“…Additional limitations not already mentioned include a lack of control groups [51,[56][57][58][59][60] possibly inappropriate or insufficient control or comparison groups [52][53][54][55]; hospital-based rather than population-based interventions [58,59]; non-random sampling [57]; limited description of the intervention [54,58]; older studies [54,59]; small sample size or being underpowered [51,52,55]; biased by being implemented with a cost to some and free to others [51]; lack of exposure to/limited dissemination of the intervention [55]; assessment of hazard reduction rather than injury occurrence [43,52,53]; possible observer bias [43,52,53]; and short time to follow up after intervention [43,52,53,57]. Most importantly, in low resource settings, only three of the studies address cost: Chamania et al address the reduction of 'cost of illumination' per year, and Jetten et al and Swart et al mention some specific costs relating to certain parts of the intervention initiative [51,52,57]. None of these studies address long-term sustainability.…”