Evaluating interventions honestly
Agree success measures and limitations before delivery. Improvement after a programme does not by itself prove the programme caused it.
The performance question
A team’s results may change because staffing, leadership, demand or recording changed at the same time. Evaluation should examine implementation and the proposed mechanism as well as outcomes. Avoid presenting a favourable testimonial as evidence of organisational effect.
What the evidence supports
NICE NG212 recommends monitoring and evaluating support at least annually and evaluating manager training through employee outcomes. It recommends validated tools for mental wellbeing. These support planned evaluation, not a guarantee that any observed change is causal.[1]
A practical approach
Document the baseline, intended change, comparison where feasible and review period. Track delivery, access and unintended effects. Record other changes that could influence results and use a suitable validated tool if wellbeing is the actual outcome. Involve a competent evaluator where the claims or investment justify it.
How to review progress
Report results with sample size, missing responses and uncertainty. Distinguish no effect from insufficient evidence and do not suppress adverse feedback. Decide whether to continue, adapt or stop using the agreed criteria, rather than choosing a success measure after seeing the data.
Sources and references
- Mental wellbeing at work Opens in a new tab
NICE
NG212 sections 1.1 to 1.6: strategy, environment, managers and individual support