European Heart Journal Advance Access originally published online on December 7, 2006
European Heart Journal 2007 28(2):154-159; doi:10.1093/eurheartj/ehl391
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Adherence to evidence-based statin guidelines reduces the risk of hospitalizations for acute myocardial infarction by 40%: a cohort study
1 PHARMO Institute, PO Box 85222, 3508 AE Utrecht, The Netherlands
2 Utrecht Institute of Pharmaceutical Sciences, Utrecht, The Netherlands
3 Academic Hospital Amsterdam, Amsterdam, The Netherlands
Received 10 February 2006; revised 1 September 2006; accepted 3 November 2006; online publish-ahead-of-print 7 December 2006.
* Corresponding author. Tel: +31 30 2345162; fax: +31 30 2345568. E-mail address: fernie.penning{at}pharmo.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
Aims To investigate the ‘real world’ effectiveness of robust statin therapy, focusing on the effect of dose and early treatment discontinuation on the risk of hospitalization for acute myocardial infarction (AMI).
Methods and results In the PHARMO database, including among others drug-dispensing and hospital discharge records for more than two million subjects in the Netherlands, 59 094 new users of statins in the period 1 January 1991 until 31 December 2004, 
18 years of age were identified. In these patients, exposure to statins, both in terms of persistence and dose, was determined over the first two treatment years. To determine the risk for AMI, patients were followed from this 2-year time point until the first hospital admission for AMI, death, or end of the study period. A total of 31 557 patients (53%) discontinued statin use within 2 years; 20 883 patients (35%) were persistent users with an average equipotent dose 4. A 30% reduction in risk of hospitalization for AMI with persistent statin use was observed. The protective effect increased with a higher dose (20 and 40% risk reduction with an equipotent dose 
3 and 4, respectively).
Conclusion These results show that statins are suboptimally used in real life for having the maximum benefit in terms of preventing AMI.
Key Words: Statins • Persistence • Dose • Acute myocardial infarction • Daily practice • Effectiveness
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
The reduction of serum cholesterol levels is associated with a considerably lower risk of coronary heart disease (CHD), both among high-risk patients and individuals without cardiovascular risk factors.1–4 The benefits of reducing cholesterol levels are obtained early, the major part after 2 years of therapy and the full benefit only after 5 years.5 In the range of cholesterol levels that are encountered in western countries, there exists no threshold value below which no further benefit regarding CHD risk reduction might be expected.5,6 These findings support a robust reduction of cholesterol, which has become feasible during recent years by the introduction of rosuvastatin, the non-statin lipid-lowering drug ezetimibe, as well as prescribing higher doses of statins.
In daily practice, however, statins appear to be less effective in reducing cholesterol levels when compared with the results of clinical trials. This may be due to lower patient compliance, high baseline cholesterol levels, prescription of sub-optimal doses as well as factors driven by reimbursement.7,8 The number needed to treat is therefore unnecessarily high and subsequently the cost-effectiveness unnecessarily low. The relationship between the dose and the duration of statin use and the magnitude of prevention of CHD has been described in both randomized clinical trials and observational studies.9–13 In these studies, it has been consistently demonstrated that without proper dosing and lipid control, prevention of cardiovascular outcomes will be sub-optimal. Furthermore, in the WOSCOPS trial, it was demonstrated that CHD risk reductions were largest among compliant patients.14
In the present study, we investigated the ‘real world’ effectiveness of robust statin therapy for primary and secondary prevention. The study focused on the effect of dose, and early treatment discontinuation, on the hospitalization rates for acute myocardial infarction (AMI).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
Setting
Data were obtained from the PHARMO Record Linkage System, which includes several databases, among which linked drug dispensing records and hospital records from more than two million individuals in defined areas in the Netherlands. The drug dispensing histories contain data on the dispensed drug, the type of prescriber, the dispensing date, the amount dispensed, the prescribed dose regimens, and the legend duration of use (prescription length). All drugs are coded according to the Anatomical Therapeutic Chemical Classification. The hospital records include detailed information concerning the primary and secondary diagnoses, procedures, and dates of hospital admission and discharge. All diagnoses are coded according to the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM).
