Arch Osteoporos (2014) 9:191 DOI 10.1007/s11657-014-0191-2
ORIGINAL ARTICLE
Impact of comorbidity, age, and gender on seasonal variation in hip fracture incidence. A NOREPOS study Siri M. Solbakken & Jeanette H. Magnus & Haakon E. Meyer & Nina Emaus & Grethe S. Tell & Kristin Holvik & Guri Grimnes & Siri Forsmo & Berit Schei & Anne Johanne Søgaard & Tone K. Omsland
Received: 7 March 2014 / Accepted: 28 July 2014 # International Osteoporosis Foundation and National Osteoporosis Foundation 2014
Abstract Summary Based on a total of 136,140 hip fractures, we found a distinct seasonal variation in hip fracture incidence present in subgroups defined by age, gender, and comorbidity. The seasonal variation was most pronounced in the youngest and the healthiest patients. Purpose The purpose of this study was to examine the possible seasonal variation in hip fracture incidence in Norway by comorbidity, age, and gender. Methods Data were retrieved from the NOREPOS Hip Fracture Database containing all hip fractures in Norway during the time period 1994–2008. Hip fractures were identified by computerized hospital discharge diagnoses. Charlson comorbidity index was calculated based on additional diagnoses and categorized (0, 1, and ≥2). Summer was defined as June, July, and August and winter as December, January, and
February. Incidence rate ratios for hip fracture according to season were calculated by negative binomial models. Results In patients aged 50–103 years, 136,140 eligible fractures were identified (72.5 % women). The relative risk of hip fracture in winter versus summer was 1.40 (95 % confidence interval (CI) 1.36–1.45) in men and 1.26 (95 % CI 1.23–1.28) in women. June had the lowest number of fractures in both genders. We found seasonal variation in all subgroups by age and gender, although least pronounced in patients >79 years. There was a significant interaction between season and comorbidity (p=0.022). When comparing winter to summer, we found relative risks of 1.40 (95 % CI 1.31–1.50) in patients with Charlson index=0, 1.29 (95 % CI 1.19–1.40) in patients with Charlson index=1, and 1.18 (95 % CI 1.08–1.28) in patients with Charlson index ≥2.
The NORwegian EPidemiologic Osteoporosis Studies (NOREPOS) is a collaboration between epidemiologic osteoporosis studies, which are substudies within large population-based surveys in four districts of Norway (Tromsø, Nord-Trøndelag, Hordaland, Oslo). The NOREPOS Hip Fracture Database includes all hospitalizations for hip fracture in Norway during the time period of 1994–2008. S. M. Solbakken (*) : J. H. Magnus : H. E. Meyer : T. K. Omsland Institute of Health and Society, Department of Community Medicine, University of Oslo, PO Box 1130, Blindern 0318, Oslo, Norway e-mail: [email protected] H. E. Meyer : K. Holvik : A. J. Søgaard Division of Epidemiology, Norwegian Institute of Public Health, Oslo, Norway N. Emaus Department of Health and Care Sciences, UiT The Arctic University of Norway, Tromsø, Norway G. S. Tell : K. Holvik Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway
G. Grimnes Department of Clinical Medicine, UiT The Arctic University of Norway, Tromsø, Norway
S. Forsmo : B. Schei Institute of Public Health and General Practice, Faculty of Medicine, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
B. Schei Department of Obstetrics and Gynecology, St. Olav’s University Hospital, Trondheim, Norway
191, Page 2 of 7
Conclusions There was a distinct seasonal variation in hip fracture incidence, present in all subgroups of gender, age, or comorbidity. This variation should be accounted for when planning health-care services. Keywords Hip fracture . Incidence . Seasonal variation . Comorbidity . Age . Gender
Arch Osteoporos (2014) 9:191
were also retrieved, together with the dates of admission and discharge. Hip fractures were identified according to the International Classification of Diseases, Ninth Revision (ICD-9): 820 with all subgroups and Tenth Revision (ICD10): S72.0–S72.2. Thus, both cervical, trochanteric and subtrochanteric fractures were included. Validation
Introduction The considerable differences in hip fracture risk between countries are well described, with more than tenfold variation in hip fracture rates worldwide [1]. Even though the incidence rates of hip fracture in Norway have decreased during the last decade [2], the country still has one of the highest risks of hip fracture in the world [1]. Whether hip fracture risk varies by season has yet to be clarified. In studies reporting differences in hip fracture rates between summer and winter [3–12], potential explanations are not much explored. Furthermore, the link of this seasonal variation to comorbidity is unknown. Although winter peaks and summer troughs in hip fracture risk have been described [3–12], not all studies have found seasonal variation [13, 14], and findings from Norway are disparate. Two studies reported higher incidence during winter [15, 16], whereas others did not report seasonal variation in every year examined [17] or found variation in outdoor fractures only [18]. Seasonal variation by age and gender is examined in some of the studies, but with dissimilar results [4–6, 8–11, 15]. Hip fractures are associated with both mortality [19] and functional decline [20] and demand considerable resources from the healthcare system [21]. A large-scale study could help determine the extent of a possible seasonal variation, and thereby provide important information when planning health-care services. The purpose of this study was to examine a possible seasonal variation in hip fracture incidence in Norway by comorbidity score, age, and gender.
In order to validate the data at an individual level, NORHip was compared to local fracture registers verified by medical records or x-ray in Oslo and Tromsø. The combined Cohen’s kappa for the comparisons was 0.95 [2]. Charlson comorbidity index A modified Charlson index was used to classify patients due to the severity and number of comorbidities, based on a total score calculated from additional diagnosis codes recorded during the hospitalization for hip fracture [22, 23]. The total score was used to categorize patients into three groups: 0=a person without any of the additional diagnosis codes, 1= patients with a score of 1 indicating some comorbidity, and 2=patients with a total score of ≥2 representing severe comorbidity. Due to change of coding system from ICD-9 to ICD-10 in 1999, the Charlson index was based on ICD-10 diagnosis codes only in order to ensure comparability of obtained categories. As information on comorbidities was available for the hip fracture patients only, we calculated the person time from 1999 until fracture (by month) to estimate the correct population time at risk in each of the three Charlson index subgroups. We included only each individual’s first recorded hip fracture during 1999–2008 in the comorbidity analyses, due to dependency between the first and second hip fracture and because the Charlson index in an individual patient may change over time. A total of 81,849 fractures were included in the comorbidity analyses. Statistical methods
Materials and methods Data collection Data were retrieved from the NOREPOS Hip Fracture Database (NORHip), containing hospital-treated hip fractures in Norway during 15 years (1 January 1994 to 31 December 2008). The data collection, classification, validation, and quality assurance of this database have previously been published as supplementary material [2]. Briefly, a system developed by the Norwegian Knowledge Centre for the Health Services was used to identify hip fractures, based on computerized discharge diagnoses. Additional diagnosis codes and surgical procedure codes during the hospitalization for hip fracture
The analyses by gender and age included 136,140 hip fractures (first or second) sustained in men and women aged 50– 103 years during 1994–2008. Data on the total Norwegian population was acquired from Statistics Norway [24]. The mid-year population of each 1-year age group ≥50 and ≤103 years was defined as the population at risk. Data analysis was performed in Stata MP13. Because our data were overdispersed, using a Poisson regression model might have underestimated the standard errors [25]. We therefore used a negative binomial model to estimate the incidence rate ratios. Winter was defined as December, January, and February; spring as March, April, and May; summer as June, July, and August; and autumn as September, October, and November.
Arch Osteoporos (2014) 9:191
Page 3 of 7, 191 4500
10.6% 9.7% 9.4%
Total number of hip fractures
4000 3500
10.2% 7.7% 7.6%
3000
8.4% 6.9% 7.3%
7.6%
7.0%
7.5%
2500 2000 1500 1000 500 0
a 12000 Total number of hip fractures
We calculated the age-adjusted incidence rate ratio for each season in both men and women. Statistical interactions between season and gender, between age and month, and between season and Charlson index were tested. Incidence rate ratios were stratified on age, gender, and comorbidity. We also performed stratified analyses on age and gender within the Charlson index subgroups. To further examine a possible cyclical pattern of hip fractures, a gender stratified time series model (Cosinor analysis) was applied using day of the year as analytical unit. The fit of the model was tested. As Norway has a latitude ranging from approximately 58 to 71° N, the country was divided in a northern and southern region at approximately 64° N (Nord-Trøndelag). We tested for statistical interaction between season and region of residence and stratified by region. Concerning person time in the analyses by region, the same person time calculation as in the analyses by Charlson index was used.
Ethics
10000 8000
9.8% 8.8% 8.8%
9.9% 7.7% 8.0%
7.2%
7.8% 7.8% 7.6%
8.2% 8.4%
6000 4000 2000 0
The Regional Committee for Medical and Health Research Ethics, the Norwegian Data Protection Authority, The Directorate of Health, the National Population Register, and Statistics Norway have approved the study and the linkages of data.
b Fig. 1 Distribution of hip fractures by month (Norwegian patients aged 50–103 years during 1994–2008). Total numbers and percentages in a men and b women. Norwegian Epidemiologic Osteoporosis Studies (NOREPOS)
Seasonal variation by gender Results Distribution of hip fractures by month The total number of hip fractures by gender and age group can be found in Table 1. Of the 136,140 hip fractures, 72.5 % were sustained in women. Mean age at first fracture was 78.5 years in men and 81.2 years in women. Figure 1a, b displays the monthly distribution of fractures by gender. January and December were the peak months, whereas June had the lowest proportion of fractures (6.9 % in men and 7.2 % in women). Table 1 Total number of hip fractures by gender and age group (Norwegian patients aged 50–103 years during 1994–2008). A NOREPOSa study
50–64 years 65–79 years >79 years a
Men (%)
Women (%)
4,120 (11.0) 13,245 (35.4) 20,061 (53.6)
5,344 (5.4) 29,033 (29.4) 64,337 (65.2)
Norwegian Epidemiologic Osteoporosis Studies
The highest number of hip fractures occurred during winter, and compared to summer, the risk of hip fracture was 40 % (95 % confidence interval (CI) 36–45) higher in men and 26 % (95 % CI 23–28) higher in women (Table 2). There was a significant interaction between gender and season (p79 years
8
0.4
6
0.2
4
0
Fig. 3 Incidence rate ratios of hip fractures by month (Norwegian patients aged 50–103 years during 1994–2008) in a men and b women. Norwegian Epidemiologic Osteoporosis Studies (NOREPOS)
2
Mean daily number of hip fractures
a
1
0
Seasonal variation by age
100
200
300
Day of year (starting at January 1st) Cosinor model
Hip fractures
a
However, the youngest age group had the largest seasonal variation in both men and women.
10
15
20
25
Seasonal variation by comorbidity
5
Mean daily number of hip fractures
1.2
100
200
300
Day of year (starting at January 1st) Cosinor model
Hip fractures
b
Fig. 2 Mean daily number of hip fractures during 1994–2008 (Norwegian patients aged 50–103 years) in a men and b women. A time series model (Cosinor model) is fitted. Norwegian Epidemiologic Osteoporosis Studies (NOREPOS)
Comorbidity analyses were based on the three subgroups corresponding to Charlson index 0 (46,575 fractures), 1 (21,787 fractures), and ≥2 (13,487 fractures), including first fractures during 1999–2008. After adjustment for age and gender, we found interaction between season and comorbidity (p=0.022). When comparing winter to summer, the higher fracture risk varied from 40 % in the most healthy patients to 18 % in patients with highest comorbidity score (Table 3). In the analysis stratified according to age group and Charlson index, the seasonal variation was most pronounced among subjects younger than 65 years in the healthy subgroup (Charlson index=0) (IRRwinter versus summer =1.67; 95 % CI= 1.54–1.82), while the patients >79 years of age with highest comorbidity (Charlson index ≥2) showed less seasonal variation (IRRwinter versus summer =1.14; 95 % CI=1.08–1.21) (adjusted for age and gender).
