Jeffrey M. Ashburner

Defining Patient Complexity From the Primary Care Physician's Perspective: A Cohort Study

BACKGROUND Patients with complex health needs are increasingly the focus of health system redesign. OBJECTIVE To characterize complex patients, as defined by their primary care physicians (PCPs), and to compare this definition with other commonly used algorithms. DESIGN Cohort study. SETTING 1 hospital-based practice, 4 community health centers, and 7 private practices in a primary care network in the United States. PARTICIPANTS 40 physicians who reviewed a random sample of 120 of their own patients. MEASUREMENTS After excluding patients for whom they were not directly responsible, PCPs indicated which of their patients they considered complex. These patients were characterized, independent predictors of complexity were identified, and PCP-defined complexity was compared with 3 comorbidity-based methods (Charlson score, Higashi score, and a proprietary Centers for Medicare & Medicaid Services algorithm). RESULTS Physicians identified 1126 of their 4302 eligible patients (26.2%) as complex and assigned a mean of 2.2 domains of complexity per patient (median, 2.0 [interquartile range, 1 to 3]). Mental health and substance use were identified as major issues in younger complex patients, whereas medical decision making and care coordination predominated in older patients (P<0.001 for trends by decade). Major independent predictors of PCP-defined complexity (P<0.001) included age (probability of complexity increased from 14.8% to 19.8% with age increasing from 55 to 65 years), poorly controlled diabetes (from 12.7% to 47.6% if hemoglobin A1c level≥9%), use of antipsychotics (from 12.7% to 31.8%), alcohol-related diagnoses (from 12.9% to 27.4%), and inadequate insurance (from 12.5% to 19.2%). Classification agreement for complex patients ranged from 26.2% to 56.0% when PCP assignment was compared with each of the other methods. LIMITATION Results may not be generalizable to other primary care settings. CONCLUSION Primary care physicians identified approximately one quarter of their patients as complex. Medical, social, and behavioral factors all contributed to PCP-defined complexity. Physician-defined complexity had only modest agreement with 3 comorbidity-based algorithms. PRIMARY FUNDING SOURCE Partners Community Healthcare, Inc.

Relationship between patient panel characteristics and primary care physician clinical performance rankings.

CONTEXT Physicians have increasingly become the focus of clinical performance measurement. OBJECTIVE To investigate the relationship between patient panel characteristics and relative physician clinical performance rankings within a large academic primary care network. DESIGN, SETTING, AND PARTICIPANTS Cohort study using data from 125,303 adult patients who had visited any of the 9 hospital-affiliated practices or 4 community health centers between January 1, 2003, and December 31, 2005, (162 primary care physicians in 1 physician organization linked by a common electronic medical record system in Eastern Massachusetts) to determine changes in physician quality ranking based on an aggregate of Health Plan Employer and Data Information Set (HEDIS) measures after adjusting for practice site, visit frequency, and patient panel characteristics. MAIN OUTCOME MEASURES Composite physician clinical performance score based on 9 HEDIS quality measures (reported by percentile, with lower scores indicating higher quality). RESULTS Patients of primary care physicians in the top quality performance tertile compared with patients of primary care physicians in the bottom quality tertile were older (51.1 years [95% confidence interval {CI}, 49.6-52.6 years] vs 46.6 years [95% CI, 43.8-49.5 years], respectively; P < .001), had a higher number of comorbidities (0.91 [95% CI, 0.83-0.98] vs 0.80 [95% CI, 0.66-0.95]; P = .008), and made more frequent primary care practice visits (71.0% [95% CI, 68.5%-73.5%] vs 61.8% [95% CI, 57.3%-66.3%] with >3 visits/year; P = .003). Top tertile primary care physicians compared with the bottom tertile physicians had fewer minority patients (13.7% [95% CI, 10.6%-16.7%] vs 25.6% [95% CI, 20.2%-31.1%], respectively; P < .001), non-English-speaking patients (3.2% [95% CI, 0.7%-5.6%] vs 10.2% [95% CI, 5.5%-14.9%]; P <.001), and patients with Medicaid coverage or without insurance (9.6% [95% CI, 7.5%-11.7%] vs 17.2% [95% CI, 13.5%-21.0%]; P <.001). After accounting for practice site and visit frequency differences, adjusting for patient panel factors resulted in a relative mean change in physician rankings of 7.6 percentiles (95% CI, 6.6-8.7 percentiles) per primary care physician, with more than one-third (36%) of primary care physicians (59/162) reclassified into different quality tertiles. CONCLUSION Among primary care physicians practicing within the same large academic primary care system, patient panels with greater proportions of underinsured, minority, and non-English-speaking patients were associated with lower quality rankings for primary care physicians.

