The cost-effectiveness of implementing a new guideline for cardiovascular risk management in primary care in the Netherlands

Abstract

Background

A new Dutch guideline for cardiovascular disease management substantially extends the number of individuals for whom treatment with statins and/or antihypertensive agents is recommended. We estimated the cost-effectiveness of implementing the new guideline at the national level.

Methods

First, the number of currently untreated individuals who would become eligible for cholesterol-lowering or antihypertensive treatment under the new guideline was estimated using data from a recent population study. Cost-effectiveness of treating this group of patients was then assessed using a mathematical model.

Results

Implementing the guideline in the age category 30-69 years would lead to an additional 465 000 individuals requiring treatment. Over a period of 20 years, the cumulative incidence of acute myocardial infarction in the whole population would drop by 3.0%, that of stroke by 3.9%, and all-cause mortality would drop by 0.9%. The lifetime cost-effectiveness ratio was calculated to be 15 000 € per quality-adjusted life year gained. In the age categories 70-79 years and 80 years or above, an additional 600 000 and 450 000 persons, respectively, would need to be treated, resulting in corresponding reductions in cumulative incidences of 14 and 18% (acute myocardial infarction), 17 and 22% (stroke), and 1.2 and 0.6% (all-cause mortality) with cost-effectiveness ratios of 20 800 and 32 300 €, respectively, per quality-adjusted life year.

Conclusion

Complete implementation of the new guideline would lead to a considerable increase in the number of individuals requiring treatment. This would be cost-effective up to the age of 70 years.

Keywords

cardiovascular disease cardiovascular risk cost-benefit analysis guideline hypercholesterolemia hypertension primary prevention Systematic Coronary Risk Evaluation

Introduction

Like in most other affluent countries also in the Netherlands cardiovascular disease is the most common cause of mortality, accounting for about one-third of all deaths [1]. Even slight shifts in the prevalence of cardiovascular risk factors such as hypertension and hypercholesterolemia would therefore have a large effect on death rates. As extensive evidence from clinical trials has shown the efficacy of statins and antihypertensive agents in reducing the incidence of cardiovascular events [2–10], there has been a tendency to start preventive treatment earlier and at lower values of blood cholesterol or blood pressure [11, 12]. This is also apparent when one compares the most recent Dutch guideline [13] for cardiovascular risk management with the ones it is meant to replace [14, 15]. An important novelty in the new guideline is the estimation of risk based on a combination of risk factors. In addition, the new guideline does not specify an age limit beyond which statins should not be prescribed, whereas the previous guidelines advised against starting statins in males aged 70 years or above (females aged 75 years or above). Finally, the cut-off value for total cholesterol was lowered from 5 to 4.5 mmol/l (LDL 2.5 mmol/l).

Currently, more than one million individuals in the Netherlands (population: 16 million) use statins as a prophylaxis against atherosclerotic disease, and approximately twice or thrice that number use antihypertensive agents (The Drug Information System of the Health Care Insurance Board estimated the number of statin users in 2004 to be 1 023 000, www.gipdatabank.nl). The criteria of the new guideline for initiating treatment imply that a considerably larger proportion of the population would be eligible for treatment with statins and/or antihypertensive agents than is currently the case. In this study, we estimated the number of additional individuals in different age categories who would require treatment if the guideline is fully implemented. On the basis of that estimate, we assessed the consequences in terms of costs, health gains, and cost-effectiveness.

Methods

Risk assessment: new versus old guideline

In line with the European guideline [16], the new Dutch guideline recommends the assessment of 10-year cardiovascular risk using an adapted version of the Systematic Coronary Risk Evaluation (SCORE) risk function [13, 14, 17]. This formula takes sex, age, smoking behavior, cholesterol level, and blood pressure as input. The adaptation presented in the Dutch guideline consists of a calculation of incidence of cardiovascular disease (i.e. including morbidity and mortality), instead of taking cardiovascular mortality as in the original score. To convert cardiovascular mortality to cardiovascular disease incidence, the Dutch guideline used a mathematical model to determine risk factor-specific conversion factors. A further complication was that the SCORE formula was restricted by its authors to ages up to 65 years but not beyond. The Dutch guideline explicitly states that no age limit applies. Hence, to determine risk for those aged 65 years or above, we assumed the SCORE function remains valid for older age groups.

