Format

Send to

Choose Destination
Trials. 2019 Mar 4;20(1):157. doi: 10.1186/s13063-019-3186-x.

Using discrete choice experiments to inform the design of complex interventions.

Author information

1
Department of Global Health and Development, London School of Hygiene and Tropical Medicine, 15-17 Tavistock Place, London, WC1H 9SH, UK. Fern.Terris-Prestholt@lshtm.ac.uk.
2
National Institute for Medical Research, PO Box 1462, Mwanza, Tanzania.
3
Centers for Disease Control and Prevention (CDC), Center for Global Health, Division of Global HIV & TB, Atlanta, GA, 30333, USA.
4
Jhpiego, 1615 Thames Street, Baltimore, MD, 21231, USA.
5
Department of Global Health and Development, London School of Hygiene and Tropical Medicine, 15-17 Tavistock Place, London, WC1H 9SH, UK.
6
Department of Clinical Research, Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Kepple Street, London, WC1E 7HT, UK.
7
University of North Carolina Chapel Hill Project-China, No. 2, Lujing Road, Guangzhou, 510095, China.
8
FHI 360, 1825 Connecticut Avenue NW, Washington, DC, 20009, USA.
9
MRC Tropical Epidemiology Group, Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, Kepple Street, London, WC1E 7HT, UK.

Abstract

BACKGROUND:

Complex health interventions must incorporate user preferences to maximize their potential effectiveness. Discrete choice experiments (DCEs) quantify the strength of user preferences and identify preference heterogeneity across users. We present the process of using a DCE to supplement conventional qualitative formative research in the design of a demand creation intervention for voluntary medical male circumcision (VMMC) to prevent HIV in Tanzania.

METHODS:

The VMMC intervention was designed within a 3-month formative phase. In-depth interviews (n = 30) and participatory group discussions (n = 20) sought to identify broad setting-specific barriers to and facilitators of VMMC among adult men. Qualitative results informed the DCE development, identifying the role of female partners, service providers' attitudes and social stigma. A DCE among 325 men in Njombe and Tabora, Tanzania, subsequently measured preferences for modifiable VMMC service characteristics. The final VMMC demand creation intervention design drew jointly on the qualitative and DCE findings.

RESULTS:

While the qualitative research informed the community mobilization intervention, the DCE guided the specific VMMC service configuration. The significant positive utilities (u) for availability of partner counselling (u = 0.43, p < 0.01) and age-separated waiting areas (u = 0.21, p < 0.05) led to the provision of community information booths for partners and provision of age-separated waiting areas. The strong disutility of female healthcare providers (u = - 0.24, p < 0.01) led to re-training all providers on client-friendliness.

CONCLUSION:

This is, to our knowledge, the first study documenting how user preferences from DCEs can directly inform the design of a complex intervention. The use of DCEs as formative research may help increase user uptake and adherence to complex interventions.

KEYWORDS:

Choice experiment; Formative research; HIV; Preferences; Tanzania; Voluntary medical male circumcision

PMID:
30832718
PMCID:
PMC6399844
DOI:
10.1186/s13063-019-3186-x
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for BioMed Central Icon for PubMed Central
Loading ...
Support Center