Did Mom Smoke? -- New Tourette Study

[Retired]

Association Between Maternal Smoking and Increased Symptom Severity in Tourette?s Syndrome

Carol A. Mathews, M.D., Brianne Bimson, B.S., Thomas L. Lowe, M.D., Luis Diego Herrera, M.D., Cathy L. Budman, M.D., Gerald Erenberg, M.D., Allen Naarden, M.D., Ruth D. Bruun, M.D., Nelson B. Freimer, M.D. and Victor I. Reus, M.D.

OBJECTIVE: Substantial evidence suggests that both environmental and genetic factors contribute to the development and clinical expression of Tourette?s syndrome. Although genetic studies of Tourette?s syndrome are common, studies of environmental factors are relatively few and have not identified consistent risk factors across studies. This study examines in a large cohort of subjects (N=180) the relationship between prenatal/perinatal adverse events with Tourette?s syndrome severity as determined by tic severity and rates of commonly comorbid disorders such as obsessive-compulsive disorder (OCD), attention deficit hyperactivity disorder (ADHD), and self-injurious behavior.

METHOD: Tic severity, OCD, ADHD, self-injurious behavior, and exposure to a variety of prenatal/perinatal events were systematically assessed in all subjects enrolled in three genetic studies of Tourette?s syndrome. Using linear and logistic regression, a best-fit model was determined for each outcome of interest.

RESULTS: Prenatal maternal smoking was strongly correlated with increased tic severity and with the presence of comorbid OCD in these Tourette?s syndrome subjects. Other variables, such as paternal age and subject?s birth weight, were significantly but less strongly associated with increased symptom severity. The authors found no association between symptom severity and hypoxia, forceps delivery, or hyperemesis during pregnancy, which have been previously identified as risk factors.

CONCLUSIONS: This study identifies prenatal maternal smoking as a strong risk factor for increased symptom severity in Tourette?s syndrome.

American Journal of Psychiatry 163:1066-1073, June 2006
doi: 10.1176/appi.ajp.163.6.1066

 

[David Baxter PhD]

What's important to remember in research of this type is that they are identifying risk factors, not demonstrating cause and effect.

For example, in forensic psychology research and criminal risk prediction, there are several risk factors and also several protective factors which can counteract the risk factors when present. These are determined via correlational statistics.

In the present case, it may well be that there is some genetic, hormonal, neurochemical, or personality factor that is causally related both to increased likelihood of smoking AND increased likelihood of having offspring with Tourettes. If so, it would not be the fact that mothers smoked during pregnancy but that other mediating factor that was the cause.

 

[Holly]

Hi TSOW,
Very interesting article about smoking, good post, thanks!

 

[Retired]

In the present case, it may well be that there is some genetic, hormonal, neurochemical, or personality factor that is causally related both to increased likelihood of smoking AND increased likelihood of having offspring with Tourettes

I'm not sure if I understand the full significance of your analysis.

We know that Tourette is a genetic disorder so one of the parents need to be a carrier, and even at that it's not certain they will have a child with Tourette.

My understanding of what this study suggests is that a mother's smoking during pregnancy increases the risk of greater severity of symptoms and/or co-morbid conditions such as OCD, ADHD and SIB.

Am I on the right track?

 

[David Baxter PhD]

Yes. My points are two:

1. To say that something is "genetic" or "inherited" doesn't tell us WHAT exactly is inherited. Schizophrenia seems to have a clear genetic component but exactly what is transmitted genetically is not clear - same thing with Tourette Syndrome as far as I am aware.

2. Studies like this one do not prove causation - they only show correlation. It is conceivable that the factors that lead to mothers smoking in pregnancy are the same factors that are related to increased severity of Tourettes.

Let's say, for example, that higher levels of a certain neurotransmitter, dopamine, cause Tourette Syndrome (this isn't true but just let's play with it hypothetically). Now we know that there are different levels of Tourette symptoms - perhaps because the higher the dopamine level, the greater the severity of Tourette sumptoms. Now what if I showed you that higher dopamine levels also cause smoking. Then what you might want to conclude is that both Tourette and smoking are caused by higher levels of dopamine, with symptoms of each increasing in direct proportion to how much dopamine is in the brain. In such a case, the higher levels of dopamine would make it more difficult for the mom to quit smoking during pregnancy and she passes on those higher dopamine levels to the baby and the baby develops Tourette, NOT because she is smoking but because she has high dopamine levels. They are correlated but not causative.

 

[Steph]

In the present case, it may well be that there is some genetic, hormonal, neurochemical, or personality factor that is causally related both to increased likelihood of smoking AND increased likelihood of having offspring with Tourettes. If so, it would not be the fact that mothers smoked during pregnancy but that other mediating factor that was the cause.

Dr. Baxter please show me the evidence where you came to this conclusion.

Tourette Syndrome is a genetic disorder. This disorder has nothing to do with a personality factor or hormones.

The research I have done has shown that TS is an increase amount of the neurochemical dopamine and this is what causes the extra movements that is characteristic of TS.

No mother is going to allow a cause and effect study on their baby!! Besides tell me what ethical board would allow such a study? Correlation is the best that we have and what is only available!

Steph

 

[David Baxter PhD]

1. It's not a conclusion. It's an observation that one cannot conclude causation from such studies, only correlation. That's what the concept of intervening or mediating variables is all about.

2. As I said, to say something is a genetic disorder does not tell us what is being inherited. We know that schizophrenia is "genetic". Why do not ALL children of a schizophrenic parent develop schizophrenia? Why is it that even in identical (monozygotic) twins the concordance is not 100%?

3. The relationship between neurochemistry and conditions like Tourette Syndrome is not nearly as simple as you have described. Dopamine levels seem to be implicated in several different conditions, not just Tourettes. What is it that makes whatever is going on be expressed as Tourettes in some people and something else quite different in another?

No mother is going to allow a cause and effect study on their baby!! Besides tell me what ethical board would allow such a study? Correlation is the best that we have and what is only available!

4. I am not trying to say research like this shouldn't be done - just that we need to be careful about what conclusions we draw from them. It is possible in some cases to get down to the level of cause-effect at the genetic level, as in the case of Down's Syndrome or "extra-Y" syndrome in XYY males or Turner's Syndrome "XX" in females. But in the case of most mental health conditions, we are not even close to that yet.

 

[Steph]

In my opinion, the big problem is that there is not enough studies done on mental illness.

Mental illness does not get the research money nor the publicity that other diseases get, including diseases that does not affect as many people.

We have to be advocates, among others, toward the underprivileged, poor, and the mentally ill in order to have a society that as Canada claims to be "a just society".

Steph

 

[David Baxter PhD]

I don't mean to be contrary but my feeling is that the problem isn't a lack of funding for research into mental health issues - there's actually quite a lot of that.

The enormous problem in this country (Canada) is that there isn't nearly enough money or resources put into TREATMENT and SUPPORT resources for mental health patients. And I'm just cynical enough about politics to believe that a big reason for that is that many of those suffering from more serious or debilitating mental disorders probably are not going to vote. Politicians worry about keeping people who do vote happy - they're not inclined to care much about what people who don't vote think or want.

Given this, it's really up to the rest of us who do vote and care about mental health issues to advocate on their behalf.

 

[Steph]

Well, I don't mean to be contrary either.

I think there is a devasting lack of research. A big wish that I have is that the gene is found for TS.

But I do agree with you that more treatment and support and don't forget education for mental health patients, their families, and society.

Politicians look at their main job as getting elected again and that focus needs to be changed.

Steph

 

[Holly]

The enormous problem in this country (Canada) is that there isn't nearly enough money or resources put into TREATMENT and SUPPORT resources for mental health patients. And I'm just cynical enough about politics to believe that a big reason for that is that many of those suffering from more serious or debilitating mental disorders probably are not going to vote. Politicians worry about keeping people who do vote happy - they're not inclined to care much about what people who don't vote think or want.

Given this, it's really up to the rest of us who do vote and care about mental health issues to advocate on their behalf.

I would agree we have to let the Politicians know about issues in respect to mental health and care, advocates have been trying for years. It just falls on deaf ears in this country!

 

[Steph]

what is being inherited

Is Tourette syndrome inherited?

Tourette syndrome is inherited as an autosomal dominant trait with incomplete penetrance and variable expression. What this means is as follows:

Autosomal: The gene for Tourette syndrome is autosomal; it is on one of the autosomes (the non-sex chromosomes). A person, male or female, with the Tourette syndrome gene has a 50-50 chance of passing on the gene to each one of his or her offspring.

Dominant: The gene for Tourette syndrome is capable of expressing itself in single copy. This means that the gene from one parent is sufficient to cause the syndrome.

Why is it that even in identical (monozygotic) twins the concordance is not 100%?

Incomplete penetrance: If a gene is completely penetrant in a population, everyone with the gene expresses it. However, not everyone with the Tourette syndrome gene has Tourette syndrome. The Tourette syndrome gene is thus incompletely penetrant.

What is it that makes whatever is going on be expressed as Tourettes in some people and something else quite different in another?

Variable expression: The gene for Tourette syndrome may express itself as a milder tic disorder, or as obsessive compulsive behaviours, or possibly as attention deficit disorder with few or no tics at all. A higher than normal incidence of milder tic disorders and obsessive compulsive behaviours has been found in families of individuals with Tourette syndrome.

Source

Steph

 

[Retired]

Thanks for those important definitions, Steph!

 

[Steph]

You are welcome. :goodjob:

Steph

 

[David Baxter PhD]

But, again, what this says to me is that, as with other mental health disorders (like schizophrenia) which appear to have a genetic component, it is not known what specifically is inherited - just that something is - and it is not known why inheriting the gene(s) or chromosome(s) does not guarantee that all members of a family will develop the disorder. To say that it is due to "incomplete penetrance" does not answer that question but merely restate it or acknowledge that it is true.

