PURPOSE: DVT prophylaxis is a widely adopted quality indicator judging the performance of healthcare delivery. The goal of this study is to evaluate DVT compliance during week days, weeknights and weekends.
METHODS: This is a retrospective mono-centric study performed at a closed tertiary center ICU. Inclusion and exclusion criteria were determined in priori. Pharmacologic and mechanical prophylaxes were recorded. Patients were categorized into groups (weekdays, weeknights and weekends) according to time of admission. Unweighted risk factor score (RFS) was calculated from 20 different risk factors. Compliance was defined as any type of prophylaxis for RFS below 3 or both types of prophylaxis for RFS above 3. Noncompliance was defined as no prophylaxis or single type prophylaxis for RFS above 3.
RESULTS: Of 248 patients reviewed, 143 were excluded and 105 were included. 80(76.19%) were compliant and 25 (23.81%) were non compliant, of which 11 (10.48%) were on no prophylaxis.There were no differences among the groups in baseline characteristics; however the weekday group had fewer female patients (38% vs. 63% and 67% p=0.029). DVT compliance was not different across the three groups. Univariate analysis using chi-square or ANOVA showed the noncompliant group had a higher RFS (3.5±1.9 vs. 2 ±1.2; p<0.0001), fewer female patients, and the admitting residents had less postgraduate training, as signified by their lower postgraduate year (PGY) level.Multivariate analysis using logistic regression revealed that only RFS and PGY level were significant (p<0.0001 and 0.004 respectively). The time of admission was not a significant factor.
CONCLUSION: We found that time of admission did not impact DVT prophylaxis compliance. Compliance improved with higher PGY level and lower RFS. Higher level of knowledge probably explains this association with PGY level; however we cannot explain the inverse relationship between RFS and compliance.
CLINICAL IMPLICATIONS: Variation in compliance appears to be the product of knowledge of the admitting resident rather than the time of admission. Applying standardized algorithms may be one solution to decrease variation and improve DVT compliance.
DISCLOSURE: Hani Jaouni, No Financial Disclosure Information; No Product/Research Disclosure Information