Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Death and Seeking Alternative Therapy Largely Accounted for Lost to Follow-up of Patients on ART in Northwest Ethiopia: A Community Tracking Survey

  • Mamo Wubshet ,

    mamo_wubshet@yahoo.com

    Affiliations Institute of Public Health, University of Gondar, Gondar, Ethiopia, Addis Continental Institute of Public Health, Addis Ababa, Ethiopia

  • Yemane Berhane,

    Affiliation Addis Continental Institute of Public Health, Addis Ababa, Ethiopia

  • Alemayehu Worku,

    Affiliations Addis Continental Institute of Public Health, Addis Ababa, Ethiopia, School of Public Health, Addis Ababa University, Addis Ababa, Ethiopia

  • Yigzaw Kebede

    Affiliation Institute of Public Health, University of Gondar, Gondar, Ethiopia

Death and Seeking Alternative Therapy Largely Accounted for Lost to Follow-up of Patients on ART in Northwest Ethiopia: A Community Tracking Survey

  • Mamo Wubshet, 
  • Yemane Berhane, 
  • Alemayehu Worku, 
  • Yigzaw Kebede
PLOS
x

Abstract

Background

Antiretroviral treatment programs in sub-Saharan African countries are highly affected by LTF. Tracking patients lost to follow-up and understanding their status is essential to maintain program quality and to develop targeted interventions to prevent LTF. We aimed to determine the outcome and factors associated with LTF.

Method

A lost to follow-up community tracking survey was conducted to determine the reasons, outcomes and factors associated with LTF at the University of Gondar Hospital, northwest Ethiopia. All patients were tracked at home to ascertain outcome status for lost to follow-up (death and non-death losses).

Result

Out of the 551 patients LTF, 486 (88.20%) were successfully tracked. Death was the most common reason accounted for 233 (47.94%) of the lost to follow-up. Reasons for non-deaths losses include: stopped antiretroviral treatment due to different reasons, 135(53.36%), and relocation to another antiretroviral treatment program by self- transfer, 118(46.64%). The rate of mortality in the first six months was 72.12 per 100 person-years (95% CI: 61.80–84.24) but this sharply decreased after 12 months to 7.92 per 100 person-years (95% CI: 4.44–14.41). Baseline clinical characteristics were strongly associated with mortality.

Conclusion

Death accounts for about half of the loss to follow up. Most deaths occur in the first six months of loss. Seeking alternative therapy is another major reason for loss to follow up. Early tracking mechanisms are necessary to prevent death.

Background

Antiretroviral treatment (ART) programs in sub-Saharan African countries are highly affected by loss to follow-up (LTF). LTF remains one of the major challenges to the success of ART programs in these settings [1][3]. A systematic review of ART programs in sub-Saharan Africa found that about 40% of patients were lost in two years, with large variations in retention rates [4]. In resource-limited settings 1 in 5 of patients are lost in the first six months of initiating ART [5]. Furthermore, it is known that LTF is the most cause of attrition in such settings [6]. On the other hand, the outcome of a third of patients LTF remained unknown in such settings [7].

The outcome of patients LTF in Africa has received relatively little attention. Patients who stopped taking antiretroviral drugs are resulting in high mortality. On the other hand, with increasing availability of ART, patients may have official or self-transferred to another ART program, for example, a program closer to their place of residence. Identifying and tracking patients who are potentially LTF and understanding their status is essential to maintain program quality and to develop targeted interventions to prevent LTF and reduce mortality and thus identify practical solutions that can improve retention.

We performed a community tracking for patients LTF from the ART clinic at the UoG Hospital. The objectives of the study were to determine the reasons and outcomes of patients LTF.

Methods

Setting

The study was conducted in the ART Program run by the UoG Teaching Hospital, in northwest Ethiopia, described in detail elsewhere [8]. The national HIV prevalence was 1.5% in 2011 [9]. Patients from other facilities are referred to the ART program. Patients routinely pick up antiretroviral drug supplies for one month. A limited number of patients may however collect their three-month supply at a time for various reasons. TB/HIV Co-trimoxazole Preventive Therapy (CPT) is recommended to all TB/HIV co-infected patients regardless of the CD4 count [10].