Arch Osteoporos (2014) 9:191
Page 5 of 7, 191
Table 3 Number of first hip fractures and incidence rate ratios in winter versus summer by Charlson comorbidity index (1999–2008). A NOREPOSa study Charlson index N (%) 0 1 ≥2 a
IRRb winter versus summer 95 % CI
46,575 (56.9) 1.40 21,787 (26.6) 1.29 13,487 (16.5) 1.18
1.31–1.50 1.19–1.40 1.08–1.28
Norwegian Epidemiologic Osteoporosis Studies
b
Winter (December, January, February); summer (June, July, August). Adjusted for age and gender
Discussion Based on a total of 136,140 hip fractures among patients aged 50–103 years, we demonstrated an overall increased fracture risk in winter compared to summer. The relative risk in winter versus summer was 1.40 in men and 1.26 in women. June had the lowest number of fractures in both genders. The seasonal variation occurred in all subgroups by age and gender, but less distinct with increasing age and extent of comorbidity. Several studies have reported seasonal variation in hip fracture incidence, but a Swedish and a Norwegian study were inconclusive [13, 17]. These studies were both based on data from individual years, making them vulnerable to year by year fluctuations. A study from Harstad, Norway, [18] found seasonal variation in outdoor fractures only, and an Italian study based on 4 years with a total of 8,978 cases reported no seasonal variation [14]. Nevertheless, our findings were consistent with results from several other studies. In addition to previous studies in Norway [15, 16], increased hip fracture rates during winter time have been reported from Spain, USA, Canada, UK, Scotland, New Zealand, Hong Kong, and Taiwan [3–12]. In agreement with our findings, the largest seasonal variations have been demonstrated in men [4, 8, 10] and in younger age groups [4, 8–10, 15]. Some studies have reported higher hip fracture risk during the winter season in older age groups [11] and higher mean age of the patients in this season [5], possibly owing to decreased mobility due to many layers of clothing, inadequate home safety precautions, decreased capacity of synthesizing vitamin D and higher vitamin D requirements to maintain bone density in the elderly. However, a British study found no differences in hip fracture patient characteristics between summer and winter [6]. Climate has been suggested as a causal factor for the seasonal variation, due to outdoor conditions such as ice and slippery pavements [26]. The association between hip fractures and season-related variables such as snow, freezing rain, and cold temperatures has been described [5, 9, 10, 12]. However, several studies have found seasonal variation at
temperate and even subtropical climates [4, 7, 8, 11]. Furthermore, the majority of fractures occur indoors [13, 18], thus reducing the impact of ice and snow. Certain environmental factors will nevertheless affect both outdoor and indoor conditions, such as temperature and number of daylight hours. These aspects might be important when trying to explain the increased fracture risk in the winter months. Lower temperatures in winter might affect conditions indoors as well as outdoors with the potential to cause hypothermia, impairment of coordination, and thereby lead to falls with hip fracture as a consequence [27]. Moreover, the number of daylight hours in Norway is greatly reduced in winter and even absent in the northern part of the country. The lack of sunshine exposure during winter will abolish cutaneous vitamin D synthesis, and seasonal variation in vitamin D levels and bone turnover markers has been found in several studies [28–32]. Low vitamin D levels are associated with increased risk of hip fractures [33–35] related to both bone loss [36] and impaired muscle function [36–38] and thereby increased risk of falling [39, 40]. This theory is supported by our findings, demonstrating the largest seasonal variation in the northern part of the country. On the other hand, the north-south difference could also be caused by climatic differences (more snow, ice, and lower temperatures in the north). In addition, impaired vision due to winter darkness might affect coordination and contribute to the seasonal variation in hip fracture incidence. To our knowledge, the association between comorbidity and seasonal variation has not been described previously. When analyzing subgroups by Charlson index, we found the greatest seasonal variation in the healthiest group. Men 50–64 years in this group displayed the greatest differences between summer and winter. In contrast, hip fractures in patients >79 years of age in the group with most comorbidity showed less seasonal variation. Dissimilarities in activity level and exposure to environmental factors might explain the differences. Healthy subjects in the youngest age group are more likely physically active, which might include outdoor activities related to risk of injuries such as winter sports. Consequently, they might be more vulnerable to environmental hazards like snow and ice. Most probably, their counterparts with highest comorbidity >79 years of age are much more inactive, institutionalized and sometimes immobilized, and will thus be affected by seasonal factors to a lesser extent. Results from the Harstad study support this theory, reporting higher mean age at fracture in those occurring indoors compared to outdoors [18]. Further research on the place and type of injury in different age groups could be useful when considering preventive measures.
191, Page 6 of 7
Strengths and limitations The strengths of this study include the large number of cases based on a validated hip fracture database. All hospital-treated hip fractures in Norway in a 15-year period were included in the analyses by gender and age (the comorbidity analyses were restricted to a 10-year period). Using Charlson index [23] to measure comorbidity has its limitations. The registration of additional diagnosis codes might have been incomplete or missing and was based on the information at the time of the hip fracture hospitalization only. Health professionals are supposed to record diagnoses relevant for the hip fracture hospitalization. Unfortunately, complete data on comorbidity is therefore lacking. However, diagnoses considered to be irrelevant are perhaps not important when assessing the current health status in the patient at the time of fracture.
Conclusions Based on a total of 136,140 fractures, we found a considerably higher risk of hip fracture in winter versus summer, present in all subgroups defined by gender, age, or degree of comorbidity. The seasonal variation was most pronounced in men, in the younger age groups, and in the healthiest with least comorbidity. The substantial seasonal variation should be accounted for when planning health-care services.
Conflict of interest None.
References 1. Kanis JA, Oden A, McCloskey EV, Johansson H, Wahl DA, Cooper C, Epidemiology IOFWGo, Quality of L (2012) A systematic review of hip fracture incidence and probability of fracture worldwide. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 23(9):2239–2256 2. Omsland TK, Holvik K, Meyer HE et al (2012) Hip fractures in Norway 1999–2008: time trends in total incidence and second hip fracture rates: a NOREPOS study. Eur J Epidemiol 27:807–814 3. Alvarez-Nebreda ML, Jiménez AB, Rodríguez P, Serra JA (2008) Epidemiology of hip fracture in the elderly in Spain. Bone 42:278– 285 4. Bischoff-Ferrari HA, Orav JE, Barrett JA, Baron JA (2007) Effect of seasonality and weather on fracture risk in individuals 65 years and older. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 18:1225–1233 5. Chiu KY, Ng TP, Chow SP (1996) Seasonal variation of fractures of the hip in elderly persons. Injury 27:333–336 6. Crawford JR, Parker MJ (2003) Seasonal variation of proximal femoral fractures in the United Kingdom. Injury 34:223–225 7. Douglas S, Bunyan A, Chiu KH, Twaddle B, Maffulli N (2000) Seasonal variation of hip fracture at three latitudes. Injury 31:11–19
