The effects of decentralising antiretroviral therapy care delivery on health outcomes for adolescents and young adults in low- and middle-income countries: a systematic review

Figures & data

Figure 1. PRISMA flowchart of studies screened and selected for systematic review.

LMIC: low- and middle-income country.

Table 1. Summary of included studies, listed in reverse chronological order.

Study author(s) and year of publication	Study design	Study location and dates	Total participant population and sample size	Setting	Mode of decentralisation and visit frequency	Cadres of healthcare providers at decentralised sites	Criteria for decentralisation eligibility	Outcomes measured
Scheibe et al., 2013 [37]	Retrospective cohort study	Iganga District, Uganda, April 2004-September 2009	All patients who were diagnosed with HIV and registered on ART in the included facilities during study period, n = 973 (decentralised n = 271) Gender: 63.0% female, 37.0% male Age (median, IQR): 36 (30–43)	1 general hospital and 4 health centres (3 level IV, 1 level III)	ART management (from December 2005) and initiation (from February 2007) Visit frequency: monthly	Nurse, clinical officer, midwife	N/A	Attrition (not attending facility for >90 days before audit, excl. transfers) Effective coverage (ratio of number of patients currently receiving ART: number of people living with HIV needing ART) Probability of retention in care
Chan et al., 2010 [32]	Retrospective cohort study	Zomba District, Malawi, October 2004-December 2008	All patients who initiated ART in the included facilities during study period, n = 8093 (decentralised n = 3440) Gender: 63.0% female, 37.0% male Age (mean, SD): 33.2 (±13.6) in centralised, 35.0 (±16.7) in decentralised	1 tertiary hospital and 16 rural health centres	ART management (from March 2007) and initiation (from April 2008) Visit frequency: monthly for first 6 months after initiation, then as per provider assessment	Physician (visiting), clinical officer, nurse, medical assistant	National criteria: 1. Stable on first-line ART for >3 months 2. No active opportunistic infection or drug intolerance 3. ART provider confidence in patient adherence 4. Patient closer to health centre than hospital	All-cause mortality Defaulting (not seen for >3 months after scheduled follow-up)

IQR: Interquartile range; N/A: Not available; SD: Standard deviation.

Table 2. Included studies’ characteristics and main findings based on age-disaggregated data (10–24 years old).

Study author(s) and year of publication	Study design	Study location and dates	Adolescent and youth sample size	Adolescent and youth characteristics	Outcomes measured	Main results
Scheibe et al., 2013 [37]	Retrospective cohort study	Iganga District, Uganda, April 2004-September 2009	n = 63 (decentralised n = 16)	Gender: 71.4% female Age (median, IQR): 19 (13–22)	Attrition rate Probability of retention on ART	In centralised arm, 55.3% (n = 26) dropped out; in the decentralised arm, 37.5% dropped out (n = 6) Overall probability of retention on ART for youths 10–24 was 0.71 (95%CI 0.58–0.80) at 6 months, 0.62 (95%CI 0.48–0.73) at 12 months, and 0.45 (95%CI 0.31–0.58) at 18 and 24 months Receiving centralised care was not significantly associated with attrition from care in multivariate Cox analysis (aHR 1.26 [95%CI 0.42–3.81], p= 0.681)
Chan et al., 2010 [32]	Retrospective cohort study	Zomba District, Malawi, October 2004-December 2008	n = 1062 (decentralised n = 436)	Gender: 75.1% female Age at initiation (median, IQR): 21 (14–23) in centralised, 21 (14–23) in decentralised	WHO stage at ART initiation (n) Mortality (n) Defaulted (n)	Using crude ORs, there was no significant difference between study arms for WHO Stage I/II vs. Stage III/IV at initiation (OR 0.92 [95%CI 0.70–1.22], p= 0.585) Using crude ORs, decentralised care was significantly associated with a lower rate of mortality (OR 0.14 [95%CI 0.07–0.29], p< 0.001) Using crude ORs, decentralised care was significantly associated with a lower rate of defaulting from care (OR 0.37 [95%CI 0.26–0.55], p< 0.001)

aHR: Adjusted hazard ratio; CI: Confidence interval; IQR: Interquartile range; OR: Odds ratio; WHO: World Health Organisation.

Scheibe FJB, Waiswa P, Kadobera D, et al. Effective coverage for antiretroviral therapy in a Ugandan District with a decentralized model of care. PLoS One. 2013 Jul 23;8:e69433.

 Web of Science ®Google Scholar

WHO. Recognizing adolescence: World Health Organization; 2014 [cited 2018 Jul 23]. Available from: http://apps.who.int/adolescent/second-decade/section2/page1/recognizing-adolescence.html

 Google Scholar