Development and Validation of the Life Expectancy Estimator for Older Adults with Diabetes (LEAD): the Diabetes and Aging Study.

BACKGROUND: Estimated life expectancy for older patients with diabetes informs decisions about treatment goals, cancer screening, long-term and advanced care, and inclusion in clinical trials. Easily implementable, evidence-based, diabetes-specific approaches for identifying patients with limited life expectancy are needed. OBJECTIVE: Develop and validate an electronic health record (EHR)-based tool to identify older adults with diabetes who have limited life expectancy. DESIGN: Predictive modeling based on survival analysis using Cox-Gompertz models in a retrospective cohort. PARTICIPANTS: Adults with diabetes aged ≥ 65 years from Kaiser Permanente Northern California: a 2015 cohort (N = 121,396) with follow-up through 12/31/2019, randomly split into training (N = 97,085) and test (N = 24,311) sets. Validation was conducted in the test set and two temporally distinct cohorts: a 2010 cohort (n = 89,563; 10-year follow-up through 2019) and a 2019 cohort (n = 152,357; 2-year follow-up through 2020). MAIN MEASURES: Demographics, diagnoses, utilization and procedures, medications, behaviors and vital signs; mortality. KEY RESULTS: In the training set (mean age 75 years; 49% women; 48% racial and ethnic minorities), 23% died during 5 years follow-up. A mortality prediction model was developed using 94 candidate variables, distilled into a life expectancy model with 11 input variables, and transformed into a risk-scoring tool, the Life Expectancy Estimator for Older Adults with Diabetes (LEAD). LEAD discriminated well in the test set (C-statistic = 0.78), 2010 cohort (C-statistic = 0.74), and 2019 cohort (C-statistic = 0.81); comparisons of observed and predicted survival curves indicated good calibration. CONCLUSIONS: LEAD estimates life expectancy in older adults with diabetes based on only 11 patient characteristics widely available in most EHRs and claims data. LEAD is simple and has potential application for shared decision-making, clinical trial inclusion, and resource allocation.

Medical Assistant Protocol Improves Disparities in Depression Screening Rates.

INTRODUCTION: Depression is a prevalent condition for which screening rates remain low and disparities in screening exist. This study examines the impacts of a medical assistant screening protocol on the rates of depression screening, overall and by sociodemographic groups, in a primary care setting. METHODS: Between September 2016 and August 2018, a quasi-experimental study of adult primary care visits was conducted at an urban academic clinic to ascertain the change in the rates of completion of the Patient Health Questionnaire-2 after the implementation of a medical assistant protocol (intervention) versus that of physician-only screening (control arm). Analyses were conducted between April 2019 and April 2020 and used interrupted time-series models with generalized estimating equations. RESULTS: A total of 45,157 visits by 21,377 unique patients were included. Overall, screening increased from 18% (physician-only screening) to 57% (medical assistant protocol) (p<0.0001). Screening increased for all measured demographics. With physician screening, depression screening was less likely to occur at visits by women (than at visits by men; OR=0.91, 95% CI=0.85, 0.98) and at visits by Black/African American patients (than at visits by White; OR=0.91, 95% CI=0.84, 0.99). However, with the medical assistant protocol, depression screening was more likely to occur at visits by women (than at visits by men; OR=1.07, 95% CI=1.0002, 1.14) and at visits by Black/African American patients (than at visits by White; OR=1.11, 95% CI=1.02, 1.20). In addition, age-related disparities were mitigated for visits by patients aged 40-64 and ≥65 years (e.g., age ≥65 years: physician, OR=0.66, 95% CI=0.59, 0.73; medical assistant protocol, OR=0.78, 95% CI=0.71, 0.85), compared with visits by patients aged 18-39 years. CONCLUSIONS: Implementation of a medical assistant protocol in a primary care setting may significantly increase depression screening rates while mitigating or removing sociodemographic disparities.

Artificial Intelligence Can Cut Costs While Maintaining Accuracy in Colorectal Cancer Genotyping.

