Altering one's diet for a short period of time — especially in old age — would not necessarily affect one's long-term health risks. It's not how much fat you eat that's important — it's what kinds.

As best scientists can tell, trans fats found in foods like margarine contribute to cardiovascular diseases, whereas unsaturated fats found in vegetable oils and fish actually have the opposite effect — lowering the risk of cardiovascular disease.

Saturated fats found in butter and red meat fall somewhere in between. Some of the biggest controversies surround saturated fats. Scientists disagree about the extent to which saturated fats contribute to important health outcomes like heart disease, stroke, and cancer.

The available research does suggest, however, that there are health benefits from replacing saturated fats with unsaturated fats in the diet, and that eating lots of nutrient-poor carbs like sugary cereals, soda, and white bread instead of fat is a bad idea.

Stay away from foods that are high in trans fats. And you're likely better off eating foods rich in unsaturated fat instead of saturated fat. But there doesn't seem to be any need to worry about your total fat intake.

So as long as you're eating a variety of real foods and not too many calories, you're doing well. Even so, the study serves as a reminder that the debate about the effects of dietary fat on the body is still very much alive among researchers, and that a lot of the thinking about "good fat" and "bad fat" over the past several decades may have been wrong.

Dietary fat is one of the most confusing — and controversial — food topics around. And no wonder: Americans have been hearing bizarrely mixed messages about whether it's okay to eat fat for more than half a century. In the s and '60s, saturated fat — the stuff found in red meat and butter —began acquiring a bad reputation.

Back then, researchers were finding that people with diets lower in saturated fat appeared to be healthier. Public health officials worried that eating too much saturated fat could lead to heart disease, a major killer in the United States.

This specific concern about saturated fats eventually mutated into a generalized panic about all types of fat. In the s, the official US dietary guidelines began warning Americans to cut their total fat intake.

This recommendation wasn't very scientifically grounded, since it didn't distinguish among types of fat at the time, researchers were also finding that unsaturated fats, such as those found in vegetable oils and fish, had health benefits.

But as Marion Nestle describes in Food Politics , the meat industry didn't want the government telling people to eat less red meat, a huge source of saturated fat.

So the message became the vague "eat less fat, period. Those warnings helped spur the "low-fat" diet craze of the past couple of decades. They also had a number of harmful unintended consequences.

Food manufacturers began replacing the fat in their products with sugar — think Snackwell's cookies — and marketed them as healthy alternatives. This turned out to be a bad idea: Those sugars and refined carbohydrates were often just as bad for health. Worse still, food manufacturers and consumers started moving away from saturated fats and toward artificial trans fats — as seen in the switch from butter to margarine.

This, too, was a disaster, since trans fats turned out to be very bad for the body. Today the conventional wisdom is shifting yet again. Some critics now argue that saturated fat isn't actually that bad for you, and that we all made a terrible mistake switching to low-fat diets that were higher in sugar.

In , the former New York Times food writer Mark Bittman declared, "Butter is back," going through the research on how saturated fat wasn't nearly as harmful as we thought and arguing that we should ditch artificial foods like margarine in favor of natural foods like, well, butter.

I decided to sift through the available evidence, interviewing eight researchers and reading more than 60 journal articles on the subject.

What I learned is that there's still a ton of controversy about fat — although there is also clarifying consensus in important areas. For starters, just about everyone agrees that the s-era recommendations about switching to a low-fat diet were not supported by science.

There isn't any high-quality evidence to back up that advice. In fact, researchers today generally don't think the total amount of fat you eat has much effect on obesity and heart health so long as you're eating healthy foods and not consuming too many calories.

Instead, they focus on what types of fat we should eat. Not all fats are created equal. More on that in the next section. Artificial trans fats appear to be extremely harmful, which is why they're now being banned from foods. Unsaturated fats, like those found in vegetable oils and fish, appear to have some health benefits.

Saturated fats fall somewhere in between. We've also learned that other types of ingredients, such as the highly refined carbohydrates that make up cookies and soda, can actually be just as unhealthy as "bad fats.

Now, this doesn't mean it's okay to eat a cheeseburger every single day. What it does mean, however, is that not all fats are bad and that fat can most certainly be a part of a healthy diet. These revelations have also triggered a debate among researchers about whether it's still useful to give dietary advice about macronutrients like fat and carbohydrates — or whether we should focus on actual foods instead.

