You are planning a study on a new type of cancer treatment and you need to create a grant proposal. In your study, cancer patients will be randomly assigned to either of two treatment groups. The standard treatment is surgery and the new treatment is surgery plus a new form of chemotherapy. The outcome is recurrence of the cancer as determined by diagnostic tests (i.e., "survival" is that they remain cancer-free). Thus, the appropriate analytic technique to use is survival analysis.
Patients will be accrued or entered into the study at a constant rate. The hazard rate or risk of dying is constant. The attrition or drop out rate is also constant. You select "Subjects entered during study, at one rate" from the Accrual menu; "Hazard rate is constant" from the Hazard rates menu; and "Attrition rate is constant" from the Attrition menu.
The initial interactive screen shows the default values for your selections.
Your study will run for a total of six years. You can enter this in years or months. Here, you choose to enter it in years. You select "Customize screen" and set "Year" as the Interval Label and "Patients" as the Subject Label.
You plan to enter patients into the study for one year and then follow them for five more years. You choose "1" for the Accrual Period and "6" for the Total Duration.
Now you want to know what the survival rates will be for each of the two groups at the end of the study. By default the program expects you to enter the hazard rate. Instead, you choose to enter the cumulative survival rate at a specific time-point. In this case, you want to know the survival rate at end of the study, so you choose "At final interval."
For patients given the standard treatment, you anticipate a six-year survival rate of 50 percent. The new treatment causes serious side effects, so you feel that it will be worthwhile only if it increases the six-year survival rate by to 70 percent. You also anticipate that one percent of patients per year will stop returning for follow-up exams. You set alpha at the .05 research standard. The test is two-tailed.
You simply click on "Find N" to find the sample size required to yield the minimum acceptable power of 80 percent. SamplePower shows that a sample of 101 (patients per group for one year) is required.
Now that you have the sample size, you can prepare a report to include in your grant proposal. Simply click the "Report" button on the toolbar and SamplePower produces a text report of the power analysis findings in a format that can be easily included in your proposal. You can also use SamplePower's Graph and Table tools to analyze and presents your findings. Graphics created with any of these tools can be inserted into programs such as Microsoft Word® or Microsoft PowerPoint®.
Using SamplePower's presentation tools, you have included a detailed research report, complete with visuals, in your grant proposal.
Predictive Analytics
can make your organization
more
successful
Key Products and Services
Resources
Buy
Demos and Downloads
