Friday, October 14, 2016

New old news about antidepressant side effects

When this story on Psyblog went through my twitter feed because @PsychologyNow shared it this week, I was intrigued and then surprised to find it was a study I knew about... published two years ago. Hardly news, then.

But important nonetheless, and sufficient time has passed that we can look at how the study was received.

The paper reports findings based on online questionnaires completed by 1829 New Zealanders who had been prescribed antidepressants in the five years prior. The side-effects they had experienced were the main focus: although a recipients are usually warned about potential side effects, what they are warned about is not necessarily what is then experienced.
The authors identify some emotional or interpersonal effects and characterize them as 'closing down'

  • feeling emotionally numb
  • feeling not like myself
  • reduction in postive feelings
  • caring less about others 

Importantly, they address the question of whether these are caused by the drugs or by the depression. Because participants were asked to rate their level of depression before they started taking the drugs, the relationship between those levels and adverse effects could be checked: the authors conclude that the simplest explanation is that these adverse effects are caused by the drugs rather than the depression, although the authors also consider the possibility that those who feel the drugs are helpful pay less attention to these adverse effects.

The concern, anyway, is that over a third of participants were not told about these adverse effects beforehand, a trend that was decreasing over time - which is a challenge to the ideal that patients are given an informed choice over medication use.

There are to date 15 citations of this article in Scopus, 15 in Science Direct, and 29 in Google Scholar, so we stick with the latter for a quick check of how the finding was received.
1. Karanges et al. (2014) , Durisko et al. (2015), Busfield (2014), Demyttenaere et al. (2015)Botella et al. (2015)  (in Spanish), Hughes et al. (2016), Ruby (2016), Kleszcz (n.d.) and Donskoy (2015) simply report the main findings - although Demyttenaere et al. have an interesting argument themselves about what patients vs physicians expect treatment of depression to result in - and Donskoy cites in an article arguing that the lack of informed choice is a violation of human rights.
2. Read, Cartwright or Gibson cite the article  (10 times)
3. Högberg et al. (2014) focus on the idea that empathy is a hindrance to suicidal impulses and medication that causes a closing down ('care less syndrome') may thereby enable suicide.
4.  A book chapter dealing with the validity of the depression diagnosis
5. A biological experiment
5. Some unclear citations in work I can't access or in student theses (5 of these)

In two years' worth of citations then, I can't find a single objection to the findings. On the contrary, there is only more and more support for the idea that doctors who prescribe antidepressants don't tell patients enough about possible adverse effects. Some argue that this is a violation of human rights and others point out this in itself could affect suicide rates.

Does this resonate with you? Did doctors tell you enough? Let us know @hashtagpsych... Read, J., Cartwright, C., & Gibson, K. (2014). Adverse emotional and interpersonal effects reported by 1829 New Zealanders while taking antidepressants Psychiatry Research, 216 (1), 67-73 DOI: 10.1016/j.psychres.2014.01.042

Friday, October 7, 2016

Gender pay gap - why the disagreements?

Not long ago I wrote in reference to a suggestion that academic feminism is hurting women. The article that had inspired my post included reference to one of the better known critics of reporting about the gender pay gap. The rather troubling piece seemed to suggest that women have already won the fight for equality and it is only radical feminists and other liberals who perpetuate the myths that, for example, there is a gender wage gap. It's been bothering me that not only did I not address this myth-not myth in my post, but that I keep hearing it again and again.

Christina Hoff-Sommers is frequently cited when people want to argue that the gap is a myth - she works for a conservative thinktank in the States and argues that many feminists are lying to us when they present statistics - so for example,

The bottom line: the 23-cent gender pay gap is simply the difference between the average earnings of all men and women working full-time. It does not account for differences in occupations, positions, education, job tenure or hours worked per week. When such relevant factors are considered, the wage gap narrows to the point of vanishing.

Wage gap comic
Cartoon from
In one of her slightly condescending youtube videos, she says that the gap represents only the total difference between all men and women in full-time work without considering differences in occupation, job tenure, positions, education, and hours worked per week. A special example she gives is in medicine, because women are more likely to choose lower paying specialties, like paediatrics rather than cardiology, because they're more likely to choose to work part-time, and because full-time female doctors work 7% fewer hours than males. And women choose to take long periods of absence to start a family. If you account for that, there's no difference. What Hoff Sommers willfully ignores is that just because accounting for these sorts of variables makes the wage gap go away does not mean that those variables themselves are not evidence of some kind of gender wage problem. For example, some questions, why isn't paediatrics paid as much as cardiology? Why are there not more females in cardiology? Is it really the case that female doctors feel the same amount of social pressure to take a break from work to start a family as their husbands? Without a comprehensive answer to these questions, it's merely shifting the blame for the wage gap across from one variable to another and pretending that means there's no problem. I would argue that as long as the statistics show there is a difference, whether the manifestation is at the level of equal pay or at stereotypical patterns of choice does not really matter: there's a problem that needs to be understood. Hoff Sommers prefers to obscure it. At the same time, she claims that only truth will really improve things for women. When she insists that there is freedom of choice, she blames women's free choices for the pay gap. For psychologists it is very hard these days to argue that decisions are made completely freely, in some sort of emotional, social and chronological vacuum; this makes her claim ideological rather than factual. What ideological bent would drive the insistence that the gender pay gap doesn't exist when so many institutions claim it does?

To its credit, the Ministry for Women agree with my assessment of the issue in their recent announcement of the latest figures, which show an increase in the gap. They address the claim made by Hoff Sommers and others, with spectacular use of mocking inverted commas to say that sometimes the gap is 'explained' by other factors. But the Ministry presents those factors as causes of the gap, so there is a difference between explaining the gap and explaining away the gap.

Difference in % between male and female pay, based on data from Ministry of Women

So now that it's been announced that the gender pay gap is up (clearly on the rise since 2012 but with very small increase year on year) - how long before Hoff Sommers gets brought out to argue that it's not a real thing? I'd give it a day or two at most...

Saturday, October 1, 2016

Does The Danish Way Exist and Does it Lead to Happiness?

This week we look at one of the highest ranking stories with the #psychology tag. Tracy Gillett, who runs her own blog, wrote “The Happiest Country on Earth is Proof Positive Parenting Works” about parenting in Denmark. It relates to a book written by Jessica Joelle Alexander and Iben Sandahl that I hadn’t heard of, but according to a one-star review on Amazon, there is not much more in the book than is on the web anyway.

I have three concerns about all of this, and I’ll work through them one by one.

1. What is this Danish way? Can there really be a ‘Danish Way’ of parenting? It’s a smallish country, but I’d be surprised if a distinctive way of parenting could be found that operates independently of other factors.

2. If there is, is it right that Gillett refers to it as ‘positive parenting’ or is that something different?

3. The word ‘proof’ suggests evidence of a solid causal link between the two ideas. In this case we can check for evidence of a solid causal link between positive parenting (or maybe Danish parenting) and being the happiest country on Earth. That’s quite a big causal claim.

4. Denmark is often referred to as the happiest country on Earth and that claim needs some attention.

1. The Danish Way

As I understand it, the book is organised round an acronym PARENT, which is remarkably English for what are supposed to be Danish ideas, and reminds us that it is written for an American audience (see for a quote from the introduction). For information about the six ideas represented by the letters PARENT, look at The Danish Way website, but be prepared to settle for ARENT on that page. The P is missing. Regardless, the six are Play, authenticity, reframing, empathy, no ultimatums, togetherness (hygge). Here’s the kind of claim I’m looking for:

Danes, on the other hand, follow a philosophy called “proximal development,” which basically states that kids need space to learn and grow (with a little help, if necessary).

I dispute most of that sentence. Although there is a researcher named Danish who has some work on proximal development, and some research on proximal development has certainly been done in Denmark, I can’t find evidence for such claim. But the claim doesn’t make much sense anyway. I’m not aware of a philosophy called proximal development and I believe the authors actually refer to Vygotsky’s “Zone of Proximal Development” (ZPD) – not a philosophy, but an incompletely explained developmental psychology concept that has implications close to the reverse of the authors’ claims. Vygotsky’s point was to emphasise the social aspect of learning: we learn how to deal with new challenges not only by ourselves, but with the presence and presumably mentorship or assistance of more competent others. Thus the suggestion that Danish parents do not intervene unless it is absolutely necessary sounds both false and anyway irrelevant to the spirit of what Vygotsky was trying to emphasise. The book supports some of its claims with studies carried out in the US and with reference to the Danish play patrol system, which does indeed seem to carry some elements of ZPD.

So I can only give a mixed answer without reading more of the book. Is there a Danish way? I’m not sure. Where there are differences, like a later starting age for primary school children than is typical in Anglo-Saxon countries, is that a difference in parenting, or is it just better to bring kids up in Denmark ? A life-swap article from The Guardian relating to this is a compelling read – and I’m inclined to agree with this review that if a country’s happiness is to be improved, it will take political upheaval before the kind of parenting suggested in this book is able to occur.

What I can say is that what little I have read of the book does not convince me that there is a Danish way (as opposed, for example, to a continental European or Scandinavian way) worthy of this kind of treatment, even if the ideas presented are in themselves quite intuitively appealing.

2. Positive parenting?

Back to Gillett. Actually her article is mostly a summary of the messages in The Danish Way, and she doesn’t really get to this point, except at the end in the final lament that there is not enough support for the choices that Gillett and friends make. What is positive parenting? Another page on Gillett’s own site is quite useful in explaining it, and she’s right, it’s not that dissimilar to what The Danish Way authors seem to suggest. So The Danish Way is mostly just ‘positive parenting’ – not especially, uniquely, or exclusively Danish.

3. Positive parenting as a cause of something

Gillett has a page dedicated to a defence of positive parenting decisions. There’s evidence that punishment causes problems, but I don’t see evidence of benefits. I understand the logic and I like it. But I’m a step further from being able to say the title claim is valid. The problem I see now is that Gillett’s claim seems to be based on a claim made in The Danish Way. So here’s the summary of the evidence so far:

Denmark is consistently the happiest country. The Danish Way suggests it could be because of parenting. I’m suggesting the book has only a weak connection to any genuinely Danish parenting or one that is not clearly demonstrated on The Danish Way’s website. I am inclined to think that what has been described is probably more representative of positive parenting as explained by Gillett, which means that positive parenting makes Denmark the happiest country, not Danish parenting.

4. Is Denmark really the happiest country?

Gillett refers us to a BBC article about the World Happiness Report 2016, which put Denmark on top and Burundi at the bottom. The article begins to explain happiness through the way people interact with each other and unhappiness through civil unrest. I wonder if The Danish Way would help the Danes stay happy in equally troublesome civil conditions as are faced in Burundi. Isn’t it a little insulting to the people of Burundi?

Have a look at a couple of images copied from the Report (pages 16-22). First, here’s the top 20 countries. Notice that there’s not actually a big difference in scores among these countries.

And the bottom 20 countries are here. Overall there is clearly a difference between the top and the bottom, but is it worth trying to explain differences between individual countries, especially if you clutch at ‘parenting’ as an explanation. If you look at the key below, you’ll notice what factors contributed to the overall index.

The Report uses these individual factors in an OLS regression to show the importance of each of them in contributing overall to happiness. Social support, freedom to make life choices, and generosity all seem to make important contributions to positive affect. Look at the orange and green bars. Much smaller for the bottom 20 countries than the top 20 countries, right? What does that tell you?

My impression, regardless for the moment of whether I think the measures used were sufficient, is that comparisons within the top 20 are silly. Comparisons across time within a country, comparisons between regions… these things are ok.

But I just can’t see that it makes any sense to tell any of the countries on top that they can do anything to compete with Denmark: it’s all so close at the top.


Positive parenting deserves more attention and I agree with many of the fundamental principles but I haven’t yet seen enough to make a comment about it that has any scientific basis.

In the meantime, I am very sceptical about The Danish Way because a) I don’t believe a specific set of parenting strategies can be labelled as Danish and b) it appears that any evidence for effect belongs to the strategies themselves rather than to an overall strategy and c) its popularity seems to be linked to the distinctly unproven claim that this parenting has a causal connection to Danish happiness. For Gillett, the book itself was proof. It isn’t.