Benefits of sharing longitudinal health data today
It is true. There are only those who have moved and those who are going to move. Vanishingly few of us remain living in one location throughout the decades of our lives. Furthermore, moving is hard, hard with children and especially hard with children with complex illnesses. Meet Emily. Emily is a mom with three little children. Her youngest has Down’s syndrome and regularly sees her primary medical provider as well as a number of different specialists. Out of necessity, Emily is extremely organized but no amount of effort on her part seems to speed up the heroics required to assemble her own and her children’s medical records. This process will take months, multiple phone calls, many faxed letters and frequent bills requiring payment… all to gain access to medical records that apparently she and her young daughters own. Emily uses Getmyhealthdata.org for help navigating the unanticipated complexities remaining in this effort today. This is just the beginning. We, like Emily, envision a future with significant benefits associated with sharing longitudinal health data; not just sharing with new medical providers but with a host of other products and services.
But wait; who needs longitudinal health data? What if you hope to avoid needing access to personal health data by not getting ill? Your chances of success are not good.
Today it is no longer the exception to be in need of intermittent or even regular medical care. In fact, humans in general seem to be increasingly ensuring they experience poor health outcomes. Females born in the year 2000 have a miraculous 40–50% chance of diabetes in their lifetime [1]. Boys have a slightly less chance. This changing pattern of disease may not be attributable to personal behaviors alone. Increasingly the suggestion of trans-generational (mother to child or possibly grandmother to grandchild) migration of disease risk is mounting [2]. Add on top of these risks the general health mishaps of daily living and soon an all inclusive future takes hold. Basically all of us are those who have either had or will have multiple medical interventions throughout our lives.
Unfortunately, not all of the interventions will go well. Why?
While we seem to suffer an increasingly complex life of accelerated human frailty, our access to in-person medical care is eroding. The US could be short 46,000 to 90,000 physicians by the year 2025. Thus the family doctor that “knows all” about us as an individual is a relic of bygone era. Most of us will end up in outpatient clinics. These are places described well by Mark Graber MD , “where the doctor doesn’t know the patient, the patient doesn’t trust the doctor, time pressures and frequent interruptions are the rule.” This sets up a third recurring scenario.
Medical misdiagnosis affects about 12 million adults every year in the United States [3].
These events are particularly problematic because a misdiagnosis data trail exists in bits and pieces throughout wide and unconnected health delivery systems. Patients will jump from one provider to another in the hopes of landing the right diagnosis. These are not necessarily exotic or rare diagnosis either. Among the most common misdiagnosed are Cancer, Heart Attack, Stroke and Depression. Other diagnosis, generally well known to be difficult to pin down, are Celiac Disease, Lyme Disease and Fibromyalgia. A 2013 Washington Post report included Graber’s work suggesting that “Diagnoses that are missed, incorrect or delayed are believed to affect 10–20 percent of cases, far exceeding drug errors and surgery on the wrong patient or body part, both of which have received considerably more attention.”
This is not good.
Securing information regarding misdiagnosis from health systems is at a minimum problematic if not largely impossible. Without good information, effective prevention strategies are difficult to craft at the enterprise level let alone the individual level. Yet individuals could play a key role in improving the collective intelligence required to avoid misdiagnosis. Longitudinal personal health records that share the mishaps as well as the successes will help others avoid retracing undesirable steps. Months and years of painful struggle could be reduced significantly if we could easily crowd-source the details that led to multiple misdiagnosis. No one health system can do this. People armed with the right technology could get this done.
Enough of the dark side, let’s move onto the brighter side for sharing longitudinal health data.
For example, getting care when you need it and where you need it is increasingly becoming common. Today many interventions can occur online. Since 2013, Omada has provided a digital solution for diabetes risk management. Virta Health is an online specialty service that reverses Type II Diabetes without medication or surgery. With a growing suite of products and services available on line, the cumbersome model of in-person care can make way for more convenient approaches. Thus convenient access and portability of our own personal health data would make using any online service more streamline. “Visiting” our data through patient portals becomes simply insufficient and unacceptable. In an ideal world, personally controlled health records would reduce the burden of sharing relevant data with online service providers to a “click here” option all within the privacy and security needed to avoid any data mishaps.
Likewise we can envision a future where a patient can receive a prescription as an addition to her personal health record. This prescription can be offered up to a prescription filling service which seeks to fill the prescription according to a number of selected parameters such as cost, speed of delivery, mode of delivery (mail, courier, drone or other) and so on. As medications near their expiry date they need to be sold quickly or lose all value. These close-to-expiration medications could be preferentially connected to valid and current prescriptions for a significant cost savings. Emily would welcome the ability to permission her and her daughter’s prescriptions to reputable prescription-filling services. Payers would welcome the lower costs and patients would welcome the convenience. Widespread secure longitudinal health records could make this benefit scale to other products and services as we expect increasingly to transact online.
For instance, sharing some personal health data with retailers is likely to become common practice. In doing so deliberately (not without our permission) our shopping experiences benefit from smart and personally customized approaches: Low glycemic load products for those reducing weight and lowering diabetes risk, specific products for high performance athletes and so on. Even our reading and media entertainment options might be permissioned to be informed to a varying degree by our personal health records. With these mass customization approaches, we can help ensure we have access to the ongoing and up-to-date intelligence needed to drive our best health expression.
With all the hassles of moving and medical warnings of misdiagnosis along with the outstanding on-line benefit potential of sharing longitudinal health data, why haven’t these records taken off yet? Why is Emily still struggling today? Is asking individuals to be the trusted guardians of their own health data simply too complex, too fraught with error to be reasonably possible? Today we still rely on legacy approaches providing a trusted central authority to control complex privacy and security concerns. In doing so, valid patient-generated requests for information continue to be a challenge. Though getmyhealthdata.org has helped hundreds since it began offering services in 2015, Erin Mackay confirmed recently that health data continues to move at a snail’s pace and points to a recent Office of the National Coordinator report outlining the data acquisition challenges as seen by individual patient narratives. Many years of significant effort have not produced the future anticipated and envisioned above.
Key barriers to this future that are often cited include poor patient engagement as evidenced by the low patient adoption of patient portals. Though most clinics and hospitals allow patients to visit their data through portals owned by the health facility, only a minority do so today [4]. Clearly the “just do it for your health” value proposition has not been enough to motivate patients sustainably. Advancing beyond the ‘for you’ model to more of a ‘for us’ model may take advantage of many of the principles of mass participation that Clay Shirky describes in “Here comes Everyone”. Combining these principles with other benefits inherent in new technology approaches that put the patient at the center of their data are beyond this post…more to come. In this post, we aim to confirm that the benefits of sharing longitudinal health are robust.
The preferred future is not here today. Personally controlled sharing of personally owned longitudinal health data will have a key role in providing a multitude of significant benefits that still elude us today while we remain locked in current approaches modelled on a bygone era.
May tomorrow hurry! Emily and the rest of us are depending on it.
Endnotes:
1) Lifetime risk for diabetes mellitus in the United States. Narayan KM, Boyle JP, Thompson TJ et al. JAMA. 2003 Oct 8;290(14):1884–90.
2) Metabolic disease: evolutionary, developmental and transgenerational influences. Gluckman PD, Hanson MA. Nestle Nutr Workshop Ser Pediatr Program. 2005;55:17–27. Review.
3) The frequency of diagnostic errors in outpatient care: estimations from three large observational studies involving US adult populations Singh H, Meyer AND, Thomas EJ, BMJ Qual Saf 2014;23:727–731.
4) HHS Should Assess the Effectiveness of Its Efforts to Enhance Patient Access to and Use of Electronic Health Information GAO-17–305: Published: Mar 15, 2017. Publicly Released: Mar 15, 2017.