Treatment is recommended for all individuals who are at a 10-year risk of 10% or higher; within the range of 5-9% risk, treatment is recommended as an option in the presence of additional risk factors.

Determining the increase in the number of treated individuals

Several studies pertaining to risk factor levels among the Dutch population were identified [18–22]. From these, the numbers of individuals were selected who would qualify for treatment with antihypertensives and/or statins according to the new guideline, as well as those already being treated. The difference between these two was then taken as the increase in the number of individuals requiring treatment. As suitable data for the age categories of 70 years and above were not available, we used the same risk factor level distributions as for the age group 65-69 years, assuming that this distribution would remain constant from 65 years of age onwards.

The RIVM Chronic Disease Model

Estimates of health gains [quality-adjusted life years (QALYs)] and cost-effectiveness were derived using the Chronic Disease Model (CDM), a mathematical model that has been developed at our institute RIVM (National Institute of Public Health and the Environment). The CDM simulates the yearly changes in prevalences of risk factors and diseases in the population. The potential effects of health care interventions can be studied by following the shifts that occur in these prevalences over time and the associated changes in health outcomes and health care costs [23–25]. Risk factors and diseases are linked by using the concept of relative risk. The remaining parameters of the model are initial prevalences, 1-year incidence and mortality probabilities, and 1-year transition probabilities of changing risk factor class. The values for these parameters, specified by age and sex, have been derived from a variety of sources, including national and local disease registries, mortality statistics, and the international literature. Demographic figures describing the Dutch population (as at the end of 2004) were obtained from Statistics Netherlands.

Blood pressure and cholesterol are modeled as categorical variables subdivided into eight classes each: first according to level [blood pressure (mmHg)]: < 120, 120- < 140, 140- < 160, and ≥ 160; cholesterol (mmol/l): < 5.0, 5.0- < 6.5, 6.5- < 8.0, and ≥ 8.0, and next according to whether or not drug treatment has been initiated. Information regarding the current distribution of the Dutch population over these eight classes was derived from the sources mentioned above [14, 18, 23, 24, 26].

Endpoints

Endpoints assessed were: cumulative incidences over a period of 20 years of acute myocardial infarction (AMI), other coronary disease, stroke, and all-cause mortality. In addition, life expectancy in years and in QALYs was calculated (by following the cohort to ‘extinction’), as were total costs.

Determination of costs, quality-adjusted life years, and cost-effectiveness

Health care costs are calculated in the CDM [27] on the basis of average costs of illness stratified by age and sex as estimated by the so-called second Dutch Cost-of-Illness Study [28]. Intervention costs, used as input to the model, were taken to include costs of diagnostics (in the first year only) and doctor's visits, repeat prescriptions, and drug costs. Using unit costs as valid for the Netherlands, this led to estimates of € 293 per year (€ 331, in the first year) for the use of statins, and € 258 (€ 286) for treatment with antihypertensives. All cost calculations were carried out from a health care perspective and did not include societal costs because of productivity losses. Apart from intervention costs, the calculation of total costs also included subtraction of cost savings from reductions in causally related diseases and addition of costs for causally nonrelated diseases in life years gained.

To adjust life years for quality of life (to derive QALYs), disease-specific disability ‘weights' were used that were determined recently in the context of a Dutch implementation of the Global Burden of Disease Study [29]. Applying these weights (or actually 1 minus the weight) yields 1 QALY in case of a year lived in full health, or a fraction of a QALY for a year spent in ill health, the more severe the disease the lesser the fraction (zero representing death).

In calculating cost-effectiveness, costs were discounted at 4% and QALYs at 1.5% [30]. Following the current Dutch recommendations for health economic assessment, a ratio of 20 000 € per QALY was taken as the threshold for cost-effectiveness [31].

Scenarios

Endpoints were compared running two different scenarios: continuation of the current situation (reference scenario) versus implementation of the new guideline. It was assumed that everyone meeting the criteria for treatment would receive treatment for their remaining lifetime.

