Multi-Institutional Validation of a Mammography-Based Breast Cancer Risk Model.

PURPOSE: Accurate risk assessment is essential for the success of population screening programs in breast cancer. Models with high sensitivity and specificity would enable programs to target more elaborate screening efforts to high-risk populations, while minimizing overtreatment for the rest. Artificial intelligence (AI)-based risk models have demonstrated a significant advance over risk models used today in clinical practice. However, the responsible deployment of novel AI requires careful validation across diverse populations. To this end, we validate our AI-based model, Mirai, across globally diverse screening populations. METHODS: We collected screening mammograms and pathology-confirmed breast cancer outcomes from Massachusetts General Hospital, USA; Novant, USA; Emory, USA; Maccabi-Assuta, Israel; Karolinska, Sweden; Chang Gung Memorial Hospital, Taiwan; and Barretos, Brazil. We evaluated Uno's concordance index for Mirai in predicting risk of breast cancer at one to five years from the mammogram. RESULTS: A total of 128,793 mammograms from 62,185 patients were collected across the seven sites, of which 3,815 were followed by a cancer diagnosis within 5 years. Mirai obtained concordance indices of 0.75 (95% CI, 0.72 to 0.78), 0.75 (95% CI, 0.70 to 0.80), 0.77 (95% CI, 0.75 to 0.79), 0.77 (95% CI, 0.73 to 0.81), 0.81 (95% CI, 0.79 to 0.82), 0.79 (95% CI, 0.76 to 0.83), and 0.84 (95% CI, 0.81 to 0.88) at Massachusetts General Hospital, Novant, Emory, Maccabi-Assuta, Karolinska, Chang Gung Memorial Hospital, and Barretos, respectively. CONCLUSION: Mirai, a mammography-based risk model, maintained its accuracy across globally diverse test sets from seven hospitals across five countries. This is the broadest validation to date of an AI-based breast cancer model and suggests that the technology can offer broad and equitable improvements in care.

Impact of Education and Peer Comparison on Antibiotic Prescribing for Pediatric Respiratory Tract Infections.

INTRODUCTION: Inappropriate prescribing of broad-spectrum antibiotics is a significant modifiable risk factor for the development of antibiotic resistance. The objective was to improve guideline-concordant care for 3 common acute respiratory tract infections (ARTIs) and to reduce broad-spectrum antibiotic prescribing in ambulatory pediatric patients. METHODS: Quality measures were developed for 3 ARTIs: viral upper respiratory infection (URI), acute bacterial sinusitis (ABS), and acute otitis media (AOM). Among 22 pediatric clinics, a collaborative of 10 was identified for intervention using baseline data for each ARTI, and 3 plan-do-study-act cycles were planned and completed. Outcomes included guideline-concordant antibiotic utilization and broad-spectrum antibiotic prescribing percentage (BSAP%). Comparison in number of diagnoses for the ARTI measures and total antibiotic prescribing over time served as balancing measures. RESULTS: Collaborative clinics had baseline medians for appropriate or first-line treatment of 70% for URI, 53% for ABS, and 36% for AOM. To reach targets for URI, ABS, and AOM required 6, 14, and 18 months, respectively. At 42 months, performance for all 3 ARTIs remained ≥90%. BSAP% decreased from a baseline of 57% to 34% at 24 months. There was a limited effect from financial incentives but a significant decrease was noted in total antibiotic utilization. Diagnosis shifting may have occurred for URI and ABS while the rates for diagnoses for AOM declined over time. CONCLUSIONS: Through education and peer comparison feedback, guideline-concordant care for 3 ARTIs in collaborative clinics improved and remained beyond above targets and was accompanied by reductions in BSAP% and total antibiotic prescribing.

Quality of care for secondary prevention for patients with coronary heart disease: results of the Hastening the Effective Application of Research through Technology (HEART) trial.

BACKGROUND: Effective therapies for reducing mortality rates in persons with coronary heart disease (CHD) remain underused. We report the results of an effectiveness trial of a quality improvement effort to increase the use of 3-hydroxy-3methylglutaryl coenzyme A (HMG CoA) reductase inhibitors, beta-blockers, and angiotensin-converting enzyme (ACE) inhibitors in patients with CHD in a network-model managed-care setting. METHODS: Patients with CHD were identified by searching a claims database. The use of therapies was assessed by linkage with a pharmacy database. An intervention, consisting of a guideline summary, peer comparison performance feedback, and patient specific chart reminders was evaluated in a randomized, practice-based effectiveness trial. RESULTS: Data were available for >700 patients per year (1999-2002) in 131 practices. At baseline (1999), 55% of patients were receiving HMG CoA reductase inhibitors, 39% of patients were receiving beta-blockers, and 24% of patients were receiving ACE inhibitors. The use of all 3 types of medications increased steadily with time, with the exception of a decrease in the use of HMG CoA reductase inhibitors in the final year (2002). No difference in medication use was observed between randomized groups. CONCLUSIONS: The observed pattern of care supports the contention that the quality of outpatient care for secondary prevention of CHD improved from 1999 to 2002 in this setting. The basis for the inconsistent pattern of use of HMG CoA reductase inhibitors is not certain, but may relate to concerns about toxicity. Centralized mailings of guideline summaries, performance feedback reports, and chart reminders had no observable impact on quality of care in this setting. More intensive intervention may be required to improve the quality of outpatient care for the secondary prevention of CHD.