As the processing power and connectivity of mobile devices have increased and the number of on-board sensors has multiplied, there has been increasing interest in the role that these advancements can play in a Learning Healthcare System. At the same time, an increasing proportion of the population are engaging with social media and other online services within the public and private sector.
These developments are driving an exponential growth in the amount of data available about individuals. There is some hope that personalisation will offer a way to improve quality and outcomes within medicine and that data generated outside the healthcare system will enable that process (Loder 2015).
New wearables technologies and the emergence of health kits, such as the Apple HealthKit and Google Fit, could produce a change in how people interact with their own data. Mobile apps allow patients to be a lot more connected to their information (Foley and Fairmichael 2015). This can then be integrated with the EHR (EMIS 2014)
There is evidence that some patients are beginning to coordinate their own data for self-care purposes and there is a huge range of health related apps already available (Dunbar-Rees 2015). Patients with chronic or serious conditions have displayed a much higher use rate of apps designed for them than has been enjoyed by the general health apps. Their primary motivation may not be to become research subjects, but that may be a secondary benefit (Loder 2015).
PatientsLikeMe (www.patientslikeme.com) is a web platform that allows patients to share their health related experiences in a highly structured format that can be used by other patients and by researchers. With over 300,000 members, it has shown that there is an appetite for this sort of service.
Many more people use general social media platforms and they post a lot of information that could be highly predictive of how well they may fare with regard to their health. Some believe that this source of information could rival genomics in terms of the value that it could provide to healthcare. Big data methodologies would need to be developed further to process such vast quantities of data, but such techniques may become feasible in the next 5-10 years (Bates 2015).
Monitoring technologies that record and analyse vital signs may see significant improvement and advances in analytics could help to reduce false positive alarms within such monitoring systems (Bates 2015).
Bringing this data into the healthcare system will be a challenge and might have implications for the doctor/patient relationship. In some instances doctors may be keen to engage with this data. In general practice, for example, doctors may only see a patient for 10 minutes once every 6 months. In this time, it is difficult to collect all of the relevant information and patients may have forgotten important issues. It may be helpful if the patient can bring a lot of data to the consultation, but only if it can be visualised quickly and in a meaningful way by the GP (Foley and Fairmichael 2015, Loder 2015).
Analysing and interpreting the data from these new potential sources may produce associations that were previously unknown, but it is not clear how researchers would get access to this data and whether it could be linked to data from within the healthcare system. In England, HSCIC could potentially provide the context, the mechanics and the safe haven where this work could be done (Manning 2015).
Ultimately, the use of data generated outside of the healthcare system will depend on its usefulness and validity (Loder 2015). This has not yet been demonstrated on a large scale and there remains significant scepticism about how useful this area will ultimately become. There are many examples of health apps that are downloaded but rarely used (Loder 2015). There are also big questions around who owns this data and how it will be used by the organisations who facilitate its collection (Manning 2015). With regard to data from social media, there needs to be a societal discussion as to what data it is acceptable to mine (Bates 2015).
There is a concern that the patients with the highest needs and highest costs in the healthcare system are also less likely to own smart phones or engage with online services (Bates 2015). Patients with the most complex needs often have a combination of physical and mental ill health coupled with sensory impairment, drug and alcohol problems, may not speak the local language or are homeless. Even if smart phones and Internet access became ubiquitous, it is not clear how such technologies would help these patient groups.
Often the most disruptive technology will come from outside of the healthcare market, but will be adapted into it (Bates 2015) . Developments in other industries offer interesting examples. In the US, Mint.com aggregates financial data at the individual level. The user provides the service with log in details for their bank accounts and credit cards. The website then analyses all of the user’s financial data and presents it back through a suite of visualisation tools, making recommendations on how they could save money by shopping elsewhere or switching to a different providers.
As patients begin to access their health records online and generate additional health related data, such a system might help them to make sense of it and might even make suggestions about treatment options to be discussed with their GP.
Mr John Loder Interview
Mr Kingsley Manning interview
Dr Rupert Dunbar-Rees Interview
IBM Watson Site Visit
Dr David W Bates Interview
Dr Know: A Knowledge Commons in Health