SEARCH

SEARCH BY CITATION

Keywords:

  • BONE DENSITOMETRY;
  • BONE;
  • EPIDEMIOLOGY;
  • POPULATION STUDIES;
  • OSTEOPOROSIS

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

Higher rates of hip fracture and all fractures combined have been observed in urban compared with rural areas, but whether there are urban-rural differences in distal forearm fracture rates is less studied. The aim of this longitudinal study was to compare the incidence of forearm fracture in postmenopausal women in urban and rural areas in Norway and to investigate risk factors that could explain potential fracture differences. The study included data from 11,209 women aged 65 years or more who participated in two large health studies, the Tromsø Health Study in 1994–1995 and the Nord-Trøndelag Health Study in 1995–1997. Forearm bone mineral density (BMD) was measured by single-energy X-ray absorptiometry in a subsample of women (n = 7333) at baseline. All women were followed with respect to hospital-verified forearm fractures (median follow-up 6.3 years). A total of 9249 and 1960 women lived in areas classified as rural and urban, respectively. Urban women had an increased forearm fracture risk [relative risk (RR) = 1.29, 95% confidence interval (CI) 1.09–1.52] compared with women in rural areas. Rural women had higher body mass index (BMI) than urban women, and the RR was moderately reduced to 1.21 (95% CI 1.02–1.43) after BMI adjustments. Rural women had the highest BMD. In the subgroup with measured BMD, adjustments for BMD changed the urban versus rural RR from 1.21 (95% CI 0.96–1.52) to 1.05 (95% CI 0.83–1.32), suggesting that BMD is an important explanatory factor. In conclusion, higher rates of forearm fractures was found in urban compared with rural women. © 2011 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

Norway has among the highest osteoporotic fracture rates ever reported,1–3 and the incidence varies between rural and urban parts of the country.4–7 Higher osteoporotic fracture rates in urban compared with rural areas also have been reported from Sweden, Australia, and North America.8–11 Bone mineral density (BMD) is an important risk factor for fracture.12 Higher BMD or bone mineral content (BMC) in the distal forearm,13, 14 hip, and spine15, 16 have been reported in rural compared with urban areas.

Forearm fractures are the most frequent osteoporotic fracture and, in contrast to hip fractures, are not associated with increased mortality.17–19 However, forearm fractures are important predictors of hip fractures.20 Only a small number of studies have investigated urban-rural differences in distal forearm fractures.7–11 Although there seems to be higher fracture rates in postmenopausal women living in urban compared with rural areas, the differences are not statistically significant,8, 9 the results are difficult to interpret,11 the study is cross-sectional,7 fractures are self-reported,7 or the study did not have access to individual data on possible confounding factors.10 Risk factors explaining possible urban-rural fracture differences are even less studied.

With prospective data from two population-based studies, the aims of this study were to examine possible differences in forearm fracture rates in urban and rural areas and to assess whether forearm BMD and other relevant risk factors can explain potential urban-rural differences in fracture risk.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

The Norwegian Epidemiological Osteoporosis Studies (NOREPOS) is a collaboration between osteoporosis substudies in four large Norwegian population-based multipurpose studies. In this study, we used data from women aged 65 years or more in the county of Nord-Trøndelag and the municipality of Tromsø. Baseline characteristics were collected in 1994–1995 in the fourth Tromsø study (Tromsø IV) and in 1995–1997 in the second Nord-Trøndelag Health Study (HUNT II). Forearm fractures sustained after inclusion in the health studies and onward were registered. Median follow-up time in Tromsø was 9.7 years (range 4 days to 10.3 years), and median follow-up time in Nord-Trøndelag was 6.2 years (range 1 day to 7.4 years).

Participants

A total of 8824 women aged 65 years or older (74% of invited) attended HUNT II in Nord-Trøndelag. Among these, 8747 answered the relevant questionnaires and were eligible for fracture follow-up. In Tromsø, 2511 women aged 65 years or older (73% of invited) attended the Tromsø IV study. Among these, 2462 answered the relevant questionnaires and were eligible for fracture follow-up.

Fracture registration

At both study sites, distal forearm fractures were defined as the first registered fracture of the distal radius, distal ulna, or both after attendance in the health studies in Tromsø and Nord-Trøndelag.

