Will Healthcare become like the Aviation Industry?

Could the delivery of General Practice (GP) Healthcare take some cues from the Aviation Industry?

I have an interest in both and it seems to me that healthcare delivery of the future is going to have a lot in common with the modern aviation industry and perhaps we should start structuring our approach to healthcare to recognise and take advantage of this.  You might also argue that physician training could draw upon similar training as pilots in managing information and human factors etc.

The idea that we will all be generating some health related telemetry is a reality.  Many of us our doing this already through the use of personal monitoring devices and apps such as "Health" on the iOS platform. Many of us also have access to the internet meaning that we can exchange videos/photos and vital signs quite readily. We also frequently diagnose/misdiagnose ourselves via the common internet search engines. What is absent is any professional input and oversight of this data.  

The next logical step is to start associating that telemetry along with some related metrics such as our genetic makeup, lifestyle and occupational inputs along with the more transient inputs (illnesses, diet, accidents, interventions etc)  into a profile that is then managed and added to centrally in much the same way as regional air traffic controllers (ATC) take the details of an aircraft, its flight plan and safely transit it through its area of coverage.  Analysis of health analytics in the circumstances above would have to be largely automated with the exceptions being dealt with as they present either through artificial intelligence taking actions if the intervention is fairly clear cut or engaging human input if appropriate and necessary to do so.    Like health maintenance, a flight plan is a unique, personal thing.  Yes it has a lot in common with the flight plans of other aircraft and routes yet it is still unique to a journey.  It is also dynamic.  A pilot has to adjust for variable conditions, weather, passenger illness etc.  I can see humans having the equivalent of a health flight plan where the variables are plugged in and a level of automation oversees the journey automatically through the use of telemetry.  What we are currently missing is the ATC component.  There are the beginnings of this thinking focused on the delivery of specialist conditions but as a general concept its still does not seem to be on the agenda.The idea of a digital health record is gaining traction for individuals but it is an opportunity lost until we get realtime analysis and intervention based  upon information contained within.  This maybe something that becomes common to the general public or a service funded by the individual or an employer for its employees.  Maybe it requires such levels of computing that it is not practical to apply this to the masses?

Health related technology and research is producing more and more data.  The individuals "health flight plan" is a life time journey.  Healthcare currently seems to assume that our GP's will somehow continue to manage our health.  This is not practical, affordable or deliverable in the current structure.  Just as pilots cooperate with ATC and jointly take charge of a largely automated flight, the potential exists for a physician led equivalent of an Air Traffic Control Centre to provide oversight to managing health flight plans for thousands if not millions of people simultaneously.  Artificial Intelligence will have to replace the GP personal interventions for the more mundane tasks such as advising on Cholesterol management and the likes and humans will have to accept that the AI has taken this responsibility on.  If they wish to engage a human physician this should come at a premium in these more mundane transactions.

Just as AI systems are beating chess grand masters it is logical to assume that an AI GP system can absorb and process a lot more information about an individual in a more timely manner.  AI can also benefit from the application of aligning symptoms with favourable outcomes based upon treatments for those patients that have a similar health flight plan and who have already undergone therapy.  AI can arguably be more aware and learn much quicker than an individual health practitioner and therefore future physicians may have more of a role maintaining these AI systems than front line healthcare delivery.  GP's and other healthcare workers are also more likely to engage with humans via digital means.  It can be more efficient and achieve the same outcomes.  Physical tasks such as blood screening, vaccinations etc will still need to be done in person but not necessarily by GP's who are the most expensive and valued asset in the delivery cycle.  It would be interesting for someone to research the level of confidence/satisfaction the public would have in a AI assisted GP provider vs the traditional provider.

As we learn more about specific conditions and treatments, GP's are expected to know more than is practical about all of the health conditions and will require some form of AI decision support capability. To some degree this is catered for by GP's with differing special interests collaborating in a common practice.  Treating physicians will also increasingly become more specialised. Specialisation leads to dilution of availability in geographies and therefore access to these specialists requires two things 1) an appropriate diagnosis and 2) access to the specialist resources which for rare conditions could be based internationally.  The future GP service will have to become adept at diagnosing and shepherding patients/clients to the most appropriate treatment resource or scientifically and ethically reaching a decision to move to a palliative response.  As a public we should no longer expect all treatments to be available on our doorstep. It is neither practical nor beneficial for a sustainable health service.  In these more specific circumstances it is arguably better to bring the patient to the specialist centre for specialist treatment where there is a greater chance of a positive outcome.

