Can geography affect depression treatment?

The Guardian published some of their own research last week, examining the variations that exist in prescription rates for antidepressants across the UK. The headline statistics were shocking in various ways. Firstly, the Guardian found that the rate of antidepressant prescriptions varied massively from area to area, such that prescription rates in some parts of the country were three times higher than rates elsewhere. The second shocking statistic was that there appeared to be a correlation between geography and prescriptions: the trend was for the highest prescription rates to emerge in northern areas, often in those characterised by socioeconomic disadvantage (such as Salford and Cleveland), with the lowest rates seen in the more affluent southern areas (such as Kensington and Chelsea). The third statistical shocker, in my mind, was the actual per capita rates of prescriptions in raw terms. The highest rates amounted to some 133,000 perscriptions annually per 100,000 population; that is to say, more prescriptions than there are people (of course, pursuing this line of interpretation would involve accounting for the fact that the population base rate will include children and babies and others unlikely to ever be prescribed antidepressants; thus, the rate of prescription per 100,000 adults will be far higher that 133,000). Even the lowest rates of prescription were strikingly high in these terms: the rate in Kensington and Chelsea, for example, was still some 30,000 per 100,000 (total) population. So what’s going on? Should we be shocked?

The impact of these statistics depend on two things: how they are interpreted, and whether they are reliable. As regards interpretation, each of the findings could be seen as alarming. The overall geographic variation in prescription rates could be taken to imply either (a) that rates of depression vary extremely widely across the UK, or (b) that prescription habits of physicians differ dramatically by region. While the apparent correlation between geography and prescriptions was imperfect (there were exceptions; i.e., poor southern areas with low prescription rates), such a link between economic factors and prescription can also be interpreted two ways: (a) it could mean that poverty heightens the risk of depression itself; or (b) it could reflect a difference in the availability of appropriate mental health care, such that northerners find it harder to access time-intensive treatments such as (“talk”) psychotherapy. The third statistic, the actual per capita rates of prescriptions in raw terms could be taken as indicating that a huge proportion of the UK population is depressed. After all, as well as the people receiving prescriptions for antidepressant medication, presumably there are large numbers of others who are successful in obtaining psychotherapy, as well as a further group who are yet to be diagnosed or who, for whatever other reason, are not being treated at all.

So finally we get to the reliability question: is everything as it seems? Well, there are a number of factors that need to be considered before taking these statistics at face value. Firstly, the headline figures refer only to the number of prescriptions per region and do not incorporate the number of prescriptions per person. It could be that the vast majority of prescriptions are for extremely small doses of antidepressants, or are repeat-prescriptions for the same relatively small cohort of patients. In other words, if an average patient receives 12 monthy prescriptions per year (admittedly a fairly rapid rate of repeat prescription) then a figure of 30,000 prescriptions would cover just 2,500 patients. Related to this is the fact that the data do not take account of the size of each perscription (i.e., how much dose or treatment is encapsulated in each script). As doctors are free to issue small or large prescriptions, there could then be sound reasons why prescription rates vary in terms of economic factors. Doctors may provide wealthier patients with larger prescriptions that require them to visit the pharmacy less frequently but which cost more money; while doctors may facilitate poorer patients by providing smaller, and therefore more numerous, prescriptions in order to minimize short-term financial burdens. As such, it is less the case that poor people are more depressed, it is just that they are more poor. Blogger AnneMarie Cunningham, a GP based in Cardiff, has provided analyses of the average costs of prescriptions which corroborates this interpretation (thanks to @amcunningham for tweeting this to me).

It can also be noted that that the Guardian’s statistics do not control for other factors that vary by region (due to migration flows) and that can influence either the occurrence of depression or its presentation (i.e., the extent to which individuals seek treatment when depressed). Such factors include, predominantly, age, but also factors like social attitudes (both toward depression itself and toward conspicuous remedies such as psychotherapy), cultural and religious norms, access to peer-level (compared to family-level) social support, engagement in exercise, and alcohol use. There may even be geographical differences in the training levels, and thus treatment attitudes, of physicians and mental health professionals. Certainly it has long been feared that the training models and age demographics of clinical psychologists may be leading to a brain drain of younger professionals (whose training is more current) into wealthier cities and away from the regions.

Prescription rates are important sources of data but are notoriously difficult to interpret. It is therefore important that they be treated with caution. In principle, when it comes to describing human behaviour, any regional or geography-based generalization is likely to be wrong.



Categories: Health, Newspapers & Magazines, Pharmaceuticals, The Guardian

»So what do you make of this?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: