Research cost analyses to aid in decision making in the conduct of a large prevention trial, CARET. Carotene and Retinol Efficacy Trial

Our plain-English summary of Research cost analyses to aid in decision making in the conduct of a large prevention trial, CARET. Carotene and Retinol Efficacy Trial, by M.D. Thornquist, N. Urban, A. Tseng, et al (Controlled Clinical Trials, 1993).

The study

• Authors: M.D. Thornquist, N. Urban, A. Tseng, C. Edelstein, B. Lund, G.S. Omenn
• Journal: Controlled Clinical Trials
• Published: 1993;14(4):325-39
• Type of article: Randomized trial
• Cited by: 11 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Because of their larger study populations and longer durations, prevention trials typically are more costly than treatment trials. Thus it is important to analyze costs systematically to aid in making cost-effective decisions during the conduct of prevention trials as well as in the original design. Cost analysis must be tied to sample size estimation because costs depend on such factors as the total number of person-years of follow-up and the number of trial outcomes, which are not basic design parameters but are derived quantities resulting from sample size estimation. We illustrate the use of cost analysis to decide among options for future conduct of an ongoing prevention trial with three issues that have arisen during the Carotene and Retinol Efficacy Trial (CARET): the trade-off between extending the duration of the trial or increasing the number of participants, the effect on costs of delay in accrual, and the cost effectiveness of particular retention activities.

In plain English

This article was a cost analysis, not a study of health outcomes. The authors used a spreadsheet model linked to sample-size calculations to estimate the direct costs of CARET. CARET was a randomized, placebo-controlled trial testing beta-carotene and vitamin A to reduce lung cancer in asbestos-exposed workers and heavy cigarette smokers. The authors compared a base-case trial design with three alternatives, each projected to cost roughly $40 million in direct costs. Dropping a planned sixth study center at Irvine, California, would have meant randomizing 4300 fewer smokers and lengthening the trial by 1.1 years. That option was projected to cost $0.8 million less. Part of the saving came from needing fewer person-years of follow-up, because the model projected a higher lung cancer risk for asbestos-exposed participants than for smokers. The center was added anyway, and the authors said cost did not drive that decision. Delaying Irvine by a year was projected to cost $0.3 million. Halving nonadherence was projected to save $3 million. The authors concluded that cost analysis linked to sample-size estimation is a key decision-making tool. Their estimates excluded indirect costs and inflation.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Research cost analysis (spreadsheet cost model linked to sample-size estimation) of an ongoing randomized, double-masked, placebo-controlled prevention trial (CARET), comparing a base-case design with three alternative designs
• Who was studied: The CARET trial: recruitment goal of 3983 asbestos-exposed participants and 13,719 heavy smokers at six US study centers (Baltimore, Irvine, New Haven, Portland, San Francisco, Seattle), plus 1845 pilot participants (816 asbestos-exposed, 1029 heavy smokers) randomized in Seattle from 1985 to 1988. Unit costs came from staff surveys done in 1991.
• Exposure: The article analyzes trial costs, not asbestos exposure. Participants were asbestos-exposed workers and heavy cigarette smokers; fiber types and jobs are not described in this article. The trial intervention was daily beta-carotene (30 mg) plus 25,000 IU vitamin A as retinyl palmitate, compared with placebo.
• Main results:
  • Not adding the Irvine Study Center (Design B) was projected to cost $0.8 million less in direct costs than the base case (Design A). Under Design B the trial would last 1.1 years longer and randomize 4300 fewer smokers. Table 6 totals were $40.9 million for the base case and $40.1 million for Design B.
  • Design B was projected to need 4800 fewer person-years of follow-up. The authors attributed this to the projected higher lung cancer risk of asbestos-exposed participants compared with smokers, since Design B had a higher proportion of asbestos-exposed participants. This is a model projection, not an observed result.
  • The Irvine center was added despite the projected saving. The authors' reasons were concerns that a longer trial would reduce adherence and that recruitment targets might not be met; cost did not drive the decision.
  • Delaying the opening of the Irvine center by one year (Design C) was projected to cost the trial $0.3 million in direct costs.
  • Halving nonadherence among efficacy participants (Design D) was projected to shorten the trial from 9.3 to 8.4 years and save $3 million in direct costs. A 10% reduction in nonadherence was projected to save $0.8 million. Spending up to $170 per participant (halving) or $45 per participant (10% reduction) on adherence activities would therefore be cost-effective.
• The authors' conclusion: The authors concluded that costs should be monitored throughout a prevention trial, that cost analysis must be linked to sample-size estimation, and that, alongside other factors, it is a key tool in decision making.
• Limitations the authors state:
  • Indirect costs were not included in the calculations.
  • Costs were in 1989 dollars, unadjusted for inflation, and unit costs were assumed constant across the design options.
  • Costs of reviewing cancer outcomes other than lung cancer were left out of the cost equation.
  • Sporadic activities, such as handling possible vitamin A toxicity, were not included in the cost equation; the authors expected the effect to be negligible.
  • The equations used the same per-year follow-up costs for all participants, although pilot participants cost more to follow; the authors said this had little effect.
  • Once a trial has started, constraints prevent optimizing its design.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer.

Related studies

• Risk factors for lung cancer and for intervention effects in CARET, the Beta-Carotene and Retinol Efficacy Trial — G.S. Omenn, G.E. Goodman, et al. (1996)
• Lobe of origin and histologic type of lung cancer associated with asbestos exposure in the Carotene and Retinol Efficacy Trial (CARET) — C.A. Brodkin, J. McCullough, et al. (1997)
• Recruitment for the beta-carotene and retinol efficacy trial (CARET) to prevent lung cancer in smokers and asbestos-exposed workers — G.S. Omenn, G. Goodman, et al. (1992)
• CARET, the beta-carotene and retinol efficacy trial to prevent lung cancer in asbestos-exposed workers and in smokers — G.S. Omenn, G. Goodman, et al. (1991)
• Changes in cholesterol and triglyceride concentrations in the Vanguard population of the Carotene and Retinol Efficacy Trial (CARET) — B. Cartmel, J. Dziura, et al. (2005)
• Vitamin A and cancer prevention II: comparison of the effects of retinol and beta-carotene — N.H. de Klerk, A.W. Musk, et al. (1998)

Browse return to medical research.