Factors contributing to defaulting on antiretroviral therapy among people living with HIV in selected health facilities in Maswa District, Tanzania

¹Department of Community Health, School of Public Health, Amref International University, Kenya, ²Department of Community Medicine, School of Public Health, Catholic University of Health and Allied Science, Mwanza, Tanzania, ³Regional Medical Office, Simiyu, Tanzania, ⁴Department of Epidemiology, Behavioral Sciences and Biostatistics, Mwanza City Council, Manza, Tanzania

^&Corresponding author
Fransiscko Fundi, Department of Epidemiology, Behavioral Sciences and Biostatistics, Mwanza City Council, Manza, Tanzania

Introduction    

Defaulting on antiretroviral therapy (ART) among people living with HIV (PLHIV) remains a significant global and local public health challenge, undermining efforts to achieve viral suppression, improve quality of life, and prevent HIV transmission [1]. ART discontinuation increases the risks of drug resistance, disease progression, and further HIV transmission, posing an obstacle to achieving the UNAIDS 95-95-95 targets, which aim for 95% of PLHIV to know their status, 95% of those diagnosed to receive ART, and 95% of those on treatment to achieve viral suppression by 2030 [2]. Globally, studies estimate that nearly 30-40% of PLHIV struggle with optimal ART adherence, with sub-Saharan Africa accounting for the highest burden due to systemic and individual-level barriers. In Tanzania, adherence challenges are evident, as national surveys report that up to one in five patients default from treatment within the first two years, and long-term retention rates fall below 60% in some regions. These figures underscore that ART adherence is not only a clinical issue but a major public health challenge requiring urgent attention. Although ART availability has increased, retention remains a challenge, particularly in resource-limited settings like sub-Saharan Africa, where 12-month retention rates range from 64% to 94%, and long-term retention drops to approximately 60% [3]. In Tanzania, approximately 1.7 million people were living with HIV in 2020, with an ART coverage of 82% [4]. Despite this progress, rural areas such as Maswa District experience low retention due to transportation difficulties, stigma, and inadequate healthcare infrastructure [5]. While Tanzania has made progress in HIV care, systemic and individual-level barriers persist, including stigma, economic constraints, and weak healthcare support systems [3]. Studies in Tanzania have reported defaulter rates ranging from 0.9% to 19%, with loss to follow-up and mortality as primary contributors [6]. ART defaulting threatens healthcare systems by increasing costs and deepening health disparities, particularly among vulnerable groups such as women, children, and rural populations [7].

Retention in ART is influenced by multiple factors, including demographic characteristics, socioeconomic conditions, and healthcare system inefficiencies [8]. Stigma associated with HIV remains a major barrier to sustained treatment adherence, as PLHIV fear disclosure and discrimination, which often leads to disengagement from care [9]. Financial constraints also affect retention, as transportation costs and lost wages deter PLHIV from attending clinic appointments, particularly in rural areas with limited access to healthcare facilities [10]. Healthcare-related challenges such as overburdened staff, medication stockouts, and inconsistent follow-up mechanisms further contribute to ART discontinuation [11]. Despite existing global and national efforts to improve ART adherence, these challenges persist, highlighting the need for localized interventions tailored to the specific context of regions like Maswa District [12].

Although various studies have examined ART retention in Tanzania, there remains a critical gap in understanding the predictors of ART defaulting in Maswa District and the broader Simiyu Region [13]. Most research has focused on urban areas or national trends, leaving a gap in knowledge regarding the specific barriers affecting retention in rural settings [14,15]. While previous studies have identified factors such as stigma, transportation challenges, and healthcare infrastructure limitations as contributors to ART discontinuation, few have systematically investigated their impact within the unique socio-economic and healthcare landscape of Maswa District [16]. Furthermore, there is limited evidence on the effectiveness of existing interventions in addressing these challenges in rural Tanzania [17,18]. Addressing this gap is crucial for developing evidence-based policies and targeted interventions that enhance ART retention and contribute to the broader goal of achieving universal HIV treatment success [19]. In view of these challenges and gaps, this study investigated factors contributing to ART default among PLHIV attending selected healthcare facilities in the district, Tanzania, to inform targeted interventions to strengthen ART adherence and advance progress toward the UNAIDS 95-95-95 targets.

Methods     

Study design: this study employed a cross-sectional design with a mixed-methods approach to examine the factors contributing to antiretroviral therapy (ART) default among people living with HIV (PLHIV) in Maswa District, Tanzania. The mixed-methods approach allowed for the integration of quantitative data from patient medical records and qualitative insights from interviews with clients and healthcare providers. This design was chosen to provide both statistical analysis of ART default predictors and an in-depth understanding of the barriers and facilitators affecting ART retention. Data collection was carried out in 2023, covering ART initiation data from January 2020 to December 2022.

Study setting: the study was conducted in four purposively selected care and treatment clinics (CTCs) in Maswa District: Maswa, Lalago, Mwasayi, and Malampaka. These sites were chosen based on geographic diversity, patient volume, and facility classification to ensure a representative sample of the district's HIV care landscape. Maswa District, located in the Simiyu Region of northwestern Tanzania, has 31 facilities offering ART services and is predominantly inhabited by the Sukuma ethnic group. The setting was appropriate due to its high ART coverage and the availability of electronic health records.

Participants: three participant groups were included. First, 2,454 HIV-positive adults aged 18 years and above who initiated ART between 2020 and 2022 formed the basis of the quantitative data. Of these, 1,519 were eligible for analysis after excluding transfers, deaths, and opt-outs. Second, nine clients who had defaulted but were later reengaged in care participated in in-depth interviews (IDIs). Third, five healthcare providers with a minimum of one year´s experience in HIV care were purposively selected for key informant interviews (KIIs). Purposive sampling was used to ensure the inclusion of participants with relevant experiences.

Inclusion and exclusion criteria: the study includes HIV-positive adults (≥18 years) who initiated ART between January 2020 and December 2022 at selected CTCs, as well as patients who were previously lost to follow-up (≥90 days of ART interruption) but successfully reengaged in care and have remained in care for at least one month before the study period. Healthcare providers eligible for the study are those who have been providing HIV care at the selected CTCs for at least one year. Exclusion criteria include HIV patients missing key medical records (such as date of birth, gender, or retention status), those who started ART before January 2020 or after December 2022, or those under 18 at ART initiation. Formerly lost-to-follow-up patients who returned to care less than one month before the study period and healthcare providers with less than one year of experience in HIV care are also excluded.

Sample size and bias management: the sample size for the quantitative component was determined using the Kish-Leslie formula, which estimated a minimum of 385 participants. However, due to the retrospective nature of the study, a census sampling approach was employed to include all 1,519 eligible patient records, thereby maximizing data completeness and minimizing selection bias. To address potential biases, trained data collectors used standardized data abstraction tools, and all entries were double-checked for accuracy. Triangulation across multiple informant types was applied to enhance the credibility and validity of the qualitative findings. Interviewer bias was minimized through the use of neutral probing techniques, while social desirability bias was mitigated by assuring confidentiality and conducting interviews in private settings.

Definition of variables/themes: the dependent variable for this study was ART defaulting, defined as missing antiretroviral therapy (ART) for more than 90 consecutive days from the last scheduled clinic appointment. This variable indicated whether a patient had experienced treatment interruption, regardless of later re-engagement. Viral load suppression was defined as achieving an HIV ribonucleic acid (RNA) viral load of less than 1,000 copies/mL, assessed at baseline, 6, 12, 18, and 24 months. Clients were classified as “yes” for ART default or viral suppression based on whether these respective thresholds were met. For the qualitative component, themes explored included stigma, service accessibility, social support, and health system challenges.