Sensitivity analyses

Sensitivity analyses were performed in which effects, discount rates, time horizon, and compliance were varied. More conservative estimates for health effects were assumed by taking the lower boundaries of the confidence intervals for the relative risks as reported in two large meta-analyses [32, 33]. Discount rates were varied between 0 and 4% for both costs and effects. Regarding compliance, we assumed that 50% of patients stopped taking medication after 2 years, thus incurring expenses without the health benefits.

Results

Table 1 shows the percentages of currently untreated individuals who would require treatment according to the new guideline. In most cases, treatment with both antihypertensives and statins turned out to be indicated. The number of additional persons requiring treatment is especially high in the age group 70 years and above. Altogether, of all people aged 30-79 years, more than one million people would require initiation of drug treatment. Table 2 provides a comparison of the numbers of first time myocardial infarctions, new cases of other coronary diseases, strokes, and numbers of deaths predicted for a period of 20 years under the two scenarios considered. In addition, shown are the respective health gains achieved by the intervention (reductions in AMIs, etc.).

Table 1

Currently untreated individuals requiring treatment according to the new guideline a


	Statins (%)		Antihypertensives (%)		Both (%)

Age (years)	Males	Females	Males	Females	Males	Females	Total b

30-39	< 0.1	< 0.1	< 0.1	0.2	< 0.1	< 0.1	2500
40-49	0.1	< 0.1	< 0.1	0.2	< 0.1	< 0.1	4000
50-59	0.6	< 0.1	0.6	0.4	4.3	0.1	68 500
60-69	10.0	0.4	0.2	1.1	34.3	5.3	391 000
Total 30-69							466 000
70-79 c	33.1	16.2	< 0.1	< 0.1	43.6	26.7	606 000
Total 30-79							1 072 000
80 + c	39.3	41.2	< 0.1	< 0.1	43.6	33.1	452 000
Total 30-80+							1 524 000

Calculated from population survey data by determining the proportions of individuals without cardiovascular disease with a 10-year cardiovascular mortality risk > 10%.

Extrapolated to the whole Dutch population as in the year 2006 (the Statistics Netherlands).

Estimated numbers, based on risk factor levels for the age group 65-69 years.

Table 3 shows QALYs gained, costs (intervention and total), cost-effectiveness ratios (CERs), and numbers of (healthy) life years gained.

The CER when treating all individuals who met the guideline criteria seemed to be reasonable (22 000 € per QALY). However, when taking a closer look into subgroups according to the age at intervention, the CER seems to be less favorable for the group of persons aged 80 years and above: 32 300 € per QALY (31 000 for males and 33 000 for females). For the age groups 30-69 years and 70-79 years, CERs were 15 000 € per QALY (14 700 for males and 15 900 for females) and 20 800 € per QALY (20 600 for males and 21 000 for females), respectively. We also show the CERs when not including all cost categories.

Sensitivity analyses

Results of the sensitivity analyses are presented in Table 4. As was to be expected a decrease in compliance or effect size yields fewer QALYs. However, for the age group 30-69 years, the intervention remains cost effective when compliance is as low as 50%. The shorter the time period considered the less cost effective the intervention.

Discussion

Implementing the recommendations of the new national guideline for cardiovascular risk management would imply the initiation of treatment with statins and/or antihypertensives in 1.5 million currently untreated individuals. Projected over a period of 20 years into the future, this would prevent a substantial proportion of the expected numbers of AMIs. Compared with the reference scenario, the cumulative incidence in the whole population would be reduced by 3.0% amongst 30-69-year-old individuals, by 14% in the age group 70-79 years, and 17.6% in those aged 80 years or above. For stroke, reductions would be comparable. This would translate into a reduction in total mortality of 0.9, 1.2, and 0.6%, respectively, compared with the situation in which current practice would continue unchanged. Intervention would be cost effective up to the age of 70 years. In all age groups, the CER was more favorable for men than for women.

