Development of a Follow-Up Measure to Ensure Complete Screening for Colorectal Cancer

Importance

The current quality performance measure for colorectal cancer (CRC) screening is limited to initial screening. Despite low rates, there is no measure for appropriate follow-up with colonoscopy after receipt of an abnormal result of a stool-based screening test (SBT) for CRC. A quality performance measure is needed.

Objective

To develop and test a quality performance measure for follow-up colonoscopy within 6 months of an abnormal result of an SBT for CRC.

Design, Setting, and Participants

This retrospective quality improvement study examined data from January 1, 2016, to December 31, 2020, with 2018 plus 6 months of follow-up as the primary measurement period to verify performance rates, specify a potential measure, and test for validity, reliability, and feasibility. The Optum Labs Data Warehouse (OLDW), a deidentified database of health care claims and clinical data, was accessed. The OLDW contains longitudinal health information on enrollees and patients, representing a diverse mixture of ages and geographic regions across the US. For the database study, adults from 38 health care organizations (HCOs) aged 50 to 75 years who completed an initial CRC SBT with an abnormal result were observed to determine follow-up colonoscopy rates within 6 months. Rates were stratified by race, ethnicity, sex, insurance, and test modality. Three HCOs participated in the feasibility field testing. Data were analyzed from June 1, 2022, to May 31, 2023.

Main Outcome and Measures

The primary outcome consisted of follow-up colonoscopy rates following an abnormal SBT result for CRC. Reliability statistics were also calculated across HCOs, race, ethnicity, and measurement year.

Results

Among 20 581 adults (48.6% men and 51.4% women; 307 [1.5%] Asian, 492 [7.2%] Black, 644 [3.1%] Hispanic, and 17 705 [86.0%] White; mean [SD] age, 63.6 [7.1] years) in 38 health systems, 47.9% had a follow-up colonoscopy following an abnormal SBT result for CRC within 6 months. There was significant variation between HCOs. Notably, significantly fewer Black patients (37.1% [95% CI, 34.6%-39.5%]) and patients with Medicare (49.2% [95% CI, 47.7%-50.6%]) or Medicaid (39.2% [95% CI, 36.3%-42.1%]) insurance received a follow-up colonoscopy. A quality performance measure that tracks rates of follow-up within 6 months of an abnormal SBT result was observed to be feasible, valid, and reliable, with a median reliability statistic between HCOs of 94.5% (range, 74.3%-99.7%).

Conclusions and Relevance

The findings of this observational study of 20 581 adults suggest that a measure of follow-up colonoscopy within defined periods after an abnormal result of an SBT test for CRC is warranted based on low current performance rates and would be feasible to collect by health systems and produce valid, reliable results.

Perspectives on Lower Extremity Peripheral Artery Disease: A Qualitative Study of Early Diagnosis and Treatment and the Impact of Health Disparities

Lower-extremity peripheral artery disease (PAD), the accumulation of atherosclerotic plaque in the arteries of the legs, causes substantial morbidity and mortality. Frequent under- and delayed diagnosis result in poor outcomes, disproportionately affecting individuals from racial and ethnic minority groups. To understand barriers to early detection and treatment and factors contributing to disparities, American Medical Group Association (AMGA) conducted roundtable discussions and semistructured interviews in 2021. Eighteen participants discussed PAD evaluation, diagnosis, early medical management, and disparities in care. A qualitative case study approach and data reduction methods were used to generate themes, draw conclusions, and make actionable recommendations. Identified themes included lack of (1) prioritization of PAD for population health; (2) engagement of primary care providers in early evaluation and referral; (3) "ownership" of lower-extremity PAD within health systems; and (4) focus on disparities in care. Participant solutions included (1) financial impact of early PAD management, in the context of value-based payment; (2) embedding an advanced practice provider into a vascular surgery practice to facilitate evaluation and provide medical therapy; and (3) leveraging care coordination, multidisciplinary clinics, and telehealth technology to provide comprehensive care for patients with PAD and address disparities. A deliberate focused effort is necessary to close gaps and the accompanying disparities in early evaluation, diagnosis, and treatment for people with lower-extremity PAD. The authors describe 3 models that can be emulated to improve care for this high-risk population. With improved reimbursement and better medical therapies, now is the time to focus on early diagnosis and management of PAD.

Rates of Follow-up Colonoscopy After a Positive Stool-Based Screening Test Result for Colorectal Cancer Among Health Care Organizations in the US, 2017-2020

IMPORTANCE

Noninvasive stool-based screening tests (SBTs) are effective alternatives to colonoscopy. However, a positive SBT result requires timely follow-up colonoscopy (FU-CY) to complete the colorectal cancer screening paradigm.

OBJECTIVES

To evaluate FU-CY rates after a positive SBT result and to assess the association of the early COVID-19 pandemic with FU-CY rates.

DESIGN, SETTING, AND PARTICIPANTS

This mixed-methods cohort study included retrospective analysis of deidentified administrative claims and electronic health records data between June 1, 2015, and June 30, 2021, from the Optum Labs Data Warehouse and qualitative, semistructured interviews with clinicians from 5 health care organizations (HCOs). The study population included data from average-risk primary care patients aged 50 to 75 years with a positive SBT result between January 1, 2017, and June 30, 2020, at 39 HCOs.

MAIN OUTCOMES AND MEASURES

The primary outcome was the FU-CY rate within 1 year of a positive SBT result according to patient age, sex, race, ethnicity, insurance type, Charlson Comorbidity Index (CCI), and prior SBT use.

