Statistics 101 for Infertility

Statistics to be valid are based on assumptions. If those assumptions are violated, then your conclusions are not true. One of those assumptions dictates that data should be collected at random and sampling should be planned in a way to avoid unnecessary bias. Later in our discussion it will become more apparent to you how the success rate numbers are collected. This cannot be accomplished in a single facility offering infertility services. Patient's age, infertility factor(s), spouse's age and specific infertility factor(s) for each partner make it very difficult to find a homogeneous population of couples sharing common infertility traits. Therefore, proper randomization cannot be accomplished and the sampling process may be invalid.

Providing that data was collected according to the rules, you can make a correct assumption on a population based on a small sample of that population. You can extrapolate from a small population to a larger one. That is how those polls that you have seen on TV or the newspaper are conducted. Then, the information collected is utilized to make inferences about the general population of interest. For example, a newspaper column says that based on a survey of 11O people, 25% of the population in St. Joe County have high blood pressure and risk of heart failure. However, we cannot conclude that Mr. Jones, a resident of St. Joe County has a 25% chance of being at risk of heart failure and high blood pressure. That is a wrong way of utilizing information correctly obtained. Many more factors may increase or decrease the likelihood of Mr. Jones having heart problems. Likewise, because you are, an individual and NOT a population you cannot assume that because a clinic advertises that 50% of the patients going through IVF get pregnant you are increasing your chances of becoming pregnant if you chose that clinic. In fact your probability could be totally different if that figure was obtained from a population that does not share any of your characteristics. It is not always true that the center with the best statistics provides the best health care.

Statistically sound conclusions are derived from UNBIASED estimates. Many, many people in the infertility arena do not have proper training in the statistical sciences. Thus, they are making honest mistakes when speaking about their data. Others simply do not care about being honest and offer figures good enough to attract more patients. Either way, both are biased.