Table 2

Prevented 20-year cumulative incidence of cardiovascular disease and mortality as a result of full guideline implementation


	30-69			70-79			80 +

Age group (years)	Total	Males	Females	Total	Males	Females	Total	Males	Females

Incidence
AMI
Reference scenario	650 000	455 000	195 000	121 000	67 600	53 400	37 800	16 000	21 800
Numbers prevented	19 500	17 700	1850	17 100	10 700	6400	6650	2750	3900
Relative reduction (%)	3.0	3.9	1.0	14.1	15.7	12.0	17.6	17.2	17.8
Other CHD
Reference scenario	870 000	480000	390000	138 600	52 400	86 200	41 300	10 150	31 150
Numbers prevented	21 200	17 900	3350	19 800	9200	10 600	7800	1950	5850
Relative reduction (%)	2.4	3.7	0.9	14.3	17.5	12.3	18.9	19.4	18.7
CVA
Reference scenario	725 000	398 000	327 000	214 000	90 800	123 200	82 200	25 300	56 900
Numbers prevented	27400	22 700	4700	37 000	18 200	18 800	17 800	5600	12 200
Relative reduction (%)	3.9	5.7	1.4	17.3	20.1	15.2	21.6	22.1	21.4
Mortality
Reference scenario	1 780 000	1 045 000	735 000	1 008 000	453 400	554 600	557 000	1 79 600	377400
Deaths prevented	16 200	13 800	2450	12 000	5000	7000	3350	700	2650
Relative reduction (%)	0.9	1.3	0.3	1.2	1.1	1.3	0.6	0.4	0.7

The reference scenario is the continuation of current practice; cardiovascular disease prevented represents the difference in numbers of incident cases between the reference scenario and the scenario in which all eligible individuals receive treatment. AMI, acute myocardial infarction; CHD, coronary heart disease; CVA, cerebrovascular accident.

Table 3

Lifetime costs and benefits of national implementation of the new guideline, because of the newly treated individuals


	30-80+	30-79	30-69			70-79			80+

Age group (years)	Total	Total	Total	Males	Females	Total	Males	Females	Total	Males	Females

Newly treated individuals (n)	1 523 000	1 071 000	465 000	402 000	63 000	606 000	354 000	251 000	452 000	156 000	296 000
Costs of intervention (× 10⁶ €)	17 840	13 390	6480	5590	790	2790	1670	1120	1150	423	730
Life years gained	720 000	575 000	266 000	226 000	40 000	265 000	133 000	132 000	105 000	33 000	72 000
Quality-adjusted life years gained	507 000	421 000	216 000	187 000	29 000	177 000	94 000	83 000	62 000	21 000	41 000
Costs per quality adjusted life year (€)	22 000	18 900	15 000	14 700	15 900	20 800	20500	21 000	32 300	31 000	33 000

Table 4

Results of the sensitivity analyses


		Age 30-69 years		Age 70-79 years		Age 80 + years

		Costs per QALY (€)	QALYs gained	Costs per QALY (€)	QALYs gained	Costs per QALY (€)	QALYs gained

Variable Reference	Values	15 000	216 000	20 800	177 000	32 300	62 000

Relative risks	Lowest bound	21 500	137 400	28 600	101 500	43 100	35 300
Discount rates a	0%, 0%	33 600	282 500	31 400	214 300	41 500	71 500
	4%, 4%	22 500	144 200	28 000	131 600	40 000	50400
Time horizon b	5	80600	10 400	59100	16 700	61 900	12 100
	10	34500	41 800	28 300	63 000	39 700	35 400
	20	19 400	134 000	22 500	148 700	−	−
Compliance	50%	17 500	106 000	20 700	93 400	44 800	31 900

QALY, quality-adjusted life year.

Discount rates for costs and effects.

In years.

Point of departure of our study was an estimation of the numbers of currently untreated individuals who would qualify for treatment according to the new guideline. This assessment was based on data from a Dutch population study [18]. Data collection for that survey took place in the period 1998-2002. Since then, the use of statins has become more common. On that account, we might have overestimated the number of people requiring treatment initiation. In contrast, it is likely that higher risk individuals, such as smokers, were under-represented in the population study. Obviously, any bias in this estimate would translate into corresponding deviations in calculated costs and benefits.

Cardiovascular risk calculation in the Dutch guideline is based on cardiovascular disease incidence, whereas the SCORE function took cardiovascular mortality as its endpoint. Transformation from one risk definition to the other has a consequence that for corresponding levels of risk factors the risk estimate is higher (maximum a doubling of risk), and hence the number of individuals requiring treatment is greater in the Dutch guideline compared with the SCORE function.

A further noteworthy aspect of the Dutch guideline pertains to the older age groups, whereas the SCORE function was limited to age groups up to 65 years; the Dutch guideline explicitly states that age is no longer a restriction. However, it entirely neglects the age restriction of the risk function. This is a shortcoming of the guideline that requires updating. Unfortunately, no data regarding risk factor levels were available for those aged 70 years or above. We therefore assumed these to remain constant from age 65 years onwards; but if they actually increase, the number of people eligible for treatment in that age group would be even greater, increasing both costs and health benefits.

In the previous guideline, treatment of the elderly and of patients with relatively low-risk profiles was not recommended on behalf of a CER that was deemed to be unfavorable. However, in the meantime the patents on the two agents that were singled out as cholesterol-lowering treatment of first choice, simvastatin and pravastatin, have run out of patent. Lowering of the prices of these drugs has shifted the balance between costs and effects. Moreover, further cost reductions could be achieved by cutting down on overtreatment, that is, by reducing the number of patients whose treatment is unjustified according to the guideline [34].

We found the relative reduction in cardiovascular disease as a result of guideline implementation to be more pronounced in the older age groups. In the elderly age groups, untreated persons qualifying for treatment initialization constitute a greater proportion of the population and the benefits to be achieved are correspondingly greater. However, as life expectancy is smaller, the CER, which is based on QALYs, is less favorable. The importance of age in determining cost-effectiveness was highlighted in a recent study among individuals aged 35-85 years with a history of occlusive arterial disease or diabetes [35]. However, that study did not take into account what we believe to be an important asset of our approach, namely the inclusion of costs of diseases that are not related to the intervention but that can occur in life years gained as a result of successful prevention (‘substitute diseases’).

When projecting future benefits and costs of treatment, we assumed that patient compliance would be optimal. However, a recent Dutch study [36] showed that 2 years after having been prescribed cholesterol-lowering agents, only 46.5% of patients were still filling their prescriptions. Similar findings have been described in a number of studies [37–40]. A lack of compliance would affect the CERs in an unfavorable manner, as patients discontinuing treatment incur costs while not, or only incompletely, benefiting from the treatment effects. Furthermore, the impact of side effects and drug-drug interactions, which we did not take into account, would probably tilt the CER in a slightly more unfavorable direction.

Conclusion

Large-scale implementation of the new guideline would imply a substantial increase in the number of individuals being treated with statins and/or antihypertensives. The effect of the intervention is most pronounced in the older age groups. However, the health gain per prevented case is largest in the younger age groups. Consequently, in terms of cost-effectiveness, the intervention is most favorable in the younger age groups. The calculations performed with our model show this to be a cost-effective intervention in the primary prevention of cardiovascular disease up to the age of 70 years.

Footnotes

Acknowledgements

The authors are grateful to Mr J. Hoekstra, MSc, Mr L. Tariq, Mrs S.M. Vijgen, Mrs P.D. Gumbs, MSc (National Institute of Health and the Environment, Bilthoven, The Netherlands) and Mrs J.C. Witteman, MSc, PhD (Erasmus University, Rotterdam, The Netherlands) for their contributions to this study. This research was sponsored by Grant V/260401/01/HK from the Ministry of Health, Welfare and Sport, The Netherlands. There are no conflicts of interest.

References

Vaartjes

van Dis

Peters

RJG

Bots

ML.

Cardiovascular disease in The Netherlands. In: Vaartjes

Peters

RJG

van Dis

Bots

, editors. Cardiovascular disease in The Netherlands [in Dutch]. Den Haag: Dutch Heart Foundation; 2006. pp. 9–21.

Ezzati

Lopez

Rodgers

Vander Hoorn

Murray

CJ.

Selected major risk factors and global and regional burden of disease. Lancet 2002; 360:1347–1360.

Lawes

Vander Hoorn

Law

Elliott

MacMahon

Rodgers

Blood pressure and the global burden of disease 2000. Part II: estimates of attributable burden. J Hypertens 2006; 24:423–430.

Briel

Studer

Glass

Bucher

HC.

Effects of statins on stroke prevention in patients with and without coronary heart disease: a metaanalysis of randomized controlled trials. Am J Med 2004; 117:596–606.

Law

Wald

Rudnicka

AR.

Quantifying effect of statins on low density lipoprotein cholesterol, ischaemic heart disease, and stroke: systematic review and meta-analysis. BM J 2003; 326:1423.

LaRosa

Vupputuri

Effect of statins on risk of coronary disease: a meta-analysis of randomized controlled trials. JAMA 1999; 282:2340–2346.

Cheung

Lauder

Lau

Kumana

CR.

Meta-analysis of large randomized controlled trials to evaluate the impact of statins on cardiovascular outcomes. Br J Clin Pharmacol 2004; 57:640–651.

Bucher

Griffith

Guyatt

GH.

Systematic review on the risk and benefit of different cholesterol-lowering interventions. Arterioscler Thromb Vasc Biol 1999; 19:187–195.

9 Medical Research Council Working Party. MRC trial of treatment of mild hypertension: principal results. Br Med J (Clin Res Ed) 1985; 291:97–104.

10.

10 SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 1991; 265:3255–3264.

11.

Van Dam

van Wissen

Kastelein

JJ.

Declaring war on undertreatment: rationale for an aggressive approach to lowering cholesterol. J Cardiovasc Risk 2002; 9:89–95.

12.

12 Antikainen

Moltchanov

Chukwuma

Kuulasmaa

Marques-Vidal

Sans

. Trends in the prevalence, awareness, treatment and control of hypertension: the WHO MONICA project. Eur J Cardiovasc Prev Rehabil 2006; 13:13–29.

13.

13 The Dutch Institute for Healthcare Improvement CBO. Dutch Guideline Cardiovascular Risk Management [in Dutch]. Utrecht: CBO; 2006.

14.

14 Treatment and prevention of coronary heart disease by lowering plasma cholesterol levels. Second revision of the Consensus Cholesterol. Behandeling en preventie van coronaire hartziekten door verlaging van de plasmacholesterolconcentratie. Consensus cholesterol tweede herziening. Utrecht: CBO; 1998.

15.

15 Revision Guideline Hypertension. Herziening Richtlijn Hoge bloeddruk. Utrecht/Den Haag: CBO/Nederlandse Hartstichting; 2000.

16.

16 De Backer

Ambrosioni

Borch-Johnsen

Brotons

Cifkova

Dallongeville

. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and Other Societies on Cardiovascular Disease Prevention in Clinical Practice. Eur Heart J 2003; 24:1601–1610.

17.

17 Conroy

Pyorala

Fitzgerald

Sans

Menotti

De Backer

. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J 2003; 24:987–1003.

18.

Blokstra

Smit

Verschuren

WMM.

Changes in lifestyle factors and risk factors for chronic diseases with ageing: the Doetinchem Study 1987–2002 [in Dutch]. Bilthoven: Dutch National Institute for Public Health and the Environment; 2006. Report No. 260401003.

19.

Smit

Verschuren

WMM

Bueno-de -Mesquita

Seidell

JC.

The Monitoring Project on Risk Factors for Chronic Diseases (MORGEN-project): general background [in Dutch]. Bilthoven: Dutch National Institute for Public Health and the Environment; 1994. Report No. 263200001.

20.

Hofman

Boerlage

Bots

den Breeijen

de Bruijn

Grobbee

DE.

Prevalence of chronic diseases among elderly people; the Rotterdam Elderly Study [in Dutch]. Ned Tijdschr Geneeskd 1995; 139:1975–1978.

21.

Hofman

Grobbee

de Jong

van den Ouweland

FA.

Determinants of disease and disability in the elderly: the Rotterdam Elderly Study. Eur J Epidemiol 1991; 7:403–422.

22.

van der Linden

Westert

de Bakker

Schellevis

FG.