Rising cancer care costs impose financial burdens on health systems. Applying artificial intelligence to diagnostic algorithms may reduce testing costs and avoid wasteful therapy-related expenditures. To evaluate the financial and clinical impact of incorporating artificial intelligence-based determination of mismatch repair/microsatellite instability status into the first-line metastatic colorectal carcinoma setting, we developed a deterministic model to compare eight testing strategies: A) next-generation sequencing alone, B) high-sensitivity polymerase chain reaction or immunohistochemistry panel alone, C) high-specificity panel alone, D) high-specificity artificial intelligence alone, E) high-sensitivity artificial intelligence followed by next generation sequencing, F) high-specificity artificial intelligence followed by next-generation sequencing, G) high-sensitivity artificial intelligence and high-sensitivity panel, and H) high-sensitivity artificial intelligence and high-specificity panel. We used a hypothetical, nationally representative, population-based sample of individuals receiving first-line treatment for de novo metastatic colorectal cancer (N = 32,549) in the United States. Model inputs were derived from secondary research (peer-reviewed literature and Medicare data). We estimated the population-level diagnostic costs and clinical implications for each testing strategy. The testing strategy that resulted in the greatest project cost savings (including testing and first-line drug cost) compared to next-generation sequencing alone in newly-diagnosed metastatic colorectal cancer was using high-sensitivity artificial intelligence followed by confirmatory high-specificity polymerase chain reaction or immunohistochemistry panel for patients testing negative by artificial intelligence ($400 million, 12.9%). The high-specificity artificial intelligence-only strategy resulted in the most favorable clinical impact, with 97% diagnostic accuracy in guiding genotype-directed treatment and average time to treatment initiation of less than one day. Artificial intelligence has the potential to reduce both time to treatment initiation and costs in the metastatic colorectal cancer setting without meaningfully sacrificing diagnostic accuracy. We expect the artificial intelligence value proposition to improve in coming years, with increasing diagnostic accuracy and decreasing costs of processing power. To extract maximal value from the technology, health systems should evaluate integrating diagnostic histopathologic artificial intelligence into institutional protocols, perhaps in place of other genotyping methodologies.

The Productivity Requirements of Implementing a Medical Scribe Program.

BACKGROUND: Economic analyses of medical scribes have been limited to individual, specialty-specific clinics. OBJECTIVE: To determine the number of additional patient visits various specialties would need to recover the costs of implementing scribes in their practice at 1 year. DESIGN: Modeling study based on 2015 data from the Centers for Medicare & Medicaid Services (CMS) and National Ambulatory Medical Care Survey. Scribe costs were based on literature review and a third-party contractor model. Revenue was calculated from direct visit billing, CPT (Current Procedural Terminology) billing, and data from the National Ambulatory Medical Care Survey. DATA SOURCES: 2015 data from CMS and the National Ambulatory Medical Care Survey. TARGET POPULATION: Health care providers. TIME HORIZON: 1 year. PERSPECTIVE: Office-based clinic. OUTCOME MEASURES: The number of additional patient visits a physician must have to recover the costs of a scribe program at 1 year. RESULTS OF BASE-CASE ANALYSIS: An average of 1.34 additional new patient visits per day (295 per year) were required to recover scribe costs (range, 0.89 [cardiology] to 1.80 [orthopedic surgery] new patient visits per day). For returning patients, an average of 2.15 additional visits per day (472 per year) were required (range, 1.65 [cardiology] to 2.78 [orthopedic surgery] returning visits per day). The addition of 2 new patient (or 3 returning) visits per day was profitable for all specialties. RESULTS OF SENSITIVITY ANALYSIS: Results were not sensitive to most inputs, with the exception of hourly scribe cost and inclusion of CPT revenue. LIMITATION: Use of Medicare data and failure to account for indirect costs, downstream revenue, or changes in documentation quality. CONCLUSION: For all specialties, modest increases in productivity due to scribes may allow physicians to see more patients and offset scribe costs, making scribe programs revenue-neutral. PRIMARY FUNDING SOURCE: University of Chicago Medicine's Center for Healthcare Delivery Science and Innovation and the Bucksbaum Institute.

Use and Discontinuation of Insulin Treatment Among Adults Aged 75 to 79 Years With Type 2 Diabetes.