Broadly speaking, there are three main types of dietary fats: saturated, unsaturated, and trans fats. Foods with fat contain some mixture of these three, and chemically they're all pretty similar chains of carbon atoms attached with hydrogen atoms. But they seem to do different things to the body.

Saturated fats are usually solid at room temperature. They're found at high levels in animal-based foods such as red meat beef, bacon , poultry, and full-fat dairy milks, butter, and cheese. Some plant-based foods, like coconuts and palm oil, are also high in saturated fat.

By contrast, unsaturated fats typically remain soft or liquid at room temperature. These are more likely to appear at high levels in fish and certain vegetables. There are two types: monounsaturated fats found in olive, peanut, and canola oils, avocados, almonds, pecans, pumpkin, sesame seeds, etc.

But both lean and fatty meats provide only trace amounts of these vitamins 1 , 2. Most conventionally raised beef today is fed a diet of grain, primarily corn, whereas a more natural diet for ruminant cows is grass. Because corn feed is rich in omega-6 fatty acids, the fatty acid profile of corn-fed beef is higher in omega-6s.

On the other hand, grass contains larger quantities of omega-3 fats, so grass-fed beef tends to contain more omega-3s 3. A diet that includes large amounts of omega-6s, without sufficient omega-3s to balance it out, may cause inflammation 4.

On the low carb, high fat, moderate protein keto diet, your body burns fat for energy rather than carbs 5. Therefore, a fatty cut of meat may be a better choice energy-wise, because it provides you with more fat that can be used for fuel.

One review that investigated several studies on saturated fat and heart disease found that the association between the two appears to be very weak 6. The recommendation to avoid saturated fat for heart health appears to arise from just a few studies that were not representative of the larger body of research.

Still, the American Heart Association recommends limiting saturated and replacing it with polyunsaturated fat 6. Fat has more calories than protein or carbs do, so fattier cuts of meat may add excess calories to your diet 1 , 2.

High fat processed meats such as bacon, sausage, and ham have also been linked to certain cancers, including colon cancer and stomach cancer 7 , 8 , 9. However, the mechanism behind this link is not yet clear, and most of the evidence for it comes from observational studies rather than high quality trials 7 , 8 , 9.

Finally, consider that leaner meats — like turkey, chicken, and fish — are also rich in nutrients and serve as excellent sources of protein. For this reason, you may consider including high fat red meats in your diet. They are rich in nutrients and ideal for people following a keto diet.

Additionally, grass-fed versions may offer higher quantities of anti-inflammatory omega-3 fatty acids. However, try to stick to unprocessed versions, since processed meats are associated with increased cancer risk.

Furthermore, to ensure that you have a balanced diet, speak with a healthcare professional such as a doctor or registered dietitian before starting or increasing consumption of red meat. Try this today: Want to dive deeper into the meat debate?

Check out this article about whether meat can fit into a healthy diet. Our experts continually monitor the health and wellness space, and we update our articles when new information becomes available.

VIEW ALL HISTORY. Many people believe red meat can harm your health. Here are the health effects of red meat, including possible benefits and downsides of adding it to…. If you're considering adding or removing meat from your diet, you may wonder whether meat is healthy.

This article explores the environmental and…. Person-times ended at the earliest of the following dates: date of pancreatic cancer diagnosis, date of death, or December 31, , the closure date of the study.

Tests based on Schoenfeld residuals showed no evidence that proportional hazards assumptions were violated for any analysis. Separate models for men and women showed similar patterns. Therefore, we present models including both sexes, adjusting for sex as a stratum variable to allow for different baseline hazard rates.

All Cox models were additionally stratified by follow-up time, categorized as 2 years or less, more than 2 to 5 years, and more than 5 years. Food group and nutrient exposures were investigated in disease models in terms of quintiles. Four dummy variables were created to represent the quintiles, which were based on the distribution of each exposure across the entire cohort men and women.

Median values for sex- and ethnic-specific quintiles were used in the respective models to test for trend. Age at cohort entry, ethnicity, history of diabetes mellitus, history of familial pancreatic cancer, smoking status never, former, or current smoker , and energy intake logarithmically transformed were used as adjustment factors in all multivariable models.

Energy was included so that the associations with foods and nutrients could be analyzed independently of their relationship to overall energy intake.

In additional analyses, we adjusted for pack-years of smoking as a more detailed measure of smoking. However, the risk estimates did not change, and therefore we chose to use only the smoking status variable because the data for this variable were more complete than those for pack-years of smoking.

