Why Isn't Predictive Analytics A Bigger Part Of Patient Care?

The NHS are lagging behind in terms of their use of data


Digitalization has been a long and arduous process for the NHS. IT systems have failed - often at huge cost to the taxpayer - and a large proportion of note-taking by doctors is still, even now, done by hand. This means both that there is far greater potential for error than there would be if notes were taken electronically, and that it has to be couriered from one area of the hospital to the other, causing unnecessary delays that may prove costly.

There are a number of reasons for the NHS’s failure to keep up with technological advancement. Investment, regulations, and the practicalities of introducing such technologies in a clinical environment have all meant that it is not as simple as dishing out a few iPads to clinical staff and telling them to get on with it. However, not keeping up has had a significant knock-on on their ability to adopt technologies coming out today that could dramatically improve patient care. One of the most notable examples of these is predictive analytics.

In medicine, the applications for predictive analytics are far reaching. The ability to predict potential health issues is, obviously, hugely beneficial. Catching something early greatly improves the chances of any disease being cured. It can also be used for things like improving chronic disease management initiatives and lowering hospital readmission rates. The move to a predictive model of healthcare should also greatly reduce costs - something vitally needed as baby boomers hit old age.

One example of predictive analytics being used effectively is at Parkland Health and Hospital System in Dallas, Texas. They have have come up with a validated algorithm based on Electronic Health Records (EHR) that predicts readmission risk in patients with heart failure. Those found to be at high risk of readmission are educated on how best to minimize the changes by a multidisciplinary team, given follow-up support via telephone within two days of discharge to ensure medication adherence, have an outpatient follow-up appointment within seven days, and a non-urgent primary-care appointment. The initiative was found to have slashed readmissions by 26%.

It is one thing providing treatment to patients though, it is quite another ensuring that they take the medicines prescribed. In 2013, medication non-adherence cost the healthcare industry approximately $337 billion. Predictive analytics tools can provide a solution for this, however, enabling providers to pinpoint when a patient will stop following a medication plan so that they can intervene accordingly.

These examples show how well predictive analytics can work, but they are still the exception and not the rule. Data is now considered a fact of life in most industries, but in healthcare it seems as if it is still often perceived as a burden. In a recent survey by HealthPrize, 24% of primary care participants reported that they did not want to acquire drug adherence information for each patient because it would lead to ‘data overload’. Meanwhile, just 44% of respondents stated they were satisfied with the helpfulness of the patient data they get from pharmacies and health insurance companies.

Such a response is revealing. Firstly, it highlights the simple fact that doctors typically don’t have a lot of time. This is either a fault in the amount of training that doctors are given around data, or a fault in the analytics tools not being well integrated into the existing workflow so they do not cause delays. Healthcare providers must ensure that processes are in place so that neither of these are a problem.

Most important is that the infrastructure is in place to collect, analyze, and leverage all possible healthcare data. For an institution as unwieldy as the NHS, in which siloed data is common place, creating a data-driven culture is no easy task. The training and tools must be in place to garner and action the insights gained from predictive analytics. This will likely require wholesale structural change, and complete upheaval in ways of working. Whether there is sufficient investment to enable this remains to be seen. 

University lecture small

Read next:

How Are Higher Education Institutions Using Analytics?