Management of ventilator-associated pneumonia in intensive care units: a mixed methods study assessing barriers and facilitators to guideline adherence

To identify barriers and facilitators of managing VAP in ICUs from the providers’ perspective, a survey was developed based on literature review and content analysis of data collected from focus groups. Two focus group discussions (FGDs) (one with physicians and one with nurses, respiratory therapists and pharmacists) were conducted at each of the two hospitals. The objective of the FGDs was to identify and discuss issues of VAP management using the SEIPS framework (Fig. 1). Content analysis of focus groups’ transcripts identified a total of 193 “unique” comments, which were further classified into 9 themes: (1) communication between providers, (2) difficulty in diagnosing VAP, (3) education related to VAP and VAP management, (4) guideline awareness and use, (5) management of the condition, (6) provider responsibilities, (7) technology and its use, (8) use of clinically indicated tests, and (9) variation in practice. Based on the comments and themes identified, a survey was developed and pilot tested on four participants, all of whom were involved in medical management of VAP. They included one physician, one nurse, one respiratory therapist and one pharmacist. Revisions to the content of the questions were made to the survey according to the feedback from pilot testing.

The final survey was adapted to the two hospitals. The version developed for hospital A consisted of five sections. Background had two questions asking the participants if they were aware of the IDSA/ATS guideline for VAP management and to what extent they were willing to follow the guideline. Methods asked participants to indicate to what extent they agree or disagree with 45 statements on VAP management, using a 5-point Likert scale (1 strongly agree, 2 agree, 3 neither agree nor disagree, 4 disagree and 5 strongly disagree). Results asked participants to indicate the frequency with which 17 situations occurred in VAP management. This used a 5-point Likert scale (1 rarely, 2 occasionally, 3 sometimes, 4 fairly often and 5 very often). Discussion had five yes/no questions. These addressed VAP management and diagnosis guidelines and assessed considerations regarding the mini-bronchoalveolar lavage (mini-BAL) procedure. Conclusion collected demographic information of participants including gender, age, job position, work shift, and average length working for the present employer in the current position.

The survey developed for hospital B similarly had five sections, but a few questions were different from those for hospital A. This was because, in addition to the IDSA/ATS guideline for management of VAP, hospital A also has its own VAP management guidelines that were largely similar to national guidelines but not identical.

The paper survey was distributed to all physicians, nurses, respiratory therapists and pharmacists who practiced in the medical-surgical ICUs at both hospitals. The study was approved by the University of Wisconsin-Madison and the Marshfield Clinic Institutional Review Boards (IRBs). We obtained informed consent from all participants involved in the study.

Survey data was entered (with verification for accuracy) in an SPSS database. Descriptive statistics were calculated to identify top-ranked barriers and facilitators of VAP management. For each question, the number and percentage of participants who chose each of the available response categories were calculated. If the wording of a question was positive (e.g., pharmacist participation on ICU rounds is beneficial), then participants who chose “agree” and “strongly agree” or “fairly often” and “very often,” (when the question was about frequency), considered it as a facilitator; and participants who chose “disagree” and “strongly disagree” or “occasionally” and “rarely” considered it as a barrier. In contrast, if a question was stated in a negative manner (e.g., having multiple groups of physicians manage patients in the ICU complicates VAP guideline use), then participants who chose “agree” and “strongly agree” or “fairly often” and “very often” considered it as a barrier; and participants who chose “disagree” and “strongly disagree” or “occasionally” and “rarely” considered it as a facilitator. Questions were ranked according to the percentage of respondents who considered them as barriers or as facilitators.

Statistical analysis was performed to compare perceptions of participants from different professional groups and to examine the impact of guideline awareness on the perceptions of participants. Three questions designed only for hospital A were excluded from the analysis because they were not applicable to hospital B and 7 questions were excluded from the analysis because they were designed only for physicians and were not applicable to the other healthcare professional groups. Analysis for the remaining questions proceeded as follows. We treated questions that used a 5-point Likert scale (1 rarely, 2 occasionally, 3 sometimes, 4 fairly often and 5 very often) or (1 strongly agree, 2 agree, 3 neither agree nor disagree, 4 disagree and 5 strongly disagree), depending on whether the question was about agreement or frequency, as continuous variables [28]. Because the dependent variables examining the differences between physicians, nurses, respiratory therapists and pharmacists were not normally distributed, we conducted one-way non-parametric ANOVA. When significant differences were identified between different professional groups, post-hoc analysis was performed using the Mann-Whitney test. This test was also used to examine differences between participants who were aware of and who were not aware of the IDSA/ATS guideline for VAP management. The Mann-Whitney test works by ranking all the values from low to high. Therefore, the hypothesis tested in these tests was that means of the ranks in the two comparison groups was zero.