Arch Osteoporos (2014) 9:191 8. Jacobsen SJ, Goldberg J, Miles TP, Brody JA, Stiers W, Rimm AA (1991) Seasonal variation in the incidence of hip fracture among white persons aged 65 years and older in the United States, 1984– 1987. Am J Epidemiol 133:996–1004 9. Jacobsen SJ, Sargent DJ, Atkinson EJ, O'Fallon WM, Melton LJ 3rd (1995) Population-based study of the contribution of weather to hip fracture seasonality. Am J Epidemiol 141:79–83 10. Levy AR, Bensimon DR, Mayo NE, Leighton HG (1998) Inclement weather and the risk of hip fracture. Epidemiology 9:172–177 11. Lin HC, Xiraxagar S (2006) Seasonality of hip fractures and estimates of season-attributable effects: a multivariate ARIMA analysis of population-based data. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 17:795–806 12. Modarres R, Ouarda TBMJ, Vanasse A, Orzanco MG, Gosselin P (2012) Modeling seasonal variation of hip fracture in Montreal, Canada. Bone 50:909–916 13. Leavy B, Aberg AC, Melhus H, Mallmin H, Michaelsson K, Byberg L (2013) When and where do hip fractures occur? A population-based study. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 24:2387–2396 14. Pedrazzoni M, Alfano FS, Malvi C, Ostanello F, Passeri M (1993) Seasonal variation in the incidence of hip fractures in EmiliaRomagna and Parma. Bone 14(Supplement 1):57–63 15. Gronskag AB, Forsmo S, Romundstad P, Langhammer A, Schei B (2010) Incidence and seasonal variation in hip fracture incidence among elderly women in Norway. The HUNT Study. Bone 46: 1294–1298 16. Diamantopoulos AP, Rohde G, Johnsrud I, Skoie IM, Johnsen V, Hochberg M, Haugeberg G (2012) Incidence rates of fragility hip fracture in middle-aged and elderly men and women in southern Norway. Age Ageing 41:86–92 17. Lofthus CM, Osnes EK, Falch JA, Kaastad TS, Kristiansen IS, Nordsletten L, Stensvold I, Meyer HE (2001) Epidemiology of hip fractures in Oslo, Norway. Bone 29:413–418 18. Emaus N, Olsen LR, Ahmed LA, Balteskard L, Jacobsen BK, Magnus T, Ytterstad B (2011) Hip fractures in a city in Northern Norway over 15 years: time trends, seasonal variation and mortality: the Harstad Injury Prevention Study. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 22: 2603–2610 19. Haentjens P, Magaziner J, Colón-Emeric CS, Vanderschueren D, Milisen K, Velkeniers B, Boonen S (2010) Meta-analysis: excess mortality after hip fracture among older women and men. Ann Intern Med 152:380–390 20. Bentler SE, Liu L, Obrizan M et al (2009) The aftermath of hip fracture: discharge placement, functional status change, and mortality. Am J Epidemiol 170:1290–1299 21. Haentjens P, Lamraski G, Boonen S (2005) Costs and consequences of hip fracture occurrence in old age: an economic perspective. Disabil Rehabil 27:1129–1141 22. Stagg V (2006) CHARLSON: Stata module to calculate Charlson index of comorbidity. http://ideas.repec.org/c/boc/bocode/s456719. html#download. Accessed January 14th 2014 23. Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 40:373–383 24. (2013) Table: 07459: Population, by sex and one-year age groups. Statistics Norway, https://www.ssb.no/statistikkbanken/selectvarval/ Define.asp?subjectcode=&ProductId=&MainTable=NY3026&nvl= & P L a n g u a g e = 1 & n y T m p Va r = t r u e & C M S S u b j e c t A r e a = befolkning&KortNavnWeb=folkemengde&StatVariant=&checked= true. Accessed January 7th 2014 25. Ismail N, Jemain AA (2007) Handling overdispersion with negative binomial and generalized Poisson regression models. Casualty Actuarial Society Forum
Arch Osteoporos (2014) 9:191 26. Ralis ZA (1981) Epidemic of fractures during period of snow and ice. Br Med J 282:603–605 27. Bastow MD, Rawlings J, Allison SP (1983) Undernutrition, hypothermia, and injury in elderly women with fractured femur: an injury response to altered metabolism? Lancet 1:143–146 28. Darling AL, Hart KH, Gibbs MA et al (2013) Greater seasonal cycling of 25-hydroxyvitamin D is associated with increased parathyroid hormone and bone resorption. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 25(3): 933–941 29. Klenk J, Rapp K, Denkinger MD, Nagel G, Nikolaus T, Peter R, Koenig W, Bohm BO, Rothenbacher D (2013) Seasonality of vitamin D status in older people in Southern Germany: implications for assessment. Age Ageing 42:404–408 30. Nurmi I, Kaukonen JP, Luthje P, Naboulsi H, Tanninen S, Kataja M, Kallio ML, Leppilampi M (2005) Half of the patients with an acute hip fracture suffer from hypovitaminosis D: a prospective study in southeastern Finland. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 16:2018–2024 31. Pasco JA, Henry MJ, Kotowicz MA, Sanders KM, Seeman E, Pasco JR, Schneider HG, Nicholson GC (2004) Seasonal periodicity of serum vitamin D and parathyroid hormone, bone resorption, and fractures: the Geelong Osteoporosis Study. J Bone Min Res : Off J Am Soc Bone Min Res 19:752–758 32. Midtby M, Magnus JH, Joakimsen RM (2001) The Tromso Study: a population-based study on the variation in bone formation markers with age, gender, anthropometry and season in both men and women. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 12:835–843
Page 7 of 7, 191 33. Bakhtiyarova S, Lesnyak O, Kyznesova N, Blankenstein MA, Lips P (2006) Vitamin D status among patients with hip fracture and elderly control subjects in Yekaterinburg, Russia. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 17:441–446 34. Holvik K, Ahmed LA, Forsmo S, Gjesdal CG, Grimnes G, Samuelsen SO, Schei B, Blomhoff R, Tell GS, Meyer HE (2013) Low serum levels of 25-hydroxyvitamin D predict hip fracture in the elderly: a NOREPOS study. J Clin Endocrinol Metab 98:3341–3350 35. Sakuma M, Endo N, Oinuma T, Hayami T, Endo E, Yazawa T, Watanabe K, Watanabe S (2006) Vitamin D and intact PTH status in patients with hip fracture. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 17:1608–1614 36. Lips P (2001) Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocr Rev 22:477–501 37. Ceglia L (2008) Vitamin D and skeletal muscle tissue and function. Mol Asp Med 29:407–414 38. Visser M, Deeg DJH, Lips P (2003) Low vitamin d and high parathyroid hormone levels as determinants of loss of muscle strength and muscle mass (sarcopenia): the Longitudinal Aging Study Amsterdam. J Clin Endocrinol Metab 88:5766–5772 39. Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, Orav JE, Stuck AE, Theiler R, Wong JB, Egli A, Kiel DP, Henschkowski J (2009) Fall prevention with supplemental and active forms of vitamin D: a meta-analysis of randomised controlled trials. BMJ 339:b3692 40. Murad MH, Elamin KB, Abu Elnour NO et al (2011) Clinical review: the effect of vitamin D on falls: a systematic review and metaanalysis. J Clin Endocrinol Metab 96:2997–3006
ORIGINAL ARTICLE
Impact of comorbidity, age, and gender on seasonal variation in hip fracture incidence. A NOREPOS study Siri M. Solbakken & Jeanette H. Magnus & Haakon E. Meyer & Nina Emaus & Grethe S. Tell & Kristin Holvik & Guri Grimnes & Siri Forsmo & Berit Schei & Anne Johanne Søgaard & Tone K. Omsland
Received: 7 March 2014 / Accepted: 28 July 2014 # International Osteoporosis Foundation and National Osteoporosis Foundation 2014
