Hoarding Patient Data is a Lousy Business Strategy: 7 Reasons Why

Leslie Kelly Hall
Vince Kuraitis


Among many healthcare providers, it’s been long-standing conventional wisdom (CW) that hoarding patient data is an effective business strategy to lock-in patients — “He who holds the data, wins”. However…we’ve never seen any evidence that this actually works…have you?

We’re here to challenge CW. In this article we’ll explore the rationale of “hoarding as business strategy”, review evidence suggesting it’s still prevalent, and suggest 7 reasons why we believe it’s a lousy business strategy:

  1. Data Hoarding Doesn’t Work — It Doesn’t Lock-In Patients or Build Affinity
  2. Convenience is King in Patient Selection of Providers
  3. Loyalty is Declining, Shopping is Increasing
  4. Providers Have a Decreasingly Small “Share” of Patient Data
  5. Providers Don’t Want to Become a Lightning Rod in the “Techlash” Backlash
  6. Hoarding Works Against Public Policy and the Law
  7. Providers, Don’t Fly Blind with Value-Based Care


In the video below, Dr. Harlan Krumholz of Yale University School of Medicine capsulizes the rationale of hoarding as business strategy.

We encourage you to take a minute to listen to Dr. Krumholz, but if you’re in a hurry we’ve abstracted the most relevant portions of his comments:

“The leader of a very major healthcare system said this to me confidentially on the phone… ‘why would we want to make it easy for people to get their health data…we want to keep the patients with us so why wouldn’t we want to make it just a little more difficult for them to leave.’ …I couldn’t believe it a physician health care provider professional explaining to me the philosophy of that health system.”

Health care data has been viewed as a competitive asset by many health care providers. The CW among many providers over the past several decades has been that hoarding patient information locks-in patients to specific care providers and deters them from going to competitors. CW adds that data also can lock-in independent physicians and other care providers.

Quantifying the magnitude of hoarding is challenging. It’s not exactly the type of guiding vision that senior management necessarily would broadcast broadly and loudly, or even acknowledge publicly. It is likely that many constituents — e.g., clinicians, local press, regulators, employees — would find the strategy to be odious. It’s more likely that senior management would publicly deny that they embrace hoarding as business strategy.  

Where it does exist it’s possible to exist primarily (or solely) at the C-level of provider organizations. It might not even be an explicit strategy — it could be implicit — something that the CEO, CIO and/or CFO hold in their heart-of-hearts or understand with a wink-and-a-nod.

Hoarding also comes in many shades and flavors, many of which might be more paternalistic or cautious than malicious. For example, one organization’s view of “stewardship” might be considered hoarding by another that wants the data. Both are right.

So how prevalent is hoarding? We have to rely on indirect evidence — and there is plenty of that available.

A Center for Studying Health System Change analysis examined barriers to health information sharing. In interviews with health system executives, they found that Hospitals’ greatest concern in weighing the costs and benefits of participation in [a health information exchange (HIE)] was losing competitive advantage by relinquishing control of ‘their’ data. They viewed clinical data as a key strategic asset, tying physicians and patients to their organizations.” (emphasis added)

A Robert Wood Johnson Foundation Study asked HIE organization respondents to characterize the degree to which potential barriers slowed their development. 83% listed “concerns about competitive position” and 78% listed “lack of stakeholder [provider] interest”.

A 2017 study by Julia Adler-Milstein and Eric Pfeifer examined the prevalence of information blocking. HIE respondents answered that 50% of EHR vendors routinely engaged in information blocking and that another 33% did so occasionally; they also answered that 25% of hospitals/health systems engaged in information blocking routinely and another 34% occasionally.

A few qualifiers. First, we don’t mean to imply that hoarding as business strategy is the only reason that’s limited appropriate exchange of patient health data; there are many other factors at play. To name a few: a fear of violating HIPAA regulations; a lack of technical standards; privacy/security concerns; technology and infrastructure gaps and others.

