Osteoporosis International

, Volume 23, Issue 7, pp 1859–1865

Mortality after vertebral fracture in Korea

Analysis of the National Claim Registry

Authors

  • Y.-K. Lee
    • Department of Orthopaedic SurgerySeoul National University Bundang Hospital
  • S. Jang
    • Inje University College of Pharmacy
  • S. Jang
    • Graduate School of Public HealthSeoul National University
  • H. J. Lee
    • Health Insurance Review and Assessment Service
  • C. Park
    • Health Insurance Review and Assessment Service
  • Y.-C. Ha
    • Department of Orthopaedic SurgeryChung-Ang University College of Medicine
    • Department of Nuclear MedicineKyung Hee University Hospital
Original Article

DOI: 10.1007/s00198-011-1833-5

Cite this article as:
Lee, Y., Jang, S., Jang, S. et al. Osteoporos Int (2012) 23: 1859. doi:10.1007/s00198-011-1833-5
  • 510 Views

Abstract

Summary

The present study evaluates the incidence and mortality of vertebral fractures in Korea, using data from the Health Insurance Review Assessment Service, which includes nationwide information entrusted by Korean government.

Introduction

A vertebral compression fracture is a serious complication associated with osteoporosis of the spine. We evaluated the incidence of vertebral fracture and subsequent mortality in South Korea, using nationwide data from the Health Insurance Review and Assessment Service (HIRA).

Methods

All new visits or admissions to clinics or hospitals for fractures were recorded in nationwide cohort by the Korean HIRA using International Classification of Disease, tenth Revision (ICD-10) code. The incidence of vertebral fracture and excess mortality associated with vertebral fracture were evaluated, in men and women aged 50 years or more between 2005 and 2008. Standardized mortality ratio (SMR) was calculated to determine excess mortality associated with vertebral fracture.

Results

The crude overall incidence of vertebral fractures was 984 per 100,000 person years from 2005 to 2008. The overall mortality rate at 3 months, 6 months, 1 year, and 2 years after vertebral fracture in men (5.56%, 9.41%, 14.6%, and 20.61%, respectively) were higher than that in women (2.41%, 4.36%, 7.16%, and 10.48%, respectively). In both genders, the age-specific mortality rates were more than those of the general population. The SMR was highest during the first 3 months and gradually declined to 2.53 in men and 1.86 in women at the 2-year period.

Conclusions

The incidence of vertebral fracture in South Korea was comparable with other countries such as Switzerland, and the mortality after vertebral fracture is higher than that of normal populations. The incidence of osteoporotic vertebral fracture and following high mortality are likely to become serious socioeconomic problems.

Keywords

IncidenceMortalityOsteoporosisVertebral fracture

Introduction

A vertebral compression fracture is a serious complication associated with osteoporosis of the spine, and a major health concern, affecting a growing number of individual worldwide. Several studies suggested that osteoporotic vertebral fracture is associated with adverse effects. In addition to increased back pain [1] and diminished overall quality of life [2], excess mortality is also well described in patients who develop osteoporotic vertebral fracture [36].

However, previous cohort studies that have examined the relation between osteoporotic vertebral fractures and mortality have had limitations that may influence their results and the generalizability of the studies, including small samples [3, 6, 7], the inclusion of only women [7], the enrollment of participants from specific areas (i.e., hospitals or certain geographic regions) [3, 5, 6], the nonrandom selection of participants and the lack of statistical adjustment for confounding factors that may influence mortality[3, 5, 7]. Although several studies have used nationwide registers or claim records to overcome these limitations of studies using regional cohort, the most used registers had also limitation, such as hospitalization-based [810]. In addition, epidemiologic studies using nationwide register have been performed only in western country [810], and no excess mortality has been shown following the osteoporotic vertebral fracture in Asia.

The purpose of this study were to investigate the incidence of vertebral fracture and excess mortality following a vertebral fracture in Korean people over 50 years of age using nationwide claim database with obligation.

