Community-led total sanitation (CLTS) is a widely used approach to reduce open defecation in rural areas of low-income countries. Following CLTS programs, communities are designated as open defecation free (ODF) when household-level toilet coverage reaches the threshold specified by national guidelines (e.g., 80% in Ghana). However, because sanitation conditions are rarely monitored after communities are declared ODF, the ability of CLTS to generate lasting reductions in open defecation is poorly understood. In this study, we examined the extent to which levels of toilet ownership and use were sustained in 109 communities in rural Northern Ghana up to two and a half years after they had obtained ODF status. We found that the majority of communities (75%) did not meet Ghana's ODF requirements. Over a third of households had either never owned (16%) or no longer owned (24%) a functional toilet, and 25% reported practicing open defecation regularly. Toilet pit and superstructure collapse were the primary causes of reversion to open defecation. Multivariate regression analysis indicated that communities had higher toilet coverage when they were located further from major roads, were not located on rocky soil, reported having a system of fines to punish open defecation, and when less time had elapsed since ODF status achievement. Households were more likely to own a functional toilet if they were larger, wealthier, had a male household head who had not completed primary education, had no children under the age of five, and benefitted from the national Livelihood Empowerment Against Poverty (LEAP) program. Wealthier households were also more likely to use a toilet for defecation and to rebuild their toilet when it collapsed. Our findings suggest that interventions that address toilet collapse and the difficulty of rebuilding, particularly among the poorest and most vulnerable households, will improve the longevity of CLTS-driven sanitation improvements in rural Ghana.