In this day and age, I and thousands of others happily, albeit subconsciously, accept that technology has consulted the weather, traffic, passenger and cargo metrics and determined the optimal route to safely and efficiently cross the Atlantic on an aircraft.  I accept the aircraft is largely flying itself, constantly adjusted elevators, ailerons, engine management to track a course.  It will avoid other aircraft and terrain and the pilot is largely providing oversight and approving what the systems are recommending, I expect the Master Alarm to sound if a fault is detected and the pilot to act in a timely manner to intervene and manage  the problem to a successful conclusion.  ATC assists in transit through geographies and provides oversight, though increasingly this will become automated as capacity exceeds what humans can safely deal with. They work with the pilots and the aircraft owners to manage issues in realtime as they occur.   Airport ATC provide specialist take off and landing instructions.  It is no coincidence that air travel is the safest way to travel.  The synergies are clear for me between healthcare delivery and delivery of thousands of safe flights each day across the globe.  Maybe we need to be a bit more proactive in embracing technology in the healthcare world?

Anthony Dooher

Healthcare Innovation Expo Manchester highlights

My reflections on Day 1 #Expo15NHS, a really interesting and well attended NHS focused conference.

Rt Hon Jeremy Hunt, Health Minister reiterated and further expanded upon the vision of a paperless NHS by 2018. New expectations for patients and care providers to be able to not only visualise their full record (rather than just a summary) but also contribute to the record were set.  It is anticipated that the new interoperability road map documentation to be published tomorrow by NHS England will provide tools and guidance on how this might be achieved.  Many of the recommendations are expected to be based around the vanguard projects that have been initiated and are providing valuable insight to how interoperability and a regional health community might work. 

It is largely expected that health apps based around given diseases or care pathways will evolve. Thus allowing patients to maintain their health in the community setting but with clinical oversight and technical algorithms essentially providing a safe guard to detect and facilitate actioning of a deteriorating condition proactively rather than reactively. This proactive approach lends itself to community based care being provided by such professionals as pharmacists, GP's, community workers etc which should be better for the patient and cheaper to provide.   The ability to provide this care by such professionals will be possible as a consequence of being able to access the record universally.  Intelligence in the platform is expected to not only facilitate interoperability but also analyse and action data.  An example in a social care setting is  a person attending multiple times to various health providers for injuries which individually would not cause concern but when considered collectively might suggest an underlying domestic abuse problem. 

Access to such records will be facilitated by open APIs which also received a lot of coverage at this event and appears to be the defacto standard for interoperability going forward.  Users should be able to access the full record from the native application that they are using via the APIs as part of a connected interoperability community.  So a pharmacist can use the pharmacy system for example to access the record.  

Discovery of record content in an aggregated manner will be a challenge so NHS England are looking into a national index type solution. Where regional solutions evolve it is anticipated that these will be incorporated into a national solution. 

Open Source is also something that is gaining traction.  Coding for health invited vendors to publish their APIs to the platform to enable developers to code against their APIs.  It will be interesting to see if vendors take up this invitation.  

Clinical Digital Maturity Index (CDMI) scores are also likely to be a measured and rewarded KPI much like CQUINS that exist today.  In addition the associated outcomes linked to the CDMI scores will be measured to ensure the investment is delivering as expected.

All in all a very well attended event with plenty of energy and thought leadership from the very top that seems to resonate with many of the stakeholders attending. 

It will be interesting to see how quickly and comprehensively this vision is turned into a reality. 

EPR vs Content Repository and Apps - Is it a losing battle?

EPR's are massive undertakings to procure, implement and maintain.  They are also hugely expensive and demanding on valuable clinical front line staff to implement.  These statements are well evidenced in many publications and research over the years, yet still we see hospitals, regions or even countries chasing EPR solutions, usually from a single vendor to get over interoperability fears.

