Scary independent studies from Australia
So far we have seen a lot of bad studies using the 3D-DXA software, but always did they have a member of Galgo Medical or CETIR Grup Mèdic as a co-author. What really scares me is the emergence of studies independent from these organisation, thus having no financial interest in this technology. You can’t quite blame clinicians that are excited about using a new technology from a seemingly legitimate company. But this is an indication how a fraudulent technology can spread throughout a research community to eventually end up in clinical practice.
I recently found a presentation with several of these studies coming from Australia with the main authors Belinda R. Beck, Amy T. Harding and Conor Lambert from the Menzies Health Institute Queensland [1-4]. The document can be found here and includes the posters referenced below. These include absurd studies such as: Is proximal femur geometry from DXA-derived 3D analysis predictive of pQCT-derived geometry of the tibia. If anyone could tell me what’s the point of this study I would greatly appreciate it.
I have previously explained that the 3D-DXA software can not reconstruct bones from anyone other than an elderly Spanish woman, and even then it provides only an educated guess. To understand what happens when you try to reconstruct the bone of Australian men and woman, please look at the figure below taken form poster .
Although this looks like a pathological heavily deformed hip, it is supposed to be from a healthy patient. It makes me nervous knowing that there are such groups of clinicians who blindly trust the software from a medical device company in such a way that they are willing to accept a result like this.
Now pay attention! The poster with an overview of these shoddy studies, seemingly showing how 3D-DXA can be used in research, is hosted by Altec Medico As., a distributor of the DXA systems in Denmark. Look here ladies and gentlemen. What you see here is a prime example of how medical device companies use gullible medics for the promotion of their quack medical products. We have seen this behaviors before in drug companies, which led to uncountable unnecessary deaths. While a misdiagnosis of osteoporosis or other bone disease might seem rather harmless in comparison, let’s not forget that hip fractures are a major cause of death in our fragile elderly population. A misdiagnosis may lead to a high risk patient remaining untreated, leading to an entirely preventable hip fracture. If this fraudulent technology ends up being used in clinical practice, I will personally lead a class action lawsuits to put on trial all those responsible for the spread of this quackery.
I also want to note that Beck et al.  cites my paper by stating: “The model was developed and validated previously by comparing 30 reconstructed DXA images against a set of 85 QCT scans (Whitmarsh et al., 2010).” I don’t quite understand what this sentence is supposed to say, but is this a confession that my model and software was used?
- Belinda R Beck, Steven L Watson, Lisa J Weis, Amy T Harding, Sean A Horan, Benjamin K Weeks. High intensity exercise improves indices of proximal femur strength in postemenopausal women with low to very low bone mass: Bone geometry and the LIFTMOR trial., Annual Meetings of the ESA-SRB 2016.
- Belinda Beck, Amy Harding, Benjamin Weeks, Steven Watson. Is vBMD from 3D Hip analysis of a standard proximal femur DXA scan related to aBMD at the hip?, ASBMR Annual Meeting.
- Amy T Harding, Conor Lambert, Benjamin K Weeks, Rossana C Nogueira, Steven L Watson, Sally F Dzera, Belinda R Beck. Is proximal femur geometry from DXA-derived 3D analysis predictive of pQCT-derived geometry of the tibia?, Annual Meetings of the ESA-SRB 2016.
- Conor Lambert, Amy T Harding, Steven Watson, Sally F Dzera, Rossana C Nogueira, Benjamin K Weeks, Belinda R Beck. Relationship of lifelong bone-specific physical activity 3 to proximal femur geometry from DXA-derived 3D analysis, Annual Meetings of the ESA-SRB 2016.