Arch Osteoporos (2013) 8:161 DOI 10.1007/s11657-013-0161-0
ORIGINAL ARTICLE
Testing for vitamin D and other trends in the clinical management of osteoporosis David Rowell & Louisa Gordon
Received: 30 September 2013 / Accepted: 11 October 2013 / Published online: 21 November 2013 # International Osteoporosis Foundation and National Osteoporosis Foundation 2013
Abstract Summary This descriptive paper identifies some salient trends in the Australian management of osteoporosis. Changes in pharmaceutical consumption and medical utilisation are analysed. The total national consumption of four pharmaceuticals is estimated. From 2001 to 2011, the consumption of bisphosphonates and vitamin D increased. After 2006, the management rate for osteoporosis decreased. Purpose During the first decade of the millennia, the total cost of vitamin D tests has increased significantly within many health care jurisdictions worldwide. Australia reports a 100fold increase. Legitimately, concerns regarding the economic effectiveness of this test exist. The purpose of this paper is to identify what other salient clinical trends could affect the efficacy of this test when treating osteoporosis. Methods Longitudinal data from two Australian datasets are analysed. The first are data obtained from Medicare Australia, which report pharmaceutical consumption. The second are data obtained from a national survey of general practitioners. Results The management of osteoporosis has been characterised by a movement away from single to combination formularies. The consumption of calcium carbonate has declined, and the use of bisphosphonates has increased. While the gross consumption of risedronate increased steadily over the decade (8.4 to 186.5 kg), the consumption of alendronate declined after 2007. The consumption of vitamin D (over-the-counter and prescription) has increased from 0.58 to 2.8 kg over the decade. While
D. Rowell (*) : L. Gordon Centre for Applied Health Economics, Griffith Health Institute, Griffith University, Logan Campus, University Drive, Meadowbrook, Queensland 4131, Australia e-mail: [email protected] D. Rowell : L. Gordon Centre for Research Excellence in Sun and Health (CRESH), Queensland University of Technology, Queensland 4029, Australia
prescription vitamin D comprised just 10 % of the total, its consumption has undergone a 20-fold increase since 2006. Importantly, we can also report that the management of osteoporosis by Australia's general practitioners increased steadily until 2007–2008, before declining by 36 %. Conclusions Further research is required to determine possible casual relationships that may exist between these data. Until a formal economic evaluation of vitamin D testing in this clinical setting is completed, no conclusions regarding its economic effectiveness should be drawn. Keywords Osteoporosis . Treatment of osteoporosis . Vitamin D . Vitamin D testing
Introduction Low levels of 25-hyrodxyvitamin D (25OHD) can cause osteoporosis [1], and treatment with vitamin D is recommended to maintain calcium and phosphate homeostasis to optimise bone and muscle function [1]. Evidence from randomised controlled trials has demonstrated that vitamin D with calcium supplementation can reduce the risk of falls and fractures [1, 2]. Furthermore, treatment protocols recommend that testing for serum levels of 25OHD should be considered in patients with a Z-score less than or equal to −2.0 on dualenergy X-ray absorptiometry (DXA) or if a cause of secondary bone loss is clinically suspected [3]. Evidence derived from observational studies has also implicated hypovitaminosis D in the pathogenesis of an increasing diverse range of diseases, including breast and colon cancer, autoimmune diseases, such as multiple sclerosis, and depression [1]. However, concern has been raised regarding the rapid increase in the costs of 25OHD testing. Sattar et al. [4] state that Scotland has seen a doubling of 25OHD tests from 2008 to 2010 and a hospital in the UK reports a sixfold increase in tests from 2007 to 2010. Bilinski and Boyages [5] report that
161, Page 2 of 5
the cost of 25OHD testing in Australia has increased from Australian $1.02 million to Australian $96.75 million from 2000 to 2010. While there is a consensus that the increased number of 25OHD tests may not be justifiable, there is some divergence in the tone of conclusions. While Sattar et al. [4] “…urge all clinicians to stop and think critically before measuring 25OHD, particularly in conditions not related to bone disease”, Bilinski and Boyages [6] assert that “[t]hese findings [unstainable growth in vitamin D testing] have widespread consequences in terms of quality of care, unnecessary cost, and potential over diagnosis”. An economic evaluation, which considers both costs and benefits, is necessary to fully appraise the net benefit of any given intervention. Conclusions regarding the costeffectiveness of a treatment derived solely from a cost-ofillness study are implicitly reliant on the caveat, all other things being equal. Given that the most established indication for prescribing vitamin D remains the treatment of osteoporosis [1], the motivation for this paper is to explore what changes may have occurred in the management of this disease. The aim of this paper is to present clinical data reporting the salient trends in the management of osteoporosis alongside financial data, which reflects the exponential increase in the cost of 25OHD tests, during the last decade.
