Antibiotic prophylaxis in the surgical management of miscarriage in low-income countries: a cost-effectiveness analysis of the AIMS trial

Goranitis, Ilias; Lissauer, David M; Coomarasamy, Arri; Wilson, Amie; Daniels, Jane; Middleton, Lee; Bishop, Jonathan; Hewitt, Catherine A.; Weeks, Andrew D.; Mhango, Chisale; Mataya, Ronald; Ahmed, Iffat; Oladapo, Olufemi T.; Zamora, Javier; Roberts, Tracy E.

Abstract

Background
There is ongoing debate on the clinical benefits of antibiotic prophylaxis for reducing pelvic infection after miscarriage surgery. We aimed to study the cost-effectiveness of antibiotic prophylaxis in the surgical management of miscarriage in low-income countries.

Methods
We did an incremental cost-effectiveness analysis using data from 3412 women recruited to the AIMS trial, a randomised, double-blind, placebo-controlled trial designed to evaluate the effectiveness of antibiotic prophylaxis in the surgical management of miscarriage in Malawi, Pakistan, Tanzania, and Uganda. Economic evaluation was done from a health-care-provider perspective on the basis of the outcome of cost per pelvic infection avoided within 2 weeks of surgery. Pelvic infection was broadly defined by the presence of clinical features or the clinically identified need to administer antibiotics. We used non-parametric bootstrapping and multilevel random effects models to estimate incremental mean costs and outcomes. Decision uncertainty was shown via cost-effectiveness acceptability frontiers. The AIMS trial is registered with the ISRCTN registry, number ISRCTN97143849.

Findings
Between June 2, 2014, and April 26, 2017, 3412 women were assigned to receive either antibiotic prophylaxis (1705 [50%] of 3412) or placebo (1707 [50%] of 3412) in the AIMS trial. 158 (5%) of 3412 women developed pelvic infection within 2 weeks of surgery, of whom 68 (43%) were in the antibiotic prophylaxis group and 90 (57%) in the placebo group. There is 97–98% probability that antibiotic prophylaxis is a cost-effective intervention at expected thresholds of willingness-to-pay per additional pelvic infection avoided. In terms of post-surgery antibiotics, the antibiotic prophylaxis group was US$0·27 (95% CI −0·49 to −0·05) less expensive per woman than the placebo group. A secondary analysis, a sensitivity analysis, and all subgroup analyses supported these findings. Antibiotic prophylaxis, if implemented routinely before miscarriage surgery, could translate to an annual total cost saving of up to $1·4 million across the four participating countries and up to $8·5 million across the two regions of sub-Saharan Africa and south Asia.

Interpretation
Antibiotic prophylaxis is more effective and less expensive than no antibiotic prophylaxis. Policy makers in various settings should be confident that antibiotic prophylaxis in miscarriage surgery is cost-effective.

Citation

Goranitis, I., Lissauer, D. M., Coomarasamy, A., Wilson, A., Daniels, J., Middleton, L., …Roberts, T. E. (2019). Antibiotic prophylaxis in the surgical management of miscarriage in low-income countries: a cost-effectiveness analysis of the AIMS trial. Lancet Global Health, 7(9), e1280-e1286. https://doi.org/10.1016/s2214-109x%2819%2930336-5

• Journal Article Type: Article
• Acceptance Date: Jun 14, 2019
• Online Publication Date: Aug 8, 2019
• Publication Date: Sep 1, 2019
• Deposit Date: Aug 9, 2019
• Publicly Available Date: Aug 9, 2019
• Journal: The Lancet Global Health
• Print ISSN: 2214-109X
• Electronic ISSN: 2214-109X
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 7
• Issue: 9
• Pages: e1280-e1286
• DOI: https://doi.org/10.1016/s2214-109x%2819%2930336-5
• Public URL: https://nottingham-repository.worktribe.com/output/2410425
• Publisher URL: https://www.sciencedirect.com/science/article/pii/S2214109X19303365?via%3Dihub#!
• Additional Information: This article is maintained by: Elsevier; Article Title: Antibiotic prophylaxis in the surgical management of miscarriage in low-income countries: a cost-effectiveness analysis of the AIMS trial; Journal Title: The Lancet Global Health; CrossRef DOI link to publisher maintained version: https://doi.org/10.1016/S2214-109X(19)30336-5; CrossRef DOI link to the associated document: https://doi.org/10.1016/S2214-109X(19)30314-6; Content Type: article; Copyright: © 2019 The Author(s). Published by Elsevier Ltd.