Bias in clinical trials

Clinical trials test new medications in very small groups of people, evaluating their effectiveness and safety so they can be released to the market for broad use. People in clinical trials often report better improvements from the medications being tested than is reported by similar people not in clinical trials after the medications become widely available. The difference is important because the cost effectiveness of the medication can be greatly exaggerated during the trial, giving it an undeserved benefit in the marketplace. People may end up wasting limited financial resources and precious time taking new medicines that just aren’t as great as they claim to be.

One interesting cause of inflated clinical trial results is the so-called Hawthorne effect. This effect was originally noticed in an industrial setting when workers who were singled out and made to feel important increased their productivity. The definition has been expanded to treatment response in medical settings. In a clinical trial, the Hawthorne effect may be defined as extra improvement caused by increased attention to participants during the clinical trial.

To test the Hawthorne effect we studied 264 RA patients who completed a commercially sponsored trial of a Food and Drug Administration approved RA treatment.We evaluated changes in patient functionality, pain, overall health, and fatigue during three periods: at the start of the trial, at the end of the trial, and 8 months after the close of the trial. Patients were receiving the same treatment the entire time.

We found that from 23% to almost half of the benefits seen in the trial had disappeared 8 months after the end of the trial. The disappearing benefits can be attributed to the Hawthorne effect. Our results suggest that independent follow-up studies, removed from the trial assessment, can and should be used to estimate real-world improvement in new medications.


Progress of fibromyalgia

Fibromyalgia (FM) is a chronic condition for which treatment is frequently ineffective. But in a 2010 study of FM, 25% of patients in the study improved so much that they were not considered to have FM at the study’s end. Other studies have also reported various degrees of improvement.

How do such improvements affect treatment? We looked at the question of symptom variability in 1,555 NDB participants over several years to determine the course of the condition.

People with FM often switch between having and not having FM based on their symptoms, and they have a wide variation in types and severity of symptoms. This makes it hard to give a useful and definite diagnosis of FM in some cases. Symptom severity, as opposed to an on-or-off diagnosis, appears to be more helpful when treating FM.

Careful evaluation of the change in symptoms could reduce unnecessary treatment and side-effects when treating FM with powerful drugs.


Urinary tract infections

The NDB questionnaires ask a lot of questions about infections.We do this because when medications suppress the immune system, which is involved in the damage done by auto-immune diseases, infections can more easily take hold.

Looking at more than 100,000 questionnaires completed by more than 17,000 NDB participants with RA who reported urinary tract infections (UTI), we found a slightly increased risk of UTIs. 11.4% of women and 3.3% of men had UTIs. Only .39% of these infections required hospitalization or intravenous antibiotic treatment.

Prednisone use or diabetes increased the risk of UTI, but biologic medications had no effect.