News from the front line

NPS MedicineWise, with funding from the Australian Government Department of Health, has released the first General Practice Insights Report using data from MedicineInsight. 

More than two million patients visit GPs every week across Australia, with GPs providing high quality and accessible services for patients. It is important to understand what is happening in general practice to ensure preventive and primary care remain strong pillars of the Australian health system.

Established in 2011, MedicineInsight collects de-identified data from the clinical information systems of participating general practices. 

MedicineInsight’s longitudinal primary care dataset contains de-identified data on 2.1 million patients and documents over 10 million encounters at Australian general practices in 2016–17, representing about 10% of all general practice patients in Australia. It shines a light on primary care in Australia, and the report is an exciting milestone on the journey towards realising the potential of MedicineInsight as a national primary care dataset.

Primary health care and why it matters

The cohort selected for the report consisted of 475 general practices, representing 5.9% of all general practices in Australia, and 2682 GPs, or 7.5% of GPs practising in Australia. The report is a window into the Australian primary health care system, where most preventive care happens and where many chronic conditions are managed. 

Every Australian will benefit from a strong primary care sector and an important contributor to this is being able to see what is actually happening on the front line of health care.

Working with general practices to improve outcomes

MedicineInsight isn’t only used to provide big-picture policy reports like the one that’s just been released.

Data from the program is also fed back to participating practices, to help them optimise the care they provide to patients, identify patients who need follow up and compare their care of patients with that provided by similar practices. This is a very concrete way in which MedicineInsight is improving primary care on the ground.

What’s happening out there? 

So what insights has the report delivered? What do we now know about Australians and their general practice visits?

Why do we go to the doctor?

According to the reason for encounter recorded in GP notes, the top five reasons that Australians went to see their GPs in 2016–17 were:

  • to obtain a prescription
  • for review/follow-up purposes
  • for an upper respiratory tract infection (URTI)
  • for immunisation, and
  • to discuss results.

What’s wrong with us?

Of 14 of the most common conditions that were identified in practice software for the 2.1 million patients, the top 5 were:

  • hypertension (high blood pressure)
  • depression
  • dyslipidaemia (high cholesterol levels)
  • anxiety, and
  • asthma.

Many patients also have health risk factors. Smoking status was recorded by doctors for 82% of patients over 16 years of age, allowing further analysis of this potent risk to health.

On the other hand, BMI was recorded, or could be calculated, for only 29% of patients over 18. Only 14% of patients aged over 16 had alcohol use recorded, of whom 42% were recorded as non-drinkers.

What are doctors prescribing?

Given that obtaining a prescription is the top reason for visiting a GP, what are GPs prescribing? This information comes from the automated systems used to print prescriptions out for patients, and the dataset doesn’t record whether the prescription was ever filled at a pharmacy. Even so, it’s striking that more than a third of patients had one or two prescriptions during the year, another third had 3 or more, and 31 % had no prescriptions recorded.

Prescribing patterns varied between men and women, and between more and less disadvantaged areas of Australia, and in general, numbers of prescriptions per patient increase with age.

The most common medicines prescribed during 2016–17 were medicines for the cardiovascular system, making up 29% of total prescriptions.

What tests are being ordered?

On average, 25 individual pathology test results were recorded per patient in the MedicineInsight data. Women had higher average test numbers than men, and averages also varied between areas of socio-economic advantage and disadvantage, with more tests ordered in disadvantaged areas.

The top 5 pathology tests carried out in 2016–17 were:

  • haemoglobin (as a proxy for full blood count (FBC)
  • ALT (as a proxy for liver function tests (LFTs)
  • sodium (as a proxy for urea, electrolytes and creatinine (UECs)
  • total cholesterol (as a proxy for lipids)
  • TSH (as a proxy for thyroid function tests).

One group of patients had more than 80 pathology tests in the study period, and these were significantly more likely to have one or more chronic conditions, particularly chronic kidney disease or cardiovascular disease. 

Using the data that doctors already collect

MedicineInsight collects de-identified data general practitioners and other practice staff gather while seeing their patients and enter into their clinical information systems. An ongoing project is to help doctors understand how to improve the quality of the data they produce.

Clinical information systems

Australian general practices have been computerised for many years, and almost all of them use one of only a handful of clinical information systems. 

MedicineInsight interfaces mainly with two of these, under strict data security and governance conditions, extracting the anonymised data on a monthly schedule that over time produces longitudinal records for all the patients in the practice. 

Practices and patients are classified by their state, regionality and by socio-economic status. Risk factors, such as smoking status and BMI, can be collected as well. However, as with many other datasets that aren’t collected with the primary intent of being used for research and analysis, the data has limitations that should be kept in mind.

Keeping the data safe

The MedicineInsight program has rigorous governance processes to ensure the program is run ethically and for the public good. Information is available in participating practices to inform their patients, and patients who do not wish to participate in the program may opt out.

Data are always encrypted during transit and storage, following government and industry best practice standards. MedicineInsight data are collected, used and stored strictly in line with Australian privacy laws.

Sharing MedicineInsight data only takes place within a robust data governance framework, including approval by an independent Data Governance Committee. Consumer advocates, privacy and security experts, general practitioners and researchers are members of the committee.

Weighting the data

MedicineInsight worked with the Australian Bureau of Statistics to develop a weighting procedure using other national data, such as Medicare claims data, to ensure the representativeness of the data for this report.

This procedure was used to weight the statistics for patients, encounters, prescriptions and pathology results, producing a representative national dataset for general practice in Australia.

Looking into the future

The report is an important milestone in working with general practices so that clinical data can be used to inform policy and research quality to improve health outcomes for all patients across Australia. Australia currently has a lot of data about the management of patients in hospital and it is also important to understand what happens in general practice where most patients receive most of their care most of the time.

As the program evolves and matures against the backdrop of the fast-moving Australian digital health environment, future reports will reflect shifts in our knowledge and understanding of the data. The report is a preliminary exploration of how MedicineInsight data can describe general practice activity in Australia. The next General Practice Insights Report for 2017–18 is already in development. Stay tuned for more insights.