The Transforming Outcomes for Patients Through Medical Home Evaluation and reDesign (TOPMED) Cluster Randomized Controlled Trial: Cost and Utilization Results.

BACKGROUND: Primary Care Medical Home (PCMH) redesign efforts are intended to enhance primary care's ability to improve population health and well-being. PCMH transformation that is focused on "high-value elements" (HVEs) for cost and utilization may improve effectiveness. OBJECTIVES: The objective of this study was to determine if a focus on achieving HVEs extracted from successful primary care transformation models would reduce cost and utilization as compared with a focus on achieving PCMH quality improvement goals. RESEARCH DESIGN: A stratified, cluster randomized controlled trial with 2 arms. All practices received equal financial incentives, health information technology support, and in-person practice facilitation. Analyses consisted of multivariable modeling, adjusting for the cluster, with difference-in-difference results. SUBJECTS: Eight primary care clinics that were engaged in PCMH reform. MEASURES: We examined: (1) total claims payments; (2) emergency department (ED) visits; and (3) hospitalizations among patients during baseline and intervention years. RESULTS: In total, 16,099 patients met the inclusion criteria. Intervention clinics had significantly lower baseline ED visits (P=0.02) and claims paid (P=0.01). Difference-in-difference showed a decrease in ED visits greater in control than intervention (ED per 1000 patients: +56; 95% confidence interval: +96, +15) with a trend towards decreased hospitalizations in intervention (-15; 95% confidence interval: -52, +21). Costs were not different. In modeling monthly outcome means, the generalized linear mixed model showed significant differences for hospitalizations during the intervention year (P=0.03). DISCUSSION: The trial had a trend of decreasing hospitalizations, increased ED visits, and no change in costs in the HVE versus quality improvement arms.

Primary care practices' ability to predict future risk of expenditures and hospitalization using risk stratification and segmentation.

BACKGROUND: Patients with complex health care needs may suffer adverse outcomes from fragmented and delayed care, reducing well-being and increasing health care costs. Health reform efforts, especially those in primary care, attempt to mitigate risk of adverse outcomes by better targeting resources to those most in need. However, predicting who is susceptible to adverse outcomes, such as unplanned hospitalizations, ED visits, or other potentially avoidable expenditures, can be difficult, and providing intensive levels of resources to all patients is neither wanted nor efficient. Our objective was to understand if primary care teams can predict patient risk better than standard risk scores. METHODS: Six primary care practices risk stratified their entire patient population over a 2-year period, and worked to mitigate risk for those at high risk through care management and coordination. Individual patient risk scores created by the practices were collected and compared to a common risk score (Hierarchical Condition Categories) in their ability to predict future expenditures, ED visits, and hospitalizations. Accuracy of predictions, sensitivity, positive predictive values (PPV), and c-statistics were calculated for each risk scoring type. Analyses were stratified by whether the practice used intuition alone, an algorithm alone, or adjudicated an algorithmic risk score. RESULTS: In all, 40,342 patients were risk stratified. Practice scores had 38.6% agreement with HCC scores on identification of high-risk patients. For the 3,381 patients with reliable outcomes data, accuracy was high (0.71-0.88) but sensitivity and PPV were low (0.16-0.40). Practice-created scores had 0.02-0.14 lower sensitivity, specificity and PPV compared to HCC in prediction of outcomes. Practices using adjudication had, on average, .16 higher sensitivity. CONCLUSIONS: Practices using simple risk stratification techniques had slightly worse accuracy in predicting common outcomes than HCC, but adjudication improved prediction.

Effect of a Pragmatic, Cluster-randomized Controlled Trial on Patient Experience With Care: The Transforming Outcomes for Patients Through Medical Home Evaluation and reDesign (TOPMED) Study.

