Variable Talk:Assessment:Exposure risks of Hg, dioxins and PCBs and nutritional benefits of fish consumption by means of Bayesian belief networks
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : In my opinion "age" is a driving variable in the fish case study. It is used to divide the target population into separate subgroups (basically children and adults or even more detailed sub-groups like 4 age groups as currently specified) and to estimate various health effects for these subgroups. Patrycja Jesionek ←#(2 P): : This is a reasonable strating point for the study. --Anna Karjalainen 12:48, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : We propose the following categories: 1) 0-2 years to describe exposure during periods of prenatal, suckling and infancy; 2) 2-18 years for childhood and puberty period; 3) 18-55 years for adulthoo; and 4) over 55 years to describe both a probable period of prevalence of many diseases such as cancer and cardiovascular diseases included as endpoints in the model and old age. --O. Leino, JT Tuomisto, Anna Karjalainen 12:48, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : Each age group has its own characteristics like exposure to fish constituents, body mass index and background incidences of health effects. For that reason "age" is connected in the diagram with "exposures", "body mass index" and "background incidences".Enter your comment between these two bars P. Jesionek ←#(2 P): : Yes, reasonable. --Anna Karjalainen 12:48, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
←#(1 P): : Body mass index for a given age group (if applicable) P. Jesionek ⇤#(2 P): : To my understanding, these should be used as means in the final interpretation of the health effecs, so we don't aim to describe any "exposure-responses of BMI on e.g. IQ." --Anna Karjalainen 12:48, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : Background incidence of that health effect in a given age group needed. P. Jesionek(EEST) ←#(1 P): : Yes, if available (should be mostly available at least in Finland in statistical databases) --Anna Karjalainen 13:23, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : Is beeing cathered currently at least by Henna Karvonen (FoodFiles), Leino (KTL), Anna Karjalainen (KTL) --Anna Karjalainen 13:23, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : In the preliminary BBN for the fish case study socioeconomic status has an influence on cancer and cardiovascular effects while smoking status affects cancer, cvd and developmental defects. Do we keep this set-up in the new BBN or do we assume that both statuses affect all health effects? P. Jesionek ←#(1 P): : Both statuses - SES and smoking affect all health effects --O. Leino, JT Tuomisto, Anna Karjalainen 13:23, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : How do we define health effects like developmental teeth defects and IQ change? Are these lifetime risks or childhood risks? P. Jesionek (EEST) ←#(2 P): : Both are childhood risks. Developmental defects can be defined as dental aberration dose-responses (exposure concentration correlated incidence of dental aberrations in humans). IQ change can be defined as unit or percentual change. --Anna Karjalainen 13:25, 5 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : Do we still assume that the background IQ change is zero? Or maybe there are some data on this and we will us it? P. Jesionek ←#(2 P): : At least for no, yes. --O. Leino, J.T. Tuomisto, Anna Karjalainen 13:23, 5 October 2007 |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : In my opinion each age group should have its own smoking status because of different percentages of smokers and non-smokers in these groups. --Patrycja Jesionek 15:46, 5 October 2007 (EEST) ←#(2 P): : Do I recall right that we agreed on to take into account only active smoking? I can surely see SES as a dependent variable of age (age-groups). --Anna Karjalainen 18:51, 8 October 2007 (EEST) ←#(3 P): : Yes, we agreed on active smoking. Thus, the differentiation between active smokers and non-smokers certainly applies to age groups 18-55yr and 55yr +. However, there might be also some percentage of people smoking in the age group 2-18yr. Do we take this into account or do we assume that there are only non-smokers in this age group? Obviously, there are no actively smoking children in age group 0-2yr. But, are we going to include information about actively smoking pregnant women and nursing mothers and examine impact of that smoking on health effects in the two youngest age groups (IQ loss and dental defects)? --Patrycja Jesionek 11:17, 9 October 2007 (EEST) ←#(number): : I would say yes for both, but ONLY if it is reasonably feasible to take into account smokers in age-group 2-18yr and passive smoking in age group 0-2yr. --Anna Karjalainen 13:33, 11 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : Do age groups differ with respect to socioeconomic status? SES could be different among age groups 18-55yr and 55yr+. What is the SES for age groups 0-2yr and 2-18yr? --Patrycja Jesionek 16:16, 5 October 2007 (EEST) ←#(2 P): : A reasonable assumption might be that the two youngest age groups's SES would be according to the parents SES, but I'm not really sure if this feasible. Anyhow, if we aim to country-specific populations, I suppose SES cannot be population specific. --Anna Karjalainen 18:51, 8 October 2007 (EEST) |
| Statements:
Resolution: Resolution not yet found. (A stable resolution, when found, should be updated to the main page.) |
| Argumentation:
--#(1 P): : During the workshop in Berlin it has been decided that socioeconomic status and smoking status should be discrete variables having respectively three (low/medium/high) and two (smoking/not-smoking) states. Moreover, as indicated in the discussion on this website, distributions of these variables are age-group and country specific. However, if socioeconomic status and smoking status are discrete variables then the modeling of dose-response functions become more complex and imposes higher requirements on data which in some cases may not be available. For example, suppose that we want to estimate the (remaining) lifetime cancer risk of people in the third age group (18-55yr) in a selected population. In order to determine the dose-response function in this case one needs to specify, among others, the background (or average) level of the (remaining) lifetime risk of cancer in the considered age group. If smoking and socioeconomic statuses impact cancer risk via background risk and are assumed to be discrete variables (with states as above) then the background risk is a function of these statuses and takes six different values. These values have to be extracted from data. If there is not enough data on background risk levels in different socioeconomic and smoking groups within a given age group we propose to replace discrete variables by continuous ones (they could be defined as daily intake of nicotine smoke and number of years of schooling received). This replacement implies that smoking and socioeconomic statuses won’t interact through background risk but they will be included in the dose-response function as additional variables, whose slopes in relation to the health effect studied have to be determined from data. This replacement will also allow capturing and quantifying effects of much larger group of interactions including interactions between intakes of fish constituents and smoking (and/or socioeconomic) status and interaction among smoking and socioeconomic statuses. --Patrycja Gradowska 17:13, 26 October 2007 (EEST) |
