What a meta-analysis found about online versus offline recruitment

Online recruitment for clinical trials is often pitched as a straightforward upgrade over traditional methods: cheaper, faster, done. A systematic review and meta-analysis, drawing on 61 studies with 23 included in the formal meta-analysis, found something more layered than a simple upgrade story, with online recruitment winning decisively on some measures and losing just as clearly on another.

Cheaper per enrollee, and faster by the day

On cost, the result was clear. Online recruitment cost an average of $72 per enrollee, compared to $199 through offline methods, a statistically significant difference. That's not a marginal saving; it's less than half the cost per person actually enrolled.

Speed showed an equally strong result, with an important qualifier attached. Measured by recruitment rate per active recruiting day, online methods were better in every single study examined for this measure, 7 out of 7, with an incidence rate ratio of 4.17, meaning online recruitment brought in participants at more than four times the rate per active day compared to offline methods. That's a decisive result on a genuinely important measure: how quickly a recruitment effort converts active effort into enrolled participants while it's actually running.

The full-timeline picture was a lot less decisive

Recruitment rate measured across the entire recruitment period in months, rather than just active recruiting days, told a different and much less one-sided story. Only 12 of 23 studies, 52%, found online recruitment outperforming offline over the full period, with an incidence rate ratio of 1.11 that wasn't statistically significant. In other words, whatever advantage online recruitment showed on a per-active-day basis often didn't translate into a meaningfully faster overall recruitment timeline once the full study period was considered.

That gap between "faster per active day" and "not reliably faster overall" is worth sitting with. It suggests something about how online recruitment campaigns are actually run in practice, plausibly with more starts, stops, and inconsistent active periods than a continuously running offline recruitment effort, or that the initial burst of online interest doesn't sustain at the same rate across a longer campaign.

Offline recruitment converted better once someone showed interest

The clearest finding in offline recruitment's favour was around conversion, the rate at which people who showed initial interest, or were screened, actually went on to enrol. Here, 9 of 13 studies, 69%, found offline recruitment achieved significantly better conversion rates than online recruitment, with a risk ratio of 0.8 favouring offline methods.

This is arguably the most operationally important finding in the whole analysis, because it changes what "recruitment success" actually means depending on which stage you're measuring. A campaign that generates a large volume of low-cost online interest, but converts a smaller share of that interest into actual enrolled participants, might still come out ahead on total enrolment and total cost, or it might not, depending on exactly how the conversion gap compares to the cost and speed advantage. A study relying purely on the cost-per-enrollee or speed headline figures, without accounting for this conversion gap, risks a genuinely misleading picture of which approach will actually hit an enrolment target.

What this means for planning a recruitment strategy

A few practical implications follow directly from a result this specific:

• Cost-per-enrollee and recruitment speed are real advantages for online methods, and worth using them for exactly that. The $72 versus $199 gap and the fourfold difference in per-active-day recruitment rate are strong, consistent findings across the reviewed studies.
• Don't assume the per-active-day speed advantage holds over a full recruitment campaign. The lack of significance in the full-period measure is a genuine caution against extrapolating short-term online recruitment performance into a full-study timeline projection.
• Budget for a lower conversion rate on online-sourced leads, and design the follow-up process accordingly. If offline conversion genuinely outperforms online by the margin this analysis found, a study relying heavily on online recruitment needs a stronger screening and follow-up process to close that conversion gap, rather than assuming volume alone will compensate.
• A blended strategy plausibly captures more of each method's strength than either alone. Given online wins on cost and per-day speed while offline wins on conversion, a recruitment plan combining both, rather than treating the choice as binary, is well supported by what this analysis actually found.

The broader lesson is a familiar one dressed in specific numbers: a recruitment channel that looks unambiguously better on one metric can look considerably more mixed once a different, equally important metric is brought into the comparison. Total enrolment, total cost, and total timeline all depend on how these separate effects, cost, per-day speed, full-period speed, and conversion, actually combine for a specific study's design, not on which single figure sounds most impressive in isolation.