Impact statement
Relapse remains one of the major challenges affecting the long-term recovery of people living with SMI, particularly in low-resource settings where access to continuous mental healthcare is limited. This study provides important evidence from Northwest Ethiopia by showing that relapse is common among people receiving psychiatric care and identifying groups who may require closer clinical attention.
The findings highlight that relapse is not only a consequence of psychiatric symptoms but is also associated with broader clinical and social circumstances, including repeated psychiatric hospitalization, longer illness duration, comorbid medical conditions, unemployment and stressful life events. Recognizing these characteristics during routine care may help clinicians identify individuals who require more intensive monitoring and tailored support.
The study has implications for mental health service planning in Ethiopia and similar low- and middle-income settings. Strengthening continuity of care, improving integration between mental and physical healthcare services and addressing psychosocial challenges may contribute to reducing repeated episodes of illness and improving functional outcomes. The findings also emphasize the need for health systems to move beyond acute episode management toward sustained, recovery-oriented care for people with severe mental illness.
By providing context-specific evidence from a region where longitudinal relapse data remain limited, this study contributes to a better understanding of the challenges faced by individuals with severe mental illness in sub-Saharan Africa and may inform future research, clinical practice and policy development aimed at improving long-term mental health outcomes.
Background
Psychiatric disorders, particularly severe mental illnesses (SMIs), contribute substantially to the global and national burden of disease and are among the leading causes of disability, morbidity and reduced quality of life worldwide (Kessler et al., Reference Kessler, Aguilar-Gaxiola, Alonso, Chatterji, Lee, Ormel, Üstün and Wang2009; Santomauro et al., Reference Santomauro, Miller, Shadid, Hanson, Vo, Roy, Hagins, Herrera, Scott and Erskine2026). SMIs, including schizophrenia, bipolar disorder and major depressive disorder with recurrent or psychotic features, are chronic psychiatric conditions characterized by persistent symptoms, marked functional impairment and a high risk of relapse and hospitalization (Organization, 2022b; Birhan et al., Reference Birhan, Rtbey and Gelaw2025). In the context of SMIs, remission refers to a sustained period during which psychiatric symptoms are absent or substantially reduced, accompanied by relative improvement in functioning (American Psychiatric Association, 2022). Relapse is defined as the recurrence or worsening of psychiatric symptoms following a period of partial or complete remission, resulting in clinical deterioration and often requiring treatment modification, intensified clinical management or psychiatric hospitalization (Andreasen et al., Reference Andreasen, Carpenter, Kane, Lasser, Marder and Weinberger2005; Kazadi et al., Reference Kazadi, Moosa and Jeenah2008). Depending on the underlying disorder, relapse may present as the recurrence of psychotic symptoms, manic or depressive episodes, behavioral disturbances or other clinically significant manifestations requiring intervention (Porcelli et al., Reference Porcelli, Bianchini, De Girolamo, Aguglia, Crea and Serretti2016). Because SMIs are typically chronic and recurrent disorders, relapse remains a major challenge to long-term management, recovery and prognosis (Ayano and Duko, Reference Ayano and Duko2017). Recurrent relapse is associated with poorer functional outcomes, reduced quality of life, increased healthcare utilization, greater treatment resistance and a higher risk of repeated hospitalization (Papakostas et al., Reference Papakostas, Dar and Ratna2020).
Relapse is common among individuals with SMIs, although reported rates vary considerably depending on the psychiatric diagnosis, relapse definition, follow-up period and study design (Kessler et al., Reference Kessler, Berglund, Demler, Jin, Merikangas and Walters2005; Birhan et al., Reference Birhan, Rtbey and Gelaw2025). Globally, relapse rates among patients with SMI range from approximately 30% to 80% (Organization, 2023). Studies from developed countries have reported relapse rates of 44% in Spain (Bernardo et al., Reference Bernardo, Amoretti, Cuesta, Parellada, Mezquida, González-Pinto, Bergé, Lobo, Aguilar and Usall2021), and 30.52% in the United States (Rivelli et al., Reference Rivelli, Fitzpatrick, Nelson, Laubmeier, Zeni and Mylavarapu2024), while a 5-year cohort study in the United States found a cumulative relapse rate of 81.9% among patients with first-episode schizophrenia or schizoaffective disorder (Robinson et al., Reference Robinson, Woerner, Alvir, Bilder, Goldman, Geisler, Koreen, Sheitman, Chakos, Mayerhoff and Lieberman1999). In Africa, studies have reported relapse rates of 61.8% in South Africa (Kazadi et al., Reference Kazadi, Moosa and Jeenah2008), 57.4% in Southwest Ethiopia (Fikreyesus et al., Reference Fikreyesus, Soboka and Feyissa2016), 41% in southwest Nigeria (Adebiyi et al., Reference Adebiyi, Mosaku, Irinoye and Oyelade2018) and 52.1% in South-East Nigeria (Eneh and Eneh, Reference Eneh and Eneh2024). Similarly, a recent systematic review of African studies reported relapse (or recurrence) rates ranging from 65% to 73% among patients with bipolar disorder and from 52% to 92% among those with schizophrenia spectrum disorders (Birhan et al., Reference Birhan, Rtbey and Gelaw2025). Among patients with major depressive disorder receiving specialized mental health care, recurrence rates have been reported as 60% after 5 years, 67% after 10 years and 85% after 15 years (Hardeveld et al., Reference Hardeveld, Spijker, De Graaf, Nolen and Beekman2010).
Relapse in SMI is associated with substantial clinical, social and economic consequences (Ayano and Duko, Reference Ayano and Duko2017). Recurrent relapse frequently results in repeated psychiatric hospitalizations, increased utilization of limited mental health services and higher direct and indirect healthcare costs for patients, families and health systems (Amha et al., Reference Amha, Getnet, Munie, Workie, Alem, Mulugeta, Bishaw, Ayenew, Gedfew and Desta2024). It is also associated with poorer long-term outcomes, including impaired social and occupational functioning, progressive functional decline, reduced likelihood of sustained remission and an increased risk of chronic disability (Birhan et al., Reference Birhan, Rtbey and Gelaw2025). Furthermore, relapse has been linked to a greater risk of psychiatric and physical comorbidities, substance use, suicidal behaviors and poorer quality of life, particularly among individuals with inadequate psychosocial support or poor treatment adherence (Zhang et al., Reference Zhang, Zhang, Zhang, Jin and Zheng2018; Lin et al., Reference Lin, Joshi, Keenan, Shepherd, Bailey, Berry, Wright, Meakin, Benson and Kim2021). Consequently, relapse remains a major barrier to sustained recovery and effective long-term management across the spectrum of SMIs (Touya et al., Reference Touya, Lawrence, Kangethe, Chrones, Evangelatos and Polson2022).
