One of the core beliefs of the antivaccine movement is that there is an “autism epidemic.” The observation that autism prevalence has been climbing for the last two to three decades led some parents with autistic children to look for a cause, specifically an environmental cause, for autism. Because several vaccines are given in the age range when children are typically diagnosed with autism, they fell victim to the all-too-human tendency to confuse correlation with causation and latch on to vaccines as the main cause of their child’s autism. Then, when these parents banded together, they naturally latched on to vaccines as The One True Cause of the Autism Epidemic because there was an expansion of the recommended vaccine schedule back in the 1990s that they viewed as coinciding with the increase in autism prevalence. Of course, it never occurred to them to blame other things that increased around the same time, such as Internet usage, cable TV viewing, and, my favorite, sales of organic food. That’s right, no one blames organic food for the “autism epidemic.”
Scientifically, the question has been whether this increase in autism prevalence is a true increase rooted in biology or whether it is an artifact of something else—or a little of both. For example, the way we diagnose diseases and conditions change with time. In the 1990s there was a broadening of the diagnostic criteria for autism spectrum disorders. That alone would be expected to increase apparent autism prevalence. Screening is another reason why the apparent prevalence of a condition or disease can increase. As I like to say, whenever you look harder for a medical condition, its incidence will increase. The example I like to use is ductal carcinoma in situ (DCIS), a premalignant condition in the breast whose incidence increased 16-fold since the 1970s. Now, there is no reason to think that this represents biology or that the incidence of DCIS has really increased that much in such a short period of time. We did, however, start to screen for breast cancer in the late 1970s and early 1980s, and a whole lot of preclinical DCIS.
Another example is hypertension. Before the 1920s doctors didn’t routinely measure systolic/diastolic blood pressure ratios; so there were few, if any, cases of hypertension because doctors weren’t looking for it and didn’t know how to easily measure it. Even over the last decade, prevalence of hypertension has increased (for instance, in Canada). Perhaps a better marker for hypertension diagnoses is the percentage of adults who have been prescribed antihypertensive medications, which has skyrocketed since 1950. Of course, over that time, the definition of what constituted “hypertension” has changed markedly, to lower and lower diastolic and systolic pressures, as evidence showed that treating lower blood pressures could prevent premature death, although that trend appears to have leveled off. Amusingly, various denialists lambaste such broadening of diagnostic criteria to claim that hypertension is an “engineered” epidemic.
Of course, there are other potential causes of the apparent increase in autism prevalence. One that I noted a long time ago is diagnostic substitution. In diagnostic substitution, children that would have received one diagnosis in the past now receive another. In the case of autism, diagnoses of mental retardation decreased at the same time diagnoses of autism spectrum disorders increased.
In other words, definitions of medical conditions matter. They can have a huge influence on prevalence rates observed. Moreover, if you don’t look for something, you won’t find it, and if you do look for something, you will find a lot more of it than you expected. It is quite possible that that is what’s happening with autism, but such a narrative doesn’t fit into the vaccine-autism idea. The antivaccine movement needs an increasing apparent autism prevalence. If there’s no “autism epidemic,” there’s no way that vaccines cause autism. Over the years, there has been quite a bit of evidence that the “true” prevalence of autism has been fairly stable for quite some time. For instance, one study examined autism prevalence using stable diagnostic criteria between 1990 and 2010 and found “no clear evidence of a change in prevalence for autistic disorder or other ASDs between 1990 and 2010. Worldwide, there was little regional variation in the prevalence of ASDs.” Steve Novella has summarized several of these studies.
Now there’s a new study out of Australia looking at the same criteria. It basically relies on one of my favorite sayings, namely that the harder you look for a medical condition the more of it you will find. Actually, it uses a corollary of that maxim, namely that, when you look for a condition and diagnose more of it (i.e., cause its apparent prevalence to increase), many of those cases will be milder cases. So, just as mammographic screening led to an increase in diagnosis of smaller tumors and premalignant DCIS, screening for prostate cancer led to milder cases of prostate cancer being diagnosed, and screening for hypertension led to milder cases of hypertension being diagnosed, if the increase in prevalence in autism is due to the broadening of the diagnostic criteria plus increased screening, we’d expect that the average severity of cases diagnosed would decline with time.
