Assessment of the Generalizability of Clinical Trials of Delirium Interventions

Delirium affects up to 7 million hospitalized older adults annually¹ and is associated with increased risk of mortality, institutionalization, and cognitive and functional impairment.² There has been a proliferation of pharmacologic and nonpharmacologic clinical trials to reduce incidence and sequelae of delirium.³ For other neuropsychiatric disorders, exclusion criteria disqualify up to 75% of individuals with the condition under study from participating.⁴ We sought to examine how often common at-risk populations are excluded from clinical trials of interventions for delirium.

We performed a cross-sectional analysis of data from ClinicalTrials.gov.⁵ As a publicly available data source that contains only summary data (rather than information on individual participants), use of data from ClinicalTrials.gov does not require institutional review board approval. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

We selected all trials in which the study type was listed as interventional and outcome measures included the keyword delirium as a condition studied (N = 89). Two independent reviewers (R.C.M., R.E.K.) manually reviewed study inclusion and exclusion criteria and extracted information on participant age, comorbidities, and setting as exclusion criteria; these data were verified by a third reviewer (Y.Z.). Frequencies were calculated for each type of exclusion criteria, stratified by study start date in 5-year intervals from January 1, 2000, to October 15, 2019. The Fisher exact test was used to examine changes in frequencies over time. P values were considered significant at P < .05 using a Hochberg step-down procedure and were 2-sided. Analyses were conducted using R, version 3.6.1 (R Project for Statistical Computing).

Delirium studies were generally small, with median enrollment of 100 participants (interquartile range, 44-283) (Table 1). Of the 89 studies, 50 (56.2%) primarily focused on prevention and 27 (30.3%) on treatment using pharmacologic interventions (52 [58.4%]). Most studies were randomized, controlled clinical trials (80 [89.9%]). Consistent with the small sample size, most trials (64 [71.9%]) were funded by sources other than the National Institutes of Health or industry.

All 89 studies enrolled only adults, with 41 (46.1%) enrolling only adults 60 years or older (Table 2). Most studies (47 [52.8%]) were restricted to surgical units (primarily elective knee and hip replacement), with a smaller percentage conducted in medical surgical (24 [27.0%]) and medical (13 [14.6%]) units. Only 1 study (1.1%) was conducted in nursing homes, 3 (3.4%) in palliative care units, and 1 (1.1%) in emergency departments. A total of 26 studies (29.2%) were restricted to intensive care units. Neurologic and psychiatric comorbidities were commonly excluded: 46 studies (51.7%) excluded individuals with preexisting dementia; 48 (53.9%), other neurological disorders; 37 (41.6%), psychiatric disorders; and 34 (38.2%), substance use. Patients with advanced or terminal illness were also commonly excluded, usually by specifying individuals with limited life expectancy or supportive care only: 12 (13.5%) excluded nursing home residents, 22 (24.7%) excluded individuals receiving palliative care, and 28 (31.5%) excluded individuals with terminal illness. There were no differences over time in exclusion of neurologic or psychiatric disorders, but exclusions based on advanced or terminal illness decreased over time.

The study found that there was an increase in clinical trials for delirium between 2000 and 2019, although the total number of trials remained small. Trials predominantly consisted of pharmacologic interventions for prevention of delirium, with a smaller number focused on treatment.

Limitations are that, although ClinicalTrials.gov captures most trials, other registries may be used. However, small clinical trials on delirium (particularly those not federally or industry funded) may not be reported in any database. Clinicaltrials.gov also does not capture reasons for exclusion criteria. In addition, many exclusion criteria allow considerable discretion on the part of the investigator; thus, actual exclusion rates may be higher than our results indicate.

The study found that most intervention studies for delirium were limited to surgical and intensive care unit populations. Most excluded individuals with neurologic and psychiatric comorbidities, which are common in hospitalized older adults⁶ and associated with increased delirium risk, with no change over time. Similar to other neuropsychiatric disorders, these findings raise concerns about exclusion of a large number of hospitalized older adults. Further work is still needed to increase inclusion in clinical trials on delirium to maximize generalizability of trial results.

Accepted for Publication: June 15, 2020.

Published: July 30, 2020. doi:10.1001/jamanetworkopen.2020.15080

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Martin RC et al. JAMA Network Open.

Corresponding Author: Richard E. Kennedy, MD, PhD, Department of Medicine, University of Alabama at Birmingham, 933 19th St South, CH19-218R, Birmingham, AL 35294 (rekenned@uab.edu).

Author Contributions: Dr Kennedy had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Martin, Kennedy.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Martin, Kennedy.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Fowler, Zhang, Kennedy.

Obtained funding: Kennedy.

Administrative, technical, or material support: DiBlasio, Zhang.

Supervision: Kennedy.

Conflict of Interest Disclosures: Dr Kennedy, Ms DiBlasio, and Ms Fowler reported receiving grants from the National Institute on Aging, National Institutes of Health during the conduct of the study. No other disclosures were reported.

Funding/Support: This work was funded by grants R21 AG057982 and R01 AG060993 from the National Institute on Aging, National Institutes of Health (Dr Kennedy).

Role of the Funder/Sponsor: The National Institute on Aging, National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: This study was presented in part at the 2019 Annual Meeting of the Gerontological Society of America; Austin, Texas; November 15, 2019.