HIV prevention costs and program scale: data from the PANCEA project in five low and middle-income countries
2007

HIV Prevention Costs and Program Scale

Sample size: 206 publication 10 minutes Evidence: high

Author Information

Author(s): Elliot Marseille, Lalit Dandona, Nell Marshall, Paul Gaist, Sergio Bautista-Arredondo, Brandi Rollins, Stefano M Bertozzi, Jerry Coovadia, Joseph Saba, Dmitry Lioznov, Jo-Ann Du Plessis, Evgeny Krupitsky, Nicci Stanley, Mead Over, Alena Peryshkina, SG Prem Kumar, Sowedi Muyingo, Christian Pitter, Mattias Lundberg, James G Kahn

Primary Institution: University of California, San Francisco

Hypothesis

Costs per unit of HIV prevention program output will initially decrease as programs expand, but may increase if expansion continues beyond an optimal size.

Conclusion

Unit costs decrease with scale across a wide range of service types and volumes.

Supporting Evidence

  • Doubling in scale resulted in reductions in unit costs averaging 34.2%.
  • The fraction of variation in efficiency explained by scale ranged from 26% to 96%.
  • Efficiency increased with scale across all countries and interventions examined.
  • Regression trends suggested an inflection point in India beyond which unit costs increased.

Takeaway

When more people use HIV prevention programs, the cost for each person goes down, making it cheaper to help more people.

Methodology

The study analyzed cost and output data from 206 HIV prevention programs across five countries using regression analysis to examine the relationship between scale and efficiency.

Potential Biases

Potential biases may arise from the varying quality of data collection across different programs and countries.

Limitations

The study is cross-sectional and does not document changes in unit costs over time within expanding programs.

Participant Demographics

The study included programs from India, Mexico, Russia, South Africa, and Uganda, targeting various populations including sex workers and individuals seeking voluntary counseling and testing.

Statistical Information

P-Value

< 0.001

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1472-6963-7-108

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication