Article Text
Statistics from Altmetric.com
We would like to thank Dr Langer and colleagues for the interest shown in our paper.1 The authors appear to have interpreted our article and the accompanying editorial as suggesting that accelerometers measure exactly the same outcomes as activity monitors. This is not the case, as evidenced by the fact that we used an activity monitor (Dynaport) to validate the recordings made with an accelerometer (Actiwatch). However, the data we presented show a close correlation between measurements of overall activity recorded by the two devices (overall activity score: r = 0.92, intensity of activity score: r = 0.83). This supports our assertion that leg activity measured by the Actiwatch is the major determinant of whole body activity measured by the Dynaport.
Accelerometers do not try or claim to specifically measure time spent walking, which is not the sole contributor to overall whole body activity. In our COPD population, time spent cycling is rarely of relevance and no patient spent any time cycling during our Dynaport recordings. We do not dispute that time spent walking is a useful measure and an easy concept for an individual to understand, but we disagree that level of physical activity is conceptually difficult for a patient to comprehend. In fact, the UK government has tried specifically to address the issue producing guidance on how to increase physical activity in the overall population.2 We believe that improving level of physical activity after an intervention is an outcome with which patients can identify. Despite the concerns raised, the Actiwatch was able to detect change in activity in slow moving patients after a standard exercise programme, even with a similar level of disease severity and improvement in walking distance compared with previously published results.3
Clearly the information obtained from precisely measuring time spent walking and cycling has to be balanced against the lower cost and ease of use of accelerometers. In effect, purchase of current activity monitors is impossible for almost all rehabilitation programmes so, although this outcome is an important one, it will not be measured. In other studies we have found that a significant number of patients considered the activity monitor cumbersome and difficult to use and, as a consequence, failed to complete adequate recording time.4 We agree that, in time, activity monitors will advance technologically and current problems will be overcome but, at present, they are likely to remain a research tool because of the additional information they supply. For these reasons we feel that accelerometers are a more appropriate device for clinical practice because they accurately measure activity, are affordable and easier to use.
Footnotes
Competing interests: None.