Make your own free website on

Dealing with Statistics

Our journey into preemiedom began with our genetics counselor advising us that mom's blood test indicated that our child had a 1:13 chance of Downs Syndrome and a 1:6 chance of open neural tube defect. Based on the statistics, she made it known to us that we were brought in quickly so we would have the option to end the pregnancy. This page is included to help bring understanding that while statistics are expressed in numbers on paper, they always begin in real life and tend to have real life implications. NONE of Ashley's current complications are related to the problems this blood test identified.


Statistics related to Preemies

Premature Infant Information provided by University of Wisconsin and The Center For Perinatal Care at Meriter Hospital Madison, Wisconsin offers an excellent overview of preemie info.

Teen Pregnancy contributes to preemie population as about one million teens per year (in the US alone) become pregnant and 70% do not receive proper prenatal care. This lack of care and the incomplete physical development of the mother contribute to premature births according to Boston University Medical Center Another site (not as well documented as the Boston University site) for looking at aspects of how adolescent pregnancies influence premature birth outcomes can be found at\par.

UT Southwestern notes that only the smallest 3% are in the greatest danger rather than the smallest 10% as previously thought. This link takes you to the article, which covered 127,000 infants.

Survival Rate For Preemies Increases With Liquid Ventilator at Thomas Jefferson University Hospital [sample size is only 13 kids though].

Growth Standards can result in Faulty Decisions Cornell University nutritionists note that "millions of healthy infants may be assessed as growing too slowly because of inadequate international child growth standards."

Understanding Statistics

What to Look for in statistics; some general observations:

How many persons were measured? Statistics are numbers describing past events, but are often used to try to predict future events. Statistics can come reasonably close to describing a population - if enough persons are measured. For example, an estimated 423,107 babies are born prematurely each year according to the United States Center for Disease Control One can enter the above number into the Sample Size Calculator and find that the group being measured must be larger than 97 individuals to have meaning for all the babies born in one specific year in the US assuming the confidence interval is 3% (the +/- figure usually quoted in polls). If you are looking for statistics for preschool preemies, then the figure would be 423,107 times five years which would be 2,115,535 individuals requiring a sample size of 357 individuals to be applicable to the entire preschool preemie population. Look for sample sizes - many professional articles mention that the article only describes the measured individuals, [not preemies or children in general] but the summary [or abstract] of the article may not point this out.

Can I use small sample studies to make decisions? Take the study you have found to your doctor. A number of mathematical methods are used to compensate for small and/or non-random samples. See Interactive Statistical Calculation Pages at for an idea of how technical and complicated statistics can be.

Where did the statistics or article come from? The source you are receiving the information from should disclose the author, the journal, the year of publication, and the page number. As an alternative, a web site address should be available. The person writing the article should be using the same questions or measurements for all the participants. In many cases, sources such as newspapers or magazines are not intended to be taken as seriously as medical journals.

Can you physically locate the person writing the article? The web allows anyone to post anything. However, schools, universities, and school affiliated medical centers usually monitor what is posted on their sites, making the info more trustworthy. You can typically reach a real person or can verify addresses and phone numbers through outside internet search engines. See for a list of United States universities.

Statistics can usually be quoted at least two different directions. For example, do you describe a glass of water as half full or half empty? The level of the glass is the same regardless of how you express it, but the wording is completely different. Statistics are always written by someone else, but you are the one who must live with the decisions you make based on those statistics. Some decisions are more reversible than others are.

Listen and look for what is NOT being said. For example, some studies feel that micropreemies tie up a major portion of NICU resources (see ). Perhaps they do, but their care also contributes to the financial resources to help other infants. Using different wording, our infant allowed a number of professionals to work additional hours due to the required 1:1 or 2:1 nurse to child ratio. Her insurance payments have contributed to the well being of over 70 non-profit and for-profit entities in the local economy.

What controls were used? Children are very complicated, responding to an incredible array of factors. Studies should mention what factors were measured for (that is what medicines were used, what machinery, what methods, what complications were present, etc.). Be careful of what influence caused what response - good studies will usually say if there are things that may have influenced outcomes, but were not measured for in the current study.

In many cases statistics do not exist. Rapid technology changes in the last five years has allowed larger number of children to survive. Longitudinal studies measure over time and provide a better perspective than a snapshot of a particular day in time. For example, would you feel that a person measuring you on one particular day of your life would be more accurate than a summary of all the days of your life? Few, if any, micropreemies such as ours have survived to be measured for statistical studies in the past, therefore statistics literally do not exist to describe her currently.

