Table 2.

Studies investigating the failure cascade - repeat VL test, EAC and switch to second-line ART

	Author, year	Study Setting (Sampling Year)	Study Population/Size	Age of Participants	Study Design	Primary Outcomes	Results of VL monitoring and cascade outcomes	VL monitoring in national guidance	Risk of Bias
14	Glass et al., 2019	Lesotho (2015–2018)	24,948 participants receiving ART in the district of Butha-Buthe (1 district hospital and 11 rural clinics) (73% female)	Median: 41 years (IQR 33–52)	Prospective cohort	Describing the VL care cascade through follow-up VL testing and decision making based on VL result	- 9,949/24,948 had VL tests (39.9%) - 1028/9,949 (11%) had VL > 1000 copies/ml - 260/1028 (25%) were managed according to guidelines: 410/1028 (40%) had follow-up VL, 260/1028 (25%) resuppressed (VL < 1000 copies/ml) or switched to 2nd line ART	2014: Yes [63]	Low risk
15	Awungafac et al., 2018	Cameroon (2013–2015)	830 participants receiving ART for 18 months, at Limbe and Buea Regional Hospital treatment centres, Bamenda Regional Hospital and Bafoussam Regional Hospital (65% female)	Median: 40 years (IQR 34.5–47.5)	Retrospective cohort	Assessing uptake and utilization of VL tests for clinical decision making	- 201/830 (24.3%) had VL test - 0/201 (0%) had second VL in time on ART - 190/201 (94.5%) had viral suppression (VL < 1000 copies/ml) - 5/11 (45%) with VF were switched to 2nd line therapy	2015: Yes (but ART based on CD4) [64]	Low risk
16	Swannet et al., 2017	Mozambique (2014–2015)	43,579 participants receiving ART for 6 months, at 6 MSF supported health centres in Maputo city (68% female)	15–25 years: 2,220 participants (5.4%) > 25 years: 39,361 (94.6%)	Retrospective cohort	Assessing VL testing scale up through uptake of VL testing and follow-up	- 16,006/41,581 (38.5%) had VL test - 903/2,651 (34%) of those with an initial result > 3000 copies/ml had a follow-up VL test - 558/903 (61.8%) had VL > 3000 copies/ml	Data in Portuguese only	At risk
17	Onyedum et al., 2013	Nigeria (2012)	4229 participants on 1st line ART and 186 participants switched to 2nd line ART, at the HIV care and treatment centre of the University of Nigeria Teaching Hospital (59.1% female)	Mean: 41.8 ± 9.6 years	Retrospective cohort	Investigating reasons behind ART switch and cascade outcomes for those on 2nd line ART	- 186/4,229 (4.4%) switched to 2nd line ART - 147/186 (79%) switched due to virological failure on VL test (VL > 1000 copies/ml) - 93/144 (81.6%) had viral suppression on VL test after switch to 2nd line ART	2010–2015: Yes [65]	Low risk
18	Kehoe et al., 2020	South Africa (2011–2016)	8058 participants on ART and enrolled in adherence clubs in Khayelitsha, Cape Town (74% female)	Median: 39 years (IQR 34–45)	Longitudinal	Assessing proportion of patients with elevated VL result and cascade of care after confirmed virologic failure	- 7,136/8,058 (89%) had VL tests - 6,621/7,136 (93%) had viral suppression (VL < 400 copies/ml) - 150/441 (34%) with virological failure (VL > 1000 copies/ml) had no repeat VL test - 120/291 (41%) with VF were successfully resuppressed with EAC or treatment switch	2007–2011: Yes [57]	At risk
19	Rutstein et al., 2015	Malawi (not stated)	1498 participants on 1st line ART for 6 months, from 5 ART clinics in central and southern Malawi (70.3% female)	Mean: 42.1 years	Prospective cohort	Determining feasibility and effectiveness of dried blood spot for VL monitoring	- 1,494/1,498 (99.7%) had VL test - 1406/1498 (93.9%) were virally suppressed (VL > 5000 copies/ml) - 54/88 (61.4%) had confirmed virological failure with VL test - 50/54 (93%) initiated 2nd line ART	2014: Yes [66]	At risk
