Cost Effectiveness of Lipid Lowering Study - CELL


The Cost Effectiveness of Lipid Lowering (CELL) study was a randomized trial designed to assess the effects of "intensive advice" from primary care providers regarding lifestyle modification to reduce cardiovascular (CV) risk, versus "usual advice," in combination with an HMG-CoA reductase inhibitor, placebo, or neither. The primary endpoint was the change in Framingham risk score over 18 months of follow-up.

The study was run out of 32 health centers in Sweden. Despite the study title, cost data were not collected and cost-effectiveness ratios were not calculated.


1) The act of taking a daily medication may create a placebo effect on CV risk reduction.
2) Patients managed with "intensive advice" will achieve greater reductions in CV risk than those managed with "usual advice."

Study Design

Study Design:

Patients Enrolled: 681
Mean Follow Up: 18 months
Mean Patient Age: Mean 49 ± 7 years
Female: 15

Patient Populations:

1) Age 30-59.
2) Moderate hyperlipidemia, defined as serum total cholesterol ≥6.50 mmol/l on three repeated measures, but <7.79 mmol/l on the last, with an LDL:HDL ratio of ≥4.0 and serum triglycerides <4.0 mmol/l.
3) At least two additional cardiac risk factors (among male gender, active smoking, BMI >30 kg/m2, hypertension, family history of coronary artery disease before age 60, or personal history of CV disease).


1) Abnormal hepatic or renal function.
2) Elevated baseline serum creatine kinase.
3) History of drug abuse.
4) Type II diabetes mellitus.
5) History of pancreatitis.

Primary Endpoints:

Change in Framingham risk score from baseline to 18 months

Secondary Endpoints:

Changes in serum total cholesterol and cholesterol subfractions, body mass index, blood pressure, exercise score, and smoking status from baseline to 18 months

Drug/Procedures Used:

Patients were randomized in a 2x3 factorial design into one of six groups: intensive or usual advice plus either pravastatin, placebo, or no drug. Pravastatin was started at 10 mg/day and titrated to achieve a 15% reduction in total cholesterol.

Usual care subjects received advice at 1, 2, 6, 12, and 18 months after randomization on fat intake, exercise, weight loss, and smoking cessation. Intensive care subjects received the same, and in addition, attended local group sessions run by general practitioners and nurses.

Concomitant Medications:

Patients with hypertension had to have a diastolic blood pressure of <95 mm Hg prior to randomization, preferably by use of "lipid-neutral" drugs (i.e., not beta-blockers).

Principal Findings:

Significant decreases in the overall Framingham risk score from 0-18 months were only seen in subjects treated with pravastatin, and were similar between usual care (-0.16, 95% confidence interval [CI] -0.23, -0.09) and intensive care (-0.13, 95% CI -0.20, -0.06) groups. No differences between the placebo and the "no drug" groups were seen in any endpoint.

The Framingham scores increased by smaller amounts in the placebo and no drug groups receiving intensive care (+0.06-0.07) than the same groups receiving usual care (+0.16-0.20) due to larger declines in serum total cholesterol (-0.18 mmol/l vs. +0.00-0.07 mmol/l) and larger improvements in smoking scores.


In this primary prevention study, pravastatin favorably altered lipid profiles and overall Framingham risk scores to a similar degree whether subjects received usual or intensive health care advice from their primary care providers. There was no apparent benefit from taking placebo pills compared with no pills.

Among subjects not taking pravastatin, intensive advice was associated with greater reductions in total cholesterol and low-density lipoprotein and smaller increases in Framingham scores than usual advice; however, the magnitude of these changes was modest.


Main study results:
Lindholm LH, Ekbom T, Dash C, Isacsson A, Schersten B. Changes in cardiovascular risk factors by combined pharmacological and nonpharmacological strategies: the main results of the CELL study. J Intern Med 1996;240:13-22.

Supplemental Results:
Lindhom LH, Ekbom T, Dash C, Eriksson M, Tibblin G, Schersten B. The impact of health care given in primary care on cardiovascular risk. CELL Study Group. BMJ 1995;310:1105-9.

Clinical Topics: Diabetes and Cardiometabolic Disease, Dyslipidemia, Prevention, Atherosclerotic Disease (CAD/PAD), Homozygous Familial Hypercholesterolemia, Hypertriglyceridemia, Lipid Metabolism, Nonstatins, Novel Agents, Statins, Hypertension

Keywords: Coronary Artery Disease, Life Style, Cost-Benefit Analysis, Hyperlipidemias, Risk Reduction Behavior, Weight Loss, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Hypercholesterolemia, Primary Prevention, Cholesterol, Sweden, Placebo Effect, Pravastatin, Triglycerides, Hypertension, Primary Health Care, Smoking Cessation

< Back to Listings