Purpose: New contraceptive methods provide greater choice in terms of effectiveness, management of side-effects, convenience and frequency of administration and flexibility, but make the decisions about contraception more complex. There are limited data on the factors that determine women’s choices among these alternatives, to inform providers about the factors which are most important to women, or to predict uptake of new products. This paper reports on choice experiments designed to elicit women’s preferences and providers’ recommendations in relation to prescribed contraception. Method: The attributes of the women’s experiment included a labelled product attribute and 7 other product characteristics. The products were the combined pill, progesterone pill; hormonal implant, hormonal intra-uterine device (IUD); hormonal injection and a conventional IUD, as well as two new products, a hormonal patch and a vaginal ring. Other attributes included frequency of administration, effect on weight, effect on bleeding, effect on acne, cost, effectiveness, and doctor’s recommendation. The providers’ experiment presented attributes that described a consultation, and asked providers to nominate which product(s) they would discuss and recommend. Women in Australia aged 18-50 who were currently using or planning to use prescribed contraception in the next 5 years were recruited via an on-line panel (n=528). Australian General Practitioners were recruited via telephone (n=161). Both DCEs were completed on-line. Mixed logit models were estimated, specifying random correlated product intercepts to allow for preference heterogeneity and to take account of the panel nature of the data. The final model was used to forecast the impact of introducing new products. Result: Overall respondents had a preference for the combined pill, the minipill or the injection over the IUD (the base case). However the large and precisely estimated standard deviation indicates substantial heterogeneity among women's preferences for the combined pill. The implant, hormonal IUD, patch and ring had large significant standard deviations with relatively smaller non-significant intercepts indicating that there was no strong consensus among respondents for preference for these products over the IUD. Conclusion: The model forecasts that the patch would be well received among women, achieving a greater market share than current non-pill products, but the ring would have limited appeal.
Candidate for the Lee B. Lusted Student Prize Competition