Table of Contents
Plain-language article summaries
The Issue in Brief
Evaluation of an Enhanced Primary Care Team Model to Improve Diabetes Care
Joseph R. Herges and colleagues
Background Researchers from the Mayo Clinic created and implemented an Enhanced Primary Care Diabetes (EPCD) model, which includes nurses, pharmacists and social workers on the medical team. They used a retrospective, interrupted time series design to evaluate the model’s impact on reported quality measures of diabetes care, known as the D5. Quality measures include indicators of glycemic control, blood pressure, cholesterol, abstinence from tobacco, and aspirin use. The team examined the EHRs from 32 primary care practices in an integrated health care delivery system that cares for adults 18-75 years with diabetes. This included staff clinician (physician and Advanced Practice Provider) practices with access to the EPCD model (5,761 patients); resident physician practices with access to EPCD (1,887 patients); and staff clinician practices without access EPCD (10,079 patients). The team compared primary outcomes in the percentage of patients meeting the D5 between pre-implementation (7 months) to post-implementation (10 months).
What This Study Found After EPCD implementation, staff clinician practices saw a significant improvement
in meeting the D5 (P=0.01). Trends in D5 achievement did not change among the resident
physician practices with access to EPCD (P= 0.14) and worsened among staff clinician
practices without access to EPCD (P= 0.05). In the direct pre-to-post-implementation
analysis, D5 indicators improved more in the EPCD-staff group and EPCD-trainee groups
compared to the non-EPCD group (IRR= 0.017; P< 0.001 and IRR= 0.017; P= 0.026 respectively).
Implications
- Implementation of the EPCD model was associated with improved diabetes care quality.
Nadia Sourial and colleagues
Background Researchers compared emergency department use and hospitalizations in people with dementia receiving care from an interprofessional primary care team and people seeking care from a physician-only group practice. A population-based, repeated cohort study design was used to extract yearly cohorts of 95,323 community-dwelling persons in Ontario, Canada, newly identified in administrative data with dementia between April 1, 2005 to March 31, 2015. Patient enrollment to an interprofessional or physician-only practice was determined at the time of dementia diagnosis.
What This Study Found Interprofessional primary care was associated with increased ED use but not hospitalizations in people recently diagnosed with dementia as compared to those receiving care from a physician-only primary care practice. The researchers found no evidence of an association between interprofessional primary care and hospitalization outcomes.
Implications
- This unexpected finding regarding the increase in urgent and non-urgent ER visits raises concerns about the possible impact of interdisciplinary teams on ER visit volume. However, the authors point out that it’s possible the observed increase in ER visits was appropriate. While interprofessional primary care may be well-suited to manage the growing and complex dementia population, a better understanding of optimal characteristics of team-based care and reasons leading to acute care episodes for persons with dementia is needed.
Bethany Shorey Fennell and colleagues
Background Researchers conducted a study examining the differences in smoking Quitline treatment enrollment, engagement and smoking cessation outcomes among almost 219,000 Spanish- and English-preferring primary care patients linked with evidence-based tobacco treatment using Ask-Advise-Connection (AAC). The AAC program was implemented between April 2013 and February 2016 in a large safety-net health system to proactively connect smokers with treatment via a link in the electronic health record. Rates of treatment enrollment, engagement, acceptance of nicotine replacement therapy (NRT) and smoking abstinence (self-reported and biochemically confirmed) were compared at six months among patients who received treatment in Spanish and English.
What This Study Found The team assessed and recorded in the EHR the smoking status of 218,915 patients. Smoking prevalence was 8.4% among Spanish- and 27.0% among English-preferring patients. Spanish-preferring patients were less likely to enroll in treatment (10.7% vs. 12.0%), yet completed more counseling calls once enrolled (median=2 vs. 1; P<.0001). Patients who received treatment in Spanish (vs. English) were twice as likely to be abstinent at six months (self-reported: 25.1% vs. 14.5%; biochemically confirmed: 7.6% vs. 3.7%). Receipt of NRT increased abstinence for all patients. Language did not interact with NRT to predict abstinence.