Patient selection
An historical follow-up study was performed including new users of statins (rosuvastatin, atorvastatin, simvastatin, pravasatin, cerivastatin, and fluvastatin), aged 18 and older from 1 January 1991 until 31 December 2004. The source population included all new users of lipid-lowering drugs in the period 1991–2004. A patient was defined as a new user of a lipid-lowering drug, if none of these drugs had been dispensed for at least 1 year before the first recorded dispensing in the PHARMO database (i.e. the index date). All patients fulfilling the following criteria were included in the study cohort: a statin prescription at the index date, a registration in the PHARMO database of at least 1 year before and 2 years after the index date, no missing data on the dose and type of all statin prescriptions during the 2 years after the index date. Starting from the fact that lipid-lowering drugs should be used for at least 2 years to benefit from treatment,5 exposure to statins was determined over the first two treatment years. To determine the risk of hospitalization for AMI, patients were followed from this 2-year time point until the first outcome of interest, death, or end of the study period (31 December 2004), whichever was first. Patients were classified into two groups according to their risk of cardiovascular disease at the start of outcome follow-up (i.e. the 2-year time point). The high-risk group included patients with a hospitalization in the preceding year due to ischaemic heart disease (ICD-9-CM: 410–414), a cerebrovascular accident (ICD-9-CM: 430–438), peripheral vascular disease (ICD-9-CM: 443.9), or a percutaneous vascular intervention (procedure codes: 8-836, 8-837, 8-839), and patients with diabetes (i.e. using antidiabetic drugs or hospitalized for diabetes (ICD-9-CM: 250, 251, 3572, 362.0, 366.41, 648.0, 962.3) in the year prior to the start of outcome follow-up. The low or intermediate-risk group included all other patients.
Exposure
Exposure to statins was determined over the first 2 years of treatment, both regarding persistence and dose. For all prescriptions during this period, the type and dose of statin were combined into a single equipotency score to control for the fact that different types and doses of statins differ with respect to the percentage reduction in total cholesterol (Appendix 1). An increase of the equipotency score of 1 indicated a doubling of the effective dose. The first generation statins pravastatin, fluvastatin, and simvastatin have a maximum equipotency of 6, 5, and 6, respectively. The second generation statins atorvastatin and rosuvastatin have a maximum equipotency of 7. Treatment episodes of continuous use of a certain equipotency were constructed based on the method of Catalan.15 A treatment episode was measured as the time span between the starting date of the first dispensing until the expiry date of the final dispensing. The latter included the permissible gap following the final dispensing within the specific episode. In case of interruptions between two prescriptions, the episode was considered uninterrupted if the duration of this gap was less than half the period of the given dispensing or 7 days, whichever was greater. In case a different equipotent dosage was started before the end of the former treatment episode, this former episode was ended at the starting date of the next episode. Cessation of statin use, either temporary or permanent, was defined by an episode with an equipotent dosage of zero.
Persistence of statin use was defined as the number of days of continuous use of any statin. Patients were categorized into 2-year persistent statin users, 18-months persistent statin users, and non-persistent statin users (i.e. < 18 months of continuous statin use). In 2-year persistent statin users, the weighed average equipotent dosage during the first 2 years of treatment was determined and categorized into ‘low’ (3), ‘intermediate’,4 and ‘high’ (5).
Outcome
The outcome of interest was defined as a hospital admission for AMI (ICD-9-CM: 410), diagnosed by the combination of medical history, ECG findings, and blood tests for cardiac enzymes.
Statistical analysis
The effect of persistence of statin use and average equipotent dose during the first 2 years of treatment on risk of hospitalization for AMI thereafter was estimated in a Cox regression model, stratified by cardiovascular risk category at the start of the outcome follow-up (SAS V8.2, PROC PHREG). Assuming that a high dosage of statins only exerts its effect in case of persistent drug use, the analysis on the effect of dosage was restricted to patients who had continuously used statins during the first 2 years of treatment. Adjustments were made for age, gender, and use of antihypertensives, nitrates, antiarrhythmics, digoxin, platelet inhibitors, and anticoagulants in the year preceding the start of the outcome follow-up. The proportional hazards assumption was checked by extending the multivariate model with a time-dependent exposure covariate (e.g. persistence * log time). Statistical significance was defined as a two-sided P-value < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
The source population included 160 147 new users of lipid-lowering drugs in the period 1991–2004. Of these patients, 59 094 met the inclusion criteria of this study and represent all new statin users, 18 years of age and older, with sufficient duration of follow-up and dose information. These patients filled 548 084 statin prescriptions during the first 2 years of treatment. Most of the patients started treatment with simvastatin (27 791, 47%), followed by pravastatin (13 656, 23%), atorvastatin (12 769, 22%), fluvastatin (3894, 6%), and cerivastatin (984, 2%). The characteristics of the study patients are shown in Tables 1 (total) and 2 (stratified by persistence and average dose of statin use during the first 2 years of treatment). In total, 12 762 patients (22%) had a high cardiovascular risk at the start of the outcome follow-up. The main cardiovascular co-medications were antihypertensives and platelet inhibitors, used by 36 606 (62%) and 26 284 (45%) patients, respectively. These co-medications were used more frequently by patients using statins persistently. Among high-risk patients, 6610 patients (52%) discontinued statin use within 2 years and 5018 patients (39%) were persistent users with an intermediate or high dose (average equipotent dose
4). The same stop and dose treatment patterns were observed among low-risk patients. With respect to dosing, differences were observed over calendar time. The percentage of 2-year persistent statin users with an intermediate or high dose (equipotency 4) almost doubled after 1998 from 22% (4600/20 958) to 43% (16 283/38 136).
