Skip to main content
SMFM 2024 Pregnancy Meeting Main banner
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Additional registration fee required
Additional registration fee required
Faculty have requested this content not be shared on social media
Faculty have requested this content not be shared on social media
CME Credit Offered
CME Credit Offered
Poster Icons
Foundation Awardee
Foundation Awardee
No Social Media
No Social Media
Back
Q&A

Labor

Poster Session 4

(1108) Sharing or not-sharing individual participant data: how valid are randomized trials evaluating induction of labor?

Wednesday, February 14, 2024
1:00 PM – 2:30 PM  
Location: Prince George's Exhibit Hall CDE, Lower Atrium Level
Objective: To compare characteristics, trustworthiness of randomized controlled trials (RCTs) and outcomes on labor induction that shared individual participant data (IPD) versus those that did not.

Study Design: In this meta-epidemiological study, RCTs included in five IPD-MAs comparing different methods of labor induction were evaluated using published data. We used Monte Carlo simulation with baseline characteristics to assess randomization. Trustworthiness of RCTs was evaluated using the TRACT checklist (pubmed.ncbi.nlm.nih.gov/37337220/). Agreement in outcomes (cesarean delivery and hyperstimulation) between shared and non-shared was done with random-effects meta-analysis.

Results:

Of 112 RCTs, 29 (25.9%) shared data and 83 (74.1%) did not. Median year of publication of the shared group was 2015 (range 1996-2023) versus 2008 (1983-2021) for non-sharing RCTs. Shared RCTs were larger median (range) compared to the non-shared; 339 (101-1859) versus 126 (38-1358).

Studies that did not share data had a small probability of having proper randomization (p=0.03), whereas shared group seems to be having proper randomization (p=0.85). In non-shared group, authors who refused to share were less likely (p=0.003) to have proper randomization than authors who could not share data as data were not available (p=0.14).

Proportion of RCTs with ‘concerns’ under seven domains of the TRACT checklist between shared vs non-shared was: Governance 36.8% vs 75.1%; Author Group 2.3% vs 11.2%; Plausibility of Intervention 0% vs 16.9%; Timeframe 3.4% vs 21.1%; Drop-Out Rates 0% vs 27.7%; Baseline Characteristics 0% vs 11.7%; Outcomes 12.6% vs 32.1%.

According to random-effects meta-analysis, only one comparison showed comparable effect estimates for cesarean delivery and uterine hyperstimulation between shared and non-shared groups (Table 1).

Conclusion:

In studies on labor induction, RCTs without IPD sharing have lower quality and more trustworthiness concerns than those who shared IPD. The availability of IPD and willingness to share is a good indicator of the quality and trustworthiness.