Veterans receive VA-funded nursing home (NH) care in VA-operated Community Living Centers (CLC), State Veterans Homes, or community nursing homes (CNH). Most direct care is provided by para-professionals or nursing assistants (Nas). Successful implementation of clinical practice guidelines in these settings requires NA participation. The HSR&D study, PDA-Based Measurement of Guideline Implementation by Nursing Assistants (SHP 08-200), demonstrated the feasibility of using a direct observation strategy to measure the frequency of NA performance of nursing care processes associated with clinical guidelines for pressure ulcer prevention (PUP).
The recent "culture change" movement in VA-funded NHs, with its emphasis on interpersonal relationships and communication, motivated a focus on and analysis of the frequency of NA interpersonal interactions with residents, peers, family, non-clinical and interdisciplinary staff, and licensed nurses.
To: (a) test a PDA-based direct observation strategy to measure the frequency of NA performance of nursing care processes associated with pressure ulcer (PU) clinical risk factors; (b) determine percentages of observed NA care processes associated with PUP; (3) determine percentages of direct, indirect, and unproductive care observed; and (4) frequency of NA interpersonal interactions with residents, peers, family, non-clinical and interdisciplinary staff, and licensed nurses.
To achieve these objectives, we conducted a descriptive observational study using industrial engineering work-sampling. Two CNHs providing long-term care and short-term rehabilitation were used, including a 174-bed not-for-profit Medicare/Medicaid and a 158-bed freestanding Medicare/Medicaid certified facility. A convenience sample of full-time or part-time Nas was used. Ninety-three discrete care processes (the unit of analysis) related to PUP were defined. Every 3 minutes, data collectors observed Nas for 30 seconds, recording all care process initiated during the interval. Care processes were observed intermittently throughout shifts and 7-day work week over 5 months. Reliability testing of 10% of total observations by 2 data collectors resulted in reliability scores of at least 95% reliability of the occurrence and non occurrence of observed care processes. Counts and percentages were used for data analysis, calculated using WorkStudy 3.0 PDA software and then verified manually.
Final samples included a total of 1,360 care processes, including 783 (site 1) and 577 (site 2) at a 95% confidence level (p<.05). Facility characteristics at both sites were similar. Staffing levels met California mandated NH staffing requirements. Day shift Nas (4 vs. 8) and evening shift (0 vs. 1) volunteered as subjects. Categories of care included direct (88% vs. 93%); indirect (5% vs. 3.5%); unproductive (7% vs. 3.5%). The percentages of clinical domains observed at both sites were similar. They included general work (65% vs. 66%); incontinence (1.5% vs. 5%); nutrition (22% vs. 16%); pain (.5% vs. 0%); and PU (pressure ulcer)/mobility (11 vs. 13%). The category of general work included generic processes such as getting equipment and removing trash.
Nas talked with residents most frequently (21% at both sites), followed by peers, licensed vocational nurses, family members, non-clinical and interdisciplinary staff, and RNs in descending order of frequency. Interpersonal interactions of Nas with RNs were rarely observed.
By demonstrating the feasibility of using a PDA-based direct observation strategy to measure the frequency of NA performance of care processes, we will be able to include this measurement strategy in the design of a supervisory intervention study that is currently under development.
External Links for this Project
None at this time.