Academic and Community ICUs Participating in a Critical Care Randomized Trial: A Comparison of Patient Characteristics and Trial Metrics

Abstract

OBJECTIVES:Clinical research in Canada is conducted primarily in "academic" hospitals, whereas most clinical care is provided in "community" hospitals. The objective of this nested observational study was to compare patient characteristics, outcomes, process-of-care variables, and trial metrics for patients enrolled in a large randomized controlled trial who were admitted to academic and community hospitals in Canada. DESIGN:We conducted a preplanned observational study nested within the Probiotics: Prevention of Severe Pneumonia and Endotracheal Colonization Trial (PROSPECT, a randomized controlled trial comparing probiotics to placebo in mechanically ventilated patients) Research Program. SETTING: ICUs.PATIENTS: Mechanically ventilated patients. MEASUREMENTS:We compared patient characteristics, interventions, outcomes, and trial metrics between patients enrolled in PROSPECT from academic and community hospitals. MAIN RESULTS:Participating centers included 34 (82.9%) academic and seven (17.1%) community hospitals, which enrolled 2,203 (86.2%) and 352 (13.8%) patients, respectively. Compared with academic hospitals, patients enrolled in community hospitals were older (mean [sd] 62.7 yr [14.9 yr] vs 59.5 yr [16.4 yr]; p = 0.044), had longer ICU stays (median [interquartile range {IQR}], 13 d [8-23 d] vs 11 d [7-8 d]; p = 0.012) and higher mortality (percentage, [95% CI] in the ICU,.4%]vs 20.5% [18.9-11.3%]; p = 0.002) and hospital (40.6% [35.6-45.8%] vs 26.1% [24.3-27.9%]; p < 0.001). Trial metrics, including informed consent rate (85.9% vs 76.3%; p = 0.149), mean (sd) monthly enrolment rate (2.1 [1.4] vs 1.1 [0.7]; p = 0.119), and protocol adherence (90.6% vs 91.6%; p = 0.207), were similar between community and academic ICUs. CONCLUSIONS:Community hospitals can conduct high-quality research, with similar trial metrics to academic hospitals. Patient characteristics differed between community and academic hospitals, highlighting the need for broader engagement of community hospitals in clinical research to ensure generalizability of study results.

If health organisations and staff engage in research, does healthcare improve? Strengthening the evidence base through systematic reviews

“…Partly, this is because the scope of the papers included has increased. There are many more studies where all the clinicians or organisations compared are engaged in research but to varying extents and/or with different levels of resources, for example within a trial [28, 50,59,73,93,103,107], or within a network [23,33,49,51,66,68,69,85,95]. The inclusion of papers about differences within trials, and the emergence of the importance of the "dose effect", have implications for both a) how the issue of research engagement is analysed, and b) how far efforts to enhance research engagement should be concentrated or spread widely across a system.…”

“…In the with the assumption that by participating as equals in the research process, community physicians would be more likely to accept and implement the results in their practices with non-protocol patients" (p.336) [37]. Similar moves to encourage wider participation in clinical trials have recently been made in Canada in the nested study described above [103]. And a recent analysis showing higher levels of research activity within the English healthcare system are associated with lower mortality, notes that although the NIHR CRN was established to promote research participation across England, there is still some way to go to ensure greater geographical equity [69].…”

“…Seaburg et al pre-planned observational study nested within a clinical trial in order to test how well traditionally nonresearch active community hospitals could participate in an RCT alongside the traditional RCT sites in academic hospitals. But while that aspect of the study did show that, in terms of adherence to trial metrics, the community hospitals could successfully participate in studies, outcomes for patients in the trial were signi cantly better in the traditional research hospitals, although the full reasons for this will need further exploration [103].…”

Barriers to participation in biosampling-based translational research: A cross-sectional survey of Canadian critical care researchers

“…Institutions can explore ways to increase enrollment success such as video-based or e-Consent strategies [3]. In Canada, the majority of ICU care is provided by community hospitals, which have limited participation in clinical trials or other research [6,7]. Given that community sites disproportionately serve racialized and socio-economically deprived Canadians, their involvement in research is critical to improve research participant diversity in Canadian critical care trials [6].…”

“…In Canada, the majority of ICU care is provided by community hospitals, which have limited participation in clinical trials or other research [6,7]. Given that community sites disproportionately serve racialized and socio-economically deprived Canadians, their involvement in research is critical to improve research participant diversity in Canadian critical care trials [6]. A recent Canadian study demonstrated that community ICUs can conduct research with comparable recruitment rates, consent rates, and protocol adherence when compared to academic ICUs [6].…”