20	Keiser et al., 2011	South Africa (not stated)	18,706 participants initiating ART, in 4 ART programmes in Khayelitsha, Gugulethu, Tygerberg and Themba Lethu (65.7% female)	Median: 34 years (IQR 30–41)	Prospective cohort	Comparing rates of treatment switch with and without VL monitoring	- 8,892/14,258 (62.4%) of eligible patients at month 6 had VL test - 1,833/18,706 (9.8%) of total patients had switched to 2nd line ART at year 3 - 243/18,706 (1.3%) were on failing 1st line ART at year 3 (VL > 10,000 copies/ml)	2007–2011: Yes [57]	At risk
21	Sunpath et al., 2022	South Africa (2017)	116 participants receiving 1st line ART with VL > 1000 copies/ml, in 3 clinics in Durban (51.5% female)	Median: Pre-implementation 36 years (IQR 23–41) Post-implementation 35 years (IQR 30–39)	Before-after	Reporting the impact of the VL champion on the virological failure cascade of care	- Pre-implementation: 37/60 (61.7%) had a 2nd VL test 3/37 (8.3%) managed according to guidelines 4/37 (10.1%) switched to second line ART - Post-implementation: 29/56 (51.7%) had 2nd VL test 3/29 (10.7%) managed according to guidelines 5/29 (16.1%) changed to 2nd line ART	2017–2022: Yes [61]	At risk
22	Mnzava et al., 2022	Tanzania (2017–2020)	4454 participants receiving ART for 6 months in the districts of Kilombero and Ulanga, Morogoro region (69% female)	Median: 42 years (IQR 35–51)	Prospective cohort	Assessing the VL monitoring cascade of care and comparing turnaround times	- 4,238/4454 (95%) had a VL test - 3,683/4,238 (88%) had viral suppression (VL < 1000 copies/ml) - 425/494 (86%) had a follow-up VL after initial failure (VL > 1000 copies/ml) - 32/55 (58%) of those not already on 2nd -line ART were switched to 2nd line ART	2017–2022: Yes [67]	At risk
23	Iwuji et al., 2020	South Africa (2010–2016)	29,384 participants initiating ART in the sub-district of Hlabisa (69.9% female)	Median: 31 years (IQR 25–39)	Retrospective cohort	Determining if guidelines on management of virological failure are being implemented	- 9,861/24,199 (40.7%) 7,765/22,807 (34%) 4,334/16,965 (25.5%) had VL tests at 6, 12 and 24 months respectively - 2,135/19,582 (10%) had a VL > 1000 copies/ml - 658/2,135(30.8%) had a repeat VL test to confirm virological failure - 250/658 (38%) were resuppressed (VL < 1000 copies/ml) and 141/391 (36%) switched to 2nd line ART	2007–2011: Yes [57]	Low risk
24	Nicholas et al., 2019	Malawi (2013–2017)	21,400 participants receiving ART for 3 months in decentralised clinics and the district hospital of Chiradzulu (65% female)	Median: 38 years (IQR 31–46)	Retrospective cohort	Investigating outcomes from the first 4 years of routine VL monitoring using point-of-care testing	- 18,182/21,400 (85%) had a VL test - 16,150/18,182 (89%) had viral suppression after routine VL test - 1,281/1544 (83%) with virological failure (VL > 1000 copies/ml) had a follow-up VL test - 434/540 (80%) with confirmed VL failure on a 3rd VL test were switched to 2nd line ART - 275/347 (79%) were re-suppressed (VL < 1000 copies/ml) on repeat VL test after switch	2014: Yes [66]	At risk
25	Warrier et al., 2019	Zambia (2016–2018)	118,266 participants enrolled in HIV care with routine VL test result in 74 facilities across 3 Zambian provinces (no data on sex of participants)	No data	Retrospective cohort	Assessing each step of the failure cascade in order to help achieve the third “90” in the 90-90-90 goals	- 14,291/118,266 (12.1%) were virally unsuppressed after first VL test - 4,978/14,291 (9.2%) had a follow up VL within 90 days - 2,459/4,978 (49.4%) had virological failure (VL > 1000 copies/ml) - 720/2,459 (29.3%) were switched to 2nd line ART	2014–2016: Yes [68]	At risk
26	Labhardt et al., 2017	Lesotho (2014–2015)	138 participants receiving 1st line ART at 10 rural facilities in Lesotho for 6 months, with a VL > 80 copies/ml (65.9% female)	Median: 41.1 years (IQR 32.4–49.9)	Prospective cohort	Describing the outcomes of the failure cascade for patients with an unsuppressed VL	- 124/138 (90%) received EAC - 116/138 (84%) had a repeat VL test - 36/116 (31%) were virally re-suppressed (VL < 80 copies/ml) - 58/80 (73%) were switched to 2nd line ART	2014: Yes [63]	At risk
27	Etoori et al., 2022	Swaziland (2013–2015)	828 participants receiving ART with a VL > 1000 copies/ml, at 23 primary care clinics in the Shiselweni region, Swaziland (65.5% female)	Median: 35 years (IQR 29–44)	Retrospective cohort	Understanding where the gaps from along the VL cascade of care	- 288/828 (34.8%) received 3 sessions of EAC - 696/828 (84.1%) had a follow up VL test - 410/696 (58.9%) had virological failure (VL > 1000 copies/ml) - 120/278 (43.2%) were switched to 2nd line ART	2009–2014: No [69]	At risk
28	Hermans et al., 2020	South Africa (2007–2008)	104,719 participants receiving ART at 52 urban and rural facilities across 4 provinces of South Africa (67.6% female)	Median: 35.7 years (IQR 29.9–43.0)	Retrospective cohort	Assessing rates of virological suppression and the clinical management of viraemia	- 93,200/104,719 (89%) were virally suppressed at 12 months (VL < 1000 copies/ml) - 20,766/104,719 (19.8%) had virological failure at follow-up - 13,210/20,766 (63.6%) had a follow up VL test - 7,180/13,210 (54.4%) had virological failure (VL > 1000 copies/ml) - 1,872/4,510 (41.5%) were switched to 2nd line ART	2007–2011: Yes [57]	Low risk
29	Ntwali et al., 2019	Rwanda (2012–2016)	775 participants initiating ART from 1 public hospital and 1 health centre in the Northern Province of Rwanda (67.0% female)	Median: 34 years (IQR 27–41)	Retrospective cohort	Looking at the use of routine VL testing after installing a VL testing platform in the study district	- 510/547 (93.2%) participants had annual VL test - 451/510 (88.5%) had viral suppression (VL < 1000 copies/ml) - 103/117 (88%) had a follow up VL after initial elevated VL test - 26/41 (63.4%) switched to 2nd line ART	2009–2012: Yes [55]	Low risk
30	Nyagadza et al., 2019	Zimbabwe (2016–2017)	9456 participants from 10 sites in Manicaland Province (no data on sex of participants)	No data on total participants Median: 32 years (IQR 15–43) for those with a high VL result	Retrospective cohort	To disseminate lessons learned from Médecins Sans Frontier’s involvement with HIV VL testing scale-up in collaboration with the ministry of health	- 5966/9456 (63%) had a VL test after 6 months on ART - 5033/5966 (84%) were virally suppressed (< 1000 copies/ml) - 205/662 (31%) of virally unsuppressed patients with documentation had 1 documented EAC session - 96/201 (47.8%) had a second VL test - 69/96 (72%) were virally unsuppressed after the repeat VL test (> 1000 copies/ml) - 32/69 (46.4%) were switched to 2nd line ART	2013–2017: Yes [59]	At Risk