Empowerment among treatment-engaged individuals living with schizophrenia in Tanzania

Abstract

Purpose

As low-resource settings move to address the treatment and social service gap for people living with schizophrenia (PLWS), person-centered and recovery-oriented interventions should monitor impacts on empowerment. This study explores empowerment and associated factors among PLWS in Tanzania.

Methods

This study uses endline data from the Culturally Adapted Family Psychoeducation for Adults with Psychotic Disorders in Tanzania (KUPAA) pilot trial. Participants included 66 dyads of PLWS and caregivers recruited from two tertiary-level hospitals. The main outcome variable of interest was empowerment (Rogers 28-item Empowerment Scale), and the main exposure variable was Participation in Society (Domain 6 of the WHO Disability Assessment Schedule, WHODAS 2.0). Key psychosocial correlates of interest included hopefulness, general self-efficacy, internalized stigma, and family functioning. Bivariate and multivariable analyses were used to explore variable relationships.

Results

There were 21 women and 40 men with a mean age of 32 years. Bivariate analyses revealed greater participation in society (p < 0.0003) was correlated with greater empowerment, higher hopefulness (p < 0.0001) and higher self-efficacy (p < 0.0001). Lower empowerment was correlated with higher self-stigma (p < 0.0001) and worse family functioning (p < .001). Multivariable models indicated more participation in society was associated with higher empowerment, but when hope, self-efficacy, internalized stigma, and/or family functioning were added to the models, those factors were more strongly correlated with empowerment than participation in society.

Conclusion

Empowerment is increasingly being recognized as an important outcome of psychosocial interventions. Understanding empowerment and its possible effects on recovery-centered outcomes is important when thinking of future interventions for PLWS in low-resource settings. Future recovery-oriented interventions and research should both consider including empowerment measurement among PLWS and incorporate their lived experiences in psychosocial treatment programming.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-025-06700-y.

Introduction

Globally, approximately 24 million people are living with schizophrenia (PLWS) [1, 2]. There is currently a renewed global focus on prioritizing treatment for psychoses in low- and middle-income countries (LMIC) and expanding the research agenda [3, 4]. Historically, evidence-based programs and interventions have focused on clinical recovery, including symptom management and medical adherence,however, there is a movement toward more patient-defined recovery. The oft-cited CHIME conceptual framework focuses on personal recovery related to mental health and identifies five recovery processes: connectedness, hope and optimism about the future, identity, meaning in life, and empowerment [5]. Meanwhile the World Health Organization’s (WHO) Comprehensive Mental Health Action Plan 2013–2030 includes a core principle called “Empowerment of lived experience” [2, 6]. With CHIME and the WHO’s action plan both focusing on empowerment as a key component of recovery, research exploring empowerment among PLSW in LMICs is timely.

Most of the growing evidence-base in the literature on personal recovery or person-centered recovery is focused on high-income country settings [5, 7–9]. While access to recovery-oriented care for PLWS in high-income countries is far from optimal, there is an even larger treatment and social service gap for PLWS in LMICs [1, 2]. Barriers to access to care include poor supply of services, limited literacy about mental health, and stigma [1, 2]. Coercive practices such as mixing medication into food without informing PLWS and involuntary physical restraint continue to occur and de-center treatment preferences of PLWS [10]. In 2021, the Lancet Psychiatry Commission on Psychoses in Global Context, has a goal of addressing challenges related to under-sourced mental health services in LMICs and expanding research beyond North America, Western Europe, and Australasia [3]. The development and testing of recovery-oriented psychosocial interventions for PLWS is currently hampered without more insight on the elements of recovery, including empowerment, in lower resource settings [11].

Empowerment as a key element of recovery for PLWS can be a specific target in psychosocial interventions [5, 12–14]. Currently, there is no consensus on a single definition of empowerment. However, at the individual level, empowerment can be understood as one’s participation in activities, knowledge of resources, and problem-solving abilities [15]. Additionally, empowerment can be understood as the belief and confidence that one has the ability to make choices and have control in one’s life, including a sense of purpose, hope, and active engagement in all aspects of one’s care [1, 2, 16]. Empowerment can range from basic skills of daily living to making informed decisions about where to live and discussions on treatment options. It is often promoted alongside informed decision-making and peer support as psychosocial interventions that involve PLWS in their treatment options and decision-making process [10]. Empowerment has been studied as an outcome of various interventions for PLWS [17, 18]. Additionally, it has been shown to mediate quality of life and recovery among PLWS [19–21]. There are only a limited number of studies looking at factors that influence empowerment in this population [22].

Globally, an estimated 69% of PLWS in LMICs do not receive treatment [4]. A significant treatment gap for schizophrenia is likely also present in Tanzania, but there are limited data available with a nationally representative study to understand the global burden of disease (GBD) and access to services [4]. In Tanzania, there are only 0.07 psychiatrists per 100,000 people [1, 2]. When combining psychiatrists, mental health nurses, psychologists, social workers, and other specialized mental health workers there are still only 1.31 per 100,000 people [1, 2]. Additionally, we know that there have been limited published studies on treatment for psychotic disorders in Tanzania and among those few, improving medication adherence is a prioritized outcome [23],however, more recent research efforts are currently examining peer support and psychosocial outcomes [24]. Broadly, there is a research gap in the literature on empowerment among PLSW in low-resource settings.

Examining empowerment, as a key component of recovery, and its correlates among treatment-engaged PLWS in Tanzania can help inform psychosocial interventions that aim to foster empowerment. To our knowledge, there are no studies that have looked at empowerment as an outcome measure among PLWS in Africa. This study, which uses data from a pilot clinical trial called KUPAA, has three aims: 1) to describe the distribution of an empowerment score among PLWS in Tanzania, 2) to test the hypothesis that among PLWS in Tanzania, increased participation in society is associated with increased empowerment, and 3) to explore other psychosocial correlates of empowerment.

Methods

Study overview

This study uses cross-sectional endline data from a parent study evaluating a psychosocial treatment intervention called Family Psychoeducation for Adults with Psychotic Disorders in Tanzania (KUPAA) (Clinicaltrials.gov ID # NCT04013932, registered July 10, 2019). The purpose of the pilot clinical trialwas to assess the feasibility and acceptability of a culturally tailored Family Psychoeducation intervention for people living with schizophrenia and their caregivers in Tanzania. The main trial enrolled both PLWS and caregivers as dyads, and collected data at three timepoints—baseline, immediate post-intervention, and four to six months post-intervention (endline).

Study setting

Endline data were collected between August and October 2020. The pilot clinical trial was carried out at two sites in Tanzania: Muhimbili National Hospital (MNH) in Dar es Salaam and Mbeya Zonal Referral Hospital (MZRH) in Mbeya city. MNH is a national referral hospital that serves as a local and national referral hospital for inpatient and outpatient psychiatric care. Staff include psychiatrists, psychiatric nurses, social workers, occupational therapists, and psychologists. MZRH is located in the southern part of the country and is the only referral hospital for approximately 2 million people. The Psychiatry and Mental Health Unit staff includes psychiatrists, general practitioners, a psychologist, psychiatric nurses, and social workers. At both facilities, mental health education and family counseling is available, but treatment is largely focused on medication management.

Participants for this analytic sample

The study enrolled 66 pairs of PLWS and their matched caregivers. The endline sample for this study analysis consists of 61 PLWS. Three people were not interviewed due to active relapse, one patient died, and one patient was lost to follow-up. Baseline inclusion criteria for PLWS included regularly attending outpatient psychiatric services at MNH or MZRH, having an ICD-10 (International Classification of Disease) diagnosis of non-organic psychotic disorder, ages 18–50 years at the time of informed consent, and hospitalization or non-hospitalized relapse within the past year.

Procedures

Study visits and data collection occurred in office facilities within MNH and MZRH. Research assistants administered all patient interviews with self-reported assessment measures, including sociodemographic information. Interviews were carried out in Kiswahili, the official language in Tanzania. Data were collected electronically on tablets using REDCap software. Additional details of study procedures have been previously published [25].

Ethical approval

All study procedures were approved by the ethical review boards at Duke University Medical Center (Protocol No. Pro00094163), the University of North Carolina at Chapel Hill (IRB # 21–3328), Muhimbili University of Health and Allied Sciences (Ref No. DA.282/298/0 I.C), Mbeya Zona Referral Hospital (Ref No. SZEC-2./39/R.E IV 11–13), and the Tanzanian National Institute for Medical Research (Ref No. NIMRJHQ/R.8a/Vol. IX/3156) and in accordance with the Declaration of Helsinki. All participants provided written informed consent for the study.

Measures

Data for this current analysis included socio-demographic data collected at endline, except when indicated otherwise, and patient self-reported scales collected at endline. All scales went through the WHO’s four-step process for translation and cultural validation for use in Tanzania, including forward translation, back-translation, pre-testing, and finalization for our Kiswahili versions.

