My Research

This study analysed the increasing trends seen in population BMI in New Zealand over recent decades, using an Age-Period-Cohort (APC) analysis, and provides medium-term predictions of mean BMI and obesity prevalence for the next 20 years. APC analysis is a statistical methodology, widely used in disease epidemiology as well as the social sciences, for decomposing longitudinal trends in outcomes such as disease prevalence into the contributions of life-course changes (age effects), generational replacement (cohort effects), and secular time trends (period effects). In the case of BMI trends in New Zealand, we found that almost all of the observed and predicted increase in BMI was due to secular period trends, reflecting the effects of the increasingly obesogenic environment in which we live (such as the increasing availability of low-nutrient, high-energy ‘junk foods’, and reduced levels of physical activity due to sedentary occupations and passive leisure activities). A journal article reporting on these findings has been published in the Australian and New Zealand Journal of Public Health.

This work was part of a project, led by Prof. Haxby Abbott at the Centre for Musculoskeletal Outcomes Research, to develop and evaluate a novel, early, coordinated programme of care for osteoarthritis in the New Zealand healthcare system. For this programme, I have led the development of a new microsimulation model of the population health impacts, healthcare costs, and treatment outcomes of osteoarthritis in New Zealand, which will be used for evaluating the cost-effectiveness of the proposed programme of care. A journal article reporting on the model development and validation was published in Osteoarthritis and Cartilage, a study applying the model to provide projections of the future burden of osteoarthritis to the NZ health system and society has been published in the New Zealand Medical Journal, and an application of the model to cost-effectiveness analysis of interventions for the management of osteoarthritis in the NZ population has recently been published in Osteoarthritis and Cartilage Open as well as being presented at several conferences (including the OARSI World Congress, ISPOR European Congress, SMDM Annual Meeting, NZ Orthopaedic Association Annual Scientific Meeting, and Australian Physiotherapy Conference).

This project, which was funded by a Jack Thomson Arthritis Grant from the Otago Medical Research Foundation, includes two studies looking at the evaluation of the health-related quality of life impacts associated with osteoarthritis. The first study provides estimates of the health utility losses across multiple dimensions of health associated with radiographic knee osteoarthritis. To evaluate the cost-effectiveness of disease prevention strategies for osteoarthritis requires such estimates; this was the first study worldwide to provide these for radiographic knee osteoarthritis. An article reporting on this study has been published in Rheumatology.

The second study provides a tool to map from a generic quality of life measure (the SF-12) to health utility values in populations with osteoarthritis. The SF-12 is widely used in outcomes research in osteoarthritis, but the outcome measures commonly reported do not provide the utility values needed for cost-effectiveness analysis. This mapping will therefore greatly increase the range of studies available to inform economic analyses. A journal article reporting on this study is currently being finalised.

I was part of a team awarded a Health Research Council project grant to investigate the prevalence, management pathways, and health and economic impacts of hip and knee osteoarthritis in primary care in New Zealand. Most osteoarthritis care and management, in New Zealand and elsewhere, is delivered in primary care, but very little is known, at a systematic level, about how many patients present to primary care with osteoarthritis symptoms, what treatments they are offered, and how much it costs both the health system and patients themselves. This project combined novel large-scale data extraction from primary care patient records with the wealth of data on secondary care utilization, pharmaceutical prescriptions, and wider economic and social indicators available from the Statistics New Zealand Integrated Data Infrastructure (IDI), to address these important questions. Final research papers on the health and economics impacts of osteoarthritis diagnosed and managed in primary care in NZ are now being completed for publication.

I have been involved in several studies using data from patients undergoing total joint replacement or non-operative management for late-stage hip and knee osteoarthritis at Dunedin Public Hospital, investigating the outcomes of each of these management pathways and the optimal selection of patients for surgery. These studies have included evaluations of the ‘Joint Clinic’, a multidisciplinary chronic disease management pathway designed to improve access to joint replacement surgery for patients with hip and knee osteoarthritis by providing optimal non-operative management for patients with lower need for immediate surgery, with two articles published in the Journal of Arthroplasty; evaluation of the relationship between preoperative health status and postoperative improvement to help inform rationing decisions, published in Arthroplasty Today; and cost-effectiveness analysis of total joint replacement surgery based on preoperative patient characteristics and duration of follow-up, with a manuscript forthcoming in the Journal of Arthroplasty.