Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website

EDU 08-417 – HSR Study

EDU 08-417
CONNECT for better falls prevention in VA Community Living Centers
Cathleen S Colon-Emeric, MD
Durham VA Medical Center, Durham, NC
Durham, NC
Funding Period: August 2009 - July 2012
The current standard to improve resident outcomes in nursing homes combines staff educational programs and quality improvement processes. These educational programs typically focus on individual behaviors and content mastery. However, social constructivism theory and complexity science suggests that learning is a social process, occurring within the context of the relationships and interactions of the individual in their environment. Thus, traditional educational programs will not result in optimal changes unless a context is present which allows social learning to occur.

We proposed that an intervention which lays the groundwork for social constructivist learning would improve the effectiveness of standard educational approaches. We randomized 4 VA Community Living Centers (CLCs) to receive either a traditional falls education program alone (FALLS), or the novel CONNECT program followed by FALLS. CONNECT is a 12 week program that trains staff to form better connections and use effective communication strategies with a diverse network of co-workers, so that problem solving about resident problems is enhanced. We tested the impact of CONNECT on staff measures of communication, fall QI measures, and resident fall rates. We conducted focus groups to explore the impact of CONNECT on social constructivist learning.

This study has direct relevance to the VA, which relies heavily on traditional educational and QI programs to improve care for its Veterans. Few studies of educational programs have described their impact on care processes or Veteran outcomes. CONNECT has the potential to improve care for other complex syndromes requiring coordination of care by diverse staff.

1. Describe the change in fall quality indicators completed within 1 month of resident falls, and determine whether it differs between facilities receiving CONNECT+FALLS or FALLS alone.
2. Describe the change in staff interaction measures and determine whether it differs between intervention groups.
3. Describe the change in facility fall-rates, and determine whether it differs between intervention groups.
4. Describe staff's perceptions of organizational learning and determine whether it differs between intervention groups.

This was a randomized, controlled, single blind study. Four VA CLCs in VISN-6 participated; all direct-care staff were eligible for participation. Overall, 271 of 480 eligible staff members participated in at least 1 educational component (56%).

CONNECT was delivered over 3 months, and included 2 interactive learning sessions, individual mentoring, and relationship mapping focused on helping staff build problem-solving networks. FALLS was delivered over 3 months and included web-based modules, weekly teleconferences, audit and feedback, and academic detailing.

Outcome measures: Facility-level fall rates were calculated for the 6 months preceding and following the intervention period. Falls were identified from incident logs, the MDS, and chart review. Bed days of care were obtained from the VISN.

Staff surveys included measures of safety culture, participation in decision making, communication openness, accuracy, and timeliness. Overall, 235 staff participated, with 104 (44%) completing all 3 waves.

Twelve falls QI measures were abstracted from individual fallers' medical records by trained study personnel blinded to intervention status for 340 unique fallers (100% of eligible Veterans). Inter-rater reliability was >90% on a random 5% sample of charts.

Eight focus groups including 46 staff members from all sites were completed to assess staff perceptions of social constructivist learning in their CLC.

Analysis used negative binomial regression models to test for significant changes in fall rates and fall quality indicator counts within sites over time, and test whether the difference in rate of change by treatment group was significant. Results were adjusted for resident fall risk factors and time at risk. SAS PROC MIXED was used to test for staff measure differences at 3 and 6 month follow-up controlling for baseline scores, with individual and site-level clustering taken into account in significance testing.

Intervention facility fall rates declined by 14% from 2.30 to 2.02 falls/bed/year, while control facilities did not change (2.93 to 2.90), effect of treatment on rate of change 0.60 (0.06-6.05, p=0.22). This trend was not explained by changes in either fall QIs or staff measures of communication. Fall QIs improved significantly in both groups (21% intervention group, p<0.0001; 29% control group, p<0.0001), with no significant treatment effect. Staff measures of communication showed no overall treatment effects; there was a significant increase in communication timeliness in the treatment group (p=0.03), and significant increases in participation in decision-making and safety culture in the sample as a whole (p<0.01). There was evidence of a ceiling effect in all staff measures.