RESULTS

This cohort study included 32 769 individuals (16 929 [51.7%] female; mean [SD] age, 63.1 [7.1] years; 2092 [6.4%] of Black and 28 832 [88.0%] of White race; and 825 [2.5%] of Hispanic ethnicity). The FU-CY rates were 43.3% within 90 days of the positive SBT result, 51.4% within 180 days, and 56.1% within 360 days (n = 32 769). In interviews, clinicians were uniformly surprised by the low FU-CY rates. Rates varied by race, ethnicity, insurance type, presence of comorbidities, and SBT used. In the Cox proportional hazards regression model, the strongest positive association was with multitarget stool DNA use (hazard ratio, 1.63 [95% CI, 1.57-1.68] relative to fecal immunochemical tests; P < .001), and the strongest negative association was with the presence of comorbidities (hazard ratio, 0.64 [95% CI, 0.59-0.71] for a CCI of >4 relative to 0; P < .001). The early COVID-19 pandemic was associated with lower FU-CY rates.

CONCLUSIONS AND RELEVANCE

This study found that FU-CY rates after a positive SBT result for colorectal cancer screening were low among an average-risk population, with the median HCO achieving a 53.4% FU-CY rate within 1 year. Socioeconomic factors and the COVID-19 pandemic were associated with lower FU-CY rates, presenting opportunities for targeted intervention by clinicians and health care systems.

Predicting Chronic Opioid Use Among Patients With Osteoarthritis Using Electronic Health Record Data

OBJECTIVE

To estimate the risk of a patient with osteoarthritis (OA) developing chronic opioid use (COU) within 1 year of a new opioid prescription by using electronic health record (EHR) data and predictive models.

METHODS

We used EHR data from 13 health care organizations to identify patients with OA with an opioid prescription between March 1, 2017 and February 28, 2019 and no record of opioid use in the prior 6 months. We evaluated 4 machine learning models to estimate patients' risk of COU (≥3 prescriptions ≥84 days, maximum gap ≤60 days). We also estimated the transportability of models to organizations outside the training set.

RESULTS

The cohort consisted of 33,894 patients with OA, of whom 2,925 (8.6%) developed COU within 1 year. All models demonstrated good discrimination, with the best-performing model (random forest) achieving an area under the receiver operating characteristic curve (AUC) of 0.728 (95% CI 0.711-0.745), but the simplest regression model performed nearly as well (AUC 0.717 [95% CI 0.699-0.734]). Predicted risk deciles spanned a range of 2% risk for the 10th percentile to 18% risk for the 90th percentile for well-calibrated models. Models showed highly variable discrimination across organizations (AUC 0.571-0.842).

CONCLUSIONS

We found that EHR-based predictive models could estimate the risk of future COU among patients with OA to help inform care decisions. Black-box methods did not have significant advantages over more interpretable models. Care should be taken when extending all models into organizations not included in model training because of a high variability in performance across held-out organizations.

Physician Practices With Robust Capabilities Spend Less On Medicare Beneficiaries Than More Limited Practices

No research has considered a range of physician practice capabilities for managing patient care when examining practice-level influences on quality of care, utilization, and spending. Using data from the 2017 National Survey of Healthcare Organizations and Systems linked to 2017 Medicare fee-for-service claims data from attributed beneficiaries, we examined the association of practice-level capabilities with process measures of quality, utilization, and spending. In propensity score-weighted mixed-effects regression analyses, physician practice locations with "robust" capabilities had lower total spending compared to locations with "mixed" or "limited" capabilities. Quality and utilization, however, did not differ by practice-level capabilities. Physician practice locations with robust capabilities spend less on Medicare fee-for-service beneficiaries but deliver quality of care that is comparable to the quality delivered in locations with low or mixed capabilities. Reforms beyond those targeting practice capabilities, including multipayer alignment and payment reform, may be needed to support larger performance advantages for practices with robust capabilities.

Assessing opportunities to advance quality measures in adult obesity

OBJECTIVES

To evaluate the methodological soundness and performance of 3 obesity quality measures aimed at promoting improvements in obesity care.

STUDY DESIGN

Retrospective, clinical, and administrative data-based observational research study to evaluate scientific soundness, feasibility, and performance of obesity quality measures.

METHODS

Four test sites (clinicians/clinician groups) submitted clinical and administrative health data including patient demographics, diagnoses, and encounter information for patient panels encompassing individuals aged 18 to 79 years with at least 1 ambulatory visit between July 1, 2017, and June 30, 2018 (measurement period). Clinician/clinician group data were supplemented by an Optum data set contributing patient information from 21 health care organizations with approximately 6 million qualifying patients to assess the impact of using a larger data set for measure testing. Patients were excluded if they met any of the following criteria: were pregnant during the measurement period or in the 6 months prior to the measurement period, had died during the measurement year, or had evidence of palliative or hospice care during the measurement period.

RESULTS

This study resulted in the identification of a clinician/clinician group-level measure, Documentation of Obesity Diagnosis, as being feasible and reliable; however, the measure requires additional evaluation and potential adjustments to determine validity. Other measures included in our evaluation had feasibility and methodological challenges due to data capture and coding limitations.

CONCLUSIONS

Findings of our current study suggest that there are emerging opportunities to capture data and advance obesity measurement incrementally. A process measure focused on obesity diagnosis has the most potential for immediate implementation by clinicians, and additional measures focused on change in body mass index over time and use of evidence-based obesity treatment remain challenging to implement due to data capture and benefit coverage.

Successful Implementation of an All-or-None Diabetes Measure in 10 U.S. Health Systems

Type 2 diabetes mellitus (T2DM) affects 31.5 million adults in the United States and is commonly treated in primary care settings. One promising approach to comprehensive care is to focus on an all-or-none diabetes bundle measure, which ensures each patient meets a set of guideline-recommended measures. This requires a practice-level coordinated strategy. The purpose of this initiative was to help health care organizations (HCOs) improve the care and outcomes of patients with T2DM using an all-or-none bundle measure. This observational study was carried out in the context of a national best practices learning Collaborative that implemented targeted interventions in primary care settings and measured success using an all-or-none bundle measure. Ten AMGA member-HCOs, across 8 states, treating nearly 300,000 adult patients with T2DM in primary care participated. The primary measure, the Together 2 Goal^® Core Bundle, included hemoglobin A1c (A1c) control (<8%), blood pressure (BP) control (<140/90 mmHg), lipid management (prescribed a statin), and medical attention for nephropathy. All 10 HCOs improved the Core Bundle measure during the 12-month Collaborative. The rate for the Core Bundle improved from 40.2% to 42.8%, an absolute increase of 2.6% (P < 0.001). In addition, 9 HCOs improved BP control, 8 improved lipid management, 6 improved attention to nephropathy, and 4 improved A1c control. Implementing interventions in primary care settings was successful in achieving comprehensive care for an estimated additional 7700 people living with T2DM who met all 4 components of the bundle measure during the 12-month intervention period.

