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

NRI 05-146 – HSR Study

NRI 05-146
Improving Quality: Registered Nurse Work Activities in VA Sponsored Nursing Homes
Mary Ellen Dellefield, PhD BSN MS
VA San Diego Healthcare System, San Diego, CA
San Diego, CA
Funding Period: January 2006 - December 2007
Improving the quality of VA-sponsored nursing homes (NH) is important to veterans and other stakeholders involved in VA-sponsored NH care, including VA operated Community Living Centers (CLC), state veteran homes, and contract community nursing homes. Although NHs tend to employ few registered nurses (RNs), research finding indicate that RNs significantly influence their quality. RNs perform essential and specialized care processes that may be categorized as direct and indirect care. The effective implementation of clinical guidelines for pressure ulcer prevention (PUP) in NHs involves RN direct and indirect care. Major clinical risk factors associated with direct and indirect care include include incontinence, mobility, nutrition, pain, and actual PU treatment. Indirect care activities include supervision, management, and documentation. PU prevention is believed to be sensitive to NH quality and is widely used as both a nurse-sensitive and an organization-level quality indicator. Better description of how RN care processes are associated with PUP in NHs is important. It will contribute to our understanding of more effective and efficient RN deployment in VA-sponsored NH settings to advance quality and prevent pressure ulcers.

Project objectives included:

1. Development and testing of the feasibility of using a personal digital assistant (PDA)-based system to data collect observed RN direct and indirect care processes associated with pressure ulcer (PU) prevention; and

2. Measurment of the frequency of performance of RN care processes related to clinical documentation, supervision, and management.

This descriptive study was conducted at 2 VA-sponsored NHs in San Diego County, including a 40-bed short stay CLC and a 174-bed contract community nursing home (CNH). A convenience sample of 7 RNs from each site participated. Resident census, acuity levels, and nurse staffing levels were collected at both sites. Average RN wages were calculated based on a California NH database. The indistrial engineerin technique of work smpling was used to both identify the size of the sample of RN care processes to be collected at to collect data. The percentages of 45% direct and 40% indirect care (CLC) and 30% direct and 60% indirect (CNH) were used to determine the final samples. Care processes observed but not categorizable into one of the 5 established clinical categories based on direct observation alone were categorized as "other." 258 RN care processes were defined, based on a literature and expert panel review. These were loaded onto PDAs using the work sampling software - WorkStudy 3.0. Four data collectors were trained in the data collection technique of observing for a 30-second period, and recording the observations, every 5 minutes. Reliability testing of 10% of total observations resulted in reliability scores of at least 95% reliability of the occurrence and nonoccurrence of observed behaviors. This measure is used to test for reliability in behavioral observation studies. Data were anlyzed using descriptive statistics, including percentages, counts, and means. The WorkStudy 3.0 software provided summary counts and percentages using standard data display templates. Data were cleaned for entry errors prior to data analysis.

The RN subjects, including 7 at each site (CLC & CNH), were similar demographically. Subjects were mostly female (86% vs 100%), included half of the total RN staff at the sites (51% vs 50%), younger at the CLC (49 years vs 60 years old), and tenure (8 vs 7 years) at the sites. They were ethnically diverse, including African Americans (1 vs 0), caucaisan (1 vs 3), and Filipino 5 vs 4). The average hourly wage for RNs differed in expected ways, with the governmental hourly wage of $44 per hour and the CNH hourly wage of $32 per hour. The average daily census (29 vs 162) and acuity levels (32 Case Mix Index vs 28)differed. Both sites had total nurse staffing levels above minimum state and federal requirements (6.61 vs 4.2 hours per resident day [HPRD]). Average total RN staffing was 2.32 HPRD vs .30 HPRD. Care processes were observed at a confidence level of 95%, included 2,855 (CLC) and 4,476). The unit of analysis was each discrete care process observed from 258 possible processes. The percentages of direct (47% vs 31%) and indirect care processes (41% vs 59%) were statistically significant at (p<.000). Clinical processes observed included incontinence (4% vs 6%), mobility (7% vs 2%), nutrition (6% vs 1%), pain (2% vs 1%), PU treatment (4% vs 1%), and other (76% vs 94%). The statistical signficance of percentages of clinically specific care processes observed and unproductive time could not be calculated because the study sample size was calculated based on direct and indirect time rather than percentages of time for each clinical category. Most observed care processes could not be ascribed to a clinical domain; most were categorized as "other." Direct care processes so categorized included doing care, getting equipment, talking with residents and family, and giving medications. Most indirect care processes included documentation (34% vs 48%), managing (7% vs 17%), and supervising (6% vs 4%). Few management and supervisory care processes occurred between RNs and certified nursing assistants.

The percentages of RN care processes performed involving direct and indirect care are significantly different in the 2 sites. The CLC used only RN and LVN staff; the CNH used the traditional nursing skill mix of RNs, LVNs, and CNAs. In other studies more RN direct care has been associated with fewer negative clinical outcomes, including incidence of PUs. Direct observation needs to be linked with RN self-report and medical record audits to clinically categorize care processes. Most care processes observed were generic. Few RN clinically targeted care processes for PUP were observed. Much time was spent in clinical documentation. Minimal supervision occured at either site, with scant supervision of CNAs by RNs at the CNH site. Study findings suggest that how RNs spend their time under different nursing skill mix conditions warrant further study. Linking this with the incidence and prevalence of PUs will be important. RNs need to educated about how to increase their performance of clinical supervision. Strategies to reduce the percentage of RN activity associated with clinical documentation are needed.

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. Dellefield ME, Harrington C, Kelly A. Observing how RNs use clinical time in a nursing home: a pilot study. Geriatric nursing (New York, N.Y.). 2012 Jul 1; 33(4):256-63. [view]
  2. Dellefield ME, Kelly A, Schnelle JF. Quality assurance and performance improvement in nursing homes: using evidence-based protocols to observe nursing care processes in real time. Journal of nursing care quality. 2013 Jan 1; 28(1):43-51. [view]

DRA: Health Systems
DRE: none
Keywords: Research measure, Research method
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.