Case series investigate the presentation, history, and/or follow-up of a group of patients and provide valuable information on the natural history or prognosis of a disease. Case series may differ from clinical trials in regard to patient selection, patient characteristics, and length and completeness of follow-up; these characteristics may establish the quality and applicability of a case series. The case series provides preliminary information for a larger study with a comparison group.
Case series can include bias if they only include patients with severe disease from tertiary referral centers such as university-based clinics, or patients with only mild cases of a disease. For example, a study examining a new minimally invasive glaucoma surgery in patients with glaucoma may show incremental lowering of IOP and the need for fewer medications 6 months after the surgery. However, in the Methods section of the study, the reader discovers that the study population came from a tertiary glaucoma center that specialized in a new procedure to reduce patient dependence on drops. Also, the study only included patients with stable glaucoma. In other words, the study was biased toward patients with mild disease and was therefore not generalizable to patients with other severities of glaucoma; it was also biased to decrease the number of medications because the providers and patients involved in the study were motivated to decrease or stop their glaucoma drops after the procedure.
Case series might not standardize the collection of patient information, measurements, tests, and other evaluations. This may result in underreporting or overreporting of results. For example, the technicians, examination rooms, lighting, and charts used to measure visual acuity may differ within the various clinics. Or study personnel may record visual acuity differently; for example, some may record the nearest whole line (20/25) while others record to the letter (20/25 + 2).
Lengths of follow-up intervals may vary within a single case series. If there are differences in follow-up time, the study should report what the specific follow-up times were, such as 1, 2, and 3 years after the initiation of treatment. When the outcome being measured is an event, such as corneal graft failure, a survival analysis can account for the varying lengths of follow-up. If the study does not follow all of its participants for the full length of the possible follow-up period, these losses to follow-up may cause the reported outlook for the case series to be biased. For example, in a case series of patients with macular edema from branch retinal vein occlusion, some patients may choose not to return because their macular edema has resolved, and their vision has improved; some patients may experience further loss of vision and seek care from another ophthalmologist; and some patients may move to another location. Overall, if a large percentage of patients do not return for complete follow-up, the study results may not be valid; the remaining subjects may have had an unusually good or unusually bad course compared with the subjects lost to follow-up.
Excerpted from BCSC 2020-2021 series: Section 1 - Update on General Medicine. For more information and to purchase the entire series, please visit https://www.aao.org/bcsc.