Factors associated with discontinuation of anti-retroviral therapy among adults living with HIV/AIDS in Ethiopia: a systematic review protocol

Review question/objective

The aim of this review is to assess the best available evidence regarding risk factors for discontinuation from anti-retroviral therapy (ART) in Ethiopia. Specifically, the review will be assessing the association between discontinuation from ART and the following:

• Socio-demographic and economic factors
• Behavioral factors
• Clinical factors
• Institutional factors.

Background

Since its emergency in the 1980s, the human immunodeficiency virus (HIV) has been infected people of all ages, sexes, races and income status, leading to poor health and socio-economic outcomes across the world.¹ Since its epidemic recognition, almost 78 million people have been infected and about half of these people have died of its sequel, AIDS.² By the end of 2013, globally, 35 million (33.2–37.2 million) people were living with HIV, of whom nearly 0.8% were adults aged 15–49 years.²

Africa, Asia and Latin America were the major continents affected by the disease.³ A total of 24.7 million people were infected with HIV whereby Sub-Saharan Africa was the hardest hit subcontinent.² The subcontinent was home for nearly 71% of people infected with HIV, and one in every 20 adults was living with the virus.² At the end of 2013, in Ethiopia, the prevalence of HIV was 1.2% with approximately 0.8 million people living with the virus.⁴ Given that there is currently no cure and any human being is susceptible to this scourge, the advent of its treatment, anti-retroviral therapy, known to prolong the life of HIV patients, was a significant achievement.⁵

Over time since its recognition, the uptake and scaling up of ART treatment has been increasing. For example, in 2013, a total of 12.9 million patients were receiving ART, of which 11.7 million were from low- and middle-income countries.⁶ There was a 5.6 million increase in the number of people receiving ART from 2010 to the end of 2013.⁶ In Ethiopia alone, the number of people enrolled in ART rose from 900 in 2005 to 300,000 in 2010, and the number of facilities providing the same increased from four in 2003 to 481 in 2009.^5,7 If the quality of life and survival of people living with HIV PLWHIV are to be improved, further effort needs to be made to ensure that once started the treatment is continued effectively to the desirable level.⁸ Discontinuation from ART treatment (hereon in referred to as discontinuation), is the major contributor, among many others, to poor quality of life and death of patients.^9–12 Discontinuation is defined as interruptions to ART therapy due to lost to follow-up, drop out or defaulting, transferring out and stopping medication while remaining in care.¹³ Treatment discontinuation reduces the immunological benefit of ART and increases HIV-related complications, including AIDS-related morbidity, mortality, admission and drug resistance.^13–18

Discontinuation is becoming a significant problem across the globe. For example, a study conducted in India showed that the overall dropout rate was 38.1 per 100 person-years.¹⁹ In Malawi, the attrition rate was 33 and 36 person-years for hospitals and health centers, respectively.²⁰ Additionally, a retrospective analysis conducted in three sub-Saharan African countries indicated that the dropout rate was 2.1 per 100 person-years.²¹ In Ethiopia, lost to follow-up was as prevalent as in other countries. Studies conducted in Aksum St Marry Hospital¹⁰, Mizan Aman General Hospital¹², Jimma University Specialized Hospital²² and University of Gondar²³ reported that the proportion of lost to follow-up was 9.8%, 26.7%, 28% and 31.4%, respectively. Another retrospective longitudinal study from Ethiopia reported that retention of patients in care was a major challenge and varied across health facilities.²⁴

Primary studies conducted in Ethiopia reported that socio-demographic and -economic, behavioral, clinical and institutional factors contributed to discontinuation.^9–12 For example, males were reported to be most affected by discontinuation from being away from home, resulting in discontinuation of medication.¹⁰ Similarly, drug addicted people may be faced with drug-drug toxicity and might discontinue from treatment.¹² The fear of HIV-related stigma has also been reported to keep patients away from treatment.10 Patients with advanced clinical stage (WHO stage 3 or 4) at entry have been recognized to have a better chance of adhering to treatment as treatment leads to improvement.² Lastly, distance from ART clinics has also been noted to discourage patients from returning for treatment due to long travel and waiting times.⁹ Patients dependent on food supplies, patients with mental problems, patients whose partners were HIV negative, patients not being provided with isoniazid prophylaxis, and patients with baseline CD4 <200 cells/mm³ were at risk of discontinuation.^9–12 However, these factors were not uncovered based on systematic reviews, and the studies did not offer recommendations on priority interventions.

There have been very few systematic reviews on patient retention rates conducted in Sub-Saharan Africa. A systematic review conducted in 2007 indicated that there was monthly weighted mean attrition rates of 3.3%/month, 1.9%/month, and 1.6%/month for studies reporting at six, 12 and 24 months, respectively.13 This systematic review assessed retention rates rather than risk factors. In Ethiopia, the issue has been given less attention. One systematic review²⁵ on ART non-adherence or non-compliance was conducted but it did not specifically identify the predictors of loss to follow-up, defaulting and total stoppage from treatment. As far as the authors know, there is no published systematic review and meta-analysis so far on his topic. Furthermore, the lack of high quality data on the association between the treatment discontinuation and its risk factors is an issue, preventing national HIV/AIDS control programs from providing accurate data to inform tailored intervention strategies. Additionally, the absence of a clear and uniform definition of discontinuation is also another challenge. Further evidence is needed to develop a consistent definition. A study from five East African countries revealed the existence of 14 different definitions of ART treatment defaulting were in use.²⁶ Currently, in Ethiopia, lost to follow-up refers to a patient discontinuing for less than three months, and dropping/defaulting if discontinued for more than three months.²⁷

