Advertisement

Understanding how deferred consent affects patient characteristics and outcomes: an exploratory analysis of a clinical trial of prone positioning for COVID-19

      What is new?

        Key findings

      • Differences in characteristics were found between patients in the deferred consent and nondeferred consent groups.
      • Primarily it was found that the patients in the deferred consent group were more likely to be enrolled on weekends and were more likely to be older.

        What this adds to what was known

      • This finding shows that deferred consent, the process of enrolling patients in clinical trials prior to obtaining consent, may allow for the enrollment of a greater number of individuals.

        What is the implication and what should change now?

      • These findings may have implications for obtaining consent when conducting clinical trials in the future.

      1. Introduction

      Deferred consent, the process of enrolling patients in a clinical trial before consent is obtained, is often employed in studies of minimal risk [
      • Grimes D.A.
      • Hubacher D.
      • Nanda K.
      • Schulz K.F.
      • Moher D.
      • Altman D.G.
      The Good Clinical Practice guideline: a bronze standard for clinical research.
      ,
      • Harron K.
      • Woolfall K.
      • Dwan K.
      • Gamble C.
      • Mok Q.
      • Ramnarayan P.
      • et al.
      Deferred consent for randomized controlled trials in emergency care settings.
      ,
      • Topolovec-Vranic J.
      • Santos M.
      • Baker A.J.
      • Smith O.M.
      • Burns K.E.A.
      Deferred consent in a minimal-risk study involving critically ill subarachnoid hemorrhage patients.
      ,
      • Honarmand K.
      • Belley-Cote E.P.
      • Ulic D.
      • Khalifa A.
      • Gibson A.
      • McClure G.
      • et al.
      The deferred consent model in a prospective observational study evaluating myocardial injury in the intensive care unit.
      ], or when patients are unable to consent at the time of study enrollment [
      The ARISE Investigators and the ANZICS Clinical Trials Group∗
      Goal-directed resuscitation for patients with early septic shock.
      ,
      • Young P.
      • Saxena M.
      • Bellomo R.
      • Freebairn R.
      • Hammond N.
      • van Haren F.
      • et al.
      Acetaminophen for fever in critically ill patients with suspected infection.
      ,
      • Myburgh J.A.
      • Finfer S.
      • Bellomo R.
      • Billot L.
      • Cass A.
      • Gattas D.
      • et al.
      Hydroxyethyl starch or saline for fluid resuscitation in intensive care.
      ,
      The NICE-SUGAR Study Investigators
      Intensive versus conventional glucose control in critically ill patients.
      ]. Deferred consent has several possible benefits including decreasing the time to study enrollment, allowing for enrollment when study personnel are unavailable [
      • Shamy M.C.F.
      • Dewar B.
      • Chevrier S.
      • Wang C.Q.
      • Page S.
      • Goyal M.
      • et al.
      Deferral of consent in acute stroke trials.
      ,
      • Menon K.
      • O’Hearn K.
      • McNally J.D.
      • Acharya A.
      • Wong H.R.
      • Lawson M.
      • et al.
      Comparison of consent models in a randomized trial of corticosteroids in pediatric septic shock.
      ], and allowing the inclusion of populations less represented in clinical trials [
      • Honarmand K.
      • Belley-Cote E.P.
      • Ulic D.
      • Khalifa A.
      • Gibson A.
      • McClure G.
      • et al.
      The deferred consent model in a prospective observational study evaluating myocardial injury in the intensive care unit.
      ,
      • Burns K.E.A.
      • Zubrinich C.
      • Tan W.
      • Raptis S.
      • Xiong W.
      • Smith O.
      • et al.
      Research recruitment practices and critically ill patients: a multicenter, cross-sectional study (the consent study).
      ,
      • Byrne M.M.
      • Tannenbaum S.L.
      • Glück S.
      • Hurley J.
      • Antoni M.
      Participation in cancer clinical trials: why are patients not participating?.
      ]. Multiple studies have identified that deferred consent is considered an acceptable substitute for pre-enrollment consent by both participants and clinicians [
      • Manda-Taylor L.
      • Bickton F.M.
      • Gooding K.
      • Rylance J.
      A formative qualitative study on the acceptability of deferred consent in adult emergency care research in Malawi.
      ,
      • Den Boer M.C.
      • Houtlosser M.
      • Foglia E.E.
      • Lopriore E.
      • de Vries M.C.
      • Engberts D.P.
      • et al.
      Deferred consent for delivery room studies: the providers’ perspective.
      ]. Our objective was to assess patient-level characteristics, adherence, and rate of withdrawal among participants enrolled with consent obtained before (nondeferred consent) vs. after (deferred consent) randomization in the COVID-PRONE randomized trial (NCT04383613).