Patient’s paper record is regularly transferred to the electronic database. After patients are referred by the ART clinician, Case managers are responsible for adherence counseling and managing the ART risk factors. Patients who fail to return on their designated appointments are tracked by Case managers. Initial tracking is attempted by telephone by the clinic nurses on the day following the missed appointment. If this is unsuccessful, Case managers are deployed to patients’ respective addresses. Case managers are certified care givers with a modest experience of community health. Patients at the ART Clinic are referred by ART clinician to Case managers for adherence and managing ART risk factors. Case managers also conduct planned outreach activities to track LTF and onsite community support. As part of the government program, the clinic provides care and support to people living with HIV free of charge according to the national and World Health Organization (WHO) guidelines [11], [12].

Study Design and Data Source

A community tracking survey was conducted to determine the outcome and factors associated with LTF at UoG Hospital, northwest Ethiopia. This study involved patients in the treatment program from March 2005 through August 2010 (Figure 1). Patients who missed their last scheduled visit by 3 months, as proposed by the National ART Program guidelines [11], were tracked to find out the reason and outcomes for LTF.

Study Population and Data Collection

Between September and October 2010, we tracked adult patients who were 3 months late for their appointment to pick-up their antiretroviral drugs. The database maintained for similar purpose had contact information and baseline characteristics. Baseline clinical characteristics were taken from the patient card. Outcomes and reasons of tracked LTF were collected using a standardized semi-structured questionnaire which was translated to Amharic and back translated to English, and was piloted in the field.

Case managers contacted patients or their family members in person in their homes to determine whether the patient’s LTF was related to death or non-death. If patients were tracked at home, they were asked whether they were still taking ART and if not, why they had stopped. If the patient had died, the relatives were asked when the patient had died. If the patient had moved away, relatives or friends were asked where they had moved. The contact details of immediate family member or friend in the patient card was taken for such consultations.

Non-death losses were the sum total of self-transfer-outs and true losses to follow-up. Self transfer-outs were patients whose care was transferred to other facilities by patients themselves, unlike to facility transfer outs whose care transfer is official. True losses were those patients whose information was not able to be tracked by the community tracking survey. For patients who were apparently alive but stopped taking ART at the time of data collection, we asked the reasons for not returning to the clinic.

Statistical Analysis

Data were analyzed using Stata, version 11(Stata Corp). The main outcome measures were patients who were lost to the program due to death and non-death reasons. Descriptive statistics was used to characterize the LTF. We calculated rates of mortality from the date of ART initiation and date of LTF registered in the database. Person-time was censored at the end of August 2010 for individuals who were found to be alive but stopped treatment. Deaths were censored on the date of death reported by family members. Then mortality rates were calculated by dividing deaths by person-year to get mortality rates per 100 person-years of observation.

We estimated time to LTF and death by Kaplan-Meier failure function. In other product-limit analyses, log rank tests were used to examine the effect of baseline clinical characteristics on LTF probabilities. Product limit analysis was used to calculate the instantaneous hazard of death or other losses through time among individuals receiving ART; we plotted smoothed hazard-function estimators using weighted kernel-density estimates based on the Gaussian function.

Cox-proportional hazard models were used to examine determinants of mortality and non-death losses among individuals LTF in the treatment program. Separate models were developed to examine factors associated with reasons for LTF (death and none death reasons). We considered the model which included all the predictors that had a P value of less than 0.20 in the univariate analyses. Covariates predetermined to be in the univariate analysis were entered to full multivariable model. Model diagnostics and the proportional hazards assumption were examined using Schoenfield residuals [13]. The following baseline covariates were entered into the model: age, sex, marital status, occupation, disclosure status, presence of tuberculosis, place of residence, functional status, WHO stage, and CD4 count.

Ethics Statement

This study obtained ethical clearance from the Institutional Review Board of the UoG, Ethiopia. All patients successfully tracked for the study provided a written informed consent. Patients found to be alive but stopped ART, counseling was given to continue their treatment by the Case managers during data collection. Care was taken to keep all patient information confidential during tracking.