Data sources: quantitative data were extracted from electronic medical records and patient files using a standardized data abstraction form. Qualitative data were collected through IDIs and KIIs using semi-structured interview guides tailored to explore personal and system-level barriers and enablers of ART retention. All tools were pre-tested and translated into Kiswahili for clarity and cultural relevance.

Data collection: quantitative data were collected retrospectively from CTC records covering the period from January 2020 to December 2022. Qualitative interviews were conducted face-to-face in private settings to ensure confidentiality and openness. Each IDI and KII was audio-recorded with consent and transcribed verbatim. Translations were done from Kiswahili to English for analysis.

Data analysis: quantitative data were cleaned, coded, and analyzed using SPSS Version 25. Descriptive statistics were used to summarize demographic and clinical characteristics. Frequencies and percentages were calculated for categorical variables. Bivariate and multivariate logistic regression models were performed to determine independent predictors of ART default, with odds ratios adjusted for potential confounders. Qualitative data were transcribed and manually analyzed using thematic analysis. Interview transcripts were reviewed iteratively, coded, and grouped into themes reflecting barriers and enablers to ART adherence and retention.

Ethical considerations: the study received ethical clearance from the Amref Ethics and Scientific Review Committee (ESRC) and the Lake Zone Institutional Review Board (LZIRB) under approval number MR/53/100/790. Authorization to access patient data was granted by the Maswa District Medical Office and the Ministry of Health through a data-sharing agreement. Written informed consent was obtained from all IDI and KII participants. Confidentiality and anonymity were strictly maintained throughout data collection, analysis, and reporting. Participants were informed of their right to withdraw at any time without consequences to their care.

Results     

Characteristics of the study participants: a total of 2,602 people living with HIV (PLWHIV) were enrolled on ART between 2020 and 2022, with 2,454 individuals meeting the eligibility criteria for this study. All participants were residents of Maswa District, Simiyu Region. The distribution across health facilities was 45% at Maswa District Hospital, 22.5% at Lalago Health Center, 17.6% at Malampaka Health Center, and 14.9% at Mwasayi Health Center. Females comprised 57.7% (1,415) of participants, and 68.8% (1,688) were married. The median age was 39.6 years (±12.4 SD). Most participants were referred from facility indexes (38.3%) and outpatient departments (23.3%), with 96.7% linked to ART within seven days of diagnosis. The majority (59.6%) were in WHO stage 1, and 99.2% were on the first-line regimen TDF/3TC/DTG. Viral suppression was achieved by 98.4%, while 65.1% were classified as unstable due to adherence challenges.

ART retention and default rates among PLHIV

The proportion of retention and default in the ART program: out of the total study population, 1,187 (48.4%) were successfully retained in care. The study recorded 127 deaths (5.2%), while 332 (13.5%) were lost to follow-up, 19 (0.8%) opted out of the program, and 789 (32.2%) transferred to other clinics. After excluding the deceased and those who transferred and opted out, 1,519 individuals were considered eligible for follow-up. Among these, 332 (21.9%) were defaulters, which included those lost to follow-up and those who opted out. This resulted in a retention rate of 78.1%, with 21.9% of the eligible population defaulting from care (Figure 1).

Bivariate analysis of factors influencing ART default: the bivariate analysis revealed that ART default varied significantly across health facilities. Mwasayi Health Center recorded the highest proportion of defaulters, with 127 (38.4%) out of 331 clients. Lalago Health Center followed with 89 (22.7%) out of 392, Maswa District Hospital had 75 (12.8%) out of 585, and Malampaka Health Center accounted for 41 (17.9%) out of 229 clients. These differences were statistically significant (Pearson Chi-square = 318.866, p < 0.001), indicating that facility type may influence default patterns. In terms of sex, males accounted for 159 defaulters, while females had 173. A statistically significant relationship was observed between sex and ART default (Pearson Chi-square = 28.727, p < 0.001). Regarding entry source, the highest number of defaulters was recorded among patients referred through facility index testing (FaIND), with 145 (24.1%) out of 601 defaulters, followed by those referred through community-based health services (CBHS), contributing 74 (32.6%) defaulters out of 227. This association between entry source and ART default was statistically significant (Pearson Chi-square = 197.6, p < 0.001) (Table 1).