Abstract Summary Based on a total of 136,140 hip fractures, we found a distinct seasonal variation in hip fracture incidence present in subgroups defined by age, gender, and comorbidity. The seasonal variation was most pronounced in the youngest and the healthiest patients. Purpose The purpose of this study was to examine the possible seasonal variation in hip fracture incidence in Norway by comorbidity, age, and gender. Methods Data were retrieved from the NOREPOS Hip Fracture Database containing all hip fractures in Norway during the time period 1994–2008. Hip fractures were identified by computerized hospital discharge diagnoses. Charlson comorbidity index was calculated based on additional diagnoses and categorized (0, 1, and ≥2). Summer was defined as June, July, and August and winter as December, January, and
February. Incidence rate ratios for hip fracture according to season were calculated by negative binomial models. Results In patients aged 50–103 years, 136,140 eligible fractures were identified (72.5 % women). The relative risk of hip fracture in winter versus summer was 1.40 (95 % confidence interval (CI) 1.36–1.45) in men and 1.26 (95 % CI 1.23–1.28) in women. June had the lowest number of fractures in both genders. We found seasonal variation in all subgroups by age and gender, although least pronounced in patients >79 years. There was a significant interaction between season and comorbidity (p=0.022). When comparing winter to summer, we found relative risks of 1.40 (95 % CI 1.31–1.50) in patients with Charlson index=0, 1.29 (95 % CI 1.19–1.40) in patients with Charlson index=1, and 1.18 (95 % CI 1.08–1.28) in patients with Charlson index ≥2.
The NORwegian EPidemiologic Osteoporosis Studies (NOREPOS) is a collaboration between epidemiologic osteoporosis studies, which are substudies within large population-based surveys in four districts of Norway (Tromsø, Nord-Trøndelag, Hordaland, Oslo). The NOREPOS Hip Fracture Database includes all hospitalizations for hip fracture in Norway during the time period of 1994–2008. S. M. Solbakken (*) : J. H. Magnus : H. E. Meyer : T. K. Omsland Institute of Health and Society, Department of Community Medicine, University of Oslo, PO Box 1130, Blindern 0318, Oslo, Norway e-mail: [email protected] H. E. Meyer : K. Holvik : A. J. Søgaard Division of Epidemiology, Norwegian Institute of Public Health, Oslo, Norway N. Emaus Department of Health and Care Sciences, UiT The Arctic University of Norway, Tromsø, Norway G. S. Tell : K. Holvik Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway
G. Grimnes Department of Clinical Medicine, UiT The Arctic University of Norway, Tromsø, Norway
S. Forsmo : B. Schei Institute of Public Health and General Practice, Faculty of Medicine, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
B. Schei Department of Obstetrics and Gynecology, St. Olav’s University Hospital, Trondheim, Norway
191, Page 2 of 7
Conclusions There was a distinct seasonal variation in hip fracture incidence, present in all subgroups of gender, age, or comorbidity. This variation should be accounted for when planning health-care services. Keywords Hip fracture . Incidence . Seasonal variation . Comorbidity . Age . Gender
Arch Osteoporos (2014) 9:191
were also retrieved, together with the dates of admission and discharge. Hip fractures were identified according to the International Classification of Diseases, Ninth Revision (ICD-9): 820 with all subgroups and Tenth Revision (ICD10): S72.0–S72.2. Thus, both cervical, trochanteric and subtrochanteric fractures were included. Validation
Introduction The considerable differences in hip fracture risk between countries are well described, with more than tenfold variation in hip fracture rates worldwide [1]. Even though the incidence rates of hip fracture in Norway have decreased during the last decade [2], the country still has one of the highest risks of hip fracture in the world [1]. Whether hip fracture risk varies by season has yet to be clarified. In studies reporting differences in hip fracture rates between summer and winter [3–12], potential explanations are not much explored. Furthermore, the link of this seasonal variation to comorbidity is unknown. Although winter peaks and summer troughs in hip fracture risk have been described [3–12], not all studies have found seasonal variation [13, 14], and findings from Norway are disparate. Two studies reported higher incidence during winter [15, 16], whereas others did not report seasonal variation in every year examined [17] or found variation in outdoor fractures only [18]. Seasonal variation by age and gender is examined in some of the studies, but with dissimilar results [4–6, 8–11, 15]. Hip fractures are associated with both mortality [19] and functional decline [20] and demand considerable resources from the healthcare system [21]. A large-scale study could help determine the extent of a possible seasonal variation, and thereby provide important information when planning health-care services. The purpose of this study was to examine a possible seasonal variation in hip fracture incidence in Norway by comorbidity score, age, and gender.
In order to validate the data at an individual level, NORHip was compared to local fracture registers verified by medical records or x-ray in Oslo and Tromsø. The combined Cohen’s kappa for the comparisons was 0.95 [2]. Charlson comorbidity index A modified Charlson index was used to classify patients due to the severity and number of comorbidities, based on a total score calculated from additional diagnosis codes recorded during the hospitalization for hip fracture [22, 23]. The total score was used to categorize patients into three groups: 0=a person without any of the additional diagnosis codes, 1= patients with a score of 1 indicating some comorbidity, and 2=patients with a total score of ≥2 representing severe comorbidity. Due to change of coding system from ICD-9 to ICD-10 in 1999, the Charlson index was based on ICD-10 diagnosis codes only in order to ensure comparability of obtained categories. As information on comorbidities was available for the hip fracture patients only, we calculated the person time from 1999 until fracture (by month) to estimate the correct population time at risk in each of the three Charlson index subgroups. We included only each individual’s first recorded hip fracture during 1999–2008 in the comorbidity analyses, due to dependency between the first and second hip fracture and because the Charlson index in an individual patient may change over time. A total of 81,849 fractures were included in the comorbidity analyses. Statistical methods
Materials and methods Data collection Data were retrieved from the NOREPOS Hip Fracture Database (NORHip), containing hospital-treated hip fractures in Norway during 15 years (1 January 1994 to 31 December 2008). The data collection, classification, validation, and quality assurance of this database have previously been published as supplementary material [2]. Briefly, a system developed by the Norwegian Knowledge Centre for the Health Services was used to identify hip fractures, based on computerized discharge diagnoses. Additional diagnosis codes and surgical procedure codes during the hospitalization for hip fracture
The analyses by gender and age included 136,140 hip fractures (first or second) sustained in men and women aged 50– 103 years during 1994–2008. Data on the total Norwegian population was acquired from Statistics Norway [24]. The mid-year population of each 1-year age group ≥50 and ≤103 years was defined as the population at risk. Data analysis was performed in Stata MP13. Because our data were overdispersed, using a Poisson regression model might have underestimated the standard errors [25]. We therefore used a negative binomial model to estimate the incidence rate ratios. Winter was defined as December, January, and February; spring as March, April, and May; summer as June, July, and August; and autumn as September, October, and November.