Tromsø

Fractures were registered between attendance in Tromsø IV in 1994–1995 and December 31, 2004. The fracture registration system in Tromsø has been validated previously and described by Joakimsen and colleagues.21 The fracture registry was based on the radiographic archives at the University Hospital in Tromsø. Fractures sustained outside the city of Tromsø are also likely to be captured because the University Hospital in Tromsø covers large areas of the county of Troms. Any radiographic examinations coded abnormal were reviewed to ascertain the fracture code, identify the exact anatomic location of the fracture, and distinguish consecutive fracture cases from each another and to capture fractures that had not been coded correctly as fractures. A total of 262 distal forearm fractures were identified during the follow-up period.

Nord-Trøndelag

Fractures were registered between attendance in HUNT II in 1995–1997 and December 31, 2002. Fractures sustained in Nord-Trøndelag are treated either at Namsos Hospital or Levanger Hospital. Computerized lists of all radiographic examinations of the upper extremity were made, and all descriptions of these examinations were read by two medical doctors. In addition, separate searches for medical records in the electronic hospital register [Patient Administrative System (PAS)] were performed. Medical records were identified from computerized lists of diagnoses [ICD-9 (813.X) or ICD-10 (S52.X)] and read by trained technicians at the hospitals, and confirmed fractures of the distal forearm were registered. Both fractures from the radiographic records and hospital register/medical records were defined as fractures. During the follow-up period, a total of 580 distal forearm fractures were identified based on radiographic records and hospital register/medical records combined; 59 of these fractures were identified exclusively by radiographic records, and the corresponding number identified by hospital register/medical records was 37.

Bone densitometry

In Tromsø all women 65 to 74 years of age and 5% to 10% of women older than 74 years of age were invited for single-energy X-ray absorptiometry (SXA) measurements.22 In Nord-Trøndelag, all women 65 years of age and older (except inhabitants of two small municipalities) were invited for SXA measurements of the forearm.23 A total of 1373 women had valid distal forearm BMD measurements in Tromsø (56% of those included in this study), whereas the corresponding number in Nord-Trøndelag was 5960 (68%).

Scans of the nondominant forearm were performed at the distal and ultradistal site by bone densitometers of the same type (DTX 100, Osteometer MediTech, Inc., Hawthorne, CA, USA). A similar protocol was followed at both study sites.22 Cross-calibration was performed by using the European forearm phantom (QRM GmbH, Moehrendorf, Germany).

Other covariates

The studies in NOREPOS are part of the CONOR (COhort of NORway) collaborative study.24 The CONOR protocol includes a common set of questions and standard anthropometric measurements that enables comparison between study sites. Body height and weight in Tromsø IV and HUNT II were measured in light clothes without shoes, height to the nearest centimeter and weight to the nearest 0.5 kg. Body mass index (BMI) was calculated as weight in kg/height2. Both studies included self-administered questionnaires regarding physical activity, smoking, disease history, hormonal therapy, reproductive history, and age at menopause. Light and hard physical activity levels were assessed as hours per week in the last year (0, <1, 1 to 2 or ≥3). Light and hard activity scores (0 to 3) were combined, and a new variable was computed (ranging from 0 to 6). If the combined activity was ≤2, it was considered sedentary. Consumption of alcohol was assessed as times of alcohol intake per month and used without further recoding. Smoking was classified as current smokers or not. Self-reported general health status was assessed as poor/fair/good/excellent and categorized into two groups (poor/fair or good/excellent). Number of childbirths was reported. Hormone therapy (HT) was assed as current, former, and never. Disease history (diabetes and myocardial infarction) was assessed as yes or no in the questionnaire and used without further recoding.

Definition of urban and rural areas

The municipality of Tromsø has approximately 65,000 inhabitants, and of these, 80% live in the urban part and 20% in the rural part. Nord-Trøndelag has 130,000 inhabitants and is generally a rural county with a few small villages. Tromsø is situated at approximately 70° north, whereas Nord-Trøndelag is situated at approximately 65° north. In the main analyses, the whole county of Nord-Trøndelag was classified as rural, and the population of Nord-Trøndelag and the rural population of Tromsø were combined and compared with the population in the urban part of Tromsø. This classification also has been used in previous studies.7, 14 In an additional analysis, Nord-Trøndelag was divided in two according to the population density in each of the 24 municipalities (rural <10,000 inhabitants and semiurban ≥10,000 inhabitants).

