Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 Prabhus Exactly which studies are we looking for? BTW Another interesting abstract. Gerald Surya Unique Identifier 97356098 Authors Gupta R. Prakash H. Institution Department of Medicine, Monilek Hospital and Research Centre, Jawahar Nagar, Jaipur. Title Association of dietary ghee intake with coronary heart disease and risk factor prevalence in rural males. Source Journal of the Indian Medical Association. 95(3):67-9, 83, 1997 Mar. Abstract To determine the association between intake of dietary fat, specifically Indian ghee, and prevalence of coronary heart disease (CHD) and risk factors as study was undertaken on a rural population in Rajasthan. Total community cross-sectional survey was done using a physician administered questionnaire; 1982 males aged 20 years and more were studied. The dietary questionnaire focused on the amount and type of fat consumed. Staple dietary fat in this area is mustard/rapeseed oil and Indian ghee. To define the role of ghee, the average amount consumed in a month was determined; 782 males (39%) consumed 1 kg or more ghee per month (group 1) and 1200 (61%) consumed less than 1 kg per month (group 2). To elicit details of fatty acid composition of the diet consumed, detailed dietary history was acquired from a random 460 (23%) males; 220 from group 1 and 240 from group 2. Group 1 males were significantly younger, more literate and had more weight and body-mass index. This group consumed significantly more calories, saturated and mono-unsaturated fats while the consumption of polyunsaturated fats was similar in the two groups. Fatty acid intake analysis showed that group 1 males consumed more mono-unsaturated (n-9) fatty acids than group 2. Intake of polyunsaturated n-3 and n-6 fatty acids was similar. There was significantly lower prevalence of CHD in men who consumed > kg ghee per month (odds ratio = 0.23, 95% confidence limits 0.18-0.30, p < 0.001). Multivariate analysis confirmed this association (p < 0.001). The prevalence of hypertension and other coronary risk factors was similar in the two groups. On 21 Jan 1999, Jayo das wrote: > AGTSP > > <<Actually, I want to order some of these studies from Medline, but they > onlysend them to Health Science Libraries and there is some charge. I can > probably hustle a relationship with a local college, but do you have that > kind of access? Do you have any idea how much these studies cost?>>> > > Actually Krishna Susarla has informed me he has (probably free) > access to Medline through UT/SW Med School and probably, so does > Madhusuadani dd at UCSF. > > YS, > Jayo Das Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 Prabhus Exactly which studies are we looking for? BTW Another interesting abstract. Gerald Surya Unique Identifier 97356098 Authors Gupta R. Prakash H. Institution Department of Medicine, Monilek Hospital and Research Centre, Jawahar Nagar, Jaipur. Title Association of dietary ghee intake with coronary heart disease and risk factor prevalence in rural males. Source Journal of the Indian Medical Association. 95(3):67-9, 83, 1997 Mar. Abstract To determine the association between intake of dietary fat, specifically Indian ghee, and prevalence of coronary heart disease (CHD) and risk factors as study was undertaken on a rural population in Rajasthan. Total community cross-sectional survey was done using a physician administered questionnaire; 1982 males aged 20 years and more were studied. The dietary questionnaire focused on the amount and type of fat consumed. Staple dietary fat in this area is mustard/rapeseed oil and Indian ghee. To define the role of ghee, the average amount consumed in a month was determined; 782 males (39%) consumed 1 kg or more ghee per month (group 1) and 1200 (61%) consumed less than 1 kg per month (group 2). To elicit details of fatty acid composition of the diet consumed, detailed dietary history was acquired from a random 460 (23%) males; 220 from group 1 and 240 from group 2. Group 1 males were significantly younger, more literate and had more weight and body-mass index. This group consumed significantly more calories, saturated and mono-unsaturated fats while the consumption of polyunsaturated fats was similar in the two groups. Fatty acid intake analysis showed that group 1 males consumed more mono-unsaturated (n-9) fatty acids than group 2. Intake of polyunsaturated n-3 and n-6 fatty acids was similar. There was significantly lower prevalence of CHD in men who consumed > kg ghee per month (odds ratio = 0.23, 95% confidence limits 0.18-0.30, p < 0.001). Multivariate analysis confirmed this association (p < 0.001). The prevalence of hypertension and other coronary risk factors was similar in the two groups. On 21 Jan 1999, Jayo das wrote: > AGTSP > > <<Actually, I want to order some of these studies from Medline, but they > onlysend them to Health Science Libraries and there is some charge. I can > probably hustle a relationship with a local college, but do you have that > kind of access? Do you have any idea how much these studies cost?