Methodological considerations for surveillance in GI practice

• Published in: Baillière's best practice & research. Clinical gastroenterology Vol. 30; no. 6; pp. 867 - 878
• Main Authors: Holme, Oeyvind, Loeberg, Magnus
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 01.12.2016; Elsevier Limited
• Subjects: Bias; Cancer; Colorectal cancer; Esophagus; Gastroenterology; Gastroenterology - methods; Gastroenterology and Hepatology; Health risk assessment; Humans; Medical prognosis; Mortality; Patients; Population; Population Surveillance - methods; Practice Patterns, Physicians; Research Design; Screening; Stomach Neoplasms - diagnosis; Studies; Surveillance; Tumors; Screening; Surveillance; Bias; Cancer
• ISSN: 1521-6918; 1532-1916
• DOI: 10.1016/j.bpg.2016.10.011

Surveillance is recommended for various GI cancers, and substantial resources are invested. However, little is known about the effect of surveillance, neither for good, nor for bad. Most evidence stems from observational studies, but observational studies of surveillance can be subject to various biases that may severely influence the results. In this chapter we discuss challenges related to various research questions, study designs, choice of endpoints, and how to deal with different forms of bias. We hope this chapter will be helpful for researchers when performing high-quality studies of surveillance, and to enable physicians and policy-makers to understand the possibilities and limitations of current evidence.