How many stratification factors

Stratified randomization refers to the situation in which strata are constructed based on values of prognostic variables or baseline covariates and a randomization scheme is performed separately within each stratum. One misconception is to think that the stratified randomization is going to require the equal number of subjects for each strata.

But we don't know how many subjects in each age group we could enroll. The purpose is to make sure that within each age group, there are equal numbers of subjects assigned to treatment A or treatment B. After the study, there may be quite different total number of subjects in each age group, but within each age group, there should be approximately equal number of subjects in treatment A or treatment B.

The strata size usually vary maybe there are relatively fewer young males and young females with the disease of interest. Strata opens the Strata window, where you map the strata from the legacy system to the corresponding target Oracle Clinical codes. You can map each target stratum to more than one legacy stratum.

Since some legacy systems use strata differently from Oracle Clinical's normal and replacement assignments, you can set a flag to indicate that Oracle Clinical merges the strata into Oracle Clinical's replacement assignments. Patterns opens the Treatment Pattern Conversions window of the Maintain Conversions window, where you map the legacy treatment pattern code Old Pattern field to the corresponding Oracle Clinical code. You can have only a one-to-one correspondence. Treatment and patient codes are character fields but are limited to a single, leading, non-numeric character.

Right-align these fields, and do not add leading zeros; for example, S Clinical Study Card Type. Treatment Pattern Card Type. Converted file structure. If required, you can run a report to read the original data file and compare it with the Oracle Clinical database to confirm that the load was correct.

Enter the name of the loaded randomization file in the Description field. You can perform the following tasks by navigating to Design , Randomization , and then Randomization Maintenance :. Replicating randomizations is a similar process to replicating clinical studies. The system records information you enter in this module into system tables, including Treatment Pattern Block Definitions, Treatment Assignments, and Randomizations.

You cannot replicate a randomization between studies. You must replicate the study first, then add the randomization to it. Use the list function to select the study to which to replicate the randomization the target study.

Mirror randomization takes an existing randomization and creates a new randomization of replacement treatment assignments starting at a new code. For example:.

The main use is when a site receives supplies for a group of patients and a corresponding set of replacement patients. Then there is an instruction that if, for example, patient 7 drops out, enroll the next patient into This maintains balance in the study's randomization. Click Show Randomizations. The Copy Randomizations window displays the selected study's randomization details. Click Copy. A pop-up window appears where you can specify the starting code.

This section describes updating the status of the randomization. The information you enter here is a basis for creating clinical study history records. To modify the study's randomization access, click the Phases button. A window appears titled Amend the Randomization Access Status. Clicking the Amend the Study Status button lets you change the planning status only, and even to do this, you must have planning attributes privileges. You can modify the Randomization Access Status only if you have access to the randomization attributes.

The studies display in study code sequence. You can scroll or query to the study of interest. You can only modify a clinical study's status; you cannot insert or delete studies here. When you click the Phases button, the system displays the phases of the live version of the current study. You can work with only the live version of a study while you are in the Amend Study Statuses module. If there is no live version of the current study, the system alerts you with a message.

An access view enables specified users to see the true randomization within a study without un- blinding the study to everybody. This technique controls who can perform an interim analysis, for which study, and when. If necessary, you can edit the view to additionally limit the data it displays. For example, you can limit it to display only the first 50 patient treatment assignments, or only patients who have completed the study.

Oracle Clinical provides the cohort view to show patients assigned to their correct cohorts groups without revealing which treatment pattern each cohort is for. The view has the same security safeguards as the Access View. You can use the PSUB submission form to create the view. The system assigns dummy codes to the study's treatment patterns by a hidden algorithm.

The algorithm decodes the real treatment pattern into letters—A, B, C, etc. To access the cohort view, the study must have randomization access status, regardless of your privileges. Take the Patients window as a starting point to manage patient positions. The system lists the patient positions in patient code sequence. Select the patient position to manage. You can only select a patient position while in this window. You cannot insert, update, or delete a patient position.

