A frustration of everyone in healthcare and technology is the unfulfilled promise of Big Data
. A study conducted by a team for NEJM Catalyst
(New England Journal of Medicine
) of 682 health care executives, clinical leaders, and clinicians indicates that at present, very few (<20 percent) believe that their healthcare organizations extremely or very effectively use data for direct patient care; 40 percent believe it is not very effective or not at all effective.
The hope comes in a trend over the next five years (NJEM chart at left above, click to enlarge). Presently, the most useful sources of data are clinical (95 percent), cost (56 percent), and claims (56 percent). In five years, they project that the top four will be clinical (82 percent) and cost (58 percent) joined by patient-generated and genomic data (both at 40 percent). How that patient-generated data will be compiled to be useful is not described, but the hope is that “With patient-generated data and genomic data, we will be able to create true “n of 1” medicine with options specific to each patient’s needs, giving a boost to priorities such as care coordination and improved clinical decision support.”
A possible roadblock is the lack of interoperability of EHRs. Less than 10 years ago, the EHR was touted as The Solution to patient records and a repository of Everything. 51 percent indicate that interoperability is weak. One-third believe that ease of use and training for EHRs are also weak.
Other findings indicated strong support for greater patient access to personal medical records (93 percent), fee/price information for comparison shopping (80 percent), and outcomes information listed by hospital (73 percent)–but not by doctor (55 percent).
The full report is available for download at the NEJM Catalyst link here. Also Mobihealthnews.
The Federal Communications Commission (FCC) has been investigating the relationship between broadband and health in the US through their Connect2Health Task Force and this week it has released an online tool “Mapping Broadband Health in America”.
It is an interactive map that allows users to visualise, overlay and analyse broadband and health data at the national, state and county levels.
This tool allows easy access to existing health and broadband access data to anyone who wants to look at the possible influence of broadband access on health over a period of time or to identify gaps which may provide opportunities to develop or expand online health services.
The interactive tool allows the user change the broadband availability measure (by say proportion of coverage or download speed for example) and select a health measure such as say obesity rate or preventable hospitalisation days and shows where the selected broadband measure is satisfied, where the selected health measure is satisfied and where both are satisfied. The types of health measures are currently limited but if users find the tool useful and feedback to the FCC there may well be further expansion.
Have a play with the map here.
Make a place in your calendar for two Royal Society of Medicine full day events coming up in May and June. Both organized by the Telemedicine and eHealth Section. Hat tip to Charlotte Cordrey, Event Team Manager, RSM
The future of medicine – the role of doctors in 2025
Thursday 19 May 2016 (Chaired by our own Editor Charles Lowe)
Big data 2016 (Clouds and the Internet of Things)
Thursday 2 June 2016
The WHO has produced an excellent report on the state of eHealth in the European region, including a review of telehealth readiness. Ericsson have produced a very interesting report confirming what I guess anyone will have realised if they’ve traveled by public transport or have children: young people downloading video content are driving a surge in data usage: there’s much detail here though. Both are well worth the read.
Mentioning Ericsson reminds that the Telegraph recently produced a summary of the 20 best-selling mobile phones of all time – takes you back, with the substantial number once produced by Nokia.
The Royal Society of Medicine has it’s fifth annual medical app conference on April 7th – numbers booked have already well exceeded last year’s sellout so they are expecting to fill this year’s much larger conference venue. The focus this year is on the many legislative, regulatory and voluntary measures being introduced that will impact medical apps – there’s still room for old favourites though, such as Richard Brady’s always-topical (more…)
Many of our recent stories have touched on ‘big (health) data’ as Achieving the Holy Grail–how it can be shared, how it can work with the Internet of Things and how poorly implemented personal health record (PHI) databases can derail national health systems (and careers) [TTA 22 Sep]. They are, after all, 1) extremely difficult to design to preserve privacy and 2) must satisfy patients’ requirements for easy use as well as privacy including opting out. But when despite all good intentions, data goes awry, the consequences can be severe.
- A daughter applies for health insurance from Aetna, and her mother’s medications, about which she had no knowledge, are attributed to her. How? Data mining off Milliman’s IntelliScript data service which mixed up the records.
- EHR exchange can spread errors such as a dropped critical health or medication record. One led to the death of an 84 year old woman. VA also had a problem with its EHR (not cited but likely VistA) slotting medication histories into the wrong patients’ files. An Australian hospital mixed up discharge files in electronically sending them to doctors. The more records are exchanged, the more possibility there is for propagation of errors.
- More information is shared with third-party suppliers; survey companies are increasingly tapping into these databases to send annoying, potentially privacy-invading treatment questionnaires to individuals.
Bloomberg Business’ conclusion is that this could be a problem, but much beyond the tut-tutting doesn’t get into solutions. The Pitfalls of Health-Care Companies’ Addiction to Big Data
Did you watch the Panorama programme yesterday on BBC (only available in the UK, I understand)? Subtitled “Could a Robot do my Job?” reporter Rohan Silva was looking at the impact of Artificial Intelligence (AI) on the workplace and jobs, primarily in the UK.
The last section of the programme was on a data analysis system at a Boston hospital (Beth Israel Deaconess Memorial Center). The reporter mentioned they use an “artifical intelligence supercomputer” (!) in their emergency department that can “forecast if you’ll die in the next 30 days”. Well, not quite, but, “forecast the probability of a patient dying with almost 96% confidence” according to the very enthusiastic doctor (and the only one featured in the programme) at the hospital. Not sure if that is all PR or verified independently.
I was very impressed when it was mentioned that the computer had 30 years of data from over 250,000 patients,so it could recognise rare deceases quicker than a doctor. After all my navigator can find me a route a 100 times faster than I can, so why not.
But then I got thinking. 30 years ago they didn’t collect patient’s blood oxygen level and blood pressure every 3 minutes like they are doing now. This was an emergency department, not the obvious place for lots of people with a rare diseases to turn up. How many rare diseases had this system diagnosed so far? So there was a fair bit of mirrors and smoke to make it look far better than it really is I think. In fact, I think the Boston system is actually just good example of what is called Big Data at work.
This tendency to exaggerate was true of the rest of the programme too which can be fairly described as sensational rather than educational.
No doubt the publicity will help the hospital. I see that the story about the dying prediction appears on many newspaper websites right now with headlines like “the supercomputer that can predict when you’ll die”!
Thanks Donna for telling me about the programme.
Another sign that mHealth is now in our rear view mirrors [TTA 24 July] is that one of the main conferences on the US and international conference calendar is changing its name. Since 2009, the mHealth Summit has closed the year. Its organizing groups have changed and it’s gone international to Europe (the recent summit in Riga). Now it has been renamed (though not on the website yet) the HIMSS Connected Health Conference-–an umbrella event comprising the mHealth Summit (including the Global mHealth Forum), and two new conferences: the Cyber Security Summit and Population Health Summit.
The shift in the industry and new concerns are clearly reflected in this reorganization. Transitions were visible last year to this Editor in covering the sessions, speaking with exhibitors and attendees. It’s not about the tech anymore, but how it fits into care models, saves money/avoids costs, improves care, improves the experience–all population health metrics–and fits with other technology and analytics. (It’s also how it fits into government payment models, an endlessly changing equation.) What is surprising is the lifting of cybersecurity to equal status, given the Hackers’ Holiday that healthcare is now (see TTA here). (Also this Editor notes that last year’s Big Buzzwords, Big Data and Analytics, has faded into where it should be–into facilitating population health and we should expect, inform data security. We also note that HIMSS has stepped forward as the organizer. HIMSS release Telehealth & Telecare Aware has been a media partner of the mHealth Summit for most years since 2009.
At the end of last week, the EU Data Protection Supervisor (EDPS) published an excellent document entitled Mobile Health – Reconciling technological innovation with data protection. To quote the press release:
Failure to deploy data protection safeguards will result in a critical loss of individual trust, leading to fewer opportunities for public authorities and businesses, hampering the development of the health market. To foster confidence, future policies need to encourage more accountability of service providers and their associates; place respect for the choices of individuals at their core; end the indiscriminate collection of personal information and any possible discriminatory profiling; encourage privacy by design and privacy settings by default; and enhance the security of the technologies used.
