, with an interquartile range of one year. About 182 (54.5%) will be sentenced for >5 years, while 118 (35.3%) will be imprisoned for 2–5 years, and 34 (10.2) will be sentenced for <1 year. Regarding physical exercise, 129 (38.6%) people engaged in physical exercise activities per week. Out of the respondents, 155 (46.4%) had not worked, while over half (53.6%) had employment options within the institution. Only 22 inmates (6.6%) reported having high perceived social support, whereas 159 inmates (47.6%) reported having low perceived social support.

Table 3: Prison Environment related variables of Woliso prison prisoners, Ethiopia, 2023 (n =334)

4.3. Clinical and Behavioural Characteristics of the Prisoners

Regarding behavioural characteristics, 307 (91.9%) of the participants had the habit of attending religious places. Among study participants, 15 (4.5%) of the prisoners had a family history of mental illness, 20 (6%) were living with chronic diseases, and 13 (3.9%) had a history of past mental illness.

Table 4. Clinical and Behavioral characteristics of the prisoners of Woliso prison prisoners, Ethiopia, April 2023 (n =334)

4.4. Substance Use

Regarding unhealthy behaviour more than one-third 113(33.8%) of prisoners consume alcohol and 31(9.3%) were smoker. From our study subjects 32(9.6%) had chewed chat while 302(90.4%) had not. (Fig.3)

Figure 3: Proportion of substance use among prisoners included the study, Woliso prison, Oromia: Ethiopia, 2023

5. Discussion

The overall magnitude of depression among prisoners was 49.7% (95% CI: 44.0-55.10), according this study. According to the quantity and intensity of symptoms, less than 1% of prisoners had moderately severe or severe depression, 11.3% had moderate depression, and 37.4% had light depression. Depression was more common among inmates who were 18–24 years old, unmarried, did not embrace their crime, used drugs, committed murder, or had been incarcerated for more than five years than inmates who were not.

According to our study, the total prevalence of depression among inmates was significantly greater than both the global depression prevalence of 27.6%(1)and Ethiopia's national estimate of depression in the general population of 11% (48). Depressive and other mental disorders can develop in prison due to a variety of unfavorable environmental factors, including a lack of social interaction, forced isolation, overcrowding, hostility, a lack of personal space, insecurities about future prospects, and limited access to social services. The outcome is consistent with Indian research. 53.3% of Odesha jail(29), Jima town prison 44.9% (40), Bahirdar prison 45.5% (39) , West Oromia Wolega prison  45.1% (19). However, this study finding was higher than a study conducted in Durban, South Africa, among 193 prisoners the prevalence of depression was 24.9% (34). In Eastern Nepal, a cross-sectional research found that 35.3% of 434 randomly chosen prisoners had depression(32). This disparity might be attributed to the fact that various depression diagnostic tools were employed to diagnose depression, such as the PHQ-9 for our study, which was used for screening purposes, and the Mini International Neuropsychiatric Interview (MINI) for previous studies, which was used for diagnostic purposes. Furthermore, differences in depression severity may be due to differences in prison environments, such as Ethiopia's high rate of inmate depression may be explained by overcrowding, poor jail infrastructure in low-income nations, and a lack of health care.

This study also discovered the risk factors for depressive illnesses among convicts. One of the main factors of depressive disorder was age. Prisoners between the ages of 21 and 25 were more likely to acquire depression than those over the age of 35. A prior research done at Arba Minch Jinka jail disputed the study showing convicts aged 48 were three times more likely to suffer depression than those aged 18-27(41). The probable discrepancy might be attributed to age classification, as this study included older people starting at 39 years old, whereas earlier studies started at 48 years old. However, our findings are consistent with those obtained from studies conducted in Jimma, Wolega, and Mekele general prisons. Individuals in this age bracket are more likely to report chewing khat and other drug usage before to imprisonment, which may contribute to depression following withdrawal (17,19,40).  

According to research from Bahir Dar jail, inmates serving sentences longer than five years had a higher likelihood of experiencing depression than those serving sentences less than a year.(39). This suggests that inmates serving longer sentences are more likely to have depressive symptoms than those serving shorter sentences, which may raise concerns about the need for mental health services for inmates serving terms longer than five years. In keeping with research from Ethiopia's Bahirdar and Debre Berhan prisons, our study also found that inmates with long-term medical conditions are more likely to experience depression.(39,44).

Similar to the current finding, a prior study found that inmates who worked within the jail had lower levels of depression. This might be as a result of their adjustment to the challenging life behind bars and the fact that they receive pay from their work, which lessens their depression.(17,19). This indicates that giving convicts employment opportunities in the detention facility would boost their production and efficiency and guarantee them a job upon their release, which in turn lowers the prevalence of depression.

The likelihood of depression was higher among inmates who did not visit any religious institutions than among those who did. This study supports the one carried out at Wolega Prison (15). Believing in a higher power can help people deal with life's challenges and find hope, which reduces risk and prevents depression.

Compared to those who were sentenced for crimes they had committed, those who had not yet received a term for no offense had a higher likelihood of experiencing depression. This study is comparable to those carried out at Mekele General Central Prison and Wolega Prison.(17,19). Imprisonment for crime never committed and injustice court cases might be a stressful period and prisoners are vulnerable to extra stress from uncertainty and proximity of possible sentences, which might precipitate depression.