I use the example of schizophrenia because again we know that is genetically transmitted. In a family where one member develops schizophrenia, you will typically see tnedencies toward anxiety, excessive worrying, perhaps obsessive-compulsive traits, depressive tendencies, an increased risk for bipolar disorder, etc., and some family members who are asymptomatic So just as with Tourette Syndrome, we can say that something is being inherited bit not what. It seems to be expressed more as a vulernability than anything else and we can surmise that it is expressed as some neurochemical deficit or imbalance or perhaps some dysfunction in certain brain areas that seem to be functionally related to the symptoms. We know that high stress and high expressed emotion in the individual's environment can be a trigger and/or can exacerbate the symptoms but not exactly why.

Part of my point is if we knew exactly what was being inherited and how and what affected vulnerability to developing the full syndrome, we'd perhaps be a good part of the way toward developing a cure, or at least better symptom management.

 

[Steph]

Could you please not use an example and just talk about Tourette Syndrome.

Also, please provide the source of your information.

Steph

 

[David Baxter PhD]

Steph, I am talking about a statistical principle here, a basic and general problem with interpreting ANY data based on correlation and how that affects the conclusions that can be drawn from those data.

I'm not questioning whether Tourette has a genetic component. I'm pointing out that deminstrating that there is a genetic component does not tell us precisely WHAT is inherited.

In the first instance, the article fingering smoking mothers as the culprit, a secondary but important point is that if such conclusions aren't clarified the effect may be to increase the guilt felt by parents in general and mothers in particular. I don't think anyone would claim that smoking or drinking during pregnancy is a healthy thing to be doing but the fact remains that many women who do, at least in moderation, nonetheless manage to have healthy children.

I've also been following this type of research for other disorders, especially schizophrenia, for about 30 years. Every few years, a new factor is identified and thrust forward as "the cause" and most of them peter out to little or nothing, precisely because of the mediating variable or confounding variable problem. One that I remember well from my student days targeted a particular nutrient as being unusually low in schizophrenics compared to various control groups. On further examination, that turned out to be a result of schizophrenic patients having longer periods of institutionalization and therefore longer exposure to institutional food than the control groups so it was basically one of the CONSEQUENCES of schizophrenia rather than a CAUSE.

 

[Steph]

I think that there are so many people that need support and encouragement, that arguing over different points of view does not seem as important.

It is ok to have different viewpoints.

I respect your opinion and I am sure that you respect mine.

Steph

 

[just mary]

I feel as though I should jump in here, even though I think Dr. Baxter has done a very good job of explaining correlation. And maybe that's all I really want to say is that I agree with him that correlation does not imply causation.

But studies like the one mentioned are important as they could give clues to a cause (Dr. Baxter's example of high dopamine levels resulting in an increased likelihood to smoke as well as TS).

 

[David Baxter PhD]

Steph, I don't want to drag this out but I do want to clarify a couple of issues. This is really NOT a matter of opinion about whether smoking in pregnancy does or does not "cause" increased severity of TS symptoms as much as a general comment about the dangers of over-interpreting correlational research.

The following is a handout for a class I taught at the University of Ottawa a few years ago which summarizes the issue:

CORRELATION

? correlation is a statistical procedure for describing the relationship between two variables
? the correlation coefficient (e.g., the Pearson r statistic) can range from -1.00 to +1.00
? a negative correlation indicates that the two variables move in opposite directions: as the value of Variable A increases, the value of Variable B decreases (and vice versa)
? a positive correlation indicates that the two variables move in the same direction: as the value of Variable A increases, the value of Variable B also increases (and vice versa)
? as the value of the correlation coefficient approaches -1.00 or +1.00, this indicates that the magnitude or strength of the correlation between the two variables is greater
? if r = 0, this means that the two variables are not related ? they are independent of one another
? if r = ? 1.00, this means that the two variables are "perfectly correlated"

Interpreting Correlation Coefficients: Rules of Thumb

? weak correlation: r between 0 and ? .29
? moderate correlation: r between ? .30 and ? .59
? strong correlation: r between ? .60 and ? 1.00

However, whether a correlation coefficient is statistically significant depends on sample size, etc., so even a "weak" correlation may be statistically significant in some cases.

Notes:
1. The correlation coefficient tells us whether two variable are related, i.e., whether they covary but correlation can not tell us anything about cause and effect

Example: suppose we find that sleep disturbance and grades in school are correlated ? does this mean that poor sleep causes low grades?

? this may perhaps be true but it could just as easily be the other way round ? maybe receiving low grades causes sleep disturbance
? or it may be that the relationship between quality of sleep and grades may exist because both of these variables are related to a third variable, like stress/anxiety, or satisfaction/dissatisfaction with current life status or relationships