Using Quality Measures to Drive Improvements in Immunization Rates: Findings from a Real-World Evaluation from 3 US Health Care Organizations

Adult immunization rates are low and continue to fail to meet national targets. The coronavirus disease 2019 pandemic halted routine health care services for many, including vaccinations. This study explored whether the National Committee for Quality Assurance's Adult Immunization Status (AIS) measure, which had previously only been evaluated for reporting by health plans, could be leveraged by health care organizations (HCOs) as a tactic to improve immunization rates. Methods included a quantitative analysis of deidentified patient electronic health record data from 3 HCOs, supplemented by qualitative interviews to further understand opportunities and barriers. The analysis indicated that the data necessary for calculation of the AIS measure are available within HCOs and that measure performance could be enhanced with supplemental data from external sources, such as state immunization registries. Although HCOs rates were consistent with national estimates, this research further validated that adult immunization rates in the United States are low and highlighted the profound disparities that exist. For instance, the likelihood of completing all age-appropriate vaccinations was lower if patients were Black or African American, enrolled in Medicaid, or without health insurance. As a result of this study, the authors concluded that the AIS measure is feasible for use in medical groups and could potentially help drive quality improvements in immunization rates; however, there are considerations for implementation particularly if providers are being held accountable for measure performance.

The Serious Illness Population: Ascertainment via Electronic Health Record or Claims Data

CONTEXT

Palliative care can improve the lives of people with serious illness, yet clear operational definitions of this population do not exist. Prior efforts to identify this population have not focused on Medicare Advantage (MA) and commercial health plan enrollees.

OBJECTIVES

We aimed to operationalize our conceptual definition of serious illness to identify those with serious medical conditions (SMC) among commercial insurance and MA enrollees, and to compare the populations identified through electronic health record (EHR) or claims data sources.

METHODS

We used de-identified claims and EHR data from the OptumLabs Data Warehouse (2016-2017), to identify adults age ≥18 with SMC and examine their utilization and mortality. Within the subset found in both data sources, we compared the performance of claims and EHR data.

RESULTS

Within claims, SMC was identified among 10% of those aged ≥18 (5.4% ages 18-64, 27% age ≥65). Within EHR, SMC was identified among 9% of those aged ≥18 (5.6% ages 18-64, 21% ages ≥65). Hospital, emergency department and mortality rates were similar between the EHR and claims-based groups. Only 50% of people identified as having SMC were recognized by both data sources.

CONCLUSION

These results demonstrate the feasibility of identifying adults with SMC in a commercially insured population, including MA enrollees; yet separate use of EHR or claims result in populations that differ. Future research should examine methods to combine these data sources to optimize identification and support population management, quality measurement, and research to improve the care of those living with serious illness.

Building Successful Models in Primary Care to Improve the Management of Adult Patients with Obesity

Although several obesity clinical practice guidelines are available and relevant for primary care, a practical and effective medical model for treating obesity is necessary. The aim of this study was to develop and implement a holistic population health-based framework with components to support primary care-based obesity management in US health care organizations. The Obesity Care Model Collaborative (OCMC) was conducted with guidance and expertise of an advisory committee, which selected participating health care organizations based on prespecified criteria. A committee comprising obesity and quality improvement specialists and representatives from each organization developed and refined the obesity care framework for testing and implementing guideline-based practical interventions targeting obesity. These interventions were tracked over time, from an established baseline to 18 months post implementation. Ten geographically diverse organizations, treating patients with diverse demographics, insurance coverage, and health status, participated in the collaborative. The key interventions identified for managing obesity in primary care were applicable across the 4 OCMC framework domains: community, health care organization, care team, and patient/family. Care model components were developed within each domain to guide the primary care of obesity based on each organization's structure, resources, and culture. Key interventions included development of quality monitoring systems, training of leadership and staff, identifying clinical champions, patient education, electronic health record best practice alerts, and establishment of community partnerships, including the identification of external resources. This article describes the interventions developed based on the framework, with a focus on implementation of the model and lessons learned.

A Retrospective Analysis of Therapeutic Inertia in Type 2 Diabetes Management Across a Diverse Population of Health Care Organizations in the USA

INTRODUCTION

If their target glycated hemoglobin (HbA_1c) is not achieved after 3 months, timely treatment intensification is recommended in people with type 2 diabetes to maintain glycemic control and minimize vascular complications. We retrospectively investigated potential therapeutic inertia in the management of type 2 diabetes in multiple health care organizations across the USA.

METHODS

Electronic health records were analyzed from 22 American Medical Group Association (AMGA) health care organizations. Bolus insulin-naïve patients with type 2 diabetes and HbA_1c ≥ 8.0% (≥ 64 mmol/mol) at baseline were followed for 24 months to identify the frequency and average duration of therapeutic inertia (no new class of glucose-lowering medication prescribed, or not achieving their target HbA_1c [< 8.0%; < 64 mmol/mol]).

RESULTS

The study cohort comprised almost 28,000 patients. Therapeutic inertia was observed in ≈ 50% of patients after 6 months, and in > 10% after 24 months. Less therapeutic inertia was observed in patients receiving one or no oral antidiabetic drugs (OADs) (36% or 28%, respectively, at 6 months), while more inertia was seen following multiple OADs or basal insulin (54% of those on baseline basal insulin at 6 months). Although an observable action was recorded for 90% of patients, many (44%) had still not achieved their target HbA_1c after 24 months.