The Second Dutch National Survey of General Practice. Symptoms and disorders in the population and in general practice [in Dutch]. Utrecht/Bilthoven: NIVEL/RIVM; 2004.

23.

Feenstra

Hamberg-van Reenen

Hoogenveen

Rutten-van Molken

MP.

Cost-effectiveness of face-to-face smoking cessation interventions: a dynamic modeling study. Value Health 2005; 8:178–190.

24.

24 Hoogendoorn

Rutten-van Molken

Hoogenveen

van Genugten

Buist

Wouters

. A dynamic population model of disease progression in COPD. Eur Respir J 2005; 26:223–233.

25.

Van Baal

Hoogenveen

de Wit

Boshuizen

HC.

Estimating health-adjusted life expectancy conditional on risk factors: results for smoking and obesity. Popul Health Metr 2006; 4:14.

26.

26 Scheltens

Verschuren

WMM

Boshuizen

Hoes

Zuithoff

Bots

. Estimation of cardiovascular risk: a comparison between the Framingham and the SCORE model in people under 60 years of age. Eur J Cardiovasc Prev Rehabil 2008; 15:562–566.

27.

Van Baal

PHM

Feenstra

Hoogenveen

de Wit

GA.

Cost effectiveness analysis with the RIVM Chronic Disease Model. Bilthoven: Dutch National Institute for Public Health and the Environment; 2005. Report No. 260706002.

28.

Polder

Takken

Meerding

Kommer

Stokx

LJ.

Costs of Illness Studyin The Netherlands [in Dutch]. Bilthoven: Dutch National Institute for Public Health and the Environment; 2002.

29.

Melse

Essink-Bot

Kramers

Hoeymans

A national burden of disease calculation: Dutch disability-adjusted life-years. Dutch Burden of Disease Group. Am J Public Health 2000; 90:1241–1247.

30.

30 Health Care Insurance Board. Guidelines for Farmaco-Economic Evaluations, Actualised Version [in Dutch]. Diemen; College voor zorgverzekeringen (Health Care Insurance Board): 2006.

31.

Casparie

van Hout

Simoons

ML.

Guidelines and costs [in Dutch]. Ned Tijdschr Geneeskd 1998; 142:2075–2077.

32.

32 Baigent

Keech

Kearney

Blackwell

Buck

Pollicino

. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90 056 participants in 14 randomised trials of statins. Lancet 2005; 366:1267–1278.

33.

33 Turnbull

Neal

Algert

Chalmers

Chapman

Cutler

. Effects of different blood pressure-lowering regimens on major cardiovascular events in individuals with and without diabetes mellitus: results of prospectively designed overviews of randomized trials. Arch Intern Med 2005; 165:1410–1419.

34.

34 Gumbs

Verschuren

Mantel-Teeuwisse

de Wit

Hofman

Trienekens

. Drug costs associated with non-adherence to cholesterol management guidelines for primary prevention of cardiovascular disease in an elderly population: the Rotterdam Study. Drugs Aging 2006; 23:733–741.

35.

35 Heart Protection Study Collaborative. Lifetime cost effectiveness of simvastatin in a range of risk groups and age groups derived from a randomised trial of 20 536 people. BM J 2006; 333:1145.

36.

Mantel-Teeuwisse

Goettsch

Klungel

de Boer

Herings

RM.

Long term persistence with statin treatment in daily medical practice. Heart 2004; 90:1065–1066.

37.

37 Perreault

Blais

Lamarre

Dragomir

Berbiche

Lalonde

. Persistence and determinants of statin therapy among middle-aged patients for primary and secondary prevention. Br J Clin Pharmacol 2005; 59:564–573.

38.

38 Avorn

Monette

Lacour

Bohn

Monane

Mogun

. Persistence of use of lipid-lowering medications: a cross-national study. JAMA 1998; 279:1458–1462.

39.

Benner

Glynn

Mogun

Neumann

Weinstein

Avorn

Long-term persistence in use of statin therapy in elderly patients. JAMA 2002; 288:455–461.

40.

Van Wijk

Klungel

Heerdink

de Boer

Rate and determinants of 10-year persistence with antihypertensive drugs. J Hypertens 2005; 23:2101–2107.