Second, we also understand that patient retention/loyalty is an important and relevant strategic initiative in most health care organizations today. It’s one of few strategies that make sense in BOTH the old world of fee-for-service payments and the new world of value-based payments. There are many viable tools and tactics to increase patient loyalty and to improve clinical care through care coordination — but “hoarding as business strategy” isn’t one of them.

Finally, the specific focus of this article is to explain why hoarding is a lousy business strategy. We also strongly believe that there are many positive reasons why appropriate sharing of patient data is a powerful and effective business strategy – but that’s a separate article.

7 Reasons: Hoarding Patient Data is NOT an Effective Business Strategy

Let’s dive in.

1) Data Hoarding Doesn’t Work — It Doesn’t Lock-In Patients or Build Affinity.

Hoarding as business strategy assumes a provider view that they’re running a tight ship — that patients are highly loyal to care providers and stay within their confines.

That view would be wrong. The reality is that care provider networks are highly “leaky” and that patients readily cross over among 2 or more health systems. This phenomenon has been studied extensively.

Some of the most extensive data on leakage and crossover (patients registered at 2 or more providers) come from public Health Information Exchanges. Maine HealthInfoNet documents their current crossover rate as being 61%. In two-hospital towns like Bangor and Lewiston, HealthInfoNet’s system has shown as much as 75% patient crossover.A study of Indiana ER visits found that more than 40% of emergency visits during the one-year study period were for patients having data at multiple institutions. That rose to 60% when multiple years of data were included.

A Massachusetts study found that significant leakage takes place even within captive referral networks. One health insurer’s referral data showed that “only 35-45% of adult inpatient care, as measured by revenue, goes to the partner hospital”.

A 2018 analysis in the American Journal of Managed Care examined specialty referral leakage in Medicare accountable care organizations (ACOs). The researchers found leakage rates of 61-72% among the most specialty-oriented ACOs (Figure 2).

The take away here is that patient lock-in strategies just haven’t worked. The natural state is for patients to go pretty much wherever they want or wherever urgency demands.

2) “Convenience is King” in Patient Selection of Providers.

Chances are you’ve seen a billboard like the one below.

Even as the hospital C-suite might be thinking “Hoarding is our business strategy”, the hospital marketing department groks that patients overwhelmingly are motivated by convenience — that they will select a specific provider simply to save a few minutes time.

Patients DO NOT think: “I want to go to ACME ER because they have all my data – they know me”.

Patients DO think: “I’ll go to ABC ER because they have a short wait time.”

The Advisory Board (AB) provides consulting, research and technology to hospitals and delivery systems. The AB asked 4,000 consumers about their on-demand care preferences and published a list of the “Top Ten Preferred Primary Care Clinic Attributes”. 6 of the top 10 attributes were related to access and convenience.

The element of whether a provider has access to a patient’s data (whether “they know me”) didn’t even make the list. They concluded that “Convenience is king”.

3) Loyalty is Declining, Shopping is Increasing.

The Advisory Board also measured patient loyalty to their primary care physicians. They surveyed 1,843 patients who had a primary care visit within the past year and found:

  • Only 53% would stay with their current PCP for the next 12 months
  • Only 9% of respondents would follow their current PCP if they changed locations
  • Only 36% would recommend their PCP to family and friends
  • Consumers were more likely to promote Sprint, Apple, Amazon, and Orbits than their PCP

The UnitedHealthcare 2018 Consumer Sentiment Survey found that 36% of respondents had used the internet or mobile apps to comparison shop for health care services — a 257% increase from 2012.

Many trends in the health care marketplace point to further decreases in patient loyalty and increased prevalence of consumers’ shopping for care:

  • The greater financial burden being shifted to patients, e.g., a report from Americas Health Insurance Plans documented that the number of people enrolled in HSA-Qualified High Deductible health plans grew from 1 million in 2005 to 22 million in 2017.
  • Patients have more choices, providers have more competition — the growth of retail clinics, the entry of tech companies (Amazon, Apple…) and retailers (Walmart, CVS…), funding for new digital health companies.
  • Increasing availability and relevance of health care quality and pricing information.
  • Improvements in patient digital tools and apps.
  • Consistently poor patient experience drives patients to seek alternatives.

