1.1. Anorexia nervosa and season of birth

Nielsen (1992) has reported an excess of births in the\udperiod February–May of patients who later developed\udanorexia nervosa in Denmark, but only those\udborn during and after 1964. Rezaul, Persaud, Takei,\udand Treasure (1996), studying nearly 2000 undifferentiated\udeating disorder patients, reported a significant\udpeak season of birth occurring in May. When\udthey also divided their population according to\udwhether birth occurred before or during/after\ud1964, they found the peak season to be March for\udthose in the later born category. These authors\udreviewed the literature thoroughly and explored\udpossible explanations of their findings, ranging from\udphysical and biological to social. The social and psychological\udconsiderations included social class, religion,\udpresence of eating disorders in the parents\udinfluencing fertility on a seasonal basis, and parental\udneurotic disorders influencing and governing their\udrelationship including frequency and timing of their\udsexual relationship.\udMETHOD\udThe present report concerns a study of the first 980\udpatients with anorexia nervosa on the St George’s\uddatabase, examined both as a total population and\udalso in relation to the same historical watershed of\udbirth in 1964. As with the study reported by Rezaul\udet al. (1996) the data were compared with those on\udseason of births in the general population obtained\udfrom the UK Office for National Statistics (ONS).\udComparisons between these data sets were conducted\udusing SPSS statistical software and the Edwards test for seasonality (Edwards, 1961). In\udaddition, season of birth in the present population\udwas examined in relation to 16 parental background\udvariables also derived from the database. Thesewere\ud(1) Paternal and maternal neurosis defined as nonpsychotic\udpsychiatric disorders for which medical\udhelp had been sought. (2) Family relationships\udcategorised as impoverished/not impoverished,\udoverprotective/not overprotective, enmeshed/not\udenmeshed, conflict avoidant/not conflict avoidant,\udsexually conflictual/not sexually conflictual. (3) Presence/\udabsence of anorexia nervosa, obesity, alcoholism,\ud‘abnormal-normal weight control’ in the father\udand mother. (4) Weight consciousness/absence of\udweight consciousness in the mother.Within the database\udall these variables were scored on a four point\udscale. These ratings were subsequently categorised\udeither as 1 compared with 2, 3 and 4, or 1 and 2 compared\udwith 3 and 4 depending on distributions of\uddata. The term ‘abnormal-normal weight control’ was\udcoined by the first author in the 1970s (e.g. Crisp,\ud1981). It has survivedinthe database, and relates very\udclosely to what, these days, is called ‘bulimia nervosa’.\udIt refers to extreme patterns of defensive behaviour\udconcerned with maintaining body weight\udwithin normal boundswhen itwould otherwise have\udescalated due to excessive food intake, causing great\uddistress. Such patterns included excessive exercise or\udvomiting/purging, either immediately or shortly\udafter eating, especially binge eating.\u