CONCLUSION

The results corroborate the presence of therapeutic inertia in people with type 2 diabetes, suggesting that treatment intensification guidelines are not being followed. Extensive variability in the presence of therapeutic inertia was observed both across and within organizations; investigating this further and sharing best practices across providers might help improve the quality of patient care at organizational and national levels.

Diagnosing Obesity as a First Step to Weight Loss: An Observational Study

OBJECTIVE

This study aimed to explore the relationship between an obesity diagnosis and weight loss as a percentage of total body weight loss over 9 to 15 months, using electronic health record data.

METHODS

An observational study of 688,878 adult patients at 15 health systems with BMI ≥ 30 kg/m² examined the relationship between weight loss and documentation of obesity diagnosis. Multivariable logistic regression models were created using a stepwise backwards elimination procedure to identify potential predictors of weight loss.

RESULTS

Of patients with BMI ≥ 30, 44.9% had an obesity diagnosis on a claim or electronic health record problem list; 16.9% and 5.9% lost ≥ 5% and ≥ 10% of their body weight, respectively. Multivariable logistic regression models revealed a diagnosis of obesity on the same day as the initial weight (odds ratio [OR] = 1.3; CI: 1.2-1.3; P < 0.001) as a predictor of ≥ 5% total body weight loss in 9 to 15 months. Other significant predictors included an antiobesity medication prescription, female sex, diagnosis of type 2 diabetes, Medicare/Medicaid insurance, and number of ambulatory visits.

CONCLUSIONS

While controlling for potentially confounding factors, documentation of an obesity diagnosis remained independently predictive of at least 5% weight loss. This suggests that documenting a diagnosis of obesity may be an important step toward engaging patients to lose weight.

Variation in Patient Smoking Cessation Rates Among Health-Care Providers: An Observational Study

BACKGROUND

Physicians play a crucial role in providing smoking cessation counseling and medications. However, it is unknown whether individual physicians' approaches affect whether patients quit.

RESEARCH QUESTION

This study assessed patient quit rates within a national quality-improvement learning collaborative to document variation in quit rates at the physician, practice, and health system levels.

STUDY DESIGN AND METHODS

A retrospective cohort study was conducted of primary care patients identified from the Optum analytics database containing longitudinal ambulatory data for patients from 22 health-care organizations between January 2012 and December 2018. The study included smokers aged ≥ 18 years who attended at least three ambulatory visits, with two visits at least 1 year apart. The primary study outcome was abstinence for ≥ 1 year. A mixed effects logistic regression model was used to predict the probability of quitting as a function of patient variables. Quit rates were then adjusted by patient factors and calculated at the level of clinician, clinic/practice, and health system.

RESULTS

Across all systems, 56% of patients had a documented smoking status in 2017. Among nearly 1 million smokers, 24% quit smoking. In the regression model, patient characteristics associated with quitting included older age, Hispanic ethnicity, being married, urban residence, commercial insurance, pregnancy, and a diagnosis of pneumonia, myocardial infarction, ischemic heart disease, cataract, or asthma. Medicaid insurance, low income, high BMI, peripheral vascular disease, alcohol-related diagnosis, and COPD were negatively associated with smoking cessation. Adjusted quit rates ranged from 14.3% to 34.5% across 20 health systems, 5% to 66% among 1,399 practice sites, and 4% to 87% among 3,803 health-care providers. Of smokers, 10.2% were prescribed smoking deterrents, and 3.9% were referred for counseling.

INTERPRETATION

Smoking cessation rates varied substantially at the practitioner, practice site, and health system levels. It is likely that individual physician approaches to smoking cessation influence patients' likelihood of quitting.

Clinician engagement in research as a path toward the learning health system: A regional survey across the northwestern United States

Introduction:

Increased research engagement of frontline, community-based clinicians could result in greater research relevancy, increased likelihood of implementation into practice, and improved health care for patients. Establishment of learning health systems within health-care organizations may facilitate this process.

Methods:

In 2016, the U.S. Northwest Participant and Clinical Interactions Network conducted a region-wide survey in four community-based health systems to identify barriers to clinician involvement in research and understand clinician interest and levels of engagement.

Results:

Survey responses indicated broad interest in research’s value to patients (77% of respondents), contribution to clinical evidence (79%), and fulfillment of intellectual curiosity (77%). Engagement was not always correlated with interest. Top barriers included time (65%), support (34%), and getting started (32%).

Conclusion:

In community health systems in the northwestern United States, clinician interest in research exists but with several significant barriers. Leveraging the learning health system movement may be one way to increase focus on research and address identified barriers.

An Adult Immunization Best Practices Learning Collaborative: Impact, Scale Up, and Spread

The research objective was to rapidly scale up and spread a proven learning collaborative approach (intervention) for adult vaccination rates for influenza and pneumococcal disease from 7 to 39 US health care organizations and to examine improvement in adult immunization rates after scale-up. Comparative analyses were conducted between intervention and nonintervention propensity score-matched providers on vaccination rates using a difference-in-differences approach. Qualitative data, collected during site visits and in-person and virtual meetings, were used to enhance understanding of quantitative results. In 2017–2018, an analysis of a subset of sites (n = 9) from 2 intervention cohorts (∼20 sites each) demonstrated greater improvement than their matched providers in pneumococcal vaccinations (PV) for patients ages ≥65 years (treatment effect range: 1.4%-3.7%, P < 0.01) and PV for high-risk patients (eg, with immunocompromising conditions) aged 19–64 years (0.8%-1.6%, P < 0.01). Significant effects were observed in one of the study cohorts for PV for at-risk patients (eg, with diabetes) aged 19–64 years (1.7%, P < 0.01), and influenza vaccination rates (2.4%, P < 0.001). Individual health systems demonstrated even greater improvements across all 4 vaccinations: 9.5% influenza; 8.7% PV ages ≥65 years; 11.8% PV high-risk; 16.3% PV at-risk (all P < 0.01). Results demonstrated that a 7-site pilot could be successfully scaled to 39 additional sites, with similar improvements in vaccination rates. Between 2014 and 2018, vaccination improvements among all 46 groups (7 pilot, 39 in subsequent cohorts) resulted in an estimated 5.5 million adult vaccinations administered or documented in 27 states.