BACKGROUND: Health reform programs like the patient-centered medical home are intended to improve the triple aim. Previous studies on patient-centered medical homes have shown mixed effects, but high value elements (HVEs) are expected to improve the triple aim. OBJECTIVE: The aim of this study is to understand whether focusing on HVEs would improve patient experience with care. METHODS: Eight clinics were cluster-randomized in a year-long trial. Both arms received practice facilitation, IT-based reporting, and financial incentives. Intervention practices were encouraged to choose HVEs for quality improvement goals. To assess patient experience, 1597 Consumer Assessment of Healthcare Providers and Systems surveys were sent pretrial and posttrial to a stratified random sample of patients. Difference-in-difference multivariate analysis was used to compare patient responses from intervention and control practices, adjusting for confounders. RESULTS: The response rate was 43% (n=686). Nonrespondent analysis showed no difference between arms, although differences were seen by risk status and age. The overall difference in difference was 2.8%, favoring the intervention. The intervention performed better in 9 of 11 composites. The intervention performed significantly better in follow-up on test results (P=0.091) and patients' rating of the provider (P=0.091), whereas the control performed better in access to care (P=0.093). Both arms also had decreases, including 4 of 11 composites for the intervention, and 8 of 11 for the control. DISCUSSION: Practices that targeted HVEs showed significantly more improvement in patient experience of care. However, contemporaneous trends may have affected results, leading to declines in patient experience in both arms.

Defining High Value Elements for Reducing Cost and Utilization in Patient-Centered Medical Homes for the TOPMED Trial.

INTRODUCTION: Like most patient-centered medical home (PCMH) models, Oregon's program, the Patient-Centered Primary Care Home (PCPCH), aims to improve care while reducing costs; however, previous work shows that PCMH models do not uniformly achieve desired outcomes. Our objective was to describe a process for refining PCMH models to identify high value elements (HVEs) that reduce cost and utilization. METHODS: We performed a targeted literature review of each PCPCH core attribute. Value-related concepts and their metrics were abstracted, and studies were assessed for relevance and strength of evidence. Focus groups were held with stakeholders and patients, and themes related to each attribute were identified; calculation of HVE attainment versus PCPCH criteria were completed on eight primary care clinics. Analyses consisted of descriptive statistics and criterion validity with stakeholder input. RESULTS: 2,126 abstracts were reviewed; 22 met inclusion criteria. From these articles and focus groups of stakeholders/experts (n = 49; 4 groups) and patients (n = 7; 1 group), 12 HVEs were identified that may reduce cost and utilization. At baseline, clinics achieved, on average, 31.3 percent HVE levels compared to an average of 87.9 percent of the 35 PCMH measures. DISCUSSION: A subset of measures from the PCPCH model were identified as "high value" in reducing cost and utilization. HVE performance was significantly lower than standard measures, and may better calibrate clinic ability to reduce costs. CONCLUSION: Through literature review and stakeholder engagement, we created a novel set of high value elements for advanced primary care likely to be more related to cost and utilization than other models.

Implementing Risk Stratification in Primary Care: Challenges and Strategies.

INTRODUCTION: Primary care risk stratification (RS) has been shown to help practices better understand their patient populations' needs and may improve health outcomes and reduce expenditures by targeting and tailoring care to high-need patients. This study aims to understand key considerations practices faced and practice experiences as they began to implement RS models. METHODS: We conducted semistructured interviews about experiences in RS with 34 stakeholders from 15 primary care practices in Oregon and Colorado and qualitatively analyzed the data. RESULTS: Three decisions were important in shaping practices' experiences with RS: choosing established versus self-created algorithms or heuristics, clinical intuition, or a combination; selecting mechanisms for assigning risk scores; determining how to integrate RS approaches into care delivery. Practices using clinical intuition found stratification time-consuming and difficult to incorporate into existing workflows, but trusted risk scores more than those using algorithms. Trust in risk scores was influenced by data extraction capabilities; practices often lacked sufficient data to calculate their perceived optimal risk score. Displaying the scores to the care team was a major issue. Finally, obtaining buy-in from care team members was challenging, requiring repeated cycles of improvement and workflow integration. DISCUSSION: Practices used iterative approaches to RS implementation. As a result, procedural and algorithmic changes were introduced and were influenced by practices' health IT, staffing, and resource capacities. Practices were most successful when able to make iterative changes to their approaches, incorporated both automation and human process in RS, educated staff on the importance of RS, and had readily accessible risk scores.