But, I see a shift in thinking starting to emerge and I for one think its both exciting and beneficial to all of the stakeholders.  The move is towards a structured repository of content which would typically include patients historical and future notes, medications, alerts and care plans etc.  Standards such as IHE XDS will facilitate development, procurement and implementation of best of breed departmental systems that can function in isolation but publish or access beneficial content such as results to/from the XDS Registry and Repository if necessary.  This is not so new now, but what is interesting is the rapid pace in which industry and clinical users are embracing mobile device based "apps" to access or record new content in real time.  This can include patient vital sign observations through to personal monitoring devices that populate an app managed by the patient in the community setting and uploads data to the central repository.

Whats really exciting about this is that clinicians and industry from start-up innovators through to more seasoned apps houses can now participate in contributing to the Health IT sector with a realistic chance of being able to sell and implement the solution, having being able to focus efforts towards addressing a specific demand in an agile and affordable manner.  The nature of the app means that implementing it, maintaining it and ultimately replacing it in the long term is a lot less disruptive to an organisation and the end to end costs are likely to be much less than having to add or modify modules to an EPR and testing/validating all of the interdependencies.  Adoption of the app is also more likely given its focus on a task or function and the familiarity that most health practitioners and patients have with this technology today.  The training and change management costs are also likely to be much less than inclusion of similar functionality in a much larger traditional EPR type solution.

If companies/developers and health providers can abide by some of the logical demands as defined by the NHS ISB for example regarding consistent location and format of content such as patient banner and also nail down consistent use of descriptive terms and nomenclature then there are few reasons why the use of apps in mainstream healthcare won't continue to flourish and start to threaten the big EPR players in the market.  Clearly this is something that Apple™ have recognised with the release of the Healthkit and enhanced capabilities of their OS.

Its great to see the innovation, and larger vendors must be mindful that customers want to see innovation and solutions that can be developed and implemented in an affordable, agile manner. Standards based exchange of content via Web API interfaces is the future and its great to see some of the companies starting to embrace this reality.

Its my belief that use of Apps that fulfill a specific requirement for a target audience will become mainstream, underpinned by standards based interoperability and a repository(s) of structured content facilitating the various apps based access points to that content.   The idea of a monolithic EPR solution seems out dated already and I think its days are numbered as is the appetite for them; financially, technically and operationally.

What do Personal Heath Records (PHR's) and George Orwells 1984 have in common?

In the past 6 months I have gained further first hand experience that supports some of the discussion points raised in my previous post around Personal Health Record (PHR) management.

I've seen an interesting posting by Dr Mohammad Al-Ubaydli where he also shares the view of the public maintaining their own records http://www.patientsknowbest.com/videos.html and has gone as far as setting up his own company which has gained some traction in the UK and the USA.  I don't accept the view that a patient has or can be asked to have an implied obligation to question the doctor about the contents of the record and in so doing the doctor is relinquished of some of the responsibility in determining that the diagnostics are correctly interpreted and acted upon in their entirety which includes incidental findings. I do however accept that incidental findings are often identified but not acted upon as the requesting physician is focused on interpretation of contextually based findings to the physicians specialty.  That said, I have also seen the development of safeguards to such things as incidental findings where a radiologist can alert a cancer control office of a likely incidental oncology finding as part of the reporting process for a non-oncology referral, enabling them to ensure the followup is provided.  Applications such as RadAlert by Rivendale Systems (http://rivendale.ie/index.php/healthcare/radalert)  have been successfully deployed in large academic hospitals such as Cork University Hospital in Ireland to address these findings with published results evidencing the success of such solutions at RSNA 2011.

NHS England is further building upon this demand by trying to enable patients to access their GP records on line by 2015 http://www.england.nhs.uk/ourwork/pe/patient-online/  but how prepared the GP practices are for this I don't know.  How complete the record is may also be questionable as results and content can be provided to the GP practice system in various forms, i.e. email content, email attachments, HL7 interfaces from distribution systems/medical gateways and of course paper based information posted or faxed to a practice. The dependency from a patient perspective is that the GP system is correctly maintained, fed information in a complete and timely manner and accessible by the patient and by persons that the patient wishes to grant consent to.

