Zimbabwe DSD Performance Review, June 2019

Dr. Clorata Gwanzura Differentiated Care Medical Officer Ministry of Health and Child Care 13 November 2019 Overview

Background and Approach

Overview of key results

Summary, lessons learned, and future steps

The CQUIN Project Background

Annual Cross-sectional data • Implementation of Review meetings part of the DSD collection for Facility M and E Plan Coverage • Method of data collection also earmarked for reporting in Sampling method for client non-partner supported districts uptake and coverage: Semi- • The MOHCC and its stakeholders i.e. implementing annually partners, and the ICAP CQUIN team developed Comprehensive collection for Facility and Client Uptake objectives of review and Coverage: Quarterly • Provincial teams from Mash West, Mat South and Electronic system for Mat North were engaged to guide the planning and collection of Uptake, Coverage and Outcome implementation of the review Indicators: Quarterly Analysis, 2020

M and E Plan Approach: Population

• 17 districts across the 4 provinces were identified for inclusion • 73 facilities were selected for data collection • 2 facilities failed to do/complete data collection • Selected via stratified random sampling • All districts in 2 provinces, 5 facilities in each provinces (District hospital, 2high volume sites, 2 low volume sites) • Selected districts and facilities (total of 3) in 2 provinces • Two patient cohorts: 12 and 24 months on ART

The CQUIN Project 4 Approach: Data collection

• Data collection via abstraction of patient data from facilities • Patient ART booklet main data source with additional data sources in some cases – pragmatic approach • Standardized data entry form in MS Excel • Data collection supervised by district teams

The CQUIN Project Approach: Sharing and discussion of results

• Convened a 3-day meeting to review results • District, Provincial, and National MOHCC teams; PEPFAR implementing partners • Provincial overview presentations and discussions led by province teams • District-level presentations and discussions led by district teams

The CQUIN Project Key Results Numbers of Patients, by Province and District

Number of patients, by province and Number of patients, by province district, both cohorts combined Both cohorts combined (N=2574) n (%) Mashonaland West 1188 Chegutu 182 (15%) Mashonaland Matabeleland Hurungwe 149 (13%) West North 138 (12%) 1188 76 Makonde 181 (15%) (46%) (3%) MhondoroNgezi 181 (15%) Sanyati 172 (14%) Zvimba 185 (16%) Matabeleland North 76 Bubi 42 (55%) Umguza 34 (45%) Matabeleland South 1310 305 (23%) Matabeleland Bulilima 165 (13%) South Gwanda 168 (13%) 1310 184 (14%) (51%) Mangwe 170 (13%) Matobo 153 (12%) Umzingwane 165 (13%) The CQUIN Project Total 2574 8 Overall Modified HIV Care Cascade, by cohort

1400

1200 85% 44% of 85% active 54% of 1000 patients active 61% of patients patients 68% of 800 with a VL 26% of patients test patients with a VL 1306 1268 46% with VLS 600 test 32% of 1114 1081 37% patients 32% 400 with VLS 607 23% 474 200 415 10% 289 6% 131 0 76 Initiated on ART Active on ART Received VL Test Suppressed Viral Enrolled in Non- Within 12 Months Load Mainstream Model* 24 month 12 month

The CQUIN Project *For all clients active on ART and with documented VLS 9 Provincial Modified HIV Care Cascades 12- and 24-month cohorts, combined

100% 90% 82% 80% 71% 79% of patients 31% of 25% of 60% with VLS patients patients with VLS with VLS 42% 42% 38% 40% 30% 32% 25% 25% 20% 8% 7%

0% MashWest MatSouth MatNorth (n=1188) (n=1310) (n=76) Active on ART Received VL Test Within 12 Months Suppressed Viral Load Enrolled in Non-Mainstream Model* The CQUIN Project *For all clients active on ART and with documented VLS 10 Overall DSD Coverage All patients active on ART - 12 and 24 month cohorts combined (n=2148)

5 (0%) 144 226 (7%) (11%) Conventional Care 103 CARG (5%) 97 Facility Club (+CATS) (4%) Family Refill 1573 (73%) Fast Track Outreach

The CQUIN Project 11 DSD Coverage by Province All patients active on ART - 12 and 24 month cohorts combined

4% MashWest 71% 6% 9% 10% (n=1017)

2% MatNorth 39% 6% 33% 20% (n=54)

3% 3% MatSouth 77% 6% 10% (n=1077)

Conventional Care CARG Facility Club (+CATS) The CQUIN Project Family Refill Fast Track Outreach Overall Fidelity of DSD Enrollment, by cohort

(N=1095) (N=1053) 100% 132 (12%) Appropriate 86 (8%) 90% Unknown 155 (15%) 80% 173 (16%) Inappropriate 13 (1%) 70% 16 (1%) 199 (19%) Missed Opportunity 60% 292 (27%)

50%

40%

557 (53%) 30% Unknown 452 (41%) 20%

10%

0% 30 (3%) Appropriate 43 (4%) 24 month* 12 month** Not eligible no DSD Unknown Eligibility no DSD Eligible no DSD Not eligible + DSD Unknown Eligibility + DSD Eligible + DSD The CQUIN Project For clients active on ART, where eligibility is defined as a either 1) a VL test ≤12 months ago and a result of <1000 copies or 2) a CD4 test ≤ 12 months ago and a result of 200+, as well as an absence of any OIs or TB 14 *Excludes 19 individuals for missing information **Excludes 28 individuals for missing information Model Switch: Current ART model and 12 months prior Among 24 month cohort patients

Prior Model, Current Model, May 2018 May 2019 CARG 19 53

Conventional Model 670 797

Facility Club (+CATS Model) 53 45 Family Refill 77 42 Model 46 Fast Track 108 1 Outreach 1 The CQUIN Project *All clients active on ART (combined cohorts, regardless of VLS)15 Numbers of clinical visits and ART pickups per year, conventional model vs. DSD models

Yearly number of clinical visits per person, Yearly number of ART pickups per person, by model type by model type

35% 60% 55% 29% 30% 30% 26% 25% 25% 50% 25% 23% 22% 20% 40% 36% 20% 28% 30% 21% 24% 15% 18% 20% 10% 11% 7% 5% 10% 0% 0% 1-2 3-4 5-6 7+ 1-2 3-4 5-6 7+ Number of clinic visits Number of ART pickups

Conventional % DSD % Conventional % DSD %

The CQUIN Project *All clients active on ART (combined cohorts, regardless of VLS) 16 Post-hoc assessment of DSD model uptake by sex: DSD Performance Review

Question: Are women more likely than men to enroll in DSD models providing less-intensive services? Current ART model coverage, by sex (May 2019) DSD Performance Review, 70 facilities in 3 provinces

DSD uptake and model of choice was similar for men and women (n=2,148) Men Women

11% 10% Fast Track 5% Facility Club 4% Family Refill 8% 6% CARG 4% Conventional ART 5%

72% Note: Patients receiving MMS/MMD but 74% not other DSD model features are classified within Conventional ART group

Source: Zimbabwe Ministry of Health and Child Care TheThe CQUIN CQUIN Project Learning Network 19 Summary of Key Results (1)

• 85% ART patients were retained at 12 and 24 months after initiation; overall, 27% of these pts were enrolled in a DSD ART model • In the post-hoc analysis, we saw no clear differences in DSD enrollment by sex • About 50% of retained patients had a VL test within the past 12 mos; approx. 65% of these pts had documented VLS • Roughly 30% of patients with VLS were in a DSD model • Fast track and family refill were the most common DSD models overall, though popularity of models varied across provinces and districts

The CQUIN Project Summary of Key Results (2)

• Fidelity of DSD model enrollment criteria was low; a small % of ART patients had a VL result and were enrolled in the expected ART model based on results • Model switch was common across the past year among the 24-month cohort patients • Most common was switch from conventional model, though switch between DSD models also occurred • A lower % of patients in DSD models (39%) than in the conventional model (52%) had 5+ ART pickups over the past year • The number of clinical visits did not differ across DSD vs. conventional models

The CQUIN Project Lessons Learned and Next Steps

• Data provided helpful insights on DSD implementation • Extensive participation in discussions by facility and district teams • Support for DSD scale-up can be targeted using these findings • Increase uptake and coverage in priority DSD models • Improve VL testing coverage and fidelity of implementation • Eligibility/enrollment, and appointment/ART pickup spacing • Monitor and manage DSD model switch • Steps can be taken to improve data review process as well as data quality

The CQUIN Project Recommendations for Future Performance Reviews

• Strengthen data collection process • Provide a more detailed job aid on indicators collected • Ensure rigorous supervision, e.g., through daily review and sign-off on data collection sheets • Ensure adequate resources are available for data cleaning, analysis, and development of slides • Ask districts and regions to develop structured Action Plans based on results • Develop strategy for following up on Action Plan progress

The CQUIN Project Acknowledgements

•Ministry of Health and Child Care • ATP • Provincial and District Teams •I-tech, OPHID, ICAP •CQUIN TA Team •DSD TWG Stakeholders

The CQUIN Project