Zachary Terner1, Andrew Long2, Marta Reviriego-Mendoza2, John W Larkin2, Len A Usvyat2, Peter Kotanko3,4, Franklin W Maddux2, Yuedong Wang1. 1. Department of Statistics and Applied Probability, University of California-Santa Barbara, Santa Barbara, California. 2. Global Medical Office, Fresenius Medical Care, Waltham, Massachusetts. 3. Research Division, Renal Research Institute, New York, New York; and. 4. Icahn School of Medicine at Mount Sinai, Mount Sinai Health System, New York, New York.
Abstract
Background: All life on earth has adapted to the effects of changing seasons. The general and ESKD populations exhibit seasonal rhythms in physiology and outcomes. The ESKD population also shows secular trends over calendar time that can convolute the influences of seasonal variations. We conducted an analysis that simultaneously considered both seasonality and calendar time to isolate these trends for cardiovascular, nutrition, and inflammation markers. Methods: We used data from adult patients on hemodialysis (HD) in the United States from 2010 through 2014. An additive model accounted for variations over both calendar time and time on dialysis. Calendar time trends were decomposed into seasonal and secular trends. Bootstrap procedures and likelihood ratio methods tested if seasonal and secular variations exist. Results: We analyzed data from 354,176 patients on HD at 2436 clinics. Patients were 59±15 years old, 57% were men, and 61% had diabetes. Isolated average secular trends showed decreases in pre-HD systolic BP (pre-SBP) of 2.6 mm Hg (95% CI, 2.4 to 2.8) and interdialytic weight gain (IDWG) of 0.35 kg (95% CI, 0.33 to 0.36) yet increases in post-HD weight of 2.76 kg (95% CI, 2.58 to 2.97). We found independent seasonal variations of 3.3 mm Hg (95% CI, 3.1 to 3.5) for pre-SBP, 0.19 kg (95% CI, 0.17 to 0.20) for IDWG, and 0.62 kg (95% CI, 0.46 to 0.79) for post-HD weight as well as 0.12 L (95% CI, 0.11 to 0.14) for ultrafiltration volume, 0.41 ml/kg per hour (95% CI, 0.37 to 0.45) for ultrafiltration rates, and 3.30 (95% CI, 2.90 to 3.77) hospital days per patient year, which were higher in winter versus summer. Conclusions: Patients on HD show marked seasonal variability of key indicators. Secular trends indicate decreasing BP and IDWG and increasing post-HD weight. These methods will be of importance for independently determining seasonal and secular trends in future assessments of population health.
Background: All life on earth has adapted to the effects of changing seasons. The general and ESKD populations exhibit seasonal rhythms in physiology and outcomes. The ESKD population also shows secular trends over calendar time that can convolute the influences of seasonal variations. We conducted an analysis that simultaneously considered both seasonality and calendar time to isolate these trends for cardiovascular, nutrition, and inflammation markers. Methods: We used data from adult patients on hemodialysis (HD) in the United States from 2010 through 2014. An additive model accounted for variations over both calendar time and time on dialysis. Calendar time trends were decomposed into seasonal and secular trends. Bootstrap procedures and likelihood ratio methods tested if seasonal and secular variations exist. Results: We analyzed data from 354,176 patients on HD at 2436 clinics. Patients were 59±15 years old, 57% were men, and 61% had diabetes. Isolated average secular trends showed decreases in pre-HD systolic BP (pre-SBP) of 2.6 mm Hg (95% CI, 2.4 to 2.8) and interdialytic weight gain (IDWG) of 0.35 kg (95% CI, 0.33 to 0.36) yet increases in post-HD weight of 2.76 kg (95% CI, 2.58 to 2.97). We found independent seasonal variations of 3.3 mm Hg (95% CI, 3.1 to 3.5) for pre-SBP, 0.19 kg (95% CI, 0.17 to 0.20) for IDWG, and 0.62 kg (95% CI, 0.46 to 0.79) for post-HD weight as well as 0.12 L (95% CI, 0.11 to 0.14) for ultrafiltration volume, 0.41 ml/kg per hour (95% CI, 0.37 to 0.45) for ultrafiltration rates, and 3.30 (95% CI, 2.90 to 3.77) hospital days per patient year, which were higher in winter versus summer. Conclusions: Patients on HD show marked seasonal variability of key indicators. Secular trends indicate decreasing BP and IDWG and increasing post-HD weight. These methods will be of importance for independently determining seasonal and secular trends in future assessments of population health.
Authors: Jennifer L Nguyen; Wan Yang; Kazuhiko Ito; Thomas D Matte; Jeffrey Shaman; Patrick L Kinney Journal: JAMA Cardiol Date: 2016-06-01 Impact factor: 14.676
Authors: Charmaine E Lok; Jyothi R Thumma; Keith P McCullough; Brenda W Gillespie; Richard J Fluck; Mark R Marshall; Hideki Kawanishi; Bruce M Robinson; Ronald L Pisoni Journal: Semin Dial Date: 2013-09-25 Impact factor: 3.455
Authors: Jeroen P Kooman; Len A Usvyat; Marijke J E Dekker; Dugan W Maddux; Jochen G Raimann; Frank M van der Sande; Xiaoling Ye; Yuedong Wang; Peter Kotanko Journal: Blood Purif Date: 2018-11-16 Impact factor: 2.614
Authors: D Weber-Mzell; P Kotanko; A C Hauer; U Goriup; J Haas; N Lanner; W Erwa; I A Ahmaida; S Haitchi-Petnehazy; M Stenzel; G Lanzer; J Deutsch Journal: Eur J Clin Invest Date: 2004-03 Impact factor: 4.686
Authors: Adrian M Guinsburg; Len A Usvyat; Michael Etter; Xiaoqi Xu; Stephan Thijssen; Daniele Marcelli; Bernard Canaud; Cristina Marelli; Claudia Barth; Yuedong Wang; Paola Carioni; Frank M van der Sande; Peter Kotanko; Jeroen P Kooman Journal: BMC Nephrol Date: 2015-08-14 Impact factor: 2.388