Risk factors for incident kidney disease in older adults: an Australian prospective population-based study

Background and Aim To determine risk factors for incident chronic kidney disease (CKD) in a large population-based cohort. Methods This prospective opt-in population-based cohort study is based on the 45 and Up Study, where New South Wales residents aged >= 45 years were randomly sampled from the Services Australia database and agreed to complete the 45 and Up Study baseline questionnaire and have their responses linked to their health data in routinely collected databases. The primary outcome was the development of incident CKD, defined as eGFR < 60 mL/min/1.73 m(2). CKD incidence was calculated using Poisson regression. Risk factors for incident CKD were assessed using Cox regression in multivariable models. Results In 39 574 participants who did not have CKD at enrolment, independent factors associated with developing CKD included: older age, regional residence (HR 1.38 (1.27-1.50) for outer regional vs major city), smoking (1.13 (1.00-1.27) for current smoker vs non-smoker), obesity (1.25 (1.16-1.35) for obese vs normal body mass index), diabetes mellitus (1.41 (1.33-1.50)), hypertension (1.53 (1.44-1.62)), coronary heart disease (1.13 (1.07-1.20)), depression/anxiety (1.16 (1.09-1.24)) and cancer (1.29 (1.20-1.39)). Migrants were less likely to develop CKD compared with people born in Australia (0.88 (0.83-0.94)). Gender, partner status and socioeconomic factors were not independently associated with developing CKD. Conclusions This large population-based study found multiple modifiable and non-modifiable factors were independently associated with developing CKD. In the Australian setting, the risk of CKD was higher with regional residence. Differences according to socioeconomic status were predominantly explained by age, comorbidities and harmful health-related behaviours.