Designing an instrument for police officers' life balance and sustainable lifestyle.

The raters disagreed so much about which domains mattered, with agreement of 0.51 in one group and 0.31 in the other where 1 means full agreement, that the model had to be built from the authors' own reading of the group discussions rather than a shared ranking. The full content validity process was not completed. Domain definitions came out ambiguous, and the authors state the whole content validity stage has to be repeated before reliability can even be tested, so the instrument is several steps away from being usable. The authors admit they should have collected qualitative comments alongside the numerical expert ratings, and should have included lay experts as well as content experts; the expert panel was nine occupational therapists only, and five of them had already taken part in the first focus group. Participants were recruited through convenience, purposive and snowball sampling, and the numbers are small, so the ratings come from a narrow group of people already interested in the topic. The officers who tested the items were Swedish frontline police, most of them men in patrol service, so how the items read to other officers or to other professional groups is unknown.

The authors declare no conflicts of interest. The study is part of a PhD research project, funded by the Department of Community Medicine and Rehabilitation at Umeå University and the Kempe Foundation. No company selling a product is involved.

Do not treat this as a finished lifestyle questionnaire you could hand to a police officer. There are no scoring rules and no interpretation of scores, the raters disagreed about which domains belong in it, and the authors say content validity has to be done again before any reliability testing can begin.