Estimated prevalence of osteoporosis from a Nationwide Health Insurance database in Taiwan

Health Policy 75 (2006) 329–337

Estimated prevalence of osteoporosis from a Nationwide Health Insurance database in Taiwan Nan-Ping Yang, Chung-Yeh Deng, Yiing-Jenq Chou, Po-Quang Chen, Ching-Heng Lin, Pesus Chou, Hong-Jen Chang ∗ Institute of Public Health, National Yang-Ming University, Community Medicine Research Center, Taipei, Taiwan 112, Taiwan

Abstract Approximately 9.24% of the Taiwanese population is aged 65 years or older. Among them, osteoporosis is a major problem, along with it associated age-related fractures. We investigated the prevalence of osteoporosis in 1996–2001 by sampling Taiwan’s National Health Insurance (NHI) database. Data from 102,763 men (51.27%) and 97,654 women (48.73%) were evaluated. In this cohort, osteoporosis was recorded in each yearly dataset if the codes 733.0 or 733.00–733.09 were found on a search of the administrative or outpatient sub-databanks. A stable estimated prevalence of osteoporosis was calculated according to Taiwan’s NHI sampling data from 1999 to 2001. The results showed a trend toward increasing proportions of coded osteoporosis with age, more predominantly in the female population. The averaged prevalence of osteoporosis, between 1999 and 2001, in those aged ≥50 years was 1.63% for men and 11.35% for women. These estimates were lower than those reported elsewhere for Taiwan and for Japan but more equal to that in the Mexican American sub-population of the United States. In conclusion, the prevalence of osteoporosis is underestimated in the NHI database. Policymakers should be aware of this finding and allocate resources accordingly. © 2005 Published by Elsevier Ireland Ltd. Keywords: Osteoporosis prevalence; Taiwan

1. Introduction Osteoporosis is a metabolic bone disease characterized by low bone mass and micro-architectural deterioration of bone tissue. It is a major public health problem because of its association with age-related fractures. The cumulative lifetime risk of fracture after the age 50 years approaches 80% for women in West∗ Corresponding author. Tel.: +886 2 28229695; fax: +886 2 28201461. E-mail address: [email protected] (H.-J. Chang).

0168-8510/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.healthpol.2005.04.009

ern populations, and the for populations in Asia and Latin America are rapidly approaching this rate [1]. Osteoporotic fractures exact a terrible toll on the population with respect to morbidity, cost, and to a lesser extent mortality. These effects can lead to psychological problems, social consequences, functional limitations, and poor quality of life [2]. Relative mortality rates increase 1.2–2.3 times after vertebral fracture, and hip fracture lead to 10–20% excess deaths, mostly because of medical complications [3,4]. In 1995, direct medical expenditures for osteoporotic fractures in the United States were an estimated $13.8 billion [5]. In

330

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

1998, total direct and indirect costs of hospitalization, nursing home care, and other costs related to osteoporosis and osteoporotic fractures in the United States were $21.9 billion [6]. Worldwide, the total number of hip fractures in 1990 was estimated to be 1.26 million [7]. If no change in the age- and sex-specific incidences is assumed, the number is expected to approximately double to 2.6 million by the year 2025 and to increase to 4.5 million by the year 2050. In 1990, 26% of all hip fractures occurred in Asia; this figure could rise to 37% in 2025 and to 45% in 2050 as major demographic changes occur. [7]. Therefore, from the perspective of health policy, the issue of osteoporosis will become increasingly important in the future. Until now, the operational definition of osteoporosis by the World Health Organization (WHO) was restricted to white women, i.e., those with bone mineral density more than 2.5 standard deviations below the normal mean for young white women [8,9]. However, the exact prevalence of osteoporosis was difficult to estimate because of differences in sex, race, approaches to normalize for bone size, specific skeletal sites assessed, and diagnostic criteria [10]. Therefore, most epidemiological studies of osteoporosis, including those conducted in Taiwan [11], have focused on osteoporotic fractures [12,13]. To promote an accurate preventive programs for osteoporosis, descriptive epidemiological studies are needed. The purpose of this study was to estimate the prevalence of osteoporosis in Taiwan based on information from the National Health Insurance (NHI) database. Taiwan’s NHI, a universal health insurance program was implemented in 1995 and covers comprehensive services. More than 97% of the 23 million people of Taiwan have benefited from this program, and the NHI consistently receives a 70% public satisfaction rate [14,15]. The NHRI provided a database of 200,432 random subjects, about 1% of the population, to perform a related health insurance study.