Importance: Among older individuals with type 2 diabetes, those with poor health have greater risk and derive less benefit from tight glycemic control with insulin. Objective: To examine whether insulin treatment is used less frequently and discontinued more often among older individuals with poor health compared with those in good health. Design, Setting, and Participants: This longitudinal cohort study included 21 531 individuals with type 2 diabetes followed for up to 4 years starting at age 75 years. Electronic health record data from the Kaiser Permanente Northern California Diabetes Registry was collected to characterize insulin treatment and glycemic control over time. Data were collected from January 1, 2009, through December 31, 2017, and analyzed from February 2, 2018, through June 30, 2019. Exposures: Health status was defined as good (<2 comorbid conditions or 2 comorbidities but physically active), intermediate (>2 comorbidities or 2 comorbidities and no self-reported weekly exercise), or poor (having end-stage pulmonary, cardiac, or renal disease; diagnosis of dementia; or metastatic cancer). Main Outcomes and Measures: Insulin use prevalence at age 75 years and discontinuation among insulin users over the next 4 years (or 6 months prior to death if <4 years). Results: Of 21 531 patients, 10 396 (48.3%) were women, and the mean (SD) age was 75 (0) years. Nearly one-fifth of 75-year-olds (4076 [18.9%]) used insulin. Prevalence and adjusted risk ratios (aRRs) of insulin use at age 75 years were higher in individuals with poor health (29.4%; aRR, 2.03; 95% CI, 1.87-2.20; P < .01) and intermediate health (27.5%; aRR, 1.85; 95% CI, 1.74-1.97; P < .01) relative to good health (10.5% [reference]). One-third (1335 of 4076 [32.7%]) of insulin users at age 75 years discontinued insulin within 4 years of cohort entry (and at least 6 months prior to death). Likelihood of continued insulin use was higher among individuals in poor health (aRR, 1.47; 95% CI, 1.27-1.67; P < .01) and intermediate health (aRR, 1.16; 95% CI, 1.05-1.30; P < .01) compared with good health (reference). These same prevalence and discontinuation patterns were present in the subset with tight glycemic control (hemoglobin A1c <7.0%). Conclusions and Relevance: In older individuals with type 2 diabetes, insulin use was most prevalent among those in poor health, whereas subsequent insulin discontinuation after age 75 years was most likely in healthier patients. Changes are needed in current practice to better align with guidelines that recommend reducing treatment intensity as health status declines.

Validation of the Computerized Adaptive Test for Mental Health in Primary Care.

PURPOSE: The US Preventive Services Task Force recommends screening for depression in the general adult population. Although screening questionnaires for depression and anxiety exist in primary care settings, electronic health tools such as computerized adaptive tests based on item response theory can advance screening practices. This study evaluated the validity of the Computerized Adaptive Test for Mental Health (CAT-MH) for screening for major depressive disorder (MDD) and assessing MDD and anxiety severity among adult primary care patients. METHODS: We approached 402 English-speaking adults for participation from a primary care clinic, of whom 271 adults (71% female, 65% black) participated. Participants completed modules from the CAT-MH (Computerized Adaptive Diagnostic Test for MDD, CAT-Depression Inventory, CAT-Anxiety Inventory); brief paper questionnaires (9-item Patient Health Questionnaire [PHQ-9], 2-item Patient Health Questionnaire [PHQ-2], Generalized Anxiety Disorder 7-item Scale [GAD-7]); and a reference-standard interview, the Structured Clinical Interview for DSM-5 (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition) Diagnoses. RESULTS: On the basis of the interview, 31 participants met criteria for MDD and 29 met criteria for GAD. The diagnostic accuracy of the Computerized Adaptive Diagnostic Test for MDD (area under curve [AUC] = 0.85) was similar to that of the PHQ-9 (AUC = 0.84) and higher than that of the PHQ-2 (AUC = 0.76) for MDD screening. Using the interview as the reference standard, the accuracy of the CAT-Anxiety Inventory (AUC = 0.93) was similar to that of the GAD-7 (AUC = 0.97) for assessing anxiety severity. The patient-preferred screening method was assessment via tablet/computer with audio. CONCLUSIONS: Computerized adaptive testing could be a valid and efficient patient-centered screening strategy for depression and anxiety screening in primary care settings.

