ObjectivesHigh-risk prescribing can have deleterious effects on the health of older people. This study aimed to assess the role of inappropriate prescribing on changes in frailty status over 3 years of follow-up. Design, settingThis is a prospective observational study nested in the GAZEL cohort. ParticipantsThe study sample included 12,405 community-dwelling people aged 58 to 73 in 2012, and followed for 3 years. MeasurementPolypharmacy and potentially inappropriate medications (PIMs) were assessed from reimbursement data by the French National Health Insurance. Frailty was evaluated each year with the Strawbridge questionnaire. PIMs were defined according to the Laroche list plus additional criteria dealing with inappropriate prolonged use of medications. The relationship between PIMs and changes in frailty status (incident frailty and recovery) was analyzed with Markov multistate modeling. ResultsThe prevalence of frailty increased from 14% in 2012 to 17% in 2014, whereas the frequency of PIMs was 29% in 2012 and 23% in 2014. Polypharmacy (5-9 drugs: aHR 1.31, 95% CI 1.14-1.50; and 10 drugs or more: aHR 1.57, 95% CI 1.28-1.92) and potentially inappropriate use of nonsteroidal anti-inflammatory drugs (aHR 1.33, 95% CI 1.04-1.71) were significantly associated with incident frailty, when the presence of at least 1 PIM presented a small association with the risk of becoming frail (aHR 1.15, 95% CI 1.01-1.32). Conclusions/ImplicationsThis study brings new elements to our knowledge regarding the association between inappropriate prescribing and frailty in older adults, which support research development to alert on inappropriate prescribing and to improve drug prescribing among old people, especially with polypharmacy.