Early prediction of end-stage kidney disease using electronic health record data: a machine learning approach with a 2-year horizon.

Objectives: In the United States, end-stage kidney disease (ESKD) is responsible for high mortality and significant healthcare costs, with the number of cases sharply increasing in the past 2 decades. In this study, we aimed to reduce these impacts by developing an ESKD model for predicting its occurrence in a 2-year period. Materials and Methods: We developed a machine learning (ML) pipeline to test different models for the prediction of ESKD. The electronic health record was used to capture several kidney disease-related variables. Various imputation methods, feature selection, and sampling approaches were tested. We compared the performance of multiple ML models using area under the ROC curve (AUCROC), area under the Precision-Recall curve (PR-AUC), and Brier scores for discrimination, precision, and calibration, respectively. Explainability methods were applied to the final model. Results: Our best model was a gradient-boosting machine with feature selection and imputation methods as additional components. The model exhibited an AUCROC of 0.97, a PR-AUC of 0.33, and a Brier score of 0.002 on a holdout test set. A chart review analysis by expert physicians indicated clinical utility. Discussion and Conclusion: An ESKD prediction model can identify individuals at risk for ESKD and has been successfully deployed within our health system.

Behavioural nudges increase COVID-19 vaccinations.

Enhancing vaccine uptake is a critical public health challenge1. Overcoming vaccine hesitancy2,3 and failure to follow through on vaccination intentions3 requires effective communication strategies3,4. Here we present two sequential randomized controlled trials to test the effect of behavioural interventions on the uptake of COVID-19 vaccines. We designed text-based reminders that make vaccination salient and easy, and delivered them to participants drawn from a healthcare system one day (first randomized controlled trial) (n = 93,354 participants; clinicaltrials number NCT04800965) and eight days (second randomized controlled trial) (n = 67,092 individuals; clinicaltrials number NCT04801524) after they received a notification of vaccine eligibility. The first reminder boosted appointment and vaccination rates within the healthcare system by 6.07 (84%) and 3.57 (26%) percentage points, respectively; the second reminder increased those outcomes by 1.65 and 1.06 percentage points, respectively. The first reminder had a greater effect when it was designed to make participants feel ownership of the vaccine dose. However, we found no evidence that combining the first reminder with a video-based information intervention designed to address vaccine hesitancy heightened its effect. We performed online studies (n = 3,181 participants) to examine vaccination intentions, which revealed patterns that diverged from those of the first randomized controlled trial; this underscores the importance of pilot-testing interventions in the field. Our findings inform the design of behavioural nudges for promoting health decisions5, and highlight the value of making vaccination easy and inducing feelings of ownership over vaccines.

Patient Navigation After Positive Fecal Immunochemical Test Results Increases Diagnostic Colonoscopy and Highlights Multilevel Barriers to Follow-Up.

BACKGROUND: The fecal immunochemical test (FIT) is a common colorectal cancer screening modality in the USA but often is not followed by diagnostic colonoscopy. AIMS: We investigated the efficacy of patient navigation to increase diagnostic colonoscopy after positive FIT results and determined persistent barriers to follow-up despite navigation in a large, academic healthcare system. METHODS: The study cohort included all health system outpatients with an assigned primary care provider, a positive FIT result between 12/01/2016 and 06/01/2019, and no documentation of colonoscopy after positive FIT. Two non-clinical patient navigators engaged patients and providers to encourage follow-up, offer solutions to barriers, and assist with colonoscopy scheduling. The primary intervention endpoint was completion of colonoscopy within 6 months of navigation. We documented reasons for persistent barriers to colonoscopy despite navigation and determined predictors of successful follow-up after navigation. RESULTS: There were 119 patients who received intervention. Of these, 37 (31.1%) patients completed colonoscopy at 6 months. In 41/119 (34.5%) cases, the PCP did not recommend colonoscopy, most commonly due to a normal colonoscopy prior to the positive FIT (19, 46.3%). There were 41/119 patients (34.5%) that declined colonoscopy despite the patient navigator and the PCP order. Male sex and younger age were significant predictors of follow-up (aOR = 2.91, 95%CI, 1.18-7.13; aOR = 0.92, 95%CI, 0.87-0.99). CONCLUSIONS: After implementation of patient navigation, diagnostic colonoscopy was completed for 31.1% of patients with a positive FIT result. However, navigation also highlighted persistent multilevel barriers to follow-up. Future work will develop targeted solutions for these barriers to further increase FIT follow-up rates in our health system.

Application of Behavioral Economics Principles Improves Participation in Mailed Outreach for Colorectal Cancer Screening.

INTRODUCTION: Colorectal cancer (CRC) is a major cause of cancer-related morbidity and mortality in the United States. Although various interventions have improved screening rates, they often require abundant resources and can be difficult to implement. Social psychology and behavioral economics principles offer an opportunity for low-cost and easy-to-implement strategies but are less common in clinical settings. METHODS: We randomized 2,000 patients aged 50-75 years eligible for CRC screening to one of the 2 mailed interventions: a previously used text-based letter describing and offering fecal immunochemical testing (FIT) and colonoscopy (usual care arm); or a letter leveraging social psychology and behavioral economics principles (e.g., implied scarcity and choice architecture), minimal text, and multiple images to offer FIT and colonoscopy (intervention arm). We compared total screening uptake, FIT uptake, and colonoscopy uptake at 1-month intervals in each group. RESULTS: There were 1,882 patients included in the final analysis. The mean age was 69.3 years, and baseline characteristics in the 2 groups were similar. Screening completion at 26 weeks was 19.5% in the usual care arm (16.3% FIT vs 3.2% colonoscopy, P < 0.01) and 24.1% in the intervention arm (22.1% FIT vs 2.0% colonoscopy, P < 0.01) (P = 0.02). DISCUSSION: Among primary care patients aged 50-75 years in an academic setting, mailed CRC outreach employing social psychology and behavioral economics principles led to a higher participation in CRC screening than usual care mailed outreach. TRANSLATIONAL IMPACT: Mailed interventions to increase CRC screening should incorporate social psychology and behavioral economics principles to improve participation.