Analysis of focus group transcripts showed that both groups believed the FALLS content was familiar, reminded them about fall risk factors, and helped them engage in teamwork. However, the CONNECT groups additionally noted that: 1) relationship mapping sessions helped them identify and correct communication weaknesses; 2) deeper, more complex connections among staff helped them more effectively manage falls; 3) improved understanding about the roles of various disciplines led to a wider fall prevention network; and 4) improved cooperation resulted in greater willingness to stick to interventions such as toileting programs. Participants mentioned that the intervention would be helpful for learning about other topics, such as behavior problems.

This project demonstrated that CONNECT is feasible, acceptable and perceived by VA staff to improve social constructivist learning. The observed decline in Veteran fall rates, while not statistically significant with the small VA facility sample size, is clinically meaningful and would be highly statistically significant (p<0.001) if confirmed in the ongoing, larger NIH-funded study with n=16 community nursing homes.

Additional practice and research implications of the study include: 1) fall QIs as measured by chart abstraction do not relate to changes in fall rates, and have limited clinical or research utility; 2) staff communication measures previously validated in community samples demonstrate a ceiling effect among VA staff and are therefore insensitive to changes in work environment.

External Links for this Project

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


Journal Articles

  1. Colón-Emeric CS, Lyles KW, Su G, Pieper CF, Magaziner JS, Adachi JD, Bucci-Rechtweg CM, Haentjens P, Boonen S, HORIZON Recurrent Fracture Trial. Clinical risk factors for recurrent fracture after hip fracture: a prospective study. Calcified Tissue International. 2011 May 1; 88(5):425-31. [view]
  2. Anderson RA, Corazzini K, Porter K, Daily K, McDaniel RR, Colón-Emeric C. CONNECT for quality: protocol of a cluster randomized controlled trial to improve fall prevention in nursing homes. Implementation science : IS. 2012 Feb 29; 7:11. [view]
  3. Toles M, Colón-Emeric C, Naylor MD, Asafu-Adjei J, Hanson LC. Connect-Home: Transitional Care of Skilled Nursing Facility Patients and their Caregivers. Journal of the American Geriatrics Society. 2017 Oct 1; 65(10):2322-2328. [view]
Conference Presentations

  1. Colon-Emeric CS, McConnell E, Pinheiro S, Simpson K, Hancock K, Lipscomb J, Beales J, Anderson RA. CONNECT for better fall prevention in nursing homes: results from a randomized controlled pilot study. Paper presented at: American Geriatrics Society Annual Meeting; 2013 May 3; Grapevine, TX. [view]
  2. Anderson RA, Corazzini K, McConnell ES, Thornlow DK, Pinheiro S, Colon-Emeric CS. CONNECT for Better Falls Prevention in Nursing Homes: An Intervention. Paper presented at: Gerontological Society of America Annual Scientific Meeting; 2010 Aug 5; Washington, DC. [view]
  3. Colon-Emeric CS, Pinheiro S, Simpson KM, Porter K, Corazzini K, Anderson RA. Improving Uptake of a Falls Educational Program by Focusing on Staff Interactions. Poster session presented at: American Geriatrics Society Annual Meeting; 2012 May 2; Grape Vine, TX. [view]
  4. Simpson K, Porter K, McConnell ES, Colon-Emeric CS, Daily KA, Stalzer A, Anderson RA. Qualitative Evaluation of a Staff Quality Improvement Program to Reduce Falls in Nursing Homes. Poster session presented at: Gerontological Society of America Annual Scientific Meeting; 2012 Nov 14; San Diego, CA. [view]

DRA: Aging, Older Veterans' Health and Care, Health Systems
DRE: Prevention, Technology Development and Assessment
Keywords: Education Research, Long-term care, Nursing
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.