Arch Osteoporos (2014) 9:191
Page 3 of 7, 191 4500
10.6% 9.7% 9.4%
Total number of hip fractures
4000 3500
10.2% 7.7% 7.6%
3000
8.4% 6.9% 7.3%
7.6%
7.0%
7.5%
2500 2000 1500 1000 500 0
a 12000 Total number of hip fractures
We calculated the age-adjusted incidence rate ratio for each season in both men and women. Statistical interactions between season and gender, between age and month, and between season and Charlson index were tested. Incidence rate ratios were stratified on age, gender, and comorbidity. We also performed stratified analyses on age and gender within the Charlson index subgroups. To further examine a possible cyclical pattern of hip fractures, a gender stratified time series model (Cosinor analysis) was applied using day of the year as analytical unit. The fit of the model was tested. As Norway has a latitude ranging from approximately 58 to 71° N, the country was divided in a northern and southern region at approximately 64° N (Nord-Trøndelag). We tested for statistical interaction between season and region of residence and stratified by region. Concerning person time in the analyses by region, the same person time calculation as in the analyses by Charlson index was used.
Ethics
10000 8000
9.8% 8.8% 8.8%
9.9% 7.7% 8.0%
7.2%
7.8% 7.8% 7.6%
8.2% 8.4%
6000 4000 2000 0
The Regional Committee for Medical and Health Research Ethics, the Norwegian Data Protection Authority, The Directorate of Health, the National Population Register, and Statistics Norway have approved the study and the linkages of data.
b Fig. 1 Distribution of hip fractures by month (Norwegian patients aged 50–103 years during 1994–2008). Total numbers and percentages in a men and b women. Norwegian Epidemiologic Osteoporosis Studies (NOREPOS)
Seasonal variation by gender Results Distribution of hip fractures by month The total number of hip fractures by gender and age group can be found in Table 1. Of the 136,140 hip fractures, 72.5 % were sustained in women. Mean age at first fracture was 78.5 years in men and 81.2 years in women. Figure 1a, b displays the monthly distribution of fractures by gender. January and December were the peak months, whereas June had the lowest proportion of fractures (6.9 % in men and 7.2 % in women). Table 1 Total number of hip fractures by gender and age group (Norwegian patients aged 50–103 years during 1994–2008). A NOREPOSa study
50–64 years 65–79 years >79 years a
Men (%)
Women (%)
4,120 (11.0) 13,245 (35.4) 20,061 (53.6)
5,344 (5.4) 29,033 (29.4) 64,337 (65.2)
Norwegian Epidemiologic Osteoporosis Studies
The highest number of hip fractures occurred during winter, and compared to summer, the risk of hip fracture was 40 % (95 % confidence interval (CI) 36–45) higher in men and 26 % (95 % CI 23–28) higher in women (Table 2). There was a significant interaction between gender and season (p79 years
8
0.4
6
0.2
4
0
Fig. 3 Incidence rate ratios of hip fractures by month (Norwegian patients aged 50–103 years during 1994–2008) in a men and b women. Norwegian Epidemiologic Osteoporosis Studies (NOREPOS)
2
Mean daily number of hip fractures
a
1
0
Seasonal variation by age
100
200
300
Day of year (starting at January 1st) Cosinor model
Hip fractures
a
However, the youngest age group had the largest seasonal variation in both men and women.
10
15
20
25
Seasonal variation by comorbidity
5
Mean daily number of hip fractures
1.2
100
200
300
Day of year (starting at January 1st) Cosinor model
Hip fractures
b
Fig. 2 Mean daily number of hip fractures during 1994–2008 (Norwegian patients aged 50–103 years) in a men and b women. A time series model (Cosinor model) is fitted. Norwegian Epidemiologic Osteoporosis Studies (NOREPOS)
Comorbidity analyses were based on the three subgroups corresponding to Charlson index 0 (46,575 fractures), 1 (21,787 fractures), and ≥2 (13,487 fractures), including first fractures during 1999–2008. After adjustment for age and gender, we found interaction between season and comorbidity (p=0.022). When comparing winter to summer, the higher fracture risk varied from 40 % in the most healthy patients to 18 % in patients with highest comorbidity score (Table 3). In the analysis stratified according to age group and Charlson index, the seasonal variation was most pronounced among subjects younger than 65 years in the healthy subgroup (Charlson index=0) (IRRwinter versus summer =1.67; 95 % CI= 1.54–1.82), while the patients >79 years of age with highest comorbidity (Charlson index ≥2) showed less seasonal variation (IRRwinter versus summer =1.14; 95 % CI=1.08–1.21) (adjusted for age and gender).
