We all know that eCRF design is more of an art than a science. That being said… there are still some general rules that should be followed when designing eCRFs. Keeping these in mind when designing forms will make them easier to use for site personnel, monitors and data managers, and biostatisticians.
Emulate Paper Forms – When eCRFs resemble forms described in the protocol (or paper forms previously used for previous clinical trials), clinicians find them easier to use. This is especially true when site personnel are relatively new to Electronic Data Capture (EDC). Familiarity breeds acceptance!
“Right-size” Each Form – Well-designed eCRFs should not contain either too little or too much data. If too little data is collected on each form, the result may be too many forms for each study visit. This makes it more difficult for sites to navigate to the correct form and increases form signing and finalization overhead. Alternatively, when there’s too much information collected on a single eCRF the information on that form tends to be unrelated, and excess navigation (scrolling) within the form itself can lead to missing data.
Consider the Site’s Workflow – If the protocol calls for the sites to perform several procedures during one study visit, data for each should be collected on a separate form. This allows the sites to complete (and sign off on) one activity before proceeding to another (possibly totally different) activity. If data is to be entered by users with different roles (e.g., Investigator and Treatment Administrator), it may be prudent to use separate forms for that data. This makes assigning form permissions simpler and can also assist in the implementation of masking.
Consider Masking Implications – The correct implementation (and testing!) of masking is a critical component of most clinical trials. If a user should be masked to a particular data element (e.g., Treatment Group), but can also view the form containing the Treatment Group be sure to use field-level masking to ensure the field can’t be viewed by the masked users. In VISION, if a user is masked to a field – that field won’t appear when the user views the audit trail. Similarly, the field won’t be included in any database searches, reports, or exports done by the user.
Group Related Information – This may seem obvious, but we’ve actually seen forms where the fields were arranged in alphabetical order. Or in one case, fields were arranged by “type” (e.g., text and numeric). Grouping related information reduces data entry difficulties for site personnel – which increases accuracy. Plus, the form will probably look prettier!
These are just a few tried-and-true rules for good eCRF design. Stay tuned for Part 2 (coming soon!)