As mentioned in previous blogs, I volunteer as medical director of a free clinic for adults without health insurance. As an interfaith-based project, we’re supported by several local churches, a synagogue, a mosque, and individual donations. Culmore Clinic (www.culmoreclinic.org) is currently housed on the second floor of Columbia Baptist Church in Falls Church, VA. On Tuesday and Thursday mornings, the church classrooms are converted to exam rooms, an office, nurses’ station, conference room, counseling rooms, and returned to their usual configuration in the evening by a group of volunteers from Goodwin House, a nearby retirement community. Since I joined five years ago, Culmore Clinic has expanded from once to twice a week with two providers (a volunteer retired internist for a half day and a nurse practitioner for a full day) to currently three part-time NPs, a PA who also acts as executive director, a counseling team of clinical social workers, a physical therapist, and another volunteer internist. We now coordinate clinical services with a dozen volunteer nurses, a paid full time nursing director, NovaSalud HIV/hepatitis screening group, Fairfax women’s clinic, George Mason University NP student training practice, and several specialty referral centers including the University of Virginia, Inova Health System, NIH, and the Northern Virginia Medical Society telemedicine pilot program. We import data from a hospital lab system and several group radiology practices. We also have interfaith prayer ministry, interpreter service, patient education, and ombudsmen/navigators for local community health resources. The community we serve is almost entirely immigrants, most of whom work jobs that do not qualify for health insurance. We have 20-30 office visits per clinic day with efficiency limited by the necessity of language interpretation, and our patients’ multiple morbidities and lack of previous primary care.
When I accepted the position, I set a goal of implementing an electronic medical record (EMR) system. Since very few of our patients qualify for federal funding, we could not take advantage of the CMS incentive program to adopt an EMR, and we could not afford the communications hardware necessary to use Inova’s EpicCare EMR system. We decided on the web-based PracticeFusion (PF) system because we do not have the resources to maintain a server, it is free of charge to the practice, and it has most of the features we wanted in an EMR. We did receive a grant from the Phillip L. Graham Foundation which we used to buy computer and communications hardware, additional software for staff and volunteer administrative management, and to pay local health informatics consultants to help in the transition from a homegrown paper-based system including staff and provider training. Most of our patients have cell phones with only voice and text, and have internet access only through the public library. Many have very limited education and literacy even in their native languages.
The terms of the grant require us to show that the funds have improved at least the process if not the outcome of health care for our patients. We selected diabetes, hypertension, and depression – the three most common chronic conditions in our population – as targets to measure our performance. We chose as outcome variables HgbA1C, arterial blood pressure, and PHQ9 score and our study is still in progress.
Because PF does not allow users to customize the clinical decision support (CDS) reminders developed for CMS Meaningful Use, I used charting templates to implement CDS for our practice. PF has a charting template editing tool which I found quite easy to learn and there is a community of PF users that have already developed and published templates for various conditions.
I reviewed current clinical practice guidelines (CPG) for the three target conditions and found freely available online versions of the following:
Standards of Medical Care in Diabetes—2017 Abridged for Primary Care Providers at
I cut and pasted text and tables from the article to the appropriate sections of the office visit template so that providers seeing a patient with diabetes can enter and edit the recommended data into a progress note for each visit of our patients with diabetes.
I used a similar approach for hypertension using Veterans Administration/Department of Defense CPG https://www.healthquality.va.gov/guidelines/CD/htn/VADoDHTNClinicianSummary.pdf. Depression has been more problematic because I could not find any one widely used CPG for US primary care: I synthesized information from USPTF on screening https://www.guidelines.gov/summaries/summary/49978/screening-for-depression-in-adults-us-preventive-services-task-force-recommendation-statement?q=depression+primary+care; from SAMHSA on suicidal risk assessment https://store.samhsa.gov/shin/content/SMA09-4432/SMA09-4432.pdf; and from AHRQ on evidence-based therapy https://effectivehealthcare.ahrq.gov/ehc/products/568/2152/major-depressive-disorder-executive-151202.pdf for our Culmore Clinic depression template.
I have only self-reported data on the extent to which the templates I developed are used by our providers, and can at best use historical audits of our patients’ outcomes before and after. Our counselors can now enter notes in the EMR and see our notes on the patients’ medical problems. Despite the limitations noted above, our preliminary data shows significant decreases in HgBA1C. We have not yet analyzed any other outcome or process data. Since we are now able to do group visits, increased use of insulins, and nutrition education with some of our diabetics it will be difficult to isolate any effect of the templated CDS, but we all agree that our EMR adoption has greatly increased our ability to share data between medical providers, counselors, and nurses and to expand our services to the community.
Charles Sneiderman, MD, PhD, DABFP
Culmore Clinic, Falls Church, VA
Published on 7/27/17