Background: Accurate measurement of physical activity and exposure time is essential for evaluating risk, prevention, and treatment strategies, but traditional accelerometry and manual tracking are costly and burdensome. Smartphone step counting offers a scalable alternative, yet its validity in orthopaedic populations is uncertain. Purpose: To test whether smartphone step count provides a valid proxy for exposure time and activity level in functionally active patients with knee osteoarthritis (OA), and young athletes post–anterior cruciate ligament reconstruction (ACLR). Study Design: Cohort study (Diagnosis); Level of evidence, 2. Methods: Participants tracked activity for 4 weeks using a smartphone app and detailed activity diaries; secondary measures included the Marx Activity Rating Scale (MARS) and Sport Participation questionnaire. Correlations were assessed using Spearman's ρ, agreement using Bland-Altman analysis, and predictability using rank regression. Results: A total of 102 participants contributed data (ACLR: n = 47; 51% female; 24 ± 4 years; OA: n = 55; 44% female; 58 ± 11 years). The OA and ACLR groups showed poor correlations between app step count and diary-reported activity (OA: ρ = 0.17; P = .20; ACLR: ρ = 0.27; P = .07). Bland-Altman plots demonstrated minimal systematic bias with mean differences near zero, but wide limits of agreement (LoA) indicating poor interchangeability for individual monitoring (OA: d̄ = −0.10, LoA [–2.49 to 2.29]; ACLR: d̄ = −0.17, LoA [–1.93 to 2.28]). MARS correlated moderately with diary-reported activity (OA: ρ = 0.33; P = .02; ACLR: ρ = 0.45; P = .002), with stronger associations in ACLR participants. Sport participation was moderately correlated with diary-reported activity in the ACLR group. MARS scores were significant predictors of diary-reported activity in both groups, and sport participation was a significant predictor in the ACLR group. Step counts did not predict diary-reported activity. Conclusion: Our study demonstrated acceptable group-level agreement between the smartphone app and diary-based activity but insufficient precision for individual patient monitoring. MARS outperformed an app-based step count as a proxy for exposure time, particularly in athletes.
