Manual or visual data review involves checks that require human logic, clinical context, and cross-domain knowledge, which are difficult to fully automate. Checking if a reported adverse event has a corresponding treatment listed in the concomitant medications log is a prime example. This requires a reviewer to logically link information across different eCRF pages (AE and ConMed) and apply clinical understanding to assess consistency and plausibility. Such complex, cross-form logical checks are characteristic of manual data review processes performed by data managers or clinical reviewers, whereas the other options represent standard, programmable edit checks.

A. Checking an entered value against a valid list of values is a standard, easily programmable edit check (e.g., a drop-down list).
C. Verifying mandatory fields for completeness is a fundamental automated check ("presence check") in any Electronic Data Capture (EDC) system.
D. Comparing a value to a predefined reference range is a common automated range check, especially for lab data.

1. Society for Clinical Data Management (SCDM). (2023). Good Clinical Data Management Practices (GCDMP).

Section 6.4.2, Data Review: This section describes data review activities that go beyond simple programmed checks. It states, "Data review is the process of reviewing data for errors, omissions, and inconsistencies... This review may be manual or computer-assisted." The example in option B, a cross-form consistency check requiring clinical context, is a classic example of the type of inconsistency identified during manual data review, as opposed to the simple, single-field checks described in options A, C, and D, which are covered under Section 6.4.1, Edit Check Programs.

2. Krishnankutty, B., Bellary, S., Kumar, N. B. R., & Moodahadu, L. S. (2012). Data management in clinical research: An overview. Indian Journal of Pharmacology, 44(2), 168–172. https://doi.org/10.4103/0253-7613.93842

Page 170, "Data validation" section: The article distinguishes between computerized checks and manual checks. It describes manual checks as a "manual review of the data for any inconsistencies, which are not covered in the computerized checks." The cross-check between adverse events and concomitant medications (Option B) is a typical example of a logical inconsistency that often requires manual review, whereas checks for valid values (A), missing data (C), and ranges (D) are standard computerized validations.

3. Lopienski, K., & Turek, P. (2017). Data Management in Large-Scale Research. Springer International Publishing.

Chapter 3, Section 3.2.2, "Validation Rules": This chapter details different types of data validation. It categorizes checks like missing values (C), value lists (A), and range checks (D) as basic, automatable validation rules. It contrasts these with more complex consistency checks that "verify logical dependencies between different data items," which often require manual review or sophisticated programming. The relationship between an AE and a ConMed (B) is a prime example of such a logical dependency.