Syndemics of depression, sick role and activation status among newly diagnosed adults with diabetes mellitus in Bahir Dar, Ethiopia: a 6-month follow-up study protocol

Abstract

Introduction

Diabetes mellitus (DM) and depression commonly coexist. Each condition increases the risk of developing the other and adversely affects treatment outcomes. Such complex interactions of diseases, referred to as syndemics, have not been well studied. This study aims to assess the syndemics of depression, sick role and activation status among newly diagnosed adults living with DM.

Methods and analysis

A prospective 6-month follow-up study will be conducted with 485 participants. Depression will be assessed with the 9-item Patient Health Questionnaire, applying a cut-off score of 10. The primary outcome will be glycaemic control, and the secondary outcomes will be health-related quality of life (HRQoL) and functional disability status. Depression, the primary outcome and the secondary outcomes, will be measured at baseline, 3 months and 6 months. The sick role, activation status and health system perspectives will be explored using qualitative methods following the second measurement. Data will be collected from adults living with DM, healthcare providers and healthcare managers. Qualitative sampling will continue until data saturation is reached.

Quantitative analysis will be done using STATA V.17. The prevalence of depression will be determined at baseline. Associated factors will be analysed using Poisson regression with a robust variance estimator. Incidence rate of depression, glycaemic control, HRQoL and disability status will be measured at 3 and 6 months. A multilevel mixed-effects generalised linear model will be fitted, with the three measurement time points nested within individuals, and individuals nested within health institutions. Qualitative data will be analysed thematically using NVivo V.12 software.

Ethics and dissemination

Ethics approval has been granted by the institutional review board of Bahir Dar University (protocol number 3098/25). Findings will be disseminated through peer-reviewed publications, conference presentations and local channels for community audiences.

Trial registration number

Protocol number 3098/25.

STRENGTHS AND LIMITATIONS OF THE STUDY.

• The study will employ data collection tools that have been previously adapted and extensively used in the study area.

• Participants will be recruited prospectively, which minimises data incompleteness.

• The study design allows identification of new depression cases occurring after diabetes mellitus diagnosis, resolving the egg and hen dilemma.

• The study will be conducted in urban healthcare facilities, which may limit the generalisability of the findings to rural settings.

• The study may face challenges related to loss to follow-up.

Introduction

Diabetes mellitus (DM) is one of the four leading non-communicable diseases (NCDs), along with cardiovascular disease, cancer and chronic respiratory disease. Globally, NCDs account for 41 million deaths each year, of which 77% occur in low- and middle-income countries.^{1 2} DM has a long history, with descriptions of the condition recorded as early as the 5th century BC.3,8 Changes in human behaviour and lifestyle have led to a significant increase in DM cases, making it a major public health threat over the past decade.^{9 10} The prevalence of DM has risen more rapidly in low- and middle-income countries than in high-income countries.¹¹ Studies have also reported an increasing prevalence of DM and its associated complications in Ethiopia.12,14 In Ethiopia, the prevalence of DM ranges from 2.0% to 6.5%, with lower rates observed in smaller rural areas.¹³ In Bahir Dar, DM was the second most commonly reported NCD (45%), following hypertension (63.5%), among patients attending chronic outpatient medical care.¹⁵

Individuals with DM are more likely to experience depression than those without the condition.^{16 17} Depression is also linked to an increased risk of developing DM.¹⁸ This co-occurrence of diseases is referred to as syndemics. Multiple health issues interact both biologically and with the environmental factors, resulting in worsened health outcomes for affected populations.19,22 Living with both DM and depression presents serious clinical challenges. It can reduce quality of life (QoL), make self-management more difficult, increase the risk of complications and ultimately shorten life expectancy.²³

Preventing DM-related complications, reducing mortality and improving clinical outcomes and QoL require effective self-care and regular monitoring of blood glucose levels.^{24 25} Individuals living with DM require more than medical support from healthcare providers; it is equally important for them to self-manage and adhere to complex self-care routines.²⁶ Many adults living with DM experience difficulties in managing their condition without sustained support.²⁷ Evidence suggests that patient activation and sick role behaviour are pivotal for the adoption of self-care practices. Improved patient activation, defined as an individual’s knowledge, skills and confidence in managing their health and healthcare,²⁸ is associated with better DM self-care and glycaemic control.29,31 The concept of the sick role, first introduced by Parsons,³² refers to the role assigned to a patient by society. Patients are granted certain rights, such as exemption from work and family responsibilities.³² However, due to the nature of the disease, individuals living with DM are expected to take an active role in managing their care, including monitoring glycaemic levels, educating others and advocating for their needs.³³ The sick role among individuals with DM is defined as a role in which they acknowledge their illness and take active, continuing responsibility for managing their condition through lifestyle modifications, adherence to treatment and regular engagement with healthcare providers.³⁴

Although the prevalence of depression among individuals living with DM is reported to be high (34.61%, 95% CI: 27.31 to 41.91),³⁵ it remains unclear whether this reflects a high incidence, the chronic nature of the disease or both. There is a scarcity of information in the study area regarding the course of DM when depression co-occurs and on sick role behaviour and activation status of individuals living with DM. Furthermore, the perspectives of healthcare professionals and managers on adapting chronic care to the needs of individuals living with DM remain unexplored. Therefore, this study aims to fill existing gaps and to propose a model of long-term care that meets the complex needs of adults living with DM.

Research questions

1. What is the prevalence of depression and associated factors among newly diagnosed adults living with DM at the time of diagnosis?

2. What is the incidence rate of depression at 3 and 6 months after diagnosis among adults living with DM?

3. What is the outcome of depression on the progression and outcomes of DM 6 months after diagnosis?

4. What are the experiences related to the sick role and patient activation status among adults living with DM?

5. What is the health system’s perspective on the changes required to meet the long-term care needs of adults living with DM?

Objectives of the study

General objective

The overall aim of the study is to assess syndemics of depression, sick role and patient activation status among newly diagnosed adults living with DM in Bahir Dar, Ethiopia.

Specific objectives

1. To determine the prevalence rate of depression and associated factors among newly diagnosed adults living with DM.

2. To determine the incidence rate of depression at 3 and 6 months after diagnosis among adults living with DM.

3. To examine the outcome of depression on the primary outcome (glycaemic control) and secondary outcomes (health-related QoL (HRQoL) and functional disability status) at 6 months after diagnosis.

4. To explore the sick role and patient activation status of adults living with DM.

5. To explore health system perspectives on the changes required to meet the long-term care needs of adults living with DM.

Conceptual framework

Depression among adults living with DM is affected by various factors, including sociodemographic variables, substance use, perceived social support (PSS) and the presence of chronic comorbidities. Perceived pain and adherence to diabetes self-care practices (taking medications as prescribed, maintaining a healthy diet and engaging in regular physical activity) are the key factors that significantly influence depression in this population. Individuals with DM and depression are more likely to experience reduced HRQoL, poorer health outcomes and an increased risk of disability. The prevalence of depression will be assessed using baseline data, while the incidence of depression and its outcomes (primary and secondary) will be evaluated at 3 and 6 months of follow-up after diagnosis (figure 1).

Figure 1. Conceptual framework illustrating how independent factors are associated with depression and how depression may subsequently affect glycaemic control, health-related quality of life and functional disability among individuals with diabetes mellitus (DM). Red arrows denote the timing for measuring the prevalence and incidence of depression. The green arrow represents the 6-month follow-up period, beginning from baseline data collection, during which the prevalence of depression will be determined, the incidence of depression will be measured at 3 and 6 months and depression outcomes will be assessed at 6 months. The blue solid lines indicate the directional relationship of independent variables with the primary and secondary outcomes, and the blue dashed lines show the relationships between independent variables.

Methods and material

Study area and settings

This study is part of the Bahir Dar University Long Care Model (BDU-LCM) study, which was originally funded by BDU. Within the BDU-LCM study, NCDs, such as DM, cancer and hypertension, as well as their co-occurrence with depression, will be investigated. The study will be conducted among newly diagnosed adults living with DM attending public healthcare facilities in Bahir Dar city. Bahir Dar, the capital of the Amhara region, is situated 565 km northwest of Addis Ababa. It is one of the country’s leading tourist destinations, offering a variety of attractions near Lake Tana and the ‘Abay’ (Blue Nile) River.³⁶ The study will be conducted from July 2025 and is expected to conclude in June 2026.

Study design

The study comprises two approaches: (1) a quantitative study with a 6-month follow-up and (2) a qualitative study employing a phenomenological design.

Quantitative: 6-month follow-up study

A 6-month follow-up study will be conducted to address the first three specific objectives of this research. The first specific objective—prevalence of depression—will be determined using baseline data. The second objective—incidence of depression—will be assessed at the 3-month and 6-month follow-ups. The third objective—the effect of depression on the primary outcome (glycaemic control) and secondary outcomes (HRQoL and functional disability status)—will be analysed at the 6-month follow-up.

Inclusion criteria

• All adults aged ≥18 years, who are newly diagnosed with DM and willing to provide informed consent, will be eligible to participate in the study.

Exclusion criteria

• Newly diagnosed adults living with DM who plan to transfer to healthcare facilities outside Bahir Dar, those with verbal communication difficulties, seriously ill patients requiring admission at the time of diagnosis and pregnant women will be excluded from the study.

Sample size determination

In this study area, the impact of depression on DM treatment outcomes has not yet been explored. However, previous research has examined its effects among patients with tuberculosis (TB).³⁷ As both TB and DM are chronic conditions associated with a high prevalence of depression and reduced QoL, the sample size for this study was determined based on the observed effect of depression on TB treatment outcomes. The sample size was calculated using the StatCalc tool in Epi Info, V.7.2.6.0, with a 95% CI, 80% power and proportions of unsuccessful treatment outcomes of 12.9% among patients with TB and depression and 3.4% among those without depression,³⁷ using an unexposed-to-exposed ratio of 4. The calculated sample size was 404. Adding a 20% contingency for loss to follow-up, the final sample size was increased to 485.

Variables and measurement

Exposure variable

The exposure variable is depression, which will be assessed using the 9-item Patient Health Questionnaire (PHQ-9). This scale has been widely used worldwide in surveys, effectiveness trials and cohort studies across diverse populations. Each PHQ-9 item is rated on a scale of 0–3, yielding a total score ranging from 0 to 27.³⁸ The PHQ-9 has been tested and validated in Ethiopia, demonstrating good reliability and validity. A cut-off score of 10 has been shown to provide optimal accuracy for detecting major depressive disorder in clinical samples in Ethiopia.³⁹ A similar cut-off point has also been used in previous studies to assess depression, including among patients with TB ⁴⁰ and individuals attending chronic outpatient medical care.¹⁵

Outcome variables (dependent variable)

1. Primary outcome: the primary outcome is glycaemic control. It will be assessed at baseline, 3 months and 6 months of follow-up. It will be measured using HbA1c testing or fasting blood glucose levels. HbA1c reflects average glycaemic levels over the preceding 3 months and is recommended by the Ethiopian Ministry of Health.^{40 41}

2. Secondary outcomes (HRQoL and functional disability status).

   • Functional disability status: it will be measured using the interviewer-administered 12-item WHO Disability Assessment Schedule, V.2.0 (WHODAS 2.0).⁴² Studies have demonstrated the usefulness of WHODAS 2.0 for assessing functional disability among patients with depression in primary healthcare settings,⁴³ as well as its ability to capture changes over time.⁴⁴ The tool has been adapted for use in Ethiopia and has shown strong convergent validity in individuals with depression.^{37 45}

   • HRQoL: it will be measured using a single-item question: ‘How would you rate your HRQoL?’ Responses will be rated on a scale from 0 (worst possible) to 10 (best possible).^{37 46} This single-item measure has been successfully used in population surveys, clinical settings and interviews^{47 48} and has been shown to validly predict vulnerability to all-cause mortality.⁴⁹ This scale has previously been used in the study area.³⁷

Independent variables

Studies reported that there were various factors associated with depression among individuals living with DM. These include sociodemographic characteristics,^{35 50 51} perceived pain,^{52 53} substance use,^{50 54} presence of comorbidities,⁵¹ PSS,^{35 50 54 55} DM-related stigma⁵⁶ and adherence to medications.^{54 57} In this study, the measurement of these factors is described in detail below.

• Sociodemographic variables: data will be collected at baseline using an interviewer-administered structured questionnaire and will include age, sex, marital status, level of education, religion, household income, occupation and place of residence (urban vs rural).

