TABLE 13.
Main characteristics of randomised controlled trials which have investigated the effect of supplemental β‐carotene on the risk of lung cancer.
| ATBC (ATBC Study Group, 1994) | PHS (Cook et al., 2000) | WHS (Lee et al., 1999) | WACS (Lin et al., 2009) | |
|---|---|---|---|---|
| Country | Finland | USA | USA | USA |
| Sex/population |
Males Heavy smokers |
Males Physicians |
Females Health professionals |
Females Post‐menopausal |
| Age at recruitment | 50–69 years | 40–84 years | ≥ 45 years | ≥ 40 years |
| Health status | No prior cancer or serious illness | Apparently healthy | Apparently healthy | History of CVD or at least three risk factors for CVD |
| Smoking status | 100% heavy smokers (~ 1 pack/day for 36 years) | 11% current and 39% past smokers | 13% current smokers | 15% current and 41% past smokers |
| Intervention |
2 × 2 factorial Vitamin E, β‐carotene (20 mg/day) |
2 × 2 factorial Aspirin, β‐carotene (50 mg e.o.d) |
2 × 2 × 2 factorial Aspirin, vitamin E, β‐carotene (50 mg e.o.d) |
2 × 2 × 2 factorial Vitamin C, Vitamin E, β‐carotene (50 mg e.o.d) |
| β‐Carotene supplement (manufacturer) | Synthetic water soluble β‐catotene (Hoffman‐La Roche) | Lurotin, microencapsulated, water‐dispersible synthetic β‐carotene (BASF corporation) | Lurotin, microencapsulated, water‐dispersible synthetic β‐carotene (BASF corporation) | Lurotin, microencapsulated, water‐dispersible synthetic β‐carotene (BASF corporation) |
| Duration of the intervention | 6.1 years (median) | 12.9 years (mean) | 2.1 years (median) | 9.4 years (mean) |
| n on β‐carotene/n not on β‐carotene | 14,560/14,573 | 11,034/11,037 | 19,937/19,936 | 3807/3820 |
| Lung cancer cases on β‐carotene/not on β‐carotene (n) | 474/402 | 85/93 | 30/21a | 41/33 |
| Serum β‐carotene levels reached | 5.59 μmol/L | 2.19 μmol/L | NR | NR |
| Results | RR 1.18 (95% CI: 1.03, 1.36) | RR 0.9 (95% CI: 0.7, 1.2) | RR 1.43 (95% CI: 0.82, 2.49)b | RR 1.26 (95% CI: 0.80, 1.99) |
Abbreviations: CI, confidence interval; CVD, cardiovascular disease; e.o.d, every other day; NR, not reported; RR, relative risk.
Lung cancer cases reported for the trial period plus the 2 year follow‐up, i.e. for a total median duration of 4.1 years.
The RR and 95% CIs were calculated by Druesne‐Pecollo et al. (2010) using the number of cases and number of participants provided in the publication.