Impact of a Learning Collaborative Approach on Influenza and Pneumococcal Immunization Rates in US Adults: A Mixed Methods Approach

The objective of this research was to test the impact of a learning collaborative model (intervention) on adult vaccination rates for influenza and pneumococcal disease. A mixed methods approach was used to identify changes in adult vaccination rates over time and organizational factors contributing to successful programs. Provider-level propensity scores were used to match intervention to non-intervention providers to control for inherent selection bias of participating organizations. Comparative analyses were conducted between intervention and non-intervention sites on vaccination rates, using a difference-in-differences approach. Qualitative data (eg, semi-structured interviews) were analyzed using a constant comparison approach to identify themes related to successful strategies. From 2014-2016, intervention providers demonstrated greater improvement than their matched providers in pneumococcal vaccinations (PV) for patients aged 65 years and older (treatment effect: 4.3%, P < 0.05) and PV for high-risk patients (eg, with immunocompromising conditions) aged 19-64 years (2.7%, P < 0.001). Significant effects were also observed for PV for at-risk patients (eg, with diabetes) aged 19-64 years (1.7%, P < 0.05). Individual health systems demonstrated even greater improvements (eg, greater increase in PV rates for patients aged 65 years and older), with treatment effects as high as 20.4% (P < 0.05). A learning collaborative approach was demonstrated to be an effective approach to improve adult vaccination rates among participating integrated delivery systems and medical groups. Factors associated with success included organization type (ie, integrated delivery systems) and systems characterized by a positive learning climate and collaborative culture.

Using EHR audit trail logs to analyze clinical workflow: A case study from community-based ambulatory clinics

To develop a workflow-supported clinical documentation system, it is a critical first step to understand clinical workflow. While Time and Motion studies has been regarded as the gold standard of workflow analysis, this method can be resource consuming and its data may be biased due to the cognitive limitation of human observers. In this study, we aimed to evaluate the feasibility and validity of using EHR audit trail logs to analyze clinical workflow. Specifically, we compared three known workflow changes from our previous study with the corresponding EHR audit trail logs of the study participants. The results showed that EHR audit trail logs can be a valid source for clinical workflow analysis, and can provide an objective view of clinicians' behaviors, multi-dimensional comparisons, and a highly extensible analysis framework.

Healthy by Design: Using a Gender Focus to Influence Complete Streets Policy

BACKGROUND

Public health leaders in Yellowstone County, Montana, formed an alliance to address community-wide issues. One such issue is Complete Streets, with its vision of safe streets for all. This case study focuses on development and adoption of a Complete Streets policy. It examines how a community coalition, Healthy By Design, infused a gender focus into the policymaking process.

METHODS

An incremental and nonlinear policymaking process was aided by a focus on gender and health equity. The focus on a large constituency helped to frame advocacy in terms of a broad population's needs, not just special interests.

RESULTS

The city council unanimously adopted a Complete Streets resolution, informed by a gender lens. Healthy By Design further used gender information to successfully mobilize the community in response to threats of repeal of the policy, and then influenced the adoption of a revised policy.

CONCLUSIONS

Policies developed with a focus on equity, including gender equity, may have broader impact on the community. Such policies may pave the way for future policies that seek to transform gender norms toward building a healthier community for all residents.

Intent-to-treat analysis of a simultaneous multisite telehealth diabetes prevention program

OBJECTIVE

Determine the effectiveness of a 16-week modified diabetes prevention program (DPP) administered simultaneously to multiple rural communities from a single urban site, as compared with a similar face-to-face intervention. A 12-week intervention was evaluated to consider minimization of staff costs in communities where resources are limited.

RESEARCH DESIGN AND METHODS

A prospective cohort study compared DPP interventions implemented in rural (via telehealth technology) and urban (face-to-face) communities using an intent-to-treat analysis. Primary outcome measures included 5% and 7% body weight loss. Logistic regression analyses were used to determine predictors of intervention success and included a variable for treatment effect.

RESULTS

Between 2010 and 2015, up to 667 participants were enrolled in the study representing one urban and 15 rural communities across Montana. The 16-week urban and rural interventions were comparable; 33.5% and 34.6% of participants lost 7% body weight, respectively; 50% and 47% lost 5% (p=0.22). Participants who were male (OR=2.41; 95% CI 1.32 to 4.40), had lower baseline body mass index (OR=1.03; 95% CI 1.01 to 1.07), attended more sessions (OR=1.33; 95% CI 1.11 to 1.58), and more frequently reported (OR=3.84; 95% CI 1.05 to 14.13) and met daily fat gram (OR=4.26; 95% CI 1.7 to 10.6) and weekly activity goals (OR=2.46; 95% CI 1.06 to 5.71) were more likely to meet their 7% weight loss goal. Predictors of meeting weight loss goals were similar for participants enrolled in the 12-week intervention.

CONCLUSIONS

Using telehealth technology to administer a modified DPP to multiple rural communities simultaneously demonstrated weight loss results comparable to those in a face-to-face intervention. Given the limitation of resources, linking rural areas to urban centers using telemedicine may increase access to much needed services to prevent or delay progression to diabetes.

Learning: Contemplating the unexamined core of Learning Health Systems

INTRODUCTION

The current Learning Health Systems literature affords insufficient attention to the process of learning. In response, Billings Clinic focused on how to advance its learning capabilities and subsequently to contribute new insights into the process of learning to the LHS literature.

METHODS

An environmental scan was conducted, including the grey literature (eg, technical reports and white papers) and peer-reviewed research publications. Semistructured interviews were also conducted with Clinic staff members to determine the motivation of their engagement in meaningful quality improvement, or learning, initiatives.

