February 28, 2020

Some Underlying Principles of Patient-Driven Healthcare

Curator’s note: “Some Underlying Principles of Patient-Driven Healthcare” is the unpublished chapter 8 of the e-Patients white paper, which Tom Ferguson wrote to follow “The Autonomous Patient and the Reconfiguration of Medical Knowledge.” I just found a printed version, dated September 2, 2020. Enjoy!

People have changed more than the organizations on which they depend. And here is the new opportunity… a new order… emerging from the disappointments, frustrations, and all-too-frequent humiliations to which today’s new individuals are routinely subjected at the hands of the old organizations.

—Shoshana Zuboff and James Maxmin

None of us started out as an e-patient advocate. We never thought of our patients as potential partners. The professionals among us were trained in the old doctor-knows-best, patient-knows-nothing tradition. And when we needed medical care, all of us adopted the same passive, dependent role we saw our patients take.

But over the past decade, our beliefs and attitudes have changed. We’ve learned from our patients, from our friends and family members’ experiences as patients, and from our own experiences as patients. We’ve also learned from a growing number of pioneering colleagues, who, like the Advisors and many of the Reviewers of this white paper, suspect that the autonomous and capable e- patient, reinforced by smart IT medical systems and supportive professionals, may offer the best possible scenario for improving healthcare.

During this time, we’ve become at least moderately Net-literate. A few of us have become extremely proficient in understanding the deeper technical levels of the networked world. We’ve also watched how the Internet has changed many other aspects of daily life. Here are some of the most important things we’ve learned so far.

The Internet Is Safer Than We Thought

In the early days of the World Wide Web, we worried that bad online advice might hurt patients. We now know this harm rarely happens. The most comprehensive review of the medical literature done to date, conducted by a distinguished international team including Reviewer Alejandro Jadad of the University of Toronto, found that for the first decade of the Net’s existence, only a single case study of a death that the authors thought might be attributed, at least in part, to the Net, had been reported in the entire international medical literature.

By contrast, a 2020 study estimated that 195,000 patients annually are killed by medical mistakes in U.S. hospitals alone.” The researchers concluded:

Medical errors seem to be the elephant in the room that no one wants to acknowledge or talk about. The lack of recognition and acknowledgment of the seriousness and urgency of the problem fosters a culture of denial and complacency.

Readers must come to their own conclusions about whether adopting the traditional passive-patient role and putting themselves in the hands of randomly chosen medical professionals might indeed be—as this data appears to suggest—a million times more dangerous than attempting to learn about their medical conditions on the Internet. I do not presume to speak for my colleagues on this contentious point. But I must admit that when I hear about clinicians who still warn their patients of the dangers of seeking health information online, a verse from deep in my childhood comes to mind:

… how can you say to your neighbor, “Friend, let me take out the speck in your eye,” when you yourself do not see the log in your own eye? You hypocrite, first take the log out of your own eye…

New Resources for Online Quality Control

Why did so many clinicians come to judge the Internet to be so much more dangerous than it is? The closed IT systems—hospital systems that report lab values, electronic medical records, medical billing systems, etc.—that most clinicians know and use may be the reason. Closed information systems use supply-side quality control measures: only the duly authorized can add or edit information. The accuracy and currency of data is closely monitored. And there is little or no tolerance for error. Thus the data in closed systems is assumed to be universally accurate.

By contrast, the Internet is an open information system. Anyone can publish anything he or she wishes. The Internet, therefore, predictably includes a good deal of inaccurate and outdated information. And while this limitation would be a major defect in a closed information system, in the Internet’s open information system, it is merely a given characteristic.

Open systems use demand-side quality control measures. Sophisticated Net users understand that one does not accept any random bit of online information with the same degree of unquestioning acceptance that might be appropriate when checking one’s bank balance or looking up a patient lab value.

For even though the Internet contains some inaccurate, outdated, biased, and blatantly self-serving information, it also offers the tools users need to tell the good from the bad. How can patients safely navigate their way through the neighborhoods of self-help cyberspace? We’ve already discussed several resources that help them: their own good common sense, their accumulated experience, their ability to compare a number of different sources, and their access to person-to-person guidance from friends and family, patient-centered networks, online patient helpers, online support groups, and Type 1 and (if they are lucky) Type 2 clinicians.

Communityware as a New Online Quality Resource

Pattie Maes, one of the most highly regarded IT visionaries and developer of the software agent,v has defined a whole new category of online quality-control systems that she calls communityware. In addition to helping loose networks of users find the high-quality online information they seek, communityware systems help online networks accomplish tasks that formerly required centralized control and coordination, for example, choosing the best doctor or medical center or deciding on the best disease treatment options. Communityware systems, Maes notes, are driving a major powershift from established authorities to online communities, not just in medicine, but also in every sector of society. The software systems that enable online support communities—listservs, Web forums, chat rooms, etc.—are communityware. The proprietary PageRank algorithm used by Goggle to identify Karen Parles’ site as the number one resource for lung cancer is communityware. Another type of communityware consists of online databases that provide user reviews and ratings (for examples, see the seller reputation databases at ebay.com, the product reviews at www.amazon.com, and the patient drug reviews at www.askapatient.com).