Health Research Institutes cooperates with the NHI Bureau (NHIB) to establish an NHI research database. How to establish the research database? First, all citizens who have established a registered domicile for at least in the Taiwan area should be enrolled in NHI. Second, all the individuals included in the entire claims database (the general population) were given various random numbers by using a random number function. Simple random sampling of about 50,000 people at a time was performed in 2000. Third, the academic sampling database consisted of four simple randomly sampled subsets and finally included 200,432 enrollees. Forth, the database of medical claims of this sample group, including ambulatory care, inpatient care, dental services, and prescription drugs, were traced back since 1996. Therefore, an academic retrospective cohort group was established. Finally, this academic cohort database was tested and noted without differences in age, sex, and medical costs from all enrollees [16]. The prevalence of osteoporosis was evaluated for the 6 years from 1996 to 2001 on the basis of the above academic cohort database. 2.2. Definition of osteoporosis Actually, the diagnostic coding of NHI in Taiwan is according to the International Classification of Diseases, Ninth Clinical Modification (ICD-9-CM). According to the definition on the Taiwan NHI’s Fee Schedule for Medical Services and the Reference List for Drugs, osteoporosis means a value for lumbar or hip BMD, tested by DXA, that is more than 2.5 S.D. below the young adult mean value. It is follow the WHO’s rules in 1994. The studied cohort population was recorded as having osteoporosis in each yearly dataset if the ICD-9-CM codes 733.0 (osteoporosis, OS) or 733.00 (OS, unspecified), 733.01 (senile OS), 733.02 (idiopathic OS), 733.03 (disuse OS), and 733.09 (OS, others) were found during a search of the administrative or outpatient sub-databanks. Only data from people aged 30 years or older were included.

2. Methods 2.1. Data source

2.3. Quality control of medical services and codes in Taiwan’s NHI system

In Taiwan, the NHI Plan has accumulated 23.75 million administrative and claims records, forming the largest such collection in the world. The National

NHIB had established a uniform system to control the quality of medical services and codes. If medical services provided by the contracted medical care in-

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

stitutions to the beneficiaries were determined by the Professional Peer Review Committee to be incompatible with the provisions of the NHI Act, the expenses thereof shall be borne by the contracted medical care institutions themselves. In case the drug, laboratory tests or diagnostic examination is provided by other contracted medical care institutions in accordance with the physician’s instruction, and the Insurer, after the examination according to the rules of examination described in the preceding article, decides not to pay the benefits due to the physician’s improper instruction, such expenses incurred thereof shall be borne by the medical institution where the physician practices. Otherwise, there shall be a Disputes Settlement Board established under the National Health Insurance to settle disputes arising from cases approved by the Insurer, and raised by the insured, the group insurance applicants or the contracted medical care institutions.

331

Fig. 1. Trends of osteoporosis prevalence based on Taiwan’s NHI database, 1996–2001.

aged 30 years or older were 45.46%, 47.28%, 48.95%, 50.75%, 52.62%, and 54.45%, respectively (Table 1). The prevalence of osteoporosis increased with age, especially after the age of 50 years, in 1996–2001, as reflected in Taiwan’s NHI data (Fig. 1). An unusually low percentage of osteoporosis, as coded in the database, was observed in the initial 3 years of the NHI program in Taiwan. This finding perhaps resulted from the unsatisfactory participation rate and the lack of public coding techniques in the early phase of the NHI program. However, the stratified prevalence of osteoporosis from the NHI data could be determined beginning