This systematic review will assess the association between known factors and discontinuation of HIV treatment. The review will use studies conducted in Ethiopia since the start of ART in 2002 using the pooled proportion of discontinuation from ART treatment variables and its factors. The authors conducted a preliminary search of databases (Medline [PubMed interface], EMBASE, CINHAL and SCOPUS), and found no current or underway systematic reviews on this or a similar topic in Ethiopia.

Inclusion criteria

Types of participants

This review will consider studies reporting on HIV-positive participants aged 15 years and older who have commenced ART. Patients who have been transferred out will be excluded. Patients should have at least one follow-up time. If studies include both adult and pediatrics, and are not stratified by age (pediatrics and adults) during analysis, they will be excluded. Besides, if the studies focus on attrition (mortality or discontinuation) and are not stratified by mortality and discontinuation during analysis, the study will also be excluded.

Types of exposure

This review will consider studies that have examined risk factors for ART treatment discontinuation. These include socio-demographic and economic risk factors such as age, sex, income and being dependent on food supplies; behavioral risk factors such as mental status, presence of bereavement, the partner's HIV status and fear stigma; clinical factors such as isoniazid prophylaxis provision, presence of side effects, baseline CD4 counts and regimen substitution; and institutional risk factors such as distance from the facility and waiting times.

Types of outcomes

This review will consider studies that include the following outcomes: ART treatment discontinuation, i.e. lost to follow up, drop out or defaulting and stopping medication while remaining in care.

• Lost to follow-up: HIV positive patient who have been on ART treatment and have missed one to three months of clinical appointments but have not yet been classified as “dead” or “transferring out”
• Drop out or defaulting: HIV positive patient who have been on ART treatment and have missed three or more monthly clinical appointments and have not yet been classified as “dead” or “transferring out”
• Stopping medication: HIV positive patient who have been on ART treatment but have stopped treatment due to any reason while they have remained in care.

Types of studies

This review will consider for inclusion both experimental and epidemiological study designs including randomized controlled trials, non-randomized controlled trials, quasi-experimental studies, before and after studies, prospective and retrospective cohort studies, case control studies and analytical cross sectional studies conducted in Ethiopia. The review will also consider for inclusion descriptive epidemiological study designs including case series, individual case reports and descriptive cross sectional studies for inclusion.

Context

The review will consider studies conducted in Ethiopia between 2002 and 2015.

Search strategy

The search strategy aims to find both published and unpublished studies. A three-step search strategy will be utilized in this review. An initial limited search of Google Scholar, MEDLINE (Pub med platform), CINAHL and SCOPUS will be undertaken followed by an analysis of the text words contained in the title and abstract, and of the index terms used to describe the article. A second search using all identified keywords and index terms will then be undertaken across all included databases. Thirdly, the reference list of all identified reports and articles will be searched for additional studies. Studies published English between 2002 and 2015 will be considered for inclusion in this review.

The databases to be searched include:

Medline [PubMed interface], EMBASE, CINHAL, SCOPUS

The search for unpublished studies will include:

Hand searches of studies and different sources of grey literatures from ProQuest Dissertations and Theses [PQDT], Google Scholar, Med Nar, World Bank, WHO and Ministry of Health Data

Initial keywords to be used will be:

ART, antiretroviral, HAART defaulting, dropout, attrition, lost to follow up, retention, linkage, engagement, transfer out, stoppage, interruption, Ethiopia

Assessment of methodological quality

Papers selected for retrieval will be assessed by two independent reviewers for methodological validity prior to inclusion in the review using standardized critical appraisal instruments from the Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) (Appendix I). Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer. Authors of primary studies will be contacted to clarify missing or unclear data.

Data extraction

Data will be extracted from papers included in the review using the standardized data extraction tool from JBI-MAStARI (Appendix II). The data extracted will include specific details about the exposures, populations, study methods and outcomes of significance to the review question and specific objectives.

Data synthesis

Quantitative data will, where possible be pooled in statistical meta-analysis using RevMan Software.28 All results will be subject to double data entry. Some variables might need to be measured by more than one item, and might vary among studies. For example, HIV related stigma might be measured using Berger HIV stigma²⁹ or The People Living with HIV Stigma Index³⁰ scales. In this case, the acceptable range of validity and reliability will be considered. Effect sizes expressed as relative risk for cohort studies and odds ratio for case control studies (for categorical data) and mean difference (for continuous data), and their 95% confidence intervals will be calculated for analysis. Heterogeneity will be assessed statistically using the standard Chi-square and I² tests, and explored using subgroup analysis. Where statistical pooling is not possible the findings will be presented in narrative form including tables and figures to aid in data presentation where appropriate.

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

Our gratitude goes to Flinders University for covering the cost of articles with fees. Thank you to Mr Andrew Craig, Topic Coordinator, Office of Graduate Research, Flinders University, for editing this protocol.