Understanding the Impact of Variations in Measurement Period Reporting for Electronic Clinical Quality Measures.

OBJECTIVE: To understand the impact of varying measurement period on the calculation of electronic Clinical Quality Measures (eCQMs). BACKGROUND: eCQMs have increased in importance in value-based programs, but accurate and timely measurement has been slow. This has required flexibility in key measure characteristics, including measurement period, the timeframe the measurement covers. The effects of variable measurement periods on accuracy and variability are not clear. METHODS: 209 practices were asked to extract and submit four eCQMs from their Electronic Health Records on a quarterly basis using a 12-month measurement period. Quarterly submissions were collected via REDCap. The measurement periods of the survey data were categorized into non-standard (3, 6, 9 months and other) and standard periods (12 months). For comparison, patient-level data from three clinics were collected and calculated in an eCQM registry to measure the impact of varying measurement periods. We assessed the central tendency, shape of the distributions, and variability across the four measures. Analysis of variance (ANOVA) was conducted to analyze the differences among standard and non-standard measurement period means, and variation among these groups. RESULTS: Of 209 practices, 191 (91 percent) submitted data over three quarters. Of the 546 total submissions, 173 had non-standard measurement periods. Differences between measures with standard versus non-standard periods ranged from -3.3 percent to 14.2 percent between clinics (p < .05 for 3 of 4), using the patient-level data yielded deltas of -1.6 percent to 0.6 percent when comparing non-standard and standard periods. CONCLUSION: Variations in measurement periods were associated with variation in performance between clinics for 3 of the 4 eCQMs, but did not have significant differences when calculated within clinics. Variations from standard measurement periods may reflect poor data quality and accuracy.

Perceptions of Risk Stratification Workflows in Primary Care.

Risk stratification (RS) in primary care is frequently used by policy-makers, payers, and health systems; the process requires risk assessment for adverse health outcomes across a population to assign patients into risk tiers and allow care management (CM) resources to be targeted effectively. Our objective was to understand the approach to and perception of RS in primary care practices. An online survey was developed, tested, and administered to 148 representatives of 37 primary care practices engaged in RS varying in size, location and ownership. The survey assessed practices' approach to, perception of, and confidence in RS, and its effect on subsequent CM activities. We examined psychometric properties of the survey to determine validity and conducted chi-square analyses to determine the association between practice characteristics and confidence and agreement with risk scores. The survey yielded a 68% response rate (100 respondents). Overall, participants felt moderately confident in their risk scores (range 41-53.8%), and moderately to highly confident in their subsequent CM workflows (range 46-68%). Respondents from small and independent practices were more likely to have higher confidence and agreement with their RS approaches and scores (p < 0.01). Confidence levels were highest, however, when practices incorporated human review into their RS processes (p < 0.05). This trend was not affected by respondents' professional roles. Additional work from a broad mixed-methods effort will add to our understanding of RS implementation processes and outcomes.

Long-term Effects of a Health Literacy Curriculum for Family Medicine Residents.

INTRODUCTION: Improving education about health literacy for health care professionals has been recommended, and many US family medicine residency programs have developed such curricula. Few studies have evaluated the effectiveness of health literacy curricula for health care professionals. This pilot study aimed to determine whether a longitudinal health literacy curriculum for family medicine residents could achieve long-term sustained improvements in health literacy knowledge and clear communication practices. METHODS: Self-reported pre- and postassessments were conducted for a series of four health literacy didactic and experiential trainings over 11 months with a cohort of 12 first-year family medicine residents (study adequately powered for cohort of 10 or more). RESULTS: Five out of five health literacy knowledge items showed sustained significant improvement immediately after the initial didactic training. Two out of eight clear communication behaviors (eliciting patients' questions through an open-ended approach, and using a teach-back method to check for patients' understanding) showed sustained significant improvements in the 11-month follow-up period. The remaining six behaviors demonstrated a saw-tooth pattern, wherein each training session produced improvements in planned behaviors, which were, however, not maintained at subsequent follow-up assessments. CONCLUSIONS: These data suggest that residents learned the cognitive material related to health literacy and clear communication easily, but had difficulty integrating many trained skills into clinical practice, despite the use of experiential learning techniques. Future studies should use an observational design to assess clear communication behaviors, and should include assessment of potential barriers to implementing clear communication skills in clinical practice.