Results

Cohort Characteristics and Tracking Patients

Five hundred fifty-one patients were lost out of the 3012 who were initiated on ART be between March 2005 and August 2010 were used for this study (Fig. 1). Follow-up time before loss to follow up ranged from 30 to 1447 days with a median follow-up time of 302 days (IQR: 152–627). Overall patients enrolled in this study contributed 624.25 person-years of observation. Baseline socio-demographic and clinical characteristics of the patients who were LTF on ART is presented in Table 1. The median age was 33 years (IQR: 28–40). The majority (71.5%) came from outside Gondar town. Their clinical characteristics also showed that 46.5% did not disclose their HIV status, 40% had tuberculosis co-infection at ART initiation. The mean CD4 cell count was 68 cells/µl (IQR: 41–100). Baseline functional status and WHO clinical stage showed that about 51% were either ambulatory or bed ridden and 39% were WHO stage III or IV, respectively.

thumbnail
Table 1. Demographic and clinical characteristics of patients included in LTF tracking study (N = 551).

https://doi.org/10.1371/journal.pone.0059197.t001

Outcomes and Reasons of LTF

Among 551 patients who were lost from follow up, 486(88.20%) were successfully tracked and the remaining 65(11.80%) were unable to be tracked (Table 2). For those tracked, death accounted for 233 (48%) of the losses. The non-death losses, (253, 52.06%), were due to relocation to another ART program as self transfer-outs (47%) and stopping ART due to different reasons (53%).

The most common reasons mentioned for not returning to the ART Clinic were: preference of traditional medicine and/or holy water instead of ART (55.56%), perceived improved (30.37%) or deteriorated (25.19%) health status, financial problem (22.22%), and stigma and social problems (9.63%). On the other hand, reasons for the unsuccessful tracking include incorrect, missing or change of addresses (Table 3).

Characteristics of Death and Non-death Losses

The overall incidence of mortality among LTF was 37.74 per 100 person-years (95% CI: 33.18–42.92). The rate of mortality in the first six months was 72.12 per 100 person-years (95% CI: 61.80–84.24) but this sharply decreased after 12 months of follow-up to 7.92 per 100 person-years (95% CI: 4.44–14.4). Mortality in patients with CD4 count <100 cells/µl, WHO stage III and IV, and the presence of tuberculosis at ART initiation was 47.66 (95% CI: 41.65–54.55), 60.40(95% CI: 51.02–71.50) and 54.61(95% CI: 46.01–64.80) per 100 person years, respectively. Kaplan-Meier analyses also showed that the probability of program loss due to death was strongly associated with baseline clinical characteristics (Figure 2), such as baseline CD4 cell count (A), WHO stage of disease (B), and the presence of tuberculosis infection (C), unlike the non-death losses.

thumbnail
Figure 2. Kaplan-Meier Estimate of Mortality among LTF by baseline clinical characteristics: CD4 cell count (A) WHO clinical stage (B) and Tuberculosis status (C).

https://doi.org/10.1371/journal.pone.0059197.g002

Cox proportional hazards model demonstrated that female sex and baseline clinical characteristics such as presence of tuberculosis infection at ART initiation, functional status (both ambulatory and bed ridden); CD4 cell count <100 cells/µL, and WHO stage III and IV were strongly associated with mortality. On the other hand, male sex, bedridden functional status and residence outside Gondar town were significantly associated with non-death losses. Where as, the presence of tuberculosis infection at ART initiation was significantly associated with both death and non-death losses (Table 4).

thumbnail
Table 4. Cox-regression model for risk factors for LTF due to death and non-death reasons.

https://doi.org/10.1371/journal.pone.0059197.t004

The risk of loss from the program changed markedly during follow-up of patients receiving ART. Risk of death had 3 distinct phases (Figure 3): high in the initial months of ART– but steeply decreasing risk thereafter (A) – and a very low risk of death after 1 year of ART (B). In contrast, the risk of program loss due to non-death causes (stopping ART and self transfer outs) was low initially, then relatively constant beyond the first year (C).

thumbnail
Figure 3. Smoothed hazard estimates for total losses to program (A), death (B), and non-death losses to program(C).

https://doi.org/10.1371/journal.pone.0059197.g003

Discussion

This study clearly demonstrated the characteristics and outcomes of patients LTF in northwest Ethiopia. Overall, losses due to death are highest in the first year while losses due to non-death reasons are highest then after. The study also showed close to half (48%) of the LTF were dead. Preference for traditional medicine and/or holy water instead of ART was the commonest reason for not returning to the ART Clinic in patients with LTF.

