Background Knee osteoarthritis (KOA) is the most common form of arthritis with an estimated lifetime prevalence of 44.7% (1). The use of orthotic devices (knee braces, insoles, wedged shoes) is a generally accepted conservative therapy for KOA patients (2). However, it is suggested that their effectiveness is detrimentally affected by poor patient compliance due to discomfort while wearing these devices (3).
Objectives The aim of this study is to objectively establish the compliance to an orthotic intervention using a thermal sensor (TS) and comparing it to patients' self-reported wear time.
Methods Ten medial KOA patients (mean±SD age 57.6±13.4 years, BMI 27.3±3.2 kg/m2), clinically diagnosed according to the ACR Guidelines, were recruited for this study. A small, light-weight TS with the size of an average lithium battery (0.5°C resolution) was placed in a newly developed ankle-foot-orthosis (AFO) that patients were asked to wear as often as possible during a period of six weeks. The TS measured the temperature every 5 minutes for 4 weeks. Patients rated the comfort of the orthosis during the first visit using a scale from 1 to 5 (1 being the most comfortable).
Additionally, the patients reported during these six weeks how many hours per day they had worn the orthosis and rated which amount of pain they felt each day (from 0=no pain to 10=most painful). To determine the patients' compliance, the AFO wear time, derived from the TS, was compared to the wear time per day recorded in the patients' diaries. The threshold to differentiate the wear and non-wear times was set at 25°C. Timeframes during which the temperature rose above 25°C were classified as wear time. A Wilcoxon signed-rank test was performed between the different wear times and Spearman's correlations between pain, comfort and wear time were determined.
Results On average, patients wore the device for 143±80 h according to the form they filled out, whereas the TS measured only 83±86 h, leading to an overestimation of 72%. Patients reported an average pain level of 3±1.4 during the six weeks period and the AFO's comfort was rated with 1.9±0.3.
Statistical differences were found between the wear time reported by the patients and the wear time derived from the TS (p=0.005). Additionally, a significant correlation between the AFO comfort and the wear time derived from the TS was found (r = -0.81; p=0.001). No significant correlation was found between pain and wear time.
Conclusions As can be expected, patients who found the AFO more comfortable were the ones who wore the orthosis for longer periods. The wear time recorded by the TS was significantly lower compared to the self-reports. This might be due to a social desire bias, overestimating the amount of hours of wearing the AFO as we asked the patients to wear it as often as possible. Secondly, patients might have trouble to accurately recall the amount of wear hours, thereby inducing an overestimation of wear time. As the reliability of self-reported measures seems to be questionable, such data should be interpreted with care.
Lee S et al. Int J Rheum Dis, 2015.
Duivenvoorden T et al. The Cochrane Library, 2015.
Brouwer RW et al. Cochrane Database Syst Rev, 2005.
Disclosure of Interest None declared