Evaluation of the Predictive Value of Routinely Collected Health-Related Social Needs Measures.

The objective was to assess the value of routinely collected patient-reported health-related social needs (HRSNs) measures for predicting utilization and health outcomes. The authors identified Mayo Clinic patients with cancer, diabetes, or heart failure. The HRSN measures were collected as part of patient-reported screenings from June to December 2019 and outcomes (hospitalization, 30-day readmission, and death) were ascertained in 2020. For each outcome and disease combination, 4 models were used: gradient boosting machine (GBM), random forest (RF), generalized linear model (GLM), and elastic net (EN). Other predictors included clinical factors, demographics, and area-based HRSN measures-area deprivation index (ADI) and rurality. Predictive performance for models was evaluated with and without the routinely collected HRSN measures as change in area under the curve (AUC). Variable importance was also assessed. The differences in AUC were mixed. Significant improvements existed in 3 models of death for cancer (GBM: 0.0421, RF: 0.0496, EN: 0.0428), 3 models of hospitalization (GBM: 0.0372, RF: 0.0640, EN: 0.0441), and 1 of death (RF: 0.0754) for diabetes, and 1 model of readmissions (GBM: 0.1817), and 3 models of death (GBM: 0.0333, RF: 0.0519, GLM: 0.0489) for heart failure. Age, ADI, and the Charlson comorbidity index were the top 3 in variable importance and were consistently more important than routinely collected HRSN measures. The addition of routinely collected HRSN measures resulted in mixed improvement in the predictive performance of the models. These findings suggest that existing factors and the ADI are more important for prediction in these contexts. More work is needed to identify predictors that consistently improve model performance.

Association of Patient and System-Level Factors With Social Determinants of Health Screening.

BACKGROUND: Health systems are increasingly recognizing the importance of collecting social determinants of health (SDoH) data. However, gaps remain in our understanding of facilitators or barriers to collection. To address these gaps, we evaluated a real-world implementation of a SDoH screening tool. METHODS: We conducted a retrospective analysis of the implementation of the SDoH screening tool at Mayo Clinic in 2019. The outcomes are: (1) completion of screening and (2) the modality used (MyChart: filled out on patient portal; WelcomeTablet: filled out by patient on a PC-tablet; EpicCare: data obtained directly by provider and entered in chart). We conducted logistic regression for completion and multinomial logistic regression for modality. The factors of interest included race and ethnicity, use of an interpreter, and whether the visit was for primary care. RESULTS: Overall, 58.7% (293,668/499,931) of screenings were completed. Patients using interpreters and racial/ethnic minorities were less likely to complete the screening. Primary care visits were associated with an increase in completion compared with specialty care visits. Patients who used an interpreter, racial and ethnic minorities, and primary care visits were all associated with greater WelcomeTablet and lower MyChart use. CONCLUSION: Patient and system-level factors were associated with completion and modality. The lower completion and greater WelcomeTablet use among patients who use interpreters and racial and ethnic minorities points to the need to improve screening in these groups and that the availability of the WelcomeTablet may have prevented greater differences. The higher completion in primary care visits may mean more outreach is needed for specialists.

Utilization of an Electronic Health Record Integrated Risk Score to Predict Hospitalization Among COVID-19 Patients.

OBJECTIVE: To evaluate the performance of an Electronic Health Record (EHR) integrated risk score for COVID-19 positive outpatients to predict 30-day risk of hospitalization. PATIENTS AND METHODS: A retrospective observational study of 67 470 patients with COVID-19 confirmed by polymerase chain reaction (PCR) test between March 12, 2020 and February 8, 2021. Risk scores were calculated based on data in the chart at the time of the incident infection. RESULTS: The Mayo Clinic COVID-19 risk score consisted of 13 components included age, sex, chronic lung disease, congenital heart disease, congestive heart failure, coronary artery disease, diabetes mellitus, end stage liver disease, end stage renal disease, hypertension, immune compromised, nursing home resident, and pregnant. Univariate analysis showed all components, except pregnancy, have significant (P < .001) association with admission. The Mayo Clinic COVID-19 risk score showed a Receiver Operating Characteristic Area Under Curve (AUC) of 0.837 for the prediction of admission for this large cohort of COVID-19 positive patients. CONCLUSION: The Mayo Clinic COVID-19 risk score is a simple score that is easily integrated into the EHR with excellent predictive performance for severe COVID-19. It can be leveraged to stratify risk for severe COVID-19 at initial contact, when considering therapeutics or in the allocation of vaccine supply.

