Which Technology Interventions Reduce Emergency Department Visits - PowerPoint PPT Presentation

Which Technology Interventions Reduce Emergency Department Visits and Hospital Admissions From Long- Term Care Facilities? Findings From a Systematic Review Deniz Cetin-Sahin, MD, PhD(s ) Department of Family Medicine, McGill University 21 April 2018 Canadian Geriatrics Society Annual Scientific Meeting Montreal, Quebec

Disclosure of Financial Support This program has received financial support from: Donald Berman Maimonides Medical Research Foundation in the  form of a research fellowship. The Fonds de recherche du Québec – Santé (FRQ-S) in the form of a  doctoral training award. This program has received in-kind support from Donald Berman  Maimonides Medical Research Foundation in the form of logistics. Potential for conflict of interest: None 

Team Members Machelle Wilchesky, PhD  McGill University, Department of Family Medicine and Division of Geriatric Medicine (Primary supervisor)  Donald Berman Maimonides Geriatric Centre  McGill University, Centre for Clinical Epidemiology, Lady Davis Institute for Medical Research, Jewish General Hospital Ovidiu Lungu, PhD  Donald Berman Maimonides Geriatric Centre  Université de Montréal, Départment de Psychiatrie Matteo Peretti, MSc(c)  McGill University, Department of Family Medicine,  Donald Berman Maimonides Geriatric Centre Genevieve Gore, MLIS  McGill University, Schulich Library of Science and Engineering Philippe Voyer, RN, PhD  Faculté des Sciences Infirmières, Université de Laval Brian Gore, MD, CCFP, Dip Epid  Donald Berman Maimonides Geriatric Centre  University of Pittsburgh School of Medicine, Department of Biomedical Informatics and Division of Geriatric Medicine Steven Handler, MD, PhD, CMD  Clinical Informatics and Long-term Care Health Information Technology, UPMC Senior Communities

Outline  Background  Knowledge gap  Review questions  Methods  Results  Conclusions 4

Background  Long-term care facility (LTCF) residents are at high risk of being transferred to acute care (Grabowski et al, 2008 )  More than 1/3 of the residents visiting emergency departments (ED) are eventually admitted to a hospital (Ackerman et al, 1998)  About 2/3 of hospital admissions (HA) are avoidable (Ouslander et al, 2010)  Significant adverse outcomes associated with avoidable ED transfers and hospitalizations (Dwyer et al, 2014)

Interventions Aimed At Reducing Potentially Avoidable Acute Care Transfers  Wilchesky M, Cetin-Sahin D, Gore G, et al. PROSPERO 2016:CRD42016048128 http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD420160481 28 Complex because they address  multi-dimentional reasons for transfers Multi-component   Training, human resources, tools, technology

Definition of “Technology”  Information and communication technology used by healthcare organizations for management or delivery of healthcare  Adapted from Effective Practice and Organization of Care (EPOC). EPOC taxonomy; 2015.

Knowledge gap  Evidence exists regarding feasibility and stakeholder satisfaction (Edirippulige et al, 2013)  Lack of evidence for their effectiveness (Edirippulige et al, 2013)  Limited number of technologies studied  Reduction in acute care transfers has not been studied  Most studies are observational and qualitative (Edirippulige et al, 2013)

Review Questions 1. What types of technology interventions exist for LTCF stakeholders in order to reduce acute care transfers in the event of an acute or complex changes in resident health status? 2. What is the effectiveness of these interventions in reducing acute care transfers as compared to usual care?

METHODS

Design: Systematic mixed studies review (Souto et al, 2015) Main inclusion criteria: Technology-centered or aided programs, models Interventions of care, innovations, or tools Comparison Usual care Outcome measures ED visits or hospital admissions Facility-based long-term care Setting (Canadian Healthcare Association) Study methods Quantitative and mixed studies Language English or French

Three-Phase Search Strategy Database search from inception to July 2016 Ovid Textwords Embase • • Backward and AMED • MEDLINE • Global Health forward Grey • CINAHL • Health and • citation literature Social Work Abstracts • Psychosocial tracking search PsycINFO Instruments • techniques Joanna Briggs The Cochrane Library • • Institute EBP Database Ovid Healthstar • Web of Science •

