Health October 4, 2019
Moving the bell curve

By Simon Swift - World Healthcare Journal

Throughout healthcare, the requirements of care will ebb and flow to the demand of populations, genetics, environment, economy and a multitude of other factors. However, what should not ebb and flow is the standard of care.

Depending on location, at both hyper-local and international levels, the ability to access the right standard of care can fluctuate drastically. Separate clinics across a country will provide different treatment methods for the same disease, which is understandable and necessary for the advancement of research and treatment development. But standards of care should never be detrimentally different.

For example, in some types of lung cancer, you can treat patients with only chemotherapy, only radiotherapy or only surgery, or varying combinations of all three. In theory, the rate of these different treatment modalities should be largely even between speciality clinics and organisations. But often an organisation will only deliver one form of treatment, irrespective of the patient’s unique condition.

Solving treatment variance through data-driven work

In order to tackle this problem which occurs through many speciality clinics and health systems at large, Methods has begun developing an approach to examining a whole health system to understand the individual standards of care delivered by clinical specialists and tease apart how well that population is served.

This approach allows us to understand what is actually occurring across a wide range of clinical specialities. What is the population receiving? What does the population need? How does that vary with what a provider is supplying? Moving forward with that data, we can then start working with organisations, governments, and care systems to start to standardise the high volume areas of care and the high variation areas of care.

If you examine the quality of care across a healthcare system, like millions of other data sets, it naturally forms a bell curve. There is a portion of care on the far left - which is admittedly a small proportion - where the quality is very low. Most of the care is in the middle of the bell curve and at the top of the bell curve is the standard quality for that country. After that, on the far right of the bell curve, quality is exceedingly good, but again is only a small volume of care.

Once you have this “state of the nation” overview at a population and provider level, you can then start to understand and determine what works.

Then we can then start engaging with the people who manage the healthcare system as a whole and help them understand where the data is showing they can improve - working through all the different elements of a care pathway and agreeing on the optimum for them, so that they can develop the best-fit care system for their population.

This data set is ultimately working towards effectively chopping off the far-left hand side of the curve and moving the bell to the right. In doing so, the poor, low-quality care provision is removed from the system altogether, and the average standard of care drastically improves.

Understanding the natural differences in care systems 

Naturally, standards of care will vary, because individuals and care systems are all inherently different. Aiming to hold the entire global health system to the exact same standard, providing the same treatments for the same conditions is a pointless – and frankly ludicrous – notion.

But a more reasonable goal to move towards is ensuring that the “standard of care” is the absolute baseline for how a certain condition is managed.

So if a system needs to deviate from that baseline, they can justify if and why they should, through analysing the data.

So if ‘hospital A’ can adhere to the baseline standard of care for 90 per cent of cases, and ‘hospital B’ can only adhere to the standard for 20 per cent of cases, there needs to be a conversation to understand why this is occurring.

Sometimes organisations use different populations and case mixes as an excuse for poor care standards. But again, with the availability of data, they can check this. When they examine the macro-level statistics, they can determine whether their population really is different, or the way they present is different. On the other side, a provider could see that they are the same but their decisions and processes are different.

The next step is developing the ability to mandate. Through looking at the different organisations, and creating a baseline for what the standard of care should be when a provider needs to deviate from the baseline, you can determine if a clinic or provider needs to follow the standard – unless it meets a certain number of conditions.

Ultimately, it’s about striking the balance between standardising care, while allowing clinicians the freedom they need to treat the patient in front of them.

Not just quick change, but long-term development

This piece of work exists not only for quickly fixing and regulating standards of care as a one-off. Within this system, a relatively small – no larger than perhaps 20 different identifiers - are the key metrics that tell us about that state of care standards and speciality in that state.

By examining these KPIs, you can continually monitor clinical safety and standards depending on the data processing within that state. If you refresh them every month, every quarter or every year, you can use them to monitor how the speciality is changing over time and developing with current trends. Through this, you can ensure a constantly evolving, personalised, and ‘high-quality’ standard of care for any healthcare system.


e: simon.swift@methods.co.uk

https://methodsanalytics.co.uk/


#whjmethods #whjdata #whjfeature #whjmedtech #whjhealthsystemsdevelopment #whjsimonswift