Monday, April 02, 2018

Overseas They Have Noticed A Downside To Contextless Results It Seems The ADHA Is Ignoring!

This appeared last week:

Without context or cushion, do online medical results make sense?

Mar 27, 2018 12:20pm
As she herded her two young sons into bed one evening late last December, Laura Devitt flipped through her phone to check on the routine blood tests that had been performed as part of her annual physical. She logged onto the patient portal link on her electronic medical record, scanned the results and felt her stomach clench with fear.
Devitt’s white blood cell count and several other tests were flagged as abnormal. Beyond the raw numbers, there was no explanation.
“I got really tense and concerned,” said Devitt, 39, a manager of data analysis who lives in New Orleans. She immediately began searching online and discovered that possible causes ranged from a trivial infection to cancer.
 “I was able to calm myself down,” said Devitt, who waited anxiously for her doctor to call. Two days later, after hearing nothing, she called the office. Her doctor telephoned the next day. She reassured Devitt that the probable cause was her 5-year-old’s recent case of pinkeye and advised her to get tested again. She did, and the results were normal.
“I think getting [test results] online is great,” said Devitt, who says she wishes she had been spared days of needless worry waiting for her doctor’s explanation. “But if it’s concerning, there should be some sort of note from a doctor.”
Devitt’s experience illustrates both the promise and the perils of a largely unexamined transformation in the way growing numbers of Americans receive sensitive—sometimes life-changing—medical information. A decade ago, most patients were informed over the phone or in person by the doctor who had ordered testing and could explain the results.
But in the past few years, hospitals and medical practices have urged patients to sign up for portals, which allow them rapid, round-the-clock access to their records. Lab tests (with few exceptions) are now released directly to patients. Studies estimate that between 15% and 30% of patients use portals.
The push for portals has been fueled by several factors: the widespread embrace of technology, incentive payments to medical practices and hospitals that were part of 2009 federal legislation to encourage “meaningful use” of electronic records, and a 2014 federal rule giving patients direct access to their results. Policymakers have long regarded electronic medical records as a way to foster patient engagement and improve patient safety. Studies have found that between 8% and 26% of abnormal lab results were not communicated to patients promptly.
Are portals delivering on their promise to engage patients? Or are these results too often a source of confusion and alarm for patients and the cause of more work for doctors because information is provided without adequate—or sometimes any—guidance?
Releasing results on portals remains “an answer with many questions,” said Hardeep Singh, a patient safety researcher at the Michael E. DeBakey VA Medical Center in Houston. “There is just not enough information about how it should be done right,” said Singh, who is also an associate professor at Baylor College of Medicine and one of the few researchers to study patients’ experiences obtaining test results from portals. “There are unintended consequences for not thinking it through.”
Although what patients see online and how quickly they see it differs—sometimes even within the same hospital system—most portals contain lab tests, imaging studies, pathology reports and less frequently, doctors’ notes. It is not uncommon for a test result to be posted before the doctor has seen it.
That means that a patient may be the first to learn of a suspicious breast mass, a recurrence of cancer or possible kidney failure. At Johns Hopkins medical system in Baltimore, for example, results of a PSA test to screen for prostate cancer come with this disclaimer: “While Johns Hopkins providers check results frequently, you may see results before your provider has seen them.”
Breast cancer specialist Lidia Schapira is an associate professor at the Stanford University Medical Center and editor-in-chief of Cancer.net, the patient information website of the American Society of Clinical Oncology. While she regards online access as beneficial, “the danger is that the patient may learn information they’re unprepared to receive and may feel abandoned if they can’t reach their doctor.”
“Those are the Friday afternoon phone calls,” she said, when “at 4:59 [p.m.] a patient has accessed the results of a scan and the doctor signs out at 5.” The recipient of the anguished inquiry that follows is typically a covering doctor who doesn’t know the patient or details of the case.

When is use meaningful?