Previous studies have identified a range of sociodemographic, clinical and psychosocial factors associated with relapse among patients with SMI. These include age, sex, educational status, marital status, employment status, participation in religious activities, types psychiatric diagnosis, age at illness onset, duration of illness, substance use, comorbid medical illness, number of previous illness episodes, perceived stress, perceived stigma, social support, stressful life events, medication adherence, previous psychiatric hospitalization, number of hospital admissions, symptom severity and living conditions (Brunette et al., Reference Brunette, Drake, Xie, Mchugo and Green2006; Kazadi et al., Reference Kazadi, Moosa and Jeenah2008; Di Capite et al., Reference Di Capite, Upthegrove and Mallikarjun2018; Dunn, Reference Dunn2018; Hayes et al., Reference Hayes, Osborn, Lundin and Dalman2019; Rasmussen et al., Reference Rasmussen, Kakuhikire, Baguma, Ashaba, Cooper-Vince, Perkins, Bangsberg and Tsai2019; Camacho-Gomez and Castellvi, Reference Camacho-Gomez and Castellvi2020; Moges et al., Reference Moges, Belete, Mekonen and Menberu2021).
In Ethiopia, SMI represents a substantial clinical and public health challenge within a context of limited mental health resources and restricted access to continuous care. Although psychiatric services have expanded in recent years, they remain predominantly centralized in specialized hospitals, with limited availability of community-based rehabilitation, psychosocial interventions and structured long-term follow-up services. These service limitations may pose challenges for the sustained management of SMI and relapse prevention. Existing Ethiopian evidence has documented a considerable burden of relapse among individuals with SMI. For example, a systematic review and meta-analysis conducted in Ethiopia reported a high relapse burden among patients with schizophrenia and identified clinical and treatment-related factors associated with relapse (Birhan et al., Reference Birhan, Rtbey and Gelaw2025). However, available evidence remains fragmented, as most Ethiopian studies have focused on specific diagnostic groups or individual clinical settings rather than examining relapse across the major categories of SMI, including schizophrenia, bipolar disorder and major depressive disorder (Moges et al., Reference Moges, Belete, Mekonen and Menberu2021). Furthermore, evidence on lifetime relapse and its associated factors among patients with SMI in Northwest Ethiopia, particularly in the Gondar catchment area, remains limited. Addressing this evidence gap is essential to inform context-specific relapse prevention strategies, strengthen continuity of care and support the planning of integrated mental health services.
Methods and materials
An institution-based cross-sectional study was conducted from April 10 to May 24, 2025, at the University of Gondar Comprehensive Specialized Hospital (UoGCSH), Northwest Ethiopia. UoGCSH is a tertiary referral and teaching hospital located approximately 750 km northwest of Addis Ababa, the capital city of Ethiopia. The hospital serves an estimated seven million people across the surrounding region and provides comprehensive inpatient and outpatient healthcare services, including specialized psychiatric care (Tefera et al., Reference Tefera, Assefa, Abebe and Rauch2016; Teni et al., Reference Teni, Gebresillassie, Birru, Belachew, Tefera, Wubishet, Tekleyes and Yimer2018). The Psychiatry Department at the UoGCSH provides specialized mental healthcare through outpatient follow-up, emergency psychiatric services and inpatient care for individuals with a wide range of mental disorders. As one of the major referral centers in Northwest Ethiopia, the department delivers multidisciplinary mental health services, including psychiatric assessment, pharmacological treatment, psychosocial interventions, follow-up care and referral services for patients from the hospital’s catchment area and neighboring regions.
The department serves approximately 13,000 outpatient visits annually and has three emergency psychiatric beds, six outpatient psychiatric clinics and four inpatient wards with a total of 25 beds. The present study was conducted among patients attending the outpatient psychiatric clinic, where individuals with SMI receive regular follow-up and ongoing treatment. For this study, SMI was defined as schizophrenia, bipolar disorder or major depressive disorder with or without psychotic features diagnosed by consultant psychiatrists or psychiatry residents according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria (Heller et al., Reference Heller, Roccoforte, Hsieh, Cook and Pickett1997; Nesvåg et al., Reference Nesvåg, Jönsson, Bakken, Knudsen, Bjella, Reichborn-Kjennerud, Melle and Andreassen2017; Baranyi et al., Reference Baranyi, Scholl, Fazel, Patel, Priebe and Mundt2019).
Participants
The source population included all adults (≥18 years) diagnosed with SMI according to the DSM-5 who were receiving outpatient follow-up care at the UoGCSH. The study population comprised eligible patients attending the outpatient psychiatric clinic during the data collection period. Participants were included if they had received psychiatric follow-up care for at least 6 months, as this duration is generally considered sufficient for achieving symptomatic remission following initiation of treatment (Fikreyesus et al., Reference Fikreyesus, Soboka and Feyissa2016). Patients attending their first follow-up visit and those who were critically ill or unable to participate in the interview were excluded.
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki (1975), as revised in 2013 (Shrestha and Dunn, Reference Shrestha and Dunn2019). Ethical approval was obtained from the Institutional Review Board of the University of Gondar, College of Medicine and Health Sciences (Ref. No. CMHSSH-UOG SOM/382/2025; approved on March 7, 2025). Written informed consent was obtained from all participants before data collection after providing information about the study objectives, procedures, potential risks and benefits and participants’ rights. Participation was voluntary, and participants were informed that they could decline to participate or withdraw from the study at any time without affecting their access to care. To ensure privacy and confidentiality, all data were anonymized, stored securely and used solely for research purposes.
The sample size was determined using the single population proportion formula, assuming a 95% confidence level, a 4% margin of error and an estimated relapse prevalence of 70% from a previous study (Agenagnew and kassaw, Reference Agenagnew and kassaw2020). After adding a 10% nonresponse rate, the final sample size was 551 participants. Participants were selected using a systematic random sampling technique from the outpatient psychiatric clinic. The sample was proportionally allocated to the three diagnostic groups based on their average patient flow during the study period. The sampling interval (K) was calculated by dividing the total number of eligible patients attending follow-up during the data collection period by the required sample size, resulting in an interval of three. The first participant was selected using a simple random (lottery) method, and thereafter, every third eligible patient was enrolled until the required sample size was achieved (Figure 1).