Statistical analysis

All analyses were performed in SPSS 16.0 (SPSS, Inc., Chicago, IL, USA) and STATA 9.2 (Stata Corporation, Inc., Collage Station, TX, USA). Person-years and incidence rates (per 10,000 person-years) in Tromsø (urban/rural) and Nord-Trøndelag (rural/semirural) were calculated. We also calculated incidence rates in urban and rural areas across 5-year age groups, and in this analysis, a subject's person-years were allowed to count in more than one age category if the person entered a new age category during follow-up.

When analyzing continuous variables, t tests and analyses of variance adjusted for age were used to compare baseline characteristics between the urban and rural populations. Logistic regression adjusted for age was used to compare dichotomous variables between the urban and rural populations.

The hazard ratios (HRs) for forearm fracture [hereafter called relative risk (RR)] and confidence intervals (CIs) were calculated by Cox proportional-hazards regression. The criteria of proportional hazards were fulfilled. Time in the Cox model was calculated as the time between the baseline examination at the screening station in each survey and first fracture or censoring (owing to end of study, death, or migration). All Cox models were adjusted for age, and in additional analyses we also adjusted for other relevant covariates. Cox models adjusted for age also were run to study effects of BMD on forearm fracture risk.

For some of the variables, a substantial amount of data was missing (eg, physical activity, alcohol intake, and BMD), but all changes in risk estimates before and after adjustments were observed in subsamples of subjects with complete data on the variables in question.

Ethics

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

The studies were evaluated by the Regional Committee for Medical Research Ethics and approved by the Norwegian Data Inspectorate. All participants gave written informed consent.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

Urban/rural differences in distal forearm fracture rates

The rate of forearm fracture was higher in urban Tromsø than in rural Tromsø (Table 1), with a corresponding age-adjusted RR of forearm fracture of 1.34 (95% CI 0.97–1.85). Dividing the population of Nord-Trøndelag in two according to population density, the difference in fracture incidence between semiurban and rural Nord-Trøndelag was small (Table 1). In the following, we have combined rural Tromsø with the predominantly rural county of Nord-Trøndelag into a rural group. In age-adjusted analyses, subjects living in urban areas had an RR of forearm fracture of 1.29 (95% CI 1.09–1.52) compared with women living in rural areas.

Table 1. Forearm Fracture Rates in Women 65 Years of Age and Older According to Urbanization in Nord-Trøndelag and Tromsø: The NOREPOS Study
 nPerson-yearsNo. of fracturesCrude fracture rate/10,000 person-yearsRR (95% CI)a
  • RR = relative risk.

  • a

    RRs are from Cox proportional-hazards models for time to first distal forearm fracture.

Nord-Trøndelag <10,000 (rural)3,58820,5142301121.00 (ref)
Nord-Trøndelag ≥10,000 (semiurban)5,15929,0603501201.09 (0.92–1.28)
Rural Tromsø5024,096441071.00 (0.72–1.39)
Urban Tromsø1,96015,1382181441.34 (1.10–1.64)
Total11,20968,808842122 

Table 2 shows fracture rates in 5-year age groups in urban and rural areas. Fracture rates generally were higher in urban compared with rural areas.

Table 2. Fracture Incidences by Age Group in Rural and Urban Areas: The NOREPOS Study
 RuralaUrbana
Age groupbNo. of person-yearsNo. of fracturesCrude fracture rate per 10,000 person-years (95% CI)No. of person-yearsNo. of fracturesCrude fracture rate per 10,000 person-years (95% CI)
  • a

    Rural, includes the county of Nord-Trøndelag and rural Tromsø; urban, urban Tromsø.

  • b

    Attained age. Participants were allowed to count in more than one category if entering a new age category during fracture follow-up.