Implications
- Automated point-of-care approaches such as AAC have great potential
to reach Spanish-preferring smokers, who seem to appreciate treatment by those who speak concordant language.
Lucinda S. Bertels and colleagues
Background Researchers assessed individuals’ reasons for not undergoing a colonoscopy after receiving an unfavorable Fecal Immunochemical Test (FIT) in a screening program to detect bowel cancer. Colonoscopy is recommended for people who receive a positive FIT test, as they have an increased risk for bowel cancer, but not all people follow up. Based on previous literature, 12.2% (N = 8,654) of FIT positives did not undergo a colonoscopy for unknown reasons. The research team distributed a nationwide, cross-sectional questionnaire to 4,009 individuals who participated in the Dutch bowel cancer screening program and received an unfavorable FIT test. For those who did not get a follow-up colonoscopy, researchers asked additional ‘agree/disagree’ questions which were based on interviews conducted in a previous study. These included, ‘I trust in my ability to self-detect bowel cancer’; ‘I think my family physician would want me to undergo the internal bowel examination’; and how often they thought about their risk of getting bowel cancer after receiving an unfavorable test result.
What This Study Found Of 2,225 respondents (56% response rate), 730 (32.8%) received no colonoscopy. Fifty-five percent of people who did not follow up (404) reported a medical reason that they did not receive a colonoscopy. Other reasons for receiving no follow-up colonoscopy included difficulty deciding to get the procedure; lack of opportunity to discuss the FIT outcome with a medical provider; and perception of having a low estimated risk for CRC. Some individuals did not follow up because they felt there was an alternative explanation for the unfavorable FIT; had trust in their own ability to self-detect CRC; and thought that polyp removal was ineffective. Other patient barriers to follow up included the perception that the colonoscopy would be uncomfortable; poor overall health behaviors; and the belief that the test result was incorrect. The belief that the family physician would support colonoscopy showed the strongest positive association with follow up. Overall, participants who receive an unfavorable screening outcome may benefit from personalized screening and the ability to speak to a physician about the FIT result.
Implications
- Researchers suggest that personalized screening counseling may help and that family practitioners should be involved. Understanding the reasons why people do not follow up with their doctors has global implications for all people at risk of CRC as it is best treated early.
Measuring the Value Functions of Primary Care: Physician-Level Continuity of Care Quality Measure
Mingliang Dai and colleagues
Background Continuity of care is foundational to the clinician/patient relationship in the family
medicine context; however, few metrics exist which allow researchers to track continuity
of care as a measurement of clinical quality. One metric developed by the American
Board of Family Medicine (AFBM) provides researchers with a means of measuring clinical
quality in the primary care setting. Called the Physician Level Continuity of Care
Measure (or CoC measure for short), the measure only recently received full National
Quality Forum (NQF) endorsement. In this paper, researchers present the results of
rigorous validity and reliability testing for the CoC measure required for NQF endorsement.
Researchers established a retrospective cohort of patients with two primary care visits
during a 12-month period (7/1/2018-6/30/2019), identified through claims data registered
in the Optum Clinformatics Data Mart. Using the Bice-Boxerman Index, researchers then
calculated how many different PCPs were seen and how many times for each patient.
The index also accounts for physician referrals, wherein visits to physicians based
on a referral are attributed to the referring physician.
What This Study Found The CoC clinical quality measure meets validity and reliability requirements for implementation in primary care payment and accountability. The reliability testing for the entire population of physicians produced a mean reliability of 0.85, a median reliability of 0.95, and a minimum reliability of 0.27. These results of reliability testing suggest that the CoC measure meets or exceeds acceptable criteria based on the majority of reliability scores above 0.7. Additionally, the researchers found that when physician CoC scores increased, the number of patients with any ED visits decreased, further suggesting empirical validity and utility of this measure.
Implications
- ABFM’s CoC clinical quality measure meets validity and reliability requirements for
implementation in primary care payment and accountability. Care continuity is correlated
with improved primary care outcomes and has implications for care access. The authors
argue that conducting research on this validated measure of continuity could help
shift physician and health system behavior to support care continuity.
