Treatment of Depression: Sometimes, It’s Hard to Beat Placebo

SOURCE: Papakostas GI, et al. The nature of placebo response in clinical studies of major depressive disorder. J Clin Psychiatry 2015;76:456-466.

In randomized, placebo-controlled medication trials for depression, meta-analyses indicate that substantial improvement occurs on placebo, averaging about 30% (range = 12-52%). Although one might expect that as rates of placebo response rise, so should active treatment response rates increase, that has been found not to be the case. Hence, when placebo-responder rates are high, the likelihood of confirming a favorable response from active treatment diminishes.

Investigation into the factors that affect placebo responses have discerned numerous potential contributors, including illness severity and chronicity, participant age, duration of the trial, and frequency of follow-up. Indeed, it has been posited that persons identified as depressed might be better categorized as “biologic” vs “environmental,” with the latter group demonstrating more potent response to placebo. There has been some suggestion in the literature that clinical trials in major depression have a substantial (up to 4 weeks) run-in phase to better identify placebo responders.

Papkostas et al direct out attention to the possibility that we may have committed a statistical Type 2 error in reference to pharmacology trials in depression: we may have dismissed an intervention when indeed it was actually effective, to some degree due to an overly robust placebo response. Clinical trials in major depression may benefit from attending to issues that tend to magnify placebo response.

NOF Guideline on Prevention and Treatment of Osteoporosis

SOURCE: Altkorn D, Cifu AS. Screening for osteoporosis. JAMA 2015;313:1467-1468.

Despite widespread public awareness campaigns, osteoporosis continues to exact a heavy toll on Americans, who suffer about 1.5 million fractures/year, including 300,000 hip fractures. Population screening for osteoporosis has not been demonstrated to reduce fracture risk or mortality in a randomized trial, but extrapolation of results from interventional osteoporosis trials in men and women treated pharmacologically show impressive reductions in vertebral fractures (approximately one-third), hence the support from various agencies with interest in osteoporosis and its consequences.

The National Osteoporosis Foundation (NOF) suggests screening be routinely performed with bone mineral density (BMD) at age 65 in women and age 70 in men, as well as in men and women aged 50-69 who have risk factors (such as menopause, corticosteroid use, positive family history, prior fracture, and positive FRAX score).

In persons with low BMD, NOF guidelines suggest vertebral imaging dependent on age and other specific risk factors (e.g., low-trauma fracture during adulthood, height loss, or long-term steroid use).

Not all voices are in harmony with NOF. For instance, the U.S. Preventive Services Task Force rated osteoporosis screening in men, regardless of age, as “I”: insufficient evidence to recommend for or against. No large randomized trial of BMD screening in men or women is on the horizon; hence, clinicians may have to rely on expert advice in the interim.

Subclinical Thyroid Dysfunction is Associated with Increased Fracture Risk

SOURCE: Blum MR, et al. Subclinical thyroid dysfunction and fracture risk: A meta-analysis. JAMA 2015;313:2055-2065.

Most commonly, when clinicians hear the words “subclinical thyroid dysfunction,” they immediately think of subclinical hypothyroidism, even though subclinical hyperthyroidism also deserves a place of recognition because it too is associated with adverse outcomes, such as arrhythmias, especially atrial fibrillation. Subclinical disease (hypo or hyper) is defined as a thyroid-stimulating hormone (TSH) outside the reference range (above or below) with a normal T4 level and no symptoms.

Blum et al investigated databases from the United States, Europe, Australia, and Japan (n = 70,298) to elucidate the relationship between subclinical thyroid dysfunction and fractures. The association of any measureable degree of hyperfunction of the thyroid with fracture risk is fairly straightforward to explain, since frank hyperthyroidism is associated with reduced bone mineral density. Based on 762,401 person-years of follow-up, Blum et al found an increased risk of fracture in persons with subclinical hyperthyroidism, and there was a “dose/response” relationship. That is, the lowest levels of TSH (indicative of relatively greater thyroid hormone excess) were associated with greater risk for fracture. Overall, incident hip fracture was 36% more common in persons with subclinical hyperthyroidism. No relationship between subclinical hypothyroidism and fracture was found. Whether treatment of subclinical hypothyroidism will reduce risk for adverse outcomes — such as fracture or atrial fibrillation — remains to be determined.