OBJECTIVE: This study investigated the cost-effectiveness of installing sidewalks to increase levels of transport-walking. METHODS: Secondary analysis using logistic regression established the association of sidewalks with transport-walking using two transport-walking thresholds of 150 and 60 min/week using Western Australian data (n=1394) from 1995 to 2000. Minimum, moderate and maximum interventions were defined, associated respectively with one sidewalk, at least one sidewalk and sidewalks on both sides of the street. Costs, average and incremental cost-effectiveness ratios were calculated for each intervention and expressed as 'the cost per person who walks for transport for more than 150 min/week (60 min/week) after the installation of new sidewalks'. A sensitivity analysis examined the robustness of the incremental cost-effectiveness ratios to varying model inputs. Costs are in 2012 Australian dollars. RESULTS: A positive relationship was found between the presence of sidewalks and transport-walking for both transport-walking thresholds of 150 and 60 min/week. The minimum intervention was found to be the most cost-effective at $2330/person and $674/person for the 150 and 60 min/week transport-walking thresholds respectively. Increasing the proportion of people transport-walking and increasing population density by 50% improved the cost-effectiveness of installing side-walks to $346/person. CONCLUSIONS: To increase levels of transport-walking, retrofitting streets with one sidewalk is most cost-effective.