Abstract
Background
Treat-all strategies improved patient outcomes, despite higher rates of loss to follow-up compared to the pre-treat era. Patients in Ethiopia experienced a higher rate of LTFU during the treat-all strategy period; however, studies did not identify contributing factors in comparison with previous strategies. This study aimed to assess the incidence and predictors of loss to follow-up before and after the start of the treat-all strategy among adults on anti-retroviral therapy in public health facilities in Hawassa City, Ethiopia.
Methods
An institution-based retrospective follow-up study was conducted among 1190 randomly selected adults on antiretroviral therapy in public health facilities in Hawassa City. Using the Open Data Kit (ODK), data were collected from medical records and exported to Stata version 16 and R 4.2.1 for analysis. A Grays test and cumulative incidence curve were used to compare the cumulative incidence function of loss to follow-up. Bivariable and multivariable competing risk regression were fitted to identify predictors of LTFU and variables with a p-value <0.05 were considered significant.
Results
The cumulative incidence of lost-to-follow-up was 4.92(3.84,6.3) and 8.67(7.26,10.3) per 100 person-years (PY) in pre-treat all and treat all cohorts, respectively. The cumulative incidence of mortality was 5.86(4.67,7.35) and 3(2.26,4.12) per 100 PY in pre-treat and treat all cohorts, respectively. Fair/poor adherence (aSHR:5.17; (95% CI 1.97, 13.51), underweight (aSHR:2.13; 95% CI: 1.15–3.93) and WHO stage III/IV (aSHR:2.69; 95% CI: 1.27, 5.71) were predictors of loss up in pre—treat all, whereas fair/poor adherence (aSHR = 2.07; 95% CI: 1.18, 3.68), underweight (aSHR:1.71; 95% CI: 1.13, 2.56), and CD4 cell >350 cell/m3 (aSHR: 1.67; 95% CI: 1.05, 2.65) predicts of loss up in treat all cohorts.
Conclusion
This study demonstrated that the incidence of loss to follow-up was considerably higher in the treat-all period as compared to the pre-treat-all era. Poor medication compliance, underweight, and a CD4 level >350 cells/m3 contributed to the higher rate of LTFU in the treat-all strategy. Targeted interventions, such as nutritional support and strengthening medication adherence counseling, should be implemented to maintain treatment retention and reduce antiretroviral therapy dropout rates.
Introduction
Loss to follow-up(LTFU) is defined as the failure of HIV-positive patients on medication refill follow-up visits for three or more months from the last appointment date and has not been documented as dead or transferred out to another HIV clinic [1]. The advantages of anti-retroviral therapies are compromised by LTFU, which also raises the risk of morbidity, death, and hospitalization due to AIDS [2–4]. It serves as a warning indicator of drug resistance, ineffective therapy, and mortality driven by suboptimal ART adherence [5, 6]. Antiretroviral therapy (ART) significantly decreases the viral load when people with HIV (PLHIV) adhere to their treatment schedule in full and for the rest of their lives [7].
Though LTFU has become a major public health concern for the ART program in developing countries [8, 9]. In sub-Saharan Africa and Asia-Pacific, more than 20% of patients fail to return to ART for more than six months annually [10]. The incidence of LTFU varies by region and setting in Ethiopia [3, 4, 11]. It was 3.7 per 100 PYO at Debre Markos [11] and 23.76 per 100 PY at Karmara General Hospital Jijiga town, Ethiopia [12]. The LTFU rate in the first six months was 15%, which significantly increased to 21% at the 12-month follow-up [13, 14].
Studies reported that several factors, such as sociodemographic factors, age [4, 11, 15], sex [16, 17], occupation [18], residence [19, 20], marital status [20], educational status [21], disclosure status [22], caregiver [17, 23, 24] and clinical and treatment-related factors like baseline WHO stage [3, 13, 25], body mass index [4], baseline CD4 count [26–28], opportunistic infection at enrollment [29], baseline functional status [30], not receiving prophylaxis [12, 29] and ART regimen [16] were predictors of LTFU.
Ethiopia implemented a universal strategy in January 2017 to enroll all HIV-infected patients in ART services [31]. Although there is sufficient data to support the benefits of starting ART early, which include increased uptake and care coordination, reduced HIV-related morbidity, slowed disease progression, and decreased HIV transmission [32], there were conflicting results regarding the association with obtaining satisfactory ART retention and optimal therapy adherence [22, 33]. To achieve the UNAIDS goals of 95-95-95, particularly the 95% viral suppression in ART-receiving populations, retention in care is essential. To end the HIV epidemic by 2030, further studies and interventions on LTFU and its contributing factors are crucial [34].
Ethiopia’s health ministry is implementing strategies to reduce follow-up rates, including standardized monitoring systems, reliable outcomes, early mortality reduction, uninterrupted drug supplies, non-toxic ART regimens, decentralization of care, reduced indirect costs, and strengthening links between health services and communities [35]. Despite government efforts, the LTFU rate remains high in the treat-all era [12], with no study conducted on it compared to the pre-treat-all era, necessitating ongoing efforts to provide updated information for effective test and treat strategies. Besides, in Ethiopia, studies on LTFU use standard Cox regression analysis [15, 36], which overlooks competing events, leading to an overestimation of LTFU incidence [37]. Fine and Gray competing-risks regression can reduce biases in estimating LTFU in competing event situations, as it considers death as a competing event [38, 39]. Thus, this study aimed to estimate the incidence and identify predictors of LTFU while considering death as a competing event in the public health facilities of Hawassa City, southern Ethiopia.
Methods
Study area and period
The study was conducted at public health facilities in Hawassa City from May 1 to 30, 2022. Hawassa City is located 273 Km far south of Addis Ababa, the capital city of Ethiopia. Currently, three governmental hospitals (Hawassa University Comprehensive Specialized Hospital, Adare General Hospital, and Tula Primary Hospital) and three health centers (Bushulo, Millennium, and Tula Health Center) are accessible for antiretroviral therapy services in Hawassa City. According to the City Health Department Health Information Management System (HIMS) report, there were 6095 HIV-infected people on ART follow-up in the health facilities at Hawassa City.
