In the first installment of this blog series, we explored the pre-project considerations to take into account before embarking on a standard response document (SRD) writing project. Having decided to proceed, in this installment we will discuss the various decisions that need to be made with respect to format and content of the SRDs.
A major component is to decide on the region for which the responses will be generated. A global SRD, (usually English language) will ensure consistency of Medical Information (MI) responses in all regions as well as preventing duplication of effort within different regions. However, this may not be necessary for all products (i.e. if not available in all regions) and with this approach comes the requirement for local translation and customization alongside the challenges of sharing and version controlling documents on a global scale. Depending on the product and its licensing status, a regional (e.g. EU) approach to SRD generation may be more appropriate. Regional responses generally would require less tailoring of responses at the local level due to the shared regulatory processes and documents but would still have the requirement for translation. Finally, local responses would be of benefit where the product licensing is limited to specific countries.
The “reach” of the SRDs, then impacts other key considerations such as sources of information and response structure.
It is essential that SRDs should be unbiased and up-to-date, and a key approach in achieving this is ensuring that the right sources are reviewed consistently and at agreed intervals. Therefore the selection of an appropriate literature database (e.g. Medline, Embase etc.) or multiple databases is important. The primary data identified and selected in accordance with a hierarchy of evidence will form the majority of a robust evidence-based SRD. However, secondary sources are still of value for specific inquiry-types or for products in the later stages of their life-cycle, when more data has been generated. Secondary sources such as pharmacopoeias or specialist texts (e.g. drug interaction or compatibility texts) are particularly useful in that large volumes of data are succinctly presented or tabulated, as well as being familiar to customers. For more localized responses, the inclusion of national guidelines or references may be considered.
Exclusion of data can be just as important as inclusion. For older products, to include all relevant data identified from a literature search would result in an overly long, possibly confusing and probably unread response. Using an evidence hierarchy, data from case reports for example, may be excluded.
A key omission you may have noticed so far is the prescribing information! This must be included in all final responses to customers, but its inclusion in SRDs will depend on the scope of the project. Due to regional differences in prescribing information, a common approach is to exclude specific information from the prescribing information in order for this to be added at the local level through editing or in an accompanying covering letter for the SRD. At the regional or local level labeling may be included directly in the SRD.
Having decided on the content of a SRD, the next task is to decide how it should be presented and whether presentation can be flexible for different products in a portfolio as a one-size-fits-all approach may not be appropriate. As discussed previously, labelling may be added in a covering document alongside the SRD. If included directly within the document, it should usually appear first and foremost in the document. For some products, particularly early in their life-cycle, it can be useful to include a summary section highlighting the key points of the response, as customers may not yet be familiar with the product. In the later stages of the life-cycle this may not be necessary and could become onerous due to data volumes which may delay response delivery. The main content of the response may then be divided into key trial data, primary data from literature search, and secondary data which should be presented in accordance with an evidence hierarchy.
In specific cases, such as generic medicines where customer knowledge of the product is established, a focus of the MI service may be to ensure responses are delivered quickly. In this case, an approach may be to reduce tasks such as data summaries and enclose a bibliographic search with the response.
In the upcoming final post, we will review further potential solutions to the various SRD requirements of different pharmaceutical companies.