|
|
The effect of persistent statin use and average equipotent dose during the first 2 years of treatment on risk of AMI is shown in Table 3. In the primary prevention group, the incidence rate of hospital admission for AMI decreased from 0.52 per 100 patient-years among non-persistent statin users to 0.42 per 100 patient-years among 2-year persistent patients (adjusted RR 0.70; 95% CI 0.60–0.81). A 30% reduction in the risk of AMI with 2-year persistent statin use was also observed in the secondary prevention group. The protective effect of statin use was larger at a higher equipotent dose. In the primary prevention group, the reduction in risk of AMI increased from 20% with an average equipotent dose
3 to 40% with an average equipotent dose 4. A 40% reduction in the risk of AMI with an intermediate or high dose was also observed in the secondary prevention group. A low dose did not, however, significantly reduce AMI risk.
|
Extrapolating these findings to the Netherlands, we roughly estimate that by persistent, high-dose treatment, 300–400 AMIs can be prevented among statin users each year. This includes 250–300 events among previously non-persistent patients, 10–20 events among previously 18-month persistent patients, and 50–70 events among previously persistent patients treated with a low dose.
In Appendix 2, the type, dose, and equipotency of the statin prescribed last, before the start of outcome follow-up in 2-year persistent users is given, stratified by the average equipotent dose during the first 2 years of treatment. Of the 6654 patients on a low equipotent dose, 6582 patients (99%) were prescribed a first generation statin last, compared with 13 245 out of 20 883 patients (63%) on a high equipotent dose. The latter group of patients was prescribed higher doses of the first generation statins, e.g. simvastatin 20 mg compared with 10 mg in patients on a low equipotent dose.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
The results of this study show that in daily practice, statins, on average, are sub-optimally dosed and used over too short a time interval to have maximum benefit regarding preventing AMI. Our findings corroborate findings from clinical trials insofar that the longer patients are treated and the higher the dose, the more successful the prevention of AMIs.10–13 We estimate that by persistent, high-dose treatment, 300–400 AMIs can be prevented each year among statin users in the Netherlands.
The persistence of statin use in our study was lower than in trials. This has also been observed in other statin studies and for other lipid-lowering drugs.16–20 The lower persistence of statin use in daily practice has significant clinical and economic implications; the population effectiveness of statin treatment based on trials is too optimistic given the poor persistence in daily practice. In primary prevention, the number of patients needed to be treated (NNT) with a statin for 5 years to prevent one case of fatal or non-fatal myocardial infarction increases from 5721 to 83, if persistence in daily practice is taken into account. Similarly, in secondary prevention, the NNT increases from 2022 to 30. The actual increase in NNT will be even larger as the estimates are based on 2-year persistence rates instead of 5-year persistence rates.
