This Kaiser Health News story published in the Washington Post delivered the core of what we expect from any article about a new treatment approach — a balanced, evidence-based look at how well the treatment works, with appropriate attention to costs and harms. We were pleased to see this appraisal linked to a larger debate about our healthcare system and what it can afford to offer in an era of skyrocketing costs. Kudos for an informative and very readable take on a complicated issue.
There are a few areas where the story might have helped readers more:
- definition of treatment resistant depression (interestingly it has differed from FDA for various treatments)
– give the proportion of depressed patients who have treatment resistant depression
– The AHRQ report cited is a recent systematic review which synthesized all the published literature on this topic. Other studies cited in this article were single studies. It would help to start educating readers about the differences in these approaches – and that not all studies are equal.
Although repetitive transcranial magnetic neurostimulation (rTMS) is approved by the FDA, evidence suggests that it offers modest clinical benefit for patients with depression. It is also much more expensive and arguably less effective than existing treatment approaches. An industry is nevertheless emerging to offer this potentially lucrative treatment to the many depressed patients who have not found relief with currently available therapies. (The story might have addressed the issue of potential conflicts of interest more directly; doctors can build or invest in rTMS centers, then refer their patients there.) There will be new pressure for insurers to cover the costs. As this story points out, devoting more resources to rTMS will have repercussions elsewhere in the health care system, so it’s very important to understand how the evidence stacks up.
The costs of rTMS, and the market forces that work to encourage or restrain use of this technology, are well described.
The use of statistics in this piece was generally appropriate, and we were pleased to see reference to the Number Needed to Treat — an intuitive measure of a treatment’s real world effectiveness. If there’s one area in which we think the story could have done a better job, it’s explaining that the benefits reported with rTMS, though larger than those associated with a placebo/sham treatment, are small in absolute terms. This could have been accomplished with less optimistic framing of certain study results. For example, the story notes that “up to 17 percent of patients” were symptom free after treatment in one study — “about twice as many as those treated with a sham rTMS machine.” The story could have qualified this by adding that the vast majority of even rTMS-treated patients (83%) were not symptom-free. In another study, rTMS was “four times as likely to get patients symptom-free as was a sham procedure,” according to the story. But again, the absolute number of patients who benefited was very small — 14.1% for active rTMS vs. 5.1% for the sham group. The authors of that study concluded that “…the overall number of remitters and responders was less than one would like with a treatment that requires daily intervention for 3 weeks or more, even with a benign adverse effect profile.”
Note: the AHRQ review referenced in the story concludes that the evidence in insufficient to determine the effects of ECT vs rTMS but the limited evidence suggest no difference.
The story adequately addresses adverse effects. It says, “Studies have shown that side effects are few and generally minor, such as headaches and scalp irritations.” In addition, “A few small studies have looked at its effects on cognitive function and found either no effect or slight improvement.”
Interesting, in the AHRQ report, rTMS vs ECT had about the same number of withdrawals due to adverse effects, so despite the scarier sounding potential adverse effects from ECT, dropouts due to actual adverse effects were about the same.
Concerns about the quality of evidence are given a good airing in this story, with appropriate emphasis on the fact that we don’t know how long the benefits last or how often patients may need to return for “booster” treatments to maintain the treatment effect. There’s one issue that the story could have addressed more specifically, but didn’t: the fact that the sham rTMS procedure used in many studies may have been easy to distinguish from the real thing, leading to biased assessments of benefits.
A couple of other issues:
- TMS is evolving (how to give, unilateral vs bilateral, number of treatments, etc) – so the number of studies for any particular mode of administration is relatively limited
– TMS has mostly been compared to sham (which is fine for proof of concept but not really what we want to know); we need more studies comparing TMS to the best available alternatives – including combination cognitive behavioral therapy + drug therapy (and some evidence suggest drug selection may matter)
The story specifies that rTMS should only be considered for patients who do not respond to or who can’t tolerate medication.
Ideally, we’d like to see the story spell out the proportion with treatment resistant depression and define more carefully so that a reader would have a better sense of how many patients affected (and if they have depression, is this story applicable to them).
The story represents both sides of this debate fairly, with the perspective of advocates from the device industry balanced against that of insurer-affiliated skeptics who say the evidence is weak. The story failed to clear the bar, however, when reporting comments from a researcher at Brown University who has conducted studies of rTMS. Although her comments are cautious and raise important questions about the treatment, the story did not disclose that she has been a consultant to, and received grant support from, rTMS device manufacturers (as indicated in this slide set from a recent presentation). It would have been nice to hear from an expert with no meaningful financial stake in this issue.
The story does a good job comparing rTMS with the major treatment alternative for patients who do not respond to antidepressant medication: electroconvulsive therapy. ECT has a much longer track record of use and is more effective, but has more risks and side effects and may not be feasible for patients who are not good candidates for anesthesia.
The story could have compared rTMS to FDA approved vagal nerve stimulation (also evaluated in the AHRQ report) – which has even weaker evidence for benefit.
The story notes that an rTMS device has been approved by the FDA and that availability of treatment is growing. It also explains that treatment usually is not covered by insurance, so patients are on the hook for the $6,000 to $12,000 cost of treatment.
Still, availability is very limited in most parts of the country
The story notes that rTMS has been studied since the 1990s. It does not oversell the novelty of this treatment approach.
This story is not based on a press release.