4) Providers Have a Decreasingly Small “Share” of Patient Data.

Providers might think that they have the corner on patient’s health data, but that mindset is increasingly inaccurate. They have a sliver of a patient’s data.

The graphic below from Travis May, CEO of Datavant, illustrates the scale and distributed nature of health data. Note that provider data in EHRs is represented only by the box in the upper left corner.

A providers “share” of a patient’s health data is declining over time. New and better sources of data are defining a patient’s experience of health and care, e.g., patient-generated health data (PGHD), data from remote monitoring devices, genetic data, data relating to social determinants of health, etc.

5) Providers Don’t Want to Become a Lightning Rod in the “Techlash” Backlash.

A recent article in the New York Times describes “the year’s biggest overarching tech narrative: the ‘techlash’ against companies like Facebook and Google over issues like privacy, data protection and antitrust concerns.”

One effect of techlash is that it is sensitizing patients to the notion that patients themselves should be in control of health data, not some 3rd party.

To-date, techlash has affected tech companies, but there’s great potential for techlash to spill over to health care organizations that treat patient data as “their” asset, not patients’ asset.

Does this sound far-fetched? Consider what’s happened at Memorial Sloan Kettering Cancer Center over just the past few weeks. Administrators acted as-if the data from 25 million patient tissue slides was “their” asset to sell to a private company…and now they are the focus of a scandal investigation and front page news.

6) Hoarding Works Against Public Policy and the Law.

The last item spelled out the dangers of providers’ being called out in the court of public opinion. Providers should also be increasingly concerned about being called out in a court of law.

Many recent legislative and policy initiatives are emphasizing patients’ rights relating to their health information:

  • Patients’ right to access their health records under HIPAA
  • Information blocking restrictions in the 21st Century Cures Act
  • The MyHealtheData initiative, which “aims to empower patients by ensuring that they control their healthcare data and can decide how their data is going to be used.

Health and Human Services Secretary Alex Azar, spelled out four areas of emphasis toward value based transformation. Item #1 on his list was “giving consumers greater control over health information through interoperable and accessible health information technology”.

7) Providers, Don’t Fly Blind with Value Based Care.

There are many ways in which information exchange will be necessary to support value based care, but we’d like to focus on one in particular.

As financial risk increasingly is shifted to care providers, providers will be responsible for the cost and quality of care provided by out-of-network care providers.

Let’s recall some of the data we presented under #1. In the last example, researchers documented that 61-72% of ACO specialty referrals went to out-of-network providers.  If data is hoarded and not shared appropriately with these out-of-network providers, two things happen:

  • Out-of-network care providers fly blind — they don’t have the best data to care for patients
  • At-risk care providers fly blind — they can’t assure that patients’ get appropriate care and they undermine their own economic self-interests

That’s flying blind for roughly 2/3 of referred patients!!


The bases of competition in healthcare are shifting – away from competing based on misguided attempts to lock-in patients and providers – and toward competing based on providing care that’s high quality, low cost, and a superior patient experience.

Sometimes conventional wisdom goes stale…or was never wise in the first place.

Vince Kuraitis, JD/MBA (@VinceKuraitis) is an independent healthcare consultant with over 30 years’ experience across 150+ healthcare organizations. 

Leslie Kelly Hall (@lesliekellyhall) is a nationally recognized leader in patient engagement advocating for patients in health IT and beyond.

18 replies »

  1. Sounds like a community approach to HOTSPOTS. Interesting. You may be interested in Mayo soc media approach to pophealth. Very cool.

  2. Patient retention/loyalty is an important aspect today. There are many viable to increase patient loyalty and the best of all is to improve clinical care through care coordination. Adapt solutions that enhance the quality of care and improve patient outcome.