Materials and methods Osteoporosis is a chronic medical condition that is treated with pharmaceuticals prescribed by Australia's general practitioners (GPs). Our analysis therefore considers both inputs (pharmaceuticals) and outputs (management rates of osteoporosis). Firstly, we analyse pharmaceutical data obtained from Australia's Pharmaceutical Benefits Scheme (PBS), which summarises the consumption of all prescription medications provided through this insurance scheme.
Arch Osteoporos (2013) 8:161
Secondly, we analyse data obtained from a national survey of Australia's general practitioners conducted by Bettering the Evaluation and Care of Health (BEACH) [7–12]. This annual survey of 1,000 general practitioners reports clinical activity per 100 doctor–patient encounters. Longitudinal data for the years 1998 to 2011 are reported conjointly to highlight changes in the management of osteoporosis in Australia.
Results Figure 1 reports on the consumption of four pharmaceuticals commonly used to treat osteoporosis, the bisphosphonates (1) alendronate and (2) risedronate, (3) calcium carbonate and (4) colecalciferol (vitamin D3). The left-hand axis reports the number of scripts consumed for each available formulation, and the right-hand axis reports the cost of the subsidy for vitamin D testing, in Australian dollars as reported by Bilinski and Boyages [5]. The data presented in Fig. 1 illustrate that while the costs of all vitamin D testing increased 100-fold, there was a significant change in the management of osteoporosis with the addition of pharmaceuticals, which combined high-dose colecalciferol and bisphosphonates, after 2006. This change was associated with a more aggressive management of osteoporosis. The use of single-agent bisphosphonates and calcium carbonate declined. The maximum available strength of risedronate increased from 35 mg in 2002 to 150 mg in 2008. While the sales of over-the-counter (OTC) colecalciferol, available in doses of 5 to 25 μg, accelerated steadily over the decade, doses of up to 140 μg became available in combination therapies available on prescription toward the end of the decade. Aggregate annual doses of risedronate, alendronate, calcium carbonate and colecalciferol supplied through the PBS were estimated and reported in Fig. 2 as follows:
Aggregate dose ¼ ΣðDose per tablet Tablets per script Scripts per yearÞ:
Figure 2 reports both the absolute and relative changes for each pharmaceutical consumed, for the years 2001 to 2012. The vertical axis reports the change in pharmaceutical consumption as a percentage of the baseline. The legend lists each pharmaceutical with aggregate consumption at commencement and completion of the time series. For example, the total consumption of risedronate increased from 8.4 to 186.5 kg from 2001 to 2012, which was a 2,200 % increase. Also included is an estimate of the consumption of OTC vitamin D. For the years 2006 to 2011, annual market sales figures provided by Euromonitor International [13] were
converted to kilogrammes per year using a retail price of Australian $0.72 per microgram derived from a convenience sample obtained from a popular supermarket. For the years 2001 to 2006, a reported 13 % growth in sales of all OTC vitamins by Blackmores®,1 a market leader, was used as a proxy for changes in vitamin D consumption. The graph for OTC colecalciferol is “dashed” for the years 2001 to 2006 to indicate a trend estimate and “solid” for the years 2006 to 2011 to indicate reported data. 1 Blackmores http://www.blackmores.com.au/about-blackmores/ investors-centre/quarterly-results
Arch Osteoporos (2013) 8:161 2.5
120
80 1.5 60 1 40
0.5
Cost of Vitamin D Tests ($m)
100
2 PBS Services (millions)