3. Results 3.1. Prevalence of osteoporosis in Taiwan A total of 200,417 insured people in Taiwan were selected for the cohort, which included 102,763 men (51.27%) and 97,654 women (48.73%). Table 1 shows the demographic information by age strata and sex. With aging, the proportion of subjects in each age strata changed: from 1996 to 2001, proportions of people Table 1 Distribution of Taiwan NHI sampling data, 1996–2001 Age stratum (years)

1996 Men

Women

<29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 ≥80

27.98 (56082) 4.56 (9142) 4.01 (8029) 3.60 (7211) 2.35 (4716) 1.78 (3558) 1.72 (3441) 1.58 (3171) 1.56 (3127) 1.13 (2266) 0.60 (1199) 0.41 (821)

26.56 (53225) 4.31 (8644) 4.01 (8045) 3.50 (7011) 2.33 (4679) 1.77 (3540) 1.74 (3484) 1.45 (2903) 1.18 (2363) 0.85 (1698) 0.52 (1051) 0.50 (1011)

Total

51.27 (102763)

48.73 (97654)

Data are percentages; data in parentheses are numbers of people.

1997

1998

1999

2000

2001

52.72 8.98 8.15 7.27 5.42 3.46 3.58 3.03 2.86 2.13 1.31 1.08

51.05 8.92 8.37 7.43 5.94 3.58 3.62 3.09 2.99 2.24 1.50 1.27

49.25 8.91 8.02 7.65 6.34 3.83 3.67 3.20 2.99 2.43 1.65 1.49

47.38 8.97 8.92 7.87 6.73 4.16 3.67 3.35 3.01 2.57 1.82 1.74

45.55 8.99 8.77 8.02 7.09 4.69 3.54 3.46 3.03 2.74 1.98 2.04

Total 54.54 (109307) 8.87 (17786) 8.02 (16074) 7.10 (14222) 4.69 (9395) 3.54 (7098) 3.46 (6925) 3.03 (6074) 2.74 (5490) 1.98 (3964) 1.12 (2250) 0.91 (1832) 100.0 (200417)

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

332

Table 2 Three-year comparison of osteoporosis selected from Taiwan NHI sampling data by ICD-9CM codes, 1999–2001 Age stratum (years)

30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 ≥80

1999

2000

2001

1999–2001

Male

Female

Male

Female

Male

Female

Male

Female

0.08 0.15 0.22 0.25 0.84 0.92 1.47 1.72 2.32 2.47 2.54

0.25 0.60 1.94 6.10 10.79 11.31 9.98 11.85 11.39 11.21 8.38

0.09 0.12 0.25 0.53 1.00 1.09 1.63 2.03 2.65 2.22 3.08

0.21 0.75 2.19 6.03 12.80 13.62 12.38 13.27 13.71 14.45 9.41

0.08 0.21 0.22 0.44 0.66 0.79 1.42 2.49 2.62 1.81 1.63

0.33 0.66 2.13 6.55 9.40 12.74 11.13 11.92 11.48 8.76 5.43

0.08 0.16 0.23 0.41 0.82 0.93 1.51 2.08 2.54 2.15 2.36

0.26 0.67 2.09 6.23 10.93 12.56 11.18 12.35 12.19 11.41 7.60

in the fourth year of the NHI program. A stable estimated prevalence of osteoporosis in Taiwan could be calculated according to the NHI sampling data from 1999 to 2001. Table 2 shows the proportions of coded osteoporosis in these 3 years from 1999 to 2001 by sex and age group. In general, we noted a trend toward increasing proportion of coded osteoporosis with age, more predominantly in the female population. Cases recoded as osteoporosis during 1999–2001 were added together to yield the estimated annual prevalence of osteoporosis in Taiwan for every age stratum.

