Big data is making huge advances in reducing hospital waiting lists

Photo by Christopher Furlong/Getty Images

Somewhere deep in the data held by Australian hospitals — the routine and dull stuff about admissions, type of treatment and the time of day — will be the solution to one of the big problems of health services.

That is waiting lists for elective surgery, the expanding period of time people have to wait to get something fixed such as a dodgy knee or a new hip installed or an eye cataract removed. This is scheduled surgery and not the emergency type when someone is in an accident and needs attention immediately.

Australia has been trying to reduce the times on its hospital waiting lists for more than two decades. Nothing seems to work and people are still waiting, according to a 2013 OECD study.

First, we tried subsidising private health insurance, via the tax system in the form of general rebates, to try to shift demand from public to private hospitals.

Then the states were given financial incentives to get their public hospitals achieving better waiting times.

“Despite these expensive efforts, waiting times barely changed, with the median even increasing slightly,” says the OECD study.

“There is some evidence that state-based programs are more effective than national ones, but their impacts have been short-lived. Several features of the current system for managing waiting lists may contribute to long waiting times, including the wide discretion given to specialists in assigning urgency to patients on the waiting list.”

The data collected by hospitals, but so far kept in different places and in a range of systems and formats, is vast. There are almost 10 million visits to the Australia’s 747 public and 612 private hospitals each year. Hospitals spend more than $55 billion and employ 287,000.

Source: Australian Institute of Health and Welfare

Of course, hospitals are at the mercy of fate, or so it would seem. Road accidents happen at odd times and the treatment of victims can unexpectedly tie up hospital resources which would have been used for elective surgery.

However, experienced traffic police officers will tell you that accidents tend to happen later in the day, as the sun starts to fade. And they are more likely to happen on days when there is a slight drizzle of rain.

Another example is testing the old saying that hospitals get busier when the moon is full. Easy to work out if its true or not if you can access the right data.

And apparently there is no evidence of a Lunar effect on hospital admission rates or births.

Analysis of the data, when this information can be extracted from hospital system and analysed, will uncover relationships to problems.

And the process will also reveal some issues the hospitals didn’t know they had, predictions about the spread of disease and more efficient uses of medical resources to help meet a so far insatiable demand for health care.

It’s all about better heath care and lower costs. Big data analysis is being taken up in healthcare across the world.

However, in Australia this big data analysis for health care is only just getting started.

Juggling resources

Down in the hospital corridors and in operating theatres, medical staff work away improving what they have.

“There are so many factors pushing and pulling on how you use a facility,” Gary Morgan, clinical director at Sydney’s Westmead Hospital, tells Business Insider.

“It’s quite a juggle at times. This morning we had 16 patients waiting in the emergency department without beds for surgery alone. A very busy day and required a bit of shuffling around. One or two patients didn’t get their planned operations because of that. It’s a complex system to manage.”

Morgan says the hospital has changed its processes a lot in the last few years which has considerably improved waiting list performance.

“There are fairly established principles about the efficient management of waiting lists — such as treating patients in turn rather than cherry picking off lists — which are not too difficult to implement and which we monitor,” he says.

“There are data driven capacity planning tools that are out there in the marketplace that we are exploring but aren’t in widespread use at this point.”

“Some of that is because we lack the data management capabilities to make them work and some of which it is expensive to introduce these tools. So there are quite a lot of impediments to introducing them.”

“But we are currently exploring that and have been for some time. I gather that there are such tools in use in other parts of the world but they are not in widespread use in Australia.”

Stay away from hospitals at the weekend

One of the simplest examples of using big data to improve health care is analysis of admission and mortality rate information.

An analysis of this data brings the simple advice that it’s a poor idea to be admitted to hospital at the weekend. If you have any choice, and are scheduled for elective surgery, make sure it’s Monday to Friday.

Researchers, in a study published in the journal BMJ Quality & Safety, found a bigger risk of death for those admitted to hospital on a Saturday or Sunday.

They used data on almost 3 million admissions between 2009 and 2012 from 28 metropolitan teaching hospitals in Australia, the UK, the US and the Netherlands.

All patients admitted at the weekend for planned surgery were more likely to die within 30 days than those admitted on other days of the week.

The researchers speculate on the reasons for the findings but no single factor appears responsible. Some surgical procedures may be sensitive to reduced access to test results and diagnostics at weekends. And there might be fewer and less experienced staff at the weekend, meaning patients needing urgent care may have to wait longer.

The risks of intensive care

Another example is Risky States, a project by the Massachusetts Institute of Technology and Aptima Inc, which estimates the risk levels at intensive care units at different hospitals.

Again, hospital records were analysed, this time to create a statistical approach to identify the drivers of risk at intensive care units.

People in intensive care are by definition at risk. They are very sick and some will die but they don’t have to be victims of preventable complications such as infections.

The Risky States created a system which analyses the risks attached to each patient by using data from medical records, past cases and information inputted by clinical staff.

In Australia, the federal government is creating digital medical records for everyone. This project, supported by state health departments, means there will be a mountain of data about health needs of Australians.

From this more targeted planning can be done, finely estimating the services, equipment and people needed in the future. The idea is to make every dollar count, reduce wastage and deliver better health care.

“We’re going to see the big data analytics,” says David Hansen, CEO of the eHealth Research Centre at the CSIRO, Australia’s national science agency.

“An example where we’ve done some work is our patient admission prediction tool. This predicts, based on historical data, how many patients are going to turn up in the emergency department and how many will go on to need a bed.”

Data analytics will drive efficiency in the health system. “That’s based on the electronic medical records and the electronic administrative systems which we’re putting in at the moment,” he says.

Business Insider Emails & Alerts

Site highlights each day to your inbox.

Follow Business Insider Australia on Facebook, Twitter, LinkedIn, and Instagram.