Representation of people with comorbidity and multimorbidity in clinical trials of novel drug therapies: an individual-level participant data analysis - PubMed (original) (raw)

Representation of people with comorbidity and multimorbidity in clinical trials of novel drug therapies: an individual-level participant data analysis

Peter Hanlon et al. BMC Med. 2019.

Abstract

Background: Clinicians are less likely to prescribe guideline-recommended treatments to people with multimorbidity than to people with a single condition. Doubts as to the applicability of clinical trials of drug treatments (the gold standard for evidence-based medicine) when people have co-existing diseases (comorbidity) may underlie this apparent reluctance. Therefore, for a range of index conditions, we measured the comorbidity among participants in clinical trials of novel drug therapies and compared this to the comorbidity among patients in the community.

Methods: Data from industry-sponsored phase 3/4 multicentre trials of novel drug therapies for chronic medical conditions were identified from two repositories: Clinical Study Data Request and the Yale University Open Data Access project. We identified 116 trials (n = 122,969 participants) for 22 index conditions. Community patients were identified from a nationally representative sample of 2.3 million patients in Wales, UK. Twenty-one comorbidities were identified from medication use based on pre-specified definitions. We assessed the prevalence of each comorbidity and the total number of comorbidities (level of multimorbidity), for each trial and in community patients.

Results: In the trials, the commonest comorbidities in order of declining prevalence were chronic pain, cardiovascular disease, arthritis, affective disorders, acid-related disorders, asthma/COPD and diabetes. These conditions were also common in community-based patients. Mean comorbidity count for trial participants was approximately half that seen in community-based patients. Nonetheless, a substantial proportion of trial participants had a high degree of multimorbidity. For example, in asthma and psoriasis trials, 10-15% of participants had ≥ 3 conditions overall, while in osteoporosis and chronic obstructive pulmonary disease trials 40-60% of participants had ≥ 3 conditions overall.

Conclusions: Comorbidity and multimorbidity are less common in trials than in community populations with the same index condition. Comorbidity and multimorbidity are, nevertheless, common in trials. This suggests that standard, industry-funded clinical trials are an underused resource for investigating treatment effects in people with comorbidity and multimorbidity.

Keywords: Comorbidity; Multimorbidity; Randomised controlled trials.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

Fig. 1

a Initial identification of individual-level participant data trials from trial repositories. See Additional file 1 for a detailed description of the selection process. Abbreviations are as follows: MeSH, Medical Subject Headings; WHO-ATC, World Health Organization Anatomic Therapeutic Chemical classification scheme; CSDR, Clinical Study Data Request repository; YODA, Yale Open Data Access repository; NIH, National Institutes of Health Biologic Specimen and Data Repository Information Coordinating Center repository; and IPD, individual-level participant data. b Definition of “denominator trials” using the US clinical trials registry (

clinicaltrials.gov

) and the effect of restricting the individual-level participant data trials to this denominator set. The height of each box on the horizontal axis corresponds to the simultaneous effect of applying the eligibility criteria to the denominator set of trials (the leftmost chart) and the three numerator sets of IPD trials. For brevity, the leftmost flowchart shows both the eligibility criteria and the trial counts whereas the other three flowcharts show only the trial counts. Of the final set of 124 trials, further 8 trials were excluded either because the index condition was either difficult or impossible to accurately identify within the primary care record or because we judged that concomitant medication may be difficult to interpret in the context of the index condition (see Additional file 1, section 1.7 for details)

Fig. 2

Fig. 2

Scatterplot of the prevalence for each comorbidity for each index condition, for the community-based sample and for clinical trial participants. Black circles indicate the community-based cohort and red circles trials. The _x_-axis is sorted according to the prevalence of the comorbidities in the community-based sample. The sort order was obtained by ranking the comorbidities from commonest to least common for each index condition, then by taking the median across all index conditions. The individual panels are sorted by the mean comorbidity count for each index condition, from highest to lowest. Where the index condition was judged to be the same as the comorbid condition, the comorbidity was not defined, which accounts for apparently missing points on the graph. So, for example, for people in the community sample who had migraine, the most common comorbidity was chronic pain with the next most common being cardiovascular disease

Fig. 3

Fig. 3

Proportion with each comorbidity count in trials and community: stratified by index condition. This plot indicates the proportion of comorbidity counts for each index condition. The height of the plot indicates the percentage of participants/patients with a particular count for each index condition. For community-based patients, the proportion of patients with each comorbidity count has been standardised to the trial populations; this was done by applying age-sex-specific proportions to the age-sex distributions of the trial participants. For the trial participants, where there were multiple trials per condition, the proportion with each comorbidity was obtained from the modelled mean comorbidity counts for each index condition (see Table 2), under the assumption that the proportion of trial participants with each comorbidity count follows a Poisson distribution. Where there was only a single trial for a given condition (e.g. osteoarthritis), raw proportions are given. See Additional file 5 for further details of these analyses

Fig. 4

Fig. 4

Ratio of mean comorbidity counts between community and trials: condition- and trial-level comparisons. Points represent the ratio of mean count between community patients and trials, and the bars represent 95% credible intervals. Trial estimates are represented by solid circles, and index-condition-level meta-estimates are represented with empty diamonds. The ratio represents the mean community comorbidity count for that index condition, divided by the mean trial comorbidity count, i.e. value of 1 indicates no difference in mean comorbidity count, value of 2 indicates the mean comorbidity count is twofold higher in community than in trials, etc. An interactive version of this figure, with links to the

clinicaltrials.gov

registration for each trial, is shown in Additional file 12

Similar articles

Cited by

References

    1. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. 2012;380:37–43. doi: 10.1016/S0140-6736(12)60240-2. - DOI - PubMed
    1. Bursi F, Vassallo R, Weston SA, Killian JM, Roger VL. Chronic obstructive pulmonary disease after myocardial infarction in the community. Am Heart J. 2010;160:95–101. doi: 10.1016/j.ahj.2010.05.004. - DOI - PMC - PubMed
    1. Quipourt V, Jooste V, Cottet V, Faivre J, Bouvier A-M. Comorbidities alone do not explain the undertreatment of colorectal cancer in older adults: a French population-based study. J Am Geriatr Soc. 2011;59:694–698. doi: 10.1111/j.1532-5415.2011.03334.x. - DOI - PubMed
    1. Smith DJ, Martin D, McLean G, Langan J, Guthrie B, Mercer SW. Multimorbidity in bipolar disorder and undertreatment of cardiovascular disease: a cross sectional study. BMC Med. 2013;11:263. doi: 10.1186/1741-7015-11-263. - DOI - PMC - PubMed
    1. Wang Philip S, Avorn J, Brookhart MA, et al. Effects of noncardiovascular comorbidities on antihypertensive use in elderly hypertensives. Hypertension. 2005;46:273–279. doi: 10.1161/01.HYP.0000172753.96583.e1. - DOI - PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources