Long-term effect of computer-assisted decision support for antibiotic treatment in critically ill patients: a prospective ‘before/after’ cohort study

From Clinfowiki
Jump to: navigation, search

This is a review of Nachtigall, Tafelski, Deja, Halle, Grebe, Tamarkin and Spies 2014 article, Long-term effect of computer-assisted decision support for antibiotic treatment in critically ill patients: a prospective ‘before/after’ cohort study.[1]


Antibiotic resistance has become a major health concern all over the world. Much of the problem can be traced back to misuse and wrong administration of antibiotics.

Antibiotic resistance leads to;:[2]

  • Longer illnesses
  • More complicated illnesses
  • More doctor visits
  • The use of stronger and more expensive drugs
  • More deaths caused by bacterial infections

The authors in this article performed a prospective cohort study to determine whether a Computer-assisted Decision Support System CDSS can have positive adherence in an ICU surgical setting. The team hypothesized that implementation of the CDSS for antibiotic therapy should decrease antibiotic use over time and improve clinical outcome.


This prospective cohort study was conducted as a clinical prospective pre-intervention/post-intervention study over four evaluation time periods within 5 years totaling 12,965 patient days in the ICU. [1]

During the pre-intervention period, every ward physician was given paper-based guidelines for antibiotic therapy. In contrast, during the post-intervention period, the CDSS for antibiotic therapy was implemented as a tool on every hospital computer. For every infection included in the CDSS, there is one main five-step algorithm. [1]


A total of 1,395 patients were treated during the study period. After exclusion criterion was incorporated, 1,316 patients were included in present analysis. It is interesting to note that the study results include two significant endpoints.

In terms of the primary endpoint, adherence to guidelines increased from 61% prior to implementation to 92% in (Post 1). This percentage then decreased in (Post 2) to 76% and decreased once again in (Post 3) with 71%. The final percentages still remained significantly higher compared with baseline values.

When the second endpoint is examined, the authors found that antibiotic-free days (AFD) increased over time. They started at 30% in all ICU days during the pre-intervention and increased to 32% (Post 1), 46% (Post 2) and finally 42% (Post 3).


Guideline adherence was highest in the period directly after implementation of the CDSS and decreased steadily in the latter years. It is noteworthy to understand that guideline adherence is a process parameter and does not provide a sufficient measure of the quality of care on its own. The data presented in the article showed that the number of AFD increased over time. This suggests that an individual patient receives shorter courses of antibiotics as guideline adherence increases. [1]


After implementing a CDSS, adherence rate to guidelines for antibiotic therapy remained above 70% over a 5-year period. In addition, another benefit correlated with the guideline adherence was reduction in individual antibiotic therapy. Finally, ICU mortality also decreased proving that CDSS can improve the quality of care in several aspects.


CDSS implementation appears to be the future direction of healthcare. Many hospitals and clinics have already begun the transition and the outcomes can be perceived as beneficial in the long run. The results obtained in this article showed that CDSS in fact could improve the quality of care. The only draw back from the article was that they didn’t maintain a long-term follow up, meaning that it is unknown if the effects stayed in place after the study was finished.

Related article review:


  1. 1.0 1.1 1.2 1.3 Nachtigall, I., Tafelski, S., Deja, M., Halle, E., Grebe, M. C., Tamarkin, A., … Spies, C. (2014). Long-term effect of computer-assisted decision support for antibiotic treatment in critically ill patients: a prospective “before/after” cohort study. BMJ Open, 4(12), e005370. doi:10.1136/bmjopen-2014-005370
  2. http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm092810.htm