3.2. The trend of estimated prevalence of osteoporosis compared to the utility of BMD examinations, 1999–2001 Based on the NHI’s cost databank, the summed volumes of DXA examinations from 1999 to 2001 had been reported [17], that was showed on Table 3. The data of self-paid DXA evaluations was not included, and whenever it had not been collected in Taiwan. Therefore, the utility of DXA scans in the NHI system was calculated [18] and, in general, markedly low utility of the BMD examinations paid by the NHI

Table 3 The nationwide utility of DXA examinations provided by NHI in Taiwan, during 1999–2001 Age stratum (years)

30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 ≥80 a b

The volumes of DXA examinationsa , 1999–2001

The utility of DXA per 10,000 personsb , 1999–2001

Male

Female

Male

Female

103 194 347 571 717 757 1089 1542 2033 1582 1351

410 899 2423 6902 22258 19503 18631 17312 14559 9771 6924

0.36 0.65 1.24 2.31 4.30 6.18 9.73 15.21 20.88 25.72 30.84

1.48 3.13 8.90 28.48 134.64 156.62 157.06 177.60 190.00 189.28 146.55

The data was calculated based on the nationwide NHI’s cost databank, coded as 33064B [17]. The total population in different gender and age stratum was based on the published data by the government in the every end of year [18].

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

Fig. 2. The trend and comparison of estimated prevalence of osteoporosis and utility of BMD examination between both genders in Taiwan based on NHI data, 1999–2001.

was surprised and unsatisfied, but a sudden increased utility of DXA scans was found after the age of 50 years, especially for the female group (Table 3). When compared with the estimated prevalence of osteoporosis between both genders (Fig. 2), there were similar trend with increased age. A nationwide homogeneous under-evaluation for osteoporosis possibly had happened and the under-estimation of osteoporosis would be occurred. 3.3. Comparison of global osteoporosis rates Nationwide surveys of osteoporosis are rarely reported. Table 4 shows the prevalence of osteoporosis from some large-sample studies in which subjects were screened by using the criterion standard dual X-ray absoptiometry (DXA) or by using more popular alternative of quantitative ultrasonography (QUS). In the present study, the annual prevalence of osteoporosis in Taiwanese men and women aged 50 years or older during the 3-year-period of 1999–2001 was 1.63% and 11.35%, respectively. The prevalence significantly differed between the sexes for every 10-year age group. To understand the validity of this estimated prevalence, we conducted a parallel comparison with results of similar and reliable studies in Taiwan, Asia, or Western country. A community-based survey of calcaneal QUS in three communities was performed in 1997–1998. The

333

design involved cluster samples in an even geographical distribution in Taiwan. The overall response rate was 50.2% among the 6322 participants [19]. In another study, investigators reviewed the medical records from annual health examinations over a 3-year-period at a single institution in Taipei, the capital city of Taiwan. The authors recruited 1514 men and 1955 women who had no major systemic disorders and who had undergone lumbar and proximal femoral DXA [20]. A multi-center survey based on calcaneal QUS at 65 healthcare centers in 11 of the 48 prefectures in Japan was performed from April 1994 to March 1996 [21]. This was a large-sample survey consisting of 12,201 Japanese female subjects. In Japan, another large-scale epidemiological study with representative samples of the Japanese women was the Japanese PopulationBased Osteoporosis Study [22]. Of the 4550 subjects selected in the original study, 3465 Japanese women underwent an evaluation of bone density with DXA and were followed up. Two cut-off values for the diagnosis of osteoporosis based on spinal or hip bond mineral density, according to the WHO criteria and the Japanese Society of Bone and Mineral Research criteria, resulted in a range in the prevalence of osteoporosis. A total of 11,359 Danish subjects (1426 men and 9933 women) who had been screened on a Hologic 2000 DXA machine from 1990 to 2000, and osteoporosis was defined as a T-score < −2.5, calculated from the published Danish reference values, in either the hip or spine [23]. In the United States, the National Center for Health Statistics and Centers for Disease Control and Prevention conducted the third National Health and Nutrition Examination Survey between 1988 and 1994. Bone mineral measurements were performed in 14,646 men and women aged 20 years or older in the full survey [24]. The total number of people aged 50 years or older was 4981 in this nationwide survey. Depending on whether sex-specific T-scores for white populations or other race/ethnic groups were used, different prevalence calculated for the African American and Mexican American sub-populations. When we compared the rates of osteoporosis from the global population surveys, our estimated prevalence from Taiwan’s NHI database was lower than those reported elsewhere in Taiwan, in Asia (specifically Japan) and in Western country. However, the rates were closer to that reported in the Mexican American sub-population of the United States.

