To the medical editor, a SAP has a different meaning than common definitions of the word that the general public might know. Typically, sap might be fluid from a plant like a maple tree, which, when turned into maple syrup and poured over a Belgian waffle, might be part of a delightful Sunday brunch. It may also refer, in a derogatory way, to a foolish and gullible person.
But to a medical editor, SAP is the statistical analysis plan of a clinical trial, the blueprint of the analysis of the trial prepared and finalized before the clinical trial data were analyzed. The SAP is one of several critical components submitted with a clinical trial manuscript, along with information as to where the clinical trial is registered within an appropriate online public registry, the trial protocol, and any amendments to these documents.
Why such requirements? Clinical trials are very special submissions. While the results of many other submissions lead to hypothesis-generation to be evaluated in subsequent studies, results of clinical trials can have direct influence on clinical practice and clinical guidelines by suggesting cause-and-effect relationships.
Not all clinical research is a clinical trial. JAMA Ophthalmology follows the International Council of Medical Journal Editors (ICMJE) definition of a clinical trial, as “any research project that prospectively assigns human participants or groups of humans to one or more health-related interventions to evaluate the effects on health outcomes.” A comparison of the visual acuity outcomes of 3 anti-VEGF agents for macular edema from a central retinal vein occlusion could be a clinical trial. But an evaluation of outcomes of patients previously treated with anti-VEGF for macular edema who were switched to corticosteroids, through a medical record review— while potentially of value for hypothesis generation—is not prospective assignment to comparison groups to study cause-and-effect relationships. Authors of the latter type of investigations might submit a protocol or statistical analysis plan, or register their study with a clinical trial registry, but such items are not required for submission.
What does the journal do with a protocol, including a statistical analysis plan, submitted with a manuscript? These supplemental materials are used by the reviewers and editors to verify that what the authors did in their methods is what they said they were going to do before analyzing their data. This approach can avoid biases that might occur when analyses are undertaken after the outcomes or results are known.
If the submission is accepted for publication, the journal then publishes the protocol, including the SAP, and any amendments, as supplemental material online. In that way, readers, if interested, may have greater confidence in the results published. If there is proprietary or other confidential information in the materials, such as telephone numbers or addresses of investigators, those may be redacted, provided that an unredacted confidential copy is provided to the editors.
Two new developments in requirements for clinical trials are on the horizon. An additional instruction has been added just this week to JAMA Ophthalmology’s Instructions for Authors, stating that if the supplemental materials for a clinical trial, such as the protocol, including the SAP, are in a language other than English, both the original and English-translated documents must be provided. Also, a recent editorial in JAMA states that as of July 1, 2018, manuscripts submitted to ICMJE journals that report results of clinical trials must contain a data sharing statement regarding deidentified data. This does not mean that all data are to be shared, but is a step towards sharing enough data to insure the veracity of the publication of the clinical trial results and perhaps to allow people other than the investigators to use the data in ways that would maximize knowledge gained from the contributions of the study participants and the study investigators.