recent study by Singh and his colleagues found that, like Devitt, nearly two-thirds of 95 patients who obtained test results via a portal received no explanatory information about the findings. As a result, nearly half conducted online searches. Many with abnormal results called their doctors.
That echoes a 2016 study led by researchers from the University of Pittsburgh. These scientists found that in addition to engaging patients, portal use may increase anxiety and lead to more doctor visits.
Among patients with low health literacy and numerical skills, confusion about the meaning of results is common. Many tests are reported in the same form that the doctor sees them, which even savvy patients may find “literally meaningless,” observed Brian Zikmund-Fisher, an associate professor in the school of public health at the University of Michigan.
“In some situations we run the risk of patients misinterpreting that there is no problem when there is one, or assuming there’s a problem when there isn’t,” said Zikmund-Fisher, lead author of a study that advocates the use of explanatory graphics to convey results. “What we need to be focusing on is giving patients context.”
A year or so ago, Geisinger Health System in Pennsylvania began making most test results—but not biopsies or HIV screening—available to patients within four hours of being finalized.
“We essentially release results twice a day seven days a week with a four-hour lag,” said Ben Hohmuth, Geisinger’s associate chief medical informatics officer. The delay, he said, gives doctors time to review results. Patients who log on over a weekend can contact an on-call physician if they can’t reach their own doctor. The goal of rapid release, Hohmuth said, is to “be patient-centered and transparent.”
“The majority [of patients] want early access to their results, and they don’t want it to be impeded” while waiting for doctors to contact them, Hohmuth said, even if the news is bad.
Patient reaction, he adds, has been “overwhelmingly positive”; the few complaints have come from physicians.
Health lawyer Kathleen Kenyon said she would have appreciated faster access to blood test results for her elderly mother, who had multiple medical problems including Alzheimer’s disease. Kenyon, who managed and closely monitored her mother’s condition, said she believes speedier access could have helped stave off a four-day hospitalization in the intensive care unit of a Washington hospital caused by her mother’s plummeting sodium level.
“It is safer for patients to have more information,” said Kenyon, formerly a senior policy analyst at the Department of Health and Human Services. “I was begging them to get my mother’s lab information in earlier.”
Lots more here:
What we see here is that there is both a real upside and a potentially  really major downside to having unmoderated pathology results (and imaging results) in the myHR.
Whenever anyone in our family gets some test results it is followed by a call to yours truly to place the results in context and to clarify if the * beside a test result is meaningful or not.
Unless the labs or the ADHA make available a hot line for patients to call if they are worried and can’t reach their own doctor sooner or later someone is going to panic and come to some real harm. There are just too many people out there who may be frightened / confused with results not to have a clearly available safety net in place.
What do you reckon?
David.

20 comments:

  1. I can only repeat part of an earlier comment of mine:

    Why are test results being put into myhr? It is supposed to be a summary record. (My answer is because NEHTA failed to address the ineteroperability problem).

    ReplyDelete
  2. Not sure what you mean, “failed to address the Interoperability problem”

    ReplyDelete
  3. To paraphrase Roy and HG's slogan: Too much health data is never enough. They were comedians. ADHA is serious.

    ReplyDelete
  4. > Not sure what you mean, “failed to address the Interoperability problem”

    I think they failed to appreciate the interoperability problem and ADHA still don't!

    Its vital the when you send some data its understood at the other end and this can literally be a matter of life and death in the right circumstances. We have badly constructed data messages being sent, often customised to be bad in a way that makes them work in the badly implemented PMS system/ If you send perfect messages they will fail so we have a N squared problem rather than a n problem because NEHTA and ADHA fail to appreciate that a system being able to faithfully receive a standards compliant message and ACK it is vital for safety. Many labs can send compliant messages, but as they say "No one has ever asked for those"

    The projects that ADHA hit this problem and they just say something like, "Can't we just send PDF" which infuriates me as that's what a modern fax system does and they then say "Lets have a bonfire of the faxes"

    They also involve a handful of the larger vendors in a messaging interoperability project and don't seem to understand that there about 30 different systems out there and having defined interoperability (dumbed down to pdf) between 3 applications they think they can claim success. When its clear that vendors can't possibly meet even basic standards compliance in the timeline, its "Can't we make it even simpler to meet a press conference deadline"

    It has not worked and will never work unless they focus primarily on systems (all systems) being able to consume standards compliant messages flawlessly as the first step and then insisting that people send compliant messages as the second. If the first step is in place you will have less trouble with the second, as developers will know that if they produce a 100% compliant message it WILL work.

    This is the fundamental building block of eHealth and its been ignored for nearly 2 decades and we have gone nowhere. The free exchange of high quality compliant data would allow a myriad of other application, including personal health records to be built with relative ease, but instead we have half baked, cobbled together repositories of pdf documents which are of very limited value and providers still can't easily, reliably exchange data, or can only exchange non atomic data which does not integrate into their system without human interpretation.

    It seems unlikely ADHA will suddenly get it, so lets hope government runs out of money and cans it, for everyone's sake.

    ReplyDelete
  5. re: "failed to address the Interoperability problem”

    Secure messaging – the problem

    https://www.digitalhealth.gov.au/get-started-with-digital-health/what-is-digital-health/secure-messaging

    "Reliable, secure provider-to-provider communication is a key component of digitally enabled integrated and coordinated care across the Australian health sector. Secure messaging is a core foundational capability required to enable interoperability and safe, seamless, secure, and confidential information sharing across all healthcare providers and consumers.

    While there are significant pockets of secure messaging in use, for example, supporting pathology communications and discharge summaries from certain hospitals, there continues to be the lack of a consistent approach to secure messaging and information exchange across Australian healthcare. This has led to significant challenges across the sector."

    The National Digital Health Strategy:

    "In Australia, there is established use of secure messaging using a range of different electronic communication methods; for example, diagnostic requesting and reporting, and sending discharge summaries from hospitals to general practice. However, these different methods are generally not compatible – meaning that these proprietary secure messaging approaches do not work with each other. Despite significant effort, there is no nationally consistent, standards-based approach to secure messaging, which limits the ability of healthcare providers to communicate effectively."

    Correct me if I'm wrong but the National Digital Health Strategy implies there is an interoperability problem and that they are trying to solve it, even though the ConOp, in 2011 said that Secure Messaging was a foundation requirement for the PCEHR System.

    ReplyDelete
  6. Before you can have interoperable secure messaging you have to have interoperable messages and this is where there is no focus. Secure messaging currently has two layers, 1. An interface engine to tweak the message so it does not fail and 2. The actual delivery.

    There are significant issues with 2. including a failed PKI infrastructure program, a failed provider identification infrastructure and a failure to reliably identify practices at a granular level.

    However without message interoperability connecting messaging systems disables the controlled interface engine component and things stop working or failures occur which threaten patient safety. ADHA/NEHTA think the problem is entirely with messaging interoperability and when the problem of message interoperability is pointed out they just run to pdf. PDF is not going to give any of the promised benefits of eHealth and really is just a fax network, which would be more efficiently handled by using fax!

    ReplyDelete
  7. Re having pathology results accessible on MyHR, a personal (and trivial) example.
    A couple of weekends back I went to an afterhours GP. She collected swabs (viral and bacterial) and prescribed an antibiotic, after I told her I had knowledge of medical microbiology and stated my preference. I understood the results would go to my regular GP. I haven't been back to the clinic yet, will try to get in this week. But I don't think that having access to those results would have been helpful, though I expect if something alarming had turned up I would have been notified.
    As I see it, those results are the property of the pathology laboratory, to be released on an understanding they will be mediated by a qualified practitioner.

    ReplyDelete
  8. I'm not sure this issue is confined to digital health, although it certainly may be exacerbated in this context. Around four years ago I was sent home with the physical images of a scan very clearly showing multiple lumps, it was 3 days before I received the email from my GP to tell me they were benign. I think this is more the case of an existing issue which will be highlighted by advances in technology and communication speeds. Is it possible this could actually be a positive use of MyHR? A clinician other than the patient's GP could have access to the patient records and background to give a faster interpretation instead of just leaving the patient hanging.

    ReplyDelete
  9. Sound familiar?
    http://dilbert.com/strip/2018-04-03

    ReplyDelete
  10. For those that might be interested in a longer explanation of what you need to standardise for pathology to work in electronic health records, I have had a go at writing about 'interoperability' for that use case here
    https://www.sciencedirect.com/science/article/pii/S0009898113004919

    ReplyDelete
  11. Michael,

    When the requester of the pathology test receives the results, those results need (often? always?) to be interpreted in context and be integrated into a larger body of data.

    Do you consider that these fall into the scope of standarised interoperability requirements?

    ReplyDelete
  12. The potential for alarm is certainly there, when patients get results they do not understand, and make no mistake most will not understand. There are people who can interpret some results better than their doctor and in this day and age there are people, with no medical qualifications sorting through the literature (via Sci Hub) and coming up with brilliant explanations for eg changes in lipids on diet. I think, at a minimum the ordering doctor should be the one to release results to the patient, as they can judge if they will understand and could potentially attach a comment in the process.

    Its not just alarm for no good reason that happens, false reassurance also occurs. Just this week I had an obstructive bowel cancer where the biopsies only showed a polyp. Would a patient feel they did not have cancer, when I would simply ignore that result and proceed with surgery knowing that that is something that happens from time to time.

    The issue is that enabling a system where the GP releases a result and attaches a comment requires more integration, quality and standards compliance to allow PMS systems to send results to MyEHR. ADHA know that would take ages to do, because no one has ever tried to ensure any standards compliance by endpoints, so they have gone for the easy fix, really at the expense of 95% of patients. I think its only fair that the ordering provider gets to review a result before its released, as its hard to answer questions about something you have never seen.

