This week Dr Kevin Ross, General Manager of Precision Driven Health and Haze White, Māori Health Researcher, spoke to Radio New Zealand about countering ethnic bias in the health system to ensure the findings of NZ health data research is meaningful for all groups of the population.
The key topic of discussion was the reality that the health gap between Māori and Pakeha in New Zealand is growing further apart. One critical impact of this is a discrepancy in access to health care. Studies have shown Māori and Pacific people have comparatively limited access to GPs. According to White this is partly due to barriers such as language, culture and varying levels of health literacy.
Inefficient triaging of referrals is part of the reason Māori patients often are delayed in seeing a specialist. Prioritising specialist referrals can have significant implications on both the utilisation of resources and patient outcomes. For instance, if a patient is seen sooner it impacts other patients and determines whether there will be delays in their treatment. You would assume that if a population group had twice the risk of developing a serious health issue, there would be twice the number of referrals coming through from primary to secondary care, however this is not the case.
Part of the challenge is that overseas studies are often based on European people, so the findings tend to be better suited to New Zealand’s pākehā population. Dr Ross explained, “There’s lots of great research going on but overall, there are more benefits happening for those who are already relatively healthy rather than those who have a lower health level to start with”. He reiterates that we need to ensure these groups are considered and accounted for accurately and engaged in the research itself, so that this bias doesn’t continue to impact Māori health in the future.
“If what we’re inputting is historically biased data then we can expect to get historically biased results”, says White. Precision Driven Health is a public-private research partnership that is committed to ensuring the health research it undertakes is going to be useful for everyone in the entire New Zealand population. This means that machine learning models, for example, must be built with accurate representation of Māori and other ethnic groups, taking into account factors such as comparatively limited access to primary healthcare.
Precision Driven Health is committed to aligning with the New Zealand Health Strategy to ensure that their research and the digital tools they develop are able to provide equivalent benefits to Māori as to the rest of the population. An ambitious project is underway, applying deep learning, a form of machine learning, to GP e-referral data to help cardiologists in triaging these referrals. The project will involve building predictive models from existing clinical data to predict high risk patients, allowing for early intervention before it’s too late.
This project will analyse referral data, from primary into secondary care. In a hospital, the specialists spend a lot of time trying to read through referrals each day to figure out who to prioritise, which takes away from time with patients. Dr Ross says where the Precision Driven Health partnership can help is using data analysis to develop decision support tools that can save specialists time and support them in deciphering which patients are most in need of treatment. One of the biggest challenges of the health system is making sure the right people get seen first, and prioritising patients for care in a short space of time can be extremely difficult in the high pressure environment of the hospital.