From the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and the Disclose button, selecting the appropriate study and patient position, to the Treatment Patterns Assigned to Patient form.

The system displays the treatment pattern records in assignment time sequence. See the next section for instructions on changing treatment pattern assignments. To change the treatment pattern assigned to a patient position without creating a new treatment assignment, you work from the Amending Treatment Patterns Assigned to Patient window.

From the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and the Change button, selecting the appropriate study and patient position.

This window is also available from the Change button in the Treatment Patterns Assigned to Patient window. Changing an assignment indicates that the treatment a patient receives is different from the treatment the randomization assigned to the patient.

This is the way you can correct mistakes in packaging or labeling. Enter the treatment code to change in the Treatment Code for this Assignment field.

Complete the Date and Reason for Change fields. You must click this button to save your work. You can replace the current patient position with a replacement patient already assigned to the same site. From the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and then the Site Replace button, selecting the appropriate study and patient position.

The system displays the Patient Replacement at the Same Site popup window, with the replacement patient controlled by the Patient Replacement Rule Type. A code on the clinical study version, the Patient Replacement Rule Type , controls how this part of the module operates. The Patient Replacement Rule Type has one of three values:. NEXT — The next sequential replacement patient position not currently enrolled, regardless of treatment. SAME — The next replacement patient position not currently enrolled that is assigned the same treatment pattern as the one being replaced.

The Replace Selected Patient With This One button confirms the patient position replacement and sets the patient enrolled date on the replacement patient to the current date. To replace the current patient position with a replacement patient not linked to any study site, from the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and then the Study Replace button, selecting the appropriate study and patient position.

The system displays the Patient Replacement not assigned to any site popup window, with the replacement patient controlled by the Patient Replacement Rule Type.

Clicking the Replace Selected Patient With This One button confirms the patient position replacement and sets the patient enrolled date on the replacement patient to the current date. To link the current patient position to an existing treatment assignment, from the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and then the Link Asg button, selecting the appropriate study and patient position.

A popup window appears, Link Treatment Assignment. This function is most useful in studies where the Kit codes or Treatment Assignment codes are not the same as the patient codes. The differences could be a result of a second randomization in the study, or of management of limited drug supplies. Clicking the Link Selected Patient To This Treatment Assignment button confirms the patient position linkage and assigns the new treatment code to the patient. You may have patients who receive more than one randomization during a study and, at the end, you have to assign the patients a reporting treatment pattern for analysis.

For these, and similar circumstances, Oracle Clinical provides a utility to create a new treatment assignment to assign to a patient position. From the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and then the Create Asg button, selecting the appropriate study and patient position.

The Create and Assign Treatment to Patient window provides a new treatment assignment code field and a list acceptable treatment patterns to assign to it. After you enter the fields and select a treatment pattern, you must click the Create a New Treatment Assignment Using This Treatment Pattern button to confirm the patient treatment assignment and create a new treatment assignment to allocate the selected treatment pattern to this patient. You can break the link between a patient position and a treatment assignment.

The treatment assignment is then available for reassignment. From the Design menu, select Randomization , choose Randomization Maintenance , select Patient Treatment Assignments , click the Patients button, and then the Unlink Asg button, selecting the appropriate study and patient position.

In the Unlink the Treatment Assigned for Patient window, you see a list of all the treatment assignments for the selected patient position. This method can be used to allow a blinded user to link patient positions to treatment assignments without knowing what treatment pattern is involved. The system creates a blind break with each disclosure. From the Design menu, select Randomization , choose Randomization Maintenance , then select Disclose Patient Treatment Assignments , click the Patients button, selecting the appropriate study and patient position.

The treatment pattern records display in assignment time sequence. Select the patient and click Disclose. Select Yes to confirm the disclosure and continue, or No to cancel the request and exit. This action always creates a blind break and should be your last choice. Use the Disclose option in the Maintain Treatment Assignments form to avoid creating a blind break whenever possible. Skip Headers. Patient positions represent potential study participants Once defined, you can use strata in any study, as needed.