Two Independent Reviewers Identification and Selection Process  Quality appraisal of selected studies:  Mixed Methods Appraisal Tool (MMAT) (Souto et al, 2015) • Scored from 0 to 4 • Data extraction:  Characteristics of studies • Descriptions of interventions • Evidence of effectiveness •

Knowledge synthesis  High heterogeneity  Most studies reported insufficient quantitative data for inclusion in a random-effects model meta-analysis

RESULTS

Identification and selection results 8,424 records identified through 3,078 additional records database searching 6,526 records after duplicates removed screened 6,382 records based on titles and abstracts excluded 144 full text articles were assessed for eligibility 29 additional records identified through 153 articles excluded other sources • Not primary studies (10) • Backward citation tracking (4) • Not technology interventions (77) • Forward citation search (22) • Not LTC setting (37) • Grey literature search (3) • No outcomes of interest (29) 16 studies included in the synthesis PRISMA-P 2015 statement (Moher et al, 2015)

Characteristics of the studies Year: Between 1998 and 2016 Country: USA (4), Australia (3), Canada (2),UK (2), Taiwan (2), China (2), New Zealand (2) Quality MMAT total score: Low scores (0-1) n=4  Other scores (2-4) n=12 

Clinical heterogeneity Design Intervention Randomized pre-post Mono vs multi-component   intervention study Various components other  Retrospective quasi-  than technology experimental study Different stakeholders  Feasibility pilot study  involved Cluster randomized  Usual care, population stepped-wedge trial under study: 2 group matched pre-post  prospective cohort study Not consistently defined  Retrospective pre-post study … 

Statistical heterogeneity ED visits Hospital Admissions • # of visits • Rate/1,000 resident days • # of annual visits • # of monthly hospital visit • # of return visits • Proportion of 30 day hospital readmissions • Proportion of 30 day return visits without hospital • # of avoidable admissions admission • # of annual admissions following ED visits • # of discharge from the ED without admission

Three types of technology 1. Web-based visual system for telemedicine (n=5)* 2. Non-visual tele-coaching (n=7) 3. Health information systems (n=6) * 2 studies also included more than 1 technology type

1. Web-based visual system for telemedicine Definition: Direct provision of a clinical service (diagnosis or management) Videoconferencing Telemedicine carts Exam cameras Digital otoscopes Electronic stethoscopes Dermatoscopes Ophthalmoscopes

1. Web-based visual system for telemedicine Effectiveness Author INTERVENTION N (setting) Hospital year ED visits admissions Grabowski 11 -- 4.4% Telemedicine for wound care 2014 (6-C; 5-I) Telemedicine for long-term 48 Hex chronic conditions and people 14% 5% (21-C; thought to be in the last 12 months 2015 27-I) of life Taiwan’s Telehealth Pilot Project: Hsu -- 25% a tele-consultation infrastructure to 3-I 2010 link the LTCF to tertiary hospitals Hui Telemedicine to provide geriatric 8.8% 10.6% 1-I 2001 services Stern Enhanced multidisciplinary 12 (exposed to 30% 20% teams via telemedicine 2014 both I and C) (advanced practice nurses)

2. Non-visual tele-coaching Definition: Clinical consultation or transfer approval process with experts from outside LTCF Telephone calls e-mails

2. Non-visual tele-coaching Effectiveness Author INTERVENTION N (setting) year Effectiveness Effectiveness Boyd Residential Aged Care Integration 54 -- 43% 2014 Program (gerontology nurse specialist) (25-C; 29-I) Codde An enhanced primary care service (ED- 15% -- 1-I 2010 based nurses) No significant Hullick The Aged Care Emergency service (ED- 12 ~35 % change ~ 2016 based nurses) (8-C; 4-I) No significant No significant Lee Care protocol (community nurse) change ~ 45 (assigned) change ~ 2002 A complex multidisciplinary No significant Sankaran -- change ~ intervention (Clinical Nurse Specialists 1-I 2010 and geriatrician) Street Residential Care In-Reach (specialist All LTCFs in a 11% 23.2% 2015 practice nurses) region Stern Enhanced multidisciplinary 12 (exposed to 2014 30% 20% teams via telemedicine both I and C)

3. Health information systems Definition: Electronic transfer of clinical information, documents, or secure messaging to either facilitate transfer of clinical data or to alert clinicians regarding resident health status changes