Introduction Despite its effectiveness, the optimal use of the combination of insecticide treated nets (ITN) and intermittent preventive treatment during pregnancy with sulfadoxine-pyrimethamine (TPIg-SP) remains low in malaria endemic areas. Therefore, we studied its variations and predictors in Guinea.MethodsWe conducted a secondary analysis of the 2012 and 2018 Guinea Demographic and Health Surveys (DHS). It included women who had given birth in the 3 years before each DHS, slept on ITN and took at least one dose of SP. Moran indices were used to determine spatial autocorrelation. A logistic regression and classification and regression tree (CART) identified the predictors of malaria preventive measures (MPM).ResultsIn 2012, 60.88% of pregnant women used MPM incompletely compared with 79.11% in 2018. Associated factors with incomplete MPM in 2012 were as follows: having an indirect link with the head of household (AOR = 2.23, 95% CI 1.08–4.61) and performing at least 4 ANCs (AOR = 0.66, 95% CI 0.44–0.99). In 2018 : live in households of 2 to 5 people (AOR = 0.54, 95% CI 0.36–0.80), have a man as the head of the household (AOR = 0.56, 95% CI 0.35–0.89), perform the first ANC in the second quarter (AOR = 0.74, 95% CI 0.54–0.99), perform at least 4 ANCs (AOR = 0.47, 95% CI 0.36–0.62), have a job (AOR = 0. 67, 95% CI 0.50–0.88), give birth in a public health facility (AOR = 0.53, 95% CI 0.39–0.72) and the middle wealth quintile (AOR = 1.56, 95% CI 1.07–2.26). A global autocorrelation (I Moran=0.0009, p = 0.2349) and high-high clusters were found in Mamou in 2012. In 2018, autocorrelation was found (I Moran=0.0169, p ≤ 0.05), with spatial clusters in regions of Boké, Mamou, Labé, and N’Zérokoré.ConclusionComplete use of MPM during pregnancy remains still low throughout years in Guinea. Strategies toward familial and healthsystem levels should be implemented and monitored, to reduce inequality in the use of MPM. In this regard, Mamou region deserves more attention.