Long‐term loss to follow‐up in the TREAT Asia HIV Observational Database (TAHOD)

Abstract

ObjectivesWith earlier antiretroviral therapy (ART) initiation, time spent in HIV care is expected to increase. We aimed to investigate loss to follow-up (LTFU) in Asian patients who remained in care 5 years after ART initiation. MethodsLong-term LTFU was defined as LTFU occurring after 5 years on ART. LTFU was defined as (1) patients not seen in the previous 12 months; and (2) patients not seen in the previous 6 months. Factors associated with LTFU were analysed using competing risk regression. ResultsUnder the 12-month definition, the LTFU rate was 2.0 per 100 person-years (PY) [95% confidence interval (CI) 1.8-2.2 among 4889 patients included in the study. LTFU was associated with age > 50 years [sub-hazard ratio (SHR) 1.64; 95% CI 1.17-2.31] compared with 31-40 years, viral load ≥ 1000 copies/mL (SHR 1.86; 95% CI 1.16-2.97) compared with viral load < 1000 copies/mL, and hepatitis C coinfection (SHR 1.48; 95% CI 1.06-2.05). LTFU was less likely to occur in females, in individuals with higher CD4 counts, in those with self-reported adherence ≥ 95%, and in those living in high-income countries. The 6-month LTFU definition produced an incidence rate of 3.2 per 100 PY (95% CI 2.9-3.4 and had similar associations but with greater risks of LTFU for ART initiation in later years (The long-term LTFU rate in our cohort was low, with older age being associated with LTFU. The increased risk of LTFU with later years of ART initiation in the 6-month analysis, but not the 12-month analysis, implies that there was a possible move towards longer HIV clinic scheduling in Asia.

Predicting the loss to follow‐up (LTFU) of HIV/AIDS patients in China using a recency‐frequency (RF) model

“…In the HIV/AIDS patient data with R ≤ 6 months retained in the previous step, the best predictive model was again used to perform k-means modelling to subdivide the patients into two categories [2,[9][10][11][12][13]: churn patients (marked as 0) and non-churn patients (marked as 1) [4][5][6][7][8]. Then, an ANN was applied to predict the LTFU in the next treatment cycle (October-December 2020) for the two types of patients [24][25][26].…”

“…Numerous studies have reported that the LTFU rates are high for people with complications of advanced HIV, a person with an AIDS-defining illness (AIDS patients) [11][12][13][14]. Dalhatu et al [14] observed that the 1-, 2-, 3-and 4-year cumulative LTFU rates for AIDS patients were 18.8%, 25.6%, 32.8% and 38.3%, respectively.…”

“…Dalhatu et al [14] observed that the 1-, 2-, 3-and 4-year cumulative LTFU rates for AIDS patients were 18.8%, 25.6%, 32.8% and 38.3%, respectively. One of the fundamental reasons for the high LTFU rate is that we still lack a suitable predictive model for LTFU [11]. Hence, we cannot take effective measures before the occurrence of LTFU to prevent it [5,[7][8][9][10][11][13][14][15][16][17][18].…”

“…One of the fundamental reasons for the high LTFU rate is that we still lack a suitable predictive model for LTFU [11]. Hence, we cannot take effective measures before the occurrence of LTFU to prevent it [5,[7][8][9][10][11][13][14][15][16][17][18].…”

Trends in Follow-Up Visits Among People Living With HIV: Results From the TREAT Asia and Australian HIV Observational Databases

“…12 In addition, combined CD4 and VL testing frequency in TAHOD declined from 2 per person-years in 2003-2005 to 1.5 per person-years in 2014-2015. 13 It is not known whether the overall follow-up visit frequency has also declined over the years in both cohorts.…”