The longitudinal impact of patient navigation on equity in colorectal cancer screening in a large primary care network.

The long‐term effects of interventions to improve colorectal (CRC) screening in vulnerable populations are uncertain. The authors evaluated the impact of patient navigation (PN) on the equity of CRC prevention over a 5‐year period.

Characteristics of “Complex” Patients With Type 2 Diabetes Mellitus According to Their Primary Care Physicians

Author Affiliations: Edmond J. Safra Center for Ethics, Harvard University, Cambridge, Massachusetts (Dr PhamKanter); Mongan Institute for Health Policy, Massachusetts General Hospital, Boston (Dr Pham-Kanter); Department of Health Systems, Management, and Policy, Colorado School of Public Health (Dr Pham-Kanter), and Department of Clinical Pharmacy, Skaggs School of Pharmacy and Pharmaceutical Sciences (Dr Nair), University of Colorado Anschutz Medical Campus, Aurora; Department of Economics (Dr Pham-Kanter), University of Colorado Denver, Denver; and Department of Epidemiology, The Johns Hopkins University, Baltimore, Maryland (Dr Alexander). Correspondence: Dr Nair, Department of Clinical Pharmacy, Skaggs School of Pharmacy and Pharmaceutical Sciences, University of Colorado Anschutz Medical Campus, Mail Stop C238, 12850 E Montview Blvd, Room V20-1202, Aurora, CO 80045 (kavita.nair @ucdenver.edu). Author Contributions: Dr Pham-Kanter had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Pham-Kanter. Acquisition of data: Pham-Kanter and Nair. Analysis and interpretation of data: Pham-Kanter, Alexander, and Nair. Drafting of the manuscript: Pham-Kanter, Alexander, and Nair. Critical revision of the manuscript for important intellectual content: Pham-Kanter, Alexander, and Nair. Statistical analysis: Pham-Kanter. Obtained funding: PhamKanter, Alexander, and Nair. Administrative, technical, and material support: Pham-Kanter and Nair. Study supervision: Pham-Kanter and Nair. Financial Disclosure: Dr Alexander is a consultant for IMS Health and has served as an ad hoc member of the US FDA’s Drug Safety and Risk Management Advisory Committee. Dr Nair is a consultant for Janssen Services. Funding/Support: This research was supported by the Edmond J. Safra Center for Ethics at Harvard University. Dr Alexander is supported by awards from the Agency for Healthcare Research and Quality (R01 HS0189960) and the National Heart, Lung, and Blood Institute (R01 HL 107345-01). Role of the Sponsors: The funding sources had no role in the design and conduct of the study; analysis or interpretation of the data; or preparation or final approval of the manuscript prior to publication. Disclaimer: The contents of and opinions expressed in this report are solely the responsibility of the authors and do not necessarily represent the official views of the funders. Additional Contributions: Richard Read Allen, MS, Igor Gorlach, BA, Matthew Lemieux, and Grace Njau, BS, acquired some of the data required for the analysis. We wish to acknowledge input from members of the Research Ethics Program of Harvard Catalyst, the Harvard Clinical and Translational Science Center (NIH Award UL1 RR 025758 and financial contributions from Harvard University and its affiliated academic health care centers).

Patient Navigation for Comprehensive Cancer Screening in High-Risk Patients Using a Population-Based Health Information Technology System: A Randomized Clinical Trial.