    I think they are taking a huge risk.

    ReplyDelete
  13. IMHO, the myhr is trying to solve two problems at once - a supposed personal health record and a desperate attempt to achieve some degree of interoperabilty.

    The pathology test results get sent to myhr so that they will end up with a GP (the interoperability bit), but the GP should see the results and discuss them with the patient first, before uploading them to myhr, if appropriate for a summary record system (The personal health record bit. Although what pathology test results are doing in a summary record is not clear).

    Unfortunately, myhr gets the whole thing dangerously wrong - the GP can't annotate the pathology results, they just sit there in the patient's myhr like a potential time bomb.

    ReplyDelete
  14. A gem of an anecdote as reported in The Disappearing Doctor: How Mega-Mergers Are Changing the Business of Medical Care is worth pulling apart.

    Eva Palmer, 22, of Washington, D.C., sought out One Medical, a venture-backed practice that is one of the nation’s largest independent groups, when she couldn’t get in to see a primary care doctor, even when she became ill. After paying the annual fee of about $200, she was able to make an appointment to get treatment for strep throat and pneumonia.
    “In 15 minutes, I was able to get the prescriptions I needed — it was awesome,” Ms. Palmer said.


    On the one hand, as suggested in the article, time-poor people directed to easier access for primary care could benefit from having a doctor perform a rapid test for antigen to pick up presence of Strep pyogenes in the pharynx.
    OTOH, though, a diagnosis of "pneumonia" inside 15 mins?

    As I see it there are three separate battlegrounds just within the context of this topic on test results in MyHR, each with own sets of players and requiring specific rules of engagement.

    1. There could be a push for "more tests" fuelled by any demand, however uninformed or misguided, from elements or entire corporations lobbying ADHA. For one, the makers of Strep antigen kits may like more sales. How do pathologists respond?
    2. "Stories" like the above anecdote will be propped up for injection into mass media infotainmant gimmicks. Responsible media outlets, especially the publicly-funded, should be able to interrogate those flashy "gems" at editorial management level.
    3. The Big Data arguments alluded to in the article are getting more strength. Teasing out the worthwhile aims (of which there are many) from the pure profit-making ones need to be addressed by regulation.

    ReplyDelete
  15. The supposed future of medicine - Patient Centric care has different meanings in the health care system and in clinical medicine (i.e. the practice of medicine).

    The danger is that (as is pointed out in Terry's article) that clinical medicine suffers in an attempt to make the health care system more Patient Centric.

    IMOH, myhr does nothing in terms of Patient Centric care in either the health care system or in clinical medicine, fact, by trying to get GPs to look after two health record systems (their own and myhr) they make things worse.

    Drawing pretty pictures with the patient in the middle (See PCEHR documentation) is a fatuous simplification and demonstrates a severe lack of understanding of what patient centric actually means.

    ReplyDelete
  16. Further to my comment re patient centric care.

    This is an excellent article by Jeremy Knibbs of the Medical Republic
    http://medicalrepublic.com.au/gps-healthcare-distributions-game-thrones/13839

    It explains very simply that the nexus of the patient/healthcare system lies in the GP/patient relationship.

    IMHO, strengthening and utilising this relationship will, eventually, lead to better patient centric medicine.

    So what did the Federal Government/NEHTA do? They tried to build a patient/government relationship with the GP feeding it with copies of their medical data.

    NEHTA techos thought the problem was all about health records - it isn't, its about processes and relationships. Acquiring, managing and utilising data in support of processes and relationships will deliver value. If you don't understand them the data you gather will be worse than useless - it will distract and confuse.

    ReplyDelete
  17. It explains very simply that the nexus of the patient/healthcare system lies in the GP/patient relationship.

    Bernard, you may have got the wrong end of the stick. My Health Record is not aimed at the patient, it is aimed at the GP. The government may actually be trying to monitor GPs. Why would they want to do that? The government's only real interest in the health care system is how much it spends. GPs are responsible for a lot of that expenditure and can also impact other expenditures. First comes the data acquisition, then comes the control.

    The difficulty the government is going to run into is that My Health Record doesn't do what the government claims it does. Getting patients and GPs to co-operate is going to be a nightmare for them. Public apathy and GP's suspicions are likely to prevail.

    Is this lot smart enough to get away with such a deception? Apart from will it work? and there are grave doubts that it will deliver what the government really wants, keeping a conspiracy a secret is always the biggest challenge.