Materials and methods

The cohort of patients with a symptomatic vertebral fracture in men and women aged 50 years or more was identified from nationwide database by the Health Insurance Review and Assessment Service (HIRA) of South Korea between 2005 and 2008.

The used HIRA database was the National Korean Health Insurance claims database, which covered about 97% of the Korean populations. All clinics and hospitals submit data on inpatients and outpatients, including data on diagnosis and medical costs, for claims. Therefore, virtually all information about patients and diseases is available from the Korean HIRA database, which has been used on several occasions for epidemiological studies [1114].

One or more claims listing an International Classification of Disease, tenth Revision diagnosis code of S22.0 (fracture of the thoracic spine), S22.1 (multiple fractures of the thoracic spine), S32.0 (fracture of the lumbar spine), M48.4 (fatigue fracture of vertebra) and M48.5 (collapsed vertebra, NEC) was required for inclusion in this study. No distinction was made between vertebral fractures arising from high energy or low energy injury.

A unique personal identifier permitted the tracking of individuals for multiple visit or admissions. Where an individual had more than three outpatient visit or one admission for vertebral fracture, the patient was followed from the first event and recounted if a further event occurred 6 months or longer after the original visit or admission [9].

We followed each patient by code to identify death date of eligible file using the Korean National Health Insurance Program.

These data were retrospectively evaluated to determine the annual age- and gender-specific incidence of vertebral fracture and excess mortality associated with vertebral fracture, in men and women aged 50 years or more between 2005 and 2008.

Using the Poisson model, the mortality at 3 months, 6 months, 1 year, and 2 years after vertebral fracture was compared to that of the general Korean population aged 50 years or more. The mortality of the general population was obtained from the web site of the Korean Statistical Information Service [15].

To determine excess mortality associated with vertebral fracture, standardized mortality ratio (SMR), was calculated as the observed mortality divided by the expected mortality for each age and gender group [16, 17]. The expected mortality rate for each group was determined from the total number of age/gender specific study population multiplied by the each age/gender-specific mortality rate of Korean population. All data of Korean men and women over the age of 50 years was based on the web site of the Korean Statistical Information Service [15]. SMR of greater than 1.0 means excess mortality associated with vertebral fracture [16]. The 95% confidence intervals (95% CIs) for the SMR of each group were calculated by Poisson method of Owen [18]. All database management and analyses were performed using SAS statistical package version 9.1.3 (SAS Institute Inc, Cary, NC)

Results

In the population aged 50 years and older, the crude overall incidence of vertebral fractures was 984 per 100,000 person years from 2005 to 2008, and the gender-specific incidence was 427 per 100,000 person years for men and 1,456 per 100,000 person years for women (Table 1).
Table 1

Annual incidence of patients with vertebral fracture (per 100,000 patient years) by sex and age

Age (years)

Men

Women

Total

Number of fractures

Population

Incidence (rate/100,000)

Number of fractures

Population

Incidence (rate/100,000)

Number of fractures

Population

Incidence (rate/100,000)

2005

  50–54

2,027

1,443,169

140

2,559

1,424,704

180

4,586

2,867,873

160

  55–59

2,273

1,127,543

202

4,725

1,142,009

414

6,998

2,269,552

308

  60–64

3,145

913,555

344

10,505

1,012,957

1,037

13,650

1,926,512

709

  65–69

3,996

757,746

527

17,378

924,764

1,879

21,374

1,682,510

1,270

  70–74

3,732

508,074

735

20,226

738,150

2,740

23,958

1,246,224

1,922

  75–79

3,174

269,782

1,177

17,849

491,902

3,629

21,023

761,684

2,760

  80–84

2,148

137,596

1,561

11,240

298,126

3,770

13,388

435,722

3,073

  85+

1,038

60,463

1,717

5,289

180,039

2,938

6,327

240,502

2,631

  Totals

21,533

5,217,928

413

89,771

6,212,651

1,445

111,304

11,430,579

974

2006

  50–54

2,143

1,555,359

138

2,826

1,533,847

184

4,969

3,089,206

161

  55–59

2,469

1,165,204

212

4,817

1,178,637

409

7,286

2,343,841

311

  60–64

2,998

918,000

327

10,020

1,006,463

996

13,018

1,924,463

676

  65–69

4,200

784,517

535

17,748

946,397

1,875

21,948

1,730,914

1,268

  70–74

4,445

547,550

812

22,155

771,730

2,871

26,600

1,319,280

2,016

  75–79

3,559

290,706

1,224

19,466

522,641

3,725

23,025

813,347

2,831

  80–84

2,325

145,268

1,600

12,284

314,088

3,911

14,609

459,356

3,180

  85+

1,216

67,059

1,813

6,115

195,746

3,124

7,331

262,805

2,790

  Totals

23,355

5,473,663

427

95,431

6,469,549

1,475

118,786

11,943,212

995

2007

  50–54

2,451

1,665,071

147

2,813

1,637,778

172

5,264

3,302,849

159

  55–59

2,522

1,201,674

210

4,898

1,214,646

403

7,420

2,416,320

307

  60–64

2,861

936,578

305

9,148

1,013,584

903

12,009

1,950,162

616

  65–69

4,431

812,552

545

17,805

971,531

1,833

22,236

1,784,083

1,246

  70–74

4,862

581,198

837

22,636

799,808

2,830

27,498

1,381,006

1,991

  75–79

3,930

316,378

1,242

21,054

556,818

3,781

24,984

873,196

2,861

  80–84

2,532

153,795

1,646

13,518

329,193

4,106

16,050

482,988

3,323

  85+

1,518

74,715

2,032

7,553

214,375

3,523

9,071

289,090

3,138

  Totals

25,107

5,741,961

437

99,425

6,737,733

1,476

124,532

12,479,694

998

2008

  50–54

2,399

1,776,842

135

2,891

1,744,397

166

5,290

3,521,239

150

  55–59

2,542

1,242,203

205

4,647

1,255,354

370

7,189

2,497,557

288

  60–64

2,958

974,150

304

8,106

1,038,552

781

11,064

2,012,702

550

  65–69

4,643

829,526

560

17,655

987,096

1,789

22,298

1,816,622

1,227

  70–74

4,927

610,634

807

22,604

825,685

2,738

27,531

1,436,319

1,917

  75–79

4,155

345,459

1,203

21,718

590,729

3,676

25,873

936,188

2,764

  80–84

2,711

163,459

1,659

14,150

345,906

4,091

16,861

509,365

3,310

  85+

1,662

83,102

2,000

8,678

234,430

3,702

10,340

317,532

3,256

  Totals

25,997

6,025,375

431

100,449

7,022,149

1,430

126,446

13,047,524

969

During study periods, the annual incidence rate of vertebral fracture was stable in both genders (Table 1). From 2005 to 2008, the annual incidence of vertebral fractures for women was three times more than that for men in South Korea. In the population aged 50 years and older, the incidence of vertebral fractures was 431 per 100,000 persons for men and 1,430 per 100,000 for women in 2008.

The overall mortality rate at 3 months, 6 months, 1 year, and 2 years after vertebral fracture in men (5.56%, 9.41%, 14.6%, and 20.61%, respectively) were higher than women (2.41%, 4.36%, 7.16%, and 10.48%, respectively). In both genders, the age-specific mortality rates after vertebral fracture were more than those of the general population (Table 2).
Table 2

Mortality rate (/1,000 persons) in the general population and in men and women among 3 months, 6 months, 1 year, and 2 years after vertebral fracture

Age (years)

Mortality rate (/1000 persons) (95% CI)

General population

Among 3 months

Among 6 months

Among 1 year

Among 2 years

Men

  50–54

2.7

16.5 (13.9–19.6)

26.5 (23.1–30.5)

40.8 (36.5–45.6)

57.1 (51.9–62.8)

  55–59

4.0

27.6 (24.2–31.4)

41.4 (37.2–46.1)

60.6 (55.4–66.3)

84.5 (78.3–91.2)

  60–64

5.7

30.3 (27–34)

54 (49.5–58.9)

82.1 (76.5–88.1)

111.2 (104.6–118.2)

  65–69

9.3

38.8 (35.7–42.3)

69 (64.7–73.6)

107.9 (102.5–113.7)

147.2 (140.7–154.1)

  70–74

15.3

55.5 (51.7–59.5)

94.3 (89.2–99.6)

148.1 (141.7–154.9)

212.9 (205–221.1)

  75–79

24.4

71.3 (66.5–76.4)

125.3 (118.9–132.1)

206.2 (197.7–215.1)

303.6 (292.9–314.8)

  80–84

38.9

107.2 (100.1–114.9)

181.1 (171.5–191.3)

288.3 (275.7–301.4)

421.4 (405.5–438)

  85–89

55.2

141.9 (129.8–155.2)

249.8 (233.1–267.8)

383.9 (362.2–407)

568.2 (540.1–597.7)

  90–94

70.5

216.5 (185.8–252.3)

322 (282.6–366.8)

493.1 (441.5–550.8)

747.6 (679–823.1)

  95–100

82.0

255.4 (180.6–361.1)

382.5 (284.7–514)

620.5 (484.7–794.3)

980.9 (791–1216.4)

  Total

 

55.6 (54–57.3)

94.1 (91.9–96.3)

146.1 (143.3–148.9)

206.1 (202.8–209.5)

Women

  50–54

1.0

7 (5.5–8.9)

12.9 (10.8–15.5)

19.1 (16.4–22.1)

23.8 (20.8–27.2)

  55–59

1.4

6.9 (5.7–8.4)

12.5 (10.9–14.5)

20.6 (18.4–23)

28.1 (25.5–31)

  60–64

2.1

8.2 (7.2–9.3)

14.1 (12.7–15.5)

23.3 (21.5–25.1)

33.2 (31.1–35.4)

  65–69

3.7

10.6 (9.7–11.5)

20.3 (19.1–21.6)

34 (32.5–35.7)

48 (46.1–50)

  70–74

7.0

16 (15–17)

30.5 (29.1–31.9)

49.6 (47.9–51.4)

73.7 (71.6–75.9)

  75–79

13.5

26.1 (24.8–27.4)

48.4 (46.6–50.2)

81.3 (79–83.7)

123.3 (120.4–126.3)

  80–84

25.5

43.9 (41.8–46)

79.8 (77–82.7)

132.4 (128.7–136.2)

197.4 (192.8–202.1)

  85–89

42.5

65.2 (61.5–69.2)

116.6 (111.5–121.9)

198.6 (191.7–205.6)

296.2 (287.5–305.1)

  90–94

60.9

110.7 (101.3–120.9)

184.7 (172.2–198.1)

290.6 (274.3–307.8)

443.4 (422.2–465.6)

  95–100

76.7

158 (130.9–190.8)

254.2 (218–296.4)

420.2 (370.7–476.3)

583.5 (521.5–653)

  Total

 

24.1 (23.5–24.6)

43.6 (42.9–44.4)

71.6 (70.7–72.6)

104.8 (103.6–106)

The SMR was highest during the first 3 months and gradually declined over time in both genders. However, at the end of the 2-year period, the risk of death was still higher than that of the general population, which showed SMR of 2.53 in men and 1.86 in women (Table 3). The SMR of each age group decreased with increase of age in women, but the peak SMR of 55–59 years showed in men (Fig. 1).
Table 3

SMR at 3 months, 6 months, 1 year, and 2 years after vertebral fracture

Age (years)

SMR (95% CI)

3 months

6 months

1 year

2 years

Men

  50–54

10.39 (8.28–12.86)

8.85 (7.46–10.43)

6.6 (5.74–7.55)

4.8 (4.28–5.35)

  55–59

12.98 (11.06–15.13)

10 (8.8–11.31)

7.17 (6.44–7.95)

5.09 (4.66–5.54)

  60–64

9.56 (8.35–10.89)

8.06 (7.27–8.92)

6.17 (5.68–6.69)

4.14 (3.86–4.43)

  65–69

7.37 (6.65–8.15)

6.46 (5.98–6.96)

5.03 (4.73–5.35)

3.38 (3.21–3.56)

  70–74

6.68 (6.14–7.25)

5.52 (5.17–5.88)

4.28 (4.06–4.51)

3.05 (2.92–3.18)

  75–79

5.12 (4.72–5.55)

4.38 (4.11–4.65)

3.49 (3.32–3.66)

2.51 (2.41–2.61)

  80–84

4.41 (4.06–4.78)

3.59 (3.36–3.82)

2.77 (2.63–2.92)

1.99 (1.9–2.08)

  85–89

3.65 (3.27–4.06)

3.09 (2.84–3.36)

2.28 (2.12–2.44)

1.68 (1.58–1.78)

  90–94

3.74 (3.09–4.47)

2.68 (2.29–3.12)

1.92 (1.68–2.19)

1.49 (1.33–1.66)

  95–100

2.52 (1.68–3.65)

1.94 (1.36–2.68)

1.6 (1.19–2.1)

1.24 (0.97–1.57)

  Total

5.46 (5.27–5.65)

4.56 (4.43–4.68)

3.53 (3.45–3.61)

2.53 (2.48–2.58)

Women

  50–54

14.12 (10.37–18.78)

13.93 (11.24–17.06)

10.22 (8.58–12.09)

6.37 (5.47–7.39)

  55–59

9.54 (7.54–11.91)

8.45 (7.1–9.98)

6.99 (6.11–7.96)

4.82 (4.31–5.37)

  60–64

6.72 (5.76–7.79)

5.83 (5.19–6.52)

4.92 (4.5–5.36)

3.46 (3.22–3.72)

  65–69

5.29 (4.79–5.83)

5.13 (4.77–5.5)

4.3 (4.07–4.53)

3.01 (2.88–3.14)

  70–74

4.23 (3.93–4.54)

4.05 (3.84–4.26)

3.29 (3.15–3.42)

2.43 (2.35–2.51)

  75–79

3.53 (3.32–3.74)

3.24 (3.1–3.38)

2.71 (2.62–2.81)

2.04 (1.98–2.09)

  80–84

2.99 (2.83–3.16)

2.71 (2.6–2.82)

2.21 (2.14–2.29)

1.64 (1.6–1.68)

  85–89

2.45 (2.28–2.63)

2.17 (2.06–2.29)

1.82 (1.75–1.9)

1.39 (1.34–1.43)

  90–94

2.37 (2.12–2.64)

1.95 (1.79–2.12)

1.51 (1.4–1.61)

1.21 (1.14–1.28)

 95–100

1.34 (1.05–1.69)

1.42 (1.17–1.7)

1.27 (1.09–1.48)

0.95 (0.83–1.08)

  Total

3.25 (3.16–3.34)

2.99 (2.93–3.05)

2.48 (2.44–2.52)

1.86 (1.83–1.88)

https://static-content.springer.com/image/art%3A10.1007%2Fs00198-011-1833-5/MediaObjects/198_2011_1833_Fig1_HTML.gif
Fig. 1

Age-specific SMR of women and men

Discussion

This is the first study, to date, to provide nationally representative incidence, mortality, and excess mortality of osteoporotic vertebral fracture in South Korea. Moreover, this is one of the few studies worldwide presenting the excess mortality of osteoporotic vertebral fractures based on national health insurance claims data [810].

In this study, incidence of vertebral fracture was not increased between 974/100,000 in 2005 and 969/100,000 in 2008.

Although Korean has been considered as one of the countries with low risk of osteoporosis and osteoporotic fracture,[19, 20] we cannot compare directly the incidence of vertebral fracture in Korea with those of other countries, because incidence of vertebral fracture varies according to the definition of vertebral fracture (Table 4). The previous study in Switzerland, which performed using ICD-10 code for vertebral fracture from nationwide database, showed lower incidences of vertebral fracture than those in South Korea [21]. However, Switzerland has been known to have high risk of osteoporosis. This may explain why the different codes of ICD-10 for vertebral fracture were used. Actually, the incidence of hip fracture has been used for comparisons among countries [19, 20], because estimates of vertebral fracture depend on survey protocol, the time frame studies and the definition of vertebral fracture.
Table 4

Comparison of incidence (per 100,000) of osteoporotic vertebral fractures over 50 years old in different populations

 

Study period

Men

Women

Switzerland [21]

2000

243

486

Australia [25]

1996

70

210

Japan [26]

2004

134.7

322.5

South Korea (current study)

2008

431

1430

There have been several previous studies that have documented an increase in mortality risk following osteoporotic vertebral fracture [5, 810, 2224]. Most previous studies agreed that vertebral fracture is more prevalent in women than in men [5, 810, 23, 24]. However, differences in mortality risk between men and women after osteoporotic vertebral fracture is a controversial factor in previous studies [5, 810, 2224]. Our results showed that men have a substantially higher risk of mortality than women, and this was maintained after adjusting for general population by SMR.

It was interesting that the SMR of 55–59 years showed peak in men. Although the cause of the peak SMR in men is unclear, to reduce the mortality and socioeconomic burden following osteoporotic vertebral fracture in South Korea, a concentrated management is necessary, especially in these populations of elderly men.

This study had several limitations. First, computerized data in national claim database provide a potential bias based in claim and coding practices. However, computerized rate was above 99% during study periods. Second, bone mineral density (BMD) of patients was not available due to the study design based on National Claim Registry in this study. It was possible that vertebral fractures due to high-energy trauma were included in this study, because distinction between high and low-energy fractures could not be made by using ICD-10 coding system. Patients with vertebral fracture by high energy injury will have high mortality, and this can influence the results in this study. However, we used additional criteria of “aged 50 years or more” to exclude non-osteoporotic fractures. In addition, these inclusion criteria including ICD code and age without BMD measurement could be found in several studies on osteoporotic vertebral fracture [810]. Therefore, our study without BMD measurement could be justified. Third, in mortality analysis, there was a lack of consideration for medical condition, which can much influence the mortality in old populations. Thus, we cannot avoid the confounding effect on results imposed by different medical condition in this study. Forth, all patients with osteoporotic vertebral fractures may not be coded in this nationwide database. We evaluated only the patients who visited in clinic or hospital and were coded in National Claim Registry. The patients with non-symptomatic or mild symptomatic vertebral fracture may not be included, and this can influence the National Claim Registry-based incidence to be underestimated. The patients treated in hospital would have severe or multiple vertebral fractures, which would probably result in poor prognosis. This selection bias can influence the results based on the registry database. However, because the real situation of clinicians is same to that of our study based on National Claim Registry, we believe our results to be more relevant and informative to clinicians in the real clinical setting.

In spite of these limitations, the results of this study present that osteoporotic vertebral fracture were associated with mortality in South Korea, and provide a basic estimates of mortality after vertebral fracture. This is informative to establish appropriate general health strategies to reduce the mortality associated with vertebral fracture.

Acknowledgements

This study was supported by the Korean Health Insurance Review and Assessment Service (HIRA) and conducted using data from the Korean National Health Insurance Claims Database.

Conflicts of interest

This study was supported by the Korean Health Insurance Review and Assessment Service (HIRA) and conducted using data from the Korean National Health Insurance Claims Database.

Copyright information

© International Osteoporosis Foundation and National Osteoporosis Foundation 2011