# A Protocol-Driven, Digital Conversational Agent at the Hospital Bedside to Support Nurse Teams and to Mitigate Delirium and Falls Risk

> **NIH NIH R44** · CARE.COACH CORPORATION · 2020 · $555,611

## Abstract

Abstract
About 700,000 to 1 million falls per year occur during hospital stays in the US, costing hospitals $3 billion to $7
billion each year to treat. These costs are not reimbursed by the Centers of Medicare and Medicaid, resulting
in significant financial losses for hospitals. Delirium, a major cause of falls, is highly prevalent (29%-64%)
among hospitalized elders. Even without causing a fall, the average case of delirium increases the length of
stay by 7.78 days, disrupting throughput and significantly reducing net hospital revenues. Including the
common sequelae of post-discharge functional decline, delirium costs the national healthcare system over
$130 billion per year. Due to the high cost and significant psychosocial needs of certain hospitalized patients, a
widely-used intervention to mitigate risk is to assign hospital staff & nurses to serve as “patient sitters” at the
bedside, at a cost of $1 million/year for a typical hospital. However, despite the workforce burden and expense,
patient sitters often do not reliably execute risk-mitigating protocols, and the literature does not support their
efficacy in preventing adverse events. In this SBIR Fast-Track proposal, we seek to develop an advanced,
human-in-the-loop artificial intelligence (AI) avatar system to enhance the wellbeing of hospitalized patients,
avoid adverse events including delirium and falls, and improve workforce efficiency by supporting nursing staff
to work at the top of their license, potentially generating savings of $2,000,000 each year for a typical 300-bed
hospital. This proposal aligns with NINR’s cross-cutting focus areas of Promoting Innovation and 21st Century
Nurse Scientists, while applying the principles of patient self-management and wellness to the acute care
environment, where the outcomes driven by our patient engagement and support platform will have an
outsized, immediate cost benefit to enable rapid scaling and dissemination.

## Key facts

- **NIH application ID:** 9829116
- **Project number:** 5R44NR017842-03
- **Recipient organization:** CARE.COACH CORPORATION
- **Principal Investigator:** Victor Wang
- **Activity code:** R44 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $555,611
- **Award type:** 5
- **Project period:** 2018-05-01 → 2022-11-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9829116

## Citation

> US National Institutes of Health, RePORTER application 9829116, A Protocol-Driven, Digital Conversational Agent at the Hospital Bedside to Support Nurse Teams and to Mitigate Delirium and Falls Risk (5R44NR017842-03). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/9829116. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*