Logistic regression analysis of factors influencing ART default: variables that had significant associations between ART default at bivariate analysis (Table 1), further analyses using logistic regression (Table 2). In unadjusted models, age group 25-35 years (OR = 1.42, p = 0.008), divorced/separated marital status (OR = 1.68, p = 0.001), VCT entry (OR = 1.47, p = 0.005), advanced disease stage (e.g., WHO stage 4: OR = 3.48, p < 0.001), unsuppressed viral load (OR = 2.18, p < 0.001), unstable clients (OR = 1.84, p < 0.001), and delayed linkage to ART (>7 days) (OR = 1.58, p < 0.001) were significantly associated with default. However, after adjusting for confounders, only WHO disease stages 2, 3, and 4, viral suppression status, client category, and delayed ART linkage remained significant. For instance, those at WHO stage 4 had nearly three times the odds of defaulting (aOR = 2.84, p < 0.001), and those with unsuppressed viral loads were also more likely to default (aOR = 1.82, p < 0.001). Being an unstable client (aOR = 1.52, p = 0.002) and linking to ART after more than 7 days (aOR = 1.32, p = 0.042) also significantly increased default risk.

Barriers to ART adherence and retention: findings from in-depth interviews (IDIs) with nine ART clients who had previously defaulted and later reengaged in care revealed several barriers and facilitators to treatment adherence. Barriers included a lack of awareness about long-term ART commitment, adverse drug effects, and stigma. As one participant shared, “I knew the drugs were important, but I didn´t understand why I had to take them every day for life” (#32-year-old male). Another emphasized side effects, stating, “the side effects were difficult sometimes, so I would skip a few doses here and there” (#45-year-old female). Stigma and fear of disclosure were also evident: “I worried people would find out about my status if I went to the clinic regularly” (#28-year-old female).

Facilitators of retention included health scares, family support, and compassionate healthcare delivery. A 45-year-old female noted, “I realized I wasn´t well when I stopped taking my medication. I felt worse, and that scared me”. Another commented, “my family encouraged me to get back into care. They reminded me of my health” (#32-year-old male). Improved provider attitudes also helped: “the healthcare workers were more understanding this time. They listened to my concerns” (#28-year-old female).

Key informant interviews (KIIs) with five healthcare providers highlighted systemic issues such as overburdened staff and inadequate follow-up structures. One nurse stated, “we have too many clients and not enough time. That affects how well we can track and support them” (#KII-Nurse, Maswa Hospital). A facility manager added, “some clients disappear before their next visit. We need a better system to follow up, especially in remote areas” (#KII-Manager, Mwasayi). Providers also emphasized the role of community support: “people respond better when community leaders and groups are involved. It reduces stigma and builds trust” (#KII-Lay Counselor, Malampaka). Despite some facilities lacking structured support groups, informants agreed on their impact: “where we have active support groups, patients are more open and committed” (#KII-clinical officer, Lalago).

Discussion     

This study aimed to investigate the factors contributing to antiretroviral therapy (ART) default among people living with HIV (PLWHIV) attending care and treatment clinics (CTCs) in Maswa District, Tanzania. The findings revealed an ART default rate of 21.9%, translating to a retention rate of 78.1%. This figure aligns with national studies conducted in Tanzania, where retention rates typically range between 75% and 80% [20-22]. However, it is slightly higher than those reported in other sub-Saharan African settings such as Mozambique and South Africa, where retention after 12 months often falls below 70% [23,24].

Several predictors of ART default were identified in the quantitative analysis. First, clients in WHO stage 3 or 4 at ART initiation were significantly more likely to default compared to those in earlier stages. This finding is consistent with evidence from Uganda and Nigeria, which indicates that patients diagnosed at an advanced stage often face increased symptom burden, fatigue, and hospitalization, all of which interfere with regular clinic attendance [13,25]. Therefore, strengthening early diagnosis and prompt linkage to care through community-based testing initiatives is critical.