In my brief trawl of the web it seems that adoption rates of PHR are low, and this I think is understandable as fortunately the vast majority of the population are well.  Interactions with medical care usually occur in an acute setting and previous medical history for the most part does not influence the care given in the acute context.  Acute care is often provided and concluded in a short period of time and can be largely summed up as an inconsequential event in the overall health for most acute encounters in most peoples eyes.  The effort vs benefit of adding all the care provided in an acute setting to a PHR does not stack up for most members of the public.   For more mid term care that arises from an Acute event such as RTA with multiple complex fractures and injuries that might involve months of complex surgeries and rehabilitation in specialist and community settings then you can more readily anticipate that the patient would benefit from a PHR that they can share with the various stakeholders throughout the care period.  There are also likely financial benefits for both the care providers and the patient as the costs associated with such an encounter are likely to be high and therefore reimbursement to the care provider or to the patient if funding the care via insurance becomes more important.  Access to the medical records also means a more speedy encounter as clinicians in the care pathway can be more readily prepared in advance and expensive and harmful diagnostics do not need to be repeated as a consequence of unavailability.

The increasing number of screening programmes that patient's participate in would in my opinion benefit from a PHR as the screening capability is increasingly more likely to be provided as an outsourced service and therefore the screening encounters will become isolated encounters, but with clinical dependencies to access previous findings for comparison (as in Breast Screening). Government agencies will increasingly run public procurement's for such services (as well as resulting elective interventions for positive findings) on a regular basis and adopting a PHR can avoid costly data migrations whilst allowing a competitive value and technology driven market to flourish. People/ Patients are increasingly becoming more elderly whilst maintaining health to a greater age and therefore remain mobile for a longer period of time.  Increasing ages for retirement will mean that these elderly people will move around and will need to maintain the continuum of screening independent of geography and provider.  It is also likely that patients will want to "shop around" for care in the future based upon quality and financial indicators and metrics.

The same can be said of chronic disease management and I increasingly anticipate with the evolution of devices located in peoples homes, about their person and even in time internally situated to be remotely monitored and even care to be in someways automated.  An example is insulin dependent diabetic patients that could wear a device that monitors blood sugar and automatically administer the appropriate dose of insulin internally from an embedded insulin reservoir  (at least conceptually as im not sure of the shelf life of insulin but you get the gist) and there are similar parallels already used in cardiac pacemakers where a shock can be delivered automatically.  The device can report the event to an external system and can be managed remotely.   It is therefore probable that in years to come we will see call centres and monitoring companies that will manage the various streams of data coming from potential PHR connected devices that run silently in the background, much like we see with security cameras and call centres protecting properties today.

In summary I believe that PHR's will become mandatory in the future and just as we get issued a social security number today I can see the day when we will be issued with a national/global PHR account that is completely web based, which will communicate with other government agencies via a Health Exchange Portal and which will provide the basis for much more predictable healthcare expenditure based upon real data, an aging population and research based upon large population numbers.  Clinical trials, pattern recognition and data analytic's will increasingly determine budgets,  treatment and preventative healthcare strategy in the future. It may also controversially provide justification for treatment denial or withdrawal (but that's a whole other can of worms to explore separately)

The effort and finance that is being invested in making hospital based systems interact in a local, regional or national context is therefore wasted in my opinion and if the same effort and revenue could be steered to a mandatory PHR for citizens instead then the needs of the future would have a much higher likelihood of being addressed in a pragmatic and affordable manner.  There would also be spin off benefits that would allow other government departments to operate in a more streamlined manner.

George Orwell's 1984 looks increasingly similar to the healthcare and associated governmental agencies landscape in my opinion based upon what I have laid out above and see of the evolving health related technology landscape and Im not sure everyone has quite considered the implications of adopting and implementing all of this technology for the future generations and what this might actually mean when you apply for health insurance, a mortgage, a job etc.....

On that sobering thought I would be interested to hear from people that have used a PHR (a product or a home made solution such as Google Plus account) and their opinions/experiences.  I would also be interested to hear from thought leaders in this area especially with a background in related technology/services provision, the ethics of data interoperability and what this might mean for the future generations and also government strategists on whether they agree or disagree with my thoughts.

Google Plus as my EPR?

So this is my first post to my own blog and this is really a space to share my thoughts and ideas and see if anyone else has some traction or view on them.