    Tim will get his nice salary and travel perks for as long as he can. He'll then slink off to the WHO, casually blaming everyone else. We'll all be left with a lost opportunity and another failed government Health/IT project.

    ReplyDelete
  18. "you may have got the wrong end of the stick"

    Quite possibly.

    If you look at the three stakeholder groups, you get different perspectives.

    Most patients aren't interested in their health records. They already have lots of advice and health messages directed at them but they choose to ignore them. Opt out will not give them an "online summary of your health information.", as the government claims. All opt out will do is register them, they've still got to see their GP and arrange to upload data. Patients who are really interested in their own health can already get at all their health information.

    Health professionals don't need another health record, they need their existing medical record systems to talk to each other better. Making GPs use two record systems makes no clinical sense at all, especially when one is only a summary system, is unreliable (the government says so), and can have data hidden or excluded. Health professionals also include state governments. Looking at their involvement, they are quite happy to go along with myhr as an adjunct to what they are doing as providers of health care services. myhr isn't replacing anything or adding anything to what they are doing.

    The government, on the other hand is something of a dark horse. I don't know why they have spent $2b so far and with much more to come, I can only guess.

    There are some indicators.

    The government is running a project called the Multi-Agency Data Integration Project (MADIP)

    http://www.abs.gov.au/websitedbs/D3310114.nsf/home/Statistical+Data+Integration+-+MADIP

    "MADIP is a partnership among Australian Government agencies to combine information on healthcare, education, government payments, personal income tax, and the 2011 Census to create a comprehensive picture of Australia in 2011.

    Authorised researchers can use confidentialised, de-identified MADIP data to look at patterns and trends in the Australian population, and provide insight into the effectiveness of government policies, programs, and services. MADIP demonstrates how combining existing public data can help target services (such as healthcare) to the people and communities who need them.

    MADIP is a core component of the Australian Government’s Data Integration Partnership for Australia (DIPA). There are six Commonwealth agencies working together on the project:

    ABS, ATO, Education and Training, Health, DHS, DSS."

    myhr is not a conspiracy - it is quite obvious what the government is trying to do.

    Where the government is being disingenuous is when they misrepresent what they are doing to the key players - patients and GPs. They are pretending that myhr is all about patient health care. They are pretending that health professionals will benefit. In both cases, there is no evidence that either claim is valid, quite the opposite.

    The question I'd like to know the answer to is, does ADHA know the full extent of the government agenda, or have they been misled as well?

    This could be a classic example of divide and hide. Individual agents don't see the full picture and so don't understand the full ramifications. GPs are being encouraged to spend time and effort for no benefit to themselves. In fact it is quite possible that the data they help to acquire could be used against them. It's an obvious conclusion if you look at the full extent of what the government is trying to do. As it says above "provide insight into the effectiveness of government policies, programs, and services". The Federal government does not provide health care services, they fund them. It is logical that their focus is on health care economics.

    I agree that, by looking only at what the government is saying about myhr, I/you will get the wrong end of the stick. The key is to look at what the government is saying and doing in its entirety. If you do that, you'll see a different stick altogether.

    ReplyDelete
  19. Nice informative insight Bernard. I would suggest the stick is in fact more akin to a large mallet

    ReplyDelete
  20. Just in case you think I'm making up all this stuff about myhr being about health economics not health care, have a look at this document, which is being used to support the MADIP initiative.

    https://www.pmc.gov.au/sites/default/files/publications/case-studies-data.pdf

    "The Data Linkage System at the WA Dept. of Health was established in 1995 to undertake systematic integration of health and administrative data for research and statistics – the first of its kind in Australia. Over 800 projects from academia, government, and hospitals have used WA’s linked data.

    Data linkage has been used to evaluate doctors and hospitals and to study the patterns, causes and effects of heart disease and cancer. Data linkage has also been used to improve planning for the flu -season and other busy times."

    Note the phrase "to evaluate doctors and hospitals".

    The Federal government wants to get in on the game. It has data, it wants more and it expects to do more with it. Opt-out is only the start. Forcing GPs to use the system and input more and more data is, in all probability, the next step. That won't help patients or GPs, but will help the government and its economists.

    IMHO, better sharing of health information among health professionals is a good thing. Reducing waste and increasing efficiency in the health care industry is a good thing.

    Doing them badly and trying to hoodwink the population destroys trust and they don't get to do either.

    As David has pointed out in today's blog posting, there are better ways of doing Digital Health than the government is currently doing. That's what exposes the fiction that it is in the patient's and GP's interests.

    ReplyDelete