Notes
Aims:
Few risk factors for rotator cuff disease (RCD) and corresponding treatment have been firmly established. The aim of this study was to evaluate the relationship between numerous risk factors and the incidence of surgery for RCD in a large cohort.
Methods:
A population-based cohort of people aged between 40 and 69 years in the UK (the UK Biobank) was studied. People who underwent surgery for RCD were identified through a link with NHS inpatient records covering a mean of eight years after enrolment. Multivariate Cox proportional hazards regression was used to calculate hazard ratios (HRs) as estimates of associations with surgery for RCD accounting for confounders. The risk factors which were considered included age, sex, race, education, Townsend deprivation index, body mass index (BMI), occupational demands, and exposure to smoking.
Results:
Of the 421,894 people who were included, 47% were male. The mean age at the time of enrolment was 56 years (40 to 69). A total of 2,156 people were identified who underwent surgery for RCD. Each decade increase in age was associated with a 55% increase in the incidence of RCD surgery (95% confidence interval (CI) 46% to 64%). Male sex, non-white race, lower deprivation score, and higher BMI were significantly associated with a higher risk of surgery for RCD (all p < 0.050). Greater occupational physical demands were significantly associated with higher rates of RCD surgery (HR = 2.1, 1.8, and 1.4 for always , usually , and sometimes doing heavy manual labour vs never , all p < 0.001). Former smokers had significantly higher rates of RCD surgery than those who had never smoked (HR 1.23 (95% CI 1.12 to 1.35), p < 0.001), while current smokers had similar rates to those who had never smoked (HR 0.94 (95% CI 0.80 to 1.11)). Among those who had never smoked, the risk of surgery was higher among those with more than one household member who smoked (HR 1.78 (95% CI 1.08 to 2.92)). The risk of RCD surgery was not significantly related to other measurements of secondhand smoking.
Conclusion:
Many factors were independently associated with surgery for RCD, including older age, male sex, higher BMI, lower deprivation score, and higher occupational physical demands. Several of the risk factors which were identified are modifiable, suggesting that the healthcare burden of RCD might be reduced through the pursuit of public health goals, such as reducing obesity and modifying occupational demands.
Application 27034
Evaluation of genetic and non-genetic risk factors for degenerative rotator cuff disease.
Aim 1: We will test associations of genetic variants with degenerative rotator cuff disease risk through a genome-wide association study (GWAS).
Aim 2: We will estimate associations of non-genetic risk factors, such as age, smoking, and occupational upper extremity demands, and build a predictive model of degenerative rotator cuff disease including all risk factors.
Aim 3: We will evaluate interactions between genetic and non-genetic risk factors for degenerative rotator cuff disease.
There is growing evidence that familial predisposition increases rotator cuff disease risk. However, few studies have evaluated associations between specific genes and rotator cuff disease. The only prior GWAS had limited statistical power with <350 rotator cuff disease patients. Furthermore, the influence of genetic risk on effects of non-genetic risk factors is unknown. Results from this study are expected to be useful in identifying individuals who are at high-risk for rotator cuff disease. These individuals might benefit from preventative strategies and early treatment of rotator cuff tears and interventions aimed at modifiable risk factors, such as smoking cessation. Individuals with rotator cuff disease will be identified using hospital ICD-9/10 codes and compared to individuals without rotator cuff disease. We will conduct a GWAS to identify genetic markers significantly associated with rotator cuff disease after adjusting for multiple comparisons. Non-genetic risk factor (ex. age, occupational demand, and smoking) associations with rotator cuff disease will be estimated. Prediction models will be developed. To evaluate interactions between genetic and non-genetic factors, we will test product interaction terms in our models and evaluate risk within strata defined by genetic and non-genetic risk factors. We propose to use the full cohort with genetic information.
Lead investigator: | Dr Elizabeth Yanik |
Lead institution: | Washington University in St. Louis |