PROMIS Scales for Assessment of Persistent Post-COVID Symptoms: A Cross Sectional Study.

OBJECTIVE: Persistent post-COVID symptoms are estimated to occur in up to 10% of patients who have had COVID-19. These lingering symptoms may persist for weeks to months after resolution of the acute illness. This study aimed to add insight into our understanding of certain post-acute conditions and clinical findings. The primary purpose was to determine the persistent post COVID impairments prevalence and characteristics by collecting post COVID illness data utilizing Patient-Reported Outcomes Measurement Information System (PROMIS®). The resulting measures were used to assess surveyed patients physical, mental, and social health status. METHODS: A cross-sectional study and 6-months Mayo Clinic COVID recovered registry data were used to evaluate continuing symptoms severity among the 817 positive tested patients surveyed between March and September 2020. The resulting PROMIS® data set was used to analyze patients post 30 days health status. The e-mailed questionnaires focused on fatigue, sleep, ability to participate in social roles, physical function, and pain. RESULTS: The large sample size (n = 817) represented post hospitalized and other managed outpatients. Persistent post COVID impairments prevalence and characteristics were determined to be demographically young (44 years), white (87%), and female (61%). Dysfunction as measured by the PROMIS® scales in patients recovered from acute COVID-19 was reported as significant in the following domains: ability to participate in social roles (43.2%), pain (17.8%), and fatigue (16.2%). CONCLUSION: Patient response on the PROMIS® scales was similar to that seen in multiple other studies which used patient reported symptoms. As a result of this experience, we recommend utilizing standardized scales such as the PROMIS® to obtain comparable data across the patients' clinical course and define the disease trajectory. This would further allow for effective comparison of data across studies to better define the disease process, risk factors, and assess the impact of future treatments.

Functional Outcomes and Health-Related Quality of Life Before and After Primary Total Knee Replacement for Patients From Diverse Geographic Regions.

BACKGROUND: The purpose of this study was to investigate whether patient-reported knee function and health status before and after primary total knee replacement (TKR) at an academic medical center differs among patients from diverse geographic regions. METHODS: We assessed patient-reported outcomes (PROs) as measured by the Oxford Knee Score (OKS) and EuroQoL-5D (EQ-5D) in 2855 TKR patients at preprimary and one year postprimary TKR procedure between January 1, 2012 and June 30, 2014. We compared the demographic characteristics, response rates, and changes in OKS, EQ-5D, and EQ visual analog scale among local, regional, and national patient groups. Patient- or hospital-related predictors of the postoperative scores were identified after controlling for preoperative scores on the PRO measures. RESULTS: Local patients had more comorbid conditions. Groups were similar in clinical outcomes such as length of stay at hospital, complication, and reoperation rates. Local, regional, and national patients had similar response rates and reported a similar level of knee function and health-related quality of life before and after TKR. Eighty nine percent had clinically important improvement on OKS, 69% on EQ-5D index, and 28% EQ visual analog scale, and there were no differences among groups in the proportions of clinically meaningful change. Better postoperative PRO scores were associated with older age, shorter hospital stay, fewer comorbidities, nonsmoking status, fewer complications, and private health insurance. CONCLUSION: Some patient characteristics such as comorbidities were greater in the local patient cohort versus the national cohort. Nevertheless, clinical outcomes and PROs were comparable across all geographic tiers.

