Individualized Comparative Effectiveness of Models Optimizing Patient Safety and Resident Education.

Title: Sleep duration (hours of sleep) as measured with an actigraph Time frame: Measured daily for 14 days

Title: Behavioral alertness as measured with Psychomotor vigilance test Time frame: Measured daily for 14 days

Title: Self perceived sleepiness as measured by the Karolinska Sleepiness Score Time frame: Measured daily for 14 days

Title: Time (hours per day) spent in direct patient care Description: Time spent in direct patient care over 2-4 wks as measured by a trained observer shadowing the intern Time frame: measured daily over 2-4 wks

Title: Trainee satisfaction with education Description: Self-reported satisfaction with education as assessed thru survey questions Time frame: measured at baseline and at end of intervention year

Title: Program director satisfaction with trainee education Description: Self-reported satisfaction with trainee education as assessed thru survey questions Time frame: measured at baseline and end of intervention year

Patient safety and costs:

Title: Total costs of patient care as measured by total Medicare payments Time frame: Measured over year of trial

Inclusion criteria

• Current internal medicine program in the United States Continued Accreditation status with the ACGME - Program director agrees to:

• Randomization to one of the two study arms.
• Develop, and share with the study team, institutional duty hour schedules and regulatory policies/procedures that will ensure adherence to, and enforcement of, the duty hour regulations that apply to their study arm
• Allow access and analysis of de-identified resident duty hour adherence data/logs, call schedules, and rotation schedules to the Study Team.
• Participate, and encourage trainee participation at your institution, in beginning and end of year iCOMPARE surveys.

Exclusion criteria

• The investigators excluded 119 programs that comprise the bottom 50% in resident-to-bed ratio and the bottom 25% in patient volume related to the diagnoses by which the patient population will be selected for evaluation of safety outcomes.
• Within the 260 programs that remain, the investigators excluded the 65 in the lowest quartile of program size to ensure we can feasibly obtain sufficient trainee measurements.
• The 195 remaining programs are eligible for inclusion.
• The investigators have also excluded children and VA hospitals.