Cost-effectiveness analysis of two-way texting for post-operative follow-up in Zimbabwe's voluntary medical male circumcision program

Abstract

Objective: Although adverse events (AEs) following voluntary medical male circumcision (VMMC) are rare, their prompt ascertainment and management is a marker of quality care. The use of two-way text messaging (2wT) for client follow-up after VMMC reduces the need for clinic visits (standard of care (SoC)) without compromising safety. We compared the cost-effectiveness of 2wT to SoC for post-VMMC follow-up in two, high-volume, public VMMC sites in Zimbabwe.

Materials and methods: We developed a decision-analytic (decision tree) model of post-VMMC client follow-up at two high-volume sites. We parameterized the model using data from both a randomized controlled study of 2wT vs. SoC and from the routine VMMC program. The perspective of analysis was the Zimbabwe government (payer). The time horizon covered the time from VMMC to wound healing. Costs included text messaging; both in-person and outreach follow-up; and AE management. Costs were estimated in 2018 U.S. dollars. The outcome of analysis was AE yield relative to the globally accepted safety standard of a 2% AE rate. We estimated the incremental cost per percentage increase in AE ascertainment and the incremental cost per additional AE identified. We conducted univariate and probabilistic sensitivity analyses.

Results: 2wT increased the costs due to text messaging by $4.42 but reduced clinic visit costs by $2.92 and outreach costs by $3.61 -a net savings of $2.10. 2wT also increased AE ascertainment by 50% (92% AE yield in 2wT compared to 42% AE yield in SoC). Therefore, 2wT dominated SoC in the incremental analysis: 2wT was less costly and more effective. Results were generally robust to univariate and probabilistic sensitivity analysis.

Conclusions: 2wT is cost-effective for post-VMMC follow-up in Zimbabwe. Countries in which VMMC is a high-priority HIV prevention intervention should consider this mHealth intervention to reduce overall cost per VMMC, increasing the likelihood of current and future VMMC program sustainability.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1. Decision analytic model (decision tree) of post-VMMC follow-up of men in Zimbabwe.

*Leads to outreach. **Note the use of clones to complete the decision tree diagram after day 2 in the SOC and 2wT arms.

Fig 2. One-way sensitivity analysis on incremental cost comparing 2wT to SoC.

The top ten influential variables are shown. FTE–full time equivalent. 2wT–two-way texting. SoC—standard of care.

Fig 3. Incremental cost-effectiveness scatterplot of comparison of 2wT vs. SoC.

AE–Adverse event.