Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Re folate intake, alcohol and breast cancer risk
- Wenbin Liang (7 October 2005)
-
Should it be encouraged to state the assumptions/inferences on measurements ?
- Wenbin Liang (9 October 2005)
-
Does Family History influence the results?
- Helen Carter (12 October 2005)
|
|
|||
|
Wenbin Liang, taking master of public health Curtin University of Technology
Send response to journal:
|
Dear Editor, 1) When the exposure of alcohol and folate were measured,[1] the exposure periods of the two substances would be needed to evaluate the total doses of exposure. 2) People who have a high folate intake (>400 micrograms/ per day) are more likely to take care of their health: they are likely to have higher physical activity level, consume a balance diet, consume “better” alcohol, pay more attention to or even overestimate their risk factors—they may overestimate the amount of alcohol that they have drunk. People who have high alcohol consumption (>40g/day) and low folate consumption may have more problems with their health, therefore they may visit their GPs more frequently, and high alcohol intake may draw the GPs’ attention to their potential breast cancer risk. This might increase the “risk” of these people to be early diagnosed. A further comparison of baseline characters between these two small groups (people who consume >40g alcoholic drinks, and consume either less than 200micrograms or more than 400 micrograms) may be appropriate. Reference: 1. Baglietto, L., et al., Does dietary folate intake modify effect of alcohol consumption on breast cancer risk? Prospective cohort study. Bmj, 2005. Competing interests: None declared |
|||
|
|
|||
|
Wenbin Liang, taking master of public health Curtin University of Technology
Send response to journal:
|
Dear Editor, It may be suitable to consider the questionaries in this article as an instrument to measure alcohol consumption and folate consumption.[1] However there are always assumptions/inferences that need to be made from what had been measured to what the exposure level really was. In this case, the subjects’ answers regarding their alcohol and food intake were used to estimate their alcohol and folate intake amount over a period of time. Therefore, it would be necessary to infer both the average level of alcohol intake and folate intake at a time, and the period that the intake level referred to (at least the length of the period) from the answers on the questionaries. Based on this assumption, a relationship between the dose of alcohol intake and folate intake can be further investigated; otherwise, the relationship being investigated is only about the relationship between the answers on the questionaries and the incidence of breast cancer. Similarly in another study on stress and breast cancer,[2] there was at least an assumption needed to make: the answers given by the subjects, reflected / strongly associated with the stress level of the subjects over a period of time (at least length of the time). Giving clear statements on the assumptions/inference on measurements, it would be much easier to evaluate the measurement method of a study based on the information provided in the study, and it would be easier to further construct a study to test the validity of that particular method. Reference 1. Baglietto, L., et al., Does dietary folate intake modify effect of alcohol consumption on breast cancer risk? Prospective cohort study. Bmj, 2005. 2. Nielsen, N.R., et al., Self reported stress and risk of breast cancer: prospective cohort study. Bmj, 2005. 331(7516): p. 548. Competing interests: None declared |
|||
|
|
|||
|
Helen Carter, Specialist Registrar in Public Health Birmingham University B15 2TT
Send response to journal:
|
Dear Editor, As a female doctor who enjoys a glass of red wine with her dinner I read with interest the article by Baglietto et al, (1), regarding dietry folate intake and whether it modifies the effect on breast cancer from alcohol. At baseline a structured interview was used to obtain information about potential risk factors including age, sex (?), country of birth, education, reproductive history and alcohol consumption. Similarly, other potential confounders were examined including eduction, Body Mass Index, age at menarche, Hormone Replacement Therapy, parity and use of multivitamins. I was surprised that missing from this list was family history of breast cancer in first degree relatives, smoking and use of oral contraceptive pill. I wondered if this information was collected and if so did it affect the results? Reference 1. Baglietto, L., et al., Does dietary folate intake modify effect of alcohol consumption on breast cancer risk? Prospective cohort study. Bmj, 2005. Competing interests: None declared |
|||