This article is more than 5 years old. Information may no longer be current.

Patient characteristics differ among RCTs, observational studies in NAFLD

Participants in randomized control trials for nonalcoholic fatty liver disease had significantly different characteristics, such as age, BMI and sex, compared with patients in observational studies, according to a systematic review.

“Given the high prevalence of NAFLD, there is a need for effective therapies to prevent progression of disease and to treat established fibrosis. However, while many treatments have been trialed, few have shown categorical benefit in [nonalcoholic steatohepatitis],” Richard Parker, PhD, from the University of Birmingham, U.K., and colleagues wrote. “Poor external validity may promote ineffective treatments or, conversely, limit the acceptance of effective treatments. These data ... show that characteristics of individuals with NAFLD in [observational studies] differ from those included in [randomized control trials].”

The researchers reviewed 143 human studies from the PubMed and Ovid EMBASE databases, published between 1948 and 2016. Eighty-one were observational studies with 40,014 patients and 67 were randomized control trials with 4,846 patients.

The characteristic differences found between randomized control trials and observational studies included mean age (50 vs. 49.4 years; P < .001), mean BMI (32.1 vs. 29.3 kg/m²; P < .001), prevalence of diabetes (8% vs. 24%; P < .001), and sex (55% vs. 50% men; P < .001).

In a subgroup analysis, the researchers found that there was a significant difference between randomized control trial participants and observational patients with advanced fibrosis regarding age (50 vs. 54 years; P < .001) and BMI (32.1 vs. 33.8 kg/m²; P < .001). BMI also significantly differed between randomized control trial participants and observational patients with NASH (32.1 vs. 32.4 kg/m²; P < .009).

Further subgroup analysis showed that while 8% of randomized control trial participants had diabetes, the prevalence of diabetes was 39% in observational patients with NASH and 45% in patients with advanced fibrosis. Additionally, regarding sex, 55% of the randomized control trial participants were men, while in the observational studies 47% of patients with NASH were men and 38% of patients with advanced fibrosis were men (P < .001).

Among the randomized control trial studies, exclusion criteria included cirrhosis (41% of 2,008 participants), diabetes (24% of 1,114 participants), medications for diabetes (34% of 1,643 participants) and medications to treat dyslipidemia (8% of 357 participants).

“These differences are statistically significant but are often small and may not be clinically significant. However, marked differences exist between [randomized control trial] cohorts and individuals with [advanced fibrosis] who are more likely to progress to liver-related morbidity,” the researchers wrote. “These differences may limit the application of [randomized control trial] data to high-risk patients with NAFLD.” – by Talitha Bennett

Disclosures: The researchers report no relevant financial disclosures.