Arch Osteoporos (2014) 9:191
Page 5 of 7, 191
Table 3 Number of first hip fractures and incidence rate ratios in winter versus summer by Charlson comorbidity index (1999–2008). A NOREPOSa study Charlson index N (%) 0 1 ≥2 a
IRRb winter versus summer 95 % CI
46,575 (56.9) 1.40 21,787 (26.6) 1.29 13,487 (16.5) 1.18
1.31–1.50 1.19–1.40 1.08–1.28
Norwegian Epidemiologic Osteoporosis Studies
b
Winter (December, January, February); summer (June, July, August). Adjusted for age and gender
Discussion Based on a total of 136,140 hip fractures among patients aged 50–103 years, we demonstrated an overall increased fracture risk in winter compared to summer. The relative risk in winter versus summer was 1.40 in men and 1.26 in women. June had the lowest number of fractures in both genders. The seasonal variation occurred in all subgroups by age and gender, but less distinct with increasing age and extent of comorbidity. Several studies have reported seasonal variation in hip fracture incidence, but a Swedish and a Norwegian study were inconclusive [13, 17]. These studies were both based on data from individual years, making them vulnerable to year by year fluctuations. A study from Harstad, Norway, [18] found seasonal variation in outdoor fractures only, and an Italian study based on 4 years with a total of 8,978 cases reported no seasonal variation [14]. Nevertheless, our findings were consistent with results from several other studies. In addition to previous studies in Norway [15, 16], increased hip fracture rates during winter time have been reported from Spain, USA, Canada, UK, Scotland, New Zealand, Hong Kong, and Taiwan [3–12]. In agreement with our findings, the largest seasonal variations have been demonstrated in men [4, 8, 10] and in younger age groups [4, 8–10, 15]. Some studies have reported higher hip fracture risk during the winter season in older age groups [11] and higher mean age of the patients in this season [5], possibly owing to decreased mobility due to many layers of clothing, inadequate home safety precautions, decreased capacity of synthesizing vitamin D and higher vitamin D requirements to maintain bone density in the elderly. However, a British study found no differences in hip fracture patient characteristics between summer and winter [6]. Climate has been suggested as a causal factor for the seasonal variation, due to outdoor conditions such as ice and slippery pavements [26]. The association between hip fractures and season-related variables such as snow, freezing rain, and cold temperatures has been described [5, 9, 10, 12]. However, several studies have found seasonal variation at
temperate and even subtropical climates [4, 7, 8, 11]. Furthermore, the majority of fractures occur indoors [13, 18], thus reducing the impact of ice and snow. Certain environmental factors will nevertheless affect both outdoor and indoor conditions, such as temperature and number of daylight hours. These aspects might be important when trying to explain the increased fracture risk in the winter months. Lower temperatures in winter might affect conditions indoors as well as outdoors with the potential to cause hypothermia, impairment of coordination, and thereby lead to falls with hip fracture as a consequence [27]. Moreover, the number of daylight hours in Norway is greatly reduced in winter and even absent in the northern part of the country. The lack of sunshine exposure during winter will abolish cutaneous vitamin D synthesis, and seasonal variation in vitamin D levels and bone turnover markers has been found in several studies [28–32]. Low vitamin D levels are associated with increased risk of hip fractures [33–35] related to both bone loss [36] and impaired muscle function [36–38] and thereby increased risk of falling [39, 40]. This theory is supported by our findings, demonstrating the largest seasonal variation in the northern part of the country. On the other hand, the north-south difference could also be caused by climatic differences (more snow, ice, and lower temperatures in the north). In addition, impaired vision due to winter darkness might affect coordination and contribute to the seasonal variation in hip fracture incidence. To our knowledge, the association between comorbidity and seasonal variation has not been described previously. When analyzing subgroups by Charlson index, we found the greatest seasonal variation in the healthiest group. Men 50–64 years in this group displayed the greatest differences between summer and winter. In contrast, hip fractures in patients >79 years of age in the group with most comorbidity showed less seasonal variation. Dissimilarities in activity level and exposure to environmental factors might explain the differences. Healthy subjects in the youngest age group are more likely physically active, which might include outdoor activities related to risk of injuries such as winter sports. Consequently, they might be more vulnerable to environmental hazards like snow and ice. Most probably, their counterparts with highest comorbidity >79 years of age are much more inactive, institutionalized and sometimes immobilized, and will thus be affected by seasonal factors to a lesser extent. Results from the Harstad study support this theory, reporting higher mean age at fracture in those occurring indoors compared to outdoors [18]. Further research on the place and type of injury in different age groups could be useful when considering preventive measures.
191, Page 6 of 7
Strengths and limitations The strengths of this study include the large number of cases based on a validated hip fracture database. All hospital-treated hip fractures in Norway in a 15-year period were included in the analyses by gender and age (the comorbidity analyses were restricted to a 10-year period). Using Charlson index [23] to measure comorbidity has its limitations. The registration of additional diagnosis codes might have been incomplete or missing and was based on the information at the time of the hip fracture hospitalization only. Health professionals are supposed to record diagnoses relevant for the hip fracture hospitalization. Unfortunately, complete data on comorbidity is therefore lacking. However, diagnoses considered to be irrelevant are perhaps not important when assessing the current health status in the patient at the time of fracture.
Conclusions Based on a total of 136,140 fractures, we found a considerably higher risk of hip fracture in winter versus summer, present in all subgroups defined by gender, age, or degree of comorbidity. The seasonal variation was most pronounced in men, in the younger age groups, and in the healthiest with least comorbidity. The substantial seasonal variation should be accounted for when planning health-care services.
Conflict of interest None.
References 1. Kanis JA, Oden A, McCloskey EV, Johansson H, Wahl DA, Cooper C, Epidemiology IOFWGo, Quality of L (2012) A systematic review of hip fracture incidence and probability of fracture worldwide. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 23(9):2239–2256 2. Omsland TK, Holvik K, Meyer HE et al (2012) Hip fractures in Norway 1999–2008: time trends in total incidence and second hip fracture rates: a NOREPOS study. Eur J Epidemiol 27:807–814 3. Alvarez-Nebreda ML, Jiménez AB, Rodríguez P, Serra JA (2008) Epidemiology of hip fracture in the elderly in Spain. Bone 42:278– 285 4. Bischoff-Ferrari HA, Orav JE, Barrett JA, Baron JA (2007) Effect of seasonality and weather on fracture risk in individuals 65 years and older. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 18:1225–1233 5. Chiu KY, Ng TP, Chow SP (1996) Seasonal variation of fractures of the hip in elderly persons. Injury 27:333–336 6. Crawford JR, Parker MJ (2003) Seasonal variation of proximal femoral fractures in the United Kingdom. Injury 34:223–225 7. Douglas S, Bunyan A, Chiu KH, Twaddle B, Maffulli N (2000) Seasonal variation of hip fracture at three latitudes. Injury 31:11–19
Arch Osteoporos (2014) 9:191 8. Jacobsen SJ, Goldberg J, Miles TP, Brody JA, Stiers W, Rimm AA (1991) Seasonal variation in the incidence of hip fracture among white persons aged 65 years and older in the United States, 1984– 1987. Am J Epidemiol 133:996–1004 9. Jacobsen SJ, Sargent DJ, Atkinson EJ, O'Fallon WM, Melton LJ 3rd (1995) Population-based study of the contribution of weather to hip fracture seasonality. Am J Epidemiol 141:79–83 10. Levy AR, Bensimon DR, Mayo NE, Leighton HG (1998) Inclement weather and the risk of hip fracture. Epidemiology 9:172–177 11. Lin HC, Xiraxagar S (2006) Seasonality of hip fractures and estimates of season-attributable effects: a multivariate ARIMA analysis of population-based data. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 17:795–806 12. Modarres R, Ouarda TBMJ, Vanasse A, Orzanco MG, Gosselin P (2012) Modeling seasonal variation of hip fracture in Montreal, Canada. Bone 50:909–916 13. Leavy B, Aberg AC, Melhus H, Mallmin H, Michaelsson K, Byberg L (2013) When and where do hip fractures occur? A population-based study. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 24:2387–2396 14. Pedrazzoni M, Alfano FS, Malvi C, Ostanello F, Passeri M (1993) Seasonal variation in the incidence of hip fractures in EmiliaRomagna and Parma. Bone 14(Supplement 1):57–63 15. Gronskag AB, Forsmo S, Romundstad P, Langhammer A, Schei B (2010) Incidence and seasonal variation in hip fracture incidence among elderly women in Norway. The HUNT Study. Bone 46: 1294–1298 16. Diamantopoulos AP, Rohde G, Johnsrud I, Skoie IM, Johnsen V, Hochberg M, Haugeberg G (2012) Incidence rates of fragility hip fracture in middle-aged and elderly men and women in southern Norway. Age Ageing 41:86–92 17. Lofthus CM, Osnes EK, Falch JA, Kaastad TS, Kristiansen IS, Nordsletten L, Stensvold I, Meyer HE (2001) Epidemiology of hip fractures in Oslo, Norway. Bone 29:413–418 18. Emaus N, Olsen LR, Ahmed LA, Balteskard L, Jacobsen BK, Magnus T, Ytterstad B (2011) Hip fractures in a city in Northern Norway over 15 years: time trends, seasonal variation and mortality: the Harstad Injury Prevention Study. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 22: 2603–2610 19. Haentjens P, Magaziner J, Colón-Emeric CS, Vanderschueren D, Milisen K, Velkeniers B, Boonen S (2010) Meta-analysis: excess mortality after hip fracture among older women and men. Ann Intern Med 152:380–390 20. Bentler SE, Liu L, Obrizan M et al (2009) The aftermath of hip fracture: discharge placement, functional status change, and mortality. Am J Epidemiol 170:1290–1299 21. Haentjens P, Lamraski G, Boonen S (2005) Costs and consequences of hip fracture occurrence in old age: an economic perspective. Disabil Rehabil 27:1129–1141 22. Stagg V (2006) CHARLSON: Stata module to calculate Charlson index of comorbidity. http://ideas.repec.org/c/boc/bocode/s456719. html#download. Accessed January 14th 2014 23. Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 40:373–383 24. (2013) Table: 07459: Population, by sex and one-year age groups. Statistics Norway, https://www.ssb.no/statistikkbanken/selectvarval/ Define.asp?subjectcode=&ProductId=&MainTable=NY3026&nvl= & P L a n g u a g e = 1 & n y T m p Va r = t r u e & C M S S u b j e c t A r e a = befolkning&KortNavnWeb=folkemengde&StatVariant=&checked= true. Accessed January 7th 2014 25. Ismail N, Jemain AA (2007) Handling overdispersion with negative binomial and generalized Poisson regression models. Casualty Actuarial Society Forum
Arch Osteoporos (2014) 9:191 26. Ralis ZA (1981) Epidemic of fractures during period of snow and ice. Br Med J 282:603–605 27. Bastow MD, Rawlings J, Allison SP (1983) Undernutrition, hypothermia, and injury in elderly women with fractured femur: an injury response to altered metabolism? Lancet 1:143–146 28. Darling AL, Hart KH, Gibbs MA et al (2013) Greater seasonal cycling of 25-hydroxyvitamin D is associated with increased parathyroid hormone and bone resorption. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 25(3): 933–941 29. Klenk J, Rapp K, Denkinger MD, Nagel G, Nikolaus T, Peter R, Koenig W, Bohm BO, Rothenbacher D (2013) Seasonality of vitamin D status in older people in Southern Germany: implications for assessment. Age Ageing 42:404–408 30. Nurmi I, Kaukonen JP, Luthje P, Naboulsi H, Tanninen S, Kataja M, Kallio ML, Leppilampi M (2005) Half of the patients with an acute hip fracture suffer from hypovitaminosis D: a prospective study in southeastern Finland. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 16:2018–2024 31. Pasco JA, Henry MJ, Kotowicz MA, Sanders KM, Seeman E, Pasco JR, Schneider HG, Nicholson GC (2004) Seasonal periodicity of serum vitamin D and parathyroid hormone, bone resorption, and fractures: the Geelong Osteoporosis Study. J Bone Min Res : Off J Am Soc Bone Min Res 19:752–758 32. Midtby M, Magnus JH, Joakimsen RM (2001) The Tromso Study: a population-based study on the variation in bone formation markers with age, gender, anthropometry and season in both men and women. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 12:835–843
Page 7 of 7, 191 33. Bakhtiyarova S, Lesnyak O, Kyznesova N, Blankenstein MA, Lips P (2006) Vitamin D status among patients with hip fracture and elderly control subjects in Yekaterinburg, Russia. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 17:441–446 34. Holvik K, Ahmed LA, Forsmo S, Gjesdal CG, Grimnes G, Samuelsen SO, Schei B, Blomhoff R, Tell GS, Meyer HE (2013) Low serum levels of 25-hydroxyvitamin D predict hip fracture in the elderly: a NOREPOS study. J Clin Endocrinol Metab 98:3341–3350 35. Sakuma M, Endo N, Oinuma T, Hayami T, Endo E, Yazawa T, Watanabe K, Watanabe S (2006) Vitamin D and intact PTH status in patients with hip fracture. Osteoporos Int: J Established Res Coop Eur Found Osteoporos Natl Osteoporos Found USA 17:1608–1614 36. Lips P (2001) Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocr Rev 22:477–501 37. Ceglia L (2008) Vitamin D and skeletal muscle tissue and function. Mol Asp Med 29:407–414 38. Visser M, Deeg DJH, Lips P (2003) Low vitamin d and high parathyroid hormone levels as determinants of loss of muscle strength and muscle mass (sarcopenia): the Longitudinal Aging Study Amsterdam. J Clin Endocrinol Metab 88:5766–5772 39. Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, Orav JE, Stuck AE, Theiler R, Wong JB, Egli A, Kiel DP, Henschkowski J (2009) Fall prevention with supplemental and active forms of vitamin D: a meta-analysis of randomised controlled trials. BMJ 339:b3692 40. Murad MH, Elamin KB, Abu Elnour NO et al (2011) Clinical review: the effect of vitamin D on falls: a systematic review and metaanalysis. J Clin Endocrinol Metab 96:2997–3006