Outcome measure

Empowerment is our dependent variable of interest. To assess empowerment among PLWS we used the Empowerment Scale (ES), a 28-item scale that measures one’s personal sense of empowerment [26]. The scale includes five factors: self-esteem/self-efficacy, power/powerlessness, community activism and autonomy, optimism and control over the future, and righteous anger. The scale is a self-report measure and responses to each item range from 1 = ‘Strongly Agree’ to 4 = ‘Strongly Disagree’. This scale has been used in inpatient and outpatient mental health service settings and in a range of countries including Sweden, Japan, Portugal, and the Netherlands [27–29] The scale had a good internal consistency (reliability) in our study with a Cronbach’s alpha value of 0.82.

Main exposure variable

The main exposure variable of empowerment in this study is the extent to which PLWS participate in society. The World Health Organization Disability Assessment Schedule-Second Version (WHODAS 2.0) is a self-report assessment that measures difficulties performing daily activities in the past 30 days [30]. It consists of 36 Likert-formatted questions, across six domains of functioning: cognition, mobility, self-care, getting along, life activities, and participation in society. We hypothesized that domain six (participation in society) would be most likely correlated with empowerment. It includes the following items asking: 1) How much of a problem did you have joining in community activities (for example, festivities, religious or other activities) in the same way anyone else can? 2) How much of a problem did you have because of barriers or hindrances in the world around you? 3) How much of a problem did you have living with dignity because of the attitudes and actions of others? 4) How much time did you spend on your health condition, or its consequences? 5) How much have you been emotionally affected by your health condition? 6) How much has your health been a drain on the financial resources of you or your family? 7) How much of a problem did your family have because of your health problems? and 8) How much of a problem did you have in doing things by yourself for relaxation or pleasure? [30]. Responses range from 1 (none) to 5 (extreme), and WHODAS 2.0 total scores range from 0 (no disability) to 100 (full disability). For WHODAS 2.0 domain six, the higher the disability scores indicate, greater difficulty participating in society. This measure has shown great internal consistency, reliability and validity across various cultural contexts [31, 32]. In our setting, we calculated a Cronbach’s alpha 0.97, for the entire WHODAS 2.0 scale and 0.91 for domain, which indicates great internal consistency.

Additional measures

Herth Hope Index (HHI): This is a 12-item scale developed to assess hope in adults in a clinical setting [33]. The HHI uses a 4-point Likert scale with responses ranging from 1 (strongly disagree) to 4 (strongly agree) with overall scores ranging from 12 to 48. There are three factors: temporality and future, positive readiness and expectancy, and interconnectedness. It has been tested and validated in low-income settings [34, 35]. In this study, the Cronbach’s alpha value was 0.94.

General Self-Efficacy (GSE): The GSE is a 10-item scale developed to test self-efficacy and predict coping strategies and adaptation to stressful events and has been shown to work in a global context [36–38]. Responses are rated on a 4-point Likert scale with overall scores ranging from 10 to 40, with a higher score indicating higher self-efficacy. This scale has high internal consistency with Cronbach’s alpha value of 0.96.

Internalized Stigma of Mental Illness (ISMI) Scale: This scale is a 29-item questionnaire measuring self-stigma among people with various psychotic disorders and asks respondents how much they agree or disagree with various statements [39]. This scale uses a 4-point Likert scale with responses ranging from 1 (strongly disagree) to 4 (strongly agree). ISMI mean scores range from 1 to 4, with scores below 2.5 considered low internalized stigma. It has been translated and validated in different languages and cultural contexts [40]. In our study, the Cronbach’s alpha value was 0.96.

The Systemic Clinical Outcome and Routine Evaluation version 15 (SCORE-15): This is a self-report measure used to measure overall family functioning, including subscales to measure strength and adaptability, sense of being overwhelmed by difficulties, and communication [41, 42]. The scale has 15 Likert scale items with responses ranging from 1 (very well) to 5 (not at all), with a lower score indicating higher family functioning. The measure was developed in Europe but has been validated in other countries outside of Europe and in multiple languages [43]. It has been shown to have good test–retest reliability with a Cronbach alpha of 0.90 during development [44]. In this present study the alpha value was 0.91.

Data analysis

Descriptive statistics for all study participants are presented in Table 1. Continuous variables were summarized by their means, standard deviation (SD), and range. Counts and percentages were used to summarize categorical variables. Crude linear regressions for empowerment scale score were conducted using various characteristics of individuals with schizophrenia (Table 3).

Table 1.