In addition, models were adjusted for body mass index, educational attainment, fruit and vegetable intake, and alcohol consumption. However, risk estimates changed only marginally data not shown and therefore these adjustments were not included in the final models.

To reduce measurement error in the dietary assessments, we analyzed daily food and nutrient intakes in terms of densities, i.

As noted above, in the validation study we found that energy-adjusted intake produced substantially higher correlation coefficients with the reference instrument than did crude intake This phenomenon has also been reported in other studies Densities measure the contribution of the food or nutrient to the overall diet and are therefore interpreted differently from absolute measures.

By contrast, the use of absolute values assumes that a specific amount of a food or nutrient will have the same effect on risk, regardless of the energy content of the remaining diet. However, we also fitted all models using absolute measurements of intake grams per day , and the results data not shown led to the same conclusions.

For nutrients, intake was further adjusted by applying sex- and ethnicity-specific calibration functions derived from regression models of hour recall intakes on intakes in the QFFQ based on the calibration substudy.

Two sets of calibrated nutrients were computed. The first set included additional covariates in the model, such as age and body mass index, as described 19 , whereas the second set did not. Individuals with extreme diets were excluded from the calibration models as described above.

The calibrated nutrients were then used in a Cox regression model to test the trend in risk with increasing intake. The results from the two sets of calibrated nutrients were identical; therefore, we present those not adjusted for other covariates because fewer individuals were excluded due to missing values.

Calibration-adjusted intakes were not computed for foods because the day-to-day variability in food consumption is too high except for very broad groupings, such as all meat as a single item.

The likelihood ratio test was used to determine the statistical significance of the interaction between smoking status and dietary variables with respect to pancreatic cancer.

The test compares a main effects, no-interaction model with a fully parameterized model containing all possible interaction terms for the variables of interest. All analyses were performed using SAS Statistical Software, version 8 SAS Institute, Inc.

Further characteristics of study participants are shown in Table 1. Pancreatic cancer patients were, on average, 5 years older than nonpatients at cohort entry and included a higher percentage of men than nonpatients. Current smoking, a prior diagnosis of diabetes mellitus, and a familial history of pancreatic cancer were statistically significantly more common among cancer patients than among nonpatients.

There also were statistically significant differences in the ethnic distributions among pancreatic cancer patients and nonpatients; higher percentages of patients than nonpatients were African-Americans, Japanese-Americans, and Native Hawaiians.

Characteristics of pancreatic cancer patients cases and subjects without pancreatic cancer non-cases in the Multiethnic Cohort Study. P value from t tests for continuous measures and chi-square tests for categorical measures.

The associations between consumption of meat, dairy products, and eggs with pancreatic cancer are shown in Table 2. In the study population, median daily meat consumption, in terms of densities, ranged from 3. In general, high intakes of red meat and of processed meat were associated with an increased risk for pancreatic cancer, whereas consumption of poultry, fish, dairy products, and eggs showed no such association.

Consumption of pork and of total red meat i. Statistically significant positive trends were observed for both variables, although the trend for total red meat was not monotonic. The overall findings for red meat and processed meat were consistent in most ethnic groups considered separately data not shown , but the numbers of cases were too small for meaningful analyses.

The incidence rates, age adjusted to the age distribution of person-years in the cohort, were In unadjusted analyses, Cox models were stratified for sex and time on study. In multivariable analyses, Cox models were stratified for sex and time on study and adjusted for age at cohort entry, ethnicity, history of diabetes mellitus, familial history of pancreatic cancer, smoking status, and energy intake.

Fat intake from red meat and processed meat was slightly higher than fat intake from dairy products data not shown. Total fat showed no association with pancreatic cancer risk data not shown.

Table 3 shows the associations between percentage of energy as fat and risk of pancreatic cancer. None of the tests for trend showed statistically significant associations, whether or not they were based on the calibration-adjusted nutrient intakes.

In the separate analysis of fat from red and processed meat and fat from dairy products, however, we found that fat from meat but not fat from dairy products was associated with increased risks for pancreatic cancer. P trend values using calibration-corrected nutrient intakes are given in parentheses.

The calibration equations were sex and ethnicity specific and did not include additional covariates. The associations with saturated fat intake were similar to those with total fat.

Overall, percentage of energy from saturated fat showed no association with pancreatic cancer risk. Separate analyses for saturated fat from meat and from dairy sources showed positive associations between the risk for pancreatic cancer and fat from red meat and processed meat and essentially no association with fat from dairy products.

Neither absolute nor relative cholesterol intake was statistically significantly related to pancreatic cancer risk, and no statistically significant trend was seen across quintiles.

The same associations for trends were seen in analyses using calibration-corrected nutrient intakes as in analyses using uncorrected measurements Table 3. We also conducted an analysis based on estimated intake of nitrosamine, the major contributor to which was processed meat.

Finally, we found no evidence for an interaction between the meat food groups and smoking on the risk of pancreatic cancer data not shown. The effect seemed to be independent of energy intake.

Because the analysis of total fat and saturated fat intakes showed a statistically significant increase in risk only for meat sources, rather than overall and for dairy sources, fat is more likely to be an indicator of meat consumption than to be directly involved in the underlying carcinogenic mechanism.

Cholesterol intake was not related to pancreatic cancer risk. To date, seven prospective studies have investigated associations between consumption of various meats and pancreatic cancer 13 — 16 , 21 — Two found statistically significant positive associations with disease risk 22 , 23 , whereas four reported no associations 13 — 16 and one found a decreased risk with pork and sausage consumption All of these studies except one 13 included fewer than pancreatic cancer patients or used limited dietary assessment methods covering only a few food items 13 , 15 , 21 — Two cohort studies, the Nurses' Health Study NHS and the Alpha-Tocopherol, Beta-Carotene Cancer Prevention cohort ATBC study , that used comprehensive dietary assessments and reported null findings for meat intake also analyzed intake of fats as an exposure variable.

Findings from the NHS 14 were null for fat or fatty acid intakes and disease risk, whereas results from the ATBC study showed increases in risk with saturated fat intake and butter consumption Dairy product and egg consumption also were studied prospectively in two studies 14 , 15 , but no association with pancreatic cancer was found.

Because these studies were undertaken in selected study populations, i. It is also possible that the different results of our study and the NHS and ATBC study reflect different patterns of meat consumption in the three cohorts. For example, Caucasian men and women in our study ate less red meat, especially pork, and more poultry than those in the NHS and ATBC study.

Case—control studies of meat consumption and pancreatic cancer have also yielded inconsistent findings. Seven case—control studies reported a positive association between intake of different kinds of meat and pancreatic cancer 24 — 31 , whereas four case—control studies did not 32 — The positive associations were found for different meat items or groups: all meat 26 , 28 , 30 , 31 , red meat 24 , beef 26 , 27 , 29 , pork 25 , 29 , pork products 25 , 30 , and chicken Studies investigating the association of the intake of various dairy products with pancreatic cancer risk generally found no convincing associations 26 , 29 , 34 , One study reported an increased risk of pancreatic cancer among men only 25 , and another reported a decrease in risk with the consumption of fermented milk products Associations with fat intake have also been investigated in five case—control studies, all of which found no association 26 , 35 — For cholesterol, three of seven case—control studies showed statistically significantly increased risks with increasing intake 36 , 38 , 39 , whereas four studies reported null findings 26 , 32 , 35 , An increased risk with cholesterol intake in one study was assumed to be due to higher consumption of eggs among case patients than among control subjects In addition to total and saturated fat intake, exposure to mutagenic compounds produced during the cooking or preservation process has been considered as a possible explanation for the link between consumption of red meat and processed meat and pancreatic cancer risk.

Heterocyclic amines are formed when meats are cooked at high temperatures, and polycyclic aromatic hydrocarbons are formed when meats are charcoal broiled or grilled 40 , Both classes of compounds have been shown to be carcinogenic in animals 42 and could account for the red meat association.

N-nitroso compounds, which are found in nitrite-preserved meats or produced endogenously in the stomach when such meats are consumed, might underlie the positive association between processed meat consumption and pancreatic cancer risk 4 , Research into associations between meat preparation methods and cancer has been carried out mainly in the setting of colorectal cancer 43 , but a few case—control studies of pancreatic cancer are also available.

Anderson et al. Other studies have reported increased pancreatic cancer risks with the consumption of fried, grilled, cured, or smoked meats or foods 28 , 31 , Intake of meat that was not fried, grilled, cured, or smoked was not associated with cancer risk 28 , 33 , suggesting the possible role of polycyclic aromatic hydrocarbons, heterocyclic amines, or nitrosamines formed during these cooking or preserving processes.

