I’m not sure if I’ve ever met Nordic Chief Medical Officer Craig Joseph, MD in person, but he’s definitely on my list of “people I’d like to have a cocktail with” at some point. His Twitter posts @CraigJoseph always have interesting tidbits, such as a recent white paper from ECRI’s Partnership for Health IT Patient Safety. He notes, “Lots of smart people with clinical and EHR vendor chops outline specific actions to consider.”
I checked out the paper, titled “Optimizing Health IT for Safe Integration of Behavioral Health and Primary Care.” It resonated with me because this is an issue I’ve had to deal with for years – navigating the intersection of those two disciplines while trying to coordinate care while maintaining privacy. Many organizations, including some of my current clients, choose to keep primary care and behavioral health records siloed. This results in fragmentation of care and lack of understanding around whole-person factors that drive both physical and mental health.
It lays out clear reasons why primary care and behavioral health need to be integrated:
- 80% of behavioral health patients will visit a primary care provider (PCP) at least annually.
- 50% of behavioral health disorders are treated in primary care settings.
- 48% of appointments for psychotropic medications are with non-psychiatrist PCPs.
- 67% of people with behavioral health disorders don’t get behavioral health treatment.
- 30-50% of patients referred from PCPs to outpatient behavioral health don’t keep their first appointment.
- Two-thirds of PCPs report being unable to access outpatient behavioral healthcare for their patients.
Additional barriers for mental healthcare access include provider shortages, health plan barriers, and coverage issues. In my major metropolitan area, we recently opened a dedicated mental health emergency department unit that is staffed full time by specialist providers. The community accepted it readily because we know we don’t do the best job for patients needing non-medical services who present to other care venues, such as the emergency department or urgent care facilities.
Even for health systems or provider-side organizations that want to try to integrate the behavioral health and primary care realms, EHRs aren’t always supportive. Psychiatry notes, therapy notes, and documentation from social workers are often kept under separate access where the primary medical team can’t see them. Especially when we’re dealing with medical conditions that can have significant behavioral components, it would be useful to be able to see all the information about the patient. The white paper does a nice job explaining different levels of integration, from “coordinated” care to that which is “co-located” to fully “integrated” care. Right now, many primary care practices are struggling to deliver even minimally coordinated care.
One of the major participants in the creation of the paper was the HIMSS Electronic Health Record Association (EHRA). I have some experience with EHRA from a past life and know many of the members of the project’s working group to be knowledgeable individuals with a deep understanding of EHRs and care delivery. EHRA has a code of conduct for EHR and health IT developers that addresses the need for collaboration described in the paper.
However, working as someone outside of an EHR vendor, I’ve found it nearly impossible to access the materials that we relied on when I was on the vendor side. This forces those of us who work on homegrown or in-house systems to re-invent the wheel trying to determine best practice as we develop our technology. Since this is a partnership with ECRI and this paper exists, I take that as positive signs. Still, non-commercial developers are going to have to do a lot of figuring out on their own unless there are maneuvers to standardize at the federal or payer levels.
The paper talks about standardizing screening and documentation tools so that data is consistent across an enterprise. Although this would be good, many patients may receive their care across multiple organizations. I cared for a patient the other day who receives medical care primarily through a county health clinic because she doesn’t have health insurance, but receives a telehealth benefit from her employer, so she’s using that for urgent care services and psychotherapy. She came to our urgent care because she needed stitches and we have an affordable self-pay program. Given the vast differences in the systems used by those entities coupled with the relative immaturity of our state’s HIE, there’s no way there will be coordination any time soon.
There are also legal barriers to sharing of data under both HIPAA and 42 CFR Part 2, especially around sensitive health information. Many organizations find these restrictions daunting and either don’t have the wherewithal or the manpower to try to tackle them, especially while simultaneously coping with a pandemic and the generalized dysfunction of healthcare delivery in the US. Patients also struggle to understand the protections and restrictions and become frustrated when we try to explain why we have to ask the patient to summarize their care because we can’t access the information that we need without recreating the proverbial wheel.
The document has some great appendices, including a literature review and tables of evidence used during its creation. The summaries of EHR challenges and existing workarounds were fascinating case studies in dysfunction: lack of integration between the EHR and tablet devices used for patient-completed screeners and surveys; copy and paste to add the same note to the PCP and behavioral health EHRs; printing and scanning of medication lists from the behavioral health EHR to the medical one; and more. My favorite is “Reliance on patient or clinical recall for inaccessible clinical information – providers describe this as ‘flying blind.’”
Due to my employer’s lack of integration with our state HIE or nearby health systems, I’ve been flying blind for the last six years, except for when patients use their phones to access MyChart and then hand them over. That’s been useful in a number of medical situations, but I have to admit I’ve never seen psychiatry notes or therapy notes in any of those encounters, and I usually can’t see a full medication list history to know what’s been tried in the past — only current medications are typically displayed.
The paper also contains pages of recommendation tables, some of which push back on ONC, CMS, and other agencies to provide easily accessible standards for developers to use when creating documentation. It also calls on ONC to drive adoption and implementation guidance for APIs to improve integration. There is also a bid for the federal government to incentivize patient care organizations to implement standardized tools. There’s a great swim-lane diagram of an ideal IT-enabled workflow for safe integration. It will be interesting to look back at this paper in a couple of years and see how far we’ve come or whether we’re still living in the land of siloed documentation.
Has your organization done work to support integration of primary care and behavioral health? Leave a comment or email me.
Email Dr. Jayne.