This section contains the following topics: Stratification Randomization Randomization Maintenance Stratification Oracle Clinical provides stratification utilities to ensure that the groups you select are mutually exclusive and represent a full cross-section of your study's population. For example, you could stratify a study according to the following groups: male smokers over 65 male nonsmokers over 65 males 40 to 65 female smokers 45 to 65 females over 65 You create strata with a view to preserving statistical significance.

Maintaining Factors As a first step in creating strata, clinical study enrollment criteria, clinical study termination criteria, and treatment regimens, you create factors. Glossary of Terms for Factors Ranges represent variables requiring further qualification to make them meaningful to the study. Maintaining Single Strata To create, modify, and delete single strata, created for use in all clinical studies, use the Maintain Single Strata window. Selecting a Factor You select a factor from the Factors window list of all active factors before you can proceed to create a single stratum.

A new stratum can only be created for an active factor. Assigning Strata to Combination or Combined Combination Strata Select a factor and then click Strata to see the strata assigned to the factor. Add or remove special factor-related strata to the selected combination stratum with these buttons: Add assigns the selected stratum to the strata.

Remove removes the selected stratum. Displaying Strata Assigned to Combination or Combined Combination Strata Use this function to display the strata already assigned to a combination strata or combined combination strata and allow them to be removed.

Maintaining a Study Stratification Factor The assignment of a stratum to a clinical study version is a study stratification factor. Maintaining Strata To reach the Study Factors window, which lists the strata assigned to the study, navigate to Design , Strata , Stratification , and click the Stratum Assigned to the Study button. Two function-specific buttons appear in the Study Stratification Factors window: Factors displays the Factors window.

The following buttons are in the Strata window: Details of the Stratum displays the window Details of Selected Stratum. Randomization A study's treatment patterns specify the drugs to be administered and the medical procedures to be performed.

Senn SJ Added Values: controversies concerning randomization and additivity in clinical trials. Stat Med 23 : — Article Google Scholar. Simon R Importance of prognostic factors in cancer clinical trials. Cancer Treat Rep 68 : — Br J Cancer 89 : — Download references. You can also search for this author in PubMed Google Scholar. Correspondence to P Silcocks. From twelve months after its original publication, this work is licensed under the Creative Commons Attribution-NonCommercial-Share Alike 3.

Reprints and Permissions. Silcocks, P. How many strata in an RCT? A flexible approach. Br J Cancer , — Download citation. Received : 06 December Revised : 16 February Accepted : 20 February Published : 13 March Issue Date : 27 March Anyone you share the following link with will be able to read this content:.

Sorry, a shareable link is not currently available for this article. Provided by the Springer Nature SharedIt content-sharing initiative. Advanced search. Skip to main content Thank you for visiting nature. Download PDF. Subjects Cancer Clinical trials. This article has been updated. Abstract Background: The need to allow for prognostic factors when designing and analysing cancer trials is well recognised, but the possibility of overstratification should be avoided by restricting the number of strata.

Methods: Given a proposed sample size, a minimum allowable number in a stratum and an acceptable risk of observing fewer than this minimum, the number of strata can then be obtained by assuming a Poisson distribution for the number of observations per stratum. Conclusion: The method proposed is flexible and based on explicit principles and may be applied in the design or analysis of both clinical trials and epidemiological studies.

Main The need to allow for prognostic factors when designing and analysing cancer trials has been recognised for many years Simon, , because the treatment effect that is being sought will often be smaller than the effects of prognostic factors. Results Consider a hypothetically proposed two-arm study of patients, with an average block size of 4.

Conclusion An advantage of the proposed method over those proposed by Hallstrom and Davis or Kernan et al is that it is not necessary to assume a particular block size. Change history 28 March This paper was modified 12 months after initial publication to switch to Creative Commons licence terms, as noted at publication. Ethics declarations Competing interests The author declares no conflict of interest.