Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Persistence with statins


Lack of use of prescribed medicines has a number of labels (compliance, concordance, persistence, adherence). For many prescribed medicines, use of the medicines is known to be low, yet good studies are infrequent. Lack of use can be due to patients failing to start therapy prescribed for them. Patients may take medicines only intermittently, or they may discontinue for several reasons, including adverse events.

For statins, we know that persistence is low, and falls with time. An on-line Bandolier review highlighted this in 2004. A new Canadian study [1] provides more information about persistence in middle-aged people, especially in primary and secondary prevention, and about different statins.


The study was in Quebec, where databases have information on patient demographics, medical services, and prescriptions. A cohort of patients with a first prescription for a statin in the years 1998-2000 was selected. They had to have no prescription for a lipid lowering drug in the previous year, and be between 50 and 64 years old. The cohort was divided into use of statin for primary prevention in subjects with no indication of cardiovascular disease by diagnosis or therapy, and for secondary prevention in those with diagnosed coronary artery disease, again by diagnosis or therapy.

The drug database was used to identify statin agents dispensed during the follow up period, which was to mid 2001. The outcome of statin persistence was having any statin prescription dispensed within 60 days of the end of a previous prescription.


There were 13,642 patients in the primary cohort and 4,316 in the secondary cohort. In both cases the mean age was 58 years, but there were expected differences between the two cohorts. The secondary cohort, for instance, had more men, and more patients with hospital admission in the year before first prescription. The most frequent initial prescription was for pravastatin (42%) or simvastatin (36%), with lower rates for atorvastatin (13%), fluvastatin (6%) and lovastatin (3%).

Persistence was reduced to 90% within the first month, mostly because only one prescription was filled. At six months and at three years, persistence was low, and below half by three years (Figure 1), though somewhat higher in the secondary cohort. At six months, significantly higher persistence was found for simvastatin (mean dose 17 mg; Figure 2) than atorvastatin (16 mg), fluvastatin (27 mg) or lovastatin (21 mg), though with limited numbers in some.

Figure 1: Persistence at 6 months and 3 years for primary and secondary cohorts

Figure 2: Six-month persistence with different statins

The likelihood of stopping statins was higher in people taking more tablets, and who used more pharmacies and had more than three prescribing physicians, but was lower in those with hospital admission in the year before statin was prescribed, who had diabetes, hypertension, or respiratory disease.


Bandolier 113 looked at evidence that stopping statins was possibly a bad thing, with higher rates of death or heart attack than in those continuing statin and even, possibly, than in those who never had a statin. Stopping statins is not good, and not just because potential benefits are removed.

This very large database study of middle-aged Canadians confirms what we know from other studies in older people, that statin persistence over several years is low. A review [2] also looks at the pharmacoeconomic consequences, but concludes that we just don't know enough to measure the impact of low persistence.

It is clear, though, that potential benefits on a population level cannot be delivered when people do not take medicines prescribed for them. Healthcare is not a dictatorship, so we need to learn not just about how effective medicines prove to be in clinical trials (where persistence is often very high compared with clinical practice), but what makes people stop taking medicines, and how to encourage them to persist. It's not all what we do, perhaps more the way that we do it. A useful thought for policy makers.


  1. S Perreault et al. Persistence and determinants of statin therapy among middle-aged patients for primary and secondary prevention. British Journal of Clinical Pharmacology 2005 59: 564-573.
  2. AM Peterson, WF McGhan. Pharmacoeconomic impact of non-compliance with statins. Pharmacoeconomics 2005 23: 13-25.

previous or next story