Tristen L. Hall and colleagues
Background Diabetes affects approximately 34 million Americans, and many do not achieve
glycemic targets. Continuous glucose monitoring (CGM) is associated with improved
health outcomes for patients with diabetes. Most adults with diabetes receive care
for their diabetes in primary care practices, where CGM adoption has been slow. Researchers
used a cross-sectional web-based survey to assess CGM prescribing behaviors and resource
needs among U.S. primary care clinicians. The authors conducted descriptive statistics
and multivariable regression to identify characteristics associated with prescribing
behaviors; openness to prescribing CGM; and to understand resources needed to support
CGM use in primary care.
What This Study Found Clinicians located more than 40 miles from the nearest endocrinologist’s office were more likely to have prescribed CGM and to have more favorable attitudes toward prescribing the devices in the future compared to clinicians located within 10 miles of an endocrinologist. Clinicians who served more Medicare patients reported more favorable attitudes toward future prescribing and higher confidence in using CGM to manage diabetes than those with fewer Medicare patients. Primary care doctors needed consultation on insurance issues and CGM training in order to best support their prescribing CGMs to their patients.
Implications
- Primary care clinicians are interested in using CGM for patients with diabetes, but
many lack the resources to do so. CGM use can be supported through clinician education via workshops. Consultation on insurance issues should be targeted toward residents and practices without a nearby endocrinologist. The authors also advocate for continued expansion of Medicare and Medicaid coverage for CGMs.
External Validation of the COVID-NoLab and COVID-SimpleLab Prognostic Tools
Mark H. Ebell and colleagues
Background Researchers developed and externally validated two simple risk score models to help clinicians more accurately estimate a hospitalized patient’s COVID-19 mortality risk. The two models are the COVID-NoLab and COVID-SimpleLab. In collaboration with the Biobanque Québécoise de la COVID-19, researchers established a retrospective cohort of patients receiving care for COVID-19 in 11 hospitals between March 8, 2020 and February 22, 2021. Researchers compared mortality accuracy in patients to evaluate prediction stability of the models over time. A validation cohort of 588 of these patient records were used to measure the predictive accuracy of the COVID-NoLab model. Another cohort of 479 were used to validate the COVID-SimpleLab model. Each patient in the retrospective cohort was categorized as low-, moderate-, or high-risk using the original cutoffs identified from both the COVID-NoLab and COVID-SimpleLab models.
What This Study Found The two scoring systems were found to predict patient mortality rates with reasonable accuracy, offering clinicians a reliable way to quickly identify low-risk patients who could potentially be managed as outpatients in the event of a bed shortage. Mortality in the low-, moderate-, and high-risk groups for COVID-NoLab were 1.1%, 9.6% and 21.2%, respectively. For COVID-SimpleLab, mortality rates were 0.0%, 9.8% and 20.0%, respectively.
Implications
- These prediction models may help decrease unnecessary hospital admissions during COVID-19
surges. The authors assert that it will be important to continue testing the models
over time, particularly in an outpatient setting, to ensure accuracy as new COVID
variants emerge. The two risk score models are available at and https://ebell-projects.shinyapps.io/COVID-SimpleLab/
Leveraging Free-Form Comments to Assess and Improve Patient Satisfaction
James Kroes and colleagues
Background Researchers created a spreadsheet tool called the HCP Review Comment Analysis Tool to analyze large comment samples, specifically to identify themes within patient comments and to measure thematic relationships with patient satisfaction. The tool was tested using 721 patient comments collected in 2019 by the Family Medicine Residency of Idaho (FMRI).
What This Study Found The tool allows users to analyze up to 5,000 comments and gives providers without
qualitative research skills or tools the ability to easily analyze patient comments
and identify actionable measures of patient satisfaction. Additionally, the tool allows
researchers to reduce vast sets of comment text into numerical data, which the authors
say is well suited for quantitative analyses. To use the tool, users simply paste
in patient comments. The tool then automatically measures the prevalence of each theme’s
keywords.
