TY - JOUR T1 - Post-consultation acute respiratory tract infection recovery: a latent class informed analysis of individual patient data from randomised controlled trials and observational studies JF - British Journal of General Practice JO - Br J Gen Pract DO - 10.3399/BJGP.2022.0229 SP - BJGP.2022.0229 AU - Hilda Hounkpatin AU - Beth Stuart AU - Shihua Zhu AU - Guiqing Yao AU - Michael Moore AU - Christin Löffler AU - Paul Little AU - Timothy Kenealy AU - David Gillespie AU - Nick A Francis AU - Jennifer Bostock AU - Taeko Becque AU - Bruce Arroll AU - Attila Altiner AU - Pablo Alonso-Coello AU - Alastair D Hay Y1 - 2022/11/22 UR - http://bjgp.org/content/early/2022/11/22/BJGP.2022.0229.abstract N2 - Background: There is a lack of evidence regarding post-consultation symptom trajectories for patients with respiratory tract infections (RTIs) and whether patient characteristics can be used to predict illness duration. Aim: To describe symptom trajectories in patients with RTIs, assess baseline characteristics and adverse events associated with trajectories. Design and setting: 9103 adults and children from 12 primary care studies. Method: Individual patient data latent class-informed regression analysis of randomised controlled trials and observational cohort studies. Post-consultation symptom trajectory (severity and duration), re-consultation with same or worsening illness and hospitalisation were assessed. Results: 90% of participants recovered from all symptoms by 28 days, regardless of antibiotic strategy. For studies of RTI with cough as a dominant symptom (n=5314), four trajectories were identified: ‘rapid[6]’ (90% of participants recovered within 6 days)’ in 52.0%; ‘intermediate[10]’ (28.9%); ‘slow progressive improvement[27]’ (12.5%); and ‘slow initial high symptom burden[27]’ (6.6%). Older age (OR: (95% CI): 2.57 (1.72-3.85), higher presenting illness baseline severity (OR) (95% CIs): 1.51 (1.12-2.03)); presence of lung disease (OR (95% CI): 1.78 (1.44-2.21)); above median illness duration prior to consultation (OR (95% CI): 1.99 (1.68-2.37)) were associated with slower recovery (>10 days) compared to faster recovery (≤10 days). Re-consultations and hospitalisations were respectively higher in those with slower recovery (ORs: 2.15 (1.78-2.60) and 7.42 (3.49-15.78)). Conclusion: Older patients presenting with more severe, longer pre-consultation symptoms, and chronic lung disease should be advised they are more likely to experience longer post-consultation illness durations, and that recovery rates are similar with and without antibiotics. ER -