Specifications of Clinical Quality Measures and Value Set Vocabularies Shift Over Time: A Study of Change through Implementation Differences.

Clinical quality measures (CQMs) aim to identify gaps in care and to promote evidence-based guidelines. Official CQM definitions consist of a measure's logic and grouped, standardized codes to define key concepts. In this study, we used the official CQM update process to understand how CQMs' meanings change over time. First, we identified differences between the narrative description, logic, and the vocabulary specifications offour standardized CQMs' definitions in subsequent versions (2015, 2016, and 2017). Next, we implemented the various versions in a quality measure calculation registry to understand how the differences affected calculated prevalence of risk and measure performance. Global performance rates changed up to 5.32%, and an increase of up to 28% new patients was observed for key conditions between versions. Updates to definitions that change a measure's logic and choices to include/exclude codes in value set vocabularies changes measurement of quality and likely introduces variation by implementation.

From Concepts and Codes to Healthcare Quality Measurement: Understanding Variations in Value Set Vocabularies for a Statin Therapy Clinical Quality Measure.

OBJECTIVE: To understand the impact of distinct concept to value set mapping on the measurement of quality of care. BACKGROUND: Clinical quality measures (CQMs) intend to measure the quality of healthcare services provided, and to help promote evidence-based therapies. Most CQMs consist of grouped codes from vocabularies - or 'value sets' - that represent the unique identifiers (i.e., object identifiers), concepts (i.e., value set names), and concept definitions (i.e., code groups) that define a measure's specifications. In the development of a statin therapy CQM, two unique value sets were created by independent measure developers for the same global concepts. METHODS: We first identified differences between the two value set specifications of the same CQM. We then implemented the various versions in a quality measure calculation registry to understand how the differences affected calculated prevalence of risk and measure performance. RESULTS: Global performance rates only differed by 0.8%, but there were up to 2.3 times as many patients included with key conditions, and differing performance rates of 7.5% for patients with 'myocardial infarction' and 3.5% for those with 'ischemic vascular disease'. CONCLUSION: The decisions CQM developers make about which concepts and code groups to include or exclude in value set vocabularies can lead to inaccuracies in the measurement of quality of care. One solution is that developers could provide rationale for these decisions. Endorsements are needed to encourage system vendors, payers, informaticians, and clinicians to collaborate in the creation of more integrated terminology sets.

Patient-Provider Communication: Does Electronic Messaging Reduce Incoming Telephone Calls?

PURPOSE: Internet-based patient portals are increasingly being implemented throughout health care organizations to enhance health and optimize communication between patients and health professionals. The decision to adopt a patient portal requires careful examination of the advantages and disadvantages of implementation. This study aims to investigate 1 proposed advantage of implementation: alleviating some of the clinical workload faced by employees. METHODS: A retrospective time-series analysis of the correlation between the rate of electronic patient-to-provider messages-a common attribute of Internet-based patient portals-and incoming telephone calls. The rate of electronic messages and incoming telephone calls were monitored from February 2009 to June 2014 at 4 economically diverse clinics (a federally qualified health center, a rural health clinic, a community-based clinic, and a university-based clinic) related to 1 university hospital. RESULTS: All 4 clinics showed an increase in the rate of portal use as measured by electronic patient-to-provider messaging during the study period. Electronic patient-to-provider messaging was significantly positively correlated with incoming telephone calls at 2 of the clinics (r = 0.546, P < .001 and r = 0.543, P < .001). The remaining clinics were not significantly correlated but demonstrated a weak positive correlation (r = 0.098, P = .560 and r = 0.069, P = .671). CONCLUSIONS: Implementation and increased use of electronic patient-to-provider messaging was associated with increased use of telephone calls in 2 of the study clinics. While practices are increasingly making the decision of whether to implement a patient portal as part of their system of care, it is important that the motivation behind such a change not be based on the idea that it will alleviate clinical workload.