This is an archived version of the Handbook. For the current version, please go to or search for this chapter here.

17.2  Patient-reported outcomes and Cochrane reviews

Systematic review authors will select PROs for inclusion depending on the scope and aims of their review. PROs are most important when externally observable patient-important outcomes are unavailable, or rare. For many conditions, including pain, functional disorders, sexual dysfunction and insomnia, no satisfactory biological measures are available. Conditions in which outcomes are known only to the patients themselves, such as pain intensity and emotions, demand PROs as primary outcomes. PROs are also important when observable outcomes are available, because they reflect directly what is important to patients.


An important early part of the systematic review process is to define and list all patient-important outcomes that are relevant to their question (Guyatt 2004) (see Chapter 5, Section 5.4.1). This step is highly germane to the measurement of PROs. Many primary studies fail to measure aspects of perceived health and quality of life that are very important to patients. When this is the case, evidence regarding impact of interventions on PROs may be much weaker than evidence regarding impact on disease indicators such as morbidity or mortality. In the extreme, there may be a line in a ‘Summary of findings’ table that is blank, that is, for instance, a line specifying health-related quality of life (HRQL) that is blank because no study addressed this issue directly. The careful prior consideration of all patient-important outcomes and inclusion as a blank row in a ‘Summary of findings’ table will highlight what is missing in outcome measurement in the eligible randomized trials and other studies.


It is important that review authors understand the nature of the PROs used in the studies included in their review, and communicate this information to the reader. In clinical trials, investigators use many instruments to capture PROs, and methods for developing, validating, and analysing PRO data are diverse.