IMPORTANCE Patient navigation (PN) to improve cancer screening in low-income and racial/ethnic minority populations usually focuses on navigating for single cancers in community health center settings. OBJECTIVE We evaluated PN for breast, cervical, and colorectal cancer screening using a population-based information technology (IT) system within a primary care network. DESIGN, SETTING, AND PARTICIPANTS Randomized clinical trial conducted from April 2014 to December 2014 in 18 practices in an academic primary care network. All patients eligible and overdue for cancer screening were identified and managed using a population-based IT system. Those at high risk for nonadherence with completing screening were identified using an electronic algorithm (language spoken, number of overdue tests, no-show visit history), and randomized to a PN intervention (n = 792) or usual care (n = 820). Navigators used the IT system to track patients, contact them, and provide intense outreach to help them complete cancer screening. MAIN OUTCOMES AND MEASURES Mean cancer screening test completion rate over 8-month trial for each eligible patient, with all overdue cancer screening tests combined using linear regression models. Secondary outcomes included the proportion of patients completing any and each overdue cancer screening test. RESULTS Among 1612 patients (673 men and 975 women; median age, 57 years), baseline patient characteristics were similar among randomized groups. Of 792 intervention patients, patient navigators were unable to reach 151 (19%), deferred 246 (38%) (eg, patient declined, competing comorbidity), and navigated 202 (32%). The mean proportion of patients who were up to date with screening among all overdue screening examinations was higher in the intervention vs the control group for all cancers combined (10.2% vs 6.8%; 95% CI [for the difference], 1.5%-5.2%; P < .001), and for breast (14.7% vs 11.0%; 95% CI, 0.2%-7.3%; P = .04), cervical (11.1% vs 5.7%; 95% CI, 0.8%-5.2%; P = .002), and colon (7.6% vs 4.6%; 95% CI, 0.8%-5.2%; P = .01) cancer compared with control. The proportion of overdue patients who completed any cancer screening during follow-up was higher in the intervention group (25.5% vs 17.0%; 95% CI, 4.7%-12.7%; P < .001). The intervention group had more patients completing screening for breast (23.4% vs 16.6%; 95% CI, 1.8%-12.0%; P = .009), cervical (14.4% vs 8.6%; 95% CI, 1.6%-10.5%; P = .007), and colorectal (13.7% vs 7.0%; 95% CI, 3.2%-10.4%; P < .001) cancer. CONCLUSIONS AND RELEVANCE Patient navigation as part of a population-based IT system significantly increased screening rates for breast, cervical, and colorectal cancer in patients at high risk for nonadherence with testing. Integrating patient navigation into population health management activities for low-income and racial/ethnic minority patients might improve equity of cancer care. TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT02553538.

Self-ratings of Health and Change in Walking Speed Over 2 Years: Results From the Caregiver-Study of Osteoporotic Fractures

Although poorer self-rated health (SRH) is associated with increased mortality, less is known about its impact on functioning. This study evaluated whether poorer SRH was associated with decline in walking speed and whether caregiving, often considered an indicator of chronic stress, modified this relation. The sample included 891 older US women from the Caregiver-Study of Osteoporotic Fractures. SRH was assessed at the baseline Caregiver-Study of Osteoporotic Fractures interview, conducted in 1999-2001, and was categorized as fair/poor or excellent/good. Rapid walking speed over 2, 3, or 6 m was measured at baseline and 2 annual follow-up interviews. Respondents with fair/poor SRH walked significantly slower at baseline than those with excellent/good health (mean = 0.8 (standard deviation, 0.3) vs. 1.0 (standard deviation, 0.3) m/second, P < 0.001). In adjusted linear mixed models of percentage change in walking speed, respondents with fair/poor SRH experienced a greater decline in walking speed than those with excellent/good SRH (-5.66% vs. -0.60%, P = 0.01). Caregivers with fair/poor SRH declined more than noncaregivers (-9.26% vs. -4.09%). High-intensity caregivers had the largest decline (-12.88%), whereas low-intensity caregivers in excellent/good SRH had no decline (2.61%). In summary, poorer SRH was associated with decline in walking speed in older women, and the stress of caregiving may have exacerbated its impact.

Non-visit-based cancer screening using a novel population management system.

Background: Advances in information technology (IT) now permit population-based preventive screening, but the best methods remain uncertain. We evaluated whether involving primary care providers (PCPs) in a visit-independent population management IT application led to more effective cancer screening. Methods: We conducted a cluster-randomized trial involving 18 primary care practice sites and 169 PCPs from June 15, 2011, to June 14, 2012. Participants included adults eligible for breast, cervical, and/or colorectal cancer screening. In practices randomized to the intervention group, PCPs reviewed real-time rosters of their patients overdue for screening and provided individualized contact (via a letter, practice delegate, or patient navigator) or deferred screening (temporarily or permanently). In practices randomized to the comparison group, overdue patients were automatically sent reminder letters and transferred to practice delegate lists for follow-up. Intervention patients without PCP action within 8 weeks defaulted to the automated control version. The primary outcome was adjusted average cancer screening completion rates over 1-year follow-up, accounting for clustering by physician or practice. Results: Baseline cancer screening rates (80.8% vs 80.3%) were similar among patients in the intervention (n = 51,071) and comparison group (n = 52,799). Most intervention providers used the IT application (88 of 101, 87%) and users reviewed 7984 patients overdue for at least 1 cancer screening (73% sent reminder letter, 6% referred directly to a practice delegate or patient navigator, and 21% deferred screening). In addition, 6128 letters were automatically sent to patients in the intervention group (total of 12,002 letters vs 16,378 letters in comparison practices; P < .001). Adjusted average cancer screening rates did not differ among intervention and comparison practices for all cancers combined (81.6% vs 81.4%; P = .84) nor breast (82.7% vs 82.7%; P = .96), cervical (84.1% vs 84.7%; P = .60), or colorectal cancer (77.8% vs 76.2%; P = .33). Conclusions: Involving PCPs in a visit-independent population management IT application resulted in similar cancer screening rates compared with an automated reminder system, but fewer patients were sent reminder letters. This suggests that PCPs were able to identify and exclude from contact patients who would have received automated reminder letters but not undergone screening.