RESULTS

Six learning principles emerged from the literature review and staff interviews: (1) draw on wisdom of groups and value connections; (2) embrace sensemaking over decision making in dealing with the unexpected; (3) bring diverse perspectives to complex challenges; (4) animate people, provide direction, update regularly, and interact respectfully; (5) appreciate the power and ubiquity of emergent change and the limitations of planned change; and (6) concentrate on small wins and characterize challenges as mere problems. Examples of how these principles are beginning to influence how learning and improvement are understood and approached at Billings Clinic are described and serve as illustrations of the principles in action.

CONCLUSION

Becoming adept in learning is essential to realizing the vision of Learning Health Systems-to harness science, clinical research, and information generated by digital technology to inform and accelerate improvement in quality health care. This article seeks to contribute to greater understanding of this process by sharing a set of principles that are proving useful at one health care organization and to a more comprehensive conceptualization of Learning Health Systems.

Improving Blood Pressure Control Using Smart Technology

BACKGROUND

The authors sought to determine if wireless oscillometric home blood pressure monitoring (HBPM) that integrates with smartphone technology improves blood pressure (BP) control among patients with new or existing uncontrolled hypertension (HTN).

METHODS

A prospective observational cohort study monitored BP control before and after an educational intervention and introduction to HBPM. Patients in the intervention group were instructed to track their BP using a smartphone device three to seven times per week. Cases were matched to controls at a 1:3 allocation ratio on several clinical characteristics over the same period and received usual care. The proportion of patients with controlled BP was compared between groups at pre- and postintervention, ∼9 months later. Results and Materials: The total study population included 484 patients with mean age 60 years (range 23-102 years), 47.7% female, and 84.6% Caucasian. Mean preintervention BP was 137.8 mm Hg systolic and 81.4 mm Hg diastolic. Mean BP control rates improved for patients who received HBPM from 42% to 67% compared with matched control patients who improved from 59% to 67% (p < 0.01).

CONCLUSION

HBPM with smartphone technology has the potential to improve HTN management among patients with uncontrolled or newly diagnosed HTN. Technology needs to be easy to use and operate and would work best when integrated into local electronic health record systems. In systems without this capability, medical assistants or other personnel may be trained to facilitate the process. Nurse navigator involvement was instrumental in bridging communication between the patients and provider.

Life after PACE (Program of All-Inclusive Care for the Elderly): A retrospective/prospective, qualitative analysis of the impact of closing a nurse practitioner centered PACE site

BACKGROUND AND PURPOSE

Caring for frail older adults is a significant healthcare concern as the frailest 10% of the population account for over 70% of healthcare expenditures. Research reveals the use of comprehensive models, such as Program of All-Inclusive Care for the Elderly (PACE), leads to improved functional outcomes for participants and cost savings through decreased utilization. This study examines how closing a PACE program impacts health outcomes of previously enrolled participants.

METHODS

Data were collected every 6 months for 2 years via phone surveys on 34 former participants enrolled in the program at the time of the closure. The survey included questions regarding satisfaction with care, activities of daily living (ADLs), instrumental ADLs (IADLs), emergency department (ED) visits, hospitalizations, and use of home health services. Deaths and nursing home placements were monitored. Outcomes were compared during and post-PACE.

CONCLUSIONS

Higher numbers of ED visits, hospitalizations, and nursing home placements occurred post-PACE. PACE/post-PACE differences in ADL and IADL scores were not significant, nor were death rates. Higher satisfaction existed with PACE versus non-PACE care.

IMPLICATIONS FOR PRACTICE

Comprehensive care programs such as PACE are effective in reducing healthcare utilization, thus limiting costs. Further work is required to maintain, develop, and support comprehensive models similar to PACE.

A qualitative analysis of patient and family perspectives of palliative care

BACKGROUND

To provide truly patient-centered palliative care services, there is a need to better understand the perspectives and experiences of patients and families. Increased understanding will provide insight into the development of health care team competencies and organizational changes necessary to improve patient care.

OBJECTIVE

Our aim was to explore patient and family perceptions of palliative care services at the end of life or during serious illness and to identify facilitators and barriers to receipt of palliative care services.

METHODS

In-depth, semi-structured patient and family interviews were conducted, transcribed, and independently reviewed using grounded theory methodology and preliminary interpretations. A combined deductive and inductive iterative qualitative approach was used to identify recurring themes. The study was conducted in a physician-led, not-for-profit, multispecialty integrated health system serving three large, western, rural states. A purposive sample of 14 individuals who received palliative care services were interviewed alone or with their families for a total of 12 interviews.

RESULTS

Presence, Reassurance, and Honoring Choices emerged as central themes linked to satisfaction with palliative care services. Themes were defined as including health care professional attributes of respect, approachability, genuineness, empathy, connectedness, compassion, sensitivity, an ability to listen, good communication, provision of information, empowerment, and timeliness. Honoring Choices included those pertaining to treatment, spirituality, and family needs.

CONCLUSIONS

At end of life or during times of serious illness, patients and families identified behaviors of Presence, Reassurance, and Honoring Choices as important. According to patients/families, health care providers must be compassionate and empathetic and possess skills in listening, connecting, and interacting with patients and families.

Telemedicine and the mini-mental state examination: assessment from a distance

The objective of this study was to determine the reliability of the Mini-Mental State Examination (MMSE) administration via telehealth with a focus on the auditory and visual test components. Reliability was assessed through use of an in-person collaborator and by assessment of faxed test copies. The MMSE was administered via telehealth with the assistance of a face-to-face collaborator. Patient responses were recorded by both the remote and in-person nurse and compared item by item; total scores for each subject were also compared. Visual items were assessed through a blinded separate scoring of a faxed copy. Percent agreement per item and total score were calculated and correlations between scores were determined by Pearson correlation coefficients. Mean score differences and associated 95% confidence intervals were calculated. Eighty percent of individual items demonstrated remote to in-person agreement of >95% and all items were >85.5% in agreement. Pearson correlation coefficients demonstrated high correlations (>0.86) between 80% of the items examined. Mean differences in scored test items were not significantly different from zero. This study demonstrates the utility of using telehealth for cognitive assessment by MMSE. It supports the use of telehealth to improve healthcare access among patients for whom distance, cost, and mobility are potential barriers to attending face-to-face clinical visits. Continued validation and reliability testing is warranted to ensure that all healthcare provided via telehealth maintains an equal quality level to that of in-person care.

