Recruitment failure was common across Dutch obstetrical and gynecological trials
A nationwide review of obstetrical and gynecological randomized controlled trials found that recruitment failure remained a frequent and predictable threat to study execution. Among 83 RCTs conducted with support from the trial centre of the Dutch Consortium of Obstetrics and Gynecology between 1 March 2003 and 1 December 2023, 46 trials — 55% — failed to reach their target enrollment within the planned study period, even when investigators allowed a maximal extension of 6 months. That is a substantial operational shortfall, not a marginal one.
The study asked a practical question with direct consequences for trial validity: which variables track with recruitment failure closely enough to help investigators design studies that actually finish on time? To answer it, the team examined variables at five levels — patient, doctor, participating centre, study organisation, and study design — and then tested which factors carried the most weight in multivariable risk prediction modelling. The result was not a diffuse list of minor contributors. Three signals stood out: a no-treatment arm in which treatment is standard clinical practice, a compensation fee of less than €200 per included patient, and funding below €350,000. A preceding pilot study lowered the risk.
Trial design features drove the strongest signal
The clearest pattern came from study architecture itself. Trials that asked clinicians and participants to accept a no-treatment arm in settings where active care is standard faced greater recruitment difficulty. In routine obstetrics and gynecology, patients often enter studies with strong expectations about intervention, symptom relief, fertility, delivery outcomes, or procedural benefit. When randomization includes a no-treatment arm in that context, equipoise may be scientifically acceptable but psychologically difficult for patients and clinicians to tolerate. Recruitment slows when the proposed comparator feels too distant from everyday care.
Financial structure also mattered. Trials with compensation below €200 per included patient and total funding under €350,000 were more vulnerable to recruitment failure. In practical terms, lower compensation can reduce the incentive and infrastructure support needed for busy centres to identify eligible participants, explain study procedures, and sustain repeated screening over long periods. Low overall funding can leave little margin for recruitment staff, coordination, or contingency planning. In multicentre research, enrollment rarely fails for a single reason alone; it usually fails when several modest pressures accumulate until the trial can no longer keep pace with its target.
The pilot study finding points in a different direction. A preceding pilot appeared to reduce the risk of missing the recruitment target. That likely reflects better calibration before the definitive trial begins: realistic screening assumptions, clearer site workflows, more accurate estimates of eligible patient flow, and early identification of barriers that would otherwise emerge only after the clock is already running. A pilot does not guarantee success. It does, however, expose weak points before they become irreversible.
Why recruitment collapses before the science does
Recruitment failure is not just a logistics problem. It can distort the entire evidentiary value of a trial.
When enrollment lags, investigators are forced to extend timelines, increase burden on participating centres, or stop early. Any of those outcomes can threaten statistical power, delay answers to clinically important questions, and waste resources already committed to the study. In obstetrics and gynecology, where conditions often involve time-sensitive decisions, fertility concerns, symptom burden, or pregnancy-related risk, prolonged recruitment can also change the patient population itself as practice patterns evolve during the study window.
The observed association with a no-treatment arm fits a well-known bedside reality. Patients are more likely to consent when every randomized option feels clinically meaningful. If one arm is perceived as doing nothing while standard care remains available outside the study, equipoise may look less like a fair test and more like an avoidable delay in treatment. That effect can be especially pronounced in disciplines where women are making decisions under stress, with symptoms or reproductive goals in the foreground.
How the findings should shape future multicentre RCT planning
The value of this analysis lies in its operational clarity. Recruitment failure was common, and the variables most tightly linked to it were modifiable at the planning stage. Trial teams designing future studies in obstetrics and gynecology should treat a no-treatment arm as a high-risk feature when standard clinical practice already offers active management. Funding levels and per-patient compensation should be checked against the true cost of screening, counselling, consent, and site coordination rather than treated as administrative afterthoughts.
A pilot phase deserves more than ceremonial status. In this cohort, it functioned as a protective feature, and that makes sense clinically and operationally. Early feasibility work can reveal whether proposed eligibility criteria are too narrow, whether centres can identify patients at the right tempo, and whether the study proposition is acceptable to participants and clinicians. When those assumptions are wrong, recruitment failure often arrives silently, then all at once.
The study does not claim that patient, doctor, or centre-level factors are irrelevant. It does show, however, that the strongest predictors of recruitment failure in this cohort were tied to the structure of the trial itself. That has direct implications for sponsors, investigators, and ethics committees reviewing multicentre studies in women’s health. A trial can be scientifically elegant and still be operationally fragile if its design makes enrollment too difficult to sustain.
The unanswered practical issue is how far these predictors can be generalized beyond this Dutch consortium and beyond obstetrics and gynecology. Even so, the core message is concrete: recruitment success is shaped early, and some of the clearest warning signs are visible before the first participant is enrolled.