Fig. 1 Changes in pharmaceutical use and cost of 25OHD testing (2001 to 2011). Risedronate Na (4443W, 4444X, 8481J, 8621R, 8972F, 9391G); Risedronate Na, CaCO3 and Colecalciferol (4380M, 8974H); Risedronate Na and CaCO3 (8899J, 8973G); Alendronate Na and Colecalciferol (9012H, 9183H); Alendronate Na, Colecalciferol and CaCO3 (39351E) for years 2001 to 2012 ([5, 17])
Page 3 of 5, 161
20
0
0 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Risedronate Na (5mg)
Risedronate Na (35mg)
Risedronate Na (30mg)
Risedronate Na (150mg)
Risedronate Na & CaCO3 (35mg/1.25g)
Risedronate Na, CaCO3 & Colecalciferol (35mg/2.5g/22ug)
Alendronate Na (40mg)
Alendronate Na (70mg)
Alendronate Na & Colecalciferol (70mg/70ug)
Alendronate Na & Colecalciferol (70mg/140ug)
Alendronate Na, Colecalciferol & CaCO3 (70mg/140ug/1.25g)
CaCO3 (1.25g & 1.5g)
Vitamin D tests ($m)
if years is beyond 2006 and an interaction term Year * Post_2006. The regression output is reported in Table 1. The null hypothesis that the coefficients for Post_2006 and interaction term Year * Post_2006 were jointly equal to 0 was rejected (F (2,10)=23.00 (p
ORIGINAL ARTICLE
Testing for vitamin D and other trends in the clinical management of osteoporosis David Rowell & Louisa Gordon
Received: 30 September 2013 / Accepted: 11 October 2013 / Published online: 21 November 2013 # International Osteoporosis Foundation and National Osteoporosis Foundation 2013
Abstract Summary This descriptive paper identifies some salient trends in the Australian management of osteoporosis. Changes in pharmaceutical consumption and medical utilisation are analysed. The total national consumption of four pharmaceuticals is estimated. From 2001 to 2011, the consumption of bisphosphonates and vitamin D increased. After 2006, the management rate for osteoporosis decreased. Purpose During the first decade of the millennia, the total cost of vitamin D tests has increased significantly within many health care jurisdictions worldwide. Australia reports a 100fold increase. Legitimately, concerns regarding the economic effectiveness of this test exist. The purpose of this paper is to identify what other salient clinical trends could affect the efficacy of this test when treating osteoporosis. Methods Longitudinal data from two Australian datasets are analysed. The first are data obtained from Medicare Australia, which report pharmaceutical consumption. The second are data obtained from a national survey of general practitioners. Results The management of osteoporosis has been characterised by a movement away from single to combination formularies. The consumption of calcium carbonate has declined, and the use of bisphosphonates has increased. While the gross consumption of risedronate increased steadily over the decade (8.4 to 186.5 kg), the consumption of alendronate declined after 2007. The consumption of vitamin D (over-the-counter and prescription) has increased from 0.58 to 2.8 kg over the decade. While
D. Rowell (*) : L. Gordon Centre for Applied Health Economics, Griffith Health Institute, Griffith University, Logan Campus, University Drive, Meadowbrook, Queensland 4131, Australia e-mail: [email protected] D. Rowell : L. Gordon Centre for Research Excellence in Sun and Health (CRESH), Queensland University of Technology, Queensland 4029, Australia
prescription vitamin D comprised just 10 % of the total, its consumption has undergone a 20-fold increase since 2006. Importantly, we can also report that the management of osteoporosis by Australia's general practitioners increased steadily until 2007–2008, before declining by 36 %. Conclusions Further research is required to determine possible casual relationships that may exist between these data. Until a formal economic evaluation of vitamin D testing in this clinical setting is completed, no conclusions regarding its economic effectiveness should be drawn. Keywords Osteoporosis . Treatment of osteoporosis . Vitamin D . Vitamin D testing