All established and emerging modes of electronic communication can also serve as platforms for communityware: e-mail networks, weblogs, instant messaging, text messaging, moblogging, wikis, P2P and group-forming technologies, social networks, and more. And many of us have been impressed by a particularly promising new type of medical communityware called the Problem-Knowledge Coupler (www.pkc.com) that is discussed in the next chapter.

From Patient Education Materials to Self-directed Learning

We’ve learned that the old, spoon-feeding model of patient education in which clinicians provide patients with all the medical information they need—no discussion required—no longer applies. Don’t get us wrong. e-Patients are delighted to receive all the medical information their clinicians may care to provide. But they no longer think of clinicians as the ultimate and unquestioned authority. They know that we are limited and fallible, as indeed we are. So no matter what we give them, they will, appropriately, wish to consult other sources as well. And once they have done so, they want to be able to discuss their options with us, ideally, at their convenience, by telephone or e-mail.

Health professionals who complain that they haven’t time for such discussions and insist their patients have no need to search the Internet because they, their clinicians, can provide all the medical information they need, have simply not been paying attention. And those who argue this position while refusing to exchange e-mail with their own patients miss the point entirely.

Again, no matter what we call it, the notion of top-down, fixed packets of “patient education materials” provided by clinicians and designed to meet all of a patient’s informational needs—with no further discussion or interaction—is obsolete. Patient education in the 21st century means two-way clinician-patient interaction. Or it means nothing at all.

We’ve learned that e-patients no longer want such one-way information packets. They no longer want doctor’s orders. They no longer want professional interventions. They want their clinicians to help them learn about their disorders and take control of their illnesses—in their own way and in their own time. And whenever possible, by e-mail.

They want choices, tools, access, and encouragement. They want invitations to collaborate and innovate. They want emotional and technical support. And if their process does not fit neatly into our accustomed intellectual constructs, clinical routines, and familiar modes of practice, this problem is ours, not theirs.

Join Them, Don’t Fight Them

A century ago, a physician’s brain could hold all the medical information required to treat most known medical conditions: there was not all that much to remember. Yet over the past century, medical information has increased exponentially. Unfortunately, as Advisor Richard Rockefeller has so eloquently observed, the capacity of the human brain to store and process it has not. As Donald Lindberg, director of the National Library of Medicine explains, “If I read and memorized two medical journal articles every night, by the end of a year I’d be 400 years behind.”

It is simply no longer possible for medical professionals to stay completely up-to-date on a large number of medical conditions—it is hard enough to do so for one or two. So unless an e-patient has been lucky enough to find a clinician who is a full-time specialist in their disorder (and as we will see, a surprising number have done exactly that) most will be able to learn more from the Internet than they can in their doctor’s office. And many will end up knowing more than their clinicians about some aspects of their conditions.

Understanding this change on an intellectual level is one thing. It is quite another to sit face-to-face with a patient newly diagnosed with a rare condition who has spent 200 hours reading up on it—while you only dimly remember reading about it in medical school many years before.

We must admit that in the beginning, dealing with patients who knew more than we did about their conditions—or at least some aspects of their conditions—came as something of a shock. So we can understand and sympathize with colleagues who may be tempted to discount their patients’ new abilities, to disparage their information sources, and to attempt to reassert their medical authority. We can also assure them that this is neither possible nor necessary. For in the end, we’re confident that they will conclude, as we have, that informed, activist e-patients are a blessing, not a curse.

Well-informed patients can take more responsibility for their care. They can assume many of the time-consuming responsibilities that currently fall to clinicians and other medical staffers: They can take their own vital signs during a clinical visit. They can test their own urine using a simple dipstick. They can examine their kids’ sore ears with a home otoscope, listen to their own hearts to detect arrhythmias, and can safely choose and self-administer many drugs currently available only by prescription.

They can help us avoid, detect, and manage medical mistakes. They can help us keep up-to-date on their conditions’ latest developments. They can, in short, become our colleagues and our partners, not mindless robots programmed to obey doctors’ orders without question or comprehension. To continue to prohibit patients from making all the healthcare contributions they can make is an irrational exercise in futility and dysfunctionality.

Inviting patients to join us in such provider-patient partnerships can be especially valuable with patients who are health professionals, are highly motivated, have rare disorders or complex problems, or with have conditions that we rarely or never see. And we have learned that most patients will respond most favorably to such invitations.

Learning to Appreciate Patient-developed Online Resources

We’ve also learned that some of the best online resources for patients are those developed by patients. When Advisor John Grohol and I reviewed the 16 medical Web sites that received his coveted “Grohol Top Site” award to date, we found that ten (62.5 percent) had been produced by e-patients while five (31.25 percent) had been developed by health professionals. (One site had been developed by a group of artists and researchers working at Xerox PARC and didn’t fit into either category.)

Clearly both patients and professionals are capable of providing useful and reliable online resources. Patient-helper Samantha Scolamiero, founder of the BRAINTMR mailing list, says:

List members—lay folk and professionals alike—have moved beyond the old, obsolete mindset that holds that only certain ‘qualified’ medical professionals may create and disseminate medical information. We lay folk are learning that we are qualified through our experience, our knowledge, and our concern. We now see that we are capable of contributions no professional can make and that by linking our efforts [with those of professionals] in a coordinated team, we can advance the well-being of all.

Why e-Health Developers Should Start in Self-Help Cyberspace

We’ve recognized the difference between forcing our patients to relate to us on our familiar professional turf—only offering face-to-face consultations in the clinic, using closed IT systems, or participating in old, professionally centered research methods—to going to meet them on their turf—visiting their Web pages and online communities, exchanging unstructured e-mail, and inviting them to co- direct our research efforts by listening to their stories in their own words. In short, we’ve learned that it is best to engage with them when and where and as they choose rather than when and where and as we have habitually been willing to interact with them.

As we saw in Chapter Six, Lester, Hoch, and their colleagues did not begin their groundbreaking project by making patients come to their offices at Massachusetts General Hospital. They began by observing what the best and brightest e-patients with neurological disorders were already doing in self-help cyberspace—then offering to help them do it more effectively. As a result, they were able to make one of the most important contributions to the new online health commons that any health professionals have made to date—on an almost absurdly small budget. Alan and Cheryl Greene did not launch their groundbreaking project by dreaming up some technically sophisticated “consumer health informatics” system that they thought pediatric patients might need. They simply asked their patients what they wanted—and gave them what they asked for. These visionary e-health developers succeeded because they ventured into self-help cyberspace and built their systems according the principles of 21st-century patient-driven care.

In their book, The Support Economy, Shoshana Zuboff and James Maxmin draw an important distinction between the traditional realm of professional/ organization space and the rapidly evolving realm of individual/consumer space in the emerging information age economy. Within the healthcare context, the former term refers to the physical and psychological space within which nearly all episodes of 20th century healthcare took place, usually the clinic and the hospital.

Professional/organization space is the clinician and researcher’s turf and the medical professional’s realm of control. When operating in professional/ organization space, patients must conform to their clinicians’ “rules, procedures, standards, requirements, and schedules, not the other way around.”

All this may feel perfectly comfortable and convenient to professionals. But the authors warn that the experiences, capabilities, and needs of our first generation of e-patients has changed so dramatically that professionals who still insist on operating exclusively within organizational space simply “do not have the capacity to perceive [these] individuals’ [new reality], let alone to respond to their [real] needs… even as a perpetual drumbeat of consumer-oriented rhetoric claims the opposite.”

Individual/consumer space is another name for what we’ve been calling self-help cyberspace, the domain within which autonomous e-patients find and use the online health resources of their choice—friends and family, patient-helpers, online communities, medical Web sites, and unstructured e-mail with Type 1 and Type 2 clinicians. But individual/consumer space lies within the e-patient’s realm of control. Zuboff and Maxmin define individual/consumer space as follows:

Individual space is always the individual’s space, and the choices that exist there are always the individual’s choices. Each individual sets the rules of relationship… [and] these relationships cannot be constructed mechanistically. They are necessarily intimate and authentic. They build overtime, based on mutual respect and interpersonal trust.

We’ve learned that the most successful online health resources—that is to say, those that e-patients like best—all operate in individual/consumer space. Most have been developed by patients, and those developed by professionals have relied on high-level patient guidance. The reason these systems work so well is that they liberate patients from the confining, limited, and subservient roles they are typically forced to play when interacting with clinicians in organizational/ professional space. For although such limits are often invisible to health professionals, most patients feel intensely uncomfortable when forced to operate within the limited and subservient roles allowed them in most contemporary healthcare institutions.

And even though some e-health developers still believe that it may be possible to develop systems that genuinely empower patients while remaining within the comfortable and familiar borders of professional/organization space, Zuboff and Maxmin suggest such efforts are doomed to failure. By way of explanation, they cite a story about a man who is looking for his keys under a streetlight instead of in the dark alley where he lost them because “this is where the light is.” They write that, like the man in the parable:

 …we professionals like to look for solutions within our accustomed organizational/professional space, even though the real area of concern is not to be found there, but in what seems to us the darkness, but in reality is simply a different world, the world of individual/community space…. [But] consumer solutions developed in organization space merely perpetuate the fundamental problems whose symptoms they seek to address… When we face discontinuity, the answers we seek cannot be found under the light from the lamppost. They are not here, but over there—in the darkness, thought the looking glass, on the other side of what [we think we know].