>>> > > Actually Krishna Susarla has informed me he has (probably free) > access to Medline through UT/SW Med School and probably, so does > Madhusuadani dd at UCSF. > > YS, > Jayo Das Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 >To determine the association between intake of dietary fat, specifically >Indian ghee, and prevalence of coronary heart disease (CHD) and risk factors >as study was undertaken on a rural population in Rajasthan. Total community >cross-sectional survey was done ..... Yes, this was an interesting study. However, the fact that this was a cross-sectional study, rather than a longitudinal one, makes it impossible to draw any conclusions re. causal relationships. Ys, Madhusudani dasi Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 >To determine the association between intake of dietary fat, specifically >Indian ghee, and prevalence of coronary heart disease (CHD) and risk factors >as study was undertaken on a rural population in Rajasthan. Total community >cross-sectional survey was done ..... Yes, this was an interesting study. However, the fact that this was a cross-sectional study, rather than a longitudinal one, makes it impossible to draw any conclusions re. causal relationships. Ys, Madhusudani dasi Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 WWW: Gerald Surya (New York NY - USA) wrote: > [Text 2067636 from COM] > > Prabhus > Exactly which studies are we looking for? > BTW Another interesting abstract. > Gerald Surya > I am meeting a local MD tommorrow (Tues) who has a health services library membership and an interest inlipids. I have a list of about 60 studies I'm tracking down now. could send you a list if seriously interested. > > > Fatty acid intake analysis showed that group 1 males consumed more > mono-unsaturated (n-9) fatty acids than group 2. Intake of polyunsaturated n-3 > and n-6 fatty acids was similar. There was significantly lower prevalence of > CHD in men who consumed > kg ghee per month (odds ratio = 0.23, 95% confidence > limits 0.18-0.30, p < 0.001). One kilo a month is about 2 1/3 tablespoons per day. Srila Prabhupada recommended 1 tablespoon a day. A book about Ayur Veda I'm reading says not more than 2 tablespoons, so the study would be consistent with those recommendations. Some ghee is beneficial, but like many things, too much of a good thing is not better. Additionally, in studies of ghee, one other variate is the terminology - sometimes vegetable oil is called ghee in common useage, and even supposed cow ghee can be adulterated with vegetable oils. I don't know if survey type studies take that into account. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 WWW: Gerald Surya (New York NY - USA) wrote: > [Text 2067636 from COM] > > Prabhus > Exactly which studies are we looking for? > BTW Another interesting abstract. > Gerald Surya > I am meeting a local MD tommorrow (Tues) who has a health services library membership and an interest inlipids. I have a list of about 60 studies I'm tracking down now. could send you a list if seriously interested. > > > Fatty acid intake analysis showed that group 1 males consumed more > mono-unsaturated (n-9) fatty acids than group 2. Intake of polyunsaturated n-3 > and n-6 fatty acids was similar. There was significantly lower prevalence of > CHD in men who consumed > kg ghee per month (odds ratio = 0.23, 95% confidence > limits 0.18-0.30, p < 0.001). One kilo a month is about 2 1/3 tablespoons per day. Srila Prabhupada recommended 1 tablespoon a day. A book about Ayur Veda I'm reading says not more than 2 tablespoons, so the study would be consistent with those recommendations. Some ghee is beneficial, but like many things, too much of a good thing is not better. Additionally, in studies of ghee, one other variate is the terminology - sometimes vegetable oil is called ghee in common useage, and even supposed cow ghee can be adulterated with vegetable oils. I don't know if survey type studies take that into account. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 > > > Yes, this was an interesting study. However, the fact that this was a > cross-sectional study, rather than a longitudinal one, makes it impossible > to draw any conclusions re. causal relationships. > > Ys, > Madhusudani dasi If you would please, a brief, accessible to the layman, definition of longitudinal and cross-sectional. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 > > > Yes, this was an interesting study. However, the fact that this was a > cross-sectional study, rather than a longitudinal one, makes it impossible > to draw any conclusions re. causal relationships. > > Ys, > Madhusudani dasi If you would please, a brief, accessible to the layman, definition of longitudinal and cross-sectional. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 At 19:51 -0800 2/1/99, COM: Madhava Gosh (das) ACBSP (New Vrindavan - USA) wrote: >If you would please, a brief, accessible to the layman, definition of >longitudinal and cross-sectional. Cross-sectional = "a snapshot in time", i.e. you measure your sample only once to see what they look like at that moment. Your sample can be diverse, but you only measure them once. If you wanted to e.g. study the relationship between age and cholesterol this way, you'd get samples of people of different ages and measure their cholesterol and then make statements about the 10 year olds, the 20 year olds, the 40 year olds etc. in your sample. You would probably also ask them about diet, exercise etc. and try to relate those to their cholesterol levels. However, the relationship between diet and cholesterol (if you found one) would be *correlational* only (note they only mentioned an *association*, that's pretty much the same thing). You couldn't make any statement re cause and effect, as A could cause B, B could cause A, or both A and B could be caused by C. Longitudinal = you recruit a sample (often referred to as a cohort) and follow them over time and take multiple measures. If you want to