65–697,2667096 (76–122)1,94535180 (129–250)
70–7415,486162105 (90–122)4,43654122 (93–159)
75–7914,485186128 (111–148)4,39557130 (100–168)
80–8410,088124123 (103–147)2,76149177 (134–235)
85+6,34582129 (104–160)1,60023144 (96–216)
Total53,670624116 (107–126)15,137218144 (126–164)

Owing to possible “loss” of fractures in Nord-Trøndelag in municipalities situated close to a larger city (Trondheim) in the neighboring county, we performed an analysis in which we excluded all subjects living in these municipalities. However, the age-adjusted RR of fracture in urban compared with rural citizens was almost unchanged (RR = 1.27, 95% CI 1.07–1.50) in this analysis (9697 subjects, 754 fractures).

Self-reported lifestyle factors and health as explanatory factors

Women living in urban areas were younger and slimmer, had higher alcohol consumption, were more often smokers, and had a higher proportion with poor/fair self-reported health and myocardial infarction (Table 3). Women residing in rural areas had higher body mass index (BMI), and more self-reported diabetes.

Table 3. Baseline Characteristics of Women Aged 65 Years or Older in Urban and Rural Areas (n = 11,209): The NOREPOS Study
 RuralaUrbanap Valueb
  • BMI = body mass index; BMD = bone mineral density; HT = hormone treatment.

  • a

    Rural; includes the county of Nord-Trøndelag and rural Tromsø; urban; urban Tromsø.

  • b

    All analyses (except age) are adjusted for age.

Continuous variables
 Age (years) (n = 11,209)74.473.0<.001
 BMI (kg/m2) (n = 10,726)27.726.2<.001
 Body height (cm) (n = 10,730)159159.397
 Number of childbirths (n = 10,228)2.92.6<.001
 Alcohol (times per month) (n = 5785)0.91.6<.001
 Ultradistal BMD (mg/cm2) (n = 7314)289256<.001
 Distal BMD (mg/cm2) (n = 7333)377343<.001
Dichotomous variables
 Cigarette smoking, current, n (%)1230 (14.8)427 (21.8)<.001
 HT, current or former, n (%)759 (8.2)149 (7.6).082
 Sedentary, n (%)3930 (64.3)1206 (61.7).187
 Poor/fair self-perceived health, n (%)4552 (50.9)1174 (60.1)<.001
 Diabetes, n (%)751 (8.4)104 (5.3)<.001
 Myocardial infarction, n (%)533 (5.9)135 (6.9).027

Adjusting for BMI at baseline in addition to age, the RR of forearm fracture in urban women decreased to 1.21 (95% CI 1.02–1.43). Similar adjustments for body height did not change the estimate (RR = 1.28, 95% CI 1.09–1.52). Self-perceived health was not significantly (p = .866) associated with risk of forearm fracture. Diabetes and myocardial infarction were the only recorded medical conditions shown to have impact on forearm fracture risk in an earlier study of the population of Nord-Trøndelag.23 The RR of fracture in urban versus rural women was unchanged (RR = 1.28, 95% CI 1.08–1.51) after adjustments for age, self-perceived health, diabetes, and myocardial infarction. Since there was a substantial amount of missing data on the HT question, missing was coded as never used, and adjustments for use of HT did not change the age-adjusted estimate (RR = 1.29, 95% CI 1.09–1.52). Analyses without recoding the HT variable did not change the conclusion (data not shown). The questions on physical activity and alcohol intake also had a substantial amount of missing data. However, the RR remained almost unchanged after adjustments for physical activity, smoking, and alcohol intake in the subgroup of subjects with valid data on these three lifestyle variables (4430 subjects, 356 fractures). Analyzing light and hard activity separately did not change the results (data not shown). In the subgroup of women with information on number of children, the age-adjusted RR was unchanged after adjustments for this variable (10,228 subjects, 774 fractures).

Bone mineral density and fracture risk

Including all subjects in analyses, each 1 SD decrease in distal BMD was associated with an RR of forearm fracture of 1.42 (95% CI 1.30–1.55). Similarly each SD decrease in ultradistal BMD was associated with an RR of 1.52 (95% CI 1.39–1.67).

BMD was significantly higher in women in rural compared with urban areas both at the distal and ultradistal sites (Table 3). In the subgroup of women with valid distal forearm BMD, the age-adjusted RR of fracture in urban versus rural residents was not statistically significant (RR = 1.21, p = .098; Table 4). The point estimate declined to 1.15 after adjustments for age and BMI, to 1.05 after adjustments for age and BMD, and to 1.04 after adjustments for age, BMI, and BMD. Adjustments for age and ultradistal BMD gave approximately the same results (RR = 1.00, 95% CI 0.80–1.26).

Table 4. Impact of BMI and BMD on the Relative Risk of Distal Forearm Fractures in Urban Versus Rural Areas (n = 7253): The NOREPOS Study
AdjustmentsRRa (95% CI)
  • BMD = bone mineral density; BMI = body mass index; CI = confidence interval; RR = relative risk.

  • a

    Restricted to women with BMI and BMD measurements. The RRs in the table are the risks of forearm fracture according to residence in an urban or rural area (referent group is residence in a rural area). RRs are from Cox proportional-hazards models for time to first distal forearm fracture.

Adjusted for age1.21 (0.96–1.52)
Adjusted for age and BMI1.15 (0.92–1.45)
Adjusted for age and distal BMD1.05 (0.83–1.32)
Adjusted for age, BMI, and distal BMD1.04 (0.82–1.31)

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

In this longitudinal study including more than 11,000 postmenopausal women, we have shown that women residing in urban areas have a nearly 30% increased risk of sustaining a forearm fracture compared with women in rural areas. A higher BMI in rural areas could explain some of the difference, whereas other measured lifestyle factors or health variables did not explain this difference. On the other hand, the study suggests that BMD is an important determinant of the urban/rural fracture differences.

Urban/rural differences in fracture risk

A number of studies report urban/rural differences in overall fracture rates,9–11 and urban/rural differences in hip fracture rates are also documented in a number of studies.10, 25–28 In contrast, studies regarding urban/rural differences in forearm fracture rates are few, and the findings are not conclusive. A study from Minnesota in the United States reported no significant urban/rural differences in rates of distal forearm fracture in either gender.9 One Australian study also found no significant urban/rural differences in distal radius fractures (Colles' fractures),8 whereas another Australian study found a borderline significant increased risk of distal forearm fractures in urban compared with rural women.10 A Swedish study did not state whether the higher rate of forearm fractures in urban areas was statistically significant or not,11 but a large cross-sectional study from Norway concluded that self-reported forearm fractures were more frequent in urban compared with rural areas in both genders.7

Possible explanatory factors

Of the covariates, BMI seemed to explain the urban/rural difference in fracture rate moderately because the RR changed from 1.29 to 1.21 after BMI adjustments. In an earlier study of urban/rural differences in forearm BMD in Norway, BMI was the only factor that was able to explain some of the variation in BMD.14 Risk factors such as smoking, physical activity, medical conditions, number of childbirths, and HT use did not explain the fracture difference. Some of the variables, such as physical activity and alcohol intake, had a considerable amount of missing data that gave us a statistical power problem. It is also possible that our information on physical activity was not detailed enough regarding recent activity, and it is a weakness that we lack information about physical activity earlier in life. We therefore cannot exclude the possibility that physical activity level does play a role—perhaps through BMD.

Moreover, there could be unmeasured factors that explain the findings, such as, for example, icy road conditions and nutritional factors. We also lack information on factors such as muscle strength, balance, and falls.

Bone mineral density and fracture risk

Although the urban/rural fracture difference was not significant in the subgroup of women who had their BMD measured (RR = 1.21, p = .098), BMD seemed to explain the entire urban/rural difference in fracture rates because the point estimate for the fracture difference changed toward 1.0 after adjustments for distal or ultradistal forearm BMD.

In accordance with our findings, a study from Sweden reported significantly lower forearm BMD in an urban compared with a rural population.13 In a Norwegian study comparing forearm BMD at different geographic sites—including the sites studied in this article—women and men living in rural areas had significantly higher BMD compared with those in urban areas.14 It was calculated that the difference in women corresponded to a 12% to 15% increased risk of fracture in urban areas. The findings of our study indicate an even larger difference in fracture rates between urban and rural areas.

Possible explanations for the urban/rural difference in BMD include higher BMI in rural areas14 and higher levels of recent and/or past physical activity in rural areas, and we cannot exclude the possibility that environmental factors such as air pollution play role.29

Strengths and limitations

The strengths of this study include the high participation rate, good fracture registration, and a long follow-up period. However, the fracture registration methods used in Tromsø and Nord-Trøndelag were not quite identical. The fracture registration method in Tromsø has been validated previously against self-reports in the 2001 survey (Tromsø V).21 The study showed good results on computerized radiographic records. In Nord-Trøndelag, however, we combined information from radiographic records and hospital register/medical records. This may have made the fracture registration in Nord-Trøndelag more complete because radiographic records alone were not used to identify fractures in Tromsø. If we included fractures from only the radiographic records in Nord-Trøndelag, the RR of fracture in urban compared with rural areas was 1.36 (95% CI 1.15–1.60). Hence the possible overestimation of fractures in rural areas may, if anything, have led to an underestimation of the urban/rural difference in fracture rates.

For both study sites, fractures never medically examined could have been missed. Fractures occurring while subjects were traveling also would be missed if no control radiographic examination was done after returning home. Such errors will only affect our conclusions if the fracture registration errors vary with degree of urbanization. We believe that this is not likely to be the case.

The main exposure variable in this article, urban versus rural residence, is a rather imprecise measure because we have classified the whole county of Nord-Trøndelag as rural. A proportion of subjects in Nord-Trøndelag live in areas that might be defined as either densely populated or urban. In an attempt to discover potential urbanization effects within Nord-Trøndelag, the county was divided in two according to population density (rural <10,000 and semiurban ≥10,000 inhabitants). The RR of fracture in rural versus semiurban areas was small and insignificant, but there was a tendency toward lower fracture rates in the less populated areas. The lack of precision in our exposure variable may have underestimated the differences in fracture risk.

High-quality registries of forearm fractures are hard to obtain because these fractures often are treated at a number of potential hospitals or outpatient clinics. In Tromsø, we had the advantage that other potential treatment facilities are located very far away; the nearest alternative radiographic service is 250 km from Tromsø. On the other hand, in Nord-Trøndelag, there could have been leakage of fracture patients to Trondheim in the neighboring county. However, if we exclude women residing in neighboring municipalities to Trondheim, the association between study site and fracture risk remains almost unchanged.

Although not statistically significant, we found a similar urban/rural difference within the municipality of Tromsø as in the main urban/rural analyses. The same fracture registration method was used for all women residing in Tromsø, suggesting that the different fracture registrations are not likely to be responsible for the fracture differences found in this study.

We only had SXA measurements in this study, but it would have been interesting to have, for example, hip BMD in addition. It has been shown that measures such as dual-energy X-ray absorptiometry (DXA) is better than SXA when predicting hip or vertebral fractures.12, 30 However, concerning forearm fracture, a meta-analysis by Marshall and colleagues showed that the ability of radius BMD to predict forearm fracture is better (RR per 1 SD decrease in BMD = 1.7 to 1.8) than other measures, such as hip BMD (RR per 1 SD decrease in BMD = 1.4) or vertebral BMD (RR per 1 SD decrease in BMD = 1.5).30 It is a limitation that we only had BMD measurements in a subgroup of the women. This is illustrated by the different point estimates for age- and BMI-adjusted risk in the whole population and in the subgroup with measured BMD (Table 4), which probably is related to different sample sizes.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

In this longitudinal study, we found a higher rate of distal forearm fracture in urban compared with rural areas in Norway. Forearm BMD is likely to be important in explaining the difference in fracture rates, whereas the regional differences persisted after adjustments for other possible explanatory factors. More studies are needed to further explore urban/rural differences in forearm fractures and BMD—and the distribution of possible explanatory factors for both.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References

This study was supported by grants from the Norwegian Research Council. The Tromsø studies were carried out by the Department of Community Medicine, Faculty of Health Sciences, University of Tromsø, in collaboration with the Norwegian Institute of Public Health, the University Hospital of Northern Norway (UNN), and the Tromsø City Council. The fractures in Tromsø were collected at UNN.

The Nord-Trøndelag Health Study (HUNT Study) is a collaboration between the HUNT Research Centre, Faculty of Medicine, Norwegian University of Science and Technology (NTNU, Verdal), the Norwegian Institute of Public Health, and the Nord-Trøndelag County Council. The fracture registration in Nord-Trøndelag was performed by the Nord-Trøndelag Hospital Trust.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Ethics
  6. Results
  7. Discussion
  8. Conclusions
  9. Disclosures
  10. Acknowledgements
  11. References