Physical Activity Guidelines Advisory Committee Report
Part F. Appendix F.3: Inclusion and Exclusion Criteria

• Inclusion Criteria
• Exclusion Criteria
• Inclusion/Exclusion Coding System
• The Flow Process

Inclusion Criteria

Studies were considered for inclusion for review, if the following criteria were met:

• English language literature,
• Date of publication between January 1, 1995 and June 30, 2007,
• Original research reports only (reviews, meta-analyses, structured reviews, previously published physical activity guidelines, and recommendations to be retrieved and stored for background information for use by Physical Activity Guidelines Advisory Committee [PAGAC]),
• Peer-reviewed literature only,
• Main antecedent or exposure variable is physical activity or exercise, and
• Main health outcome variable or risk factor is clearly described and fits into one of our health outcome categories.

Exclusion Criteria

The purpose of the Exclusion/Inclusion flow process was to classify efficiently and accurately references for the abstraction process. The overall process is a 2-phased system: Studies of physical activity and a diagnosable health outcome were abstracted during Phase 1, followed by studies of physical activity and risk factors for the health outcomes, possibly abstracted at a later date (Phase 2), if requested by the PAGAC. Abstracts that were initially excluded were coded using a system that will allow for later retrieval, should it be necessary.

The goal of the Exclusion/Inclusion process was to code references identified from the "Physical Activity AND Health Outcome" search strategy (see description of Search Strategy for specific search terms) into mutually exclusive categories, indicating if they will be included or excluded for abstraction. Upon initial review, the abstract was screened by a team leader and assigned one of the following Exclusion/Inclusion codes:

1 = Yes (Meets screening criteria, will be abstracted);

2 = No (Does not meet screening criteria, will be excluded from further review); or

3 = Phase 2 (Does not meet screening criteria for inclusion in Phase 1 abstraction process, but does meet screening criteria for Phase 2 abstraction process).

Inclusion/Exclusion Coding System

As abstracts were reviewed for exclusion/inclusion, team leaders worked through the following hierarchical flow system. If an abstract was "stopped" at any point in the flow process, then the appropriate exclusion code was entered into Reference Manager in the user-defined "Abstract" cell. The flow system is described below.

If the reference was coded "1" (YES) or "3" (Phase 2) for abstraction, the reviewer also coded the "Age group" and "Health outcome" cells in Reference Manager.

If one team leader was unsure how to triage an abstract, he/she would consult with another team leader for their input.

The Flow Process

In the first step of the flow process, the reviewer will determine if the abstract contains an appropriate measure of physical activity. If YES, then the reviewer will move on to step 2. If NO, the reviewer will enter one of the following exclusion codes into Reference Manager: 11 = Missing physical activity; 12 = Single acute bout of exercise; 13 = Specific therapeutic exercise (Kegel, inspiratory muscle training); 14 = Single measure of physical fitness (cardiovascular fitness, strength, flexibility); 15 = Not primary exposure variable or used solely as confounding variable; 16 = Sedentary behavior (TV viewing, computer games, sleep, other); or 17 = Other.

Definitions of the exclusion codes and examples to illustrate each of the codes are provided below.

Once an eligible abstract is passed to the second step, the reviewer will next determine if the health outcome is appropriate for abstraction. We abstracted in two phases, where "Phase 1" consisted of the diagnosable or so-called "hard" health outcomes, and "Phase 2" consisted of the "risk factors" for the Phase 1 outcomes. If the health outcome was deemed appropriate for Phase 1 abstractions, the reviewer moved to step 3. If not, the reviewer entered one of the following exclusion codes into Reference Manager: 21 = Missing health outcome; 22 = Inappropriate health outcome (see Phase 1 health outcome list); 23 = Phase 2 health outcome (see Phase 2 health outcome list); 24 = Study of physical activity correlates; or 25 = Other.

In the third step, the reviewer will determine if the article is of an appropriate format to abstract. If not, the reviewer will enter one of the following exclusion codes into Reference Manager: 31 = Review, meta-analysis, no abstract; 32 = Case report; or 33 = Other.

Once the reviewer determines that an article is NOT excluded for inappropriate physical activity, health outcome, or format, the article is coded as 1= "YES" in the Abstraction cell of Reference Manager. The reviewer will also need to code for Age Group (1 = Youth; 2 = Adult; 3 = Older Adult; or 4 = Age Unknown) and Health Outcome Categories (1 = Cardiovascular/Respiratory Health; 2 = Metabolic Health; 3 = Musculoskeletal Health; 4 = Cancer; 5 = Functional Health; 6 = Mental Health; 7 = All-Cause Mortality; and 8 = Adverse Events/Risk of Physical Activity).

If multiple health outcomes are found in the paper, the reviewer will code for the most prevalent or most relevant.

Table 1 provides exclusion codes for the initial review of references.

Table 1. Exclusion Codes* for Initial Review of References: Physical Activity Guidelines for Americans

1. Physical Activity

2. Health Outcome

3. Inappropriate Format

Phase 2 Abstraction Criteria From Federal Advisory Committee

Phase 2 Abstraction Codes Specific to Mental Health Subcommittee

*There are exceptions to this model. Studies with an appropriate health outcome and with physical activity classified as shown below will be included and coded:

If the study:

(1) Is an experimental study where physical activity + diet (non-drug) is a treatment arm.

(2) Is an experimental study where the treatment arms are able to test the singular effect of physical activity. For example, in a study that included two treatment arms: (1) Physical Activity + Treatment X versus (2) Treatment X.

(3) Studies where the physical activity exposure is based on group membership (not individually assessed — e.g., occupational category; athletic/sport participation, etc.) will be included.

top of page

Continue to G1. All-Cause Mortality
Back to F. Scientific Literature Search Methodology Appendix 2
Back to Physical Activity Guidelines Advisory Committee Report