334

Table 4 Prevalence of osteoporosis investigated by using various methods in the most recent studies

Age stratum (years) 50–59 60–69 ≥70 Overall, ≥50 Race (overall, ≥50) African American Mexican American White

NHI data, Taiwan 1999–2001

Taiwan published in 2003, 2004

Asia published in 1999, 2001

Western country published in 2004, 2005

Male

Female

Reference

Male

Female

Reference

Female

Reference

Male

Female

Reference

0.9 1.8 2.4 1.6

11.7 11.7 10.5 11.4

Current study Current study Current study Current study

1.5 3.0 6.5 3.4

7.9 21.7 34.5 19.5

[19]a [19]a [19]a [20]d

5.2, 1.5–18.3 18.7, 9.2–34.5 43.6, 26.7–45.7 NA, 11.6–38.0

[21]b , [22]c [21]b , [22]c [21]b , [22]c [22]c

17.7

40.8

[23]e

2.3–6.0 1.5–3.3 3.9

4.8–9.8 10.4–14.8 13.7

[24]f [24]f [24]f

Data are percentages (range) unless otherwise noted. NA, not available. a Community-based survey of 6322 people (3691 F, 2631 M) in Taiwan; measurements based on QUS. b Nationwide data collection in 12,204 female people in Japan; measurements based on QUS. c Cohort population survey of 3465 female people in Japan; measurements based on DXA. d Population-based health examination of 3469 people (1955 F, 1514 M) in Taiwan; measurements based on DXA. e Screened data of 11,359 people (9993 F, 1426 M) in Denmark; measurements based on DXA. f Third National Health and Nutrition Examination Survey of 4981 people (2590 F, 2391 M) in the United States; measurements based on DXA.

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

Group

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

According to the Denmark National Hospital Discharge Register, only 5925 Danish subjects having a diagnosis of osteoporosis (ICD10: M80, M81, and M82) in 1999, although the Danish population totalled 5,313,568 in 1999. It seems that osteoporosis is markedly under-diagnosed [23]. The prevalence of osteoporosis form the NHI database maybe was underestimated.

4. Discussion 4.1. Osteoporosis identified by ICD codes Previous investigators have used the International Classification of Diseases codes to select cases of osteoporosis from global databases. For example, to estimate the cost of osteoporosis in California, one group identified osteoporosis by using ICD-9CM codes 733.01–733.09 from the California Hospital Discharge data and combined the information with that from databases from the 1987 National Medical Expenditure Survey, the 1996 Medical Expenditure Panel Survey, and the 1997 National Nursing Home Survey [25]. In France, a similar study to estimate the annual costs of osteoporosis and osteoporotic fractures was based on cases identified by using codes M80.2–M81.9 from the International Classification of Diseases, Tenth Revision (ICD10) [26]. Another nationwide study about osteoporosis in Denmark was still based on the database of National Hospital Discharge Register by the ICD10 codes [23]. 4.2. Role of QUS to study the prevalence of osteoporosis For almost 2 decades, QUS has proven to be widely and clinically useful for assessing the status of bone, especially the calcaneus [27]. Ultrasonography has great potential for widespread use owing to its portability, low cost, and lack of ionizing radiation. By 1999, the US Food and Drug Administration (FDA) had approved five QUS devices for routinely diagnosing osteoporosis, for determining fracture risk, and for monitoring bone changes [28]. Now, QUS instruments are used in hospitals, clinics, and communities. QUS measurement of the calcaneus is also

