ehealth Week 2016 took place in Amsterdam earlier this month and brought together the European eHealth community to focus on three main themes: Empowering People; Trust and Standards; and Social Innovation and Transition.
As part of the event, EHTEL ran a thought leadership debate titled, ‘More data, better care?’, and I was given the opportunity to contribute to the discussion as a panelist. My initial response to the question was simple - yes, you do get better care with more data – but it comes with a caveat.
Collecting data alone doesn’t result in better care, it needs to be securely stored, analysed, easily accessed with fast retrieval for those authorised, and presented in an easy-to-use format to support decision-making. Then more data starts to add value.
To support my claim I shared some stories that demonstrate how more data does, or potentially could, help to deliver better care - two of which I’ll outline below.
Identifying the risk of developing heart failure
First I referred to a trial carried out at Penn Medicine hospital in the US. Here, the team are able to identify patients on admission that are at risk of developing heart failure – irrespective of what condition they originally came in with.
This is based on Intel’s Trusted Analytics Platform (TAP) which enables the analysis of all sorts of data including; patients’ postcodes, electronic medical records and historic healthcare utilization. Armed with this insight, doctors are able to investigate the risk to the patient, decide on a treatment plan and effectively pre-empt the condition from worsening – thereby minimising disruption to the patient and reducing the strain on emergency admissions.
Predicting the likelihood of developing sepsis
Another example that I highlighted was the effective use of data to improve the identification of sepsis, an area of care that NCEPOD have concluded that, “Early recognition, better documentation and prompt treatment of sepsis would all lead to improved care for this group of patients.”
In relation to this, Cerner has produced a predictive model which constantly monitors information from lab results, vital signs and prescribed medication to enable at-risk patients to be bought to the attention of the clinical team as part of their daily routine. For me, it’s really interesting to see electronic medical vendors, like Cerner, using big data analytics to develop applications such as this and is a great example of using data effectively to enable early-action.
Risks of data sharing
To give balance to the success stories, I expressed caution too. For example, we need to ensure we get value out of the data without causing risk to individuals – in particular ensuring their data doesn’t fall into the wrong hands. At Intel, we have developed a security breach assessment model that helps healthcare organisations plan for better data security.
The industry is also learning how to make sense of the large volumes of data coming from many disparate sources, developing the tools and skills over time to analyse, interpret and communicate meaning.
I concluded by saying that I think the reality of where we are in the ‘more data, better care?’ story is potentially yes, you do get better care with more data, but there’s still a lot of work to be done to actually realise the benefits.
Getting thousands of patients delivering information for research and reports to clinicians so they can make decisions on treatment and measure the effects of medication is still a while off – but it’s a start on that journey.