      2. Methods

      We conducted a secondary analysis of the COVID-prone international, pragmatic randomized clinical trial, which assessed prone positioning in patients with COVID-19 [
      • Fralick M.
      • Colacci M.
      • Munshi L.
      • Venus K.
      • Fidler L.
      • Hussein H.
      • et al.
      Prone positioning of patients with moderate hypoxaemia due to covid-19: multicentre pragmatic randomised trial (COVID-PRONE).
      ].
      Deferred consent was allowed as the benefit of prone positioning was thought to be greatest when immediately implemented, because patients may be in respiratory distress at the time of enrollment, and because we expected potential harms from prone positioning to be minimal. Deferred consent was allowed at all except two hospitals. The decision on timing of consent was made by the site lead at the time of recruitment.
      The primary outcome of this analysis was the difference in patient characteristics, time spent prone, and rate of withdrawal among participants who were enrolled with consent obtained before vs. after randomization. We did not compare the primary outcome between the two groups because our study was stopped for futility and our primary outcome was similar between patients randomized to prone positioning compared to standard of care.

      3. Results

      Among 248 total patients, 125 (50.5%) were enrolled after consent, and 123 (49.5%) were enrolled with deferred consent. The median time between randomization and consent was 1 day (interquartile range [IQR] 0–2 days) in the deferred consent group. Patients in the deferred consent group were more likely to be enrolled on weekends (14.6% vs. 9.6%), male (67.5% vs. 60.8%), and older (median age 60 vs. 54 years) Table 1. The frequency of deferred consent varied significantly by hospital (median 41.5%; range: 0–100%). Patients in the deferred consent group were more likely to require oxygen via face mask (9.8% vs. 2.4%) and less likely to have received remdesivir (27.6% vs. 56%).
      Table 1Baseline characteristics
      CharacteristicDeferred (N = 123)Nondeferred (N = 125)
      COVID status
       COVID-19 test result120 (97.6%)122 (97.6%)
      Randomization timing
       Number of days between admission and randomization1 [1, 1]1 [1, 1]
       Days between randomization and consent1 [0, 2]0 [0, 0]
       Randomized on a saturday or sunday18 (14.6%)12 (9.6%)
      Age
       Median [IQR]60 [49.5, 68]54 [39, 61]
       <5031 (25.2%)50 (40%)
       50–7069 (56.1%)64 (51.2%)
       >7023 (18.7%)11 (8.8%)
      Sex
       Female40 (32.5%)49 (39.2%)
      Date of randomization
       Before Sept 1, 20206 (4.9%)2 (1.6%)
       Sept 1, 2020 to Feb 28, 202184 (68.3%)60 (48%)
       After Feb 28, 202133 (26.8%)63 (50.4%)
      Comorbid conditions
       Diabetes35 (28.5%)32 (25.6%)
       Hypertension47 (38.2%)51 (40.8%)
       Current smoker4 (3.3%)3 (2.4%)
       COPD or asthma8 (6.5%)19 (15.2%)
       Heart failure3 (2.4%)3 (2.4%)
      Illness severity
       Lymphocyte count0.82 [0.6, 1.1]0.9 [0.68, 1.2]
       Creatinine81 [66, 100]76 [63, 93]
       Systolic blood pressure123 [114, 133]122.5 [115, 130]
       Oxygen saturation94 [93, 96]94 [93, 95]
       FiO232 [28, 36]32 [28, 36]