335

accepted as a useful tool for gross surveys of bone mass [19,21,29,30]. 4.3. The barriers for screening and intervention of osteoporosis in Taiwan Aging of the populations in Taiwan and elsewhere have renewed interest in the diagnosis, treatment, and costs of osteoporosis. An island country of about 36,000 km2 , Taiwan has a population of over 22 million, most of whom are Chinese. Nearly 9.24% of this population is elderly (aged 65 years or older), and their life expectancy is rapidly increasing [31]. In this older population, osteoporosis and related fractures are a major health problem. There were different barriers for the screening and intervention of osteoporosis in Taiwan, including uneven geographic location of DXA scanners, varied qualified staff to operate the scanners, and inadequate running cost, especially provided by the insurance et al. Osteoporosis is sometimes termed the “silent epidemic” because osteoporosis is usually asymptomatic, unless significant bone loss became evident only after a hip or vertebral fractures occurred. In Taiwan, accounting for the economic loading for the NHI, screening of osteoporosis was not supported in policy. Most of the enrollees of Taiwan NHI must pay the cost (about $US 20.0/time) of DXA evaluation by themselves. According to the rules of the Taiwan NHI’s Fee Schedule for Medical Services and the Reference List for Drugs, DXA examination paid by the insurance was limited to the people with established osteoporosis (it means potential osteoporosis combined with obvious spinal or hip fractures) or the defined osteoporosis subjects who had been receiving the treatment of antiosteoporotic drugs. However, most physicians in Taiwan, especial the orthopedic surgeons, often diagnosis the fractures of the older population and not the underlying disease (osteoporosis), so the actual number of osteoporosis may be even higher, especially in the older population. 4.4. Underestimation of osteoporosis and osteoporotic fractures In Taiwan, antiosteoporotic drugs were provided and paid by NHIB stepwise. The earliest therapeutic drug for osteoporosis was salmon calcitonin (paid since October, 1996), but that was just assigned to

336

N.-P. Yang et al. / Health Policy 75 (2006) 329–337

the osteoporotic fractured subjects. More open policy about the treatments for osteoporosis was processed after 1999. Those drugs for osteoporosis, including alendronate (paid since November, 1998), calcium supplement (paid since February, 2000), and raloxifene (paid since April, 2001), maybe promote the diagnosis of osteoporosis more popular, because they must be ordered at least with one of the codes of 733.0 and its series. That maybe the major one of the causes of a sudden increase in the prevalence of osteoporosis in Taiwan in the latter 3 years (1999–2001), compared to the former 3 years (1996–1998). In this nationwide epidemiological study, we have demonstrated that osteoporosis is underestimated in Taiwan when based on the NHI database. In the past, osteoporosis received less attention and was under-diagnosed and under-treated. A crosssectional survey of 850 primary care physicians in Spain [32] showed that only 4% of physicians reported that their primary care center had implemented specific programs for osteoporosis and that the diagnosis of osteoporosis was made always personally in 25% of cases. In Europe, a survey conducted by the International Osteoporosis Foundation in 2000 revealed that less than 50% of women with osteoporosis were aware of their risk before the diagnosis. Therefore, the European Union Osteoporosis Consultation Panel, formed in 2002, published the document entitled “Osteoporosis in the European Community: Action Plan” with financial support from the European Commission [33]. Vertebral fractures, the most common type of osteoporotic fractures, are associated with significant morbidity and mortality. The fractures frequently do not come to clinical attention, being recognized in only one-third to one-fourth of patients in clinical trials [34]. Under-diagnosis of vertebral fractures is also a worldwide problem in North America (45%), Latin America (43%), and Europe (27%) [34]. In Canada, Hajcsar et al. [35] performed 2694 fracture clinic visits in 1996–1997 and identified 228 patients (8.4%) with fragility-type fractures. Only 18.5% of these patients had undergone examination and adequate treatment of osteoporosis at 1-year follow-up. In the Netherlands, Wilkinson et al. [36] evaluated 231 patients with hip fracture and found that only 10% had undergone a bone density examination and that 9–12% were receiving medical therapy. In a 6-year-study (1995–2000) based on a national phar-

