Responding to a new study in JAMA Internal Medicine linking certain drugs to dementia, The Daily Mail ran with the headline: “Popular sleep remedies and hayfever pills ‘increase risk of Alzheimer’s by more than 50%.’” It’s not clear whom the Daily Mail is quoting, or citing, as the research in question never uses the attributed words. Neither do any of the experts.

The study in question, “Cumulative Use of Strong Anticholinergics and Incident Dementia,” reports the results from ten years of tracking older adults and their use of a class of drugs called “anticholinergic drugs.” This class includes a number of popular drugs meant to reduce symptoms of allergies, inability to sleep, anxiety, depression and bladder over-activity. According to the Daily Mail, woe unto those who use these medications: “For those taking the biggest doses of anticholinergic drugs, the risk of dementia was 54 percent higher than those using none,” said the paper, and “the risk of Alzheimer’s was 63 percent higher.”

The incendiary style leaves little doubt that these pills cause Alzheimer’s and other forms of dementia. A list of anticholinergic drugs is provided, a biological mechanism is promoted, and several insets of brand-named drugs are presented with colorful pictures of older people.

The rest of the article does hedge the paper’s bets by turning to a passive voice. The anticholinergic medications are “linked” to dementia. The “risk was higher,” or “The risk of Alzheimer’s —the most common form of dementia—was increased by 63 percent. “ Yet despite this seemingly cautious wording, the medications have been convicted; no one could read this article and still wonder if the link is real, strong, and dangerous. No caveats are given, no room is made for other explanations, no doubt is cast that these medications are the agents of mind deprivation.

The only two scientists quoted reflect doubt about the convictions of causality. One Alzheimer’s research doctor, Dr. Simon Ridley, noted that the study added to existing evidence that there is a “small increased risk of dementia, but the results don’t tell us that these drugs cause the condition.” The lead author of the study noted that no one should stop taking their medications without discussing it with their doctors. It’s almost as if the researchers are being portrayed as overcautious villains in a tale of pharmaceutical betrayal.

Does the study really say that this class of medicines directly leads to significantly increased risk of Alzheimer’s? Not quite.

The study tracked 3434 people over 65, with some replacements for deaths over the years of the study. During a mean follow-up period of just over 7 years, 23.2 percent of the patients (797 participants) developed dementia of some form. Those who had the highest exposure to this class of drugs over the past 10 years also had the highest rates of dementia—and the comparisons seem stark. Sure enough, those who took over 1095 Total Standardized Daily Doses (TSDD) had a statistically significant increased risk of dementia, and in particular Alzheimer’s. TSDD is a sum total of exposure to the class of drugs as measured in the number of units of a drug taken over the previous ten years.

Does this mean that the drugs cause dementia? Maybe. But the key to the caveats is in who the participants were, and whether there are differences among those who use those drugs compared to those who do not. Those taking the drugs were more likely to be depressed, and more likely to have sleeping problems before they even began the study. These issues are independently linked to dementia, suggesting the possibility that patients who took the drugs may have been suffering from underlying issues that could impact the health outcomes down the road.

The data were also not perfect; many over the counter sales could not be recorded for the purposes of determining how much of these drugs the participants took. Limited data were reported on substrata of the data; for example, the study mentions that the effect (a higher risk of dementia for those who take anticholinergic drugs) persists even among participants who only took the drugs to treat allergies. Yet these data are not reported, and we do not know if the difference in dementia rates reaches statistical significance.

Even some silly explanations cannot be ruled out: perhaps those with initially undetected dementia are more likely to lose or forget their prescription and get it refilled, or more likely to have someone helping them make sure they take their medications (resulting in the regular use of drugs that could be taken only when symptoms occur). This would also result in a concurrence of dementia and heavy use of the drugs.

Certainly the evidence that this class of drugs may have an impact on dementia is worth considering, as there seem to be a number of studies finding an observed relationship. But the leap to causality without couching the study in context may result in an unnecessary rush to throw away a whole class of drugs that has other benefits.

Perhaps rather than vilifying the scientists for their words of caution around their research, the Daily Mail could retain the caution for its readers: it’s about a correlation, but only the most irresponsible scientists jump to find reasons (drugs cause dementia) behind what they can only observe (the drugs and dementia are correlated).

Rebecca Goldin is Professor of Mathematical Sciences at George Mason University and Director of STATS.org. She received her undergraduate degree from Harvard University and her Ph.D. from the Massachusetts Institute of Technology. She taught at the University of Maryland as a National Science Foundation postdoctoral fellow before joining George Mason in 2001. Her academic research is in symplectic geometry, group actions and related combinatorics. In 2007, she received the Ruth I. Michler Memorial Prize, presented by the Association for Women in Mathematics. Goldin is supported in part by NSF grant #1201458.

Please note that this is a forum for statisticians and mathematicians to critically evaluate the design and statistical methods used in studies. The subjects (products, procedures, treatments, etc.) of the studies being evaluated are neither endorsed nor rejected by Sense About Science USA. We encourage readers to use these articles as a starting point to discuss better study design and statistical analysis. While we strive for factual accuracy in these posts, they should not be considered journalistic works, but rather pieces of academic writing.