Mammography FastTrack: An Intervention to Facilitate Reminders for Breast Cancer Screening across a Heterogeneous Multi-clinic Primary Care Network

Health care information technology can be a means to improve quality and efficiency in the primary care setting. However, merely applying technology without addressing how it fits into provider workflow and existing systems is unlikely to achieve improvement goals. Improving quality of primary care, such as cancer screening rates, requires addressing barriers at system, provider, and patient levels. The authors report the development, implementation, and preliminary use of a new breast cancer screening outreach program in a large multicenter primary care network. This installation paired population-based surveillance with customized information delivery based on a validated model linking patients to providers and practices. In the first six months, 86% of physicians and all case managers voluntarily participated in the program. Providers intervened in 83% of the mammogram-overdue population by initiating mailed reminders or deferring contact. Overall, 63% of patients were successfully contacted. Systematic population-based efforts are promising tools to improve preventative care.

Changes in Use of Anticoagulation in Patients With Atrial Fibrillation Within a Primary Care Network Associated With the Introduction of Direct Oral Anticoagulants

Atrial fibrillation (AF) and the decision to anticoagulate is a common problem faced by primary care physicians. Oral anticoagulation (OAC) is underused, despite its clear benefits with regard to stroke prevention. We examined OAC usage between 2010 and 2015, following the introduction of direct oral anticoagulants (DOACs) and specifically assessed whether more patients were anticoagulated over time. The study cohort included adult patients aged 18 and older with AF cared for in an 18-practice primary care network between 2010 and 2015. AF status was assigned each calendar year using a validated electronic health record algorithm. We examined OAC usage over time in all patients with AF, and in patients at high risk of stroke (CHA2DS2-VASc ≥ 2). The proportion of the population with AF increased over time (2010: 4,920 patients [3.5%], 2015: 6,452 patients [4.0%]). There was no increase in the proportion of patients prescribed any OAC treatment from 2010 (57.0%) to 2015 (57.4%) (p = 0.41). Similarly, in patients at high risk of stroke, the proportion anticoagulated did not increase over time (2010: 61.1%, 2015: 61.7%, p = 0.51). Over the study period, DOAC usage increased from 0.31% of all patients with AF in 2010 to 18.3% in 2015 (p < 0.001). Patients prescribed DOACs were younger, with lower risk of stroke. In conclusion, this study showed an increasing proportion of patients with AF over time in a primary care network. The use of DOACs increased over time; however, the proportion of patients treated with OAC did not increase over time.

Patient navigation to improve follow-up of abnormal mammograms among disadvantaged women.

BACKGROUND Patient navigation (PN) can improve breast cancer care among disadvantaged women. We evaluated the impact of a PN program on follow-up after an abnormal mammogram. METHODS Between 2007 and 2010, disadvantaged women with an abnormal mammogram (Breast Imaging-Reporting and Data System [BI-RADS] codes 0, 3, 4, 5) cared for in a community health center (CHC) with PN were compared to those receiving care in 11 network practices without PN. Multivariable logistic regression and Cox proportional hazards modeling were used to compare the percentages receiving appropriate follow-up and time to follow-up between the groups. RESULTS Abnormal mammography findings were reported for 132 women in the CHC with PN and 168 from practices without PN. The percentage of women with appropriate follow-up care was higher in the practice with PN than in non-PN practices (90.4% vs. 75.3%, adjusted p=0.006). RESULTS varied by BI-RADS score for women in PN and non-PN practices (BI-RADS 0, 93.7% vs. 90.2%, p=0.24; BI-RADS 3, 85.7% vs. 49.2%, p=0.003; BI-RADS 4/5, 95.1% vs. 82.8%, p=0.26). Time to follow-up was similar for BI-RADS 0 and occurred sooner for women in the PN practice than in non-PN practices for BI-RADS 3 and 4/5 (BI-RADS 3, adjusted hazard ratio [aHR], 95% confidence interval [CI]: 2.41 [1.36-4.27], BI-RADS 4/5, aHR [95% CI]: 1.41 [0.88-2.24]). CONCLUSIONS Disadvantaged women from a CHC with PN were more likely to receive appropriate follow-up after an abnormal mammogram than were those from practices without PN. Expanding PN to include all disadvantaged women within primary care networks could improve equity in cancer care.