OA affects one in eleven Australians and is the leading cause of disability. It is crucial to provide appropriate care for this large population. In Australia, 98% of knee replacements are attributable to pain and mobility problems associated with osteoarthritis (OA). The rate of total knee replacement (TKR) was one of the highest in the world; it also varies substantially by areas. Some of the variations may not be related to patients’ needs or preferences, or unwarranted variation. Unwarranted variation raises concerns about equity and appropriateness of care. To reduce the potential harms and inappropriate patient care for people with OA, we must identify unwarranted variation, understand the contexts influencing utilisation and contributing factors in order to properly address the problem and reduce inappropriate care.
Our research is the first Australian-wide study that investigates contexts influencing utilisation and contributing factors to geographical variation in TKR surgeries, taking into the local health demand or healthcare supply contexts. Understanding these factors holistically is an important step to reducing inappropriate use of TKR and optimising orthopaedic care to people who are in genuine need. The evidence derived from the research has a direct impact on the appropriateness levels and quality of OA care and contributes to the development of overuse of TKR by people with OA.
Our research explored the magnitude of the variation between local government areas (LGA), a much more granular geographical scale than what has been investigated in the current literature. We hope to understand where, for whom, and why the rates of knee replacement vary, and provide evidence for optimal approaches to responding to variation in TKR.
We found that there was a significant variation in TKR at the LGA level and people living in areas with higher arthritis prevalence and higher surgeon density received more TKR. The evidence suggests that demographic patterns and disease prevalence in the local areas have great implications for predicting and planning OA service needs.
The grant aimed to investigate the magnitude of and associated drivers for variation in TKR surgeries in Australia. There is currently relatively little research in the existing literature and hence a lack of evidence supporting the best methods or models to respond to the TKR variation. The variation in TKR revealed in the Australian Atlas of Healthcare Variation Series was at a very broad regional level and did not investigate the potential drivers. Measuring variation and investigating the contributing factors are important steps to reduce inappropriate use of knee replacement. Previous methods to improve OA service delivery have largely focused on changing individual clinician and patient behaviour, but strategies targeting specific regions to implement change may be more efficient.
We found that the TKR rate for OA in Australia was one of the highest in the world (227.1 per 100,000 population), more than double the rate in New Zealand (111 per 100,000). There is a significant variation across LGAs in Australia: over 30-fold difference between the highest and lowest areas, after the exclusion of the highest and lowest 10%. We also found that older Australians, areas with higher arthritis prevalence rates and higher density of orthopaedic surgeons had higher TKR utilisation rate. This suggested that inequalities in the utilisation of TKR were associated with local demographics, health needs and service supplies.
We also hypothesized that areas with high socioeconomic status, higher private health insurance and higher prevalence of obesity, one of the important risk factors for OA, may have high TKR rates. However, this is not the case when we adjusted patients’ age and sex within their local LGAs. There remained a significant variance in TKR utilisation after considering all the factors investigated in our research, suggesting other factors, relating to individual patients, surgeons or hospitals also contribute to the variations in TKR. It is important to establish whether provision matches actual need across socio-demographic groups to determine potential areas of service inequity across the country, which is what we plan to extend the research in the future.
This is the first Australian-wide study that investigates factors associated with variation in TKR surgeries at a small geographical scale. By examining patterns of OA and hence the demand for TKR, we can understand where unwarranted variation in TKR and potentially inappropriate care are provided for people with OA. The research is not only necessary but also timely for health policymakers to take actions to ensure right treatments and health services are provided to and accessed by the right people in the right areas, and thereby enhance health outcomes for people with OA in Australia.
We investigated variations between patient residential areas in the current research and plan to extend the research in several ways. We will improve the measurement for orthopaedic care supply through GIS approaches to measuring distance-to-surgeon GIS approaches. Additionally, we plan to examine variations between hospitals, taking into care supply factors. This will provide a clearer and more complete picture on the variation from both where patient lives and where they undertake surgeries, so as to ensure service provision matches actual need across socio-demographic groups in the local areas across the country.