Outcome | Justification and/or further description | Data analysis |
---|---|---|
Primary outcomes | ||
ART initiation within 7 days of study enrollment | WHO calls for initiation within 7 days for all patients | Intention-to-treat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals |
Initiated ART within 28 days of study enrollment and alive, in care, and retained on ART 8 months after study enrollment | Retention at 8 months captures early attrition on ART, in case the manner of ART initiation affects longer-term outcomes (8 months was selected to allow up to 1 month to initiate ART, 6 months of follow-up after treatment initiation, and up to 1 month to return for the 6 months routine clinic visit) | Intention-to-treat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals. Reasons for not achieving this outcome will also be described to the extent that routinely collected follow-up data allow |
Secondary outcomes | ||
ART outcomes | ||
ART initiation within 14 days of study enrollment for TB suspects | SLATE II aims to avoid delay of ART initiation in patients with mild TB symptoms | Intention-to-treat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals |
ART initiation within 1, 14, and 28 days of study enrollment | Both national and global guidelines recommend same-day initiation (1 day). Since other published studies have used 14 and 28 days, maintaining this secondary outcome will allow comparison with SLATE II results | Intention-to-treat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals |
Time to initiation, in days | Time to initiation captures any effect of SLATE II on accelerating initiation | Intention-to-treat analysis; comparison of time to initiation presented as survival curves with log rank test |
Viral suppression by 8 months after study enrollment, for all patients and for TB suspects | Allows ≤1 month (28 days) to initiate ART, 6 months of follow-up after treatment initiation, and ≤ 1 month to return for the 6 months routine clinic visit | Intention-totreat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals. Reasons for not achieving this outcome will also be described to the extent that routinely collected follow-up data allow |
Retention in care 14 months after study enrollment, for all patients and for TB suspects | Allows ≤1 month (28 days) to initiate ART, 12 months of follow-up after treatment initiation, and ≤ 1 month to return for the 12 months routine clinic visit; any visit 12–14 months after study enrollment will represent the 12 months visit | Intention-to-treat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals |
SLATE II algorithm performance | ||
Proportions of study patients who screen in and screen out for immediate ART initiation using SLATE II algorithm criteria | Will provide guidance on proportions of patients who could be initiated under SLATE II if adopted as routine care | Intention-to-treat analysis; comparison of proportions between groups presented as a risk difference and 95% confidence intervals |
Reasons for ineligibility | Will provide guidance on types of referral services required from clinics | Descriptive analysis of proportions of patients screening out for each possible reason indicated on SLATE II screens |
Frequency and types of adverse events | Will indicate probability of adverse events related to the algorithm and guidance on what to expect | Descriptive analysis of adverse events reported in medical records after ART initiation for each follow-up period |
Patient preferences on the speed and timing of ART initiation | Baseline questionnaire data | Descriptive analysis of medians and interquartile ranges (IQRs) for continuous outcomes and proportions and corresponding 95% confidence intervals for categorical outcomes |
TB outcomes | ||
Proportion of symptomatic patients who test positive for TB using the LAM test | Will determine usefulness of using a point-of-care LAM test as part of ART initiation | Descriptive analysis of proportion of tests found positive |
Proportions of symptomatic and asymptomatic patients who test positive for TB using Xpert MTB/RIF | Will determine usefulness of testing asymptomatic patients with Xpert as part of ART initiation | Descriptive analysis of proportion of tests found positive |
Health system outcomes | ||
Costs to patients of ART initiation under standard and intervention procedures | SLATE II is hypothesized to reduce the number of clinic visits required for ART initiation and thus costs to patients | Sum of clinic visit costs and time spent from enrollment visit to visit at which ARVs are dispensed, calculated from questionnaire responses |
Costs to providers of ART initiation under standard and intervention procedures and cost-effectiveness of intervention | SLATE II is hypothesized to reduce the number of clinic visits required for ART initiation and thus costs to providers | Estimate of provider costs using previously described [33] bottom-up costing methods, with resource utilization extracted from medical records and case report forms (CRFs) and unit costs obtained from study sites. The average cost to the provider per patient achieving each primary outcome will be compared between intervention and standard initiation groups to provide an estimate of the cost-effectiveness of the two strategies. Costs will be reported as means (standard deviations) and medians (IQRs) in local currencies and US dollars |
Comparison of SLATE I and SLATE II results | Assess changes between the original and revised algorithms and to look for secular changes in patient characteristics, clinic procedures, and patient outcomes over time | Direct comparison of findings between the two studies wherever possible (Note: If standard arm results change significantly between SLATE I and SLATE II, may conduct a difference-in-differences analysis, with SLATE I serving as the “pre” period and SLATE II as the “post” period) |