Tuesday, October 28, 2008


Most research findings include a lot of statistical analysis of data, and many of the conclusions or assertions made in research papers are based on the statistical analysis.

This is a major advance in the science of analyzing and interpreting data.

Yet, there are a few complaints I have about the way statistical analyses are reported:

The application of statistics is meant to give the reader a very clear, objective summary of what data show, or what data mean. The spirit is neutral objectivity, without the biases of arbitrary subjective opinion or judgment, of people "eyeballing" the data and concluding there is something meaningful there, when in fact there is not.

Yet, in most statistical summaries of research data, the words "significant" and "not significant" are frequently used. The criterion for "significance", however, is arbitrarily determined. It is part of the research, or the statistical, culture, to consider that a "significant" difference means that the data shows a difference that could be due to random chance only 5% of the time or less. If the data show a difference which could be due to randomness with a probability of 6%, then the difference would be reported as "non-significant".
This is an intrusion of human-generated arbitrariness into what is supposed to be an objective, clear analysis of data.

What I feel is a much more accurate way to report on a statistical analysis in a research paper is the following:

the probability ("P value") of a difference being due to chance, rather than to a real difference, should always be given prominently in the paper, and in the abstract, rather than the words "significant" or "non-significant". The reader can then decide whether the finding is significant or not.

As far as I'm concerned, any P value less than 0.5 (50%) carries some degree of significance to it, and the reader of a paper or abstract deserves to see this value prominently given. And it seems absurd to me that results showing a P value of 0.06 would be deemed "non-significant" while results with a P value of 0.05 would be "significant".

**note: there are more rigorous and precise definitions for the statistical terms above, I use a somewhat simplified definition to make my general point more clear and accessible; I encourage the interested reader to research the exact definitions.

Another thought I've had is that, when it comes to clinical decision-making, "eyeballing" the data-- provided the data are fairly represented (for example, on a clear graph which includes the point {0,0} ) --can often lead to more intuitively accurate interpretations than some kind of numerical statistical summary. There is more information represented visually in a graph than in a single number which summarizes the graph, in the same way that there is more information in a photograph than in a number which summarizes some quality about the photograph.

The biggest advantage of sophisticated statistical summaries lies in optimizing research resources, such that we can re-direct our attention away from treatments that work less well, and focus instead on treatments that work better, particularly if there are limited resources, and if a given treatment could determine survival (or not). Also, if there is abundant data, but little way of understanding the data well, then a good statistical analysis can guide treatment decisions. It may help to choose the best chemotherapy drug for cancer, or the best regimen to manage a heart attack. For depression, though, and perhaps other mental illnesses, the statistical analyses can often add more "fuzziness" and distortion to clinical judgment, unless the reader has a sharp eye to recognize the many sources of bias.

Monday, October 27, 2008


Many people believe that sugar (sucrose) intake causes behavioural problems. The two most common specific beliefs are that eating sugar causes worsened hyperactivity; or that eating sugar causes a rush of energy, followed by a plunge into fatigue as the sugar level "crashes".

In fact there have been a lot of very good studies looking at this, and the evidence is quite clear that sugar does not cause hyperactivity. There is an association between high sugar intake and antisocial behaviour, but the relationship is probably not causal. It is much more plausible that those with more antisocial behaviour in the first place happen to choose to consume more sugar.

In some of the prospective, randomized studies, in fact, individuals consuming sugar (instead of a placebo) did better, particularly in terms of learning tasks.

The evidence is also very clear that so-called "reactive hypoglycemia" is very rare, even in people who insist that they have it. However, there may be some individuals who become more irritable as their blood glucose level drops, even if the drop is not down to clinically hypoglycemic levels.

There is evidence that some individuals may respond adversely, in terms of their behaviour, to certain foods, but actually sugar is not a common such food, according to well-controlled studies.

I do affirm that moderating sugar intake, and also eating meals with a lower glycemic index, is part of overall good long-term health.

But most of the claims about sugar influencing behaviour adversely are part of a myth, not supported by clear evidence.

For an excellent review of the evidence on this matter, see this article by David Benton (May 2008):

Friday, October 24, 2008

Chocolate is Good for You

There is evidence that dark chocolate is healthy for you in a variety of ways:

1) May reduce high blood pressure and improve other cardiac risk factors:


Here's a reference to an article in JAMA, one of the world's leading medical journals:

2) Here's a whimsical reference to a study suggesting that pregnant women who eat chocolate end up having babies with more positive temperaments, also these women may be more resilient to stress during the pregnancy

However, it should be acknowledged that many people have trouble moderating their chocolate intake. For some people with atypical depression or eating disorders, excessive chocolate consumption can be part of their unhealthy eating behaviours. Yet, it pleases me to know that chocolate (dark chocolate, or cocoa, in particular) is good for you (in moderation of course)

Saturday, October 18, 2008


Cymbalta (duloxetine) is one of the newer antidepressants on the market. It is being quite aggressively marketed. I notice various lecturers speaking enthusiastically about its merits.

I am immediately wary of this phenomenon:
-new drug enters the market
-lots of advertising
-lots of money to pay for psychiatrists (some of whom, to my embarrassment, are mood disorders specialists) to give educational lectures about the new drug


-down the road, it is often found that the new drug is not quite as spectacular or good as everyone was thinking...
-most often, the new drug is effective and useful, but not qualitatively better than any other related drug
-the new drug becomes simply another option to try, that might work, or might not, for someone struggling with depression
-the new drug otherwise does not become a "first choice" for someone wanting to start a medication, because the many other drugs in its category (e.g. antidepressants) work just as well, on average

--my reading of the evidence is that duloxetine IS an effective antidepressant. It is an appropriate option for treatment-resistant patients who have tried many other medications. But it is no more effective than its competitors on average. I also do not see any compelling advantages with respect to side-effect profile. And it is more expensive! (of course it has more marketing dollars behind it right now, but should this be a reason to consider it first?)
--it is also being marketed for treating pain syndromes -- yet there are few studies comparing duloxetine with other existing antidepressants, for treating these pain syndromes -- so its apparent advantage may simply be due to the fact that no one has done adequate comparative studies including other treatments.

When looking at references regarding a new treatment, the interested reader should take a look at who the authors are, who sponsored the research (was it the drug company?), how often the authors have been involved by industry-funded research, and how frequently the authors have received money for giving lectures supporting the use of the new treatment. Also, look carefully at what the paper is saying as a conclusion vs. what the data from the paper shows: do the authors try to aggrandize the new drug using words (e.g. "this verifies that... is effective and safe, ... ") while the data from the very same study actually show that the new drug is not actually any better than the old drug?
Also, if the study is comparing one drug with another, are the results biased? For example, did the study protocol only allow a limited, less-than-optimal dose of the comparison drug?

Here are some references:

Friday, October 17, 2008

Vitamins & other nutritional supplements

There are many people who believe that nutritional supplements can help with a variety of health problems. There is a field called "orthomolecular psychiatry", in which the practioners believe that vitamin supplements and other nutritional adjuncts can treat mental illnesses. I consider many of these beliefs to be spurious, and to be lacking a significant evidence base. I think many of those who benefit from these treatments are experiencing a combination of a placebo effect, and are perhaps benefiting from the psychotherapeutic care involved as well. Perhaps some of these individuals are also benefiting from not having the side-effects of other conventional therapies that were not working for them.

Here is a look at the evidence regarding vitamin supplementation; I restrict any references to journals that I consider to have a high professional standard:

1) Folic acid. There are a few studies which suggest that folic acid supplements may help augment the effectiveness of antidepressants. Here are a few references:
(in this study, adding 0.5 mg of folic acid daily to the antidepressant fluoxetine, improved depression severity, in women only)


(In these studies, low folic acid levels were strongly associated with resistance to fluoxetine treatment and more frequent depressive relapse. However, I am curious to know if this is merely an association--that is, whether the low-folic acid level group had other factors such as generalized poor nutrition or self-care, etc. that caused them to be more treatment-resistant.)

2) Thiamine (vitamin B1)
Thiamine can treat and prevent an extremely severe neurological syndrome called Wernicke-Korsakov encephalopathy; this syndrome occurs in nutritionally-deficient alcoholics, and causes devastating, permanent, sometimes total inability to form short-term memories. But there is not a lot of evidence about using thiamine to help other psychiatric disorders. Mind you, it hasn't been studied much. There are a few studies in the elderly, which suggest that thiamine supplementation helps with energy and well-being:

3) Other B vitamins:

Here's a study showing a relationship between folic acid levels--but not any other B vitamins or omega-3 fatty acid levels--and depressive symptoms in a group of Japanese adults.

Vitamin b-6: High doses may cause damage to sensory nerves, and I would not recommend taking more than 10 mg daily. Most daily multivitamins have about 3 mg. The toxicity is quite clear for doses over 100 mg/d.



Here is a reference to a new study including over 5000 women, over 7 years of follow-up, from Archives of Internal Medicine. It showed significantly reduced rates of macular degeneration in a large cohort of people taking high-dose b-vitamins (2.5 mg/d of folic acid; 50 mg/d of b-6; 1000 mcg/d of b12):

4) Vitamin D
There is accumulating evidence that higher doses of vitamin D are beneficial for a variety of health variables. Also, it is becoming apparent that many people are vitamin-D deficient, especially those who live in northern climates. Most supplements contain 400 IU, but probably a dose of at least 1000 IU daily could be recommended. Doses less than 10 000 IU have not been associated with toxicity, according to my review of the evidence, but we could conservatively say that doses up to 4000 IU daily are very likely to be safe, unless there is some medical disorder present (e.g. sarcoidosis) that causes a disturbance in calcium metabolism. Here is a very interesting and promising recent study suggesting beneficial mood effects from higher-dose vitamin D supplementation; the study is from a major, highly respected internal medicine journal:

It should be emphasized that more is not always better! Many vitamins cause toxicity if they are taken in excess. Also, some of the studies are showing that groups who took certain vitamin supplements (such as vitamin E and beta-carotene) actually fared more poorly, rather than better.


5) Omega-3 fatty acids
There is some evidence that omega-3 supplementation (containing the fatty acids EPA & DHA) can help reduce depressive symptoms, particularly in those with bipolar disorder. Omega-3 fatty acids can be found most abundantly in fish such as salmon. Plant sources include canola oil, flax, and walnuts (however, the plant sources only have one of the 3 types of omega-3 fatty acids). Here's a link to recent Cochrane Review abstract:
Here are some links to other reviews:

However the evidence appears to be fairly weak at this point, there may be some publication bias (i.e. studies showing no effect may not have been published) so more research really needs to be done. In the meantime, though, omega-3 supplementation (usually in the form of fish oil capsules) appears to be harmless, and potentially beneficial. The dose corresponding to what many of the studies used is about 3-6 grams of salmon oil daily (usually 3-6 capsules, each of which containing 1 gram).

Omega-3 supplementation may be beneficial in other ways--it may help protect against macular degeneration (vision deterioration during old age). Also there is some data showing that higher omega-3 intakes, or fish consumption, may slow the rate of cognitive decline in old age. Here's the best such study I could find showing this:


I suspect that some of the benefits from increasing omega-3 intake could be augmented by consuming a diet in which other unhealthy lipids are minimized--this would involve reducing omega-6 intake, eliminating trans fatty acids, and moderating the intake of saturated fats. I invite the interested reader to research this subject further.