As a person who majored in political science and has been engaged actively in public
Recently, I re-read Michael Lewis’ great book Moneyball, which, on the surface, is a book about baseball, and, particularly about Billy Beane, the General Manager of the Oakland Athletics.
Lewis, who wrote books such as Liar’s Poker, Panic, and The Big Short, is clearly intrigued by fields of endeavor in which individuals succeed because they recognize the value of data when others are operating more by the seat of their pants. Lewis described a baseball talent evaluation marketplace in which Billy Beane, who was obsessively driven by performance statistics, battled baseball scouts, managers, and coaches who tended to evaluate players based either on their visible physical and athletic skills or the performances these individuals observed. As a result, when Beane overruled his organization and made decisions based on his statistical analyses, he tended to acquire undervalued talent and get a premium price when he disposed of overvalued talent.
Moneyball was a great book for many reasons, including its insights about how people trained and developed in a system in which a particular world view predominates have difficulty adapting to a different world view, even in the face of compelling facts.
However, what I picked up this time was how a handful of dedicated baseball fans, many of whom were engineers, rose up against the power structure of Major League Baseball in the 1980’s to gather statistical data that so-called “experts” had never collected. For example, Lewis described Project Scoresheet, an informal baseball fan-based data collection system managed by a writer named Bill James, who wrote a series of books called Bill James’ Baseball Abstracts.
For example, James and the Project Scoresheet fans understood that, while it is clearly relevant in evaluating hitters to collect data enabling fans and teams to calculate batting average, determining where each hit was placed was valuable supplementary data, because it enabled talent evaluators to determine whether a player’s pattern of hitting would work across a wide variety of Major League Baseball fields, given their non-uniform dimensions.
However, what Lewis reported is that not only did Major League Baseball not capture some extremely relevant performance data, it was uncooperative in sharing the information it collected. Lewis commented that it seemed absurd that, for an event for which individuals paid varying amounts of money to watch, they could not access the data that described that event.
There is a parallel to the health care industry. Government agencies and others who pay for the health care we receive, whether it is a large employer or a large insurance company, get a great deal of data about our health care transactions. The hospitals, clinics, and, sometimes, the medical practices we patronize, also collect a lot of data as well. What they share with us, however, is delayed, incomplete, sometimes inaccurate, and not provided to us in a user-friendly form.
I chair a personal, patient-controlled, portable, private, secure health record company called Dossia, based in Cambridge, MA. When this initiative was started four years ago, the founders, one of which was Pitney Bowes, could not have imagined how difficult it would be to get some health care providers and insurance companies to give our participants, the employees who participate in employer health plans, the health data that had been collected as a result of their health care system encounters. Insurance companies report individual transactional data to patients with the same kind of obscure coding that health care providers and some government administrators use for their multiple purposes, not of which are centered around the needs of patients.
Moreover, this data is shared with us in paper form when we receive an Explanation of Benefis, or the EOB. This EOB is not particularly helpful to someone who is not a health care professional in terms of explaining to us what happened when we visited the doctor. While many insurance companies and some integrated health plans and payers like Kaiser-Permanente have made their electronic health records accessible to patients, these records are only usable when a patient is part of that health care system or health plan. Moreover, the data that originates from health care encounters outside the system does not get captured.
For example, since Kaiser-Permanente is in seven states, if you were a Kaiser member who needed to consult a physician in New York or Connecticut when traveling there, you would have to consult with a physician unconnected to the Kaiser system. It is not automatic that the transaction record will become part of the Kaiser record.
Similarly, if you moved from California, in which Kaiser is licensed to do business, to a state in which it is not, your record would not follow you. The onus would be on you rather than Kaiser to get a copy of that record.
The same issue exists with respect to insurance plans. Insurers and third-party administrators all have electronic health records usable when you are in their health plan or when you make a claim they process from an out-of-network encounter. However, they do not capture pharmacy, behavioral health, or other health data from a system they do not administer if an employer carves out a piece of its benefits program to be administered by a different benefits manager.
At Pitney Bowes, multiple insurance companies share the benefits management for the medical claims, but the company has a separate pharmacy benefits manager, and behavioral health services provider. None of the insurance companies would include the pharmacy and behavioral health events in their electronic health records.
That is why ten companies, including Pitney Bowes, started Dossia and continue to build its capabilities. The record is electronic and may not even have as much clinical data capability as Kaiser’s record or those of the different insurance plans, but it has four huge advantages in terms of its capabilities:
- It can be comprehensive, pulling in data from all different sources, including biometric data from the patient’s daily activities;
- It is portable, meaning that it will not stay with the plan or clinical care provider that owns it. This is a patient-controlled record.
- It is designed to be easy for patients to use. Clinical electronic health records, as well as insurance-centric records, were designed for the benefit of clinicians and to help insurance companies and health care providers manage relationships with payers, not for ease of use by patients. I do not blame them for this; the systems were created for different purposes and are hard to retrofit for patient use.
- It is designed to be usable by the caregiver for a family. Clinical and insurance records treat every patient or plan member as if he or she were unconnected to any other patient or plan member. Dossia is designed to help the caregiver, usually a mother, sign on to the record, and get immediate access to all of the family’s health information, not just her own.
This is why I am passionate about doing what the baseball Project Scoresheet people did, with the help of Bill James, with respect to baseball records in the health records space. Patients need to take control of their own information. This information does not belong to the insurance companies or the health care providers. It is a record of a part of our life, so we should own and control it, and get easy and free access to it in the way we want it, not the way the government, providers and insurance companies want to provide it to us.