Our results are in line with another general finding from the larger literature on depressive disorder, which states that inmates who smoke cigarettes have a higher risk of developing depression than those who do not. Lifetime khat chewing, cigarette smoking, and depression were significantly correlated.(40). According to different research conducted in Jimma and Bahirdar Prison, inmates who chewed khat before being imprisoned had a higher risk of developing depression than those who did not. It's possible that the withdrawal symptoms of smoking and chewing khat are the cause of depression in these individuals. This suggests that prisoners with a history of substance abuse ought to get alternative treatment.(18,39).

This study found a statistically significant correlation between depression and chronic illness, which is consistent with research conducted in Southern Ethiopia(18). One possible explanation is that depression is more prevalent in persons with long-term physical illnesses. This might be because chronic diseases are stressful conditions that persist for a long period and may not be curable.

In line with a study conducted in West Oromia Wolega prison (19)this study found that inmates with low social support were more likely to experience depression. This finding may be due to the fact that depressed inmates are more likely to experience problems in many areas of their lives and seem less able to adjust to prison or life after release.

Strengths and Limitations

The quantitative part of the study was supplemented by qualitative study and standard tools were used to assess the prisoners level of depression and its related variables among prisoners.

This study employed an interviewer-administered questionnaire to assess risk factors and the extent of depression among prisoners. Social desirability bias is the tendency for people to attempt to provide a more favorable image of themselves as a consequence. Furthermore, in order to precisely analyze the influence of the political setting on the level of depression among prisoners, it would have been better to triangulate the study using qualitative approaches. Time and financial limitations prevented this from being completed, though.

6. Conclusion

Generally, in our study prevalence of depression is high. Half of prisoners were affected by depression.  Level of social support, lifetime drug use, no attending religious place, no availability of work in prison, human right violation, remission restriction, sentences no crime charge due to unfair trail, living conditions like poor food quality, bedding scarcity and poor healthy prison environment were associated with depression.

7. Recommendation

Depression is associated with prison environment; substance use and behavioural and healthy condition of prisoners. Furthermore, magnitude of depression among prisoners is high in relation to general population that need intervention of inter sectorial collaboration and community participation as a whole. As per the findings of this study:

Prisoners

• Prisoners need to empower themselves through participating in work and cooperate with any activities that deal with prisoners’ affairs.
• Refrain from Substance use and addiction.

Community and Prisoner’s family

• Refrain from crime
• Visit on regular base and empower their prisoned family

Police and Judicial Institution

• Works on crime prevention.
• Exercise corruption free and fair trial.
• Protect the right of prisoners.

Prison Health Center

• Screen, early diagnosis and treat for depressed prisoners.
• Strengthen psychiatric and social works at prison level.
• Regular follow up for prisoners on chronic care.

Zonal Prison Administration

•  Give people who were sentenced to a long period of time and lacked social support extra consideration.
• Provide job opportunity and train prisoners on entrepreneurship.
• Create conducive, safe and secure prison environment, recreational and reading facility.

Regional Prison Commission

• Improve housing condition, bedding and quality of food prepared for prisoners.
• Develop strategy for phone call and usage of social media in prison.

Regional Government

• Create productive and crime free society.
• Advance in crime investigation, police and security.
• Give remission and parole release on trend schedule.

Researchers

Case control study needed because this our study expected to be prone to limitation of cross- sectional study (temporal relationship) to know depression baseline due to prison.

Acknowledgment

We would like to thank the Department of Public Health and the College of Medicine and Health Sciences at Ambo University for their administrative support in enabling postgraduate study. Following that, I'd like to thank my advisers, Mr. Abera Shibiru and, Dr. Eshetu Ejeta, for their insightful recommendations, advice, and relevant remarks throughout the report writing process. Third, we would want to express our heartfelt gratitude to everyone participating in data collecting for giving their time. Finally, I'd like to thank the Woliso town health office and the convicts who took part in this study.

Availability of data and Materials

The corresponding author may provide the data used to support the findings of this study upon request..

Funding statement

No funding received for conducting this study.

Author information

Authors and Affiliations

Department of Public Health, Dawo District Health office, Southwest Shoa, Woliso, Ethiopia

Department of Public Health, College of Medicine and Health Sciences, Ambo University, P.O.Box 240, Ambo, Ethiopia.

Author’s contribution

Ararsa Gonfa has contributed to idea inception, design methodology, data entry, data analysis, and manuscript preparation for publications.

Eshetu Ejeta has contributed in advising the proposal and thesis writing, and assisting manuscript writing.

Abera Shibiru has contributed in advising the proposal writing, and assisting manuscript writing.

Kumsa Negasa has contributed to data collection, data entry, data analysis, manuscript preparation for publications.

Corresponding author

Correspondence to Kumsa Negasa

Ethics Declarations

Ethics approval and consent to participate

The Institutional Review Board (IRB) of Ambo University's College of Medicine and Health Sciences granted ethical permission for the study after the advisor approved it; the reference number for Woliso Prison is Ref. No. AU/PGC/571/2015 E.C. Individual research participants submitted written agreements for their willingness to participate after obtaining permission from jail administration to conduct the study and explaining the goal of the study as well as the risks and benefits; the interview was held in a quiet location to protect the privacy of the prisoners; they were informed that there are no advantages or disadvantages to participating in this study; lastly, we gave them our word that the data they provided would be kept totally private. Last but not least, we reassured them that every detail they had supplied would be kept totally confidential, emphasizing that no information was shared and that inmates may opt out of the interview at any moment without it having an impact on the services that medical facilities provide.

Consent for publication

Not applicable

Competing interest

The authors do not have any competing interest.