A high risk of death in the first few months after starting ART is characteristic of resource-limited settings where most patients start therapy late with advanced disease [7], [14], [15]. However, mortality in this study is substantially higher than the mortality commonly reported in the first year of ART based on routinely recorded deaths [1], [5], [16], [17]. Our study can thus be used to correct estimates of mortality derived without complete tracking of losses to follow up [5], [18][20].

In contrast, non-death losses are heterogeneous in this setting. Our finding suggested that in those patients who stopped ART, the risk of losses to the program remain constant after ∼1 year of treatment. This finding is consistent with similar studies [21] except differences in the definition of non-death losses. The effectiveness of ART beyond one year is likely to relate to issues of long-term patient retention rather than death. As patients remain healthy on long-term medication, their motivation to continue treatment in the longer term may diminish. LTF patients who were self-transferred in this program are also substantially higher. In programs with high rates of LTF, those LTF might thus include a sizeable group of low-risk patients who self-transfer out to another program, for example, because of a more convenient location of the new clinic or to avoid stigma or due to work-related reasons [7]. Because the number of patients who are physically well and are receiving long-term medication is increasing, the number of patients moving out of an area for social or economic reasons may continue to increase. Patients might have requested preconditions for official transfer by the health care provider. Further research should understand why patients tempted to self transfer. This finding emphasizes the importance of systems within national ART programs that can ensure the continuity of care for the highly mobile populations. It is important to address patients’ needs to transfer to other treatment centers by providing referral on request without any pre-condition.

The most common reasons mentioned for not returning to the ART Clinic was preference for traditional medicine and/or holy water, instead of ART. Previous studies have found that HIV-infected patients who distrust their health care providers are more likely to use complementary and alternative medicine (CAM) as a substitute for conventional HIV therapy [22]. In addition, HIV-infected patients with a greater desire for medical information and a negative attitude toward the effectiveness of antiretrovirals are more likely to use CAM with potential for adverse effects [23]. Retaining patients who have treatment dilemmas requires HIV clinical personnel to work with religious personnel to provide a model of faith and healing that precludes an “either/or” situation between ART and holy water [24]. Focus should also be placed on sensitizing spiritual leaders to the need for patient adherence to ART concomitant to any faith-based healing practices. Perceived improved or deteriorated health status by the patient is also another reason for not returning to care because of poor understanding of the chronic nature of the disease and the need for continued, life-long ART [7]. Our study also showed that the commonest reasons for unsuccessful tracking were incorrect, missing or change of addresses, which is consistent with reports from similar settings in Ethiopia [16]. Appropriate system to track patients should be a priority in the context of scaling ART. But these issues may not have received sufficient attention from the governmental and nongovernmental organizations driving the scale-up of ART in resource limited settings [25].

Our finding showed that advanced disease stage was a strong predictor of early death is inline with other studies in sub-Saharan Africa [14], [21]. Thirty nine percent of the patients were in Stage III and IV. This finding is different from findings from many other cohort profiles in Ethiopia where more than 50% of the patients are in Stages III and IV combined [26]. This difference might be smaller sample size in our study. We also observed, like other studies in similar settings, that tuberculosis infection at ART initiation was a strong predictor of death and non-death losses [27], [28]. Males were more likely to contribute for death and non death losses [21]. Place of residence outside the Gondar town where the treatment was initiated was associated only with non-death losses. Distance between patients’ residence and the ART Clinic is known to contribute to LTF [29]. Moreover, financial problem for transportation and other related expenses were some of the reasons for stopping medication, as reported in previous studies 7,30.

Some limitations of our study are worth mentioning. Firstly, the study was based in a single ART treatment center so may not be generalisable to a wider context. Secondly, the study was limited to characterizing the outcomes of LTF by physical tracking. Whether the tracking contributed to the re-engagement of patient LTF was not assessed. Future studies should determine the effects of patient tracking on estimates of LTF, mortality and retention in ART programs. Thirdly, the outcome of self-transferred patients was not determined in this study. This might have affected the true estimate of the outcome of LTF. Finally, the study didn’t investigate how and where LTF patients died. Further research should be directed to determine the cause and place of death of patients LTF.

In conclusion, death accounts for about half of the LTF. Most deaths occur in the first six months of loss. Preference of alternative therapies such as traditional medicine and holy water are another major reason for loss to follow up. Early tracking mechanisms are necessary to prevent death. It is also important to address patients’ needs to transfer to other treatment centers by providing referral on request without any pre-condition. Addressing the issues on concurrent use of alternative therapies can also substantially reduce loss to follow up. The findings of the study have implications for patient support and monitoring in ART programs.

Acknowledgments

The authors would like to thank the study participants, staffs of University of Gondar, Addis Continental Institute of Public Health and I-TECH-Ethiopia, Gondar team, for their support during the conduct of this study.

Author Contributions

Approved the final version: MW YB AW YK. Conceived and designed the experiments: MW YB AW YK. Performed the experiments: MW YB AW YK. Analyzed the data: MW YB AW YK. Contributed reagents/materials/analysis tools: MW YB AW YK. Wrote the paper: MW YB AW YK.

References

  1. 1. Assefa Y, Kiflie A, Tesfaye D, Mariam DH, Kloos H, et al. (2011) Outcomes of antiretroviral treatment program in Ethiopia: retention of patients in care is a major challenge and varies across health facilities. BMC Health Serv Res 11: 81.
  2. 2. Cornell M, Grimsrud A, Fairall L, Fox MP, van Cutsem G, et al. (2010) Temporal changes in programme outcomes among adult patients initiating antiretroviral therapy across South Africa, 2002–2007. AIDS 24: 2263–2270.
  3. 3. Egger M, Spycher BD, Sidle J, Weigel R, Geng EH, et al. (2011) Correcting mortality for loss to follow-up: a nomogram applied to antiretroviral treatment programmes in sub-Saharan Africa. PLoS Med 8: e1000390.
  4. 4. Rosen S, Fox MP, Gill CJ (2007) Patient retention in antiretroviral therapy programs in sub-Saharan Africa: a systematic review. PLoS Med 4: e298.
  5. 5. Brinkhof MW, Dabis F, Myer L, Bangsberg DR, Boulle A, et al. (2008) Early loss of HIV-infected patients on potent antiretroviral therapy programmes in lower-income countries. Bull World Health Organ 86: 559–567.
  6. 6. Fox MP, Rosen S (2010) Patient retention in antiretroviral therapy programs up to three years on treatment in sub-Saharan Africa, 2007–2009: systematic review. Trop Med Int Health 15 Suppl 11–15.
  7. 7. Brinkhof MW, Pujades-Rodriguez M, Egger M (2009) Mortality of patients lost to follow-up in antiretroviral treatment programmes in resource-limited settings: systematic review and meta-analysis. PLoS One 4: e5790.
  8. 8. Wubshet M, Berhane Y, Worku A, Kebede Y, Diro E (2012) High Loss to Followup and Early Mortality Create Substantial Reduction in Patient Retention at Antiretroviral Treatment Program in North-West Ethiopia. ISRN AIDS 2012.
  9. 9. Central Statistical Agency AA, Ethiopia (2011) Ethiopia Dmographic and Health Survey 2011. Addis Ababa, Ethiopia.
  10. 10. FMOH (2007). Implementation Guideline for TB/HIV Collaborative Activities in Ethiopia. Addis Ababa, Ethiopia.
  11. 11. FMOHE (2007) Antiretroviral treatment program implementation guideline in Ethiopia. In:. editor. Addis Ababa (Ethiopia) Ministry of Health.
  12. 12. WHO/UNAIDS (2006) Patient monitoring guidelines for HIV care and Antiretroviral Therapy(ART). Geneva, Switzerland WHO press.
  13. 13. Abeysekera WWM, Sooriyarachchi M.R. (2009) Use of Schoenfeld’s global test to test the proportional hazards assumption in the Cox proportional hazards model: an application to a clinical study. JNatnSciFoundation Sri Lanka: 41–51.
  14. 14. Geng EH, Bangsberg DR, Musinguzi N, Emenyonu N, Bwana MB, et al. (2010) Understanding reasons for and outcomes of patients lost to follow-up in antiretroviral therapy programs in Africa through a sampling-based approach. J Acquir Immune Defic Syndr 53: 405–411.
  15. 15. Yu JK, Chen SC, Wang KY, Chang CS, Makombe SD, et al. (2007) True outcomes for patients on antiretroviral therapy who are “lost to follow-up” in Malawi. Bull World Health Organ 85: 550–554.
  16. 16. Deribe K, Hailekiros F, Biadgilign S, Amberbir A, Beyene BK (2008) Defaulters from antiretroviral treatment in Jimma University Specialized Hospital, Southwest Ethiopia. Trop Med Int Health 13: 328–333.
  17. 17. Zwahlen M, Harris R, May M, Hogg R, Costagliola D, et al. (2009) Mortality of HIV-infected patients starting potent antiretroviral therapy: comparison with the general population in nine industrialized countries. Int J Epidemiol 38: 1624–1633.
  18. 18. Yiannoutsos CT, An MW, Frangakis CE, Musick BS, Braitstein P, et al. (2008) Sampling-based approaches to improve estimation of mortality among patient dropouts: experience from a large PEPFAR-funded program in Western Kenya. PLoS One 3: e3843.
  19. 19. Geng EH, Emenyonu N, Bwana MB, Glidden DV, Martin JN (2008) Sampling-based approach to determining outcomes of patients lost to follow-up in antiretroviral therapy scale-up programs in Africa. JAMA 300: 506–507.
  20. 20. Brinkhof MW, Spycher BD, Yiannoutsos C, Weigel R, Wood R, et al. Adjusting mortality for loss to follow-up: analysis of five ART programmes in sub-Saharan Africa. PLoS One 5: e14149.
  21. 21. Lawn SD, Myer L, Harling G, Orrell C, Bekker LG, et al. (2006) Determinants of mortality and nondeath losses from an antiretroviral treatment service in South Africa: implications for program evaluation. Clin Infect Dis 43: 770–776.
  22. 22. Hsiao AF, Wong MD, Kanouse DE, Collins RL, Liu H, et al. (2003) Complementary and alternative medicine use and substitution for conventional therapy by HIV-infected patients. J Acquir Immune Defic Syndr 33: 157–165.
  23. 23. Namuddu B, Kalyango JN, Karamagi C, Mudiope P, Sumba S, et al. (2011) Prevalence and factors associated with traditional herbal medicine use among patients on highly active antiretroviral therapy in Uganda. BMC Public Health 11: 855.
  24. 24. I-TECH (2011) Patients lost to follow-up: research into the tradition of holy water and implications for patient adherence to antiretroviral therapy in Ethiopia.: University of Washington.
  25. 25. Forster M, Bailey C, Brinkhof MW, Graber C, Boulle A, et al. (2008) Electronic medical record systems, data quality and loss to follow-up: survey of antiretroviral therapy programmes in resource-limited settings. Bull World Health Organ 86: 939–947.
  26. 26. Mulissa Z, Jerene D, Lindtjorn B (2010) Patients present earlier and survival has improved, but pre-ART attrition is high in a six-year HIV cohort data from Ethiopia. PLoS One 5: e13268.
  27. 27. Nansera D, Bajunirwe F, Elyanu P, Asiimwe C, Amanyire G, et al.. (2012) Mortality and loss to follow-up among tuberculosis and HIV co-infected patients in rural southwestern Uganda. Int J Tuberc Lung Dis.
  28. 28. Hermans SM, Kiragga AN, Schaefer P, Kambugu A, Hoepelman AI, et al. (2010) Incident tuberculosis during antiretroviral therapy contributes to suboptimal immune reconstitution in a large urban HIV clinic in sub-Saharan Africa. PLoS One 5: e10527.
  29. 29. Conley NJ, Pavlinac PB, Guthrie BL, Mackelprang RD, Muiru AN, et al. (2012) Distance from Home to Study Clinic and Risk of Follow-Up Interruption in a Cohort of HIV-1-Discordant Couples in Nairobi, Kenya. PLoS One 7: e43138.
  30. 30. Weigel R, Hochgesang M, Brinkhof MW, Hosseinipour MC, Boxshall M, et al. (2011) Outcomes and associated risk factors of patients traced after being lost to follow-up from antiretroviral treatment in Lilongwe, Malawi. BMC Infect Dis 11: 31.