While writing about health IT, I routinely find myself at legal conferences. Regulatory issues about patient privacy and safety arise everywhere health IT tries to have an impact, so people promoting change must keep in touch with policy-makers and lawyers in the health care area.
Thus I went this past Friday to Harvard for a one-day symposium, “Putting Patients at the Center of Research: Opportunities and Challenges for Ethical and Regulatory Oversight,” sponsored by Harvard’s Petrie-Flom Center.
Involving patients in patient care is a surprisingly recent concern. There was a time when doctors made all the decisions, delivering them as if they had come directly from the entrails of an oracular temple. Visitors were severely limited at hospitals, because family members just got in the way of the professional staff. And although the attitude toward engaging patients and their families has softened somewhat in health care, rigid boundaries still exist in research.
As project leader Joel Weissman pointed out at the beginning of the Petrie-Flom conference, patient rights weren’t considered by health care professionals until the 1980s, as outgrowths of the civil rights and women’s rights movements. Patient engagement languished still longer. It received a legal toehold in the 2010 Affordable Care Act, which set up the Patient-Centered Outcomes Research Institute. Although more researchers over the past eight years have warmed to the idea of engaging with patients in other ways than subjects of clinical trials, the Petrie-Flom conference highlighted how little progress we have made.
In a “nothing about us without us” era, it would seem odd to an outsider like me that patients should be excluded from the roles now being tentatively offered:
- Joining the research team in some capacity
- Recruiting subjects for trials and engaging the patient community
- Helping disseminate results
- Acting as consultants in some other way
But risks are certainly entailed by inserting non-professionals of any stripe into the research environment, so some criteria and processes need to be set up. Before filling non-traditional roles, patients should be required to undergo training in ethics, the science behind the study, and some of the methodology. There are particular risks when the patients have access to personally identifiable data. (I don’t see why this should ever be necessary, but the possibility was raised several times during the day.)
The panelists also cited conflicts of interest as a risk. Many researchers recruit engaged patients from the companies that make related drugs or other products, simply because those are easy places to recruit. This problem highlights the importance of casting a wide net and recruiting diverse populations as engaged patients. However, one could argue that merely suffering from the condition that the researchers are investigating leaves one with a conflict of interest: you want the research to produce a cure, so you may not be even-handed in your acceptance of negative results.
What spurred this conference? The Petrie-Flom Center and PCORI have spent the past academic year doing a study of patient-centered research, and recently published an article by a team led by Weissman. The center presented the results at Friday’s conference to an audience of some 80 members of the health care field and interested observers.
The study was narrow and intensive. It focused on the attitudes of those running Institutional Review Boards, which are notoriously conservative. Thus, in my opinion, the results focused on what was holding back patient-centered research rather than what was already working well. The process was quite drawn out: questionnaires sent to hundreds of medical schools, public health schools, and hospitals; six focus groups with an iterative process for evaluating recommendations; and a modified Delphi consensus process among 17 experts, including (of course) representative patients.
Respondents to the survey expressed strong support for patient-centered research, believing (at a rate of about 90%) that it would benefit patients and clinicians, as well as (at a rate of about 80%) researchers. Those IRBs who tried out patient-centered research were especially enthusiastic, likely to say that it improved the quality of research results.
But IRB heads also openly expressed confusion and frustration about the pressure to include patients in the “non-traditional” roles listed earlier. Some of their reactions were productive: for instance, large majorities of respondents called on the federal government to provide standards, guidelines, and training for patient engagement. But some of the immediate measures IRBs put in place were irrelevant and even counterproductive. For instance, some required patients to sign informed consent forms, even though these patients were not the subjects of trials and therefore had no reason to need to consent. As patient advocate Jane Perlmutter pointed out, patients in non-traditional roles don’t require protection but require training to ensure that they protect the subjects of the research.
Perlmutter emphasized the importance of financial compensation. Without it, researchers will recruit mostly unemployed patients with independent incomes. To reach out to multiple ethnic groups, age ranges, and economic strata, payment must be offered for the work performed.
Unfortunately, I didn’t see much at Friday’s conference about topics directly related to health IT, such as privacy and ownership of data. Researcher Luke Gelinas mentioned that patient-centered research is more likely to use sensors, networking, social media, and other modern technology than more traditional research, and that these raise issues of informed consent, privacy, and ownership of data.
On the whole, the Petrie-Flom researchers thought there was no need for a whole new approach. But they are working on several recommendations to improve the current situation. In summary, the takeaways I derived from the symposium include:
- The value of patient-centered research is widely appreciated, and its benefits have been demonstrated where it has been tried.
- However, progress implementing patient-centered research is slow.
- Training for patients in non-traditional roles is required, but not so much as to be daunting and make it difficult to participate.
- Researchers have not devoted enough effort to diversity.
- Governments can offer support in typical ways, such as setting standards and funding programs.
I also predict that the growth of patient-centered research will place additional strains on IT systems. Bringing in new team members in scattered environments will require multiple systems to interact without friction. Data will need to be segmented and released carefully to just the right people. Interfaces will have to be intuitive (if such a thing exists) and easy to use without much training and without risk of errors. So the field has its work cut out.