Reporting of treatment heterogeneity proves challenging

A review of randomised controlled trials (RCTs) that had been published in five prominent medical journals has revealed that heterogeneity of treatment effects (HTE) is frequently ignored or incorrectly analysed. The results of this study were published last week in Trials.

Some patients will experience more or less benefit from treatment than the averages reported from clinical trials; the magnitude of such variation in therapeutic outcome across a population is termed HTE. Highly variable treatment response rates are known to exist for many common conditions, including ischemic stroke and diabetes. Identifying HTE is therefore necessary to individualise treatment.

Gabler et al., conducted a review of the prevalence of HTE analyses in 319 RCTs published in Annals of Internal Medicine, BMJ, Journal of the American Medical Association, The Lancet, and New England Journal of Medicine. They found that just 29% of studies reported HTE analysis and were only marginally better in 2004 than in 1994. Another 28% reported subgroup-only analyses, without the formal statistical tests of heterogeneity that are recommended by the CONSORT guidelines.  

The authors conclude that HTE reporting in the general medical literature is neither rigorous nor routine and suggest it may be time to develop new standards for reporting.

Dealing with heterogeneity of treatment effects: is the literature up to the challenge?

Nicole B Gabler, Naihua Duan, Diana Liao, Joann G Elmore, Theodore G Ganiats, Richard L Kravitz

These results follow those of a another study published last year in Trials, which revealed that only 31% of RCTs published in the same leading medical journals reliably accounted for missing data when analysing quality of life outcomes.

In addition to original research relating to RCTs, Trials also encourages the publication of study protocols, recognizing that this reduces risk of non-publication of trial results and facilitates methodological discussion. Such published study protocols, while important to the scientific record, are unlikely to be heavily cited. It is therefore all the more impressive that Trials has increased its Impact Factor in the latest 2008 Journal Citation Reports to 1.74 (up from 1.44 last year). For the first time the journal is ranked above competitors such as the official journal of the Society for Clinical Trials, Clinical Trials (2008 Impact Factor 1.69)  and the Elsevier title Contemporary Clinical Trials (2008 Impact Factor 1.42).

For more information about the journal Trials, please contact the editorial office.

Abigail Jones
Senior Assistant Editor – Trials

View the latest posts on the On Medicine homepage