Second, clients with unsuppressed viral loads were also more likely to default. This supports previous findings from Kenya and Malawi, where failure to achieve viral suppression often discourages patients, leading to disengagement from care [26]. Thus, integrating frequent viral load monitoring and providing timely feedback can serve as both a motivational and clinical tool to support adherence.

Marital status emerged as another significant predictor. Specifically, individuals who were divorced or separated were at a higher risk of defaulting compared to their married counterparts. This trend mirrors studies from Mali and Ethiopia, which suggest that marital disruption can reduce emotional support and increase social isolation, thereby undermining ART adherence [27,28]. Consequently, there is a need for tailored psychosocial support and community linkage interventions for clients facing marital stress.

Additionally, entry into care through voluntary counseling and testing (VCT) was associated with a higher likelihood of default. While this finding contrasts with studies from Mozambique and Tanzania that have shown VCT to be a pathway to early diagnosis and retention [23,29], it may reflect poor post-test counseling or weak linkage mechanisms in the study area. Enhancing follow-up procedures and strengthening the quality of post-test counseling could bridge this critical gap. Beyond the quantitative predictors, qualitative findings provided deeper insight into personal and systemic barriers affecting ART retention. A major theme was limited knowledge regarding lifelong ART adherence. Several participants expressed that they did not initially understand the long-term commitment required, a finding consistent with prior studies in Ethiopia and Uganda [30,31]. This underscores the importance of comprehensive adherence education at ART initiation. In addition, adverse drug effects such as nausea, dizziness, and fatigue were reported by clients as reasons for non-adherence. Similar to findings by Gengiah et al. (2016) [32], these side effects can erode patient confidence and reduce willingness to continue treatment. Regular clinical assessments, counseling, and timely regimen adjustments are essential to address this barrier.

Stigma and fear of disclosure also played a central role in ART default. Participants shared concerns about being seen at HIV clinics or having their status discovered by family or community members. This aligns with studies from Malawi and Kenya, which highlight the persistent role of stigma in reducing treatment engagement [33,34]. Interventions must therefore focus on integrating HIV services into general care and conducting anti-stigma campaigns. Furthermore, economic and logistical barriers such as transportation costs, distance to facilities, and lack of childcare were commonly cited. These issues reflect broader systemic challenges documented in Mozambique and Zambia [23,24,35]. To address this, decentralized ART delivery models, mobile outreach clinics, and transport subsidies should be prioritized.

Healthcare providers interviewed (KIIs) corroborated many of these challenges, reporting overburdened staff, lack of follow-up tools, and poor service integration. They emphasized the need for additional staff, improved communication systems, and community engagement to retain clients in care. These systemic gaps can be mitigated through investments in digital patient-tracking systems, task-shifting strategies, and strengthened referral pathways. In conclusion, ART default in Maswa District is driven by an interplay of clinical severity, psychosocial vulnerability, inadequate counseling, and systemic weaknesses. To improve retention, targeted interventions must address both individual-level barriers and institutional limitations through a multifaceted, patient-centered approach.

Limitations: while interpreting the findings of this study, some limitations should be taken into consideration. The research was conducted in a single district (Maswa District), which may limit the generalizability of the results to other regions with different healthcare systems or socio-demographic contexts. The small number of in-depth interviews (n=9) may not fully capture the range of experiences among ART defaulters. Additionally, the use of retrospective data from electronic medical records may have been affected by missing or inaccurate entries, potentially influencing the measurement of key variables. There is also a possibility of recall bias, as participants were asked to reflect on past experiences, and social desirability bias may have led to underreporting of sensitive issues such as stigma or non-adherence. Moreover, the study did not evaluate the effectiveness of current ART retention strategies or assess the long-term sustainability of the identified facilitators. These limitations highlight the need for caution in interpreting the findings and suggest that further research is needed across broader settings and populations.

To mitigate the limitation of focusing on a single district, the study employed a mixed-methods approach that combined a large quantitative dataset with qualitative insights, enhancing internal validity and providing a comprehensive understanding of ART retention within the local context; the findings can still offer valuable lessons for similar settings with comparable health system characteristics. Although the number of in-depth interviews with formerly defaulted ART clients was small, data triangulation through the inclusion of key informant interviews with experienced healthcare providers and the integration of quantitative retention data helped strengthen the credibility and depth of the qualitative findings. Additionally, the use of routinely collected electronic medical records from a large cohort (1,519 individuals eligible for follow-up) increased the robustness of the quantitative results. While the cross-sectional design limited assessment of long-term sustainability of retention facilitators, the study mitigated this by capturing recent ART initiation cohorts (2020-2022), reflecting current programmatic practices; the findings therefore provide a strong foundation for future longitudinal studies to examine sustainability over time.

Conclusion     

This study found higher rates of ART default compared to the national average, highlighting significant challenges in retention among PLHIV in Maswa District. Key factors contributing to default included older age, divorced or separated marital status, advanced disease stages, and unsuppressed viral loads. Barriers such as socio-economic challenges, health complications, and weak support systems exacerbated default rates, while strong local support systems and prompt linkage to care facilitated retention. To address these issues, the study recommends enhanced patient education and counseling, strengthened healthcare systems through service integration, and robust psychosocial support systems, including support groups and stigma reduction initiatives. Regular viral load monitoring and targeted interventions for at-risk individuals are also critical. For future research, longitudinal studies on ART retention trends and comparative analyses across urban and rural settings are recommended to inform tailored interventions. These efforts, led by healthcare providers, policymakers, and researchers, can significantly improve ART retention and health outcomes in Maswa District.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Kidenya Sita (principal investigator): led the overall research design, conceptualization, data collection, and analysis, contributed significantly to the writing of the manuscript, particularly in the introduction, literature review, and conclusion sections; Tom Marwa (first supervisor): provided guidance on the research methodology, data analysis, and interpretation of results, reviewed and critically assessed the manuscript, offering insights to improve the quality of the research and the final write-up; Anthony Kapesa (second supervisor): offered expert advice on the theoretical framework and study design, assisted in refining the research questions and ensuring the relevance of the study´s findings to the broader field of reproductive health; Khamis Kulemba (third supervisor): contributed to study design, tool development, and field supervision, provided key input in data analysis and interpretation, and critically reviewed the manuscript to ensure scientific rigor and relevance to public health policy; Fransiscko Fundi (technical issues, proofreading and formatting, data analysis): contributed to the data analysis, including both quantitative and qualitative aspects, assisted in proofreading and formatting the manuscript, ensuring the document adhered to academic standards, additionally, played a key role in synthesizing the peer review sections and polishing the final manuscript. All the authors read and approved the final version of this manuscript.

Acknowledgments     

The authors wish to acknowledge the contributions of all individuals and institutions that supported this study. We extend our appreciation to healthcare workers and facility in-charges in Maswa District for facilitating access to participants and data. We are grateful to the clients and key informants whose insights enriched the study findings. Special thanks to Dr Sarah Kweyamba for her academic support. The collective efforts of all authors were integral to the successful completion of this manuscript.

Tables and figure     

Table 1: bivariate analysis of factors associated with antiretroviral therapy (ART) default among clients enrolled in care and treatment clinics across four facilities in Maswa District, Tanzania, from January 2020 to December 2022 (N=1,519)

Table 2: multivariate logistic regression analysis of factors associated with antiretroviral therapy (ART) default among clients enrolled in care and treatment clinics (CTCs) in Maswa District, Tanzania, from January 2020 to December 2022 (N=1,519)

Figure 1: retention status of clients enrolled in care and treatment clinics (CTCs) across four health facilities in Maswa District, Tanzania, from January 2020 to December 2022 (N=1,519)

References