It is unclear whether suboptimal adherence contributes to adverse clinical outcomes in patients with chronic hepatitis B (CHB). Moreover, there is no consensus regarding the optimal level of drug adherence. This was a population-based historical cohort study including 51 975 adult CHB patients treated with entecavir (0.5 mg/d orally). Data were obtained from the Korean national health insurance service claims database, which covers >99% of the entire population, between 2007 and 2015. Medication adherence was categorized as high (proportion of days covered [PDC], ≥90%; n = 32 089), intermediate (PDC, 80%-89%; n = 10 197) and low (PDC, <80%; n = 9689). During a median 4.5 years (maximal 9 years) of follow-up in 51 975 CHB patients treated with entecavir, multivariable analyses revealed that the risk of mortality/transplantation was significantly greater in the low-adherers (adjusted hazard ratio [HR], 1.38; P < .001) and intermediate-adherers (adjusted HR, 1.44; P < .001) than the high-adherers (P for trend < 0.001). The risk of renal failure in the low-and intermediate-adherence groups was also significantly higher than the high-adherence group (P for trend < 0.001). By contrast, the risk of hepatocellular carcinoma (HCC) was not significantly different between groups (P for trend = 0.70). The higher risk of mortality/transplantation and renal failure but similar risk of HCC for low-and intermediate-adherers compared with high-adherers was consistent in inverse probability treatment weighting analysis of the entire cohort and subcohorts with or without cirrhosis. In conclusion, high medication adherence (≥90%) is required to significantly lower risk of mortality and renal failure in patients with CHB during long-term treatment with entecavir.