My thoughts this morning turn towards maintaining and accessing a patients record, more particularly my own health record.  There are many solutions/apps now to capture all sorts of  information from a smart phone or tablet and also very mature messaging standards such as HL7.   There are also many technologies that allow information to be stored and shared on line and increasingly with more capability around access control.

So, am I being simplistic when I explore some of these technologies (as I have started to do more recently) and join these dots together .........the first dot if you will is the recognition that in most western countries we can access the internet almost anywhere.  I can have an account such as google plus, create some circles which effectively manages my access control and consent issues and upload data from my health practioner to my google plus account and share it as I see appropriate.  Ok it may benefit from some dedicated indexing within the google space which I am sure google could manage.  I can use the camera to capture data but I think health providers should record my google plus account information in their PAS systems and upload either my content or hyperlinks to my content which they maintain on my behalf.  The fact is that most data is digital or can be made digital.  Why not make it bi-directional so that I can upload content and get an opinion or even refer my record to get a second opinion. I could also save trips to the hospital to get my wound checked by uploading images from the comfort of my own home and steer them towards my care provider even using notifications to alert people of their availability.  It should even be possible to start utilising home based diagnostic or monitoring equipment docked or worn about the person and wirelessly upload this data periodically or in real time to my on line health engine which in the future has the ability to create workflows automatically as abnormal results/observations are detected.  You can see the benefits for those with disabilities, of chronic ill health, for the elderly or for those who want to keep close tabs on their health maybe as part of a training plan for a sporting event like a marathon.

So let's look at the benefits.  Well most headaches around data regard access and consent and to my mind we are building increasingly more difficult hurdles to jump as those guardians responsible for data protection feel obliged to protect data almost now at the detriment of being able to usefully access it when required .  I travel a lot and God forbid it is possible that I may need to seek medical attention anywhere in the world.  I also want my health record to be longitudinal and reflect my history from birth to death, so yes I am suggesting that parents create accounts for their children, even if they start with capturing jabs etc as it really helps in later life when the child becomes the adult and doesn't have to hunt around the attic looking for an inoculation card from 20 years ago.  It would also be useful to capture boosters, travel meds, tetanus etc.  I'm sure you get the drift and can also then link this record to driver license maintenance, employment screening for occupational health such as eye sight tests.   When you actually put your mind to it the benefits of linking data availability to consumption can have loads of benefits.   Just one other example that springs to mind.  I had an uncle with diabetes who wore a bracelet indicating as much. If you replaced this bracelet with a physical digital key that a paramedic can use to access this patients record on line in the shopping centre should they become disorientated due to  hypoglycaemia then much more specific and relevant care could be provided much earlier which is vital in an acute siutation.  Sadly as my uncle become more elderly and living alone he became less able to manage his blood sugars and ultimately died as a consequence of entering a diabetic coma.  If technology such as the contact lens that will check blood sugar levels automatically or other similar monitoring technology was available and more widely used then people like my uncle as well as those that cared for him would undoubtedly have benefitted.

It is also demonstrable that health providers will benefit. You could potentially shift a lot of the ownership of records to the patient along with the responsibility.  Most patients will ensure they keep this well maintained.  Billing and compensation could also be speeded up as the data along with potential e-claim forms could be completed and submitted on line resulting in faster compensation.

I would also argue that google is already part of many peoples health plan, how many of us google an ailment and self treat as a consequence .... I don't have the data other than anecdotal through conversation but since becoming a parent I can testify that googling a symptom and acting accordingly not only saves me a fortune in GP visits but no doubt saves the state as well.  Don't get me wrong, visiting a GP or A&E is always going to be the backstop should you be concerned and common sense is always required in judging when to consult a medical practitioner in person.

So, this is my brain dump on this matter for the moment at least.  I know some big companies such as Microsoft have considered Health Vault technology but my feeling is that stakeholders from within the industry and the benefactors such as those suggested above need to drive this forward as arguably a national initiative is required with some IT giants of this world such as google and let the industry players align accordingly.  Interoperability between hospital systems which themselves are maintained in isolation is not the future in my opinion!  The patient and the public need to become more responsible for their data and how it gets used in this world where increasingly the funding for care will be allocated directly to the patient or their GP and they will pick and choose where to be treated based upon availability, competence and budget.