Cost-effectiveness of gastric band surgery for overweight but not obese adults with type 2 diabetes in the U.S.

AIM: To determine the cost-effectiveness of gastric band surgery in overweight but not obese people who receive standard diabetes care. METHOD: A microsimulation model (United Kingdom Prospective Diabetes Study outcomes model) was used to project diabetes outcomes and costs from a two-year Australian randomized trial of gastric band (GB) surgery in overweight but not obese people (BMI 25 to 30kg/m2) on to a comparable population of U.S. adults from the National Health and Nutrition Examination Survey (N=254). Estimates of cost-effectiveness were calculated based on the incremental cost-effectiveness ratios (ICERs) for different treatment scenarios. Costs were inflated to 2015 U.S. dollar values and an ICER of less than $50,000 per QALY gained was considered cost-effective. RESULTS: The incremental cost-effectiveness ratio for GB surgery at two years exceeded $90,000 per quality-adjusted life year gained but decreased to $52,000, $29,000 and $22,000 when the health benefits of surgery were assumed to endure for 5, 10 and 15 years respectively. The cost-effectiveness of GB surgery was sensitive to utility gained from weight loss and, to a lesser degree, the costs of GB surgery. However, the cost-effectiveness of GB surgery was affected minimally by improvements in HbA1c, systolic blood pressure and cholesterol. CONCLUSIONS: GB surgery for overweight but not obese people with T2D appears to be cost-effective in the U.S. setting if weight loss endures for more than five years. Health utility gained from weight loss is a critical input to cost-effectiveness estimates and therefore should be routinely measured in populations undergoing bariatric surgery.

Racial and Ethnic Disparities in Diabetes Screening Between Asian Americans and Other Adults: BRFSS 2012-2014.

BACKGROUND: Although Asian Americans are at high risk for type 2 diabetes, it is not known whether they are appropriately screened for this disease. OBJECTIVE: To assess racial and ethnic disparities in diabetes screening between Asian Americans and other adults. DESIGN: Analysis of pooled cross-sectional data from 45 U.S. states and territories using the 2012-2014 Behavioral Risk Factor Surveillance System. We calculated the weighted proportions of adults in each racial and ethnic group who received recommended diabetes screening. To assess for racial and ethnic disparities, we used multivariable logistic regression to model receipt of recommended diabetes screening as a function of race and ethnicity, adjusting for demographics, healthcare access, survey year, and state. PARTICIPANTS: A total of 526,000 adults who were eligible to receive diabetes screening according to American Diabetes Association guidelines from 2012 to 2014 (age ≥ 45 years or age < 45 years with a body mass index [BMI] ≥ 25 kg/m2). MAIN MEASURES: Self-reported receipt of diabetes screening (defined as a test for high blood sugar or diabetes within the past 3 years) and self-reported race/ethnicity (non-Hispanic white, non-Hispanic Asian, non-Hispanic Pacific Islander, non-Hispanic American Indian or Alaskan Native, non-Hispanic black, Hispanic or Latino, and non-Hispanic multiracial or other). KEY RESULTS: Asian Americans were the least likely racial and ethnic group to receive recommended diabetes screening. Overall, Asian Americans had 34% lower adjusted odds of receiving recommended diabetes screening compared to non-Hispanic whites (95 % CI: 0.60, 0.73). In subgroup analyses by age and weight status, disparities were widest among obese Asian Americans ≥ 45 years (AOR = 0.56; 95 % CI: 0.39, 0.81). Disparities persisted among Asian Americans who completed other types of preventive cancer screening. CONCLUSIONS: Despite their high risk of diabetes, Asian Americans were the least likely racial and ethnic group to receive recommended diabetes screening.

Strategies for improving cardiovascular health in women with diabetes mellitus: a review of the evidence.

Knowledge about cardiovascular (CV) disease in women with diabetes mellitus (DM) has changed substantially over the past 20 years. Coronary artery disease, strokes, and peripheral vascular disease affect women with DM at higher rates than the general population of women. Lifestyle therapies, such as dietary changes, physical activity, and smoking cessation, offer substantial benefits to women with DM. Of the pharmacotherapies, statins offer the most significant benefits but may not be well tolerated in some women. Aspirin may also benefit high-risk women. Other pharmacotherapies, such as fibrates, ezetimibe, niacin, fish oil, and hormone replacement therapy, remain unproven and, in some cases, potentially dangerous to women with DM. To reduce CV events, risks to women with DM must be better publicized and additional research must be done. Finally, advancements in health care delivery must target high-risk women with DM to lower risk factors and effectively improve cardiovascular health.

Ethnic Differences in Geriatric Conditions and Diabetes Complications Among Older, Insured Adults With Diabetes: The Diabetes and Aging Study.

OBJECTIVE: The aim of this study was to evaluate ethnic differences in burden of prevalent geriatric conditions and diabetic complications among older, insured adults with diabetes. METHOD: An observational study was conducted among 115,538 diabetes patients, aged ≥60, in an integrated health care system with uniform access to care. RESULTS: Compared with Whites, Asians and Filipinos were more likely to be underweight but had substantively lower prevalence of falls, urinary incontinence, polypharmacy, depression, and chronic pain, and were least likely of all groups to have at least one geriatric condition. African Americans had significantly lower prevalence of incontinence and falls, but higher prevalence of dementia; Latinos had a lower prevalence of falls. Except for end-stage renal disease (ESRD), Whites tended to have the highest rates of prevalent diabetic complications. DISCUSSION: Among these insured older adults, ethnic health patterns varied substantially; differences were frequently small and rates were often better among select minority groups, suggesting progress toward the Healthy People 2020 objective to reduce health disparities.

Health care utilization and receipt of preventive care for patients seen at federally funded health centers compared to other sites of primary care.

OBJECTIVE: To compare utilization and preventive care receipt among patients of federal Section 330 health centers (HCs) versus patients of other settings. DATA SOURCES: A nationally representative sample of adults from the Medical Expenditure Panel Survey (2004-2008). STUDY DESIGN: HC patients were defined as those with ≥50 percent of outpatient visits at HCs in the first panel year. Outcomes included utilization and preventive care receipt from the second panel year. We used negative binomial and logistic regression models with propensity score adjustment for confounding differences between HC and non-HC patients. PRINCIPAL FINDINGS: Compared to non-HC patients, HC patients had fewer office visits (adjusted incidence rate ratio [aIRR], 0.63) and hospitalizations (aIRR, 0.43) (both p < .001). HC patients were more likely to receive breast cancer screening than non-HC patients (adjusted odds ratio [aOR] 2.78, p < .01). In subgroup analyses, uninsured HC patients had fewer outpatient and emergency room visits and were more likely to receive dietary advice and breast cancer screening compared to non-HC patients. CONCLUSIONS: Health centers add value to the health care system by providing socially and medically disadvantaged patients with care that results in lower utilization and maintained or improved preventive care.

Does a higher frequency of difficult patient encounters lead to lower quality care?

BACKGROUND: Difficult patient encounters in the primary care office are frequent and are associated with physician burnout. However, their relationship to patient care outcomes is not known. OBJECTIVE: To determine the effect of difficult encounters on patient health outcomes and the role of physician dissatisfaction and burnout as mediators of this effect. DESIGN: A total of 422 physicians were sorted into 3 clusters based on perceived frequency of difficult patient encounters in their practices. Patient charts were audited to assess the quality of hypertension and diabetes management and preventive care based on national guidelines. Summary measures of quality and errors were compared among the 3 physician clusters. RESULTS: Of the 1384 patients, 359 were cared for by high-cluster physicians (those who had a high frequency of difficult encounters), 871 by medium-cluster physicians, and 154 by low-cluster physicians. Dissatisfaction and burnout were higher among physicians reporting higher frequencies of difficult encounters. However, quality of patient care and management errors were similar across all 3 groups. CONCLUSIONS: Physician perception of frequent difficult encounters was not associated with worse patient care quality or more medical errors. Future studies should investigate whether other patient outcomes, including acute care and patient satisfaction, are affected by difficult encounters.

Classification of older adults who have diabetes by comorbid conditions, United States, 2005-2006.

