To the best of our knowledge, this systematic review is the first to formally aggregate the research evidence on the measurement properties of questionnaires assessing the use of CAM in pediatrics. This work will help to better understand the use of CAM in various pediatric clinical and research settings by informing future validation studies of CAM questionnaires. It will also help clinicians, researchers and the general public to better appraise the scientific literature on CAM in pediatrics.
The results of the current review suggest that many studies, mostly cross-sectional and conducted in the United States, have used CAM questionnaires to identify and understand the use of CAM among pediatric populations. However, CAM questionnaires varied greatly in terms of their characteristics. Some were disease-specific while others were generic, which may reflect that both types of instruments are seen as helpful in order to understand the use of CAM by researchers. Some succeed in targeting specific types of CAM used by patients with a certain condition while others have the advantage of assessing the use of CAM in many different conditions as well as the public at large, making it easier to compare use in different populations. Most of the questionnaires targeted parents and very few mentioned using a questionnaire specifically designed to gather children and adolescents self-report, which may be important to use in order to gather valid data about CAM use according to children. Such a self-report questionnaire may provide researchers with a valid tool to compare children’s and parents’ perceptions, which may lead to changes in the way health care providers discuss CAM with families. The high variability in the content of CAM questionnaires was also problematic and may explain great variability in use in the pediatric studies, such as in cancer care 
. This may be explained by the fact that there is no consensus on a definition of CAM and the therapies that it encompasses as well as items that should be assessed in order to understand thoroughly the use of CAM 
. Very few questionnaires were reported to have a completion time under 10 minutes, which might pose a problem if used in a research setting and routine clinical practice.
Studies included in our review were judged to be of poor methodological quality when evaluated by the COSMIN checklist since methods used to test measurement properties of CAM questionnaires were not described in enough detail to reach pre-defined criteria. Poor methodological quality of the studies may be explained by the fact that none of them were specifically designed to validate a CAM questionnaire, but rather sought to describe the use of CAM in specific populations. However, in order to do this, questionnaires still need to meet validity and reliability criteria and be described in a comprehensive manner.
None of the included CAM questionnaires met the Terwee criteria for measurement properties. The lack of thoroughly valid CAM questionnaires may be explained by the relative novelty of research in CAM compared to other fields and the small amount of funding provided to do such research, especially before the creation of various funding agencies in the early 1990’s such as the National Center for Complementary and Alternative Medicine (NCCAM) in the United States 
. Moreover, funding agencies as well as the general public have been interested primarily in evaluating CAM use in specific populations possibly because it was most needed data to show the relevance of research in CAM. Furthermore, conducting validation studies takes time, money and energy, which may explain why researchers often engage in studies without using properly validated tools, even in fields other than CAM 
. However, the fact that CAM questionnaires are not valid is a problem since results may not reflect the true use of CAM when such therapies are not clearly and comprehensively defined and presented to parents and children. An international CAM questionnaire was developed for adults and is currently being validated 
but results may be difficult to extrapolate to children since perceptions concerning CAM may differ in these two populations 
. Our results highlight the need for further research on defining and conceptualising CAM as well as developing valid questionnaires of its use, especially in pediatrics.
Using a research protocol similar to Terwee et al., helped to ensure credibility of our findings. However, the protocol as well as the COSMIN checklist and the Terwee score have been used for health status measures and not for CAM questionnaires. Because CAM questionnaires have been less extensively studied than health status measures, we encountered several difficulties in applying the same criteria. For example, the COSMIN checklist evaluates the methodological quality of validation studies, which were not conducted in CAM research and rendered parts of our assessment less pertinent. Furthermore, quality of reporting was very poor in CAM studies, which we tried to counteract by communicating with authors. Unfortunately, many did not reply and some replied without giving more details than their published work, which precludes us from knowing exactly how they proceeded to ensure their CAM questionnaires were adequate. Despite these limitations, we feel that currently available measures are inadequate and a new questionnaire is required.
In conclusion, our systematic review showed that many surveys have been conducted to document and understand the use of CAM in pediatrics without having properly validated their tools. Such tools, if valid, may prove to be useful in pediatric clinical settings to document the use of CAM as well as their perceived benefits and harms. This study highlights the need for research to define and conceptualise CAM as well as develop valid questionnaires of its use. Such tools may in turn lead to improved research and knowledge translation about CAM benefits and harms in clinical practice.