The document itself contains much of interest. To this editor, who has heard many people poo-poo the importance of wellbeing data, it was good to see:
Lifestyle and well-being data will, in general, be considered health data, when they are processed in a medical context (e.g. the app is used upon advice of a patient’s doctor) or where information regarding an individual’s health may reasonably be inferred from the data (in itself, or combined with other information), especially when the purpose of the application is to monitor the health or well-being of the individual (whether in a medical context or otherwise). (Page 5)
As someone who gets concerned at turning people off sharing their health data, it was nice to see the recognition that: (more…)
Suicides by US active duty soldiers have more than doubled since 2001, according to a January Pentagon report, and current prevention programs have not been that effective in reducing the over 200 reported suicides per year. Enter a huge database program called STARRS–Army Study to Assess Risk and Resilience in Service–to identify risk factors for soldiers’ mental health. The US Army not only likes acronyms, but also never does anything small–a five-year, $65 million program analyzing 1.1 billion data records from 1.6 million soldiers drawn from 39 Army and Defense Department databases. Researchers are looking at tens of thousands of neuro-cognitive assessments, 43,000 blood samples, more than 100,000 surveys, hospital records, criminal records, previous risk studies, family and job histories plus combat logs. The study, also using resources from the National Institute of Mental Health, the University of Michigan and other educational institutions, will conclude this June–and researchers are now wrestling with the privacy and moral consequences of responsibly using this data for health and in leadership. NextGov
The following is a brief summary of a joint Royal Society of Medicine/Institute of Engineering & Technology event held at the Academy of Medical Sciences on 6th May. The event was organised, extremely professionally, by the IET events team. The last ticket was sold half an hour before the start, so it was a genuine sell-out.
The speakers for the event were jointly chosen by this editor and by Prof Bill Nailon, who leads the Radiotherapy Physics, Image Analysis and Cancer Informatics Group at the Department of Oncology Physics, Edinburgh and is also a practising radiological consultant. As more of those invited by Prof Nailon were available than those invited by this editor, the day naturally ended up with a strong focus on advances in the many aspects of radiology as applied to imaging & treating cancer, as a surrogate for the wider examination of how medicine is changing.
The event began with a talk by Prof Ian Kunkler, Consultant Clinical Oncologist & Professor in Clinical Oncology at the Edinburgh Cancer research Centre. Prof Kunkler began by evidencing his statement that radiotherapy delivers a 50% reduction in breast cancer reappearance, compared with surgery alone. He stressed the importance of careful targeting of tumours with radiotherapy – not an easy task, especially if the patient is unavoidably moving (eg breathing) – Cyberknife enables much more precise targeting of tumours as it compensates for such movement. Apparently studies have shown that 55% of cancer patients will require radiotherapy at some point in their illness.
This was followed by Prof Joachim Gross, Chair of Systems Neuroscience, Acting Director of the Centre for Cognitive Neuroimaging & Wellcome Trust Senior Investigator, University of Glasgow, talking about magnetoencephalopathy (MEG), which enables excellent spatial & temporal resolution of the brain. However it currently uses superconducting magnets that in turn require liquid helium, so is very expensive to run. He then showed an atomic magnetometer which apparently is developing fast and will be a much cheaper alternative to MEG – he expects people will be able to wear sensors embedded in a cap soon. He then went on to show truly excellent graphics on decoding brain signals with incredible precision; he explained that the 2025 challenge is understanding how the different brain areas interact. Finally he described neurostimulation, using an alternating magnetic field with the same frequency as brain waves to change behaviour; whence the emergence of neuromodulation as a new therapy. Both exciting, and just a little scary.
Dr David Clifton, Lecturer, Dept of Engineering Science & Computational Informatics Group, University of Oxford, followed with a talk on real-time patient monitoring. He began by explaining the challenges that clinicians face with this wall of patient data coming towards them: only “big data in healthcare” enables all the data generated by patients to be analysed to identify the early warning signals that are so important to minimise death and maximise recovery. (more…)
Awash in a rising sea of data generated by devices and analytics–around treatments, population health, costs–there’s a struggle to make sense of it. We’ve noted the high value and merchandisability of 23andme
‘s genomic data (gained by individual user consent) [TTA 5 Mar
], but our healthcare institutions which should be codifying and sharing disease and treatment data, largely do not. Those with rare or ‘orphan’ diseases struggle to find information, diagnosis, fellow patients, treatments. They sometimes win breakthroughs by, believe it or not, blogging, and having their articles widely disseminated. Reasons why? According to David Shaywitz in Forbes
, they are:
- Hospitals, even research based centers, struggle to codify their genotype and phenotype data of their patients in a meaningful way that would be usable for clinical decision making. We’ve also noted (oddly not Mr Shaywitz) the long implementation process of IBM Watson cognitive processing/decision making tools in healthcare, the concentration on single diseases and their spread into other industries plus third-party integration outside of healthcare [TTA 9 Oct 14]. (more…)
The Association of British Healthcare Industries (ABHI) is looking for companies to share the British Pavilion at the CMEF trade show from 15th – 18th May 2015 in Shanghai, China. It is apparently the the leading Healthcare trade show in China and is now the largest Medical Equipment exhibition in the Asia Pacific region attracting over 60,000 visitors. Details here.
Still need to see some more predictions for 2015? – try these 12 for telecoms, which does include the odd interesting nod towards subjects we cover, including interconnection of wearables and connected homes.
Prompted by our mention of V-Connect in our review of our 2014 predictions, MD Adam Hoare has pointed out that his company also won the Medilink ‘partnership with the NHS’ award for their renal project with The Lister Hospital in Stevenage. Congratulations!
Accenture has produced an interesting (more…)
Editor Charles has treated you to a look back on his 2014 predictions, daring Editor Donna to look back on hers. Were they ‘Decidedly so’, ‘Yes’, ‘Reply hazy, try again’ or ‘My sources say no’? Read on…
On New Year’s Day 2014, it looked like “the year of reckoning for the ‘better mousetraps’”? But the reckoning wasn’t quite as dramatic as this Editor thought.
We are whipping past the 2012-13 Peak of Inflated Expectations in health tech, diving into the Trough of Disillusionment in 2014.
There surely were companies which turned up ‘Insolvent with a great idea’ in Joe Hage’s (LinkedIn’s huge Medical Devices Group) terms, but it was more a year of Big Ideas Going Sideways than Crash and Burns.
Some formerly Great Ideas may have a future, just not the one originally envisioned. (more…)
As intimated in our review of last year’s predictions, we feel little need to change course significantly, however some are now done & dusted, whereas others have a way to go. The latter include a concern about doctors, especially those in hospitals, continuing to use high-risk uncertified apps where the chance of injury or death of a patient is high if there is an error in them. Uncertified dosage calculators are considered particularly concerning.
Of necessity this is an area where clinicians are unwilling to be quoted, and meetings impose Chatham House rules. Suffice to say therefore that the point has now been well taken, and the MHRA are well aware of general concerns. Our first prediction therefore is that:
One or more Royal College/College will advise or instruct its members only to use CE-certified or otherwise risk-assessed medical apps.
The challenge here of course is that a restriction to CE-certified apps-only would be a disaster as many, if not most, apps used by clinicians do not meet the definition of a Medical Device and so could not justifiably be CE-certified. And apps are now a major source of efficiencies in hospitals – (more…)
11 November, New York
The annual event that is CES Unveiled in New York City is meant to be a nanoparticle-scale preview of International CES in Las Vegas, 6-9 January. It’s a smörgåsbord of what used to be called ‘consumer electronics’ and now is all about innovation–a taste of everything from ever-smarter video and audio to sensors, smarter homes with IoT (the cutely named Internet of Things), Big Data, robotics and (drum roll) Digital Health and the Quantified Self (QS). This Editor regrettably missed the opening briefing by Shawn DuBravac, CEA’s Chief Economist and Senior Director of Research which would likely touch on his areas of the innovation economy and disruption along with the other four 2015 trends to watch: big data analytics, immersive entertainment content, robotics and digital health. (CEA helpfully provides the 30-page white paper here.)
The exhibitors at the Metropolitan Pavilion did not fully represent the trends, however. (more…)
Another charming use for Big Bad Data. Hospitals are investigating whether available data on patients–prospective and current–on shopping patterns and other purchase behavior such as gym memberships can be used to predict patient risk of disease. Leading the way is Carolinas HealthCare System, which operates the largest group of medical centers in North and South Carolina. With more than 900 care centers including nursing homes, they have 2 million patients to analyze for risk, using data points such as purchases a patient has made using a credit card or store loyalty card, to create predictive models on patient risk and eventually to reach out to patients. Of course this data crunching has a purpose, and that is to meet quality metrics imposed by HHS and CMS. The goal would be to change the risk curve (more…)