Sampling procedure and participant selection process for assessing lifetime relapse and associated factors among patients with severe mental illness at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025.
Figure 1. Long description
The flowchart begins at the top with a box labeled Severe mental illness with a value of 1693.
An arrow points down to a horizontal line that branches into three categories
* Schizophrenia = 856
* M D D = 459
* Bipolar disorder = 378
Arrows from these three boxes point down to a wide horizontal box labeled Proportional allocation.
From the Proportional allocation box, three arrows point down to individual blue ovals containing the allocated sample sizes
* 275 under Schizophrenia
* 167 under M D D
* 109 under Bipolar disorder
These three ovals converge with arrows pointing to a green box labeled N = 551.
A final arrow points down to a grey oval at the bottom which states Completed the assessment and included in the final analysis n = 535.
Study variables
The dependent variable was lifetime relapse of SMI (yes/no). The independent variables included sociodemographic characteristics (age, sex, marital status, religion, educational status, employment status, economic status, living arrangement and place of residence); clinical and treatment-related characteristics (psychiatric diagnosis, duration of illness, history of psychiatric hospitalization, number of previous psychiatric admissions, medication adherence, family history of mental illness and comorbid medical illness); substance use-related characteristics (current alcohol use, khat chewing and cigarette smoking) and psychosocial characteristics (perceived stress, stressful life events, perceived stigma and social support).
Data collection tools and procedures
Data were collected using a structured, interviewer-administered questionnaire developed by the investigators based on a review of the relevant literature and the study objectives. The questionnaire captured sociodemographic, clinical, psychosocial and treatment-related information. The English version of the questionnaire is provided as a Supplementary Material 1. Lifetime relapse was determined through a review of participants’ medical records and a semi-structured interview with patients and, where appropriate, their caregivers. The assessment was based on the DSM-5-TR definition of relapse (First et al., Reference First, Yousif, Clarke, Wang, Gogtay and Appelbaum2022). Participants were asked about the occurrence and number of relapse episodes during follow-up. Relapse was defined as the recurrence or worsening of psychiatric symptoms following a period of partial or complete remission, as evidenced by one or more of the following: reemergence or exacerbation of symptoms, the need for psychiatric consultation or intensified case management, modification of psychiatric medication due to clinical deterioration or admission to a psychiatric unit (Almond et al., Reference Almond, Knapp, Francois, Toumi and Brugha2004; Kazadi et al., Reference Kazadi, Moosa and Jeenah2008).
Medication adherence was assessed using the 10-item Medication Adherence Rating Scale (MARS), a self-reported instrument with good internal consistency (Cronbach’s α = 0.86) (Thompson et al., Reference Thompson, Kulkarni and Sergejew2000). Each item is scored dichotomously (yes/no), yielding a total score ranging from 0 to 10, with higher scores indicating poorer medication adherence (Hardeman et al., Reference Hardeman, Harding and Narasimhan2010). For this study, participants with total MARS scores of 0–5 were classified as having good adherence, whereas those with scores of 6–10 were classified as having poor adherence.
Social support was assessed using the 3-item Oslo Social Support Scale (OSSS-3), which yields a total score ranging from 3 to 14 (Dalgard et al., Reference Dalgard, Dowrick, Lehtinen, Vazquez-Barquero, Casey, Wilkinson, Ayuso-Mateos, Page and Dunn2006). Scores of 3–8, 9–11 and 12–14 indicate poor, moderate and strong social support, respectively. The OSSS-3 has acceptable internal consistency (α = 0.640) and has been used in Ethiopian settings (Azale et al., Reference Azale, Fekadu and Hanlon2016; Menberu et al., Reference Menberu, Mekonen, Azale, Ayano, Yimer, Getnet, Belete, Kerie and Fekadu2018; Zewdu et al., Reference Zewdu, Hanlon, Fekadu, Medhin and Teferra2019). Stressful life events experienced in the past 6 months were assessed using the List of Threatening Experiences (LTE), which includes 12 categories of adverse life events (Brugha et al., Reference Brugha, Bebbington, Tennant and Hurry1985). The LTE demonstrates good test–retest reliability (κ = 0.61–0.87), acceptable predictive validity and low internal consistency (α = 0.44). The tool has also been adapted and validated for use in Ethiopia (Fekadu et al., Reference Fekadu, Medhin, Selamu, Hailemariam, Alem, Giorgis, Breuer, Lund, Prince and Hanlon2014).
Perceived stress was assessed using the 10-item Perceived Stress Scale (PSS-10), which measures the degree to which individuals perceive situations in their lives as stressful during the preceding month. Total scores range from 0 to 40 and were categorized as low (0–13), moderate (14–26) or high (27–40) perceived stress. The PSS-10 has demonstrated good internal consistency (Cronbach’s α = 0.88) (Cohen et al., Reference Cohen, Kamarck and Mermelstein1983). A three-item screening tool was used to assess participants’ perceived stigma related to mental illness. A total score of less than one indicates the absence of perceived stigma, while a score of one or more indicates the presence of perceived stigma (Sartorius and Janca, Reference Sartorius and Janca1996).
Current and lifetime substance use were assessed using an adapted version of the World Health Organization Alcohol, Smoking and Substance Involvement Screening Test, which has demonstrated acceptable reliability (Cronbach’s α = 0.73) (Onifade, Reference Onifade2014). Participants were classified as current users if they had used alcohol, khat or tobacco within the previous 3 months, and as lifetime users if they had ever used any of these substances (Humeniuk et al., Reference Humeniuk, Ali, Babor, Farrell, Formigoni, Jittiwutikarn, De Lacerda, Ling, Marsden and Monteiro2008; Newcombe et al., Reference Newcombe, Tanielu-Stowers, McDermott, Stephen and Nosa2016). Besides, a semi-structured questionnaire was used to collect sociodemographic characteristics and some clinical factors.