  3. Dr. Palmer and Vince,
    A CMIO I met with last week is very hopeful that we will get new ways to filter information to what is relevant in the moment. In the meantime, the ON/OFF can be mitigated. Most providers want their questions answered. So, a patient’s response to a question posed electronically can be very helpful, relevant and engaging.

  4. Congratulations on having so much rich information available. You are very fortunate. I agree that EMR short comings are many. Interoperability is just one. The people most likely to effect change in EMRs usability will be providers and other stakeholders, not the government. (which is good).

  5. First place transparency to occur will likely be medication price transparency that is patient specific. Platforms like @surescripts are sure to step in.

  6. Just as an aside, market-share management by the large hospital enterprises has been largely horizontal. This will probably dissipate once vertical integration begins to occur. Somewhere amid this process of change to vertical as well as horizontal integration, antitrust issues or nationalization may occur. This would further shield the healthcare industry from its obligation to participate in any Social Capital improvement strategies within a community. This includes their institutional, social responsibility to participate in both the upstream issues aka Social Capital and public health, as well as the down stream, especially mental health, community responsibilities.

  7. Especially like this section: “Loyalty is Declining, Shopping is Increasing.”. Nice summary of trends. Pundits dismiss HSA plans and shopping, even as they continue to expand. This will exponentially increase once we wrangle price transparency from health systems and insurers. It is coming, perhaps with legislation or maybe just patient demands.

  8. “That’s flying blind for roughly 2/3 of referred patients!!”

    That’s simply not the case. With a well-trained staff and current technology, I have all the info I need ( and usually more) when I see the patient 99.9% of the time.

    Dr. Palmer is right. From the physician’s point of view, “interoperability” is being used as a red herring to distract us from far more serious problems with EMRs.

    Why do I think the free flow of information is being pushed by those who stand to make a buck off the data?

  9. Dr. Palmer, Your issue certainly is valid. TMI can be a problem.

    Can we agree that there are multiple levels of issues to work out, not just that the data spigot is “off” (hoarded) or “on” (shared)?

    #1 — getting data to flow in the first place? This is what we’re primarily addressing in our article by suggesting that hoarding isn’t a good strategy. We’re not implying that the data sharing spigot should go from being entirely “off” to entirely “on”.

    #2 — getting the right data, right time, right amount, etc. — which gets to your comments. This will vary based on unique needs of patient AND clinicians.


  10. Bill,

    My own dream involves two simultaneous efforts:

    .1) Start a community defined measurement tool to account for the total number of hospital bed-days/citizen used by the community’s citizens that is publicly reported monthly within 15 days of month-end (reported possibly by age groupings).

    .2) Form a healthcare group and a neighborhood coalition group to promote within their respective group the trends and underlying issues for the monthly utilization that is occurring. The neighborhood coalition would begin first with an evolving plan to promote the micro-neighborhood network of each citizen’s family. This would eventually be slowly taken on by the community, such as, neighborhood, churches, service groups and volunteer groups (Habitat). Eventually, this effort would evolve into community based processes to help resolve pockets of entrenched poverty.

    .My view is that eventually a steadily declining level of monthly bed-days/citizen trends would level off at a 20-30% lower level over 10-15 years with periodic spikes ( such as influenza ) and periodic lows ( a recession). The only other monitoring tool, eventually, would be the benefit accrued from a quarterly reported, 5 year running average of maternal mortality. The reporting processes should be managed by the medical-staffs of each hospital to assure its accuracy and security.


  11. Good, pj. I would think a great way to test and possibly abet community social capital would be to try an experiment somewhere, sometime where we make non-ambulatory care—hospital care—a public good. No billing. No billing records. Medical records only. Everything paid for by community resources, hopefully augmented by national. Everyone on salary. Drugs obtained by world-wide shopping. Patients admitted by triage algorithm that looks for “is this process one that could kill the patient eventually? Is this process one that could disable the patient? Is this process one that could possibly cause bankruptcy in the patient?”

    Ambulatory care would be left alone. People could do whatever they wanted here. They could purchase insurance, allow employer contributions, use cash….whatever.

  12. Good article. Sound logic. But….
    You have to shift your thinking such that you picture patients, not as race horses whose every particle of medical care history might be important in winning the next race by a fraction of a second, but toward the concept that patients are more like the family jalopy car that, when it gets mechanical troubles, the mechanic doesn’t really care as to what went on before; he only cares about what he finds today. So the brakes needed fixing three years ago and the radiator leaked last year. Who cares? I want to find out why there is a rumble coming from the transmission now! This whole concern over interoperability and spreading of data to all providers is a ho-hum to most providers*. These guys want to gather their own data and size up the patient’s problems NOW.

    Besides, the human body is always changing. It is a dynamic mecurial ball of complex chemistry. It is good to repeat lab tests and imaging and always obtain more history as time moves forward.

    *Sometimes surgeons really do need to have access to old records when they go into a messy abdomen screwed up by an old roux en Y procedure, for example. Pretty rare in the big picture, however.

  13. Most of the symptoms are lifestyle associated, for which one visits to the clinician. In modern practice, a visit to clinician often results in a prescription with some diagnostic test, prophylactic (preventive) pills (disease for which symptoms) to attain health. The prescription fulfills the mutual desire that ‘‘something be done’’ to ritually and therapeutically justify the visit. The patient tracks down the prescription and struggles to remember to take them on time, and also pay-off for them to restructure their lives. Continue reading…

  14. Please accept that it might be possible the following population health problems have an underlying cause of causes: homelessness, obesity, maternal mortality, opiod overdose mortality, young adult homicide/suicide, substance addiction, mid-life depression/disability and mass shootings. Looking at the last 30 years beginning in 1985, the number of mass shootings between the first 15 years (88) and the second 15 years (206) increased by 234%. During the same interval, our nation’s maternal mortality incidence (6.84 to 16.36) increased by 239%. Our nation’s worsening level of Social Capital is the likely process disturbing the stability of each citizen’s health and producing a disproportionate effect for certain population groups. It is UNLIKELY that the current efforts underlying healthcare reform can compensate for the unstable health that affects every citizen before they encounter our nation’s commitment to health care.

    I realize that this assertion is likely to encounter a high degree of cognitive dissonance. Remember…there is very little evidence that our nation’s over-all strategy for healthcare reform demonstrates a significant effect on the cost and gaps in quality within our nation’s healthcare. To help with the cognitive dissonance, I offer my own version as a basis to understand SOCIAL CAPITAL.

    .a community’s norms of TRUST, COOPERATION and RECIPROCITY that
    .its citizens spontaneously express for resolving the social dilemmas
    .they encounter daily within the community’s civil life
    .WHEN Caring Relationships are persistently nurtured
    .within the social networks of the community’s citizens, especially
    .the generational Caring Relationships originating
    .within the micro-neighborhood network of each citizen’s Family.

    The significance of Social Capital is that it is generational and ultimately community driven. We have no nationally instituted and locally managed plan to institute such an effort. To clarify this further, I offer a definition of a COMMUNITY.

    .A cluster of persons identifiable by certain uniform attributes,
    .TYPICALLY as Families residing within a geographically defined region,
    .who share a valued awareness about their interconnected identity
    .that is borne out of mutually experienced events and
    .each person’s memory of the ecological and cultural traditions
    .associated with these mutually experienced events.

    A nationally sanctioned, county by county, collective action process has existed for our nation’s agriculture industry since “1914.” This industry is the most efficient and effective among the OECD nations, by a wide margin.

    Best study for mass shootings https://doi.org/10.1002/jip.1491
    Robert Putman best review of social capital http://smg.media.mit.edu/library/putnam.pdf
    Nine definitions for healthcare reform at https://nationalhealthusa.net/paradigm-shift/rationale/