INTRODUCTION: Older adults who have diabetes vary widely in terms of comorbid conditions; these conditions help determine the risks and benefits of intensive glycemic control. Not all people benefit from intensive glycemic control. The objective of this study was to classify by comorbid conditions older American adults who have diabetes to identify those who are less likely to benefit from intensive glycemic control. METHODS: We used latent class analysis to identify subgroups of a nationally representative sample of community-dwelling older adults (aged 57-85 y) who have diabetes (n = 750). The subgroups were classified according to 14 comorbid conditions prevalent in the older population. Using the Akaike Information Criterion, the Bayesian Information Criterion (BIC), the sample-size adjusted BIC, and the χ(2) goodness-of-fit statistic, we assessed model fit. RESULTS: We found 3 distinct subgroups. Class 1 (63% of the sample) had the lowest probabilities for most conditions. Class 2 (29% of the sample) had the highest probabilities of cancer, incontinence, and kidney disease. Class 3 (9% of the sample) had the highest probabilities (>90%) of congestive heart failure and myocardial infarction. Class 1 had only 0, 1, or 2 comorbid conditions, and both class 2 and class 3 had 6 or more comorbid conditions. The 5-year death rates for class 2 (17%) and class 3 (33%) were higher than the rate for class 1 (9%). CONCLUSION: Older adults who have diabetes, cardiovascular disease, and 6 or more comorbid conditions may represent a subgroup of older adults who are less likely to benefit from intensive glycemic control.

Impact of delaying blood pressure control in patients with Type 2 diabetes: results of a decision analysis.

BACKGROUND: In patients with diabetes, delays in controlling blood pressure are common, but the harms of delays have not been quantified. OBJECTIVE: To estimate the harms of delays in controlling systolic blood pressure in middle-aged adults with newly diagnosed Type 2 diabetes. DESIGN: Decision analysis using diabetes complication equations from the United Kingdom Prospective Diabetes Study (UKPDS). PARTICIPANTS: Hypothetical population of adults aged 50 to 59 years old with newly diagnosed Type 2 diabetes based on characteristics from the National Health and Nutrition Examination Surveys. INTERVENTION: Delays in lowering systolic blood pressure from 150 (uncontrolled) to 130 mmHg (controlled). MAIN MEASURES: Lifetime complication rates (amputation, congestive heart failure, end-stage renal disease, ischemic heart disease, myocardial infarction, and stroke), average life expectancy and quality-adjusted life expectancy (QALE). KEY RESULTS: Compared to a lifetime of controlled blood pressure, a lifetime of uncontrolled blood pressure increased complications by 1855 events per 10,000 patients and decreased QALE by 332 days. A 1-year delay increased complications by 14 events per 10,000 patients and decreased QALE by 2 days. A 10-year delay increased complications by 428 events per 10,000 patients and decreased QALE by 145 days. Among complications, rates of stroke and myocardial infarction increased to the greatest extent due to delays. With a 20-year delay in achieving controlled blood pressure, a baseline blood pressure of 160 mmHg decreased QALE by 477 days, whereas a baseline of 140 mmHg decreased QALE by 142 days. CONCLUSIONS: Among middle-aged adults with diabetes, the harms of a 1-year delay in controlling blood pressure may be small; however, delays of ten years or more are expected to lower QALE to the same extent as smoking in patients with cardiovascular disease.

A resident-led quality improvement initiative to improve obesity screening.

Instruction on quality improvement (QI) methods is required as part of residency education; however, there is limited evidence regarding whether internal medicine residents can improve patient care using these methods. Because obesity screening is not done routinely in clinical practice, residents aimed to improve screening using QI techniques. Residents streamlined body mass index (BMI) documentation, created educational materials about obesity, and launched an obesity screening QI initiative in a residency clinic. Residents designed plan-do-study-act cycles focused on increasing awareness and maintaining improvements in screening over a 1-year period. Documentation rates were collected at baseline, 2 weeks, 6 months, and 1 year post-intervention. At 1 year, obesity treatment rates also were collected. BMI documentation rates after 1 year were higher than baseline (43% vs 4%, P < .0001). In obese patients, BMI documentation was associated with lifestyle counseling (34% vs 14%, P < .01). An internal medicine resident-led QI project targeting obesity can improve screening.