The questionnaire was pretested on 5% of the sample at a facility not included in the main study, and necessary modifications were made before data collection. Data were collected through face-to-face interviews conducted by five BSc-qualified psychiatry professionals and supervised by two MSc-qualified psychiatry professionals. Prior to data collection, all data collectors and supervisors received 2 days of training on the study objectives, interview techniques, ethical principles and standardized administration of the study instruments. Throughout the data collection period, supervisors closely monitored the data collection process and reviewed completed questionnaires daily for completeness, consistency and accuracy to ensure data quality.
Data analyses
Data were checked for completeness, coded and entered into Epi-Data version 7.2.6.0 before being exported to SPSS version 25 for analysis. Descriptive statistics, including frequencies, percentages, means with standard deviations (SDs) or medians with interquartile ranges, were used to summarize the data, as appropriate. The dependent variable was lifetime relapse of SMI (yes/no). Bivariable binary logistic regression analysis was performed to examine the association between each independent variable and lifetime relapse and crude odds ratios (CORs) with 95% confidence intervals (CIs) were estimated. Variables with a p-value <0.20 in the bivariable analysis were entered into the multivariable binary logistic regression model (Hosmer Jr et al., Reference Hosmer, Lemeshow and Sturdivant2013). Adjusted odds ratios (AORs) with 95% CIs were used to quantify the strength of the associations, and statistical significance was declared at p < 0.05. The multivariable binary logistic regression model was statistically significant (Omnibus test: χ 2 = 159.58, df = 21, p < 0.001). The model explained 25.8% of the variance in lifetime relapse according to the Cox and Snell R 2 and 35.0% according to the Nagelkerke R 2. The Hosmer–Lemeshow goodness-of-fit test was not statistically significant (χ 2 = 7.13, df = 8, p = 0.523), indicating good model fit.
Result
Sociodemographic characteristics of participants
Of the 551 eligible participants, 535 completed the study, resulting in a response rate of 97.1%. The mean age of the participants was 33.0 years (SD ± 11.3; ages range from: 18 to 75 years), and 288 (53.8%) were female. Regarding marital status, 199 (37.2%) participants were single. The majority of participants were followers of the Orthodox Christian religion 440 (82.2%), while 167 (31.2%) had no formal education. More than half of the participants were unemployed 302 (56.4%), most lived with their families 390 (72.9%), and nearly two-thirds resided in urban areas 332 (62.1%) (Table 1).
Sociodemographic characteristics of patients attending the psychiatric outpatient clinic at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025 (n = 535)
Table 1. Long description
The table contains four columns: Variables, Category, Frequency, and Percentage.
* Sex: Female (288, 53.8 percent), Male (247, 46.2 percent).
* Age: 18 to 24 (109, 20.4 percent), 25 to 34 (185, 34.6 percent), 35 to 44 (128, 23.9 percent), greater than or equal to 45 (113, 21.1 percent).
* Religion: Orthodox (440, 82.2 percent), Muslim (75, 14.0 percent), Others (20, 3.8 percent).
* Marital status: Married (186, 34.8 percent), Single (199, 37.2 percent), Divorced (150, 28.0 percent).
* Educational status: No formal education (167, 31.2 percent), Primary school (123, 23.0 percent), Secondary school (137, 25.6 percent), Diploma and above (108, 20.2 percent).
* Occupational status: Employed (233, 43.6 percent), Unemployed (302, 56.4 percent).
* Current living condition: With family (390, 72.9 percent), Alone (145, 27.1 percent).
* Residency: Rural (203, 37.9 percent), Urban (332, 62.1 percent).
Clinical and psychosocial factors of participates
Among the 535 participants, 268 (50.1%) had schizophrenia and 253 (47.3%) had an illness duration of ≤5 years. Overall, 328 (61.3%) had a history of psychiatric hospitalization, including 133 (24.9%) with two or more previous admissions. More than half of the participants 301 (56.3%) had poor medication adherence, while 98 (18.3%) had a comorbid medical illness and 184 (34.4%) reported a family history of mental illness. Regarding psychosocial characteristics, 129 (24.1%) participants had experienced three or more stressful life events, 188 (35.1%) had high perceived stress, 330 (61.7%) reported perceived stigma and 290 (54.2%) had poor social support (Table 2).
Clinical and psychosocial characteristics of patients attending the psychiatric outpatient clinic at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025 (n = 535)
Table 2. Long description
The table consists of four columns: Variables, Category, Frequency, and Percentage.
* Types of diagnosis: Bipolar disorder (104, 19.4 percent), Schizophrenia (268, 51.1 percent), M D D (163, 30.5 percent).
* Previous history of hospitalization: No (249, 46.5 percent), Yes (286, 53.5 percent).
* Admission status: No admission (249, 46.5 percent), Single admission (153, 28.6 percent), Two or more admission (133, 24.9 percent).
* Duration of illness: Less than 5 years (250, 46.7 percent), 5 to 10 years (192, 35.9 percent), Greater than 10 years (93, 17.4 percent).
* Medication adherence: Good adherence (234, 43.7 percent), Poor adherence (301, 56.3 percent).
* Comorbid medical illness: No (437, 81.7 percent), Yes (98, 18.3 percent).
* Family history of mental illness: No (357, 65.6 percent), Yes (184, 34.4 percent).
* Stressful life event: Absence (228, 42.6 percent), One or two (168, 31.4 percent), Three or more (139, 26.0 percent).
* Social support: Poor social support (290, 54.2 percent), Moderate social support (173, 32.3 percent), Strong social support (72, 13.5 percent).
* Perceived stress: High perceived stress (188, 35.1 percent), Moderate perceived stress (161, 30.1 percent), Low perceived stress (186, 34.8 percent).
* Perceived stigma: No (205, 38.3 percent), Yes (330, 61.7 percent).
Substance use history of the participants
Overall, 198 (37.0%) participants reported lifetime substance use, while 166 (31.0%) were current substance users. Among current substance users, 134 (25.0%) reported alcohol use, 62 (11.6%) reported khat use and 47 (8.8%) reported cigarette smoking (Figure 2).
Patterns of substance use among patients with severe mental illness attending the psychiatric outpatient clinic at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025 (n = 535).
Figure 2. Long description
The y-axis is labeled Substance use prevalence (%) with a scale from 0 to 35 in increments of 5. The x-axis is labeled Substance type. A legend on the right indicates that blue bars represent Ever use and orange bars represent Current use.
