Juan Ramón Barrada
Psychologist in the Behavioural Sciences Methodology Department at the University of Zaragoza
Public health prevalence studies like this one involve a host of decisions—both big and small—that must be made in order to draw overarching conclusions from previous research conducted using methods that are not always fully comparable. One can read them by focusing on their shortcomings (an approach I do not share) or by considering what we can learn from them and what assurances the authors provide.
As for what we can learn, the data are clear. Mental health problems are becoming increasingly common. This aligns with previous findings. It is yet another point in a pattern that has been clear for some time, with the advantage of being the most up-to-date and employing the most sophisticated analytical techniques currently available. These increases are not seen across all disorders but are concentrated in anxiety, depression, eating disorders, and autism. The press release perhaps highlights the data that most strikingly illustrates this pattern: the change between 1990 and 2023 in the total number of people worldwide living with a mental health condition. When one takes into account the change in the Earth’s population over these more than thirty years and changes in, for example, the distribution of age pyramids, the increase remains, admittedly, very substantial.
The research team clearly seeks to account for and neutralize, as much as possible, any factors that might be skewing these results. They detail their decisions regarding data collection and analysis and justify them. If the question is whether the study is free of limitations, the answer here (and in almost any study we can think of) is no; if the question is whether they did the best they could with the available data and whether there are reasons to trust these estimates, I believe the answer is yes.
Among the limitations, I would point out two. On the one hand, one inherent to this type of mental health research. A possible analogy would be trying to estimate how many tall (or short) people there are in the world. The answer to this is inseparable from where we set the thresholds, and these will always be somewhat arbitrary. Something similar happens with mental health. In contrast, when assessing deaths, there is less uncertainty regarding who is classified as deceased and who is not. That said, the authors adhere to the cutoff points agreed upon by the scientific community. On the other hand, the need to expand research in this field is clear. The uncertainty intervals for some specific results are relatively wide in some cases, reflecting that more data would be needed to achieve more refined estimates.
The press release itself indirectly highlights why mental health disorders, despite their prevalence and rising trend, remain relatively neglected public health issues. They are more prevalent among women and minors. These are not the groups that typically drive public policy. What is clear is that the notion that mental health disorders “have nothing to do with me or anyone around me” is an idea that is becoming increasingly unfounded.