Validity Of Smartphone App To Measure Activity In Patients Post-ACL Reconstruction Or With Knee OA

PURPOSE: Determine if myrecovery© app can be a valid proxy for activity in patients with knee osteoarthritis (OA) and ACL reconstruction (ACLR) by comparing app activity measures with self-report journals. METHODS: We prospectively selected 200 active patients (patients with OA or patients at least 9 months following ACLR), of various sex, age, sport, and activity level. Participants tracked their daily activity for 4 weeks and used the myrecovery© app for the same period. Participants were encouraged to carry their phone unless not permitted by their activity. Activities from the diaries were converted to metabolic equivalent of task (1 kcal/kg/hr) per minute (MET/min). After the 4-week period, patients competed a Marx Activity Rating Scale (MARX) to denote their activity intensity over the past month. To investigate the association between activity measures, we constructed scatterplots and used Pearson’s r for linear data. Associations were interpreted as follows; poor: < 0.2, fair: 0.2 < 0.4, moderate: 0.4 < 0.6, strong: 0.6 < 0.8, very strong: 0.8 < 1.0. RESULTS: We consented 100 OA patients (39% female; age 58 ± 8 years) and 100 post-ACLR patients (43% female; age 25 ± 8 years). The average association between MET/min and step count for the 4-week period was r = 0.51, p = 0.002 for the OA group and r = 0.40, p = 0.03 for the ACL group. The association between MET/min and MARX for the 4-week period was r = 0.01, p = 0.95 for the OA group and r = 0.45, p = 0.01 for the ACL group. CONCLUSIONS: Associations between self-reported journals (MET/min) and app data (step count) were statistically significant for both groups, however the correlation was stronger for the OA patients compared to the ACLR patients, likely due to ACLR patients’ limited ability to carry their phones during sport. There was a strong association between MET/min and MARX scores (both self-reported) in the ACLR group, but the correlation between these measures for the OA group was weak. The type of activities being performed should be considered when determining the best method of measuring exposure. An activity score like the MARX may be a more valid measure for highly active patients participating in sport, such as those undergoing ACLR, whereas a app measuring step count such as myrecovery© may be preferred for older, less active patients with OA.