M Leu Agelii1, S Lehtinen-Jacks2, H Zetterberg3, V Sundh4, C Björkelund5, L Lissner4. 1. Section for Epidemiology and Social Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. Electronic address: monica.leu@gu.se. 2. Section for Epidemiology and Social Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; Faculty of Social Sciences, Health Sciences, University of Tampere, Tampere, Finland. 3. Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölndal, Sweden; UCL Institute of Neurology, Queen Square, London, United Kingdom. 4. Section for Epidemiology and Social Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 5. Section for Primary Health Care, Institute of Medicine, University of Gothenburg, Gothenburg, Sweden.
Abstract
BACKGROUND AND AIMS: The impact of vitamin D concentrations on subsequent cardiovascular disease (CVD) and overall mortality has been generally examined for periods under two decades. The magnitude of the association may depend on follow-up length. We aimed to investigate the relationship between baseline vitamin D and risk of total CVD, stroke and all-cause mortality over three decades of follow-up. Secondly, we aimed to assess how follow-up affects the associations. METHODS AND RESULTS: Concentrations of 25-hydroxyvitamin D (25D) were measured in a population-based sample of 1227 middle-aged women using serum collected at baseline and categorized into low (lowest 25D quartile) vs high 25D status (upper three 25D quartiles). Hazard ratio (HR) of the endpoints was estimated for low 25D. The impact of follow-up was examined in intermediary analyses where follow-up was interrupted up to four times, each time decreasing it by five years. There were 596 cardiovascular events and 635 participants died. During the first 17 years, the low 25D group experienced a 29% higher CVD risk and 3.3-fold higher stroke risk after accounting for confounders. Longer follow-up diminished significantly these risks and 25D status had no contribution at 32 years. For mortality, the decline over time was less dramatic, with HR = 1.96 (1.25; 3.08) at 17 years and HR = 1.42 (1.17; 1.72) at 37 years. CONCLUSION: Low 25D status increased the risk for all endpoints, but a lengthy follow-up diminished these risks towards the null. The impact of follow-up depends on the outcome. Future studies of 25D and disease should use repeated 25D assessments.
BACKGROUND AND AIMS: The impact of vitamin D concentrations on subsequent cardiovascular disease (CVD) and overall mortality has been generally examined for periods under two decades. The magnitude of the association may depend on follow-up length. We aimed to investigate the relationship between baseline vitamin D and risk of total CVD, stroke and all-cause mortality over three decades of follow-up. Secondly, we aimed to assess how follow-up affects the associations. METHODS AND RESULTS: Concentrations of 25-hydroxyvitamin D (25D) were measured in a population-based sample of 1227 middle-aged women using serum collected at baseline and categorized into low (lowest 25D quartile) vs high 25D status (upper three 25D quartiles). Hazard ratio (HR) of the endpoints was estimated for low 25D. The impact of follow-up was examined in intermediary analyses where follow-up was interrupted up to four times, each time decreasing it by five years. There were 596 cardiovascular events and 635 participants died. During the first 17 years, the low 25D group experienced a 29% higher CVD risk and 3.3-fold higher stroke risk after accounting for confounders. Longer follow-up diminished significantly these risks and 25D status had no contribution at 32 years. For mortality, the decline over time was less dramatic, with HR = 1.96 (1.25; 3.08) at 17 years and HR = 1.42 (1.17; 1.72) at 37 years. CONCLUSION: Low 25D status increased the risk for all endpoints, but a lengthy follow-up diminished these risks towards the null. The impact of follow-up depends on the outcome. Future studies of 25D and disease should use repeated 25D assessments.
Authors: Alicia K Heath; Iris Y Kim; Allison M Hodge; Dallas R English; David C Muller Journal: Int J Environ Res Public Health Date: 2019-01-29 Impact factor: 3.390