Inclusion of short-term care patients affects the perceived performance of specialists: a retrospective cohort study.

BACKGROUND: Current publicly reported quality performance measures directly compare primary care to specialty care. Specialists see short-term patients referred due to poor control of their disease who then return to their local provider. Our study looked to determine if outcomes measured in short-term care patients differed from those in long-term care patients and what impact those differences may have on quality performance profiles for specialists. METHODS: Retrospective cohort from a large academic medical Center. Performance was measured as "Optimal Care"--all or none attainment of goals. Patients with short-term care (<90 days contact) versus long-term care (>90 days contact) were evaluated for both specialty and primary care practices during the year 2008. RESULTS: Patients with short-term care had significantly lower "Optimal Care": 7.2% vs. 19.7% for optimal diabetes care in endocrinology and 41.3% vs. 53.1% for optimal ischemic vascular disease care in cardiology (p < 0.001). Combining short and long term care patients lowered overall perceived performance for the specialty practice. CONCLUSIONS: Factors other than quality affect the perceived performance of the specialty practice. Extending current primary care quality measurement to short-term specialty care patients without adjustment produces misleading results.

Comparison of the quality of patient referrals from physicians, physician assistants, and nurse practitioners.

OBJECTIVE: To compare the quality of referrals of patients with complex medical problems from nurse practitioners (NPs), physician assistants (PAs), and physicians to general internists. PATIENTS AND METHODS: We conducted a retrospective comparison study involving regional referrals to an academic medical center from January 1, 2009, through December 31, 2010. All 160 patients referred by NPs and PAs combined and a random sample of 160 patients referred by physicians were studied. Five experienced physicians blinded to the source of referral used a 7-item instrument to assess the quality of referrals. Internal consistency, interrater reliability, and dimensionality of item scores were determined. Differences between item scores for patients referred by physicians and those for patients referred by NPs and PAs combined were analyzed by using multivariate ordinal logistical regression adjusted for patient age, sex, distance of the referral source from Mayo Clinic, and Charlson Index. RESULTS: Factor analysis revealed a 1-dimensional measure of the quality of patient referrals. Interrater reliability (intraclass correlation coefficient for individual items: range, 0.77-0.93; overall, 0.92) and internal consistency for items combined (Cronbach α=0.75) were excellent. Referrals from physicians were scored higher (percentage of agree/strongly agree responses) than were referrals from NPs and PAs for each of the following items: referral question clearly articulated (86.3% vs 76.0%; P=.0007), clinical information provided (72.6% vs 54.1%; P=.003), documented understanding of the patient's pathophysiology (51.0% vs 30.3%; P<.0001), appropriate evaluation performed locally (60.3% vs 39.0%; P<.0001), appropriate management performed locally (53.5% vs 24.1%; P<.0001), and confidence returning patient to referring health care professional (67.8% vs 41.4%; P<.0001). Referrals from physicians were also less likely to be evaluated as having been unnecessary (30.1% vs 56.2%; P<.0001). CONCLUSION: The quality of referrals to an academic medical center was higher for physicians than for NPs and PAs regarding the clarity of the referral question, understanding of pathophysiology, and adequate prereferral evaluation and documentation.

The SAGE Guideline Model: achievements and overview.

The SAGE (Standards-Based Active Guideline Environment) project was formed to create a methodology and infrastructure required to demonstrate integration of decision-support technology for guideline-based care in commercial clinical information systems. This paper describes the development and innovative features of the SAGE Guideline Model and reports our experience encoding four guidelines. Innovations include methods for integrating guideline-based decision support with clinical workflow and employment of enterprise order sets. Using SAGE, a clinician informatician can encode computable guideline content as recommendation sets using only standard terminologies and standards-based patient information models. The SAGE Model supports encoding large portions of guideline knowledge as re-usable declarative evidence statements and supports querying external knowledge sources.