Introduction Low levels of 25-hyrodxyvitamin D (25OHD) can cause osteoporosis [1], and treatment with vitamin D is recommended to maintain calcium and phosphate homeostasis to optimise bone and muscle function [1]. Evidence from randomised controlled trials has demonstrated that vitamin D with calcium supplementation can reduce the risk of falls and fractures [1, 2]. Furthermore, treatment protocols recommend that testing for serum levels of 25OHD should be considered in patients with a Z-score less than or equal to −2.0 on dualenergy X-ray absorptiometry (DXA) or if a cause of secondary bone loss is clinically suspected [3]. Evidence derived from observational studies has also implicated hypovitaminosis D in the pathogenesis of an increasing diverse range of diseases, including breast and colon cancer, autoimmune diseases, such as multiple sclerosis, and depression [1]. However, concern has been raised regarding the rapid increase in the costs of 25OHD testing. Sattar et al. [4] state that Scotland has seen a doubling of 25OHD tests from 2008 to 2010 and a hospital in the UK reports a sixfold increase in tests from 2007 to 2010. Bilinski and Boyages [5] report that
161, Page 2 of 5
the cost of 25OHD testing in Australia has increased from Australian $1.02 million to Australian $96.75 million from 2000 to 2010. While there is a consensus that the increased number of 25OHD tests may not be justifiable, there is some divergence in the tone of conclusions. While Sattar et al. [4] “…urge all clinicians to stop and think critically before measuring 25OHD, particularly in conditions not related to bone disease”, Bilinski and Boyages [6] assert that “[t]hese findings [unstainable growth in vitamin D testing] have widespread consequences in terms of quality of care, unnecessary cost, and potential over diagnosis”. An economic evaluation, which considers both costs and benefits, is necessary to fully appraise the net benefit of any given intervention. Conclusions regarding the costeffectiveness of a treatment derived solely from a cost-ofillness study are implicitly reliant on the caveat, all other things being equal. Given that the most established indication for prescribing vitamin D remains the treatment of osteoporosis [1], the motivation for this paper is to explore what changes may have occurred in the management of this disease. The aim of this paper is to present clinical data reporting the salient trends in the management of osteoporosis alongside financial data, which reflects the exponential increase in the cost of 25OHD tests, during the last decade.
Materials and methods Osteoporosis is a chronic medical condition that is treated with pharmaceuticals prescribed by Australia's general practitioners (GPs). Our analysis therefore considers both inputs (pharmaceuticals) and outputs (management rates of osteoporosis). Firstly, we analyse pharmaceutical data obtained from Australia's Pharmaceutical Benefits Scheme (PBS), which summarises the consumption of all prescription medications provided through this insurance scheme.
Arch Osteoporos (2013) 8:161
Secondly, we analyse data obtained from a national survey of Australia's general practitioners conducted by Bettering the Evaluation and Care of Health (BEACH) [7–12]. This annual survey of 1,000 general practitioners reports clinical activity per 100 doctor–patient encounters. Longitudinal data for the years 1998 to 2011 are reported conjointly to highlight changes in the management of osteoporosis in Australia.