Be careful to note when and where data was collected. In some cases, the article is written long after the data was collected. Treatment methods vary widely between countries and even between states and counties. Both time and location may change outcomes. For example, our child survived in part due to the availability of artificial surfactant and oscillating respirators that have been used widely in only the last five years.

Sometimes more than one test is needed to get an accurate reading. The first measurement is often referred to as a baseline. Sometimes more than one measurement is needed to get more accurate results. See Lab Values in Nutritional Assessment which cautions about who drew the lab, who performed the tests on the sample, was sample contaminated during transportation, etc.

Be careful of terms or labels such as optimal or normal. The definitions vary over time. We know that regardless of what the going definition, normal, or optimal is in fifteen years, our little preemie will still be our daughter who is greatly loved. See for more info about labels are for cans, not for kids.

Statistics ultimately encounter treatment ethics. See this well referenced University of Pennsylvania site for more insight:

See the University of Michigan's Evaluating Health Resources for additional information related to evaluating health resources.

Describing statistics in terms of a journey

Most journals print articles that have a confidence level of 95%. Confidence Level means how sure you are that the information has correctly measured the folks you studied. Most studies and polls have a Confidence Interval of 3% to be considered meaningful. The Confidence Interval is the +/- figure seen in most studies or polls. There is a direct relationship between how many persons were measured and how many people the study would describe. In general, the smaller the number of persons studied, the smaller the number of persons the study should be considered to describe. See for more information.

Using the example of driving a vehicle during my journey, the following would be roughly true:

When I am traveling in an area where the confidence level is 99%, I am on a one-lane road and my vehicle is very likely to pass over every portion of the road. However, most of the time I am driving in areas where the confidence level is 95%. This road has four more lanes of width, so I can only pass over one small section of the road - meaning I don't pass over most of the available surface of the road. With the additional lanes, I feel that I can maneuver around any obstacles that appear. When I am in territory where the confidence interval is 3%, the shoulders of the road is about three car widths wide. I can usually see any obstacles that are on the shoulder or fixing to intersect with my road. However, when the confidence interval grows to 40% (as the result of trying to apply a too small of a study to too large of a population) then the shoulders of the road are forty car widths wide. I can no longer tell where the shoulder stops. I have no way of knowing how level the shoulder is in the event I have to veer off the paved road. This type of territory is similar to driving in sugar cane - no one stalk is big enough to stop you by itself, but a batch of them together can sure slow you down and block your vision. With shoulders so wide, you can't easily see any rest stops, restaurants, or stores. In short, don't become rattled by studies that are not big enough to have any meaning for your child. Conversely, if a method worked for the children in the study, then you may want to ask your doctor about the study.

Some folks have better vehicles than others do. (Better vehicles would be good insurance, well-equipped medical centers, knowledgeable doctors). Some of us are driving a small compact that can't travel on unpaved surfaces - others are driving four wheel drives that are not hindered when our journey takes us onto the shoulder of the road. It makes a difference when someone travels with you and helps you navigate your way along your journey. It has been an interesting journey thus far - and we are still traveling on the shoulder. Maybe one of these days our little preemie will allow us steer back onto the pavement.

**** These are suggestions and are not to be construed as medical advice. ****

**** The above suggestions were prepared by a preemie parent with some stats training, not a statistician. ****

Medical Ethics

How Small Is Too Small? an excellent article by nurse Linda Bellig, first printed by the Journal of Christian Nursing (journal's link is also included with article).

Baby Dear This article is written by a preemie parent and addresses some of the ethics issues that confront parents of preemies.

Research sites offers a variety of informations about counseling and emotional issues. This links lands you on the differences between healthy and unhealthy shame.

Association for the Care of Children's Health offers an excellent collection of links for parents in general and special needs families in particular. presented by the American Society of Gene Therapy.

Boston University Medical Center Community Outreach Health Information System includes information about teen pregnancy, a key element which skew premature baby statistics.

Child Advocacy and Medical Links by Lisa K. Lands on a photo of a one pounder.

Child Neurology Home Page The main purpose of this site is to coordinate the available internet resources in Child Neurology, both for professionals and patients.

Comfort Connection family resource center for families with special needs children.