• Perceived pain: it will be measured using a single-item question: ‘How is your pain this week?’ Responses will be scored from 0 (absence of symptom) to 10 (worst level of symptom severity), as perceived by the respondent for that week. Scores will be categorised as no symptom (0), mild (1–3), moderate (4–6) and severe (7–10), using a visual analogue scale to indicate intensity.⁵⁸ This method has also been applied in a study conducted in Ethiopia.³⁷

• Substance use: alcohol, tobacco and khat use will be assessed using the Alcohol, Smoking and Substance Involvement Screening Test, V.3.1 (ASSIST 3.1).⁵⁹ The ASSIST risk score ranges from 0 to 31 for tobacco and from 0 to 39 for alcohol and khat. In Ethiopia, the ASSIST V.3.1 tool was translated and successfully used in the study area. The cut-off risk score values for low, moderate and high risk for smoking and khat use are 0–3, 4–26 and above 26, respectively. For alcohol, the corresponding values are 0–10, 11–26 and above 26, respectively.⁵⁹ In this study, the different types of tobacco and khat will be considered equivalent. The local alcoholic drinks ‘Tella’, ‘Tej’ and ‘Araki’ will be treated as standard drinks, with alcohol content by volume of 3.84%–6.48% for Tella, 8.94%–13.16% for Tej and 33.95%–39.9% for Araki.⁶⁰

• Chronic illnesses: data on the presence of chronic diseases other than DM will be collected by asking respondents whether they have previously received a diagnosis.

• Depression treatment: if a patient is receiving antidepressant treatment, the type of medication will be identified from their medical chart and recorded using a customised form.

• PSS: refers to an individual’s expectation of whether support will be provided by their social network.⁶¹ PSS will be measured using the 3-item Oslo Scale of PSS,⁶² which has previously been used in the study area.⁶³

• DM-related stigma: it will be assessed using the Chronic Illness Anticipated Stigma Scale (CIASS). This scale measures anticipated stigma, encompassing expectations of prejudice, stereotyping and discrimination among individuals living with chronic illnesses. The CIASS comprises 12 items organised into three subscales, each addressing anticipated stigma from friends and family, workplace colleagues and healthcare professionals.^{64 65} It has been translated into Amharic and used in Ethiopia.⁶⁵

• Adherence to medications: adherence to medications will be measured using the Morisky, Green and Levine Adherence Scale. The total score will be calculated from all four items and ranges from 0 (complete non-adherence) to 4 (complete adherence).⁶⁶

Operational definitions

1. Glycaemic control level: will be categorised as controlled or uncontrolled based on the following cut-off points.

   • Uncontrolled: defined as an HbA1c measurement of ≥7%,⁶⁷ or when the average fasting blood glucose level measured over three consecutive follow-up months falls below 70 mg/dL or above 130 mg/dL.⁶⁸

   • Controlled: defined as an HbA1c measurement below 7%, or if their average fasting blood glucose over three consecutive follow-up months falls within the range of 70–130 mg/dL.⁶⁸

2. Presence of depression: will be determined using PHQ-9 scores and classified as either ‘depressed’ or ‘not depressed’ according to cut-off points supported by previous studies in Ethiopia.^{39 69}

   • Depressed: defined as an individual scoring 10 or above on the PHQ-9 scale.

   • Not depressed: defined as an individual scoring below 10 on the PHQ-9 scale.

Data collection and management

Trained data collectors will be assigned to collect the quantitative data. Supervisors will check the completeness and consistency of the collected data, while the principal investigator will conduct further double-checks. The data will be entered into EpiData V.4.6 by trained data entry clerks. Then, it will be exported to STATA Windows V.17 for analysis. Hard copies of the data will be kept securely in a locked cabinet, and the consent forms will be stored separately from the data.

Method of analysis

Analysis will be carried out using STATA Windows V.17. Descriptive statistics will be employed to describe the study participants and to summarise both dependent and independent variables. The prevalence of depression at baseline will be determined by calculating the proportion of participants who score 10 or above on the PHQ-9 scale. The factors associated with baseline depression will be examined using both bivariable and multivariable Poisson regression with a robust variance estimator. The necessary assumptions will be checked before conducting the regression analyses. Similarly, the incidence of depression will be determined by computing the proportion of study participants who score 10 or above on the PHQ-9 scale at 3 and 6 months of follow-up. To assess the impact of depression on HRQoL and functional status, a multilevel mixed-effects generalised linear model will be fitted. The three measurement points (baseline, 3 months and 6 months) will be treated as nested within individuals, with individuals nested within health institutions. Time will be centred at baseline, which will be coded as ‘zero’, with the 3rd and 6th months coded as ‘three’ and ‘six’, respectively.

The association of independent variables (sociodemographic characteristics, perceived pain, substance use, comorbidities, PSS, DM-related stigma and medication adherence) with prevalence and incidence of depression, the primary outcome (glycaemic control) and the secondary outcomes (HRQoL and functional disability status) will be analysed in a bivariable analysis. The adjusted role of independent variables will be determined in the multivariable analysis. The inclusion of independent variables in the multivariable analysis will be based on (a) their theoretical importance and (b) the adequacy of the number of participants in each category.⁷⁰ P values less than 0.05 will be considered statistically significant.

Qualitative study

Sick role, activation status of adults living with DM and the health system perspectives on the changes required to meet the long-term care needs of individuals living with DM will be addressed using a qualitative study.

Study approach

A phenomenological study design will be employed. This design is used to gain an in-depth understanding of participants’ experiences, views and perceptions regarding a specific issue.⁷¹

Participants of the qualitative study

• Sick role: this will be explored through interviews with adults living with DM, focusing on their perceptions and lived experiences in relation to four areas: the right to exemptions from responsibilities within the family and/or community due to reduced capacity; exemptions from normal social role obligations; the duty to make efforts to get well and the duty to seek competent help and cooperate with treatment.

• Patient activation: this will be explored through interviews with adults living with DM, focusing on their perceptions and lived experiences regarding the knowledge, skills and confidence needed to manage their condition.

• Health system perspectives: Health system perspectives on the changes needed to meet the long-term care needs of adults living with DM will be explored by interviewing healthcare planners and providers about the acceptability and perceived feasibility of changes required in the organisation and delivery of DM care.

Sample size determination and sampling techniques

Study participants for the qualitative study will be selected purposively. The initial sample size for qualitative objectives will consist of 15 adults diagnosed with DM, 5 healthcare professionals and 5 healthcare managers. The final sample size will be determined by data saturation, based on the following three criteria: objectives are addressed, responses are complete and detailed and no new information emerges from further data collection.⁷²

Data collection tool and management

Qualitative data will be collected following the second quantitative measurement using a prepared set of in-depth interview guiding questions in the local Amharic language. Data will be collected after obtaining informed consent from the study participant. Interviews will be conducted by experienced professionals holding a master’s degree and who have completed 2 days of training. Throughout the data collection, digital audio recording and detailed note-taking will be employed.

Method of analysis of the qualitative data

Thematic analysis will be conducted using NVivo V.12 to identify patterns in the data. Responses will be coded and organised into themes and subthemes to compare and highlight key similarities and differences.

Trustworthiness

To ensure the credibility of the study, the data collector and principal investigator will build trust with participants by spending extended periods engaging with them. Data collection will be supervised, and the team will meet daily to discuss findings and insights. Analysis will be conducted concurrently with data collection, as this approach is essential for refining the in-depth interview guides and uncovering new insights. The research team will also dedicate considerable time to reading and analysing the transcribed data. Detailed descriptions of participants’ experiences, settings and cultural contexts will be provided. Once transcription and initial analysis are completed, key themes will be shared with participants to verify whether the interpretations accurately reflect their experiences. To further refine interpretations and validate conclusions, the findings will be shared with colleagues, PhD supervisors and members of the BDU-LCM project for feedback and discussion.

Ethics and dissemination

Ethical clearance was obtained from the institutional review board of the College of Medicine and Health Sciences, BDU, on 2 April 2025 (protocol number 3098/25). Before data collection, permission will be sought from local administrations and health institutions by showing the ethical clearance and letter of support. Informed written consent will be obtained from each study participant. For study participants who are unable to read or write, fingerprints will be taken after the information sheet and consent form have been explained to them. Study participants who score 10 and above on PHQ-9 or endorse suicide items will be referred to health institution staff for further evaluation and care. Study participants will be informed of their right to withdraw from the study at any time, whenever they want to do so.

Findings will be disseminated through presentations at research conferences and publication in peer-reviewed journals. A summary of the research findings will be submitted to the Regional Health Bureau and the participating health institutions. The main findings will be translated into the local Amharic language and will be shared at a meeting with study participants, the local community and healthcare providers.