* Khat chewing. Ever use is 12.7 percent and current use is 11.6 percent.
* Alcohol use. Ever use is 29.3 percent and current use is 25 percent.
* Cigarette smoking. Ever use is 9.3 percent and current use is 8.8 percent.
Across all three categories, the prevalence of ever use is slightly higher than current use, with alcohol being the most prevalent substance and cigarette smoking the least.
The lifetime prevalence of relapse
The lifetime prevalence of relapse among patients with SMI was 61.5% (329/535; 95% CI: 57.0%–66.0%). Among participants who experienced relapse, 232 (43.4%) had one relapse episode, 60 (11.2%) had 2 episodes and 37 (6.9%) had three or more episodes (Figure 3). Lifetime relapse prevalence varied significantly across diagnostic groups, occurring in 75.0% (78/104) of participants with bipolar disorder, 69.0% (185/268) of those with schizophrenia and 40.5% (66/163) of those with major depressive disorder (Table 3).
Prevalence of lifetime relapse among patients with severe mental illness attending the psychiatric outpatient clinic at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025 (n = 535).
Comparison of sociodemographic, clinical and psychosocial characteristics among patients with schizophrenia, bipolar disorder and major depressive disorder attending the psychiatric outpatient clinic at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025 (n = 535)
Table 3. Long description
The table compares three patient groups across 18 variables.
Key findings include:
* Sex: Schizophrenia has the highest male percentage (51.1 percent), while Major depressive disorder (M D D) has the highest female percentage (60.7 percent).
* Occupational status: Employment is significantly higher in M D D (55.2 percent) compared to Schizophrenia (36.9 percent), with a p-value of 0.001.
* History of hospitalization: Significantly higher in Bipolar disorder (62.5 percent) and Schizophrenia (59.0 percent) than in M D D (38.7 percent), with a p-value less than 0.001.
* Admission status: 51.0 percent of Bipolar patients had two or more admissions, compared to only 11.7 percent of M D D patients.
* Comorbid medical illness: Highest in M D D (23.9 percent) and lowest in Bipolar disorder (7.7 percent), p-value 0.003.
* Current substance use: Most prevalent in Schizophrenia (36.9 percent) and least in M D D (20.9 percent), p-value 0.002.
* Perceived stress: High stress is most common in Schizophrenia (44.4 percent) compared to Bipolar (25.0 percent) and M D D (26.4 percent), p-value less than 0.001.
* Lifetime relapse: Reported by 75.0 percent of Bipolar and 69.0 percent of Schizophrenia patients, but only 40.5 percent of M D D patients, p-value 0.000.
Other variables like Residency (p-value 0.975), Educational status (p-value 0.374), and Social support (p-value 0.317) showed no statistically significant differences between the three groups.
Factors associated with lifetime relapse
Variables with a p-value <0.20 in the bivariable logistic regression analysis, including marital status, occupational status, psychiatric diagnosis, history of psychiatric hospitalization, admission status, illness duration, medication adherence, comorbid medical illness, family history of mental illness, current substance use, perceived stigma, perceived stress, stressful life events and social support, were entered into the multivariable logistic regression model. In multivariable binary logistic regression analysis, employment status, psychiatric diagnosis, history of psychiatric hospitalization, two or more prior admissions, illness duration, comorbid medical illness and stressful life events were significantly associated with lifetime relapse.
Unemployed participants had 1.65 times higher odds of lifetime relapse than employed participants (AOR = 1.65; 95% CI: 1.07–2.55; p = 0.023). Psychiatric diagnosis was also significantly associated with lifetime relapse. Compared with participants diagnosed with major depressive disorder, those with bipolar disorder had 3.71 times higher odds of lifetime relapse (AOR = 3.71; 95% CI: 1.92–7.19; p < 0.001), while those with schizophrenia had 2.73 times higher odds (AOR = 2.73; 95% CI: 1.68–4.46; p < 0.001). Participants with a history of psychiatric hospitalization had 3.62 times higher odds of lifetime relapse than those without a previous hospitalization (AOR = 3.62; 95% CI: 2.26–5.80; p < 0.001). Participants with two or more previous psychiatric admissions had 2.32 times higher odds of relapse than those with no previous admissions (AOR = 2.32; 95% CI: 1.27–4.23; p = 0.006). Likewise, participants with an illness duration of more than 10 years had 2.04 times higher odds of relapse than those with an illness duration of less than 5 years (AOR = 2.04; 95% CI: 1.07–3.88; p = 0.030). Furthermore, participants with a comorbid medical illness had 2.42 times higher odds of relapse than those without a comorbid medical illness (AOR = 2.42; 95% CI: 1.27–4.58; p = 0.007). Regarding psychosocial factors, participants who experienced one or two stressful life events had 1.81 times higher odds of lifetime relapse than those who reported no stressful life events (AOR = 1.81; 95% CI: 1.03–3.18; p = 0.038). Similarly, participants who experienced three or more stressful life events had 2.16 times higher odds of lifetime relapse (AOR = 2.16; 95% CI: 1.16–4.05; p = 0.016) (Table 4).
Factors associated with lifetime relapse among patients with severe mental illness attending the psychiatric outpatient clinic at the University of Gondar Comprehensive Specialized Hospital, Northwest Ethiopia, 2025 (n = 535)
Table 4. Long description
A table with nine columns: Explanatory variables, Category, Relapse (Yes/No), B, df, C O R, A O R, and p-value. Significant factors (p < 0.05) identified via multivariable binary logistic regression include:
* Occupational status: Unemployed patients have an A O R of 1.65 (1.07, 2.55), p = 0.023.
* Types of diagnosis: Bipolar disorder (A O R 3.71, p < 0.001) and Psychosis (A O R 2.73, p < 0.001) compared to Depression.
* History of hospitalization: Yes (A O R 3.62, p < 0.001).
* Admission status: Two or more admissions (A O R 2.32, p = 0.006).
* Duration of illness: Greater than 10 years (A O R 2.04, p = 0.030).
* Comorbid medical illness: Yes (A O R 2.41, p = 0.007).
* Stressful life events: One or two events (A O R 1.81, p = 0.038) and three or more events (A O R 2.16, p = 0.016).
Other variables like Marital status, Medication adherence, Family history, Current substance use, Perceived stigma, Perceived stress, and Social support did not show statistically significant adjusted associations with relapse in this model.
