Dear reader, this is a most interesting topic. Since I already had a text finished about it prior to starting this weblog, I reproduce it here, now. Perhaps it is written in a different style, if you compare it to the material I hitherto brought up. Please don’t worry, I will edit it here and there and smoothen some sections, so that the text fits better with the other stuff. Well, here goes:
OCD and the impact of life events
Because traditional psychology had it that OCD was rooted in the patient’s inner conflicts, and next to nothing was known about how the genetic make-up potentially is involved, it is fitting that the possible impact of life events gets mentioned here. A generally used abbreviation is TLE’s or TLEs, which stands for ‘traumatic life events’, i.e. experiences that have such an impact on the subject that they don’t so much form a normal part of a sequence of individual memories (and thus are integrated in one’s personal history), but instead are more likely to be the cause of anxiety and stress, up to the point where vivid ‘flashbacks’ occur, or depressive disorder is the end point.
Maina et al (1999) reported an interesting observation: in women who gave birth and who experienced a postpartum onset of OCD, severe aggressive symptoms could be seen, in the form of an inner urge or fear to harm the child. In psychiatric terms their giving birth was a TLE that led to a rapid consecutive onset of obsessive thinking. This study also notes that in a larger cohort of subjects, there was no clear association between TLEs over 12 months and eventual occurrence of OCD. No gender differences were detected here. Limitations of the study are its relativel small number of subjects and the short time span researched before possible OCD symptomology. McLaren et al (2001) performed a study on individual perception of OCD; they concluded that high OCD ratings were accompanied by a personal conviction that the subject had a very limited sense of control over stressful life events, and a tendency to extremely suppress anxious thinking. Goodyer et al (2001) note a related and intriguing result: depression-dependent stressful events are more likely to happen in persons with hypercortisolism and OCD. At first this seems to be a ‘chicken-or-egg’ riddle, until one realizes that depression, especially in adolescents, often goes undetected. And then their conclusion might be interpreted as follows: harmful events to a person (as a result of unnoticed depression) can precede onset of OCD. An important study by Lochner et al (2002) shows that severe childhood trauma is associated with later occurrence of OCD and a related condition, named trichotillomania (the compulsive pulling out of one’s hair). Also it should be noted that here a link is made between very early experiences and OCD, as opposed to the method of Maina et al (1999), above. Gothelf et al (2007) presented results that partially contradict those of Maina et al (1999), but altogether form a very useful addition to the above: in a pediatric study, they found that children with OCD reported more ‘total life events’ (in a quantitative sense), more negative life events, and a more negative perception of these events compared to healthy children, as remembered from the 12 months preceding the investigation. They conclude that in general children with anxiety disorders experienced more negative events in a quantitative, qualitative, and specific sense than did healthy children, and this effect was especially severe in OCD. Finally, Cromer et al (2007) performed two elegant studies in adults concerning a possible association between traumatic life events and later onset of OCD symptoms in adults. Firstly (2007a), they found that indeed there is a significant association between negative life events and consecutive OCD, especially where OCD manifests itself in the dimension clusters ‘obsessions/checking’ and ‘symmetry/ordering’. In a related report (2007b), the same group writes that patients, once they are diagnosed with a ‘hoarding’ obsession, are much more likely to have undergone at least one TLE prior to OCD onset when compared to patient that suffer from other OCD symptom dimensions. And moreover, when ‘hoarders’ are taken as a separate group and again questioned and observed, the hoarders that do report at least one TLE show a symptom severity that is very robustly greater than in the subgroup of hoarders that acquired OCD without prior TLEs.
It must be realized that the knowledge about the influence of life events on the occurrence of OCD is limited and fragmented. Partly this is because the potential list of environmental risk factors is huge, and in all probability will never be fully described. Purcell (2002) even goes so far as to state, based on twin studies, that familial life, parent-child interaction, and divorce have a heritable component, apart from these obviously being environmental by their nature. This type of conclusions has its drawbacks. Accounts of familial life are narrative; interviewing children from the same family can produce as many different and contradictory stories as there are children. So a major confounding factor may be interviewer bias, and friction between reports may be resolved erroneously by the interviewer’s interpretation; which in turn can lead to false associations between live events and genetic makeup.
Some of the relevant information has been obtained from patients with OCD/TS comorbidity, so hidden confounding factors may not be ruled out (see, for instance, Sheppard et al, 1999). Apart from psychosocial experiences, also perinatal injuries, infections, and toxic pathogens are associated with higher OCD rates (Moffitt et al, 2005). Vasconcelos et al (2007), for instance, showed that prolonged labour and edema during pregnancy have predictive value for OCD in later life. The study by Moffitt et al is an excellent overview of the technicalities involved in assessing G x E interactions in psychiatric disorders.
Finally, a very intriguing aspect of pediatric OCD might be that several groups independently reported a probable external cause for OCD: namely that it might be a long-term result of a streptococcal infection in children, followed by an auto-immune reaction in the central nervous system. For this putative causative relation the acronym PANDAS has been coined, meaning: ‘pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections’ (Swedo, 2002). OCD and TS symptoms were seen in children that had been exposed to streptococcus infections (Group A Beta Hemolytic Streptococcus, GABHS); and the possibility was raised that these infections might lead to auto-immune crossreactivity with brain epitopes (e.g. in the basal gaglia), and symptoms akin to those of inflammation. Although PANDAS research has been conducted broadly and for several years now, reports still are conflicting. A recent report by Morris et al (2008), for instance, states that serum antibody assessment with immunofluorescence fails to differentiate between Tourette’s, PANDAS, and controls subjects. Also, familial aggregation for OCD/TS has been reported in relatives of PANDAS probands.
So, it is still a moot point whether there actually exists such a causative chain of events, if only because in an epidemiological sense it is impossible to determine how many children did contract a streptococcal infection but never presented OCD in their life story. Or, to phrase it differently: in theory all young children might have encountered some strep infection – and if, say, 2% of them later show OCD-like symptoms, there is no ground whatsoever to link the two states.
At any rate, OCD and the immune system is a separate topic that will be treated in more depth below.
Finally, the possibility may not be ruled out that two disorders nosologically demarcated from one another are linked: PTSD (post-traumatic stress disorder) and OCD. Brown et al (15) found that, when PTSD preceded the onset of OCD, lifetime PTSD and OCD show a statistically significant covariation; so the trauma that is at the basis of PTSD may also be the cause of later OCD symptoms. Lochner et al (2002) reported that physical and sexual abuse in childhood are linked to OCD and related disorders, such as trichotillomania. This latter result is interesting because it offers a possible way out of a classification problem: if phenomenologically similar types of behaviour show a marked difference in genetic linkage or association, they may on the other hand be close relatives because of a highly similar (series of) early traumatic event(s), and therefore still be ranked within the now-acknowledged ‘spectrum of OCD disorders’.
The same caveats are valid pertaining to interpretations of TLE (environmental) factors as they are in the genetics of OCD: limitations involve factors such as sample size, individual report exactness, interpretarive qualities of the researcher, and distinctive features of the control group (or even a lack of same).
The tentative upshot of the observations in the above must be that there are strong indications that TLEs, or stressful life events, or perhaps even a rapid succession of major life events in a relatively short time period, can predispose a person to develop symptoms of obsessive-compulsive disorder.