IPD Meta-Analysis of Cohort Studies

Overview:

Osteoarthritis (OA) is one of the most common causes of disability in the world. The prevention and management of OA is dependent on the understanding of modifiable risk factors for OA in the population at earlier stages of disease.

To fully understand the risk factors for OA as well as its long-term effects, there is a need to combine data from population-based cohorts to provide sufficient statistical power. Traditional meta-analyses on OA rely on aggregate data obtained from study publications. These are vulnerable to outcome reporting and publication bias, and the quality and availability of data may vary across studies. An increasingly popular alternative to traditional meta-analysis is individual participant (IPD) meta-analysis, which utilises original raw data for the analysis. The key benefits of this type of analysis are the ability better to harmonise primary risk factors and outcomes between studies, the adjustment of identical confounders, the application of consistent inclusion and exclusion criteria, and the ability to include previously unpublished datasets into the analysis.

Aims and Objectives:

• To establish the methodology for harmonising physical activity risk factors and osteoarthritis outcomes between prospective population-based cohort studies
• To undertake three large International IPD meta-analysis studies to:
  • Quantify the association between physical activity and incident knee osteoarthritis in an individual patient data meta-analysis 
  • Quantify the association between occupation and incident knee osteoarthritis in an individual patient data meta-analysis
  • Assess the risk of premature mortality in individuals with lower-limb osteoarthritis in an individual patient data meta-analysis
• To establish collaborative links with international cohorts to develop osteoarthritis-based research questions with increased generalizability across subjects from different countries and races 

Key Findings:

• Using data from six observational studies of knee OA and pooling over 5000 participants, our findings suggest that whole-body, physiological energy expenditure during recreational activities was not associated with incident knee OA. Similarly, time spent in light or moderate intensity leisure activities was not associated with incident knee OA. A weak association between increasing time spent in vigorous leisure activity and knee Pain and Radiographic Osteoarthritis (PROA) was found, and therefore this warrants further investigation with clear disaggregation between all components of physical activity (PA).
• Data for 7391 participants were included in the occupation and knee OA IPD meta-analysis. There was more than a two-fold increase in the odds of having Radiographic Knee Osteoarthritis (RKOA), knee pain and symptomatic RKOA in those with heavy manual compared to sedentary occupations ((odds ratio (OR): 2.14; 95% confidence interval (CI): 1.79, 2.58), (OR: 2.19; 95% CI: 1.78, 2.70), (OR: 2.41; 95% CI: 1.94, 2.99) respectively). This large international multi-cohort study demonstrated an association of heavy manual work with radiographic knee osteoarthritis, symptomatic radiographic knee osteoarthritis and knee pain.  These findings suggest that measures that protect workers and are designed to reduce heavy manual related activities remain a priority to reduce the risk of knee OA. Findings also suggest that occupation is an important consideration in the relationship between physical activity and knee OA, and where possible should be considered within analysis. 
• 10,723 participants in six cohorts from four countries were included in the OA and mortality IPD meta-analysis. Multivariable models (adjusting for age, sex, race, BMI, smoking, alcohol consumption, cardiovascular disease and diabetes) showed a pooled HR, compared to pain and ROA-free participants, of 1·03 (0·83, 1·28) for ROA, 1·35 (1·12, 1·63) for POA, and 1·37 (1·22, 1·54) for PROA. Participants with painful knee OA or painful radiographic had a 35 to 37% increased association with reduced time-to-mortality, independent of confounders. Radiographic OA showed no association with mortality, suggesting that OA-related knee pain may be driving the association with time-to-mortality.

Outputs:

• Leyland, KM, Gates, LS, Nevitt, M, Felson, D, Bierma-Zeinstra, SM, Conaghan, PG, Engebretson, L, Hochberg, Hunter, DJ, Jones, G, Jordan JM, Judge A, Lohmander LS, Roos EM, Sanchez-Santos MT, Yoshimura N, van Meurs JBJ, Batt ME, Newton J, Cooper C, Arden NK. Harmonising measures of knee and hip osteoarthritis in population-based cohort studies: an international study. Osteoarthritis and Cartilage. July 2018; 26 (7) 872-879. Doi: 10.1016/j.joca.2018.01.024
• Leyland, K. M., Gates, L., Sanchez-Santos, M. T., Nevitt, M., Felson, D., Jones, G., Jordan, J. M., Judge, A., Alhambra, D. P., Yoshimura, N., Newton, J., Callahan, L. F., Cooper, C., Batt, M., Lin, J., Liu, Q., Cleveland, R., Collins, G. & Arden, N. K. Knee osteoarthritis and time-to all-cause mortality in six community-based cohorts: an international meta-analysis of individual participant-level data. Aging Clinical and Experimental Research. February 2021.  DOI: 10.1007/s40520-020-01762-2
• Parsons, C., Gates, L., Perry, T., Nevitt, M., Felson, D., Sanchez-Santos., M., Jones, G., Golightly, Y., Nelson, A., Walker-Bone, K., Cooper, C., Arden.  The association between predominant lifetime occupation and risk of knee osteoarthritis: an international cohort collaboration. Osteoarthritis and Cartilage (2020). doi.org/10.1016/j.ocarto.2020.100085
• Perry TA, Wang X, Gates L, Parsons CM, Sanchez-Santos MT, Garriga C, Cooper C, Nevitt MC, Hunter DJ, Arden NK. Occupation and risk of knee osteoarthritis and knee replacement: A longitudinal, multiple-cohort study. Semin Arthritis Rheum. 2020 Oct;50(5):1006-1014. doi: 10.1016/j.semarthrit.2020.08.003. Epub 2020 Aug 8. PMID: 33007601.

Work Package	Early Disease and Risk Prediction: Prevent
Objective	1.1, 1.2i, 1.2ii
Lead	Nigel Arden/Lucy Gates
Investigators	Camilla Parsons, Maria Sanchez-Santos, Thomas Perry
Institutions	University of Oxford/ University of Southampton