What Can We Believe?

Gretchen Becker Health Guide
  • When can we believe what we read? It’s not always easy to know.


    One problem is “news” releases and blogposts by people who have something to gain if we follow their advice. The Internet is full of “diabetes cures” that are essentially scams. Most of us know to ignore such fanciful claims. If diabetes could be cured by some nutritional supplement, no one would have diabetes. But others may be less critical and waste time and money on bogus cures.


    Another problem is “scientific studies” funded by some group that sells a particular type of food. Let’s say a study funded by the National Association of Wheat Growers  showing that wheat would reduce blood glucose levels. (I don’t know of any such studies; this is a hypothetical example.)

    Add This Infographic to Your Website or Blog With This Code:


    Such studies may appear to follow scientific methods, but perhaps they did several studies and because of random fluctuation two showed a negative effect and weren’t published, and the third showed a positive effect and hence was published and promoted by the professional organization.


    Another problem is that the population used in some study, even if it’s funded by the government and carefully controlled, may be different from you. Let’s say the study was in Caucasian men aged 40 to 60 years and you’re a 65-year-old black woman. Do you pay attention to the study or ignore it? Not easy to say.


    The problem for us is that the popular press accounts of such studies, either the evening news or science sites like Science Daily, emphasize the results and don’t go into the statistics or the details of the study. And sometimes we don’t have time to do the digging that teases out what the study really showed. Neither do our doctors, who are dealing with increasing masses of paperwork required by the government.


    Now comes a study of studies that suggests that as many as a third of clinical trials could be reanalyzed in ways that modified the conclusions in the original paper. They point out that because much of the raw data that goes into a study is not available to everyone, it’s difficult to do such reanalyses.


    "Without this access, and possibly incentives to perform this work, there is increasing lack of trust in whether the results of published, randomized trials are credible and can be taken at face value,” said John Ioannidis, MD, DSc, professor of medicine and director of the Stanford Prevention Research Center.


    Ioannidis is the author of a widely read 2005 paper claiming that most published research results are false. Now he’s suggesting how this situation can be improved.


    What are we to do in this situation? In most cases, our doctors are too busy to carefully analyze the results of some recent study, and in many cases their background in statistics is not strong even if they have the time.


    I think most physicians are very intelligent and won’t blindly prescribe some drug for you without trying to make sure it’s the best for your particular situation. For example, a clinical study might show that most patients with newly diagnosed type 2 diabetes did best with metformin. On the basis of this, they might start you off with a prescription for metformin.


    Add This Infographic to Your Website or Blog With This Code:

    But they wouldn’t stop there. They’d have you come back in a few weeks to see how it was working for you. If it didn’t help you, or if it helped you but caused severe nausea, then they’d try something else, even if the study showed benefit from the drug for the largest number of patients.


    In the future, they will be able to test people’s genes and predict on the basis of that which drug and which diet would work best for that patient. This is called nutrigenomics. However, this approach isn’t likely to be implemented soon.


    We have to keep this uncertainty in mind. If we read a new study saying that drug/supplement X does something, we shouldn’t ignore it. The study may suggest that the drug works or that it has harmful effects. But we shouldn’t accept it as a fact. We should file it away and keep alert to more information about the drug.


    Sometimes user groups are helpful. The experience of one other patient doesn’t mean much, but if 95% of the people posting have the same experience with the drug, that’s very suggestive that this result is real. It’s still not proof, as perhaps only the people with positive or negative experiences took the time to post.


    It was difficult to have diabetes back in the day when there were only two drugs for type 2 diabetes: insulin and then after World War II the sulfonylureas. But with the myriad drugs available today, not to mention the controversies about diet, it can become very confusing to be a type 2 diabetes patient.


    Despite this, things are definitely improving. Many patients today are able to keep their hemoglobin A1c levels, as well as blood pressure and lipid levels, within normal limits, almost unheard of 25 years ago. Complication rates are down. And people are living longer.


    A Spanish proverb says, “Life is problems. Without problems there would be no life.” Dealing with the complexity of type 2 diabetes is a problem. But at least we’re alive.









Published On: October 22, 2014