Cost-effectiveness analysis of lifestyle intervention in obese infertile women

What is the cost-effectiveness of lifestyle intervention preceding infertility treatment in obese infertile women?

Lifestyle intervention preceding infertility treatment as compared to prompt infertility treatment in obese infertile women is not a cost-effective strategy in terms of healthy live birth rate within 24 months after randomization, but is more likely to be cost-effective using a longer follow-up period and live birth rate as endpoint.

In infertile couples, obesity decreases conception chances. We previously showed that lifestyle intervention prior to infertility treatment in obese infertile women did not increase the healthy singleton vaginal live birth rate at term, but increased natural conceptions, especially in anovulatory women. Cost-effectiveness analyses could provide relevant additional information to guide decisions regarding offering a lifestyle intervention to obese infertile women.

The cost-effectiveness of lifestyle intervention preceding infertility treatment compared to prompt infertility treatment was evaluated based on data of a previous RCT, the LIFEstyle study. The primary outcome for effectiveness was the vaginal birth of a healthy singleton at term within 24 months after randomization (the healthy live birth rate). The economic evaluation was performed from a hospital perspective and included direct medical costs of the lifestyle intervention, infertility treatments, medication and pregnancy in the intervention and control group. In addition, we performed exploratory cost-effectiveness analyses of scenarios with additional effectiveness outcomes (overall live birth within 24 months and overall live birth conceived within 24 months) and of subgroups, i.e. of ovulatory and anovulatory women, women <36 years and 36 years of age and of completers of the lifestyle intervention. Bootstrap analyses were performed to assess the uncertainty surrounding cost-effectiveness.

Infertile women with a BMI of 29 kg/m(2) (no upper limit) were allocated to a 6-month lifestyle intervention programme preceding infertility treatment (intervention group, n = 290) or to prompt infertility treatment (control group, n = 287). After excluding women who withdrew informed consent or who were lost to follow-up we included 280 women in the intervention group and 284 women in the control group in the analysis.

Total mean costs per woman in the intervention group within 24 months after randomization were a,notsign4324 (SD a,notsign4276) versus a,notsign5603 (SD a,notsign4632) in the control group (cost difference of -a,notsign1278, P <0.05). Healthy live birth rates were 27 and 35% in the intervention group and the control group, respectively (effect difference of -8.1%, P <0.05), resulting in an incremental cost-effectiveness ratio of a,notsign15 845 per additional percentage increase of the healthy live birth rate. Mean costs per healthy live birth event were a,notsign15 932 in the intervention group and a,notsign15 912 in the control group.

Exploratory scenario analyses showed that after changing the effectiveness outcome to all live births conceived within 24 months, irrespective of delivery within or after 24 months, cost-effectiveness of the lifestyle intervention improved. Using this effectiveness outcome, the probability that lifestyle intervention preceding infertility treatment was cost-effective in anovulatory women was 40%, in completers of the lifestyle intervention 39%, and in women 36 years 29%.

In contrast to the study protocol, we were not able to perform the analysis from a societal perspective. Besides the primary outcome of the LIFEstyle study, we performed exploratory analyses using outcomes observed at longer follow-up times and we evaluated subgroups of women; the trial was not powered on these additional outcomes or subgroup analyses.

Cost-effectiveness of a lifestyle intervention is more likely for longer follow-up times, and with live births conceived within 24 months as the effectiveness outcome. This effect was most profound in anovulatory women, in completers of the lifestyle intervention and in women 36 years old. This result indicates that the follow-up period of lifestyle interventions in obese infertile women is important.

The scenario analyses performed in this study suggest that offering and reimbursing lifestyle intervention programmes in certain patient categories may be cost-effective and it provides directions for future research in this field.