The story reports findings from a trial comparing a newer diabetes medication, rosiglitazone, to older medications to treat newly diagnosed people with type 2 diabetes. Unfortunately, most of the reporting criteria were not met. The story does provide absolute rates for treatment effects, which is to be commended, but importantly, the treatment effects between rosiglitazone and metformin have uncertain clinical importance, which is not mentioned. While on the surface it appears that because this study is a randomized controlled trial the findings should be robust, there are a couple of important design or methodological weaknesses that really limit the conclusions and were not discussed. One was the outcome measure chosen, which could have made the treatment effects appear more important than they are, and the second was the large number of people who withdrew from the study, so that outcomes for the entire group are not known. Additionally, harms of treatment are minimized and not even named. Other areas for improvement include the lack of an independent source, lack of cost info, lack of availability info, lack of any information about what diabetes is or its seriousness, and lack of information on lifestyle changes, which are an important intervention to make in addition to medications.
The story reports that the new drug, rosiglitizone, is more expensive than older diabetes drugs, yet does not provide any cost information or cost comparisons.
The story does report absolute numbers of people who needed a second medication after five years for each of the three treatment groups. However, the study used fasting blood glucose levels as the unit of measurement for definining when a second medication was needed rather than the more appropriate measure of A1C, a measure of average blood sugar levels over the past 2 to 3 months. Because of this, the treatment effects seem larger than they are if one looks at the absolute difference in A1C between the rosiglitazone and metformin groups. There is no mention of this, nor is there mention that the results should probably be interpreted with some caution since a large number of patients withdrew from the study and these outcomes don't reflect outcomes for the entire group.
The story mentions "heart problems" as a side effect of the new drug and does provide an absolute number of people who experienced this problem in each group. However, the story only provides the number experiencing this problem and readers don't know what proportion of the group this represents or how significant this is. The heart problem in question is heart failure, which should be named and isn't. Other side effects included weight gain and fluid retention and these were also not mentioned.
Although the story does describe the number of participants involved and that they were randomized into three groups that then received different medications, the story does not discuss any study design or methodological weaknesses. Even though this study is a randomized, controlled trial, there are some important limitations: 1) the primary outcome selected could have made treatment effects seem bigger and more clinically important than they really are (at least for comparing treatment effects of rosiglitazone and metformin) and 2) a high proportion of patients withdrew from the study, making any findings weaker, since the outcomes of the entire group cannot be known.
There is no description of what diabetes is or how it affects people. This is a serious condition with serious complications and the number of people with diabetes is only increasing, particularly as obesity rates increase. The treatment effects are overestimated, e.g. the story does not explain that the clinical significance of the effects are marginal, at least when rosiglitazone is compared to metformin. This could be considered "treatment mongering" but will be covered elsewhere.
The only source of information is the lead author, who has received funding from the drug's manufacturer and potentially has a conflict of interest. There is no independent comment on the study's findings.
There is mention of two other diabetes drugs which might be used in a newly diagnosed person. However, there is no mention of lifestyle changes, which is an important intervention to coincide with medication.
The story does not state whether rosiglitazone (Avandia) is widely available to the public, FDA-approved, or only available in research settings.
The story only describes the drugs as "newer" or "older" and doesn't really provide any details of its novelty, e.g. when did this come on the market, etc.
We can't be sure if the story relied solely or largely on a news release, although the only person quoted is the principal investigator.