Beyond health information technology: critical factors necessary for effective diabetes disease management

BACKGROUND

Electronic health records (EHRs) have been implemented throughout the United States with varying degrees of success. Past EHR implementation experiences can inform health systems planning to initiate new or expand existing EHR systems. Key "critical success factors," e.g., use of disease registries, workflow integration, and real-time clinical guideline support, have been identified but not fully tested in practice.

METHODS

A pre/postintervention cohort analysis was conducted on 495 adult patients selected randomly from a diabetes registry and followed for 6 years. Two intervention phases were evaluated: a "low-dose" period targeting primary care provider (PCP) and patient education followed by a "high-dose" EHR diabetes management implementation period, including a diabetes disease registry and office workflow changes, e.g., diabetes patient preidentification to facilitate real-time diabetes preventive care, disease management, and patient education.

RESULTS

Across baseline, "low-dose," and "high-dose" postintervention periods, a significantly greater proportion of patients (a) achieved American Diabetes Association (ADA) guidelines for control of blood pressure (26.9 to 33.1 to 43.9%), glycosylated hemoglobin (48.5 to 57.5 to 66.8%), and low-density lipoprotein cholesterol (33.1 to 44.4 to 56.6%) and (b) received recommended preventive eye (26.2 to 36.4 to 58%), foot (23.4 to 40.3 to 66.9%), and renal (38.5 to 53.9 to 71%) examinations or screens.

CONCLUSIONS

Implementation of a fully functional, specialized EHR combined with tailored office workflow process changes was associated with increased adherence to ADA guidelines, including risk factor control, by PCPs and their patients with diabetes. Incorporation of previously identified "critical success factors" potentially contributed to the success of the program, as did use of a two-phase approach.

The economic and clinical impact of an inpatient palliative care consultation service: a multifaceted approach

BACKGROUND

While there has been a rapid increase of inpatient palliative care (PC) programs, the financial and clinical benefits have not been well established.

OBJECTIVE

Determine the effect of an inpatient PC consultation service on costs and clinical outcomes.

DESIGN

Multifaceted study included: (1) interrupted time-series design utilizing mean daily costs preintervention and postintervention; (2) matched cohort analysis comparing PC to usual care patients; and (3) analysis of symptom control after consultation.

SETTING

Large private, not-for-profit, academic medical center in San Francisco, California, 2004-2006.

SUBJECTS

Time series analysis included 282 PC patients; matched cohorts included 27 PC with 128 usual care patients; clinical outcome analysis of 48 PC patients.

MAIN OUTCOME MEASURE(S)

Mean daily patient costs and length of stay (LOS); pain, dyspnea, and secretions assessment scores.

RESULTS

Mean daily costs were reduced 33% (p < 0.01) from preintervention to postintervention period. Mean length of stay (LOS) was reduced 30%. Mean daily costs for PC patients were 14.5% lower compared to usual care patients (p < 0.01). Pain, dyspnea, and secretions scores were reduced by 86%, 64%, and 87%, respectively. Over the study period, time to PC referral as well as overall ALOS were reduced by 50%.

CONCLUSIONS

The large reduction in mean daily costs and LOS resulted in an estimated annual savings of $2.2 million in the study hospital. Our results extend the evidence base of financial and clinical benefits associated with inpatient PC programs. We recommend additional study of best practices for identifying patients and providing consultation services, in addition to progressive management support and reimbursement policy.

An Evaluation of the Advanced Illness Management (AIM) Program: Increasing Hospice Utilization in the San Francisco Bay Area

BACKGROUND

In the Medicare population, end-of-life care is fragmented and hospice is underutilized. Evidence suggests that hospice care is associated with increased patient/family satisfaction and reductions in overall health care costs. Advanced Illness Management (AIM) is a home-based program established to ease the transition between curative and comfort care for seriously ill patients who lack coordinated hospital, home health, and hospice care.

OBJECTIVE

Measure the impact of the AIM program on the discharge disposition of participating home health patients.

DESIGN

Retrospective cohort study.

SETTING/SUBJECTS

A total of 435 patients who received home health services through the Sutter Visiting Nurse Association and Hospice (SVNA&H) and who were discharged between 2003 and September 2005 were included in this study. Participants were selected based on a life expectancy of 6 months or less and additionally matched on prognosis of current episode of illness and symptom status. Intervention patients were compared to those receiving usual home health care at two SVNA&H branches.

RESULTS

In the within-branch comparison, a 28% difference was observed in the number of hospice referrals between patients who received the AIM intervention and Usual Care I (47% AIM; 33% UCI, p = 0.003). When patients receiving the AIM intervention were compared to patients from another branch, Usual Care II, a 67% difference emerged (47% AIM; 16% UCII, p < 0.0001). Among African American patients, 60% and 73% more patients were referred to hospice when AIM patients were compared to Usual Care I and Usual Care II patients, respectively (p < 0.01). These differences persisted after controlling for symptom status, patient demographics and home health length of stay.

CONCLUSIONS

The AIM program was successful at increasing hospice utilization through a targeted intervention focused on palliative and end-of-life care, increased patient education and decision making, and a dynamic treatment approach. The finding of increased utilization by African Americans, a population traditionally reluctant to use hospice, was particularly noteworthy.

The effect of parity-induced copayment reductions on adolescent utilization of substance use services

OBJECTIVE

The purpose of this study was to determine if the reduction in copayment amount by a large self-insured state employer increased utilization of adolescent services. Specifically, the study sought to discover if the number of unique adolescent users of substance use outpatient services increased as a result of reductions in cost-sharing arrangements.

