Inclusion Criteria

Inclusion criteria are the predefined characteristics that a study participant, data source, or published study must possess to be eligible for enrollment in a research study or for selection in a systematic review. They are specified in the study protocol before data collection begins to ensure that the selection process is consistent, transparent, and reproducible.

Inclusion criteria directly determine who enters the study and, consequently, the population to which findings can be generalized. In clinical trials, overly narrow criteria may yield a homogeneous sample that produces clean results but cannot be applied to the diverse patients seen in routine practice. Overly broad criteria may introduce excessive variability that obscures treatment effects. In systematic reviews, inclusion criteria define the scope of the evidence base and set out which studies are relevant to the review question. Poorly specified criteria lead to inconsistent screening decisions, irreproducible reviews, and challenges during peer review.

Inclusion criteria are established during the protocol development phase and documented in the study registration (such as ClinicalTrials.gov or PROSPERO) or protocol paper. For systematic reviews, inclusion criteria typically follow the PICO framework, specifying the acceptable population characteristics, intervention types, comparators, outcomes, and study designs. Some reviews add additional elements such as setting, time frame, or language restrictions.

For clinical trials, inclusion criteria define the demographic, clinical, and procedural requirements that participants must meet. Examples include age range, confirmed diagnosis using a specified set of criteria, minimum or maximum values for a clinical parameter (such as HbA1c above 7.5%), willingness to provide informed consent, and ability to attend follow up visits. Each criterion should have a clear rationale linked to the study's objectives, safety considerations, or regulatory requirements. When the trial is reported, CONSORT 2025 asks for the eligibility criteria for participants to be stated (item 12a).

In systematic reviews, screening against inclusion criteria is typically performed in two stages. Title and abstract screening eliminates clearly irrelevant records, and full text screening assesses the remaining records against each criterion. Best practice calls for two independent reviewers to screen records, with disagreements resolved through discussion or by a third reviewer. Documenting the reasons for exclusion at the full text stage and reporting them in the flow diagram, as PRISMA 2020 asks, makes the selection process transparent.

Define inclusion criteria at the protocol stage of any research study or systematic review. They should be specific enough to create a coherent study population but not so narrow that they exclude clinically relevant subgroups or make recruitment infeasible. Pilot testing the criteria on a sample of records (in reviews) or potential participants (in trials) helps identify ambiguities before the study begins. Revisiting inclusion criteria after data collection has started is discouraged because it introduces the risk of bias, though protocol amendments with documented justification are sometimes necessary.

One common error is defining criteria that are vague or open to interpretation, such as "patients with significant comorbidities" without specifying which comorbidities or how they are measured. This leads to inconsistent application across screeners or sites. Another mistake is setting criteria so restrictive that the study population does not represent the broader patient group, limiting generalizability. In systematic reviews, a frequent error is applying inclusion criteria inconsistently between reviewers due to inadequate training or unclear operational definitions. Some review teams also fail to prespecify all criteria in the protocol, adding criteria after seeing the search results, which introduces a risk of bias in which studies are selected.

A systematic review of cognitive behavioral therapy for insomnia specifies the following inclusion criteria: randomized controlled trials enrolling adults aged 18 or older with a clinical diagnosis of insomnia according to DSM 5 or ICSD 3 criteria; interventions consisting of at least four sessions of individual or group cognitive behavioral therapy; any comparator including waitlist, sleep hygiene education, or pharmacotherapy; and primary outcomes including sleep onset latency, wake after sleep onset, or sleep efficiency measured by validated instruments. Studies of self reported sleep difficulty without a formal clinical diagnosis are excluded.