Maintaining individuals with health limitations in the labour force is a challenge of growing importance. Determining the effect of health on occupation may tell us how people adapt to their limitations, and what types of jobs make this harder or easier. This paper uses the first 14 waves of the Household, Income and Labour Dynamics in Australia Survey to examine the effect of health and changes in health on occupation for the working-age population. I use dynamic panel models which account for selection into employment. Two measures of occupation are used to capture two aspects of occupation highlighted in the literature as being linked to health: physical job demands and status. The results of the analyses provide some evidence that a health shock reduces the likelihood of manual employment for younger men, suggesting that younger men may adapt to a health shock by reducing physical job demands. Worsening health and work-limiting long-term conditions are found to have a negative effect on occupational status for men and women, suggesting health selection into lower-status jobs, and an adverse effect of poor health on occupational mobility.