I went armed with printouts of the traces and showed them to my GP who said, “looks like you did your own sleep study”. They then referred me on to a specialist to look at the Apnea events that the data seemed to be showing. When I arrived at the specialist armed with my printouts, he didn’t seem at all interested and we went back to basics, the upshot of which was that I’d need a proper sleep study. Was mine not accurate enough?
What notice should healthcare practitioners take of data that is not commissioned/collected in a managed medical process?
This experience made reflect on a few issues and difficulties that the stakeholders in the process in this case patient, GP and specialist have when dealing with the the explosion of medical and well-being data that is easily collected by patients investigating their own symptoms. What notice should healthcare practitioners take of data that is not commissioned/collected in a managed medical process? Certainly for the GP there is low risk in looking at the data and making a referral. The risks lie mainly in wasted time and resources arising from spurious referrals driven by dodgy data.
From the specialist’s point of view, they are taking more responsibility for decisions on care and understandably don’t wish to take unnecessary risks. And there are certainly risks in taking the data presented by a member of the public at face value. What is the quality of the data? is it medical grade measurement? is the data even from the person presenting it as their own?
At present, in the majority of cases the patient will not present “self-generated” data. For the few times it does happen it may be easiest to ignore it. But the wellness and health data that people increasingly gather from their wearable devices and are able to provide clearly does have value. We need to find a way for healthcare professionals to tap into this data, alongside other information sources, to support their decision making processes. The goal would be faster decision making, better quality outcomes and (of course) with no increase in workload.
Perhaps a key enabler will be better automatically generated metadata to describe the provenance of the self-generated data and provide some assurance about it. This could include the data’s source, quality factors about the data together with standards for data exchange. Scoring or assessing the relevance of the data in this way will provide the practitioner with the evidence to assess risks and either ignore or act upon the information received with confidence.