There are a number of limitations to this study. First, information was missing on some of the other risk factors for AMI. The 30% reduction in AMI risk with 2-year persistent statin use, compared with non-persistent statin use is likely to be overestimated. Part of the risk reduction in persistent patients will be attributable to the management of other risk factors; persistent patients, for example, will be more prone to quit smoking or lose weight. On the other hand, the difference in risk reduction attributable to dose of statins might be larger than observed assuming that highly-dosed patients will have, for example, higher cholesterol levels and consequently a higher risk of AMI. Secondly, the effect of persistence and dose of statin use on risk of AMI was adjusted for use of anti-hypertensives, nitrates, antiarrhythmics, digoxin, platelet inhibitors, and anticoagulants at baseline, i.e. in the year before start of outcome follow-up. Such approach is common to cohort studies. However, patients who had stopped statin therapy at the start of outcome follow-up also may have discontinued the use of these other drugs. Indeed, a post hoc analysis showed that 16% (for platelet inhibitors) to 45% (for nitrates) of non-persistent statin users using comedications at baseline did not use these drugs anymore in the 6-month period after start of outcome follow-up. In persistent statin users, these percentages ranged from 7–41%. However, 6983 patients (12%) had less than 6 months of outcome follow-up. Adjusting the RRs in Table 3 for comedication use after instead of before the start of outcome follow-up did not change the results. Thirdly, patients with an AMI who die before hospitalization were not included, as information on mortality was not available. In the Netherlands, this concerns about 20% of all patients with an AMI. The association between statin use and AMI, however, is likely to be similar in these patients. Fourthly, the lack of biological results. It is not known whether the patients with the largest risk reduction, i.e. 2-year persistent statin users with an intermediate or high equipotent dose, also had the largest LDL cholesterol reduction. From 5580 new statin users, 18 years of age and older, with at least 6 months of follow-up, LDL cholesterol data were available. Unpublished results from these patients show that the higher the equipotent dose at the start of treatment, the higher the percentage of patients attaining goal during the first 6 months of therapy. In low-risk patients, goal attainment was increased five times with an equipotent dose 5 compared with an equipotent dose 3 (OR 4.9; 95% CI 3.8–6.3). In high-risk patients, the positive effect of an equipotent dose 5 on LDL goal attainment increased from four times at a low baseline LDL cholesterol (OR 3.7; 95% CI 1.0–13.4) to 12 times at a high baseline LDL cholesterol (OR 12.3;95% CI 7.9–19.4).
A major problem we anticipated was the interaction between the dose and type of statin. We therefore used the equipotency score. As shown in Appendix 1, the first generation statins, pravastatin, fluvastatin, and simvastatin, are dosed at a lower equipotency than atorvastatin and rosuvastatin. The largest reduction in hospitalization rates for AMI was observed among 2-year persistent patients treated with an equipotent dose 4. In line with the ‘dose’ assumption, these patients were treated relatively more often with atorvastatin and rosuvastatin. Furthermore, these patients were prescribed higher doses of the first generation statins. Increasing the dose of older statins is, however, limited by the maximum safe dose. Besides, persistence of statin use is expected to be lower, and consequently the investment loss will be higher, with an increased frequency of drug use. A high equipotent statin dose should therefore be realized preferably by using new, highly potent statins. However, because of the considerable economic impact of these new statins on pharmaceutical budgets, the opposite trend is being encouraged in the Netherlands and in Germany. In these countries, reimbursement measures promote the use of relatively inexpensive generic older statins. Such measures may be embraced if, together with the promotion of older statins, guidelines for higher dosing of these statins are also implemented, taking into account the aforementioned limitations. We observed in this study that restricting the use of statins to generic, inexpensive, standard low doses will make the problems worse.
In conclusion, the results of this observational study support robust cholesterol lowering, as recommended on the basis of trials. The reduction in risk of AMI was highest in 2-year persistent statin users with an intermediate or high equipotent dose, both in primary and secondary prevention. To improve the population effectiveness of statin treatment, persistent drug use and the use of new, potent statins should be encouraged. Today in the Netherlands, about 300–400 statin users each year experience an unnecessary AMI because of sub-optimal dosing and early discontinuation of statins.
|
Appendices |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
|
|
|
Acknowledgement |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
This study was financially supported by an unrestricted grant from Nefarma, the Netherlands.
Conflict of interest: none declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendicesAcknowledgementReferences |
|---|
- Scandinavian Simvastatin Survival Study Group. (1994) Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 344:1383–1389.[CrossRef][ISI][Medline]
- Sever PS, Dahlof B, Poulter NR, Wedel H, Beevers G, Caulfield M, Collins R, Kjeldsen SE, Kristinsson A, McInnes GT, Mehlsen J, Nieminen M, O'Brien E, Ostergren J. (2003) Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial–Lipid Lowering Arm (ASCOT-LLA): a multicentre randomised controlled trial. Lancet 361:1149–1158.[CrossRef][ISI][Medline]
- Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, McKillop JH, Packard CJ. (1995) Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 333:1301–1307.
[Abstract/Free Full Text] - Anonymous. (2002) MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 360:7–22.[CrossRef][ISI][Medline]
- Law MR, Wald NJ, Thompson SG. (1994) By how much and how quickly does reduction in serum cholesterol concentration lower risk of ischaemic heart disease? BMJ 308:367–372.
[Abstract/Free Full Text] - Grundy SM, Cleeman JI, Merz CN, Brewer HB Jr, Clark LT, Hunninghake DB, Pasternak RC, Smith SC Jr, Stone NJ. (2004) Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 110:227–239.
- Frolkis JP, Pearce GL, Nambi V, Minor S, Sprecher DL. (2002) Statins do not meet expectations for lowering low-density lipoprotein cholesterol levels when used in clinical practice. Am J Med 113:625–629.[CrossRef][ISI][Medline]
- Goettsch WG, Yin DD, Alemao E, Klungel OH, Stalenhoef AF, Herings RM. (2004) Statins are less effective in common daily practice among patients with hypercholesterolemia: the REALITY-PHARMO study. Curr Med Res Opin 20:1025–1033.[CrossRef][ISI][Medline]
- Wei L, Wang J, Thompson P, Wong S, Struthers AD, MacDonald TM. (2002) Adherence to statin treatment and readmission of patients after myocardial infarction: a six year follow up study. Heart 88:229–233.
[Abstract/Free Full Text] - Pedersen TR, Faergeman O, Kastelein JJ, Olsson AG, Tikkanen MJ, Holme I, Larsen ML, Bendiksen FS, Lindahl C, Szarek M, Tsai J. (2005) High-Dose Atorvastatin vs. Usual-Dose Simvastatin for Secondary Prevention After Myocardial Infarction: The IDEAL Study: A Randomized Controlled Trial. JAMA 294:2437–2445.
[Abstract/Free Full Text] - LaRosa JC, Grundy SM, Waters DD, Shear C, Barter P, Fruchart JC, Gotto AM, Greten H, Kastelein JJ, Shepherd J, Wenger NK. (2005) Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med 352:1425–1435.
[Abstract/Free Full Text] - Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R, Joyal SV, Hill KA, Pfeffer MA, Skene AM. (2004) Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 350:1495–1504.
[Abstract/Free Full Text] - Baigent C, Keech A, Kearney PM, Blackwell L, Buck G, Pollicino C, Kirby A, Sourjina T, Peto R, Collins R, Simes R. (2005) Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 366:1267–1278.[CrossRef][ISI][Medline]
- The West of Scotland Coronary Prevention Study Group. (1997) Compliance and adverse event withdrawal: their impact on the West of Scotland Coronary Prevention Study. Eur Heart J 18:1718–1724.
[Abstract/Free Full Text] - Catalan VS and LeLorier J. (2000) Predictors of long-term persistence on statins in a subsidized clinical population. Value Health 3:417–426.[CrossRef][Medline]
- Andrade SE, Walker AM, Gottlieb LK, Hollenberg NK, Testa MA, Saperia GM, Platt R. (1995) Discontinuation of antihyperlipidemic drugs–do rates reported in clinical trials reflect rates in primary care settings? N Engl J Med 332:1125–1131.
[Abstract/Free Full Text] - Benner JS, Glynn RJ, Mogun H, Neumann PJ, Weinstein MC, Avorn J. (2002) Long-term persistence in use of statin therapy in elderly patients. JAMA 288:455–461.
[Abstract/Free Full Text] - Mantel-Teeuwisse AK, Goettsch WG, Klungel OH, de Boer A, Herings RM. (2004) Long term persistence with statin treatment in daily medical practice. Heart 90:1065–1066.
[Free Full Text] - Eagle KA, Kline-Rogers E, Goodman SG, Gurfinkel EP, Avezum A, Flather MD, Granger CB, Erickson S, White K, Steg PG. (2004) Adherence to evidence-based therapies after discharge for acute coronary syndromes: an ongoing prospective, observational study. Am J Med 117:73–81.[CrossRef][ISI][Medline]
- Jackevicius CA, Mamdani M, Tu JV. (2002) Adherence with statin therapy in elderly patients with and without acute coronary syndromes. JAMA 288:462–467.
[Abstract/Free Full Text] - Pignone M, Phillips C, Mulrow C. (2000) Use of lipid lowering drugs for primary prevention of coronary heart disease: meta-analysis of randomised trials. BMJ 321:983–986.
[Abstract/Free Full Text] - LaRosa JC, He J, Vupputuri S. (1999) Effect of statins on risk of coronary disease: a meta-analysis of randomized controlled trials. JAMA 282:2340–2346.
[Abstract/Free Full Text] - Illingworth DR and Tobert JA. (1994) A review of clinical trials comparing HMG-CoA reductase inhibitors. Clin Ther 16:366–385.[ISI][Medline]
- Maron DJ, Fazio S, Linton MF. (2000) Current perspectives on statins. Circulation 101:207–213.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||