Darcy Ummels

Counting steps in activities of daily living in people with a chronic disease | 23 2 Introduction The use of activity tracking to self ‐ monitor physical activity is gaining popularity. In 2015, 1 out of 3 Dutch inhabitants was using apps, wearables, or activity trackers. 1 Physical activity is the most popular variable measured with these devices followed by nutrition, weight, and body functions (e.g., blood pressure). 1 Initially, these activity trackers were developed for athletes and the healthy population, but they could potentially also be useful in treating people with medical conditions (e.g., physiotherapy treatments). The Royal Dutch Society for Physical Therapy composed physical activity intervention guidelines for the most common chronic diseases seen by a physiotherapist 2 : cardiovascular disease 3 , diabetes mellitus 4 , chronic obstructive pulmonary disease (COPD) 5 , chronic pain 6 , cancer 7 , and osteoarthritis. 8 In all these guidelines, it is recommended to objectively measure the physical activity level of a patient outside of guided therapy. 2 Frequently used measurement tools by physiotherapists are questionnaires or diaries, but they have limited reliability and validity, tend to overestimate most activities while underestimating low intensity activities, and are time consuming to fill out. 9,10 For patients and physiotherapists, more objective and feasible measurement tools are useful, and activity trackers seem to be a good alternative. 11 To provide guidance in choosing an appropriate activity tracker for people with a chronic disease, we performed a literature search on the validity of activity trackers, preferably commercially available ones. The following criteria were taken into account. First, step count was considered to be the most important outcome, since it is specific to ambulation and easily interpreted by patients and physiotherapists. 11 Second, people with a chronic disease should be the target population of the study, as they often have impaired ambulatory abilities (e.g., shuffling) 12 , and activity trackers may measure incorrectly due to these altered walking patterns. 13 ‐ 15 Third, activities of daily living should be assessed (no laboratory settings), as insight into these specific activities (e.g., vacuum cleaning, walking stairs) is needed to monitor and coach participants in daily life, and activity trackers are not able to measure validly during low walk speeds (<0.8 m/s) 16 , which is often the case in activities of daily living. Last, published articles were screened on standardization of the performed activities of daily living by means of an activity protocol. Although the literature on clinometric quality of commercially available activity trackers is growing 17 ‐ 19 , only a few recent studies were found in which almost all criteria were met (validity of step count of commercially available activity trackers during free living conditions). 19,20 However, the target population in those studies consisted of healthy participants.