Use of declarative statements in creating and maintaining computer-interpretable knowledge bases for guideline-based care.

Developing computer-interpretable clinical practice guidelines (CPGs) to provide decision support for guideline-based care is an extremely labor-intensive task. In the EON/ATHENA and SAGE projects, we formulated substantial portions of CPGs as computable statements that express declarative relationships between patient conditions and possible interventions. We developed query and expression languages that allow a decision-support system (DSS) to evaluate these statements in specific patient situations. A DSS can use these guideline statements in multiple ways, including: (1) as inputs for determining preferred alternatives in decision-making, and (2) as a way to provide targeted commentaries in the clinical information system. The use of these declarative statements significantly reduces the modeling expertise and effort required to create and maintain computer-interpretable knowledge bases for decision-support purpose. We discuss possible implications for sharing of such knowledge bases.

Women with diabetes have poorer control of blood pressure than men.

OBJECTIVE: Diabetes is associated with a higher coronary heart disease (CHD) mortality in women compared with men. Less aggressive control of the CHD risk factors in women can contribute to this excess mortality. Because hypertension has a high prevalence in subjects with diabetes, we compared the control of this risk factor between men and women. METHODS: This was a retrospective cohort study comparing blood pressure levels and trends over a 1-year period between men and women with diabetes receiving primary care. Using a chronic disease registry database, subjects with type 1 and type 2 diabetes, aged >or=18, were identified for inclusion. Mean weighted systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated for subjects with multiple longitudinal readings. Subjects were classified into four blood pressure control categories based on the mean weighted blood pressure. Comparisons were made between men and women before and after controlling for baseline characteristics. RESULTS: A total of 3181 subjects (46% women) were included in the study. More women than men were in the moderate and severely elevated blood pressure categories (40% and 6% compared with 32% and 5%, respectively, p<0.001). The unadjusted mean SBP was 3 mm Hg higher in women (139 mm Hg in women compared with 136 in men, p<0.001). These differences remained significant after controlling for baseline variables. CONCLUSIONS: In subjects with diabetes receiving medical care, women had poorer control of blood pressure and a significantly higher mean SBP compared with men. These findings might partially explain the excess CHD mortality in women with diabetes.

Modeling guidelines for integration into clinical workflow.

The success of clinical decision-support systems requires that they are seamlessly integrated into clinical workflow. In the SAGE project, which aims to create the technological infra-structure for implementing computable clinical practice guide-lines in enterprise settings, we created a deployment-driven methodology for developing guideline knowledge bases. It involves (1) identification of usage scenarios of guideline-based care in clinical workflow, (2) distillation and disambiguation of guideline knowledge relevant to these usage scenarios, (3) formalization of data elements and vocabulary used in the guideline, and (4) encoding of usage scenarios and guideline knowledge using an executable guideline model. This methodology makes explicit the points in the care process where guideline-based decision aids are appropriate and the roles of clinicians for whom the guideline-based assistance is intended. We have evaluated the methodology by simulating the deployment of an immunization guideline in a real clinical information system and by reconstructing the workflow context of a deployed decision-support system for guideline-based care. We discuss the implication of deployment-driven guideline encoding for sharability of executable guidelines.

Multiple cardiovascular complications in a patient with relapsing polychondritis.

Relapsing polychondritis is an uncommon disease of unknown etiology, usually manifested by inflammatory changes of cartilaginous tissues. Cardiovascular complications are rare but have been associated with adverse prognosis. Aortitis, vasculitis of large- and medium-sized arteries with aneurysm formation, valvulitis, pericarditis, and atrioventricular conduction disturbances have been reported as late complications of relapsing polychondritis. We describe a 42-year-old man who developed all the known cardiovascular complications of relapsing polychondritis except for clinically evident pericarditis. This case illustrates the multiple, varied, and potentially fatal cardiovascular complications that can occur with this disorder. Patients with relapsing polychondritis should be monitored closely for development of such complications.