Implications
- Measuring how comments reflect themes provides insights ranging from specific, addressable actions by health care practitioners to breakdowns of structural processes. The authors assert that the tool can be used to measure other aspects of a comprehensive quality program beyond patient satisfaction.
Donovan T. Maust and colleagues
Background Despite the continuing growth of benzodiazepine (BZD)-related overdoses, BZD prescription rates have held constant. Little is known about whether a doctor’s own beliefs about BZD prescribing and the potential harms to patients is associated with how much BZD prescribing they actually do. Using a Medicare database, researchers identified primary care providers who had prescribed a BZD in 2017 and surveyed a random national sample of 100 doctors on their attitudes around BZD prescribing.
What This Study Found Among the 61 respondents, they prescribed a BZD to 11.5% to their respective patient panels. Approximately 62% of clinician respondents reported that they disagreed or strongly disagreed with the statement, “If a patient has been prescribed a benzodiazepine for years, the potential harms from continuing the benzodiazepine are low,” while 18.0% agreed or strongly agreed. When clinicians believed that the potential harms from continuing BZD were low for patients already prescribed the drug for years, they were more likely to have prescribed a BZD to patients in their panel. Other questions about physician beliefs about harms from BZD use were not correlated with BZD prescriptions.
Implications
Competencies for the Use of Artificial Intelligence in Primary Care
Winston Liaw and colleagues
Background Family Medicine professional associations have sought to identify how Artificial Intelligence
(AI) can support primary care. The associations have launched initiatives that bring
together AI experts and primary care clinicians to tackle these challenges. While
training has attempted to facilitate technology-enabled primary care, more work is
needed. In the
United States, family medicine professional milestones call for residents to use information
technology. Study authors note that without proper training, AI could impose harms,
exacerbate health care inequalities, worsen patient outcomes and population health,
and degrade clinician well-being.
What This Study Found Researchers propose six domains of competency for the effective deployment of AI-based tools in the primary care setting: 1) foundational knowledge (what is this tool?), 2) critical appraisal (should I use this tool?), 3) medical decision making (when should I use this tool?), 4) technical use (how do I use this tool?), 5) patient communication (how should I communicate with patients regarding the use of this tool?) and 6) awareness of unintended consequences (what are the “side effects” of this tool?).
Implications
Abigail Driscoll
Background Medical school student Abigail Driscoll writes a personal reflection on her first day of family medicine rotation in a resident-run clinic situated in a diverse, historically underserved community in Minneapolis. Though she was instructed by a medical resident to keep her patient interactions brief, she found herself stopping to take time to bear witness to the grieving process of a patient who had come in for “mood problems.” She listened to the patient’s plea for help in face of the very recent loss of her husband, which had led to housing instability, loss of income, and the threat of losing a beloved dog.
What This Study Found Driscoll was able to relate to the patient’s life crises due to her own recent life experiences among her own family, including cancer, suicide, COVID-19 and dementia. Driscoll empathized with the patient, expressing sorrow, asking her about her family, and simply handing her a tissue. After 40 minutes, Driscoll felt she had done something good for the world. She would encounter many more patients with similar stories and lives that had been affected by broken systems that give rise to gun violence, lack of health literacy, accidental drug overdoses, food insecurity and barriers to health.
Implications
My Role in Health Care's Existential Crisis
Curtis Kommer
Background Curtis Kommer, M.D., medical director for Accord Hospice in Sedona, Arizona who is now contemplating retirement, reflects back on his 38-year career as a family physician.
What This Study Found Kommer realizes that while he was singularly focused on taking care of patients, he neglected to advocate for himself and his patients. He also neglected to push back against what he deems are an endless series of misguided policy decisions that have had adverse impact on the health and well-being of his patients. These same policy decisions have also made his job much more difficult and stressful. Those include being required to schedule patient appointments every 10 minutes; consulting with patients via telehealth; or, more specifically, denying a wife permission to see her dying husband in the hospital during the COVID-19 pandemic.
Implications
Primary Care Research Is Hard to Do During COVID-19: Challenges and Solutions
Jodi Summers Holtrop and colleagues
Background Holtrop and Davis write a commentary to surface common challenges for family medicine researchers working in primary care practices and completing research projects during the COVID-19 pandemic.