The authors had the advantage of a good tool to look at this question, a register of newly diagnosed cases of autism spectrum disorder (ASD) in Western Australia. Data are collected prospectively and include information on the “diagnostic criteria met by each new case, as well as the severity of behaviors relevant to each criterion.” They then examined all new cases that received the diagnosis of Autistic Disorder, diagnosed using DSM-IV-TR guidelines) from 2000 to 2006, a time period when the diagnostic criteria for Autistic Disorder was stable and selected for:
These years were selected for analysis because during this period: (1) diagnostic assessments were predominantly conducted through four Government centres, and there was a central administrative portal for archival reviews, which provided confidence in achieving near-complete ascertainment of new cases of Autistic Disorder; (2) diagnosticians were strongly encouraged to complete severity scores for each criterion, and a near complete data-set was available; and (3) numerous studies [Boyle et al., 2011; Maenner & Durkin, 2010], including those in Western Australia [Parner et al., 2011], have found that this period corresponded to a particular increase in the prevalence of Autistic Disorder.
The authors studied 1303 children who met the inclusion criteria for the study. 51 cases were excluded because complete data on the criteria met and/or severity ratings by the diagnosing clinician(s) were lacking, leaving 1252 cases which were analyzed thusly:
Analyses included investigations of: (a) the age and sex distribution of the sample across the study period; (b) change in percentage of cases deemed to meet diagnosis criteria (individually); (c) change in percentage of individuals deemed to be displaying extreme behaviors for diagnosis criteria (individually and aggregated), and (d) the number of criteria diagnosed as extreme (per case). Percentage change across calendar years within the study period was quantified using linear regression.
The answer was striking. There was a statistically significabnt reduction in the percentage of diagnoses meeting two of 12 criteria, as well as a reduction in the proportion of new cases having extreme severity in six criteria. Concurrently, the percentage of new cases with no “extreme” rating in any criteria increased from 58.5% to 86.6%.
Here’s a representative graph, this one showing the percentage (a) and count (b) of Autistic Disorder diagnoses in each calendar year who are rated as having “extreme” behavior on DSM-IV-TR criteria (click to embiggen):
This is pretty dramatic evidence. Basically, over a seven year period, among children diagnosed with autistic disorder, the severity of symptoms at diagnosis decreased considerably, and this happened during a period of time where there were no changes in diagnostic criteria to confound the analysis. The authors conclude:
The current findings may inform our understanding of why the prevalence of ASD diagnoses has increased markedly over the past 50 years from 0.05% in the 1960s [Lotter, 1966] to at least 1% in the 2010s [Elsabbagh et al., 2012]. In the current study, we found that the number of individuals with no diagnostic criteria rated as “extreme” in severity increased by an average of 21 cases per year from 2000 to 2006, whereas there was no statistically significant increase in the number of individuals with at least one diagnostic criterion rated as “extreme.” This observation indicates that the majority of the increase in Autistic Disorder diagnoses in Western Australia during this period [Parner et al., 2011] is likely due to an increase in the diagnoses of individuals with less severe behavioural phenotypes. A range of sociological factors may have contributed to changes over time in the behavioural profiles of individuals presenting for a diagnostic assessment, including greater awareness of ASD among clinicians and the community [Charman 2002; Leonard et al., 2010] and the introduction of Government funded services specific to individuals with a diagnosis of Autistic Disorder [Gurney et al., 2003; Leonard et al., 2010; Mandell & Palmer, 2005; Nassar et al., 2009].
One more time, if you look for something more intensely, you will find more of it, particularly if you broaden the criteria for what constitutes what you’re looking for. In the case of medical conditions, you will find a lot of more mild cases of what you are looking for.
Is this definitive evidence? On its own, of course it’s not. Taken in the context of existing evidence, however, it is yet another study that adds to the growing body of evidence that autism prevalence has been fairly stable for the last few decades. There is no “autism epidemic” (or, as antivaccinationists like to call it, “autism tsunami”), and vaccines still don’t cause autism.