The P Value And Its Significance

n diversifying an investment portfolio, one needs to consider investing in the healthcare industries. Pharmaceutical companies that have strong pipelines of investigation drugs have strong growth speculations. In addition to examining its pipelines, one also needs to be able to evaluate the results of its clinical trials. When a news program reports the results of a clinical trial for a new drug to be "statistically significant," what does it mean? And how did they come up with that conclusion? One of the most important values to look for when interpreting the results of a study is the P value.

Elenbaas et al defines P value as "the probability that a type I error has been committed, and that the null hypothesis has been falsely rejected." Let's briefly talk about the null hypothesis, type I error, and type II error. Lets say a new drug A is being compared to a currently available drug B. The first assumption that investigators will make is that the new drug A is equivalent or is equal in efficacy to the currently available drug B. This assumption is called the null hypothesis (H0). A contradicting assumption to the null hypothesis is the alternative hypothesis (H1). H1 says that the new drug A is more effective than the currently available drug B. In fact, the alternative hypothesis (H1) is what the investigators want to prove because the pharmaceutical companies developing the new drugs fund most of its clinical trials. Type I error occurs when H0 is rejected when it is really true. In other words, investigators believe that the new drug A is more effective than the currently available drug B (rejecting H0), when drug A is equal in efficacy to drug B, or worst, not as effective as drug B. Type II error occurs when H1 is actually true (drug A is more effective than drug B), but the investigators failed to reject H0 (did not reject the hypothesis that drug A is equal in efficacy as drug B).

With the technical terminologies defined, lets go back to the P value. To reiterate, P value is "the probability that a type I error has been committed, and that the null hypothesis has been falsely rejected." (Elenbass et al). The P value is said to be "statistically significant" when it is < 0.05. For example, a clinical trial comparing the effectiveness between drug A and drug B, found that drug A worked better at lowering the blood pressure than drug B, and the P value was calculated to be 0.04. This means that there is a 4% chance that a type I error is being committed (investigators falsely believe that drug A is more effective than drug B, when they are equal in efficacy). 0.05 is the largest acceptable P value in clinical literature. When investigators claim that drug A is more effective than drug B, they were correct 95% of the time, with only less than 5% chance error.

There are many other important statistical values, but they are beyond the scope of this article. Now, whenever a reporter on television reports the results of a clinical trial for a new drug to be "statistically significant," one can perform their own verification by obtaining a copy of that study and look for the P value.

References:

1. Elenbaas RM, Elenbass JK, and Cuddy PG, et al. Evaluating the medical literature part II: statistical analysis. Annals of Emergency Medicine. October 1983;12(10):35-45.

2. Anderson HG, Kendrach MG, and Trice S, et al. Understanding statistical and clinical significance: hypothesis testing. Journal of pharmacy practice. June 1998;11(3):181-195.

http://www.onlinefreedomwork.com