maceutical benefits program, Solomon et al. [37] noted an average 3% increase in the use of osteoporosis medications during the 6 months before a fracture versus the 6 months after the fracture. Only 6–22% of elderly patients with hip or wrist fractures filled their prescription for an osteoporosis medication in the 6 months after a fracture. Lastly, a National Ambulatory Medical Care Survey from 1993 to 1997 conducted by Gehlbach et al. [38] revealed that less than 2% of women 60 years or older received diagnoses of osteoporosis or vertebral fracture, though the expected prevalence was 20–30%. Appropriate drug treatment was offered to only 36% of the patients in whom a diagnosis was made. 4.5. Future work on osteoporosis Life expectancy is increasing worldwide, and the number of elderly individuals is rising in every geographic region. Therefore, health-related costs of osteoporosis and osteoporotic fractures are also expected to rise in future generations. The issue of osteoporosis will truly become a global problem over the next half century. In Taiwan, more than 200 DXA machines are in use [39]. The Taiwan Osteoporosis Association, an academic society, was established in 1997 [40] and is also a member of the International Osteoporosis Federation and the Asian Pacific Osteoporosis Society. However, no national project focuses on the aforementioned problems in Taiwan. The under-diagnosis of osteoporosis in Taiwan, as shown in our study, underscores the need for policymakers to attend to these problems and to allocate resource accordingly. References [1] Cooper C. Osteoporosis: an increasing public health problem. Medicographia 2002;24:277–84. [2] Silverman S. Minimizing pain, maximizing quality of life. Osteoporosis International 2002;13(Suppl. 1):S158. [3] Melton III LJ. Adverse outcomes of osteoporotic fractures in the general population. Journal of Bone and Mineral Research 2003;18:1139–41. [4] Lin JT, Lane JM. Osteoporosis: a review. Clinical Orthopaedics 2004;425:126–34. [5] Ray NF, Chan JK, Thamer M, Melton III LJ. Medical expenditures for the treatment of osteoporotic fractures in the United States in 1995: report from the National Osteoporosis Foundation. Journal of Bone and Mineral Research 1997;12:24–35.

N.-P. Yang et al. / Health Policy 75 (2006) 329–337 [6] Max W, Sinnot P, Kao C, Sung HY, Rice DP. The burden of osteoporosis in California. Osteoporosis International 2002;13:493–500. [7] Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporosis International 1997;7:407–13. [8] Kanis JA, Melton III LJ, Christiansen C, Johnston CC, Khaltaev N. Perspective: the diagnosis of osteoporosis. Journal of Bone and Mineral Research 1994;9:1137–41. [9] Genant HK, Cooper C, Poor G, Reid I, Ehrlich G, Kanis J, et al. Interim report and recommendations of the World Health Organization task-force for osteoporosis. Osteoporosis International 1999;10:259–64. [10] Melton III LJ. The prevalence of osteoporosis: gender and racial comparison. Calcified Tissue International 2001;69:179–81. [11] Tsai KS, Tai TY. Epidemiology of osteoporosis in Taiwan. Osteoporosis International 1997;8(Suppl. 3):S96–8. [12] Cooper C, Melton III LJ. Epidemiology of osteoporosis. Trends in Endocrinology and Metabolism 1992;3:224–9. [13] Cooper C. Epidemiology of osteoporosis. Osteoporosis International 1999;10(Suppl. 2):S2–8. [14] Lu JF, Hsiao WC. Dose universal health insurance make healthcare unaffordable? Lessons from Taiwan. Health Affairs 2003;22(3):77–88. [15] Cheng TM. Taiwan’s new National Health Insurance Program: genesis and experience so far. Health Affairs 2003;22(3):61–76. [16] National Health Insurance Research Database. Available at: http://www.nhri.org.tw/nhird/date cohort.htm, accessed November 12, 2004. [17] NHI Dispute Mediation Committee. The argument about examinations and treatments of osteoporosis. Available at: http://www.dmc.doh.gov.tw/doh.html, accessed April 17, 2005 [in Chinese]. [18] Department of Statistics, Ministry of the Interior, Taiwan. Statistical yearbook of interior, 1999–2001. Available at: http://www.moi.gov.tw/stat/english/index.asp, accessed April 17, 2005 [in Chinese]. [19] Yang NP, Lin T, Wang CS, Chou P. Community-based survey of low quantitative ultrasound values of calcaneus in Taiwan. Journal of Clinical Densitometry 2003;6:131–41. [20] Chan WP, Liu JF, Chi WL. Evaluation of bone mineral density of the lumbar spine and proximal femur in population-based routine health examinations of healthy Asians. Acta Radiologica 2004;45(1):59–64. [21] Ishikawa K, Ohta T. Radial and metacarpal bone mineral density and calcaneal quantitative ultrasound bone mass in normal Japanese women. Calcified Tissue International 1999;65:112–6. [22] Iki M, Kagamimori S, Kagawa Y, Matsuzaki T, Yoneshima H, Marumo F. Bone mineral density of the spine, hip and distal forearm in representative samples of the Japanese female population: Japanese Population-Based Osteoporosis Study (JPOS). Osteoporosis International 2001;12(7):529–37. [23] Vestergaard P, Rejnmark L, Mosekilde L. Osteoporosis is markedly underdiagnosed: a nationwide study from Denmark. Osteoporosis International 2005;16:134–41.