Characteristics of Participants Living with Schizophrenia (N = 61), Tanzania

	N	%
Gender
Men	40	65.6
Women	21	34.4
Relationship Status
Partnered and living together	10	16.4
Partnered and not living together	2	3.3
Single, not in a relationship	49	80.3
Highest Education Level (at baseline)
Primary or less	22	36.1
Secondary or higher	39	63.9
Worked in past 3 months
No	26	42.6
Yes	35	57.4
Type of Work
Retail/Sales	11	31.4
Other	6	17.1
House chores	5	14.3
Construction	4	11.4
Agriculture	3	8.6
Government or Parastatal	3	8.6
Manufacturing/Industry	2	5.7
Student	1	2.9
	Mean (SD)	Range
Age (at baseline)	32.46 (8.03)	18–50
Empowerment Scale (ES)	2.8 (0.35)	1.6–3.4
Herth Hope Index (HHI)	37.1 (7.7)	14–48
General Self-Efficacy Score (GSE)	28.3 (8.2)	10–40
Self-Stigma (ISMI)	2.0 (0.7)	1.2–3.9
Family Functioning (SCORE-15) (n = 58)	2.4 (0.8)	1.0–4.2
WHODAS 2.0 Domain 6 (WD6)	17.4 (8.3)	8–37

Table 3.

Crude linear regressions for empowerment scale scores on characteristics of individuals with Schizophrenia (N = 61)

Variables	β (95% CI)
Age
18 yrs (Ref)	2.87 (2.68, 3.06)
Slope (linear)	−0.0044 (−0.016, 0.0070)
Gender
Men (Ref)	2.77 (2.66, 2.88)
Women	0.10 (−0.086, 0.29)
Study Arm
Control (Ref)	2.68 (2.55, 2.80)
Intervention	0.24 (0.065, 0.41)*
Site
MNH-Dar es Salaam (Ref)	2.86 (2.75, 2.97)
MZRH-Mbeya	−0.17 (−0.36, 0.022)
Education
Primary or Less (Ref)	2.81 (2.66, 2.96)
Secondary or More	−0.010 (−0.20, 0.18)
Worked in past 3 months
No (Ref)	2.77 (2.64, 2.91)
Yes	0.053 (−0.13, 0.24)
Hopefulness (HHI)
Intercept (HHI = 0)	1.52 (1.23, 1.81)
Slope (linear)	0.035 (0.027, 0.042)***
Self-efficacy (GSE)
Intercept (GSE = 0)	1.91 (1.69, 2.13)
Slope (linear)	0.031 (0.024, 0.039)***
Self-Stigma (ISMI)
Intercept (ISMI = 0)	3.53 (3.31, 3.75)
Slope (linear)	−0.36 (−0.46, −0.25)***
Family Functioning (SCORE-15)
Intercept (SCORE-15 = 0)	3.34 (3.099, 3.59)
Slope (linear)	−0.22 (−0.32, −0.13)***
WHODAS 2.0 Domain 6 (WD6)
Intercept (WD6 = 0)	3.13 (2.94, 3.33)
Slope (linear)	−0.019 (−0.029, −0.0091)***

^***p < 0.0001

^**p < 0.001

^*p < 0.01

In preparation for the multivariable analysis, we examined the correlation matrix of all variables and separated variables with correlations over 0.75 to reduce multicollinearity between independent variables. This led to the decision to run eight independent models with four additional variables: age, gender, study arm, and work status in the past three months, which were included as confounders. Mean imputations were performed to handle missing values. Assessment of model fit included examination of R², adjusted R², and correlation coefficients.

Due to the slight non-normality of empowerment scores (Fig. 1), we also conducted nonparametric tests for these associations as a sensitivity analysis. STATA 18.0 was used to perform all statistical analyses [45].

Fig. 1.

Total Empowerment Scale Scores for PLWS, Tanzania, 2020 (N = 61)]

Results

Socio-demographic characteristics

The mean age of the participants is 32.46 (SD: 8.0) and almost two-thirds (65.6%) were men and 34.4% were women. The majority of respondents (80.3%) were single and not in a relationship at endline. Ten respondents were partnered and living with their partner, while two, were partnered, but not living with their partner. Over half of the participants completed secondary education level or higher (63.9%), and over half, (57.4%) worked in the past three months leading up to endline. Most participants worked in retail/sales (31.4%), doing house chores (14.3%), or construction work (11.4%).

Measurement scores

The average empowerment scale score of the respondents was 2.8 (SD: 0.35). All items had some variability, but overall, people have high empowerment scores (Table 2). Participants had a mean score of 37.1 (SD: 7.7) on the hope scale (HHI), which is on the lower end (lower hopefulness), with scores in this sample ranging from 14 to 48. For self-efficacy (GSE), the scores ranged from 10–40, which are the lowest and highest scores that can be calculated, from summing the responses to all items. The average GSE score was 28.3 (SD: 8.2), which is on the higher end, indicating stronger beliefs in one’s ability to handle various situations. The mean internalized stigma scale (ISMI) was 2.0 (SD: 0.7), which is considered low internalized stigma, with scores ranging from 1.2–3.9. The average family functioning (SCORE-15) score was 2.4 (SD: 0.8) with a range of 1–4.2 and the WHODAS 2.0 Domain Six (WD6), mean was 17.4 (8.3; range 8–37). A WD6 score between 5–24 indicates mild difficulty.

Table 2.

Distribution of empowerment scale response items

Item	1. Strongly Agree N (%)	2. Agree N (%)	3. Disagree N (%)	4. Strongly Disagree N (%)	Total
E1. I can pretty much determine what will happen in my life.**	21 (34.4)	32 (52.5)	6 (9.8)	2 (3.3)	61
E2. People are only limited by what they think is possible. **	18 (29.5)	29 (47.5)	9 (14.8)	5 (8.2)	61
E3. People have more power if they join together as a group. **	46 (75.4)	14 (23.0)	1 (1.6)	0	61
E4. Getting angry about something never helps	42 (68.9)	11 (18.0)	8 (13.1)	0	61
E5. I have a positive attitude toward myself. **	29 (47.5)	24 (39.3)	7 (11.5)	1 (1.6)	61
E6. I am usually confident about the decisions I make. **	26 (42.6)	23 (37.7)	10 (16.4)	2 (3.3)	61
E7. People have no right to get angry just because they don't like something	18 (30.0)	21 (35.0)	12 (20.0)	9 (15.0)	60
E8. Most of the misfortunes in my life were due to bad luck	27 (45.0)	19 (31.7)	3 (5.0)	11 (18.3)	60
E9. I see myself as a capable person. **	21 (34.4)	28 (45.9)	9 (14.8)	3 (4.9)	61
E10. Making waves never gets you anywhere	44 (72.1)	14 (23.0)	3 (4.92)	0	61
E11. People working together can have an effect on their community. **	44 (72.1)	17 (27.9)	0	0	61
E12. I am often able to overcome barriers. **	20 (32.8)	21 (34.4)	15 (24.6)	5 (8.2)	61
E13. I am generally optimistic about the future. **	27 (44.3)	23 (37.7)	9 (14.8)	2 (3.3)	61
E14. When I make plans, I am almost certain to make them work. **	12 (19.7)	30 (49.2)	15 (24.6)	4 (6.6)	61
E15. Getting angry about something is often the first step toward changing it. **	24 (39.3)	20 (32.8)	11 (18.0)	6 (9.8)	61
E16. Usually I feel alone	15 (25.0)	15 (25.0)	12 (20.0)	18 (30.0)	60
E17. Experts are in the best position to decide what people should do or learn	40 (65.6)	15 (24.6)	3 (4.9)	3 (4.9)	61
E18. I am able to do things as well as most other people. **	29 (47.5)	25 (41.0)	4 (6.6)	3 (4.9)	61
E19. I generally accomplish what I set out to do. **	16 (26.2)	26 (42.6)	15 (24.6)	4 (6.6)	61
E20. People should try to live their lives the way they want to. **	36 (59.0)	23 (37.7)	2 (3.3)	0	61
E21. You can't fight city hall	27 (44.3)	20 (32.8)	11 (18.0)	3 (4.9)	61
E22. I feel powerless most of the time	11 (18.0)	12 (19.7)	18 (29.5)	20 (32.8)	61
E23. When I am unsure about something, I usually go along with the rest of the group	29 (47.5)	23 (37.7)	7 (11.5)	2 (3.3)	61
E24. I feel I am a person of worth, at least on an equal basis with others	33 (54.1)	17 (27.9)	7 (11.5)	4 (6.6)	61
E25. People have the right to make their own decisions, even if they are bad ones.**	20 (32.8)	14 (23.0)	8 (13.1)	19 (31.2)	61
E26. I feel I have a number of good qualities. **	27 (44.3)	24 (39.3)	7 (11.5)	3 (4.9)	61
E27. Very often a problem can be solved by taking action. **	37 (60.7)	19 (31.1)	3 (4.9)	2 (3.3)	61
E28. Working with others in my community can help to change things for the better. **	44 (72.1)	14 (23.0)	3 (4.9)	0	61

^** = reversed items

Univariable Analysis

There was little evidence for a association between mean empowerment score and observed sociodemographic characteristics. Crude linear regressions showed a significant (p = 0.008) relationship between empowerment scores and study arm (β = 0.24 95% CI (0.065, 0.41)) [Table 3]. There were strong positive relationships between general self-efficacy (GSE) and empowerment (β = 0.31, 95% CI (0.024, 0.039)), and between hope (HHI) and empowerment (β = 0.35, 95% CI (0.027, 0.042)). Higher self-efficacy (GSE) and hopefulness (HHI) totals among this sample were associated with higher empowerment. There were strong negative relationships between empowerment scores and scores for self-stigma (ISMI) (β = −0.36, 95% CI (−0.46, −0.25)), family functioning (SCORE-15) (β = −0.22, 95% CI (−0.32, −0.13)), and participation in society (WD6) (β = −0.019, 95% CI (−0.029, −0.0091)), and all were statistically significant. Greater empowerment was associated with lower self-stigma, better family functioning, and greater participation in society (less disability). Mann–Whitney tests for categorical variables and Spearman’s rank correlations for continuous variables showed consistent results with the linear regressions (Additional File 1).

Multivariable analysis

Table 4 shows the eight multivariable models. Participation in society (WD6), the main exposure variable, remained in each model. Model 1 included only participation in society (WD6) and the four additional sociodemographic variables; model 2 included hope (HHI); model 3 included self-efficacy (GSE); model 4 included internalized stigma (ISMI); and model 5 included family functioning (SCORE-15). The last four models combined multiple psychosocial variables that did not have correlations over 0.75 in the correlation matrix. Model 6 included hope (HHI) and family functioning (SCORE-15); model 7 included self-efficacy (GSE) and internalized stigma (ISMI); and model 8 included self-efficacy (GSE) and family functioning (SCORE-15).

Table 4.

Multivariable linear regression model for empowerment scale scores using characteristics of individuals with Schizophrenia (N = 61)

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)
R2	0.27	0.64	0.61	0.50	0.39	0.63	0.64	0.61
R2 adjusted	0.21	0.60	0.57	0.44	0.32	0.58	0.59	0.55
WD6	−0.018 (−0.029, −0.0056)**	−0.00020 (−0.0095, 0.0099)	−0.0070 (−0.016, 0.0024)	−0.0075 (−0.018, 0.0033)	−0.013 (−0.025, −0.0011)*	−0.000017 (−0.010, 0.010)	−0.0046 (−0.014, 0.0047)	−0.0060 (−0.016, 0.0040)
HHI		0.035 (0.026, 0.045)***				0.035 (0.023, 0.048)***
GSE			0.029 (0.020, 0.037)***				0.023 (0.013, 0.032)***	0.026 (0.016, 0.036)***
ISMI				−0.31 (−0.43, −0.18)***			−0.14 (−0.27, −0.013)*
SCORE-15					−0.17 (−0.27, −0.064)**	−0.0012 (−0.10, 0.10)		−0.056 (−0.15, 0.040)
Sociodemographic
Age	−0.0035 (−0.014, 0.0075)	−0.0046 (−0.012, 0.0032)	−0.0042 (−0.012, 0.0039)	−0.0047 (−0.014, 0.0045)	−0.0038 (−0.014, 0.0066)	−0.0046 (−0.013, 0.0035)	−0.0046 (−0.012, 0.0032)	−0.0040 (−0.012, 0.0044)
Gender	0.071 (−0.11, 0.25)	0.057 (−0.068, 0.18)	0.066 (−0.064, 0.20)	0.077 (−0.070, 0.22)	0.061 (−0.10, 0.23)	0.055 (−0.075, 0.19)	0.070 (−0.056, 0.20)	0.65 (−0.069, 0.20)
Study arm	0.15 (−0.023, 0.33)	0.091 (−0.036, 0.22)	0.066 (−0.067, 0.20)	0.022 (−0.14, 0.18)	0.071 (−0.10, 0.25)	0.089 (−0.049, 0.23)	0.024 (−0.11, 0.16)	0.056 (−0.086, 0.20)
Worked past 3 months	−0.087 (−0.28, 0.11)	−0.11 (−0.25, 0.024)	−0.11 (−0.25, 0.034)	−0.040 (−0.20, 0.12)	−0.093 (−0.28, 0.091)	−0.11 (−0.26, 0.031)	−0.084 (−0.22, 0.057)	−0.11 (−0.26, 0.036)

^***p < 0.001

^**p < 0.01

^*p < 0.05

Model 1 indicates that greater participation in society (lower score), on its own, has an association with more empowerment (higher score) (p = 0.005, 95% CI (−0.029,−0.0056)); however, when adjusting for other psychosocial measures in multivariable regression, this effect was attenuated. The only other model where participation in society retained its association with empowerment was Model 5 where family functioning was also strongly associated with empowerment. In Models 2 and 6, which included hopefulness (HHI), with and without family functioning, participation in society lost significance but a strong association between higher empowerment and higher hopefulness is illustrated. Likewise, for Models 3, 7 and 8 which included generalized self-efficacy (GSE), all three models indicated higher self-efficacy was associated with higher empowerment; and for Models 4 and 7, which included internalized self-stigma (ISMI), both models showed lower self-stigma was associated with higher empowerment. Hope, self-efficacy, and self-stigma all appear to account for more variation in the empowerment scores as indicated by their higher R² values, even more so than participation in society. Sociodemographic variable coefficients remained largely the same across all models.

Discussion

The current study examined empowerment among PLWS in Tanzania. To the best of our knowledge, there is no previous study looking at empowerment as an essential component of treatment and interventions for PLWS in Tanzania. The mean empowerment score in our sample was slightly higher, but still similar to empowerment scores of PLWS in other countries (United Kingdom, Croatia, Sweden) and study settings which speaks to the universality of empowerment as a construct [12, 18, 46]. A few studies focus on the extent to which person-centered interventions improve one’s empowerment. Allerby et al. [46], found that their person-centered intervention did not lead to greater empowerment, but that it did improve patient satisfaction. Their findings question whether the empowerment scale is best used in clinical contexts since the scale asks questions that require a whole-life perspective such as the item, “People are only limited by what they think is possible.” [46]. In our study, the high correlations between hope, self-efficacy, and self-stigma variables indicate that they may be components of a larger empowerment construct. The Rogers’ empowerment scale is a promising scale that worked well in our study population with largely expected correlations,and a rigorous validation analysis of the empowerment scale both in and out of clinical contexts would be beneficial in East Africa.

Among PLWS, high empowerment was correlated with higher participation in society; however, when hope, self-efficacy, internalized stigma, and/or family functioning were added to the models, those factors were more strongly correlated with empowerment than participation in society. Participation includes being involved in the community, being respected, and having control over one’s life. That said, our findings align with Omigbodun et al.’s [4] call for expansion of a social model of disability where services move beyond clinical treatment and focus on ensuring full participation in society, including peer support. Our findings revealed expected relationships between the variables of interest indicating that the Empowerment Scale (ES) could be a useful measure to include for future client-centered and recovery-oriented clinical trials with PLWS in LMIC settings. However, we do also want to note that the lack of association between empowerment and work status could reflect that PLWS have more flexible opportunities to work in the Tanzanian setting (e.g. agriculture and informal small business) and recovery should be considered as both a process and an outcome.

Social participation has been shown to impact not only prognosis among people with disabilities but also impact patient’s mid- and long-term well-being [47]. A study of 424 adults with disabilities in Spain found that engagement in meaningful activities was important when measuring perceived satisfaction with participation, good quality of life, and emotional well-being [47]. There are limited studies measuring participation in society using only domain six of WHODAS 2.0 among our population, and this could be a limitation of what we are able to conclude about this hypothesized exposure variable.

Correlations between selected variables in our study are largely consistent with previous studies. For example, a study among 257 outpatients diagnosed with severe mental illness in Iran found a negative relationship between internalized stigma and hope (r = −0.55) and self-efficacy (r = −0.64) [48]. Their study did not publish results looking directly at hope and self-efficacy. In our correlation matrix, hope and self-efficacy had a strong correlation, which is why we needed separate models. We can also gain insights from these psychosocial relationships pertaining to caregivers. For example, in a U.S. sample, among caregivers of individuals with serious mental illness, stigma-related caregiving experiences were inversely related to general family functioning and caregiver empowerment [49].

In Tanzania, the availability and cost of medications are severe barriers to participation in clinic-based mental health care, with medication management being the main purpose for PLWS to seek outpatient care [50]. In qualitative interviews with PLWS in Tanzania, employment, family support, peer support, and psychoeducation were found to be protective factors against relapse [51]. Per the government of Tanzania’s Persons with Disability Act, family members are required to provide social support for family members with disabilities, including financial support [52].

In a non-individualistic society, social support from caregivers of PLWS is invaluable; however, there is a greater need for community-level social support and there is a potential for both benefits and conflicts due to increasing individual level empowerment [53]. As a society with more collectivistic values, families in Tanzania often in engage in more shared decision-making about healthcare, but a PLWS who is not empowered, might miss being part of that shared decision-making process. Alternatively, more empowered PLWS could affect family dynamics if all do not agree on treatment choices. We saw some hints about this in our recent qualitative paper on social support [53]. This paper also showed that both PLSW and caregivers were often yearning for PLSW to take a more active economic role in the family (working and making financial contributions) which could be tied to more individual level empowerment. Currently, many PLWS rely on families for their basic needs [53].

This study has several limitations. As a pilot study our sample size was 66 dyads at baseline, with only 61 dyads at endline, and we only focused on data from the PLWS participants. Second, all the measures used self-report, which could lead to bias in responses and potential under-reporting of problems because the interviews took place within treating hospitals. Third, these measures were not originally developed in Tanzania nor validated with this specific population (PLW in Tanzania). However, all measure that had not ever been used in Tanzania previously went through a thorough translation process (forward translate, back-translate, pre-test, expert review, and finalization). Fourth, due to the observational and cross-sectional nature of the data, we cannot draw causal conclusions regarding the estimated relationships between study variables. The Empowerment Scale was only used at endline so we are unsure if and how empowerment scores may have changed over time. Fifth, our sample size is relatively small and it was not originally powered to examine empowerment as an outcome. Lastly, there is weak external validity since the entire sample includes treatment-engaged individuals, and is thus missing people in the community who are untreated. This is an outpatient population receiving medication. In general, our study sample has higher levels of education compared to the general population. This may not be generalizable to people residing in rural settings in Tanzania who may not be aware of what is possible for mental health treatment or of existing services.

These findings have practice implications. Providers and researchers often focus on reducing stigma among PLWS, but, alongside this, we may need to think how to enhance or foster greater empowerment, hope, and self-efficacy in individual treatment sessions and in group-based psychoeducation, both of which could serve as a mutual reinforcing mechanism. With increases in mental health clinical trials across Africa, the field should consider more person-centered outcomes beyond medical and clinical definitions of recovery. Empowerment is one such outcome that captures a conceptual component of recovery. Patient education and counseling may be empowering in of itself. Empowered clients may be more likely to also engage in interventions such as the process of insight orientation, occupation therapy, and home-based or facility-based rehabilitation.

Conclusion

Empowerment is an important element of person-centered recovery for PLWS. By understanding the underlying factors that contribute to or correlate with empowerment, we can better understand how to tailor person-centered recovery for PLWS. However, more research that measures empowerment in East Africa, and LMICs more broadly, can help inform programming.

Abbreviations

GSE

General Self Efficacy

HHI

Herth Hope Index

ISMI

Internalized Stigma of Mental Illness Scale

LMIC

Low- and middle-income countries

PLWS

Person/People living with schizophrenia

SCORE-15

The Systemic Clinical Outcome and Routine Evaluation version 15

WHODAS

World Health Organization Disability Assessment Schedule-Second Version

WD6

WHODAS Domain 6: Participation in society

Authors’ contributions

MJV analyzed the data and wrote the first draft of the manuscript. JRE contributed to the data analysis and revisions. PS, PL, BT, AM, and JH contributed with manuscript edits and study implementation. SK assisted with editing and leading the implementation of the study in Tanzania. JNB contributed with study design, implementation, manuscript conceptualization, and revisions. All authors read and approved the final manuscript.

Funding

The research presented in this manuscript was supported by the National Institute of Mental Health (NIMH) under award number R34MH106663. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Data availability

The dataset used during this current study is available from the corresponding author on reasonable request and with IRB approval.

Declarations

Ethics approval and consent to participate

All study procedures were approved by the ethical review boards at Duke University Medical Center (Protocol No. Pro00094163), the University of North Carolina at Chapel Hill (IRB # 21–3328), Muhimbili University of Health and Allied Sciences (Ref No. DA.282/298/0 I.C), Mbeya Zona Referral Hospital (Ref No. SZEC-2./39/R.E IV 11–13), and the Tanzanian National Institute for Medical Research (Ref No. NIMRJHQ/R.8a/Vol. IX/3156) and in accordance with the Declaration of Helsinki. All participants provided written informed consent for the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.