Correction by ProQuest


A recent News article1 mistakenly identified Derek Jones as an ex-officio member of the Inter-Agency Advisory Panel on Research Ethics.

More Info

positive and false-negative findings          and 1.1 fractures per 100 patient-years                      3.   Bradburn MJ, Deeks JJ, Berlin JA, et al. Much ado
                                                                                                                about nothing: a comparison of the performance of
rapidly increases as multiple subgroups       in the comparator group in 19                                     meta-analytical methods with rare events. Stat Med
are evaluated.2 Subgroups should only         company-conducted trials).4 Hence, the                            2007;26:53-77.
                                                                                                           4.   Grossman LD. Increased incidence of fractures in
be evaluated for the direction of effect:     available biologic, clinical and epi-                             female patients who received long-term treatment
in our study there was an increased           demiologic evidence confirms that                                 with ACTOS (pioglitazone hydrochloride) tablets
                                                                                                                for type 2 diabetes mellitus. Toronto (ON): Eli
risk of fractures in women with dia-          fractures are a class effect of both of                           Lilly Canada Inc.; 2007. Available: www.hc-sc.gc
betes who received either pioglitazone        the thiazolidinediones: rosiglitazone                             .ca/dhp-mps/medeff/advisories-avis/prof/_2007/actos
                                                                                                                _hpc-cps_2-eng.php (accessed 2009 Mar. 12).
or rosiglitazone.                             and pioglitazone.
    In addition, in implementing a                                                                         DOI:10.1503/cmaj.1090004
random-effects model, Toulis and col-         Yoon K. Loke MBBS MD
leagues not only reduced the power of         School of Medicine, Health Policy and
the meta-analysis3 but also shifted the       Practice, University of East Anglia,                         Correction
                                              Norwich, United Kingdom
weight of the analysis toward smaller,        Sonal Singh MD MPH
short-term studies that recorded rela-        Section on General Internal Medicine,                        A recent News article1 mistakenly iden-
tively few events. When one is dealing        Department of Medicine, Wake Forest                          tified Derek Jones as an ex-officio
with a rare adverse event that occurs         University School of Medicine, Winston-                      member of the Inter-Agency Advisory
only with prolonged therapy, it is only       Salem, North Carolina                                        Panel on Research Ethics. In fact, Jones
with adequately powered trials of                                                                          is the former executive director of the
                                              Competing interests: None declared.
longer duration that one will be able to                                                                   panel, who was interviewed for the arti-
discern these effects, as seen with the                                                                    cle but was not present at the launch of
thiazolidinediones and fracture risk. In-                            
To top