Letter to the Editor From Watts and Leslie: Comparisons Between Different Antiosteoporosis Medications on Postfracture Mortality: A Population-based Study

In their recent article, “Comparisons between different antiosteoporosis medications on postfracture mortality: a population-based study.” Wu et al (1) state that “the usage of osteoporotic medication…may lower postfracture mortality.” They use the word “may,” which does not preclude “may not.” In our view, they underestimate the importance of treatment selection bias and especially channeling bias, a form of allocation bias in which drugs with similar therapeutic indications are prescribed to patients with prognostic differences, with the consequence that outcomes may be incorrectly attributed to the drug. Without a rigorous strategy for examining and addressing these biases, no conclusions can be drawn. They compared women users of bisphosphonates and denosumab with those using raloxifene and bazedoxifene. Yes, they found lower mortality in bisphosphonate and denosumab users, but the difference may be more related to who prescribes these drugs (internists vs gynecologists) and by extension who gets these drugs. Their findings in men—higher mortality with denosumab (after any fracture) and ibandronate (after vertebral fracture)—are astounding. Their comparator group is zoledronic acid (raloxifene and bazedoxifene are not used in men)—the only antiosteoporosis drug shown in a placebo controlled trial to reduce mortality risk. The very short duration of medication exposure (mean 1.05 years in survivors vs 0.76 in nonsurvivors) hardly seems sufficient to affect long-term survival. Survivors started treatment later than nonsurvivors (22.42 months vs 15.34 months)—paradoxically suggesting that delaying treatment may actually be protective. Moreover, this might shorten the observation time, and therefore potential for death, among the former. Leslie et al (2) found that women in Manitoba who had a bone density test had lower mortality compared with age-matched controls. It is unlikely that having the bone density test is the reason, but rather the selection of healthier women for investigation and treatment for osteoporosis. Only a randomized trial with an appropriate control group can establish a causal relationship between treatment and outcomes, such as mortality. To that end, Cummings et al (3) performed a meta-analysis of randomized placebo-controlled clinical trials of drug treatments for osteoporosis that included 101 642 unique participants. No significant association was found between mortality and all drug treatments (38 clinical trials) or a bisphosphonate treatment (21 clinical trials). Given such strong evidence, observational studies, such as that of Wu et al, provide little new information or understanding but may add to confusion (4). The authors have nothing to disclose.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.