Study design and population
An institution-based retrospective follow-up study was conducted at Hawassa City public health facilities from January 1, 2012, to December 31, 2021. All HIV-infected adults who enrolled at six public health facilities ART clinics in Hawassa City during the study period were the source population. All randomly selected HIV-infected adults who were enrolled at six public health facilities and ART clinics in Hawassa City were included in the study. Whereas, HIV patients with incomplete records regarding baseline characteristics (age, undefined outcome, date of ART initiation, and LTFU) and those who enrolled in ART through the prevention of mother-to-child transmission (PMCT) program were excluded.
Sample size calculation and sampling procedure
The sample size was calculated based on the double population formula using Epi- info 7.2.1.0 considering the following assumptions: 95%CI, power 80%, the ratio of unexposed to exposed1:1, loss to follow up among rural residence = 13.7% and hazard ratio = 0.6 [11]. The final sample size after adding a 10% for chart incompleteness rate was 1190. By using proportional allocation, 434, 556,26,49,29 and 96 patients’ records were selected from Adare General Hospital, Hawassa University Comprehensive Specialized Hospital, Tula Hospital, Millennium Health Center, Tula, and Bushulo Health Center, respectively. STATA version 16 was used to generate a random sample. Subsequently, 1190 patient cards were selected at random using the subsequent unique number from the database registry.
Variables of the study
The outcome variable in this study was the incidence of LTFU with death as a competing event. The predictor variables assessed were baseline socio-demographic factors (sex, age, marital status, educational status, occupation, residence, functional status, and disclosure status) and baseline clinical and treatment-related factors (baseline opportunistic infection, baseline CD4 count, type of regimen at the start, isoniazid and co-trimoxazole preventive therapy, baseline and last known WHO clinical stage, BMI and comorbidity).
Operational definitions of variables
Event: loss to follow-up (LTFU) defined as not taking an ART refill for three months or longer from the last attendance for the refill and not yet classified as dead or transferred out [1]. Censored: individuals who were formally transferred- out to other health institutions after initiating ART and or individuals who remain active on ART follow-up at the end of the study.
Drug adherence: means taking the drug according to instructions given by the providers and estimated using three levels [40]. Good—equal to or greater than 95% adherence i.e., missing 1 dose from 30 doses or 2 out of 60 doses. Fair—85–94% adherence, i.e., missing 2–4 tablets from 30 tabs or 4 to 9 out of 60 doses. Poor—less than 85% adherence, i.e., missing more than five tablets from 30 tabs or > 10 tabs from 60 tablets.
Data collection methods and tool
The data extraction tool was developed from the WHO standardized ART entry and follow-up chart, which is currently in use in ART clinics [4, 36, 41]. The entire cohort was organized into two separate cohorts of PLHIV based on the ART initiation date. These are pre-treating all cohorts (Jan1/2012 to Dec 31/2016) and treating all cohorts (Jan1/2017 to Dec 31/2021) which corresponds to the period before and after the start of implementation of the treat-all strategy. The ART register, SMART care, and patient follow-up card were used to determine the clients’ status. It was recorded as being dead (from whatever cause), lost and dropped, stopped, or retained in care. The data were collected from the patient charts by six BSc nurses and two supervisors using ODK.
Data quality control
To control the data quality, the data extraction tool was pretested on 5% of the sample. The data collectors and the supervisor were trained on how to extract the data based on the study objectives. The data were checked for consistency and completeness daily by the supervisor and principal investigator. The codebook contained data cleaning for any missing values or data errors.
Data processing and analysis
Data were collected by using ODK version 3.4 and then exported Stata 16 and R 4.2.1 software for analysis. Descriptive summaries like median, proportion, and graphs were used to describe baseline sociodemographic characteristics and clinical characteristics. A competing regression analysis was fitted to identify predictors of LTFU. Variables in the bi-variable proportional sub-distribution hazard model with a p-value ≤ 0.25 were included in the final multivariate analysis. The proportional sub-hazards assumption was checked using the global test, and it was met because it was 0.4304 and 0.845 in the pre-treat and treat all cohorts, respectively. Gray’s test and cumulative incidence curve were employed to compare the cumulative incidence function of LTFU. Variables with p-value < 0.05 were considered significant predictors of LTFU.
Ethical consideration
Ethical clearance was obtained from the ethical review committee of the College of Health Sciences of Wolaita Sodo University (Ref No: CRCSD86/02/2014). A formal letter including the study objectives was submitted to the health departments of Hawassa City and HUCSH and permission was obtained from the City health department, the respective hospitals, and the health centers where the study was conducted. As the study was conducted through a review of patient records, no consent was obtained from the patients. Information about specific personal identifiers like the patient names was not collected and the personal informant was kept confidential throughout the study process.
Results
Baseline socio-demographic characteristics
A total of 1145 (96.2%) patient charts were included in the analysis; the remaining 45 (3.78%) were excluded due to eligibility criteria. The entire cohort was divided into 565 (49.3%) pre-treat-all cohorts and 580 (50.7%) treat-all cohorts. Nearly sixty percent (678) of the study participants were females. A large proportion of participants 196(34.89%) of pre-treat all cohort and 183(31.55%) of treat all cohort patients were in the age category of 25–34 years. The median age of 35 years (IQR = 28 to 40) and 32 years (IQR = 25 to 42.5) in pre-treat and treat all cohorts, respectively. The majority of the study participants 359 (63.54) of pre-treat all and 561(96.72) of treat-all cohorts were from urban and there were not many differences in socio-demographic backgrounds of clients between the two cohorts (Table 1).
Table 1. Baseline socio-demographic backgrounds of adult HIV patients on ART before and after the start of treat-all strategies in Hawassa City between Jan 2012, and Dec 2021.
| Variables | Category | Pre-treat all n = 565 |
Treat all n = 580 |
p-value |
|---|---|---|---|---|
| Sex | Male | 201(35.58) | 266(45.86) | < 0.001 |
| Female | 364(64.42) | 314(54.14) | ||
| Residence | Urban | 359 (63.54) | 561(96.72) | < 0.001 |
| Rural | 206(36.46) | 19(3.28) | ||
| Type of facility | Hospital | 472 (83.54) | 499 (86.03) | 0.954 |
| Health center | 93(16.46) | 81(13.97) | ||
| Age | 15–24 | 86(15.22) | 137(23.62) | < 0.001 |
| 25–34 | 196(34.89) | 183(31.55) | ||
| 35–44 | 194(34.69) | 146(25.17) | ||
| >/ = 45 | 89 (15.75) | 114(19.66) | ||
| Marital status | Single | 136(24.07) | 210(36.21) | < 0.001 |
| Married | 331(58.58) | 290 (50) | ||
| Divorced | 71(12.57) | 68 (11.72) | ||
| Widowed | 27(4.78) | 12 (2.07) | ||
| Educational status | No education | 75(13.27) | 59(10.17) | 0.420 |
| Elementary | 137(24.25) | 242(41.72) | ||
| Secondary | 146(25.84) | 98(16.9) | ||
| College& above | 207(36.64) | 181(31.21) | ||
| Occupation | Daily labourer | 138(24.42) | 148(25.52) | 0.359 |
| Merchant | 59(10.44) | 68(11.72) | ||
| Student | 143(25.31) | 123(21.21) | ||
| Employer | 105(18.58) | 136(23.45) | ||
| Housewife | 83(14.69) | 72(12.41) | ||
| Others | 37(6.55) | 33(5.69) | ||
| Disclosure | Yes | 452(80) | 530(91.38) | <0.001 |
| No | 113(20) | 50(8.62) | ||
| Caregiver | Yes | 446(78.94) | 454(78.28) | 0.785 |
| No | 119 (21.06) | 126 (21.72) | ||
| Cell phone | Yes | 444(78.58) | 519(89.48) | 0.048 |
| No | 121(21.42) | 61 (10.52) |
Baseline clinical and treatment-related characteristics
The proportion of PLHIV commencing ART with CD4> 350 cells/increased from pre-treat all period to treat all period (5.84% vs. 53.97%, p < 0.001). The median baseline CD4 count of 195 (IQR = 126–310) in pre-treat all cohorts was lower than 452 (IQR = 342–550) in treat all cohort participants(p<0.001). The majority 316 (54.48%) of participants in the treat-all cohort began treatment at baseline WHO stage II, whereas more than two-thirds (66.55%) of pre-treat all cohort participants had initiated treatment at baseline WHO stage III. Regarding Prophylaxis of opportunistic infection (94.69% vs. 93.62%) participants received CPT respectively. More patients in the treat-all cohort started ART when they were functionally working (54.51% vs. 62.59%, p <0.001) (Table 2).
Table 2. Baseline clinical and treatment status of adult HIV patients on ART before and after treat all strategies in Hawassa City between Jan.2012 and Dec.2021.
| Variables | Category | Pre-treat-all n = 565 | Treat-all n = 580 | p-value |
|---|---|---|---|---|
| Baseline CD4 count | <200 cell/m3 | 299 (52.92) | 71 (12.24) | <0.001 |
| 201–350 cell/m3 | 233 (41.24) | 196 (33.79) | ||
| >350 cell/m3 | 33 (5.84) | 313 (53.97) | ||
| WHO stage | I | 13(2.30) | 203(35.0) | <0.001 |
| II | 117(20.71) | 316(54.48) | ||
| III | 376(66.55) | 37(6.37) | ||
| IV | 59(10.44) | 24(4.14) | ||
| Opportunistic infection | Yes | 188(33.27) | 191(32.93) | 0.902 |
| No | 377(66.73) | 389(67.07) | ||
| Baseline TB | Yes | 45 (7.96) | 46(7.93) | 0.728 |
| No | 520 (92.04) | 534(92.7) | ||
| Baseline functional status | Working | 308(54.51) | 364 (62.76) | <0.001 |
| Ambulatory | 126 (22.30) | 187(32.24) | ||
| Bedridden | 131(23.19) | 29(5.00) | ||
| CPT given | Yes | 535 (94.69) | 543 (93.62) | 0.262 |
| No | 30 (5.31) | 37 (6.38) | ||
| IPT given | Yes | 481 (85.13) | 546(94.14) | <0.001 |
| No | 84 (14.87) | 34(5.86) | ||
| Initial Regimen | TDF-3TC-EFV | 372(65.84) | 379(65.34) | 1.00 |
| AZT-3TC-NVP | 118(20.88) | 119(20.51) | ||
| AZT-3TC-EFV | 48(8.50) | 52(8.96) | ||
| TDF-3TC-NVP | 27(4.78) | 30(5.17) | ||
| Drug adherence | Good | 384(67.96) | 538(92.76) | <0.001 |
| Fair | 62(10.97) | 26(4.48) | ||
| Poor | 119(21.06) | 16 (2.76) | ||
| Baseline BMI | 18.5–24.9 | 424(75.04) | 436 (75.17) | 0.770 |
| <18.5 | 129(22.83) | 130 (22.41) | ||
| >or = 25 | 12 (2.12) | 14 (2.41) | ||
| Comorbidity | Yes | 172(30.44) | 176(30.34) | 0.971 |
| No | 393(69.56) | 404(69.66) |
The incidence of LTFU
The participants were followed for 1279.27 and 1406.78 person-years. The median follow-up was 26.49 months (IQR: 17.03–37.93) and 29.53 months (IQR: 17.16–40.38) for the pre-treat-all and treat-all cohorts, respectively. In the pre-treat-all cohort, 63 (11.15%; 95% CI: 8.8, 14) lost, 75 (13.27%; 95% CI: 10.7, 16.3) died, and 427 (75.58%; 95% CI: 71.8, 78.9) were censored during the five-year study period. While in the treat-all cohort, 122 (21.03%; 17.9, 24.5) lost, 43 (7.41%; 95% CI: 5.5, 9.8) died, and 415 (71.5; 95% CI: 67.7, 75.0) were censored. The cumulative incidence of LTFU was 4.92(3.84,6.3) and 8.67(7.26,10.3) per 100 person-years (PY) in pre-treat all and treat all cohorts, respectively. The cumulative incidence of mortality was 5.86(4.67,7.35) and 3(2.26,4.12) per 100 PY in pre-treat and treat all cohorts, respectively (Fig 1).
Fig 1. The cumulative incidence curve of LTFU and death of adult HIV patients by treatment strategy (pre-treat-all versus treat-all) in Hawassa City July 2022.
The incidence of LTFU in patients with baseline CD4 counts > 350 cell/m3 was 10 per 100 person-year observation it was 6.5 and 4.7 per 100 person-year for CD4 < 200 cell/m3 and 200–350 cell/m3, respectively. While the incidence of mortality among patients with baseline CD4 counts < 200 cell/m3 was 9.2 per 100 person-year observation it was 3 and 1.8 per 100 person-days for CD4 >350 cell/m3 and 200–350 cell/m3, respectively (Fig 2).
Fig 2. The cumulative incidence function of LTFU and death based on adherence to care and CD4 cell count at ART initiation using ART data of public health facilities in Hawassa City Jan 2012 to Dec 2021.
The difference between the two estimates of the cumulative probability of LTFU (one minus Kaplan—Meier and cumulative incidence by considering competing for the effect of death) showed that the 1-km from the standard Cox model overestimates the probability of experiencing LTFU at each subsequent follow-up time, except in the case when there is no competing event at the beginning of the follow-up period (Fig 3).
Fig 3. The estimate of cumulative incidence curves of LTFU by standard Cox model and competing risk regression model.
Predictors of LTFU in pre-treat-all cohort
In multivariable competing risk analysis revealed: that the sub-hazard of LTFU among adults whose ART drug adherence was fair/poor was about 5 times higher than those with good adherence (aSHR: 5.17; (95% CI 1.97, 13.51). The sub-hazard of LTFU among adults who started ART at baseline WHO stage III/ IV was 2.7 times higher than those who started with baseline WHO stage I/II (aSHR:2.69; 95% CI: 1.27, 5.71). The sub-hazard of LTFU among adults who did not take CPT was 3 times higher than those who took CPT (aSHR: 3.05; (95% CI 1.27, 7.34). Furthermore, the sub-hazard of LTFU was about 2 times higher among adults who started ART with baseline underweight (BMI<18.5) compared to those who started at normal nutritional status (BMI = 18.5–24.999) (aSHR:2.13; 95% CI: 1.15–3.93) (Table 3).
Table 3. Bivariable and multivariable competing risk regression for predictors of LTFU among adult HIV patients taking ART in pre-treat all period in Hawassa city public health facilities, Hawassa, Ethiopia, 2022.
| Variables | Category | pre-treat all | ||||
|---|---|---|---|---|---|---|
| Survival status | cSHR (95% CI) | aSHR (95% CI) | ||||
| Death | LTFU | Censored | ||||
| Marital status | Single | 18 | 29 | 89 | 3.08(1.77,5.33) ** | 0.95(0.40,2.23) |
| Married | 27 | 23 | 281 | 1 | 1 | |
| Divorced | 30 | 11 | 57 | 1.36(.67, 2.77) | 0.51(0.21,1.24) | |
| Disclosure | Yes | 56 | 27 | 369 | 1 | 1 |
| No | 19 | 36 | 58 | 6.10(3.72,10.0) ** | 2.82(0.60,13.7) | |
| Base functional status | Working | 8 | 15 | 285 | 1 | 1 |
| Ambulatory | 12 | 15 | 99 | 2.27 (1.10, 4.66) * | 1.0(0.44,2.38) | |
| Bedridden | 55 | 33 | 43 | 4.13(2.25,5.57) ** | 0.88(0.37,2.09) | |
| CPT given | Yes | 72 | 48 | 415 | 1 | |
| No | 3 | 15 | 12 | 6.36(3.65,11.0) ** | 3.05(1.27,7.34) | |
| IPT given | Yes | 58 | 43 | 380 | 1 | 1 |
| No | 17 | 20 | 47 | 2.84(1.68,4.80) ** | 0.64(.27,1.54) | |
| Drug Adherence | Good | 13 | 12 | 359 | 1 | |
| Fair or poor | 62 | 51 | 68 | 8.70(4.64,16.29) ** | 5.17(1.97,13.5) ** | |
| Baseline WHO stage | stage I or II | 15 | 8 | 165 | 1 | 1 |
| stage III or IV | 60 | 55 | 262 | 2.96(1.40, 6.23) * | 2.69(1.27,5.71) * | |
| Baseline CD4 count cell/m3 | <200 | 59 | 42 | 198 | 1.77(1.03,.3.04) | 1.34(.67,2.67) |
| 201–350 | 10 | 19 | 204 | 1 | 1 | |
| >350 | 6 | 2 | 25 | 0.78(.17,3.45) | 0.54(0.12,2.34) | |
| Age | 15–24 | 18 | 53 | 152 | 0.52 (.28,.94) * | 1.45(0.61,3.43) |
| 25–34 | 31 | 68 | 280 | 0.31 (.16,.61) * | 2.00(0.61,6.55) | |
| 35–44 | 34 | 41 | 265 | 0.27 (.10,.68) * | 1.46(0.38,5.63) | |
| >/ = 45 | 35 | 20 | 148 | 1 | 1 | |
| BMI | Normal | 34 | 25 | 361 | 1 | 1 |
| Underweight | 40 | 37 | 57 | 4.33(2.61, 7.20) * | 2.13(1.15,3.93) ** | |
| Overweight | 1 | 1 | 9 | 1.49(.20–10.85) | 1.87 (0.32,10.8) | |
| Caregiver | Yes | 55 | 26 | 365 | 1 | 1 |
| No | 20 | 37 | 62 | 5.77(3.51,9.49) ** | 1.15(0.24,5.35) | |
| Cell phone | Yes | 59 | 38 | 347 | 1 | |
| No | 16 | 25 | 80 | 2.55(1.54,4.21) ** | 0.81(0.38,1.72) | |
**significant at p-value <0.001
*significant at p-value <0.05,
CSHR = crude sub hazard ratio, ASHR = adjusted sub hazard ratio.
Predictors Of LTFU in treat-all cohort
In multivariable competing risk analysis revealed: that the sub-hazard of LTFU among patients whose ART drug adherence was fair/poor was 2 times higher than those with good adherence (aSHR: 2.07; 95% CI: 1.18, 3.65). The sub-hazard of LTFU among adults who started ART at baseline underweight (BMI<18.5) was 1.71 times higher compared to those who started at normal nutritional status (BMI = 18.5–24.999) (aSHR:1.71; 95% CI: 1.13, 2.56). Furthermore, the sub-hazard of LTFU among adults who started ART at baseline CD4 cell counts of >350cell/m3 was 1.61 times higher than those started at baseline CD4 cell count of 201-350cell/m3 (aSHR: 1.67; 95% CI: 1.05–2.65) (Table 4).
Table 4. Bivariable and multivariable competing risk regression analysis for predictors of LTFU among adult HIV patients on ART in treat-all periods in Hawassa city public health facilities from Jan1, 2017 to Dec 31, 2021.
| Variables | Category | Treat all | ||||
|---|---|---|---|---|---|---|
| Survival status | cSHR (95% CI) | aSHR (95% CI) | ||||
| Death | LTFU | Censored | ||||
| Sex | Male | 19 | 62 | 185 | 1.26 (0.8, 1.79) * | 1.28 (.88, 1.85) |
| Female | 24 | 60 | 230 | 1 | 1 | |
| Marital status | Single | 22 | 42 | 146 | 0.83(0.56, 1.23) | 0.83(.56, 1.23) |
| Married | 17 | 66 | 207 | 1 | 1 | |
| Divorced | 4 | 14 | 62 | 0.68(.38, 1.19) * | 0.72(.39, 1.33) | |
| Disclosure | Yes | 36 | 106 | 388 | 1 | 1 |
| No | 7 | 16 | 27 | 1.49 (0.90,2.46) * | 1.20(.68, 2.12) | |
| Base functional status | Working | 29 | 69 | 266 | 1 | 1 |
| Ambulatory | 10 | 42 | 135 | 1.25 (.85, 1.85) | 1.27 (.83, 1.92) | |
| Bedridden | 4 | 11 | 14 | 2.22(1.20,4.11) ** | 1.97 (.80, 4.88) | |
| CPT given | Yes | 4 | 13 | 23 | 1 | |
| No | 39 | 109 | 392 | 1.91 (1.09,3.34) * | 1.21(.58, 2.50) | |
| IPT given | Yes | 5 | 10 | 18 | 1 | |
| No | 38 | 112 | 397 | 1.60 (.85, 3.01) * | 1.12 (.45, 2.77) | |
| Drug adherence | Good | 42 | 106 | 390 | 1 | |
| Fair or poor | 1 | 16 | 25 | 2.10 (1.25,3.54) * | 2.07(1.18, 3.65) * | |
| Baseline WHO stage | I or II | 31 | 102 | 386 | 1 | 1 |
| III or IV | 12 | 20 | 29 | 1.67 (1.04, 2.68) * | 1.76 (0.96, 3.29) | |
| Baseline CD4 count | <200 cell/m3 | 16 | 11 | 44 | 0.93(.46,1.84) | 0.57(.24, 1.34) |
| 201-350cell/m3 | 8 | 31 | 157 | 1 | 1 | |
| >350 cell/m3 | 19 | 80 | 214 | 1.71(1.13,2.59) * | 1.67 (1.05, 2.65) * | |
| Age | 15–24 | 7 | 33 | 97 | 1.90(1.07, 3.36) * | 1.78 (.97, 3.26) |
| 25–34 | 7 | 46 | 130 | 1.78 (0.97, 3.27) * | 1.75 (.93, 3.28) | |
| 35–44 | 15 | 29 | 102 | 1.39 (.75, 2.57) | 1.35 (.71, 2.57) | |
| >/ = 45 | 14 | 14 | 86 | 1 | 1 | |
| BMI | Normal | 29 | 82 | 325 | 1 | 1 |
| Underweight | 13 | 38 | 79 | 1.66(1.13, 2.43) * | 1.71(1.13, 2.56) * | |
| Overweight | 1 | 2 | 11 | 0.80(.18, 3.55) | 0.89(0.23, 3.68) | |
**significant at p-value <0.001
*significant at p-value <0.05,
CSHR = crude sub hazard ratio, aSHR = adjusted sub hazard ratio.
Discussion
This study investigated the incidence and predictors of LTFU in HIV-positive adults initiated on ART before and after the implementation of a test-and-treat strategy. As illustrated in Fig 3, the cumulative incidence function estimates the valid rate of LTFU, adjusting for mortality’s competing effect, as those who die are no longer at risk, which is in line with other similar studies [42, 43]. The estimated rate of LTFU in this study was 4.92 and 8.67 per 100 person-years in the pre-treat-all and treat-all cohorts, respectively.
The incidence of LTFU was 8.67 per 100 person-years in the treat-all period was consistent with 8.9 per 100 person-year reported from North Shewa zone, public hospitals [44]. However, it was lower than the previous studies conducted at Felege Hiwot Comprehensive Specialized Hospital, Northwest Ethiopia [45] and Burundi [46]. The differences might be due to the differences in baseline clinical, the occurrence of opportunistic infection, and nutritional characteristics of the patients, as low immunological status was a key criterion for treatment initiation in the pre-treat-all era, impacting care retention.
The LTFU rate increased significantly in the treat-all strategy compared to the pre-treat-all era, emphasizing the need for ongoing monitoring and follow-up. It is essential to implement a patient-centered approach and transportation assistance to retain patients on the program, particularly those who are more at risk for LTFU [47, 48]. Retention in care is essential for achieving optimal health outcomes and reducing the spread of HIV within communities [49].
This study found the advanced WHO clinical stage of diseases at baseline was the strongest predictor of LTFU. This finding was similar to findings reported in the Hadiya zone, Pawi General Hospital, Arbaminch General Hospital, and Gondar University Hospital [4, 15, 30]. This can be explained by the fact that in advanced WHO clinical disease, patients with lowered immunity face increased opportunistic infections and are bedridden, leading to severe illness and unreported deaths.
Underweight individuals are at a higher risk of developing LTFU. This is in line with different previous studies in Ethiopia [20, 50] and Central Kenya [50]. This is because undernourished individuals with HIV and AIDS can progress diseases to severe stages, which increases the occurrence of opportunistic infections. This dual burden can lead to being bedridden and ultimately dying [51].
The sub-hazard of LTFU among respondents with fair or poor adherence was higher than among respondents with good adherence. This is comparable to findings from other similar studies [15, 19]. The reason for this may be due to low awareness of lifelong drug adherence and insufficient physiological preparation for long-term treatment follow-up, which may result in high viral loads, treatment failure, increased risk of OIs, and ultimately mortality in ART patients [6]. Therefore, maintaining adherence and retention in care is crucial to determining the optimal benefits of ART programs.
Furthermore, individuals who started ART with a high CD4 count were more likely to develop LTFU. The finding was in line with other similar studies [52, 53]. However, this finding is in contrast to many studies that reported poor immunologic status (low CD4+ cell count) at ART initiation as a predictor of increased hazard of LTFU [12, 14, 54, 55]. Patients with a good immunological state may fail to adhere to early induction programs because they feel well and do not think about their medication needs. Lack of psychological preparation before treatment initiation could also be a factor. Therefore, targeted interventions and support may be needed to reduce LTFU and promote the potential benefits of early ART introduction at higher CD4 counts.
Limitations of the study
The findings of this study might suffer from the fact that it is a retrospective study. Predictors like viral load, hemoglobin level, clinical factors, institution-level factors, and contextual factors were not included as they were not well documented on patient cards. Moreover, the effect of silent transfer out and unreported death was not considered, and the distance from health facilities was reported as one factor for loss to follow-up but not addressed.
Conclusion
The findings of this study showed that the incidence of loss to follow-up was significantly higher during the treat-all era than in the pre-treat-all era. Underweight (BMI <18.5), WHO stages III and IV, fair or poor adherence, and CD4 cells > 350 cells in the treat-all cohort were independent predictors of loss to follow-up, as were underweight (BMI <18.5), fair or poor adherence, and underweight in the pre-treat all cohort. Therefore, targeted interventions, such as nutritional support and strengthening medication adherence counseling, should be implemented to maintain treatment retention and reduce antiretroviral therapy dropout rates.
Supporting information
(DTA)
(DOCX)
(DOCX)
(DOCX)
Acknowledgments
We would like to say thanks to Wolaita Sodo University College of Health Sciences and Medicine for facilitating this study. Our special thanks go to the data collectors for their kind and excellent cooperation during data collection.
Abbreviations
- AIDS
Acquired Immunodeficiency Syndrome
- ART
Antiretroviral therapy
- CD4
Cluster of Differentiation 4
- CPT
Cotrimoxazole prophylactic therapy
- IPT
Isoniazid preventive therapy
- LTUF
loss to follow up
- sAHR
sub-distributed Adjusted hazard ratio
Data Availability
All relevant data are within the paper and its Supporting information files.
Funding Statement
The author(s) received no specific funding for this work.
References
- 1.ICAP. Facality level standard oprating procedures for HIV care and treatment service. 2018.
- 2.Ayele W, Mulugeta A, Desta A, Rabito FA. Treatment outcomes and their determinants in HIV patients on Anti-retroviral Treatment Program in selected health facilities of Kembata and Hadiya zones, Southern Nations, Nationalities and Peoples Region, Ethiopia. BMC public health. 2015;15:826. doi: 10.1186/s12889-015-2176-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gezae KE, Abebe HT, Gebretsadik LG. Incidence and predictors of LTFU among adults with TB/HIV co-infection in two governmental hospitals, Mekelle, Ethiopia, 2009–2016: survival model approach. BMC infectious diseases. 2019;19(1):107. Epub 2019/02/06. doi: 10.1186/s12879-019-3756-2 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mekonnen N, Abdulkadir M, Shumetie E, Baraki A, Yenit M. Incidence and predictors of loss to follow-up among HIV infected adults after initiation of first line anti-retroviral therapy at University of Gondar comprehensive specialized Hospital Northwest Ethiopia, 2018: retrospective follow up study BMC research notes. 2019;12(1):111. Epub 2020/09/19. doi: 10.1186/s13104-019-4154-y . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mekuria L, Prins J, Yalew A, Sprangers M, Nieuwkerk P. Retention in HIV Care and Predictors of Attrition from Care among HIV-Infected Adults Receiving Combination Anti-Retroviral Therapy in Addis Ababa. PloS one. 2015;10(6):e0130649. doi: 10.1371/journal.pone.0130649 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Koole O, Denison JA, Menten J, Tsui S, Wabwire-Mangen F, Kwesigabo G, et al. Reasons for Missing Antiretroviral Therapy: Results from a Multi-Country Study in Tanzania, Uganda, and Zambia. PloS one. 2016;11(1):e0147309. doi: 10.1371/journal.pone.0147309 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.UNSAIDS. UNAIDS Data 2018. Epub 2018.
- 8.Asiimwe SB, Kanyesigye M, Bwana B, Okello S, Muyindike W. Predictors of dropout from care among HIV-infected patients initiating antiretroviral therapy at a public sector HIV treatment clinic in sub-Saharan Africa. BMC Infect Dis 2015;16(14). doi: 10.1186/s12879-016-1392-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Leshargie CT, Demant D, Burrowes S, Frawley J. The proportion of loss to follow-up from antiretroviral therapy (ART) and its association with age among adolescents living with HIV in sub-Saharan Africa: A systematic review and meta-analysis. PloS one. 2022;17(8):e0272906. doi: 10.1371/journal.pone.0272906 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhou J, Tanuma J, Chaiwarith R, Lee CK, Law MG, Kumarasamy N, et al. Loss to followup in HIV-infected patients from Asia-Pacific region: results from TAHOD. AIDS research and treatment. 2012;2012. doi: 10.1155/2012/375217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Birhanu M, Leshargie C, Alebel A, Wagnew F, Siferih M. Incidence and predictors of loss to follow- up among HIV-positive adults in northwest Ethiopia: a retrospective cohort study BMC research notes. 2019;12(1):111. Epub 2020/09/19. doi: 10.1186/s41182-020-00266-z . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Seifu W, Ali W, Meresa B. Predictors of loss to follow up among adult clients attending antiretroviral treatment at Karamara general hospital, Jigjiga town, Eastern Ethiopia, 2015: a retrospective cohort study. BMC infectious diseases. 2018;18(1):280. Epub 2016/05/14. doi: 10.1186/s12879-018-3188-4 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Telele NF, Kalu AW, Marrone G, Gebre-Selassie S, Fekade D, Tegbaru B, et al. Baseline predictors of antiretroviral treatment failure and lost to follow up in a multicenter countrywide HIV-1 cohort study in Ethiopia. PloS one. 2018;13(7):e0200505. doi: 10.1371/journal.pone.0200505 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Melaku Z, Lamb MR, Wang C, Lulseged S, Gadisa T, Ahmed S, et al. Characteristics and outcomes of adult Ethiopian patients enrolled in HIV care and treatment: a multi-clinic observational study. BMC public health. 2015;15:462. Epub 2015/05/03. doi: 10.1186/s12889-015-1776-4 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Assemie MA, Muchie KF, Ayele TA. Incidence and predictors of loss to follow up among HIV-infected adults at Pawi General Hospital, northwest Ethiopia: competing risk regression model. BMC research notes. 2020;15(1):e0227473. Epub 2018/05/12. doi: 10.1186/s13104-018-3407-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Teasdale CA, Yuengling K. Persons living with HIV with advanced HIV disease: need for novel care models. 2018;21(12):e25210. doi: 10.1002/jia2.25210 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gesesew HA, Ward P, Woldemichael K, Mwanri L. Prevalence, trend and risk factors for antiretroviral therapy discontinuation among HIV-infected adults in Ethiopia in 2003–2015. PloS one. 2017;12(6):e0179533. Epub 2017/06/18. doi: 10.1371/journal.pone.0179533 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Abrams EJ, El-Sadr WM, Teshome W, Belayneh M, Moges M, Mekonnen E, et al. Do loss to follow-up and death rates from ART care vary across primary health care facilities and hospitals in south Ethiopia? A retrospective follow-up study. Journal of the International AIDS Society. 2015;7:167–74. Epub 2018/12/15. doi: 10.2147/hiv.s85440 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dejen D, Jara D, Yeshanew F, Fentaw Z, Mengie Feleke T, Girmaw F, et al. Attrition and Its Predictors Among Adults Receiving First-Line Antiretroviral Therapy in Woldia Town Public Health Facilities, Northeast Ethiopia: A Retrospective Cohort Study. Hiv/Aids. 2021;13:445–54. doi: 10.2147/HIV.S304657 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dessu S, Mesele M, Habte A, Dawit Z. Time Until Loss to Follow-Up, Incidence, and Predictors Among Adults Taking ART at Public Hospitals in Southern Ethiopia. 2021;13:205–15. doi: 10.2147/hiv.s296226 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bucciardini R, Fragola V, Abegaz T, Lucattini S, Halifom A, Tadesse E, et al. Predictors of attrition from care at 2 years in a prospective cohort of HIV-infected adults in Tigray, Ethiopia. BMJ global health. 2017;2(3):e000325. Epub 2017/10/31. doi: 10.1136/bmjgh-2017-000325 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Akilimali P, Musumari P, Kashala-Abotnes E, Kayembe P, Lepira F, Mutombo P. Disclosure of HIV status and its impact on the loss in the follow-up of HIV-infected patients on potent anti-retroviral therapy programs ina (post-) conflict setting: A retrospective cohort study from Goma, Democratic Republic of Congo. PloS one. 2017. doi: 10.1371/journal.pone.0171407 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Eshun-Wilson I, Rohwer A, Hendricks L, Oliver S, Garner P. Being HIV positive and staying on antiretroviral therapy in Africa: A qualitative systematic review and theoretical model. PloS one. 2019;14(1):e0210408. doi: 10.1371/journal.pone.0210408 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Tiruneh YM, Omar G, Becky G, Wilson IB. Retention in Care among HIV-Infected Adults in Ethiopia, 2005–2011: A Mixed-Methods Study. PloS one. 2016. doi: 10.1371/journal.pone.0156619 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kamacooko O, Mayanja Y, Bagiire D, Namale G, Hansen CH, Seeley J. Predictors of lost to follow-up in a “test and treat” programme among adult women with high-risk sexual behavior in Kampala, Uganda. BMC public health. 2020;20(1):1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Onoka C, Uzochukwu B, Onwujekwe O, Chukwuka C, Ilozumba J, Onyedum C, et al. Retention and loss to follow-up in antiretroviral treatment programmes in southeast Nigeria. Pathogens and global health. 2012;106(1):46–54. doi: 10.1179/2047773211Y.0000000018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mberi MN, Kuonza LR, Dube NM, Nattey C, Manda S, Summers R. Determinants of loss to follow-up in patients on antiretroviral treatment, South Africa, 2004–2012: a cohort study. BMC health services research. 2015;15:259. doi: 10.1186/s12913-015-0912-2 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tweya H, Oboho IK, Gugsa ST, Phiri S, Rambiki E, Banda R, et al. Loss to follow-up before and after initiation of antiretroviral therapy in HIV facilities in Lilongwe, Malawi. PloS one. 2018;13(1):e0188488. doi: 10.1371/journal.pone.0188488 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Berheto TM, Haile DB, Mohammed S. Predictors of Loss to follow-up in Patients Living with HIV/AIDS after Initiation of Antiretroviral Therapy. North American journal of medical sciences. 2014;6(9):453–9. Epub 2018/06/20. doi: 10.4103/1947-2714.141636 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bikoro B, Oljira L, Gobena T, Erkalo D. Incidence and predictors of loss to follow-up among human immunodeficiency virus-infected adult patients on anti-retroviral therapy at Hadiya zone public hospitals, southern Ethiopia: a retrospective cohort study. Journal of Public Health. 2020;30(1):229–40. doi: 10.1007/s10389-020-01268-1 [DOI] [Google Scholar]
- 31.FMOH. National guidelines for comprehensive HIV prevention, care and treatment 2017. 2017.
- 32.Girum T, Yasin F, Dessu S, Zeleke B, Geremew M. "Universal test and treat" program reduced TB incidence by 75% among a cohort of adults taking antiretroviral therapy (ART) in Gurage zone, South Ethiopia. Tropical diseases, travel medicine and vaccines. 2020;6:12. Epub 2020/08/31. doi: 10.1186/s40794-020-00113-3 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kiwanuka J, Mukulu Waila J, Muhindo Kahungu M, Kitonsa J, Kiwanuka N. Determinants of loss to follow-up among HIV positive patients receiving antiretroviral therapy in a test and treat setting: A retrospective cohort study in Masaka, Uganda. PloS one. 2020;15(4):e0217606. doi: 10.1371/journal.pone.0217606 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Assefa Y, Gilks CF. Ending the epidemic of HIV/AIDS by 2030: Will there be an endgame to HIV, or an endemic HIV requiring an integrated health systems response in many countries? International Journal of Infectious Diseases. 2020;100:273–7. doi: 10.1016/j.ijid.2020.09.011 [DOI] [PubMed] [Google Scholar]
- 35.FMoH. National guidelines for comprehensive hiv prevention, care and treatment. 2017.
- 36.Teshale AB, Tsegaye AT, Wolde HF. Incidence and predictors of loss to follow up among adult HIV patients on antiretroviral therapy in University of Gondar Comprehensive Specialized Hospital: A competing risk regression modeling. PloS one. 2020;15(1):e0227473. doi: 10.1371/journal.pone.0227473 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Austin PC, Lee DS, Fine JP. Introduction to the Analysis of Survival Data in the Presence of Competing Risks. Circulation. 2016;133(6):601–9. Epub 2016/02/10. doi: 10.1161/CIRCULATIONAHA.115.017719 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.He PE, Frank, Scheike Thomas H. Zhang Mei-Jie. A Proportional Hazards Regression Model for the Subdistribution with Covariates-adjusted Censoring Weight for Competing Risks Data. Scandinavian Journal of Statistics. 2016;43(1):103–22. doi: 10.1111/sjos.12167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Donoghoe MW, Gebski V. The importance of censoring in competing risks analysis of the subdistribution hazard. BMC medical research methodology. 2017;17(1):52. doi: 10.1186/s12874-017-0327-3 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Aberg JA, Gallant JE, Ghanem KG, Emmanuel P, Zingman BS, Horberg MA. Primary care guidelines for the management of persons infected with HIV: 2013 update by the HIV Medicine Association of the Infectious Diseases Society of America. Clinical Infectious Diseases. 2014;58(1):e1–e34. doi: 10.1093/cid/cit665 [DOI] [PubMed] [Google Scholar]
- 41.WHO. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach, 2nd ed. 2016. [PubMed] [Google Scholar]
- 42.Assemie MA, Muchie KF, Ayele TA. Incidence and predictors of loss to follow up among HIV-infected adults at Pawi General Hospital, northwest Ethiopia: competing risk regression model. BMC research notes. 2018;11(1):287. doi: 10.1186/s13104-018-3407-5 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Matsena Zingoni Z, Chirwa T, Todd J, Musenge E. Competing risk of mortality on loss to follow-up outcome among patients with HIV on ART: a retrospective cohort study from the Zimbabwe national ART programme. BMJ open. 2020;10(10):e036136. Epub 2020/10/09. doi: 10.1136/bmjopen-2019-036136 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Shiferaw WS, Belete AM, Adela A, Getnet M, Aynalem YA. Incidence and predictors of loss to follow-up among adult HIV-infected patients taking antiretroviral therapy at North Shewa zone public Hospitals, Northeast Ethiopia: A retrospective follow-up study. Afr Health Sci. 2022;22(2):12–26. doi: 10.4314/ahs.v22i2.3 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Bantie B, Seid A, Kerebeh G, Alebel A, Dessie G. Loss to follow-up in "test and treat era" and its predictors among HIV-positive adults receiving ART in Northwest Ethiopia: Institution-based cohort study. Front Public Health. 2022;10:876430. doi: 10.3389/fpubh.2022.876430 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Nshimirimana C, Ndayizeye A, Smekens T, Vuylsteke B. Loss to follow-up of patients in HIV care in Burundi: A retrospective cohort study. Tropical Medicine & International Health. 2022;27(6):574–82. doi: 10.1111/tmi.13753 [DOI] [PubMed] [Google Scholar]
- 47.Rachlis B, Bakoyannis G, Easterbrook P, Genberg B, Braithwaite RS, Cohen CR, et al. Facility-Level Factors Influencing Retention of Patients in HIV Care in East Africa. PloS one. 2016;11(8):e0159994. Epub 2016/08/11. doi: 10.1371/journal.pone.0159994 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Chinyandura C, Jiyane A, Tsalong X, Struthers HE, McIntyre JA, Rees K. Supporting retention in HIV care through a holistic, patient-centred approach: a qualitative evaluation. BMC Psychology. 2022;10(1):17. doi: 10.1186/s40359-022-00722-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Holtzman CW, Brady KA, Yehia BR. Retention in care and medication adherence: current challenges to antiretroviral therapy success. Drugs. 2015;75(5):445–54. Epub 2015/03/21. doi: 10.1007/s40265-015-0373-2 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Wekesa P, McLigeyo A, Owuor K, Mwangi J, Nganga E, Masamaro K. Factors associated with 36-month loss to follow-up and mortality outcomes among HIV-infected adults on antiretroviral therapy in Central Kenya. BMC public health. 2020;20(1):328. doi: 10.1186/s12889-020-8426-1 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Naidoo K, Yende-Zuma N, Augustine S. A retrospective cohort study of body mass index and survival in HIV infected patients with and without TB co-infection. Infectious diseases of poverty. 2018;7(1):35. doi: 10.1186/s40249-018-0418-3 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Grimsrud A, Cornell M, Schomaker M, Fox MP, Orrell C, Prozesky H, et al. CD4 count at antiretroviral therapy initiation and the risk of loss to follow-up: results from a multicentre cohort study. Journal of epidemiology and community health. 2016;70(6):549–55. doi: 10.1136/jech-2015-206629 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Schoni-Affolter F, Keiser O, Mwango A, Stringer J, Ledergerber B, Mulenga L, et al. Estimating loss to follow-up in HIV-infected patients on antiretroviral therapy: the effect of the competing risk of death in Zambia and Switzerland. PloS one. 2011;6(12):e27919. doi: 10.1371/journal.pone.0027919 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Amera TG, Bogale KA, Tefera YM. Time to initial highly active antiretroviral therapy discontinuation and its predictors among HIV patients in Felege Hiwot comprehensive specialized hospital: a retrospective cohort study. AIDS research and therapy. 2021;18(1):93. doi: 10.1186/s12981-021-00418-z . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Arnesen R, Moll AP, Shenoi SV. Predictors of loss to follow-up among patients on ART at a rural hospital in KwaZulu-Natal, South Africa. PloS one. 2017;12(5):e0177168. doi: 10.1371/journal.pone.0177168 . [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(DTA)
(DOCX)
(DOCX)
(DOCX)
Data Availability Statement
All relevant data are within the paper and its Supporting information files.