       S/F ratio303 [261, 339]305 [264, 337]
      FiO2 delivery method
       Nasal prong109 (88.6%)113 (90.4%)
       High-flow nasal cannula1 (0.8%)6 (4.8%)
       Face mask12 (9.8%)3 (2.4%)
      Medication
       Dexamethasone115 (93.5%)121 (96.8%)
       Remdesivir34 (27.6%)70 (56%)
       Tocilizumab0 (0%)2 (1.6%)
      Code status
       Full code110 (89.4%)119 (95.2%)
       Do not resuscitate5 (4.1%)0 (0%)
       Other8 (6.5%)5 (4%)
      Abbreviations: COPD, chronic obstructive pulmonary disease; FiO2, fraction of inspired oxygen; IQR, interquartile range; S/F ratio, ratio of saturation of oxygen to fraction of inspired oxygen.
      Missingness for all variables was <2%.
      There was no difference in the rates of study withdrawal ([3%] in each group) or median number of hours spent in prone position within the group randomized to prone positioning (nondeferred = 7 [IQR 2.3–16.8], deferred = 4 [IQR 1.3–12.0]). The rate of serious adverse events was 4.9 % in the deferred consent group and 1.6 % in the nondeferred consent group Table 2.
      Table 2Secondary outcomes and rate of adverse events among patients enrolled with deferred vs. nondeferred consent
      OutcomeDeferred (N = 123)Non-deferred (N = 125)
      Secondary outcomes
       S/F ratio after 72 hours332 [207, 423]345 [260, 446]
       Change in S/F ratio in first 72 hours (median [IQR])7 [−58, 73]60 [−20, 108]
       Change in FiO2 (%) in first 72 hours (median [IQR])0 [−7, 8]−4 [−8, 4]
       Days to discharge (median [IQR])6 [4, 10]4 [3, 6]
       Discharged114 (92.7%)119 (95.2%)
      Serious adverse events
       SAE composite6 (4.9%)2 (1.6%)
      Abbreviations: IQR, interquartile range; SAE, severe adverse event; S/F ratio, ratio of saturation of oxygen to fraction of inspired oxygen.

      4. Discussion

      In this secondary analysis of an international randomized controlled trial of prone positioning for noncritically ill patients with COVID-19, patients who underwent deferred consent were more likely to be male, older, and be enrolled on a weekend. We observed similar rates of both participant withdrawal and protocol adherence.
      Our study has several limitations. First, data were unavailable for why deferred consent was chosen. Second, we did not ascertain patients' and providers’ perspectives on deferred consent, though prior literature has shown that both groups find it to be an acceptable alternative [
      • Manda-Taylor L.
      • Bickton F.M.
      • Gooding K.
      • Rylance J.
      A formative qualitative study on the acceptability of deferred consent in adult emergency care research in Malawi.
      ,
      • Den Boer M.C.
      • Houtlosser M.
      • Foglia E.E.
      • Lopriore E.
      • de Vries M.C.
      • Engberts D.P.
      • et al.
      Deferred consent for delivery room studies: the providers’ perspective.
      ]. Third, this was an exploratory and post-hoc analysis, and thus we did not test for statistical significance and our findings require replication.
      Our results suggest that the use of deferred consent may allow for the inclusion of patients who would not otherwise have been enrolled (e.g., patients hospitalized on weekends), potentially improving the external generalizability of randomized trials.

      References

        • Grimes D.A.
        • Hubacher D.
        • Nanda K.
        • Schulz K.F.
        • Moher D.
        • Altman D.G.
        The Good Clinical Practice guideline: a bronze standard for clinical research.
        Lancet. 2005; 366: 172-174
        • Harron K.
        • Woolfall K.
        • Dwan K.
        • Gamble C.
        • Mok Q.
        • Ramnarayan P.
        • et al.
        Deferred consent for randomized controlled trials in emergency care settings.
        Pediatrics. 2015; 136: e1316-e1322
        • Topolovec-Vranic J.
        • Santos M.
        • Baker A.J.
        • Smith O.M.
        • Burns K.E.A.
        Deferred consent in a minimal-risk study involving critically ill subarachnoid hemorrhage patients.
        Can Respir J. 2014; 21: 293-296
        • Honarmand K.
        • Belley-Cote E.P.
        • Ulic D.
        • Khalifa A.
        • Gibson A.
        • McClure G.
        • et al.
        The deferred consent model in a prospective observational study evaluating myocardial injury in the intensive care unit.
        J Intensive Care Med. 2018; 33: 475-480
        • The ARISE Investigators and the ANZICS Clinical Trials Group∗
        Goal-directed resuscitation for patients with early septic shock.
        N Engl J Med. 2014; 371: 1496-1506https://doi.org/10.1056/NEJMoa1404380
        • Young P.
        • Saxena M.
        • Bellomo R.
        • Freebairn R.
        • Hammond N.
        • van Haren F.
        • et al.
        Acetaminophen for fever in critically ill patients with suspected infection.
        N Engl J Med. 2015; 373: 2215-2224
        • Myburgh J.A.
        • Finfer S.
        • Bellomo R.
        • Billot L.
        • Cass A.
        • Gattas D.
        • et al.
        Hydroxyethyl starch or saline for fluid resuscitation in intensive care.
        N Engl J Med. 2012; 367: 1901-1911
        • The NICE-SUGAR Study Investigators
        Intensive versus conventional glucose control in critically ill patients.
        N Engl J Med. 2009; 360: 1283-1297
        • Shamy M.C.F.
        • Dewar B.
        • Chevrier S.
        • Wang C.Q.
        • Page S.
        • Goyal M.
        • et al.
        Deferral of consent in acute stroke trials.
        Stroke. 2019; 50: 1017-1020
        • Menon K.
        • O’Hearn K.
        • McNally J.D.
        • Acharya A.
        • Wong H.R.
        • Lawson M.
        • et al.
        Comparison of consent models in a randomized trial of corticosteroids in pediatric septic shock.
        Pediatr Crit Care Med. 2017; 18: 1009-1018
        • Burns K.E.A.
        • Zubrinich C.
        • Tan W.
        • Raptis S.
        • Xiong W.
        • Smith O.
        • et al.
        Research recruitment practices and critically ill patients: a multicenter, cross-sectional study (the consent study).
        Am J Respir Crit Care Med. 2013; 187: 1212-1218
        • Byrne M.M.
        • Tannenbaum S.L.
        • Glück S.
        • Hurley J.
        • Antoni M.
        Participation in cancer clinical trials: why are patients not participating?.
        Med Decis Making. 2014; 34: 116-126
        • Manda-Taylor L.
        • Bickton F.M.
        • Gooding K.
        • Rylance J.
        A formative qualitative study on the acceptability of deferred consent in adult emergency care research in Malawi.
        J Empir Res Hum Res Ethics. 2019; 14: 318-327
        • Den Boer M.C.
        • Houtlosser M.
        • Foglia E.E.
        • Lopriore E.
        • de Vries M.C.
        • Engberts D.P.
        • et al.
        Deferred consent for delivery room studies: the providers’ perspective.
        Arch Dis Child Fetal Neonatal Ed. 2020; 105: 310-315
        • Fralick M.
        • Colacci M.
        • Munshi L.
        • Venus K.
        • Fidler L.
        • Hussein H.
        • et al.
        Prone positioning of patients with moderate hypoxaemia due to covid-19: multicentre pragmatic randomised trial (COVID-PRONE).
        BMJ. 2022; 376: e068585