|
The user interface
Once loaded onto the user’s browser, SSP is ready for parameter values to be
entered. There are 12 number fields to be entered, and all must be filled in before
calculation can begin. Positioning the arrow with the mouse over the first number field
and then left-clicking anywhere inside that field will place the cursor inside. The user
can then enter the appropriate value and tab over to the next field to resume parameter
entry. Next to each field is a help button which can be pressed any time by left-clicking
on it with the mouse, or by tabbing over until the appropriate button is highlighted and
hitting the "Enter" key on the keyboard. Brief instructions will appear just
below the "Calculate" and "Reset Values" buttons. At the bottom of SSP
is the number field where the calculation is output.
The methods and assumptions for SSP
SSP calculates power and sample size for studies involving two groups with a
survival outcome. It is assumed that the patients will have an exponential survival
distribution. Each patient is followed until either the event occurs in that person, the
person is lost to follow-up or administratively censored, or the study is terminated. It
allows for the user to specify the distribution of patient recruitment, whether they enter
the study uniformly or non-uniformly. Specifically, non-uniform entry follows a truncated
exponential distribution. The user can also specify the loss to follow-up rates for each
of the two groups, assuming the loss rates are exponentially distributed. Further, the
user can specify the rates at which patients cross over to the other group. The methods of
Lachin & Foulkes (1986) are implemented for the calculations.
The parameters to calculate sample size or power
Sample size in the control and treated groups. Any number greater than zero is
allowed, as well as unequal sample sizes.
Power. The probability of rejecting the null hypothesis given the
alternative hypothesis is true. It is a value between 0 and 1, though common values
include 0.80 and 0.90.
Ratio of sample size, control:treated. The ratio of the sample size in the
control group to the sample size in the treated group. Examples: for equal sample size,
enter 1. For a 3:1 control:treated ratio, enter 3. In general, if R=n/m, then the total
sample size N is n(R+1)/R, where n and m are the sample sizes of the control and treated
groups, respectively.
Control-group survival rate. The expected cumulative proportion surviving
at the end of the study in the control group. A value greater than or equal to 0 and less
than or equal to 1 must be entered.
Relative risk, treated:control. The relative risk of the treated group
compared to the control group. A value greater than 0 must be entered.
Length of accrual period. The number of time periods required to recruit
all of the patients into the study. It must be a value greater than 0, usually expressed
as months or years.
Minimum follow-up time. The number of periods of follow-up that the last
patient recruited must have. Hence, the length of the entire study is the sum of the
length of accrual period and minimum follow-up time. The exponential hazard rates are
calculated based on the sum of these two values. This must have the same unit of time as
the length of the accrual period.
Two-sided type I error. The probability of rejecting the null hypothesis
given the null hypothesis is true. It must be a value greater than 0 and less than 1,
though common values include 0.01 and 0.05.
Non-Uniform accrual parameter. This specifies the distribution of the
patients being recruited into the study. Valid values are between –10 and 10. A value
of 0 implies a uniform distribution. Otherwise, a truncated exponential distribution is
assumed: negative values imply that more patients are recruited later in the accrual
period, whereas positive values imply that more patients are recruited early in the
accrual period. For example, -9 implies that almost all of the patients are recruited at
the end of the accrual period, and 2 implies that slightly more patients are recruited
early in accrual period.
Control-group and treated-group loss rates. The expected cumulative
proportions at which those in the control and treated groups are lost to follow-up by the
end of the study. Different rates for each group is allowed. It is assumed that these
losses are distributed exponentially. Values must be greater-than or equal to 0 and less
than 1.
Drop-in and drop-out rates. The expected cumulative proportion of patients
who cross over to the treatment assigned to the other group. It is assumed that the
drop-ins and drop-outs are subject to the same hazard rate for the treated and control
groups, respectively, from enrollment until the end of the follow-up period.
Executing the calculation, modifying the parameters, and exiting SSP
Once all parameter values are entered, and assuming the values are valid, simply
positioning the arrow with the mouse over the "Calculate" button and
left-clicking on it, or tabbing over until the button is highlighted and pressing
"Enter" on the keyboard, executes the calculation process. If any of the values
are invalid, an error message will appear prompting the user to modify the particular
parameter. To modify one or more of the parameter values, the user can place the cursor in
the appropriate number field by positioning the arrow with the mouse and left-clicking
anywhere inside the number field, and modifying as necessary with the delete or backspace
keys. The user can also clear all values by pressing the "Reset Values" button.
At any time, the user can close SSP by simply navigating to another web page.
|