Results Figure 1 reports on the consumption of four pharmaceuticals commonly used to treat osteoporosis, the bisphosphonates (1) alendronate and (2) risedronate, (3) calcium carbonate and (4) colecalciferol (vitamin D3). The left-hand axis reports the number of scripts consumed for each available formulation, and the right-hand axis reports the cost of the subsidy for vitamin D testing, in Australian dollars as reported by Bilinski and Boyages [5]. The data presented in Fig. 1 illustrate that while the costs of all vitamin D testing increased 100-fold, there was a significant change in the management of osteoporosis with the addition of pharmaceuticals, which combined high-dose colecalciferol and bisphosphonates, after 2006. This change was associated with a more aggressive management of osteoporosis. The use of single-agent bisphosphonates and calcium carbonate declined. The maximum available strength of risedronate increased from 35 mg in 2002 to 150 mg in 2008. While the sales of over-the-counter (OTC) colecalciferol, available in doses of 5 to 25 μg, accelerated steadily over the decade, doses of up to 140 μg became available in combination therapies available on prescription toward the end of the decade. Aggregate annual doses of risedronate, alendronate, calcium carbonate and colecalciferol supplied through the PBS were estimated and reported in Fig. 2 as follows:
Aggregate dose ¼ ΣðDose per tablet Tablets per script Scripts per yearÞ:
Figure 2 reports both the absolute and relative changes for each pharmaceutical consumed, for the years 2001 to 2012. The vertical axis reports the change in pharmaceutical consumption as a percentage of the baseline. The legend lists each pharmaceutical with aggregate consumption at commencement and completion of the time series. For example, the total consumption of risedronate increased from 8.4 to 186.5 kg from 2001 to 2012, which was a 2,200 % increase. Also included is an estimate of the consumption of OTC vitamin D. For the years 2006 to 2011, annual market sales figures provided by Euromonitor International [13] were
converted to kilogrammes per year using a retail price of Australian $0.72 per microgram derived from a convenience sample obtained from a popular supermarket. For the years 2001 to 2006, a reported 13 % growth in sales of all OTC vitamins by Blackmores®,1 a market leader, was used as a proxy for changes in vitamin D consumption. The graph for OTC colecalciferol is “dashed” for the years 2001 to 2006 to indicate a trend estimate and “solid” for the years 2006 to 2011 to indicate reported data. 1 Blackmores http://www.blackmores.com.au/about-blackmores/ investors-centre/quarterly-results
Arch Osteoporos (2013) 8:161 2.5
120
80 1.5 60 1 40
0.5
Cost of Vitamin D Tests ($m)
100
2 PBS Services (millions)
Fig. 1 Changes in pharmaceutical use and cost of 25OHD testing (2001 to 2011). Risedronate Na (4443W, 4444X, 8481J, 8621R, 8972F, 9391G); Risedronate Na, CaCO3 and Colecalciferol (4380M, 8974H); Risedronate Na and CaCO3 (8899J, 8973G); Alendronate Na and Colecalciferol (9012H, 9183H); Alendronate Na, Colecalciferol and CaCO3 (39351E) for years 2001 to 2012 ([5, 17])
Page 3 of 5, 161
20
0
0 2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Risedronate Na (5mg)
Risedronate Na (35mg)
Risedronate Na (30mg)
Risedronate Na (150mg)
Risedronate Na & CaCO3 (35mg/1.25g)
Risedronate Na, CaCO3 & Colecalciferol (35mg/2.5g/22ug)
Alendronate Na (40mg)
Alendronate Na (70mg)
Alendronate Na & Colecalciferol (70mg/70ug)
Alendronate Na & Colecalciferol (70mg/140ug)
Alendronate Na, Colecalciferol & CaCO3 (70mg/140ug/1.25g)
CaCO3 (1.25g & 1.5g)
Vitamin D tests ($m)
if years is beyond 2006 and an interaction term Year * Post_2006. The regression output is reported in Table 1. The null hypothesis that the coefficients for Post_2006 and interaction term Year * Post_2006 were jointly equal to 0 was rejected (F (2,10)=23.00 (p
