JMIR Publications

ABSTRACT

Background:

Many couples who undergo fertility treatment have multiple lifestyle risk factors that reduce their chances of becoming pregnant. The MyFertiCoach (MFC) app was developed as an integrated lifestyle program with modules on healthy weight, nutrition and exercise, and quitting smoking, alcohol, drugs, and reduction of stress. We hypothesized an improvement in lifestyle.

Objective:

The primary outcome is the change in Total Risk Score (TRS) at three- and six-months follow-up. The TRS per individual is defined as the sum of all risk scores per behavior (i.e., based on vegetable/fruit/folic acid intake, smoking, and alcohol use after three and six months). The higher the TRS, the unhealthier the nutrition and lifestyle habits that are present, and the lower the chance of pregnancy. The secondary endpoints are changes in BMI, activity score, preconception dietary risks (PDR) score, Distress-score (e.g., burden), smoking habits, alcohol intake, and program adherence.

Methods:

This was a retrospective, observational, single-center study including patients between the 1st of January 2022 and the 31st of December 2023. We invited subfertile female patients, and their partner, aged 18-43 with a referral to a gynecologist to follow the online lifestyle coaching via the MFC app. Relevant lifestyle modules were chosen by the gynaecologist using the results of the integrated screening questionnaires. We used (hierarchical) linear mixed-models (LMMs) to estimate the change in outcomes. Where the pattern of missing data was MNAR (missing not at random), we deployed joint modelling. Statistical significance was set at p≤0.05 considering methods to retain the same false-positive rate.

Results:

A total of 1805 patients received an invitation to participate in this study, of which on average 737 (40.8%) filled out a screenings questionnaire at baseline. For the TRS score, we included 798 (44.2%) patients at baseline, of which 517 (64.8%) patients included their partner. An average of 282 patients (37.9%) sent in at least one follow-up questionnaire. Patients rated the app above average (N=137, median=7 on a 1-10 scale) at days 7 and 14. The TRS scored decreased 1.5 points (P<.001) on average at T3 and T6 compared to baseline, which is clinically relevant. All secondary outcomes changed significantly (statistically) in a positive direction for patients who used a relevant lifestyle module. Most of the gains were achieved at three months yet remained statistically significant at six months (except for alcohol). These findings are similar across LMMs and joint models.

Conclusions:

Our results show the immediate benefit of a m-health application for women wanting to become pregnant. However, to maintain and even improve these results there is a need to further tailor the patient-specific programs.