Laura does Medical Research

An hour later they were at home. By then Laura was now not having any significant pain. The Novocain and the pills had both kicked in and were working. Laura went to her bedroom as soon as they got home. She logged on her computer and brought up the hospital web page. She found a link to a state medical quality and statistics page. She went to the page and found a section that had statistics on sarcoma.

The information was appalling. There were three hospitals in town, one had a seventy eight percent survival rate, the second a seventy four percent survival rate. Laura looked at the numbers. The survival of one more person would have made the seventy four percent jump to more than seventy seven percent. Her math teacher had told them any time a change of a small number in the sample would so much impact the statistics; you could not make conclusions from the answers. There was probably no difference in the care of the two hospitals.

When she got to the third hospital she was stunned. The survival rate was forty seven percent, just over half of the rate of the other hospitals. There was no way that a change of a small number of cases would change the result. Clearly the other two hospitals were superior in treating this disease. How could this possibly be so? How could a hospital be so bad and the state not step in. She looked at the number of cases. There were thirty nine. Only eighteen survived. It was a bleak view, twenty one deaths. With the other two hospitals together, there were one hundred forty-three cases, of them one hundred nine survived. That meant thirty four died. She looked at the number of hospital admissions. The number of deaths was about the same compared to the admissions for all three hospitals, the number of survivors much higher in two of the hospitals. “Why the big difference?” she wondered.

She closed out the internet and stretched out on the bed. “What had Mr. Sprenkle said about selecting samples?” In their Math class he had given more than a few examples where statistics were way off. One of the reasons was because of a bad sample. Her head was fuzzy. She couldn’t remember them. If she could just think. The codeine was really making it hard for her to concentrate. She turned on the TV. She rarely watched it but right now she needed some diversion. She told her friends she didn’t watch it because TV stood for totally vacant. She flipped to the Discovery Channel. At least it and the History Channel had some value. They were talking about a plant study. The researcher noted that there was some concern about the accuracy of the testing of the sample. She was about half asleep when the comment was made. She was fully awake in seconds. What if the sample selection was flawed? But how could this sampling be flawed?