do the study above, you might recruit them at age 10, measure their cholesterol and then check back every 10 years to look for changes (if you could find them - big if). You'd probably also measure stuff like diet, exercise etc. each time too. Then you could use those measures at time 1 to try to predict cholesterol at time 2 etc. This would be a much stronger design, especially if you controlled for other factors (such as family history) already known to predict cholesterol levelss. Of course, since in this example, the group of people with different dietary habits would still be self selected (i.e. you didn't tell them what to eat), there could still be confounds. The only way you could really make definite statements about cause and effect is if you randomly assigned particpants to different diets and then followed them over time. That's not easy though, especially from an ethical point of view (e.g. imagine if you assigned a vegan to the ghee group). So even longitudinal relationships from non-randomized studies are often published, but with lots of statements about possible limitations. However, no one would take seriously a study that claimed a causal relationship but that had used a cross-sectional design only. Hope that helps. Ys, Madhusudani dasi Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 At 19:51 -0800 2/1/99, COM: Madhava Gosh (das) ACBSP (New Vrindavan - USA) wrote: >If you would please, a brief, accessible to the layman, definition of >longitudinal and cross-sectional. Cross-sectional = "a snapshot in time", i.e. you measure your sample only once to see what they look like at that moment. Your sample can be diverse, but you only measure them once. If you wanted to e.g. study the relationship between age and cholesterol this way, you'd get samples of people of different ages and measure their cholesterol and then make statements about the 10 year olds, the 20 year olds, the 40 year olds etc. in your sample. You would probably also ask them about diet, exercise etc. and try to relate those to their cholesterol levels. However, the relationship between diet and cholesterol (if you found one) would be *correlational* only (note they only mentioned an *association*, that's pretty much the same thing). You couldn't make any statement re cause and effect, as A could cause B, B could cause A, or both A and B could be caused by C. Longitudinal = you recruit a sample (often referred to as a cohort) and follow them over time and take multiple measures. If you want to do the study above, you might recruit them at age 10, measure their cholesterol and then check back every 10 years to look for changes (if you could find them - big if). You'd probably also measure stuff like diet, exercise etc. each time too. Then you could use those measures at time 1 to try to predict cholesterol at time 2 etc. This would be a much stronger design, especially if you controlled for other factors (such as family history) already known to predict cholesterol levelss. Of course, since in this example, the group of people with different dietary habits would still be self selected (i.e. you didn't tell them what to eat), there could still be confounds. The only way you could really make definite statements about cause and effect is if you randomly assigned particpants to different diets and then followed them over time. That's not easy though, especially from an ethical point of view (e.g. imagine if you assigned a vegan to the ghee group). So even longitudinal relationships from non-randomized studies are often published, but with lots of statements about possible limitations. However, no one would take seriously a study that claimed a causal relationship but that had used a cross-sectional design only. Hope that helps. Ys, Madhusudani dasi Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 > So even longitudinal > relationships from non-randomized studies are often published, but with > lots of statements about possible limitations. However, no one would take > seriously a study that claimed a causal relationship but that had used a > cross-sectional design only. > > Hope that helps. > > Ys, > Madhusudani dasi I guess they could be used to possibly determine a direction for more focused studies. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 > So even longitudinal > relationships from non-randomized studies are often published, but with > lots of statements about possible limitations. However, no one would take > seriously a study that claimed a causal relationship but that had used a > cross-sectional design only. > > Hope that helps. > > Ys, > Madhusudani dasi I guess they could be used to possibly determine a direction for more focused studies. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 At 4:42 -0800 2/2/99, COM: Madhava Gosh (das) ACBSP (New Vrindavan - USA) wrote: >I guess they could be used to possibly determine a direction for more focused >studies. Absolutely. Often less rigorous (by faster and cheaper) studies are done as pilots for larger, more comprehensive studies. Quote Link to comment Share on other sites More sharing options...
Guest guest Posted February 2, 1999 Report Share Posted February 2, 1999 At 4:42 -0800 2/2/99, COM: Madhava Gosh (das) ACBSP (New Vrindavan - USA) wrote: >I guess they could be used to possibly determine a direction for more focused >studies. Absolutely. Often less rigorous (by faster and cheaper) studies are done as pilots for larger, more comprehensive studies. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.