Note: Ref, reference category; β, regression coefficient; df, degrees of freedom; COR, crude odds ratio; AOR, adjusted odds ratio; CI, confidence interval. p < 0.05; p < 0.01. The multivariable binary logistic regression model demonstrated adequate fit (Hosmer–Lemeshow goodness-of-fit test: χ 2 = 7.13, df = 8, p = 0.523).
Discussion
The present study assessed the lifetime prevalence of relapse and factors associated with lifetime relapse among adults with SMI attending outpatient follow-up in Northwest Ethiopia. Overall, 61.5% (95% CI: 57.0%–66.0%) of participants had experienced at least one lifetime relapse. In the multivariable analysis, unemployment, bipolar disorder, schizophrenia, previous psychiatric hospitalization, illness duration greater than 10 years, two or more previous psychiatric admissions, comorbid medical illness and stressful life events were independently associated with lifetime relapse. These findings indicate that lifetime relapse remains a substantial clinical and public health challenge among people living with SMI and highlight several factors that may help identify individuals at increased risk of relapse and guide targeted relapse-prevention strategies. The lifetime prevalence of relapse observed in this study is consistent with previous studies conducted in Southwest Ethiopia (57.4%) (Fikreyesus et al., Reference Fikreyesus, Soboka and Feyissa2016), Johannesburg, South Africa (61.8%) (Kazadi et al., Reference Kazadi, Moosa and Jeenah2008) and a recent systematic review and meta-analysis conducted in Africa, which reported a pooled relapse prevalence of 60.66% (Birhan et al., Reference Birhan, Rtbey and Gelaw2025). This consistency may reflect comparable mental health service settings, similar socioeconomic contexts and shared challenges related to the long-term management of SMI in sub-Saharan Africa (Patel et al., Reference Patel, Saxena, Lund, Thornicroft, Baingana, Bolton, Chisholm, Collins, Cooper, Eaton, Herrman, Herzallah, Huang, Jordans, Kleinman, Medina-Mora, Morgan, Niaz, Omigbodun, Prince, Rahman, Saraceno, Sarkar, de Silva, Singh, Stein, Sunkel and UnÜtzer2018; Organization, 2022a).
The prevalence observed in the present study was higher than that reported in Nigeria 41% (Adebiyi et al., Reference Adebiyi, Mosaku, Irinoye and Oyelade2018), South-East Nigeria 52.1% (Eneh and Eneh, Reference Eneh and Eneh2024), Spain 44% (Bernardo et al., Reference Bernardo, Amoretti, Cuesta, Parellada, Mezquida, González-Pinto, Bergé, Lobo, Aguilar and Usall2021) and the United States 30.52% (Rivelli et al., Reference Rivelli, Fitzpatrick, Nelson, Laubmeier, Zeni and Mylavarapu2024). These differences should be interpreted cautiously, as relapse estimates may vary according to differences in study design, study populations, diagnostic composition, relapse definitions, follow-up periods and healthcare settings. In addition, variations in access to mental health services, continuity of psychiatric care, availability of specialized mental health resources, socioeconomic conditions and the distribution of clinical characteristics, such as psychiatric diagnosis, illness duration, history of hospitalization and comorbid medical illness, may have contributed to the observed differences across studies (Moges et al., Reference Moges, Belete, Mekonen and Menberu2021; Zwide et al., Reference Zwide, Dewet and Sokudela2025). Compared with high-income countries such as Spain (Bernardo et al., Reference Bernardo, Amoretti, Cuesta, Parellada, Mezquida, González-Pinto, Bergé, Lobo, Aguilar and Usall2021) and the United States (Rivelli et al., Reference Rivelli, Fitzpatrick, Nelson, Laubmeier, Zeni and Mylavarapu2024), resource limitations and reduced access to comprehensive mental health services in Ethiopia may also influence long-term management of SMI and lifetime relapse.
The prevalence of lifetime relapse observed in the present study was lower than that reported in Nairobi, Kenya (77.5%) (Mwaura, Reference Mwaura2018), and in the United States, where the relapse rate was 81.9% (Robinson et al., Reference Robinson, Woerner, Alvir, Bilder, Goldman, Geisler, Koreen, Sheitman, Chakos, Mayerhoff and Lieberman1999). These differences may reflect variations in study populations, clinical characteristics and healthcare contexts. The Kenyan study identified medication nonadherence, missed follow-up appointments and substance use as factors associated with relapse (Gathaiya, Reference Gathaiya2011; Mwaura, Reference Mwaura2018), whereas the distribution and impact of these factors may differ across settings because of differences in healthcare delivery and patient characteristics. Similarly, the higher relapse rate reported in the United States may be explained by the inclusion of patients with first-episode psychotic disorders, the longer follow-up period and differences in the operational definition of relapse, including psychiatric rehospitalization and emergency service utilization (Rivelli et al., Reference Rivelli, Fitzpatrick, Nelson, Laubmeier, Zeni and Mylavarapu2024). Variations in healthcare systems, continuity of care and clinical management may have also contributed to the observed differences (Rivelli et al., Reference Rivelli, Fitzpatrick, Nelson, Laubmeier, Zeni and Mylavarapu2024).
Moreover, the highest relapse prevalence was observed among individuals with bipolar disorder (77.8%), which is consistent with patterns reported globally (Belete et al., Reference Belete, Ali and Legas2020; Bai et al., Reference Bai, Fan, Lu, Wei and Wu2025). Bipolar disorder is characterized by a recurrent course involving repeated mood episodes, with relapse risk remaining substantial, particularly when maintenance treatment is inconsistent or medication adherence is suboptimal (Kishi et al., Reference Kishi, Matsuda, Sakuma, Okuya, Mishima and Iwata2021). Similarly, the relapse prevalence among patients with schizophrenia (69.0%) is comparable with findings from South Africa (Kazadi et al., Reference Kazadi, Moosa and Jeenah2008). The high relapse burden in schizophrenia may be related to the chronic and relapsing nature of the disorder, impaired insight and challenges with sustained medication adherence (Ayano and Duko, Reference Ayano and Duko2017).
In this study, unemployment was significantly associated with lifetime relapse. Unemployed participants had 1.65 times higher odds of lifetime relapse compared with employed participants (AOR = 1.65; 95% CI: 1.07–2.55). This finding is consistent with previous studies reporting an association between unemployment and poorer outcomes among individuals with SMI (Waghorn and Lloyd, Reference Waghorn and Lloyd2005). Unemployment may be linked to relapse through several interconnected pathways, including socioeconomic disadvantage, reduced social functioning, financial stress and limited opportunities for structured daily activities (Kate et al., Reference Kate, Grover, Kulhara and Nehra2013). Previous evidence suggests that socioeconomic difficulties may affect continuity of care and access to treatment resources, which may influence illness outcomes (Luciano and Meara, Reference Luciano and Meara2014). In addition, employment provides routine, social engagement and a sense of purpose, which may support psychological well-being and recovery among individuals with SMI (Garety et al., Reference Garety, Kuipers, Fowler, Freeman and Bebbington2001; Carmona et al., Reference Carmona, Gomez-Benito, Huedo-Medina and Rojo2017).
Participants diagnosed with bipolar disorder and schizophrenia had significantly higher odds of lifetime relapse compared with those diagnosed with major depressive disorder. Individuals with bipolar disorder had 3.71 times higher odds of lifetime relapse, while those with schizophrenia had 2.73 times higher odds. These findings are consistent with previous evidence indicating that bipolar disorder is characterized by a recurrent course, with repeated episodes occurring over the illness trajectory (Bai et al., Reference Bai, Fan, Lu, Wei and Wu2025). In Ethiopia, a previous study reported a high relapse prevalence among patients with bipolar disorder, with longer illness duration (>5 years) associated with increased odds of relapse (Kebede et al., Reference Kebede, Alem, Shibire, Deyassa, Negash, Beyero, Medhin and Fekadu2006; Belete et al., Reference Belete, Ali and Legas2020). The higher odds of relapse observed among individuals with schizophrenia may be related to the chronic and recurrent nature of the disorder, which is often associated with persistent symptoms, functional impairment and vulnerability to clinical deterioration over time. Previous studies have identified several factors associated with relapse in schizophrenia, including medication nonadherence, illness severity and clinical characteristics (Emsley et al., Reference Emsley, Chiliza, Asmal and Harvey2013; Olivares et al., Reference Olivares, Sermon, Hemels and Schreiner2013; Smit et al., Reference Smit, Luckhoff, Phahladira, Du Plessis, Emsley and Asmal2024). Additionally, impaired insight, particularly during acute psychotic or manic episodes, may affect early recognition of symptom worsening and timely treatment-seeking (Leucht et al., Reference Leucht, Tardy, Komossa, Heres, Kissling, Salanti and Davis2012). These findings highlight the importance of diagnosis-specific relapse prevention strategies and closer monitoring of individuals with bipolar disorder and schizophrenia.
Participants with a history of previous psychiatric hospitalization had 3.62 times higher odds of lifetime relapse than those without a history of hospitalization. This finding is consistent with previous studies reporting a significant association between prior psychiatric hospitalization and relapse or rehospitalization among individuals with SMI (Olivares et al., Reference Olivares, Sermon, Hemels and Schreiner2013; San et al., Reference San, Bernardo, Gómez and Peña2013; Jørgensen et al., Reference Jørgensen, Bøg, Kabra, Simonsen, Adair and Jönsson2021). Previous psychiatric hospitalization may reflect greater illness severity, a more recurrent clinical course or the occurrence of acute episodes requiring intensive treatment. However, because hospitalization and relapse are closely related clinical events, this association should be interpreted within the context of the study design (Jørgensen et al., Reference Jørgensen, Bøg, Kabra, Simonsen, Adair and Jönsson2021). Nevertheless, a history of psychiatric hospitalization may help identify individuals who require closer follow-up, continuity of care and relapse-prevention interventions.
Participants with two or more previous psychiatric admissions had 2.32 times higher odds of lifetime relapse than those with no previous psychiatric admissions (AOR = 2.32; 95% CI: 1.27–4.23). This finding is consistent with previous studies from Spain, South Africa and other settings that reported a significant association between repeated psychiatric admissions and relapse among individuals with SMI (Kebede et al., Reference Kebede, Alem, Shibire, Deyassa, Negash, Beyero, Medhin and Fekadu2006; Kazadi et al., Reference Kazadi, Moosa and Jeenah2008; Vigod et al., Reference Vigod, Kurdyak, Seitz, Herrmann, Fung, Lin, Perlman, Taylor, Rochon and Gruneir2015). Recurrent psychiatric admissions may reflect a more severe or recurrent illness course, greater symptom burden, treatment resistance and challenges in achieving sustained remission and continuity of care (Olfson et al., Reference Olfson, Marcus, Wilk and West2006; Patel et al., Reference Patel, Saxena, Lund, Thornicroft, Baingana, Bolton, Chisholm, Collins, Cooper, Eaton, Herrman, Herzallah, Huang, Jordans, Kleinman, Medina-Mora, Morgan, Niaz, Omigbodun, Prince, Rahman, Saraceno, Sarkar, de Silva, Singh, Stein, Sunkel and UnÜtzer2018). Because repeated admissions are closely related to relapse, patients with a history of multiple hospitalizations may require closer outpatient follow-up, adherence support, psychoeducation, family involvement and other relapse-prevention interventions following discharge (Rylander et al., Reference Rylander, Colon-Sanchez, Keniston, Hamalian, Lozano and Nussbaum2016).
In this study, a longer duration of illness was independently associated with an increased likelihood of relapse. Participants with an illness duration of more than 10 years had 2.04 times higher odds of relapse than those with less than 5 years. This finding is consistent with previous studies identifying prolonged illness duration as a predictor of relapse among individuals with SMI (Olivares et al., Reference Olivares, Sermon, Hemels and Schreiner2013; Bioque et al., Reference Bioque, Mezquida, Amoretti, García-Rizo, López-Ilundain, Diaz-Caneja, Zorrilla, Mané, Rodriguez-Jimenez, Corripio, Pomarol-Clotet, Ibáñez, Usall, Contreras, Mas, Vázquez-Bourgon, Cuesta, Parellada, González-Pinto, Hidalgo-Figueroa and Bernardo2022). Prolonged illness is often accompanied by cumulative psychosocial adversity, including impaired social and occupational functioning, social isolation and diminished coping capacity, which may increase vulnerability to relapse (San et al., Reference San, Bernardo, Gómez and Peña2013). In addition, a longer illness course is frequently marked by recurrent episodes and repeated psychiatric hospitalizations, reflecting greater illness severity and persistent challenges in achieving sustained remission. A large cohort study similarly demonstrated that relapse risk increases with the number of prior psychiatric hospitalizations, indicating a progressively less favorable illness trajectory (Andreasen et al., Reference Andreasen, Liu, Ziebell, Vora and Ho2013). In low- and middle-income settings such as Ethiopia, these effects may be compounded by limited access to mental health services, inadequate continuity of care and declining social support, further increasing the risk of relapse (Moges et al., Reference Moges, Belete, Mekonen and Menberu2021).