METHOD

The data utilized in this study were 31,585 records from administrative claims data on utilization of mental health and substance abuse services from members of a state indemnity plan fromJuly 1998 through December 2001, translating to 36 months of pre-intervention data and 6 months of postintervention data. Monthly longitudinal data before and after benefit design change were analyzed using a quasi-experimental time series design, using Box and Jenkins' autoregressive, integrated, moving-average time-series modeling methods. The primary outcome measure was the number of unique users of services.

RESULTS

The hypothesis that service utilization would increase following the implementation of a reduction in copayment amount (the intervention) was supported in these analyses for adolescents' substance use service utilization. A significant increase in the number of unique adolescent users of substance use services was detected in the month following the intervention (p < .01).

CONCLUSIONS

The results of this study suggest that a reduction in adolescents' substance use service copayment requirements to a level equal to those for general medical services may be a step toward assuring full parity between such types of services. These findings provide potentially important information regarding the possible effects of broader policy changes, as parity in benefit design is a common component of laws that attempt to ensure "full parity."

Unemployment and the detection of early stage breast tumors among African Americans and non-Hispanic whites

PURPOSE

To test the hypothesis that high unemployment predicts reduced detection of local breast tumors among African American and non-Hispanic white women in the Detroit, Michigan and Atlanta, Georgia SEER catchment areas.

METHODS

We test the hypothesis with data for the 156 months from January 1985 through December 1997.

RESULTS

In situ and local breast tumors in African American and non-Hispanic white women were less likely to be detected during periods of high unemployment.

CONCLUSIONS

Contracting labor markets may impede women with symptoms from getting proper medical attention or distract women from discovering symptoms they would otherwise detect. African American women appear at greatest risk of having a tumor going undetected by virtue of labor market performance.

Reexamining the prevalence of Chlamydia trachomatis infection among gay men with urethritis: implications for STD policy and HIV prevention activities

BACKGROUND

Evidence of an STD-HIV interaction and the availability of noninvasive urine-based screening tests have resulted in an increased focus on chlamydial infections in men.

GOAL

To evaluate the prevalence of chlamydial infections among men with urethritis at the San Francisco City Clinic (SFCC).

STUDY DESIGN

In 1997, male SFCC patients diagnosed with urethritis were tested for chlamydia using urine-based ligase chain reaction and for gonorrhea using urethral culture.

RESULTS

Gonorrhea was identified in 45% of men who have sex with men (MSM) versus 26% of men who have sex with women (MSW). Among men with gonorrhea, chlamydia coinfection was found among 15.2% of MSM and 8.4% of MSW. Among men with nongonococcal urethritis, 18% and 20% of MSM and MSW had chlamydial infection, respectively. Young age was associated with chlamydial infection in MSM.

CONCLUSION

After a period of low chlamydial infection rates in MSM during the pre-AIDS era, infection rates are increasing among this population. SFCC's revised clinical practice guidelines include chlamydia testing of MSM with urethritis.

Evaluation of chlamydia and gonorrhea screening criteria: San Francisco sexually transmitted disease clinic: 1997 to 1998

BACKGROUND AND OBJECTIVES

The advent of more sensitive diagnostic testing technologies and competition in public healthcare spending have resulted in a reevaluation of sexually transmitted disease (STD) screening practices in an attempt to target populations at greatest risk. Screening among populations with a < 2% prevalence of chlamydia and a < 1% prevalence of gonorrhea may not be cost-effective.

GOAL

To identify subpopulations with a low prevalence of chlamydia or gonorrhea.

STUDY DESIGN

The prevalence of genital chlamydia and gonorrhea among asymptomatic STD patients screened from 1997 to 1998 at San Francisco City Clinic was stratified by demographic and behavioral risk factors.

RESULTS

The prevalence of chlamydia and gonorrhea was 3.4% and 1.1% among asymptomatic women and 4.0% and 1.0% among asymptomatic men, respectively. Two low-prevalence subpopulations identified among asymptomatic patients were women older than 29 years (chlamydia, 1.2%) and men who have sex with women (gonorrhea, 0.8%).

CONCLUSIONS

These data identified low-prevalence subpopulations among asymptomatic STD patients. As a result, the STD screening criteria at San Francisco City Clinic were changed accordingly.

Comparisons of cost and accuracy of DNA probe test and culture for the detection of Neisseria gonorrhoeae in patients attending public sexually transmitted disease clinics in Los Angeles County

BACKGROUND

Strict handling and transport requirements for the successful use of culture in the detection of Neisseria gonorrhoeae warrant investigation of accurate and cost-effective test alternatives such as the Gen-Probe PACE 2 DNA probe assay (Gen-Probe, Inc., San Diego, CA).

STUDY DESIGN

The Gen-Probe PACE 2 DNA probe assay for N. gonorrhoeae was compared with conventional culture methods in the principal Los Angeles County (LAC) Department of Health Services (DHS) Public Health Laboratory and three of its branch laboratories. Urethral and endocervical samples were collected from 1,566 patients (921 males; 645 females) attending six LAC DHS sexually transmitted disease clinics. Cost analysis was performed comparing material and labor costs of the two test methods.

RESULTS

The overall prevalence based on culture was 11.8% (15.7% for males; 6.4% for females). Nine samples were culture positive, Gen-Probe negative and four samples were culture negative, Gen-Probe positive and remained discordant after discrepant analysis. The sensitivity and specificity were 94.6% and 99.7%, respectively, for the PACE 2 assay compared with culture. The positive and negative predictive values were 97.8% and 99.3%, respectively. No statistically significant difference was found between the two tests. A cost analysis found an average cost of $3.11/test for culture and $3.85/test for PACE 2, given the approximate 12% disease prevalence in this population.

CONCLUSIONS

Gen-Probe's PACE 2 assay may provide an acceptable, cost-effective alternative to culture, especially among high-risk males.