What This Study Found The authors describe three challenges to conducting primary care research during a pandemic, including: 1) having to potentially rework project plans and protocols in response to changes in the primary care landscape due to COVID; 2) Less staffing for research efforts, hiring challenges, and additional work required of existing employees; 3) Daily challenges of life during a pandemic, including caregiving activities, determining where and when to get vaccinated, addressing others who do not comply with mask mandates or choose not to be vaccinated, as well as increased inequities in lower-resourced communities. The authors outline strategies in conducting research in the present milieu including stopping the work temporarily; expanding the region served by the research; reducing expectations for research processes or outcomes; altering how research is conducted; altering the scope of the intervention to include a COVID-19 response; or pivoting to a new research effort within the overall scope of project goals. The pandemic also has provided researchers the opportunity for research to be conducted in a more efficient and less expensive way through the use of telehealth; has drawn attention to the social and behavioral factors in research; and serves as a call to remind researchers of what health care topics and issues they are passionate about.
Implications
Resculpting Professionalism for Equity and Accountability
Jessica P. Cerdena and colleagues;
Background A team of authors describes the ways nebulous definitions of “professionalism” imbue
White, cisgender, straight and able-bodied standards to police the boundaries of belonging
in medicine. They argue that within the current state of the medical profession, success
requires assimilation into a culture of White supremacy that harms minoritized trainees.
They assert that marginalized trainees remain ‘unfinished sculptures,’ forced to reach
much farther than a white medical coat hanger for affirmations of medical professionalism.
For the authors – an Italian-Chilean mother, Black woman, White lesbian woman, Black
man, and Black trans woman – it is not enough to show up on time, don white coats
and stethoscopes, and invest their knowledge and humanity to help patients. Often,
they must also polish their speech, ‘carve’ their bodies and dress, and sever parts
of their lives to meet the ‘professional’ expectations their mentors and patients
set for them.
What This Study Found Through this essay, the authors seek to re-sculpt professionalism in a way that centers patients and trainees usually kept at the margins to ensure the increasingly diverse physician workforce can effectively address the needs of patients often excluded from quality care. They propose new standards for professionalism and specific interactional examples of equity, which center the trainees typically on the margins.
Implications
Patients Deserve Great Service: The Waiting Room Concierge
Susan Lee
Background Staff at Stony Brook Primary Care collaborated with the Patient Family Advisory Council
(PFAC) to help them improve information about delays in visit and wait times in the
clinic. PFAC members created the “Waiting Room Concierge,” which provided waiting
patients with personalized service. The concierge person helps with patient check-in
and interacts with them by assisting with office navigation, form completion, patient
portal sign up and assistance for patients with mobility issues. They also monitor
wait times. The program also included a prominent notification board that indicated
whether a particular doctor was on time or running late.
What This Study Found After the implementation of the Waiting Room Concierge and several Plan-Do-Study-Act Cycles, the health system’s Clinician and Group Consumer Assessment of Healthcare Providers and Systems surveys in the two areas of Wait time in the Office and Communication About Delays showed improvement. Scores went from 60% to almost 90% of patients indicating top scores of satisfaction. Patients commented that they felt more respected by being informed and welcomed in the waiting room. Based on feedback from both patients, staff, and clinicians, the team is continuing to fine tune the role of the “waiting room concierge.”
Implications
Tobacco Cessation Champions: Recognizing Physicians Who Ask, Advise, and Refer
Niharika Khanna and colleagues;
Background University of Maryland Medical System researchers studied the effects of positive reinforcement to encourage physicians and health practitioners to promote referral to tobacco quitline resources. As part of the implementation, physicians who referred patients to the quitlines received positive reinforcement in the form of acknowledgement certificates. Before physicians and practitioners received letters and certificates, 203 e-referrals were submitted to the Maryland Quitline in the fourth quarter of 2021.
What This Study Found The researchers observed an uptick in referrals after physicians and health practitioners received letters and certificates acknowledging their referral efforts in January 2022.
Implications