Old Wine in New Bottles

In conclusion, we have learned that there is only one effective strategy for building IT systems that genuinely empower patients: patients must be the developers or co- developers of such projects. In the classic analysis of a pernicious and pervasive type of IT-based medical paternalism that still frustrates many such efforts, the late Diana Forsythe, a medical anthropologist at the University of California-San Francisco School of Medicine, described what happened when a distinguished team of medical informatics professionals set out to build a computerized patient-education system for migraine patients—without seeking the guidance and collaboration of those patients. Her paper, “New Bottles, Old Wine: Hidden Cultural Assumptions in a Computerized  Explanation System for Migraine Sufferers,” should be inscribed on the cubicle walls of all medical informatics professionals.

The senior developers assumed that the patients knew nothing about migraines, that what they needed to know was what their doctors wished to tell them, and that the only reason their clinicians hadn’t provided the information the patients wanted was a lack of time. Forsythe found that all of these hypotheses were mistaken:

  • Many of the migraine patients had read extensively in the medical literature and most knew a great deal about their condition, including many things that their physicians did not know, e.g., exactly what happened during a migraine attack and the best practical ways to manage it. She concluded that some of the information the patients needed could best be supplied by other patients.
  • The things the patients wanted to know were radically different from the things the neurologists wanted to tell them. Forsythe identified a long list of “nonmedical” questions that concerned patients greatly, but which were systematically ignored by the neurologists, e.g., the possible connection between their headaches and the domestic violence several patients had reported.
  • The neurologists’ failure to answer their patients’ questions had little to do with time. They offered little explanatory material and dominated the patient interviews so completely that patients had few opportunities to ask questions.
  • When patients attempted “… to bring up concerns that their neurologists do not see as strictly medical, physicians often appear not to ‘hear’ them, ignored or disregarded all but a few “approved” types of questions.

The developers disregarded Forsythe’s advice in much the same way the neurologists had disregarded their patients “inappropriate” questions. The physicians’ perspective, and no one else’s, was included. The developers remained convinced that in spite of impressive evidence to the contrary, while the patients had much to learn from their neurologists, the clinicians had nothing at all to learn from their patients.

The final system presented the physicians’ knowledge of migraines to patients, but did not allow patients to present to professionals the many things they knew that the clinicians did not. It did not allow patients to raise the questions the clinicians habitually ignored. It did not permit patients to describe their experiences in their own words, but  forced them to choose from a limited list of doctor-defined symptoms. It did not permit patients to inform other patients on topics they knew more about than the doctors. And it did not permit patients to provide the doctors with feedback on their services or to tell the doctors other things they felt their doctors they should know.

Forsythe also observed that even though several non-physician members of the development team suffered from migraines and, in private interviews, these patient- professionals agreed with the patient-subjects and “…had eloquently described to me the pain and fear of their migraine attacks and the failure of biomedicine to provide them with significant relief, a story heard over and over again in the course of the study…”, they never spoke up about their personal experiences and convictions during team meetings, but habitually deferred to the physicians, neither of whom had ever experienced migraines.

You might also find the following blog post interesting - How To Get What You Want From Health Care Workers (1980)

Like many of their medical informatics colleagues, the developers were unwilling to address any of the important issues of knowledge, power, and control among doctors, non-physician health professionals, and patients that are vital to building effective IT systems, even though Forsythe raised these issues repeatedly.  The experiences, knowledge, needs, and guidance of the patients were completely ignored. So in the end, like many of their medical informatics colleagues, the developers built a technically impressive system that reflected all the same profound defects of medical care the patients had experienced repeatedly in the clinic.

Related Blog Post - ADHD Treatments, Discipline, and Lifestyle

Forsythe’s untimely death—she was swept away by a raging river while mountaineering in Alaska—deprived us of one of the most insightful observers in the emerging realm of e-patient studies. But we can honor her memory by remembering what she taught us. Even though many IT developers think of the systems they build as objective, neutral, and unbiased examples of technology, medical IT systems are always shaped by powerful cultural assumptions about healthcare—whether the developers are aware of this or not.

“In practice,” she once confided, late at night after a long AMIA meeting, “Most medicine is ten parts culture to every one part real science. As strange as it may seem, from a medical anthropologist’s point of view, much of what happens in the healthcare system we’ve all grown up with is invisible to most clinicians. Imagine that.”

Related Blog Post - Understanding Hospice Care