Participants with a comorbid medical illness had 2.41 times higher odds of lifetime relapse than those without a comorbid medical condition. This finding is consistent with previous studies suggesting that physical comorbidities are associated with poorer psychiatric outcomes and an increased likelihood of relapse (Šprah et al., Reference Šprah, Dernovšek, Wahlbeck and Haaramo2017). Several mechanisms may explain this association. Chronic medical conditions may increase treatment complexity, psychological stress and functional impairment, which can adversely affect the management of SMI. They may also complicate pharmacological treatment through potential drug–drug interactions, increased medication burden and reduced adherence to psychotropic medications. Furthermore, persistent physical illness may contribute to ongoing psychological distress and reduced quality of life, factors that have been associated with unfavorable psychiatric outcomes (Chiang et al., Reference Chiang, Chang, Liu and Tzeng2021). Evidence from patients with depression and type 2 diabetes also indicates that greater medical disease burden, reflected by polypharmacy, is associated with a substantially higher likelihood of relapse (Jeffery et al., Reference Jeffery, Bhanu, Walters, Wong, Osborn and Hayes2023).
Participants who experienced stressful life events had significantly higher odds of lifetime relapse. Specifically, participants reporting moderate stressful life events had 1.81 times higher odds of lifetime relapse, whereas those reporting high levels of stressful life events had 2.16 times higher odds compared with participants who reported no stressful life events. This finding is consistent with previous studies demonstrating an association between stressful life events and relapse among individuals with SMI (Samuel et al., Reference Samuel, Nigussie, Mirkena and Azale2022; Nshimiyimana, Reference Nshimiyimana2023). Stressful life events may adversely affect psychological well-being by overwhelming coping resources, disrupting social and occupational functioning and complicating illness management (Brown and Harris, Reference Brown and Harris2012). In low- and middle-income settings such as Ethiopia, where socioeconomic adversity and limited access to mental health services remain common, the impact of psychosocial stressors may be greater. A recent meta-analysis also reported that stressful life events were associated with an increased likelihood of relapse among individuals with schizophrenia, particularly in the presence of limited social support (Norman and Malla, Reference Norman and Malla1993).
Strengths and limitations of the study
This study has several strengths. It is among the few studies in Ethiopia to assess lifetime relapse across schizophrenia, bipolar disorder and major depressive disorder within a single population. The high response rate, systematic random sampling and use of both interviews and medical records for relapse assessment enhanced the validity of the findings. Additionally, standardized validated instruments improved the reliability of measurements.
Despite its strengths, this study has several limitations. First, the cross-sectional design limits the ability to establish temporal relationships between lifetime relapse and associated factors. Second, lifetime relapse was assessed retrospectively through participant interviews supported by medical record review, which may have been affected by recall bias and incomplete documentation of previous clinical episodes. Third, although relapse was assessed using an operational definition based on DSM-5-TR criteria, the retrospective nature of the assessment and variations in the documentation of remission periods and relapse episodes may have introduced potential outcome misclassification. Fourth, the study was conducted at a single tertiary referral hospital, and therefore, the findings may not be fully generalizable to individuals with SMI receiving care in other healthcare settings or community contexts. Finally, because the study included only patients who remained engaged in outpatient psychiatric follow-up, individuals who discontinued treatment, achieved sustained recovery or died before assessment were not captured, which may have introduced survivorship bias.
Conclusion
Lifetime relapse among patients with SMI was common, affecting nearly two-thirds of the study population. Factors significantly associated with relapse included unemployment, psychiatric diagnosis, previous psychiatric hospitalization, multiple prior admissions, longer illness duration, comorbid medical illness and stressful life events. These findings underscore the importance of continuous, integrated and patient-centered mental healthcare for individuals with SMI. Routine psychiatric follow-up should consider individuals with these clinical and psychosocial characteristics for closer monitoring and early identification of relapse. Strengthening continuity of care, improving the management of comorbid medical conditions and systematically assessing and addressing psychosocial stressors may contribute to reducing the burden of relapse and improving long-term outcomes among people with SMI.
Abbreviations
- AOR
-
Adjusted Odds Ratio,
- CI
-
Confidence Interval,
- COR
-
Crude Odds Ratio,
- DSM
-
Diagnostic Stastical Manual,
- MARS
-
Medication Adherence Rating Scale,
- MDD
-
Major Depressive Disorder,
- OSS-3
-
Oslo Social Support Scale,
- PSS
-
Perceived Stress Scale,
- SMI
-
Severe Mental Illnesses,
- UoGCSH
-
University of Gondar Comprehensive Specialized Hospital,
- US
-
United States
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/gmh.2026.10318.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/gmh.2026.10318.
Acknowledgments
We would like to acknowledge University of Gondar for all forms of nonfinancial support provided for the study. Furthermore, we also extend our gratitude to data collectors, supervisors and study participants for their time and effort.
Author contribution
ZY carried out the manuscript from its conception, analysis, interpretation of data, drafted the manuscript and review of the manuscript. YW, GK, DD, HT, MA, DG and MK revised the proposal, checked the data analysis and revised the manuscript. All authors read and approved the final draft of the manuscript.
Competing interests
The authors declare that they have no conflict of interests.
Data availability statement
The data for this study are available from the corresponding author (Principal investigator) on reasonable request.
Ethics and approvals
Ethical clearance was obtained from the ethical review board of University of Gondar, collage of medicine and health science. Participation in the study was entirely voluntary, and individuals who were unwilling to participate were not coerced. Written informed consent was obtained from all participants after they were provided with detailed information regarding the study’s purpose and expected outcomes. All data collected were handled with strict confidentiality and stored securely to protect participant privacy.
Funding statement
No funding.
Clinical trial number
Not applicable.
Consent for publication
Not applicable.
Open access
