544. Observer Effect on Hand Hygiene (HH) Compliance Reporting
Session: Poster Session: Hospital-acquired and Transplant Infections
Friday, October 30, 2009: 12:00 AM
Room: Poster Hall A
Introduction: Although HH is critically important in limiting spread of hospital infections, compliance among healthcare workers is poor. One factor complicating the interpretation of HH rates are non-standardized reporting methods. For instance, centers may use observers who work on the same unit (unit based), differing units, or utilize “third party” individuals such as infection control professionals (non-unit based). We hypothesized that observed compliance would be higher when performed by unit based observers as compared to third party individuals.
Methods: HH data from 2 hospitals in the DMC was reviewed for the year of 2008. 19 units including acute care, intensive care, and ER were included. Observations were categorized as unit based, if observers worked on the same unit where observations were occurring; and non-unit based if observers did not work on the same unit. HH observations were conducted using standard definitions by trained observers. HH compliance rates between unit and non-unit based observers were compared.
Results: A total of 6984 observations from the 19 units were reviewed. Of these, 3325 were unit and 3659 were non-unit based. HH compliance was 79% and 58.6% respectively (p<0.001, OR 2.7, CI 2.43-3.01). A sub-analysis was performed including only units that utilized both unit and non-unit based observers and results were similar (Table).
Table:
Unit ObserversNon-Unit ObserversDifference in Compliance (%)p valueRelative Risk*CI (95%)
unit# CompliantTotal Observations% compliant# CompliantTotal Observations% Compliant
110712287.712521957.130.6<0.0011.541.35-1.75
21361608520230566190.0251.131.03-1.24
3669172.522440655.217.30.0041.311.13-1.53
49513669.94317524.645.3<0.0012.842.14-3.77
59212971.318829663.57.80.151.120.98-1.29
630138478.43201563.1<0.0015.231.84-14.85
Total797102278785142155.2

* Likelihood unit based observers will judge observation as compliant.
Conclusion: Compared to non-unit based observers, unit-based observers rated HH compliance to be significantly higher from units using a single observer type (unit or non unit based) and from units that used both (20.4% and 22.6% respectably). The unit-affiliation of the observer should be incorporated into the analysis, benchmarking, and interpretation of HH compliance rates.
Connie Bohlinger, MS, M(ASCP)1, Thomas Chevalier, RN, BSN, CIC1, Teena Chopra, MD1, Sorabh Dhar, MD, Beth Dziekan, RN, MS, CIC1, Michelle Fitch, BS1, Elaine Flanagan, CIC3, Keith Kaye, MD, MPH, FIDSA, FSHEA4, Beth Toftey, MT(ASCP), CIC, MPH5 and  S. Dhar, None..
B. Toftey, None..
B. Dziekan, None..
T. Chevalier, None..
C. Bohlinger, None..
M. Fitch, None..
E. Flanagan, None..
T. Chopra, None..
K. Kaye, None., (1)Detroit Medical Center, Detroit, MI, (2)Detroit Medical Center/Wayne State University, Detroit, MI, (3)Detroit Medical Center, Wayne State University, Detroit, MI, (4)Detroit Medical Center, Commerce, MI