Autism epidemic in Sri Lanka?

8 May

Well, that’s what you might read if/when some other blogs see this new study:Screening of 18-24-Month-Old Children for Autism in a Semi-Urban Community in Sri Lanka. Soon to come out in the Journal of Tropical Pediatrics.

Take a look at the abstract:

All children aged 18-24 months in a defined geographical area were initially screened for autism, using ‘Red Flag’ criteria. All the children with one or more positive ‘Red Flag’ signs were further screened using Modified Checklist for Autism in Toddlers (M-CHAT) translated to Sinhala, followed by a comprehensive clinical assessment. Of a sample of 374 children, ‘Red Flag’ signs were positive in 28 (7.4%). Four children received a diagnosis of autism on clinical assessment giving a prevalence of 1.07% or 1 per 93 in the 18-24-month age group. Sensitivity of M-CHAT was only 25%, and specificity 70%. The high prevalence detected strongly justifies early community-based screening, but a culturally sensitive screening tool needs to be developed for Sri Lanka.

Let me start by congratulating the researchers. We need a lot more information about autism around the world. So far, most of the data is from the US, Canada and Western Europe. We need to know more about autism in other countries, and, more importantly, they need to know more about what is happening in their own countries.

The idea that a “culturally sensitive screening tool needs to be developed” is one that I would like to see explored. The IACC Strategic Plan had initiatives which were directed at screening diverse populations.

I find it interesting that they worked with children so young (18-24 months old). It will be interesting to see how stable those diagnoses are over time as well as if they missed anyone.

The autism prevalence is about 1.07%. One reason I decided to blog this is because it fits with a prediction made by Joseph over 2 years ago in his blog post Moving Toward a New Consensus Prevalence of 1% or Higher.

We do, indeed, appear to be moving towards a consensus of about 1% (or somewhat higher) for autism prevalence. It’s quite interesting to see, and kudos to Joseph for pointing this out 2 years ago.

9 Responses to “Autism epidemic in Sri Lanka?”

  1. Sullivan May 8, 2009 at 20:20 #

    I will point out that I need to get the actual paper, not just the abstract. 1.07% for autism rather than for ASD’s would be very high.

  2. Sullivan May 8, 2009 at 20:23 #

    Also, I count 18 “vaccines” (using the Generation Rescue counting method), and no MMR.

    So, high autism prevalence, low number of childhood vaccines, no MMR.

    I guess this paper won’t end up in future versions of the GR “study”.

  3. Joseph May 8, 2009 at 23:46 #

    I think the reason they ended up with a prevalence of 1.07% is that the group they screened had only 374 children. They probably took the time to look at every single child in some detail. The same is not true of autism screenings involving huge populations.

    The autism prevalence is about 1.07%. One reason I decided to blog this is because it fits with a prediction made by Joseph over 2 years ago in his blog post Moving Toward a New Consensus Prevalence of 1% or Higher.

    Thanks for mentioning that. You will note that one of the studies I mentioned then, Kadesjö et al (1999), found a prevalence of 1.21% among 7 year olds. This was a study done in one small town in Sweden, and the primary author personally evaluated 50% of the children in the town. It’s difficult to miss children that way.

    These findings also contradict ideas to the effect that more industrialized and heavily populated regions have a higher real prevalence of autism.

  4. Matt May 11, 2009 at 03:47 #

    Joseph makes an excellent point. The same basic point was made very well by Honda et al in their 2006 study. A quote from the study follows:

    “Accuracy is defined as the exactness of a measurement to the true value, and is influenced by systemic error. In epidemiological measurements of frequency, accuracy is difficult to measure quantitatively. A study can achieve a high degree of accuracy only by clearly indicating and implementing improved methods to reduce the inevitable undetected cases in the initial screening. Precision is the degree to which random error is reduced. Epidemiological measurements of frequency usually estimate precision by calculating a 95% confidence interval (“CI”); the larger the cohort, the smaller the value. However, no matter how small the 95% CI value is made, it is accuracy, rather than precision, that ensures a true value in that range. Some recent studies of the frequency of childhood autism seem to have put too much emphasis on precision, because the range of a 95% CI is easily calculated by adjusting the number of cases and size of the selected cohort. Some use registry data of cases receiving service at special facilities for developmental disorders in the community; this type of case-finding method produces a precise study because the cohort can easily be expanded in size. However, because the process by which cases were identified before the entry of registry data and the degree to which screeners were familiar with early detection of childhood autism in the first screenings are unclear, the possibility of undetected cases among children not listed in the registry data is not addressed. Consequently, the accuracy of these findings remains suspect. A narrower CI will lead to incorrect results in the absence of accurate screening procedures. Among 22 studies on the frequency of childhood autism published from 1996 to 2003, only four discuss measures for developing an enhanced methodology to avoid undetected cases in the screening (Honda, 1996, Arvidsson, 1997, Kadesjo, 1999, Baird, 2000). These studies yielded the most accurate case findings, i.e. with a minimum of undetected cases, but the limitation of precision remained because of the small cohort size. Thus, none of the previous studies has succeeded in conducting both an accurate and precise study.”

    What we are finding, as the studies that are being done become more accurate (as they stop using convenient and precise but inaccurate case registry methods and instead focus on improved case finding), is that measured autism prevalence is coming closer to its true level.

    I don’t know if people have seen the work by Baron-Cohen (http://www3.interscience.wiley.com/journal/118727214/abstract?CRETRY=1&SRETRY=0), but in one study the ASSQ questionaire was completed by parents and/or teachers to just under 10,000 kids. 2.1 – 2.7% were considered to be high scorers. This figure included kids all across the spectrum, includng kids who are not conventionally diagnosable. This would seem to be a very upper limit for the spectrum from chilhood autism up to the broad autism phenotype.

  5. Joseph May 11, 2009 at 14:01 #

    I don’t know if people have seen the work by Baron-Cohen (http://www3.interscience.wiley.com/journal/118727214/abstract?CRETRY=1&SRETRY=0), but in one study the ASSQ questionaire was completed by parents and/or teachers to just under 10,000 kids. 2.1 – 2.7% were considered to be high scorers. This figure included kids all across the spectrum, includng kids who are not conventionally diagnosable. This would seem to be a very upper limit for the spectrum from chilhood autism up to the broad autism phenotype.

    That URL doesn’t work if you open it directly. But I think you’re talking about Posserud et al. (2006) and Posserud et al. (2009). The results are similar anyway. Based on the second study, I estimated the prevalence of autism in Norway as 90 in 10,000 (in this comment). Incidentally, Norway is one of the countries JB Handley claims has 1/10th of the US autism rate.

  6. Mrs.Jamie October 4, 2011 at 06:54 #

    This’s Jamie, i would like to get more info on the number of autistic children in Sri Lanka. My friend is a qualified instructor therapist from abroad who is currently in SL. She would like to setup a school for the autistic children. We can we get info? contact person.

    thnx
    Jamie

    • miriam June 27, 2014 at 05:46 #

      Mrs. Jamie, There are about more than 10 autisitc blind childre n who attend the blind school at Ratmalana. have you started a school for autistic children in sri lanka. if you have done it it’s a great thing. please let me know. Thank you.
      God bless you.

  7. Ligaya Basalo November 3, 2011 at 18:11 #

    @Mrs. Jammie,I am also doing a research right now about the prevalence of Autism in Sri Lanka. Different sources says different figure. Did you get any lead?

    Thanks
    Ligaya

  8. miriam July 25, 2013 at 04:53 #

    we have a nephew who is autistic and blind. no help from Sri Lanka. we take him to the blind school at Ratmalana. There is only one teacher who has trained for this purpose but she has to cater for about 15 children.

    And the saddest thing is the blind school is going to close this unit recently and give the children to another place where there are no specialized personnel to do this charity. our child is 7 years old now. We were waiting until we got resisdential facilities and some more specialized teachers in the blind school. but what happened to us is very pathetic.

    This child’s mother left him when he was two months old. the child was born at Durdens hospital and they did not send the child to an eye specialist after keeping him in the NICU since he was premature(6 months and 10 days). In government hospitals this kind of children are sent it seems straight from the NICU to the eye clinic.

    we were not able to file a case against the Durdens hospital since we are in a mess with the child. The child is fully blind. Dr. Charith Fonseka is having an eye clinic in Durdens at that time too. we never knew that premature children after keeping them in NICU become blind if they were not taken to an eye specialist within two weeks. when we took the child to Dr. Charith Fonseka he told it was too late to restore child’s sight.

    without this child’s mother we look after this child with many difficulties. we have tried so much to get a place to train this child.

    now we are at the edge of dying with the child since we are nearly in our fifties and one sister of mine is mentally down. The father of the child is not co-operative to look after the child. he spends for the child.professor Hemamali diagnosed him autistic when he was 3 years old. We took the child to ‘Daddy’s Lanka” . we were told many times by doctors to take this child to a country where there are so many facilities and specialized persons to train him. we couldnt find anybody who can sponsor for this child and the child is growing and it is too difficult to look after him by women. He is not able to do anything on his own or doesn’t communicate at all.

    I don’t know whether you can help or even this is the correct place to make our appeal. please see whether you can do something to save our lives.

    Thank you.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: