(c)

From the data you presented in part (a), explain whether there is any evidence for a change in autism prevalence rates in children aged 8 years during the years 2004 to 2009.

Table 1
Year	Number of boys in the study population	Number of cases	Prevalence	Number of new cases in this year	Incidence
2004	145483	521	48%	172	12%
2005	143721	535	49%	170	12%
2006	147049	568	52%	190	13%
2007	142229	540	50%	173	12%
2008	138847	543	51%	170	12%
2009	138317	566	54%	180	13%

Table 2
Year	Number of girls in the study population	Number of cases	Prevalence	Number of new cases in this year	Incidence
2004	136752	109	10%	27	2%
2005	135511	112	10%	30	2%
2006	138548	112	11%	34	2%
2007	134083	125	12%	41	3%
2008	130876	107	10%	29	2%
2009	130367	106	10%	30	2%

this is the completed table from part a)

Within the 6 year period the rate of prevalence increased among the 8 year old boy data set ? This exercise has demonstrated the difficulty of interpreting prevalence estimates: criteria and methods for autism ascertainment must always be taken into account. Remember too, that a genuine increase in prevalence, as shown from 1979 to 2006, does not necessarily mean an increase in the number of people with autism. Current evidence points to increased autism awareness, professional training for autism diagnosis, and changes in the diagnostic criteria playing a major role. Countries such as the US and UK have more public awareness of autism and relatively well-developed diagnostic services. The figures of 1.1% and almost 1% for US and the UK probably do show that at least 1 in 100 children are affected by autism in these countries. Indeed, a recent US prevalence report (Baio et al., 2018) estimates the number of 8-year-olds meeting diagnostic criteria considerably higher at 1.68%. The 2011 figure for South Korea is higher still. One of the co-researchers, the anthropologist Roy Richard Grinker (1961-), had some years earlier documented high levels of autism stigma in South Korean families (Grinker, 2008) - so much so that families would opt for their autistic child to have a different diagnosis of 'Reactive Attachment Disorder', which implicates mothers' cold behaviour in causing their children's social anxiety, removing what would otherwise be a 'genetic taint' from the family. Consequently, as part of the 2011 study, Grinker conducted a campaign of public awareness raising and stigma reduction: Parent and teacher focus groups were conducted to identify local beliefs that might undermine or deter symptom reporting and to address stigma and misunderstandings related to ASC. Kim et al., (2011) It may be that these steps by Grinker to address cultural bias have led to over-identification and thus inflated prevalence estimates. 1.4.5 Autism in girls and women You may well recall from the quiz at the beginning of this week, that autism is diagnosed far more often in boys than in girls. The ratio often quoted, especially from older sources, is 4 : 1 boys to girls. This ratio may rise as high as 9 : 1 if only 'high-functioning' individuals are considered (Mandy et al., 2012). Does this mean that boys are more at risk of autism than girls? The discussion of prevalence differences in the previous section should have alerted you to the pitfalls of inferring differing levels of risk of autism in boys and girls from different prevalence estimates. What other factors might account for the fact that more boys are diagnosed? Hide answer It could be that current diagnostic criteria are more readily applied to male behaviour; it could also be that autism in girls is less overtly expressed. There is evidence that both these points are correct. William Mandy, an expert on autism in females, has highlighted a different autism phenotype in girls (Mandy et al., 2012). An increasing number of personal accounts from parents, and from autistic girls themselves, chime with this finding, suggesting that autistic girls may be less obviously 'different' and may indeed strive to actively conceal their autism. For instance, the parent quoted in Section 1.3.4 said of her 13-year-old daughter: She had developed coping and masking strategies to disguise her sense of difference and to fit in with her peer group. N (2017, personal communication) The autistic writer Liana Holliday Willey said: My mother tells me I was very good at capturing the essence and persona of people. At times I literally copied someone's looks and their actions. I was uncanny in my ability to copy accents, vocal inflections, facial expressions, hand movements, gaits and tiny gestures. It was as if I became the person I was emulating. Willey (1999, p. 22) Yet, under-diagnosis is unlikely to account for all of the male-female discrepancy. Loomes et al. (2017) conducted the first systematic review and meta-analysis (see Research methods box) of male/female prevalence differences. Their finding of an overall ratio of 3 : 1 suggests that both a diagnostic bias and different biological susceptibility play a role in the ratio. We will return to this focus on females in Week 3. Research methods: systematic reviews and meta-analysis A systematic review is a research method involving evaluation of the existing evidence for a particular topic or theory by critical review of all relevant findings by other researchers. A meta-analysis is a research method involving evaluation of the existing evidence for a particular topic or theory by pooling and statistically analysing the pooled findings from all relevant research by other researchers. The aim is to establish a coherent overall picture from individual findings which may appear contradictory