When patients with rheumatoid arthritis want to know their mortality risk, their clinicians must rely on models generated in studies that consider a wide range of risk factors. These models are only as good as the data and methods used to generate the models, so researchers must take into consideration all possible influencing factors of a patient’s health. For example, a study with cancer patients determined that rheumatoid arthritis does not increase mortality, but Dr. Theodore Pincus, of Rush University Medical Center in Chicago, believes the study may have been flawed.
“Could the inclusion of quantitative measures of inflammation, physical function or socioeconomic status have changed these findings?” asked Dr. Pincus in a viewpoint article, “Rheumatoid arthritis: Predicting mortality in RA: the quest for useful information,” that was published in Nature Review Rheumatology.
“In a hypothetical scenario, if only a small fraction of the study patients with rheumatoid arthritis and cancer had high levels of systemic inflammation, which seems a reasonable assumption, an effect of rheumatoid arthritis on cancer progression might have been detectable only in these patients,” explained Dr. Pincus, “whereas any differences in progression between all cancer patients who had or did not have rheumatoid arthritis might not be significant.” In other words, since so few patients with cancer have documented quantitative inflammation (such as 28-joint disease activity score) in rheumatoid arthritis, the effects of inflammation may have been underestimated in the model.
Adding to missed opportunities to enhance the power of predicting mortality in cancer patients with rheumatoid arthritis is the fact that physical function and socioeconomic status were not considered. According to Dr. Pincus, more than 30 years of research have brought attention to the involvement of these two factors in rheumatoid arthritis. Yet in clinical care, data gathered from patients remains largely ignored and is shadowed by laboratory tests and imaging data.
Dr. Pincus believes the quality of physical function and the level of formal education attained (an indicator of socioeconomic status) needs to be considered when predicting mortality. “Poor physical function predicts severe outcomes of rheumatoid arthritis, such as work disability, joint-replacement surgery ad mortality, at far higher levels of significance than laboratory tests and imaging data,” stated Dr. Pincus. “[Furthermore], a a patient’s formal education level… is a major risk factor for progression and mortality of rheumatoid arthritis and many chronic diseases, including cancer.”
Thus to improve the predictions from the model that was generated to evaluate the chance of mortality in patients with cancer and rheumatoid arthritis, researchers need to acquire reliable, complete data that involves factors such as inflammation. Including these data into the model will lead to more accurate predictions.