337

[24] Broussard DL, Magnus JH. Risk Assessment and screening for low bone mineral density in a multi-ethnic population of women and men: dose one approach fit all? Osteoporosis International 2004;15:349–60. [25] Max W, Sinnot P, Kao C, Sung HY, Rice DP. The burden of osteoporosis in California. Osteoporosis International 2002;13:493–500. [26] Levy P, Levy E, Audran M, Cohen-Solal M, Fardellone P, Le Parc JM. The cost of osteoporosis in men: the French situation. Bone 2002;30:631–6. [27] Stewart A, Reid DM. Quantitative ultrasound in osteoporosis. Seminars in Musculoskeletal Radiology 2002;6:229–32. [28] Gonnelli S, Cepollaro C. The use of ultrasound in the assessment of bone status. Journal of Endocrinological Investigation 2002;25:389–97. [29] Thuy VT, Chau TT, Cong ND, De DV, Nguyen TV. Assessment of low bone mass in Vietnamese: comparison of QUS calcaneal ultrasonometer and data-derived T-scores. Journal of Bone and Mineral Metabolism 2003;21:114–9. [30] Kim CH, Kim YI, Choi CS, Park JY, Lee MS, Lee SI, Kim GS. Prevalence and risk factors of low quantitative ultrasound values of calcaneus in Korean elderly women. Ultrasound in Medicine and Biology 2000;26:35–40. [31] Ministry of the Interior. 2003 Taiwan-Fukien demographic fact book. Taipei: Ministry of the Interior; 2004. [32] Perez-Edo L, Recasens MC, Castelo-Branco C, Lopez PO, Marques AG, Perez C, Dalmau JM. Management of osteoporosis in general practice: a cross-sectional survey of primary care practitioners in Spain. Osteoporosis International 2004;15: 252–7. [33] Compston J. Action plan for the prevention of osteoporotic fractures in the European community. Osteoporosis International 2004;15:259–62. [34] Delmas PD. Under-diagnosis of vertebral fractures. Osteoporosis International 2002;13(Suppl. 3):S3 [Abstract]. [35] Hajcsar EE, Hawker G, Bogoch ER. Investigation and treatment of osteoporosis in patients with fragility fractures. Canadian Medical Association Journal 2000;163:819–22. [36] Wilkinson TJ, Sainsbury R, Gilchrist NL. Osteoporosis followup after fracture. Archives of Internal Medicine 2001;161:2633 [Comment]. [37] Solomon DH, Finkelstein JS, Katz JN, Mogun H, Avorn J. Underuse of osteoporosis medications in elderly patients with fractures. American Journal of Medicine 2003;115(5):398– 400. [38] Gehlbach SH, Fournier M, Bigelow C. Recognition of osteoporosis by primary care physicians. American Journal of Public Health 2002;92(2):271–3. [39] Taiwanese Osteoporosis Association. Brief Communications of the Taiwanese Osteoporosis Association. Taipei: Taiwanese Osteoporosis Association; 2003. p. 10. [40] Chen PQ. The current strategy on the prevention and treatment of osteoporosis in Taiwan. Journal of Bone and Mineral Metabolism 2002;20:131–5.