CME Evaluation Incentives: Put the Gift Card in the Methods

An evaluation incentive is part of how feedback was collected. Put it in the methods section of a CME outcomes report, alongside who was invited, when they were asked, and which questions they answered. A larger pile of evaluations may be useful. Interpreting it still requires knowing whose views are in the pile and what those answers can establish.

AOE Consulting's August 3 question about gift cards for activity evaluations prompts an outcomes question beyond eligibility: if an incentive brings in more responses, what can the report legitimately claim?

Keep the eligibility decision with your accreditation lead. ACCME's dated April 2026 newsletter reports a policy prohibiting personal remuneration associated with purchasing or registering for accredited CME, regardless of who pays. An evaluation-recruitment plan needs its own review; a finding about survey response is not permission for a registration promotion.

More returns, different claims

Consider this invented report comparison, with no observed program results:

• An activity previously received 40 evaluations from 200 invited learners. The average satisfaction rating was 4.6 out of 5.
• After an incentive was introduced, a comparable activity received 80 evaluations from 200 invited learners. The average rating was again 4.6.

The report does not say how many respondents answered the satisfaction item. The evaluation return rate rose from 20% to 40%. The mean reported satisfaction stayed the same. Neither line shows that learning improved, that all learners were equally satisfied, or that the incentive caused the increase. If the team also changed survey length, reminders, or timing, the before-and-after comparison cannot isolate the incentive's contribution.

A defensible report would say that more invited learners returned evaluations under the revised collection process. It would describe that process and identify any important changes in the respondent group. It would reserve a learning or practice-change claim for a measure designed to address that claim.

A 2023 review of 46 randomized survey trials found that monetary incentives increased response rates. Its endpoint matters: the review did not examine the accuracy or reliability of the answers collected. It covered varied populations and survey settings, with limited evidence for online surveys. It supplies a reason to consider recruitment effects, not a forecast for a particular CME evaluation.

Response gain and respondent mix can be examined separately. A 2005 randomized physician survey sent a small scratch-lottery ticket with the first questionnaire mailing to Australian general practitioners and specialists. This was an upfront incentive, not a reward promised after an evaluation. After respondents were told the incentive research question and allowed to withdraw their data, the researchers reported rates of 49.7% with the incentive and 40.1% without it, using retained replies against the invited groups. The researchers reported no systematic demographic differences between respondents and nonrespondents. Those checks are useful context from a mailed physician survey. They do not establish representativeness or answer validity for an online CME program.

Read the denominator before the rating

The AMEE guide on response rates and nonresponse bias, published online in 2015, treats both as survey-validity concerns. Completing the whole survey and answering an individual question are also different events. A high overall return rate can coexist with an important item that few people answer.

For an outcomes team, that suggests a practical reading order: invitations, usable returns, answers to the particular item, then the result. Check whether intended learner roles or practice settings are missing from the respondents, using appropriate aggregate information already available to the team. A demographic match is useful context; it cannot prove that nonrespondents would have answered the substantive questions identically.

Our earlier discussion of making routine CME survey questions useful asks each item to support a decision. Add the collection context to that discipline. A question with a clear purpose can still be overinterpreted if its denominator or recruitment method disappears from the report.

There are also ways to reduce the effort of responding. In a January 2024 SACME coffee chat, a participant described sharing the survey near the end of the program, announcing it, and giving learners time to complete it before returning to clinical work. This was an operational account, not a controlled comparison. It makes timing worth examining alongside an incentive budget.

When the next evaluation report reaches your desk, locate the collection method and the denominator for its headline item. If they are missing, ask for them before using the rating to justify a program decision. The extra responses may improve what the team can learn; the report should show how, rather than letting their number carry the whole argument.