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

Switching statins

Cholesterol levels in trials
Studies of switching
US study [2]
UK study [3]
A note on costs

When Bandolier departs its little ivory tower (fake, of course) and tastes the real world, questions are never about esoteric issues of evidence. Rather, they are worldly issues of real import to everyday practice. Most often asked is whether switching from atorvastatin to simvastatin is really a good thing. A quick search for the evidence seemed useful.

Cholesterol levels in trials

When you look at cholesterol and cholesterol subfractions in trials of statins lasting 12 weeks or longer (91 trials and 43,000 patients [1]), two important findings stand out.

The first is that atorvastatin 10 mg and simvastatin 40 mg are virtually identical in terms of their average effect in lowering total cholesterol and LDL cholesterol (Figures 1 and 2). Both were some way better than lovastatin, pravastatin and fluvastatin. Both were less good than rosuvastatin 5 or 10 mg. However, there were 10 times more patients analysed for simvastatin than atorvastatin, and 20 times more than for rosuvastatin. The weight of evidence is with simvastatin.

Figure 1: Percent reduction in total cholesterol from baseline in studies of at least 12 weeks. Numbers of patients in brackets

Figure 2: Percent reduction in LDL cholesterol from baseline in studies of at least 12 weeks

The second point made was that these longer (>12 week) trials showed little in the way of a significant dose-response. This is different from trials of up to six weeks, when a highly significant dose response was seen for all statins. This observation may be important. Some clinicians insist that higher concordance rates are obtained and target cholesterol levels attained when statins are prescribed at lower doses or with careful upward titration, but there aren't trials to show it.

Studies of switching

Searching for studies of switching statins found only two. One was from a large institute in the USA, the other in primary care in the UK. Both showed it to be feasible and cost-effective.

US study [2]

Conducted at the Walter Reed Army Medical Centre in Washington, the study planned to change all the patients on statins to cerivastatin - at that time both licensed and the cheapest. It was done with a full consultation and an educational programme, and with a purposely-created clinic to arrange the change. The clinic was staffed by a full time pharmacist working to an agreed algorithm, an analyst, and a research administrator, all with clinical support.

In four months it saw 1,356 patients, 942 of whom agreed to be part of the efficacy and monitoring study. Their mean age was 68 years, and 60% were men.

Statins used changed from predominantly atorvastatin and pravastatin (86%) to predominantly cerivastatin (96%). There were small average improvements in LDL and HDL cholesterol, with the proportion at or below their target cholesterol rising from 65% to 75%. In the 942 patients who were monitored, 28 (3%) experienced minor adverse events, mainly of muscle origin.

Despite administration costs of almost $200,000, the conversion policy saved $115 per patient in the first year.

UK study [3]

Switching of statins in a UK primary care setting was monitored in a practice of 9,050 patients with an annual drug budget of £1.3 million. The aim was to switch patients being prescribed 10 mg or 20 mg atorvastatin to simvastatin. All those taking 20 mg atorvastatin were switched to 40 mg simvastatin.

Initially, 122 patients were identified. Of these, 43 were excluded by a pharmacist (cholesterol greater than 5 mmol/L on 10 mg, or greater than 4.6 mg on 20 mg atorvastatin; previous use of simvastatin failing to reduce to target, renal failure, transplantation, on warfarin or amiodarone). A further nine were excluded by their GP on medical, administrative, or social grounds.

The seventy remaining patients were contacted by letter; none refused. All were switched; one experienced an adverse event so was switched back.

The average total and HDL cholesterol values were the same before and 3-4 months after the switch. Before the switch three patients had total cholesterol above 5 mmol/L; after the switch six had total cholesterol above 5 mmol/L.

The total cost of switching was £2,000, with full year savings of £14,700, or just over 1% of the practice drug budget. The cost per patient switched was £213 per year.


There is one important difference between these studies, otherwise so similar in their success and cost savings. It is how the changes were made. At Walter Reed the initiative came from staff themselves, from a team of interested clinical specialists formed to plan and build support for the project. It was inclusive, and took account of some of the more complicated clinical areas. In the UK the practice was asked (euphemism for told, very often) by its Primary Care Trust to make the change.

The top-down method can be effective, but is not always so, so an interesting and important caveat is instructive. A corporate decision was made by a cardiology department in a UK hospital to give high dose statin (atorvastatin 40 mg or 80 mg) to patients after myocardial infarction or revascularisation. This was an evidence-based decision, and increasingly we know that getting LDL cholesterol below a certain target is beneficial, with substantial atheroma regression for lower LDL cholesterol and HDL cholesterol raised by more than 7.5% [4].

The administrative bosses (a group of Trusts) over-ruled the policy, limiting prescribing to simvastatin 20-40 mg. An audit over the same calendar period under successive policies provided some insight into the advisability of this [5]. As Figure 3 shows, the simvastatin policy resulted in more deaths, and cardiac and non-cardiac readmissions.

Figure 3: Audit of high dose statin (40, 80 mg atorvastatin, n=100) and low dose statin (20, 40 mg simvastatin, n=121) after infarction

A note on costs

There is much discussion about cost savings, some of it quite sensible [6,7]. When properly done in primary care in the UK, the implication is that if a similar proportion of patients were switched successfully it would save £630 million to the NHS over the five-year period until atorvastatin comes off patent. This is worth saving.

But it is interesting that the rather limited evidence to support the switch came after the orders, and not before. In primary care, all seemed to work out well over the short term. The caveat is that we have no outcome data over the long term, and now a wake up call that all may not be as simple as it looked. The buyer might be saving on one budget, and spending on the other.

There is also a question here about responsibility, which no one is answering. If, as in the example above, an administrative order (request) results in patient harm, and the patient sues, who carries the can and is up before the beak?


  1. JE Edwards, RA Moore. Statins in hypercholesterolaemia: A dose-specific meta-analysis of lipid changes in randomised, double blind trials. BMC Family Practice 2003, 4:18.
  2. KA Grace et al. Implementation of a therapeutic-interchange clinic for HMG-CoA reductase inhibitors. American Journal of Health-System Pharmacy 2002 59:1077-1082.
  3. JA Usher-Smith et al. Evaluation of the cost savings and clinical outcomes for switching patients from atorvastatin to simvastatin and losartan to candesartan in a primary care setting. International Journal of Clinical Practice 2007 61:15-23.
  4. SJ Nicholls et al. Statins, high-density lipoprotein cholesterol, and regression of coronary atherosclerosis. JAMA 2007 297: 499-508.
  5. R Butler, J Wainwright. Cholesterol lowering in patients with CHD and metabolic syndrome. Lancet 2007 369: 27.
  6. JC Moon, RG Bogle. How to lose a billion pounds. International Journal of Clinical Practice 2007 61: 2-3.
  7. R Minhas. Statin utilisation - recognising the role of the invisible hand. International Journal of Clinical Practice 2007 61: 3-6.

previous or next story