Background Child death review teams (CDRTs) are multi-agency, multidisciplinary teams that review the circumstances surrounding child deaths. Although the potential of CDRTs to promote systems improvement and prevention is well recognised, teams often struggle to translate their findings into effective preventive actions.

Objective To present results from a study assessing the quality of written recommendations in published CDRT reports; and provide guidelines for improving the quality and effectiveness of these written recommendations.

Methods A descriptive, non-experimental design was used to analyse a set of 1093 recommendations from 21 randomly selected, publicly available state and local CDRT reports. An assessment instrument, modelled on the public health approach, was developed to score the quality of recommendations. It consists of three components divided into 10 dimensions: problem assessment; written recommendations; and action on recommendations.

Results CDRT reports scored highest in the problem assessment component (mean score: 2.7/dimension), followed by written recommendations (2.2/dimension), and action on recommendations (1.9/dimension). Even among the highest ranked dimensions, the average scores were only in the mid range of quality on our assessment scale.

Conclusions The results suggest that CDRTs are doing a better job of ‘assessing the problem’ than in ‘proposing solutions’ as indicated by their written recommendations. CDRT reports often do not address follow-up of their written recommendations. Guidelines are offered for use as a practical tool to help CDRTs enhance the likelihood of producing effective recommendations that prevent future child injuries and deaths.