Hypertension in BC Sawmill workers exposed to high noise levels Hind Sbihi, Hugh W. Davies, Paul A. Demers Outline 1. Background 2. Methods 3. Analyses 4. Results 5. Strengths and limitations 1. Background Noise and Cardiovascular Diseases Noise Æ Stress Æ “non auditory” health effects Cognitive processes Pathology Physiology • Hormonal and metabolic perturbations • Cardiovascular diseases: • Annoyance • Sleep disturbance • Chronic heart disease • Hypertension 1. Background Motivation • Burden of Hypertension. • Ubiquitous nature of noise. • Epidemiological evidence still inconsistent (van Kempen et al, 2000) Public health issue Limitations in methodology 2. Methods Population • Settings – Employed at least 1 year in one of 14 sawmills in BC – follow up: April 1991 to July 1998 • Characteristics – Linkage with Hospital discharge files and medical billing system Î 10832 subjects. • Race: Caucasian (89%) - South Asian (9%) - Chinese (2%) • Gender : 99% male – 1% female 2. Methods Definitions • Case definition – ICD 9 classification codes 401-405. – 3 doctor visits in a 2 month period and/or 1 visit at the hospital • Quantitative assessment of noise exposure: – 1900 noise dosimetry measurements Æ regression models to predict each unique mill/job/time combination (n=3700) (Davies et al., 1998) 2. Methods • Metrics for noise exposure – Cumulative exposure – Duration of exposure above thresholds of 85, 90, and 95 dB(A). 10 10 10* log *Tj Leqj j 1 n =∑⎡⎣⎢ ⎤ ⎦⎥ 3. Analyses • Life Table analyses (PC LTAS, NIOSH) – Apportion person years • Poisson regression – Lowest exposed group used as baseline – Only men included – Adjusted for race, age, and calendar year 4. Results – Hypertension in Males (n=828) 0.5 1.0 1.5 2.0 3.0 R e l a t i v e R i s k ( l o g s c a l e ) < 9 5 95 - 99 10 0 - 10 4 10 5 - 10 9 11 0- 11 5 >1 15 cumulative noise exposure (dBA*Year) • Risk of hypertension rises monotonically with increasing exposure • significant test for trend 4. Results – Hypertension incidence 0.5 1.0 1.5 2.0 3.0 R e l a t i v e R i s k ( l o g s c a l e ) < 3 3 - 9 10 - 1 9 20 - 2 9 > 2 9 < 3 3 - 9 10 - 1 9 >1 9 < 3 3 - 9 10 - 1 9 >1 9 Threshold > 85dB(A) Threshold >90 dB(A) Threshold >95 dB(A) Duration of Exposure (years) 5. Strengths • Methodology – Longitudinal design – Large cohort – Administrative Data • Exposure assessment 5. Limitations Individual risk factors smoking rates and lung cancer analyses Case definition Sensitivity analyses Conclusion • Noise is associated with an increased risk of hypertension when exposed to high levels and/or over a long period. • Results were robust across case definitions • South Asian (mostly Sikhs) had double the risk (RR=2.1 CI: 1.7 – 2.5) compared with Chinese (RR=1.3 CI: 0.8 – 2.0) and Caucasian subjects (baseline group). Acknowledgments • Weiwei Du • the Heart and Stroke Foundation