2011: Chasing Better Autism Data

Finnish data model offers new insight on autism in children

By Melanie Hundley

A major consideration in any research project or audit is selecting the data to analyze.  Once the data population is determined, various methods of sampling are used to spot trends and draw conclusions.  Deciding where to obtain the data is critical to the quality, trust, and peer acceptance of the results.

That is why a new report coming out in the August 2011 issue of the Journal of Autism and Developmental Disorders should prove very interesting for future studies of autism.  Lampi, Sourander, Brown, et al's excitement about the Finnish Prenatal Study of Autism and Autism Spectrum Disorders (FIPS-A) appears justified.  Having a comprehensive collection of medical data together with blood samples sounds wonderful for any such research. 


Lampi and the other researchers are affiliated with the University of Turku, Columbia University, or the National Institute of Health and Welfare in Finland.  I will reference the team as Lampi et al in this article for brevity.


What is the FIPS-A?
According to their report, FIPS-A is a data rich collection of medical-related information on patients they believe represent Autism Spectrum Disorders (ASD). The collection or study has been derived from cross-linking various national registers focused on Autism Spectrum Disorders (ADSs), the Finnish Medical Birth Register (FMBR), and the Finnish Maternity Cohort (FMC). The FMC, for example, is described by Lampi et al as containing "virtually all births in Finland from 1983 to the present and...includes archived maternal serum [blood] samples."


An extra benefit of deriving FIPS-A from these sources is the child patient content. The researchers say the data contains a wealth of developmental information from newborn to age seven. Based on this comprehensiveness, some may predict FIPS-A will provide a refreshing and sound basis for drawing new conclusions about ASD.


Also disclosed in the report, FIPS-A was supported by a grant from the Autism Speaks organization.  From a 2007 grant, Autism Speaks references the FIPS-A on their website.


How Might FIPS-A Help?
The quality of research results is, among other things, a reflection of the quality, accuracy, and comprehensiveness of the data. The logical steps used in the scientific method to prove results and conclusions must stand up to juried scrutiny and tests of reason.  For experimental sciences, lab results must prove reproducible. 


I selected this article for review because I found it interesting, and at times, I have sensed something was lacking in other published ASD-related research. Autism and ASD are emotionally charged topics.  A few research articles appear to have a difficult time convincing their results. What causes legitimate research to fail to be accepted in some professional circles?  I suggest that some research can be hampered by many of the issues and needed improvements mentioned by Lampi and team.  Their report describes how a "new" research data model could best be designed.  Those improvements may also increase confidence in final results from the scientific and academic communities.


As readers walk through the overview and design of FIPS-A, they should notice how the material works to build the case for validating the quality, accuracy, and comprehensiveness of their selected data set.  It is ultimately up to the reviewer to determine how successful the authors have been. However, I submit that the FIPS-A piece provides some interesting insight to the "secret" language and building blocks of academic and technical research.  Without these types of details, some prior published research appears to fall short of establishing credibility.



Here Lampi et al cite prior studies that have reported a "substantial increase in the prevalence of ASD."  And they stand on the shoulders of prior researchers when they list several possible reasons for the increase, including:

• Changes in how ASD is diagnosed
• Changes in how ASD is observed, and/or
• Differences in the methods used in various studies.

Lampi's team also mentions that environmental factors or other reasons have not been ruled out.  As more researchers clamor to link adverse events with ASD causation, there seem to have been weaknesses in their scientific arguments which the Lampi team recognize as problematic, including:

• Missing maternal biomarkers
• Failure to track kids long enough into ages where ASD tends to emerge
• Enough predictive data
• Sound and well-controlled reasoning lacking bias, and 
• Simply too small of sample sizes.

Without overcoming these weaknesses, it seems unlikely the larger medical and scientific community would fully embrace emerging findings in some ASD research.  Lampi et al offer up the FIPS-A study as a larger, detailed population of data that includes serum (blood) sampling to further prove any results.


Some of the Lampi team have already published articles supporting "the validity of Finnish-based diagnoses of childhood autism."  The meticulous documentation process of autism diagnosis in Finland is explained as are details about the "controls" of the study (those children compared without ASD diagnoses).



Types of Data in the Finnish Medical Birth Register
A medical detective trying to establish links between mother and child health events would likely be pleased to see the types of data available in the FMBR.  Here is a portion of the variables table listed in the Lampi team report:

• Maternal characteristics, e.g., maternal age
• Reproductive history, e.g., number of previous births including stillbirths, neonatal deaths, etc.
• Health-related behaviours, e.g., maternal smoking, maternal weight/BMI
• Complications during delivery, e.g., pre-eclampsia, maternal hypertension, maternal proteinuria, Rh incompatibility, placenta previa, gestational diabetes, excessive bleeding, and
• Baby health, e.g., birth weight, how long baby was carried, head circumference, premature rupture of membranes, prolonged delivery, anaesthesia during delivery, fever before delivery, pH of umbilical cord.

Efforts to identify and track congenital defects for a period of time later into the newborn's life are also discussed including maintaining identity information on file with hospitals.


Blood Speaks
According to Lampi and team, the FIPS-A includes physical data, specifically, blood samples drawn in the "first trimester of pregnancy" with the consent of the mothers for screening. I was amazed when I read that Finland maintains a blood (serum) bank with archived vials of maternal blood "on 99% of all live births." These vials of blood are cataloged and stockpiled in freezers for further study.


The abundance of data points in research is statistically important in proving results. Accordingly, the Lampi team reports that in 2010, there were over 1.6 million samples drawn from about 810,000 mothers which they say is about 98% "of all pregnancies in the country."  Testing the blood, the FIPS-A studies can dig into everything from toxins to hormones to immunity functions.  Lampi et al advise that the validity of the blood samples has been exhaustively established in prior publications.


Additional data has been collected through the Ministry of Social Affairs and Health, collaborators on the Lampki and team study.  They report that Finnish public health clinics are involved in the first seven years of Finnish children's lives.  They assert that their children are seen in those clinics once a month in their first year and annually thereafter until age 7. These visits provides the opportunity for monitoring "physical, cognitive, and social development as well as implementing the national vaccination programme."


While on the topic of public health and Lampi and team's mention of vaccinations, I also tried to identify possible vaccinations researchers might expect to find in Finland, I found the following list of vaccines offered in the national vaccination program on the Hospital District of Helskini and Uusimaa website and dated October 2010:

• Tuberculosis (the vaccination is only recommended for children with an increased risk of tuberculosis)
• Pertussis
• Tetanus
• Diphtheria
• Polio
• Diseases caused by the bacteria Hemophilus influenzae (e.g. meningitis)
• Measles
• Mumps
• Rubella
• Tick-borne encephalitis (TBE; in the Aland area)
• Hepatitis A (for risk groups)
• Hepatitis B (for risk groups)
• Influenza vaccinations for all children aged 6-35 months and for risk groups in autumn
• Rotavirus vaccine
• Pneumococcal vaccine.

The hospital district's website makes additional comments about chicken pox. [Note: advance searches for "thimerosal" on the linked National Institute for Health and Welfare in Finland did not return any detailed information.  The manufacturers or ingredients of the recommended vaccines was not found in this search.  It may exist elsewhere for the Finns.  In the U.S., the CDC has responded to inquiries about some vaccine ingredients.  For more related research and information, see also my review article New Research: Autism and Vaccines.]

Early FIPS-A
Lampi and team have already done some slicing and dicing of the overall Finnish data. They report that Finland has had about 60,000 annual births in the 1987 to 2005 timeframe. Working from over a million offspring, they down selected to just over 5000 ASD cases based on the Finnish medical codes associated with their ASD diagnosis methods.  More details of the efficacy of Finnish ASD diagnoses are discussed in the article.


Details of the diagnoses among the patients they selected for FIPS-A included:

• Childhood autism (about 22%)
• Atypical autism (about 4%)
• Rett's syndrome (less than 1%)
• Asperger's syndrome (about 36%), 
• Other pervasive developmental disorders (14%), and
• Pervasive developmental disorders (about 22%), unspecified.

Overall, the ASD-related patients consisted of about 4000 males and about 1000 females.


Some Limitations
Lampi et al include some of FIPS-A's limitations as normal research disclosure.  For example, they relied on the accuracy of the prior reporting as verification for the diagnoses of ASD and give their justification.  Also, the blood samples were not collected along other intervals of the pregnancies; however, they rely on prior research that indicates the first trimester is a reasonable timing for ASD-related analysis.


Conclusion
People impacted by ASD will likely hear more from FIPS-A as more analysis can be completed.  This article seems to have laid the groundwork for the FIPS-A model.  Researchers digging for better data or more successful acceptance may benefit from reviewing the overview and design of this study.  All in all, I join Lampi et al in hopes that FIPS-A will help identify changes that may need to be considered in:

• hormonal deficiencies
• exposures to environmental toxins
• vaccinations
• hygiene, and
• infection prevention.

They believe that the FIPS-A study is also strong contender for further research in onset and early "indicators of ASD."


A special thanks to Katja M. Lampi, Clinical Psychologist, at the University of Turku for providing additional insight in these research efforts.


Disclaimer: This is an editorial review only, written by a non-medical author, and is subject to change or update.  The information and links contained in this article are for educational purposes only to support further inquiry and should not be used for diagnosis or to guide treatment. Any reader who is concerned about his or her health should contact a licensed medical doctor for advice.  While every effort is made to ensure accuracy, readers should always refer to the original cited sources for verification and/or any interpretation of source material. Author assumes no liability for content, errors, and/or omissions.

Sources

• Airoldi, J., Weinstein, L. (2007). Clinical significance of proteinuria in pregnancy. Obstretrical & Gynecological Survey. Retrieved August 2, 2011, from PubMed & National Institutes of Health online database.
• Autism Speaks. (2011). Developmental Risk Factors for Autism in a National Birth Cohort.  Retrieved August, 2, 2011, from organization website.
• Central Intelligency Agency (U.S.). (2011). Finland. The World Factbook. Retrieved August 2, 2011, from government website.
• Columbia University. (2011). Columbia University. Retrieved August 2, 2011, from education website. 
• Hospital District of Helskini and Uusimaa. (2011, October). Vaccine information for parents of a newborn baby. Retrieved August, 2, 2011, from corporate website. 
• Lampi, K. M., Banerjee, P., Gissler, M., Hinkka-Yli-Salomäki, S., Huttunen, J., Kulmala, U., & ... Sourander, A. (2011). Finnish Prenatal Study of Autism and Autism Spectrum Disorders (FIPS-A): Overview and Design. Journal of Autism & Developmental Disorders, 41(8), 1090-1096. Retrieved July 29, 2011, from EBSCOhost online database.  DOI:10.1007/s10803-010-1132-6.
• MedicineNet.com. (2011, April 27). Definition of Anesthesia. Retrieved August, 2, 2011, from corporate website. 
• MedicineNet.com. (2011, April 27). Definition of Serum. Retrieved August, 2, 2011, from corporate website. 
• MedicineNet.com. (2003, September 4). Definition of Thimerosal. Retrieved August, 2, 2011, from corporate website. 
• Medscape Reference. (2011, March 29). Hypertension and Pregnancy. Retrieved August 2, 2011, from corporate website.
• Merriam-Webster Online Dictionary. (2011). Bias. Retrieved August 2, 2011, from corporate website. 
• Merriam-Webster Online Dictionary. (2011). Controls. Retrieved August 2, 2011, from corporate website. 
• Merriam-Webster Online Dictionary. (2011). Data. Retrieved August 2, 2011, from corporate website. 
• Merriam-Webster Online Dictionary. (2011). Maternal. Retrieved August 2, 2011, from corporate website.
• Merriam-Webster Online Dictionary. (2011). Scientific Method. Retrieved August 2, 2011, from corporate website. 
  
• National Center for Biotechnology Information (U.S.). (2010, April 26). Autism. Retrieved August 2, 2011, from government website. 
• National Center for Biotechnology Information (U.S.). (2010, April 26). Finnish Prenatal Study of Autism and Autism Spectrum Disorders (FIPS-A): Overview and Design. Retrieved August 2, 2011, from government website.  Retrieved August 2, 2011, from PubMed online government database.
• National Center for Biotechnology Information (U.S.). (2010, September 11). Gestational Diabetes. Retrieved August 2, 2011, from government website. 
• National Center for Biotechnology Information (U.S.). (2009, September 12). Placenta Previa. Retrieved August 2, 2011, from government website. 
• National Center for Biotechnology Information (U.S.). (2009, September 2). Preeclampsia. Retrieved August 2, 2011, from government website. 
• National Center for Biotechnology Information (U.S.). (2011, February 10). Rh incompatibility. Retrieved August 2, 2011, from government website.
• National Institute of Environmental Health Sciences (U.S.). Biomarkers. Retrieved August 2, 2011, from government website.
• National Institute For Health and Welfare (Finland). (2010, October). Finnish Medical Birth Record: Quality of the OSF Publication Series. Retrieved August 2, 2011, from government website.
• National Institute For Health and Welfare (Finland). (2011). National Institute For Health and Welfare. Retrieved August 2, 2011, from government website.
• University of Turku. (2011). University of Turku. Retrieved August 2, 2011, from education website. 
• U.S. National Institutes of Health. (2011). Autism Spectrum Disorders. Retrieved August 2, 2011, from government website.

Article Copyright 2011 by Melanie Hundley, U.S.A.  All Rights Reserved. No claim is made to other copyright materials cited.

Photo credit: Tina Phillips.

For more information about Autism, consider: