Background:  Epidemiological studies and surveillance systems of pregnant women often rely on collection of physical activity through self‐report. This systematic review identified and summarised self‐reported physical activity assessments with evidence for validity and reliability among pregnant women.

Methods:  Peer‐reviewed articles published through 2011 were included if they assessed validity and/or reliability of an interviewer‐ or self‐administered physical activity questionnaire or diary among pregnant women.

Results:  We identified 15 studies, including 12 studies that assessed questionnaires and 4 studies that assessed diaries, conducted in Australia, Finland, Norway, the UK, the US and Vietnam. For questionnaires, 92% (11/12) assessed mode, all assessed frequency and/or duration and 58% (7/12) collected information on perceived intensity. All but one study (92%) assessed validity of the questionnaires. Questionnaires compared with objective measures (accelerometers, pedometers) ranged from slight to fair agreement, while comparison with other self‐reported measures ranged from substantial to almost perfect agreement. Five studies (42%) assessed test–retest reliability of the questionnaires, ranging from substantial to almost perfect agreement. The four studies on diaries were all assessed for validity against objective measures, ranging from slight to substantial agreement.

Conclusions:  Selection of valid and reliable physical activity measures that collect information on dose (type, frequency, duration, intensity) is recommended to increase precision and accuracy in detecting associations of physical activity with maternal and fetal outcomes.