|Title:||Fenland Measurement Reactivity (FMR) – A randomised test of reactivity to objective measurement of physical activity embedded in the Fenland study: the Fenland Measurement Reactivity (FMR) study|
|Principal research question:||The Fenland study (an ongoing population-based cohort study of participants born between 1950 to 1975 to investigate the influence of lifestyle, diet, and genetic factors on the development of diabetes, obesity and other metabolic disorders) provides an opportunity to embed a randomised test of reactivity to the Actiheart (which is currently used in the Fenland Study), a combined heart rate and movement sensor, by giving participants different (but true) information about the purpose of the device and how the data will be used. The hypothesis is that participants who are told that the device measures physical activity as well as heart rate will have a higher level of physical activity on average during the 6-day period of use compared with participants who are told that the device measures heart rate. The size of the reactivity effect (difference in means between groups) will be estimated with a given degree of precision.|
|Background:||Because of the well-known limitations of self-report measures of physical activity, objective measurement is being increasingly recommended and used in epidemiological studies and trials of physical activity interventions to replace or complement self-report measures. The participant is asked to wear a device such as an accelerometer or heart rate monitor for a period of time (e.g. 7 days), and the data are used by the researchers to estimate the individual’s level of physical activity or energy expenditure over that period. However, even if the device does not have a visual display that provides feedback to participants, it is possible that it may be ‘reactive’ i.e. participants may increase their physical activity levels while wearing the device. The device may act as a salient prompt or reminder to engage in physical activity and participants know that the research team will be able to tell how active they have been and they cannot easily influence the data without making real changes in their physical activity.
Depending on the size of the reactivity effect, it could affect the validity of conclusions drawn from studies of physical activity. In descriptive epidemiological studies, the mean physical activity level would be over-estimated. If there are systematic individual differences in reactivity (i.e. if some kinds of people react more to measurement than others), this could affect the observed associations between the measure of physical activity and measures of determinants or consequences. It is therefore important to try to estimate the size of the reactivity effect to enhance interpretation of study findings. Finally, a reactivity effect might inform the development of interventions to promote physical activity.
|Methodology description:||Potential participants to the Fenland study will be randomised to the two conditions (A = comparison; and B = experimental) prior to being sent a letter inviting them to take part in the Fenland study, enclosing version A or version B of the Fenland Study Participant Information Sheet and a reply slip. If possible, stratification or minimisation will be used to ensure that the two groups are balanced on relevant characteristics (e.g., gender; practice). Randomisation will be conducted by a statistician or researcher with no direct involvement in the study.
The two versions of the Fenland Study Participant Information Sheet will be very similar to the existing one but with changes to the bullet point currently called “Measure your fitness and physical activity level”. The guiding principle is that, although different participants will be given different information (which is essential to the study design), all information provided to participants is true. In Version A of the Fenland Study Information Sheet, the heading will be changed to “Measure your fitness and heart rate” and the monitor will be referred to as a “heart rate monitor”. In version B, the bullet heading will be left unchanged, the monitor will be described as a “combined heart rate and movement monitor” and the following sentence will be added: “By analysing the information stored in the monitor, the research team will be able to tell exactly how active you have been during these six days”. Corresponding changes will be made to the visit instruction sheet and the Actiheart instruction sheet that are sent to Fenland participants before their baseline visit.
Two weeks before the Fenland baseline measurement visit, participants will be sent an Actiheart and an instruction sheet explaining the purpose of the device and how to wear it. After they have worn the Actiheart for six days, they will be asked to post it back to the measurement centre together with a brief diary sheet noting when they started and finished wearing the monitor and to do this before they attend the Fenland baseline measurement visit. At the measurement visit, participants will be asked to complete a short Actiheart Evaluation Sheet to find out what they thought the device was measuring, whether they read the instruction sheets, whether they thought that they had increased or decreased their physical activity while they were wearing the monitor, and measures designed to assess possible mediators of any reactivity effect. The measurement staff will then fully debrief the participant, explaining the Fenland study and the reactivity component, and ask for written consent for the FMR research team to use the Actiheart data, Actiheart Evaluation Sheet data and relevant baseline characteristics (gender, age, social class, practice, measurement centre).
|Outcome measure description:||The primary outcome measure will be daily accelerometer activity counts as measured by the Actiheart, averaged over the six days of Actiheart use. The secondary outcome measure will be physical activity energy expenditure (PAEE, measured in kJ/kg/day), using heart rate monitoring with individual calibration for the heart-rate energy expenditure relationship, again averaged over the six days. These measures are already obtained in the Fenland study so no additional measurement is required for the proposed sub-study, apart from the Actiheart Evaluation Sheet referred to in the preceding section. Information on participants’ gender, age, social class, practice and measurement centre (which are already collected as part of the Fenland study) will be needed to describe the sample, check for comparability between the two groups and to increase the sensitivity of the main analysis.|
|Start date:||3rd January 2011|
|End date:||31st March 2013|
|Contact person:||Professor Stephen Sutton|
|Contact Details:||Primary Care Research Unit
Institute of Public Health
University Forvie Site, Robinson Way
Telephone: (01223) 330594
Fax: 01223 762515
|Collaborative:||Simon Griffin (MRC Epidemiology Unit) Co-Investigator|
|Funding Organisation:||NIHR School for Primary Care Research|
|References and Publications|