Cluster randomised clinical trial results: clinical effectiveness

Primary outcome: clinical effectiveness

At 6 months, the mean PSS-SR score for surviving patients in the intervention group had decreased from 11.8 (SD 11.2) points in the baseline period to 11.5 (SD 11.5) points in the intervention period. In the usual care group, the mean PSS-SR score had increased slightly from 10.1 (SD 10.6) points in the baseline period to 10.2 (SD 10.0) points in the intervention period. After adjustment, this corresponded to a primary treatment effect estimate (i.e. the interaction between treatment group and time period) of –0.03 (95% CI –2.58 to 2.52; p = 0.98) (Table 25). For full results of the primary outcome model, see Appendix 12.

Secondary outcomes: clinical effectiveness

There were no significant differences between the groups in any of the secondary outcomes, including days alive and free from sedation (mean difference 0.47, 95% CI –1.03 to 1.96; p = 0.54), duration of critical care unit stay (mean difference –0.28, 95% CI –3.45 to 2.88; p = 0.86), PSS-SR score of > 18 points at 6 months (odds ratio 1.32, 95% CI 0.66 to 2.67; p = 0.43), HADS anxiety score at 6 months (mean difference –0.24, 95% CI –1.50 to 1.01; p = 0.70), HADS depression score at 6 months (mean difference –0.22, 95% CI –1.40 to 0.95; p = 0.71) and HRQoL at 6 months (mean difference 0.007, 95% CI –0.063 to 0.076; p = 0.85) (Table 26).

Secondary analyses of the primary outcome

Of all patients surviving to 6 months, 21.2% were missing the primary outcome of PSS-SR score (21.2% in the usual care group and 21.1% in the intervention group). Sensitivity analyses, using complete-case analysis, which assumed that missing outcome data were MCAR had minimal effect on the primary outcome, reporting a treatment effect estimate of –0.02 (95% CI –2.52 to 2.47; p = 0.99).

Among patients with an IPAT score of ≥ 7 points who received at least two stress support sessions, the adherence-adjusted causal effect of the intervention on the mean PSS-SR score was –0.18 (95% CI –5.50 to 5.14; p = 0.95).

In addition, adjusting for site-level standardised mortality ratio in a post hoc analysis also had minimal effect on the primary outcome, reporting a treatment effect estimate of –0.14 (95% CI –2.66 to 2.38; p = 0.91).

Subgroup analyses of the primary outcome

There was no statistically significant interaction between the effect of treatment allocation and time period on PSS-SR scores at 6 months in any of the prespecified subgroups: age, sex, socioeconomic status – quintile of IMD (IMD 2015), duration of delirium, STAI-6 score, surgical status, overall intervention implementation score (for intervention group sites based on data from the process evaluation) or predicted PSS-SR score (heterogeneity of treatment effect). The p-values ranged from 0.08 to 0.93 (Figure 20).

FIGURE 20 Subgroup analyses of the primary outcome. p-values are for tests of interaction. The x-axis is presented on a log scale. The solid line represents no difference between the groups. a, The IMD (2015) is reported by quintiles, with higher values indicating greater deprivation; b, duration of delirium is reported as the number of days on which patients were assessed as positive on the CAM-ICU for delirium; c, scores on the STAI-6 range from 20 to 80, with higher scores indicating greater anxiety; d, predicted PSS-SR score (heterogeneity of treatment effect) is reported by quintiles, from a prediction model for the primary outcome derived using data from patients receiving usual care and adjusted for a priori important covariates (i.e. age, sex, socioeconomic status, duration of delirium, STAI-6 score, surgical status);85 e, site implementation grade encompasses dose, fidelity and reach and is derived from data collected as part of the process evaluation (see Chapter 6 for more information). Reproduced with permission from JAMA 2019;321(7):665–75.71 Copyright © 2019 American Medical Association. All rights reserved. (Figure not reproduced here; see the original article.)

Co-interventions received in the critical care unit

As the POPPI online training was delivered at the site level, it was possible that co-medical interventions received in the critical care unit by patients recruited at the intervention group sites during the intervention period could have been influenced by the online training. For all patients, receipt and duration of a number of critical care medical interventions (i.e. sedatives/anxiolytics/anaesthetics, sleep medications, benzodiazepines, antipsychotics, analgesics, antidepressants, vasoactive agents and mechanical ventilation) are summarised by treatment group and time period (Table 27).