Hospital-Level Care at Home for Acutely Ill Adults: A Randomized Controlled Trial

Brief Summary

Detailed Description

Hospitals are the standard of care for acute illness in the United States, but hospital care is expensive and often unsafe, especially for older individuals. While admitted, 20% suffer delirium, over 5% contract hospital-acquired infections, and most lose functional status that is never regained. Timely access to inpatient care is poor: many hospital wards are typically over 100% capacity, and emergency department waits can be protracted. Moreover, hospital care is increasingly costly: many internal medicine admissions have a negative margin (i.e., expenditures exceed hospital revenues) and incur patient debt. The investigators propose a home hospital model of care that substitutes for treatment in an acute care hospital. Studies of the home hospital model have demonstrated that a sizeable proportion of acute care can be delivered in the home with equal quality and safety, 20% reduced cost, and 20% improved patient experience. While this is the standard of care in several developed countries, only 2 non-randomized demonstration projects have been conducted in the United States, each with highly local needs. Taken together, home hospital evidence is promising but falls short due to non-robust experimental design, failure to implement modern medical technology, and poor enlistment of community support. The home hospital module offers most of the same medical components that are standard of care in an acute care hospital. The typical staff (medical doctor \[MD\], registered nurse \[RN\], case manager), diagnostics (blood tests, vital signs, telemetry, x-ray, and ultrasound), intravenous therapy, and oxygen/nebulizer therapy will all be available for home hospital. Optional deployment of food services, home health aide, physical therapist, occupational therapist, and social worker will be tailored to patient need. Home hospital improves upon the components of a typical ward's standard of care in several ways: Point of care blood diagnostics (results at the bedside in \<5 minutes); Minimally invasive continuous vital signs, telemetry, activity tracking, and sleep tracking; Automated alerting of MDs by mobile phone for any worrisome vital sign patterns; On-demand 24/7 clinician video visits; 4 to 1 patient to attending MD ratio, compared to typical 16 to 1; Ambulatory/portable infusion pumps that can be worn on the hip; Optional access to a personal home health aide. Should a matter be emergent (that is, requiring in-person assistance in less than 20 minutes), then 9-1-1 will be called and the patient will be returned to the hospital immediately. In previous iterations of home hospital this happens in about 2% of patients. Clinical parameters measured will be at the discretion of the physician and nurse, who treat the participant following evidence-based practice guidelines, just as in the usual care setting. In addition, the investigators will be tracking a wide variety of measures of quality and safety, including some measures tailored to each primary diagnosis.

Primary Outcomes

Secondary Outcomes

• Direct margin, $(From date of admission to date of discharge, an expected average of 4 days)
• Direct margin, modeled with backfill(From date of admission to date of discharge, an expected average of 4 days)
• Total cost, 30-day post discharge(Day of admission to 30-days post-discharge)
• Length of stay, days(From date of admission to date of discharge, an expected average of 4 days)
• Imaging, #(From date of admission to date of discharge, an expected average of 4 days)
• Lab orders, #(From date of admission to date of discharge, an expected average of 4 days)
• All-cause readmission(s) after index, #(Day of discharge to 30 days later)
• All-cause readmission(s) after index, y/n(Day of discharge to 30 days later)
• Unplanned readmission(s) after index, #(Day of discharge to 30 days later)
• Unplanned readmission(s) after index, y/n(Day of discharge to 30 days later)
• Emergency Department observation stay(s) after index hospitalization, #(Day of discharge to 30 days later)
• Emergency Department observation stay(s) after index hospitalization, y/n(Day of discharge to 30 days later)
• Emergency Department visit(s) after index hospitalization, #(Day of discharge to 30 days later)
• Hours of sitting upright per day, #(From date of admission to date of discharge, an expected average of 4 days)
• Steps per day, #(From date of admission to date of discharge, an expected average of 4 days)
• EuroQol -5D-5L, composite score(At admission, at discharge (the day the patient leaves the hospital environment), and at 30 days after discharge)
• Short Form 1(30 days prior to admission (asked on day of admission), at admission, at discharge (the day the patient leaves the hospital environment), and at 30 days after discharge)
• Activities of daily living, score(30 days prior to admission (asked on day of admission), at admission, at discharge (the day the patient leaves the hospital environment), and at 30 days after discharge)
• Emergency Department visit(s) after index hospitalization, y/n(Day of discharge to 30 days later)
• Delirium, y/n(From date of admission to date of discharge, an expected average of 4 days)
• Transfer back to hospital, y/n(From date of admission to date of discharge, an expected average of 4 days)
• Hours of sleep per day, #(From date of admission to date of discharge, an expected average of 4 days)
• Hours of activity per day, #(From date of admission to date of discharge, an expected average of 4 days)
• Instrumental activities of daily living, score(30 days prior to admission (asked on day of admission), at admission, at discharge (the day the patient leaves the hospital environment), and at 30 days after discharge)
• 3-item Care Transition Measure, score(30 days after discharge)
• Picker Experience Questionnaire, score(30 days after discharge)
• Global satisfaction with care, score(30 days after discharge)
• Qualitative interview(30 days after discharge)