This Article  • References  • Full text PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citation map  • Citing articles on HighWire  • Contact me when this article is cited  Related Content  • Similar articles in this journal  Social Bookmarking   What's this?

The Trade-off Between Cost and Simplicity

Gustavo R. Heudebert, MD; Jerome Van Ruiswyk, MD, MS; Jan Hiatt, PharmD; Gordon Schectman, MD

Arch Intern Med. 1993;153(15):1828-1837.

Abstract
Background
The National Cholesterol Education Program recommends achievement of a defined target level of low-density lipoprotein cholesterol (LDL-C) for the treatment of hypercholesterolemia. They endorse the use of niacin and/or sequestrants as the first line of therapy to achieve such target LDL-C level. This recommendation has not been compared with the use of lovastatin as initial therapy if multidrug regimens are required to achieve goal LDL-C.

Methods
Prospectively collected data on tolerance and effectiveness for niacin, sequestrants, and lovastatin were obtained from a lipid clinic at a large mid-western Veterans Affairs medical center. We used a decision tree to compare the complexity and cost of three sequential drug algorithms used for the following initial LDL-C levels: 4.14, 4.91, 5.69, and 6.47 mmol/L (160, 190, 220, and 250 mg/dL). Algorithm 1 was niacin followed by a sequestrant and then lovastatin; algorithm 2, a sequestrant followed by niacin and then lovastatin; and algorithm 3, lovastatin followed by niacin and a sequestrant. Drug and laboratory costs were obtained from the pharmacy and pathology services at the same institution. Sensitivity analyses were performed on the tolerance and effectiveness of each drug as well as drug and laboratory cost estimates.

Results
The probability of achieving target LDL-C level (3.36 mmol/L [130 mg/dL]) decreased as initial LDL-C level increased. As a rule, algorithm 3 required fewer drugs in combination to achieve the target level for all initial LDL-C levels modeled. In addition, the use of lovastatin was high across all algorithms at all initial LDL-C levels modeled. Algorithm 1 was less expensive than algorithm 2 or 3 at a low initial LDL-C level (4.14 mmol/L [160 mg/dL]), with an average cost of $375 vs $454 vs $585, respectively. At all other initial LDL-C levels (4.91, 5.69, and 6.47 mmol/L [190, 220, and 250 mg/dL]), algorithm 2 was slightly less expensive than algorithm 1. Algorithm 3 became relatively less expensive as initial LDL-C level increased: 56% more expensive than algorithm 1 at an initial LDL-C level of 4.14 mmol/L (160 mg/dL) as compared with 7% more expensive than algorithm 1 at an initial LDL-C level of 6.47 mmol/L (250 mg/dL).

Conclusions
Fulfillment of the target LDL-C approach recommended by the National Cholesterol Education Program often requires the use of multiple drugs. When lovastatin is used initially, the regimen becomes simpler, albeit more expensive. At initial LDL-C levels of 4.91 mmol/L (190 mg/dL) or higher, this difference in cost becomes progressively smaller (7% at 6.47 mmol/L [250 mg/dL]), making algorithm 3 the better alternative; at low initial LDL-C levels (4.14 mmol/L [160 mg/dL]), a niacin-first regimen is reasonably simple and substantially less expensive. At moderate and severe initial LDL-C levels (4.91 and 5.69 mmol/L [190 and 220 mg/dL]), the lovastatin-first regimen may be advantageous.

(Arch Intern Med. 1993;153:1828-1837)

Author Affiliations
From the Department of Medicine, Medical College of Wisconsin, Zablocki Veterans Affairs Medical Center, Milwaukee.

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?

THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES

Forecasting Patient Outcomes in the Management of Hyperlipidemia
Brier et al.
Arch Intern Med 1999;159:569-575.
ABSTRACT | FULL TEXT