Disability Resources provided by the University of Texas at Austin offers a number of links related to disability and rehabilitation. Medical and Pediatric Encyclopedia offers a wealth of information about symptoms, diseases, tests, drugs, nutrition and other subjects.

Doctor's Guide to Internet offers a layperson and professional version of information about a broad variety health and medical topics.

Doctor Sears Online offers good tips on how to handle minor health problems of your baby.

Emergency Nursing World is aimed at nursing professionals, but helps parents understand nurses as well.

Excite Health Medical Encyclopedia offers an incredible number of links to major diseases / disorders.

Health Answers a good research site by the American Academy of Pediatrics. Includes drug database as well.

Health Communication Network of Australia. provides a large amount of information and links in both Spanish and English.

Health Information Resources presented by the University of Michigan. helps you locate medical and healthcare resources; provides a series of forums for healthcare discussions. offers a variety of health information,CHN|~st,3460|~r,WSCHN000|~b,*|

Internet Mental Health allows you to check side effects of medications.

Family Education Network offers good educational information related to parenting. Has an easy to read format similar to a magazine.

Frontiers in Bioscience Database offers a way to look up enzyme related issues, although you can click to go elsewhere.

Future of Children The primary purpose of The Future of Children is to disseminate timely information on major issues related to children's well-being, both negative and positive.

Mayo Clinic this link takes you to the pet therapy page, where you will have an article and index to jump to other good spots.

Med Engine offers an extensive variety of ways to look up information, associations, products and other resources.

Medical Health Encyclopedia by Excite offers numerous links to hundreds of diseases.

Medical Matrix offers a way to search for medical information at no cost. offers access to a number of journal articles. offers an extensive site of medical information in an well-organized presentation.

Medwatch the FDA's Medical Products Reporting Program including drugs, biologics, medical and radiation-emitting devices, and special nutritional products (e.g., medical foods, dietary supplements and infant formulas).

MedWebPlus offers a variety of search opportunities. offers good resources and links in an well-organized manner.

Mount Sinai Medical Library in Cleveland, Ohio The Lillian and Robert Ronder Resource Center for Education and the George H. Hays Medical Library offers a number of research links and information.

Neonatal Diseases and Abnormalities offers HUNDREDS of links to an EXTENSIVE list of diseases and abnormalities.

Neonatology on the Web provided by Ahmanson Pediatric Center at Cedars-Sinai Medical Center. Nice collection of articles, links, clinical information and more.

Net Psychology offers a way to look up info on mental health issues.

Neurosciences by the Mining Company has an excellent collection of search engines.

NICU-Web excellent comprehensive site provided by the University of Washington.

National Organization of Rare Disorders allows you to look up information about an incredible number of rare diseases.

OnLine Pediatric Surgery HANDBOOK for Residents and Medical Students An excellent site for surgery related information.

Parent Resource Links offers a number of additional informational links.

Parenting on the Net has over 3,000 links for parents to look up information.

Pediatric Points of Interest good information site provided by John Hopkins University.

Pediatric Neurosurgery, presented by a community practice office, offers info about shunts and other neurological topics.

Pediatric Neurosurgery of Columbia University gives a good overview of neurological conditions. This link lands you on craniostenosis or the premature fusion of skull bones.

Pediatric Points of Interest by John Hopkins University provides an EXTENSIVE list of pediatric resources.

Pharmaceutical Information Network provides a good overview of medicines and links to disease information centers.

Psych central offers a VERY EXTENSIVE listing of links for general emotional support topics and psychology topics.

Sleep Disrupted? Check out UCLA's sleep pages offers resources and information related to sleep.

Social Work and Social Service Web Sites provided by the George Warren Brown School of Social Work at Washington University offers links to literally hundreds of sites in a well organized presentation.

Social Work Search Site by Tom Cleereman offers a variety of resources for those seeking help and those providing help.

Stepfamilies inFormation gives a variety of links, statistics, and information for stepfamilies.

Texas Children's Hospital - Houston

ThriveOnline offers a wide variety of health related resources in a search engine format.

Values Parenting offers resources for instilling values in your children.

Voices of women journal and resource guide for women.

Women's Health Interactive offers a broad collection of resources for women. This link lands on the subject of infertility.



Lamb's Way offers a fair sized collection of links related to special children.

These links are provided as resources for your information, but should not replace medical advice or consultation. Links are left visible so you may more easily print or forward the page if you desire, but please give credit where due.

©2004 Greg & Christa Hall

home page

Contact us at