Background
During intensive care unit (ICU) admission, patients and their carers experience physical and psychological stressors that may lead to increased anxiety, depression, and post-traumatic stress disorder (PTSD). Improving communication among patients, their carers, and doctors, nurses, and other ICU staff may improve these outcomes. Communication may include information or educational interventions, in different formats, which aim to improve knowledge of the patient's condition, their treatment plan, or challenges they may face after ICU discharge.
Study characteristics
The evidence is current up to 10 April 2017. We included eight studies with 1157 ICU patients and 943 carers of ICU patients. Seven studies are awaiting classification because we could not assess their eligibility, and three studies are ongoing. We included studies that assessed information given to patients or their carers compared to no information, and studies that assessed information as part of a more complex intervention compared to a complex intervention that did not include information or education. Studies included varied information: standardised or tailored to the individual, given regularly or on a single occasion, and that included verbal or written information, audio recordings, multimedia information, and interactive information packs.
Key results
Overall, it is uncertain whether information or education (given alone or as part of a more complex intervention) improves outcomes for patients and their carers following a stay in the ICU. For patients, it is uncertain whether or not information or education reduces anxiety or depression, or improves health-related quality of life. One patient asked to withdraw from the study because they believed that their mental health worsened when they completed a questionnaire to assess anxiety and depression, but it is not clear whether this person received the information intervention or not. No studies reported PTSD in patients. For carers, it is uncertain whether or not information or education reduces anxiety or depression or improves carers' knowledge acquisition or their satisfaction with information provided.
Quality of the evidence
It was not possible for researchers to mask patients and carers to the intervention they received, and it was unclear whether this would affect the results, which relied on self assessments. Study authors did not consistently report rigorous methods for carrying out randomised trials, and we noted some losses of patients and carers during the studies. We found few small studies for this review, reporting limited data for many outcomes of interest. It is uncertain whether information or education is effective due to very low-certainty evidence.
Conclusion
We are uncertain about the effects of information or education interventions given to adult ICU patients and their carers. The evidence was of very low certainty, and our confidence in the evidence was limited. We are aware of three ongoing studies and seven studies that were recently completed but not yet published. These studies may provide additional evidence or improve the certainty in the findings in future updates of the review.
We are uncertain of the effects of information or education interventions given to adult ICU patients and their carers, as the evidence in all cases was of very low certainty, and our confidence in the evidence was limited. Ongoing studies may contribute more data and introduce more certainty when incorporated into future updates of the review.
During intensive care unit (ICU) admission, patients and their carers experience physical and psychological stressors that may result in psychological conditions including anxiety, depression, and post-traumatic stress disorder (PTSD). Improving communication between healthcare professionals, patients, and their carers may alleviate these disorders. Communication may include information or educational interventions, in different formats, aiming to improve knowledge of the prognosis, treatment, or anticipated challenges after ICU discharge.
To assess the effects of information or education interventions for improving outcomes in adult ICU patients and their carers.
We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, CINAHL, and PsycINFO from database inception to 10 April 2017. We searched clinical trials registries and grey literature, and handsearched reference lists of included studies and related reviews.
We included randomised controlled trials (RCTs), and planned to include quasi-RCTs, comparing information or education interventions presented to participants versus no information or education interventions, or comparing information or education interventions as part of a complex intervention versus a complex intervention without information or education. We included participants who were adult ICU patients, or their carers; these included relatives and non-relatives, including significant representatives of patients.
Two review authors independently assessed studies for inclusion, extracted data, assessed risk of bias, and applied GRADE criteria to assess certainty of the evidence.
We included eight RCTs with 1157 patient participants and 943 carer participants. We found no quasi-RCTs. We identified seven studies that await classification, and three ongoing studies.
Three studies designed an intervention targeted at patients, four at carers, and one at both patients and carers. Studies included varied information: standardised or tailored, presented once or several times, and that included verbal or written information, audio recordings, multimedia information, and interactive information packs. Five studies reported robust methods of randomisation and allocation concealment. We noted high attrition rates in five studies. It was not feasible to blind participants, and we rated all studies as at high risk of performance bias, and at unclear risk of detection bias because most outcomes required self reporting.
We attempted to pool data statistically, however this was not always possible due to high levels of heterogeneity. We calculated mean differences (MDs) using data reported from individual study authors where possible, and narratively synthesised the results. We reported the following two comparisons.
Information or education intervention versus no information or education intervention (4 studies)
For patient anxiety, we did not pool data from three studies (332 participants) owing to unexplained substantial statistical heterogeneity and possible clinical or methodological differences between studies. One study reported less anxiety when an intervention was used (MD -3.20, 95% confidence interval (CI) -3.38 to -3.02), and two studies reported little or no difference between groups (MD -0.40, 95% CI -4.75 to 3.95; MD -1.00, 95% CI -2.94 to 0.94). Similarly, for patient depression, we did not pool data from two studies (160 patient participants). These studies reported less depression when an information or education intervention was used (MD -2.90, 95% CI -4.00 to -1.80; MD -1.27, 95% CI -1.47 to -1.07). However, it is uncertain whether information or education interventions reduce patient anxiety or depression due to very low-certainty evidence.
It is uncertain whether information or education interventions improve health-related quality of life due to very low-certainty evidence from one study reporting little or no difference between intervention groups (MD -1.30, 95% CI -4.99 to 2.39; 143 patient participants). No study reported adverse effects, knowledge acquisition, PTSD severity, or patient or carer satisfaction.
We used the GRADE approach and downgraded certainty of the evidence owing to study limitations, inconsistencies between results, and limited data from few small studies.
Information or education intervention as part of a complex intervention versus a complex intervention without information or education (4 studies)
One study (three comparison groups; 38 participants) reported little or no difference between groups in patient anxiety (tailored information pack versus control: MD 0.09, 95% CI -3.29 to 3.47; standardised general ICU information versus control: MD -0.25, 95% CI -4.34 to 3.84), and little or no difference in patient depression (tailored information pack versus control: MD -1.26, 95% CI -4.48 to 1.96; standardised general ICU information versus control: MD -1.47, 95% CI -6.37 to 3.43). It is uncertain whether information or education interventions as part of a complex intervention reduce patient anxiety and depression due to very low-certainty evidence.
One study (175 carer participants) reported fewer carer participants with poor comprehension among those given information (risk ratio 0.28, 95% CI 0.15 to 0.53), but again this finding is uncertain due to very low-certainty evidence.
Two studies (487 carer participants) reported little or no difference in carer satisfaction; it is uncertain whether information or education interventions as part of a complex intervention increase carer satisfaction due to very low-certainty evidence. Adverse effects were reported in only one study: one participant withdrew because of deterioration in mental health on completion of anxiety and depression questionnaires, but the study authors did not report whether this participant was from the intervention or comparison group.
We downgraded certainty of the evidence owing to study limitations, and limited data from few small studies.
No studies reported severity of PTSD, or health-related quality of life.