Because fat, saturated fat, and cholesterol do not seem to be responsible for the increased risk of pancreatic cancer we observed with increasing intake of red meat or processed meat, our results also support the hypothesis that preparation methods such as grilling, frying, or curing play a role in the etiology of the disease.

The findings for nitrosamines in our study to some extent affirm this theory; because detailed information about preferences for doneness and preparation methods of meat have been obtained in a more recent follow-up questionnaire on the Multiethnic Cohort Study subjects, further pursuit of this hypothesis will be undertaken in the future.

There are some potential limitations to this analysis. The study population is from Hawaii and California only. However, the cohort was population based in design to maximize the generalizability of findings to the U.

population Also, the use of frequency questionnaires as assessment instruments can cause diet to be measured with error 44 , and measurement error is certainly present in our data. We attempted to minimize this limitation by rigorous design of the questionnaire; by emphasizing nutrient densities in the analyses, which resulted in better correlations between the food frequency questionnaire and more accurate comparison measurements of dietary intake 19 , 20 ; and by incorporating calibration-adjusted nutrient variables into the analyses.

The use of densities results in another possible limitation to the study, in that associations based on absolute intakes could differ. However, when we analyzed the data using absolute amounts, the findings were similar.

Another potential limitation relates to the fact that our efforts to correct for nutrient measurement errors were probably incomplete because the hour dietary recall method used as a standard was no doubt imperfect Nevertheless, this adjustment should result in relative risks closer to the true value.

We found that the effect of calibration was negligible. Another possible limitation pertains to the findings of statistically significant associations with foods but not nutrients. Although measurement error could be greater for nutrients than foods, it should be noted that our foods incorporated broad categories composed of many different items, each of which might have been measured with some error.

Also, a statistically significant association was found for fat from meat but not fat from dairy products, and there is no reason to suspect that the measurement error for these two fat sources would be substantially different. Another limitation could result from the fact that we adjusted our analyses for diabetes mellitus as a confounding factor.

Both red meat and processed meat intake have been positively associated with diabetes mellitus 45 — 47 , which itself is considered as a risk factor for pancreatic cancer 8 ; therefore, diabetes mellitus could be an intermediate rather than a confounding factor.

If so, adjusting for it could have distorted the observed associations. However, after exclusion of all participants with self-reported diabetes mellitus from the analysis, red meat and processed meat were still statistically significantly positively associated with pancreatic cancer in our study data not shown.

Our study also has several strengths. One is the large sample size, which resulted in the largest number of incident pancreatic cancer patients yet analyzed in a prospective study and therefore considerable statistical power.

Second, the prospective design ruled out the problem of recall bias, which can influence the findings from case—control studies. Third, due to the multiethnic background of the participants, the cohort included considerable dietary heterogeneity, facilitating the identification of meaningful associations.

Indeed, keeping in mind the fact that comparisons between studies have to be made cautiously owing to the different dietary assessment protocols, the ranges across quintiles were large when compared with the corresponding ranges for the NHS 14 and the ATBC study 16 , although the absolute intake of meat and fat was relatively low in our population.

Fourth, unlike many earlier studies, the food frequency questionnaire used was quantitative and comprehensive and therefore permitted adjustment for energy intake. In conclusion, our findings suggest that intakes of red meat and processed meat are positively associated with pancreatic cancer risk and thus are potential target factors for disease prevention.

The results raise the possibility that individuals might reduce their risk of pancreatic cancer by reducing consumption of red and processed meat. The age-adjusted incidence rates were However, because the fat components of the meats did not seem to account for the findings, other compounds in these foods that are responsible for the association need to be identified.

Future analyses of meat and pancreatic cancer risk should focus on meat preparation methods and related carcinogens. This work was supported in part by grant R37 CA from the National Cancer Institute, U. Department of Health and Human Services.

Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the agency. Jemal A, Clegg LX, Ward E, Ries LA, Wu X, Jamison PM, et al.

Annual report to the nation on the status of cancer, —, with a special feature regarding survival. Cancer ; : 3 — American Cancer Society. Atlanta GA ; Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet ; : — Risch HA. Etiology of pancreatic cancer, with a hypothesis concerning the role of N-nitroso compounds and excess gastric acidity.

J Natl Cancer Inst ; 95 : — Vimalachandran D, Ghaneh P, Costello E, Neoptolemos JP. Genetics and prevention of pancreatic cancer. Cancer Control ; 11 : 6 — Klein AP, Brune KA, Petersen GM, Goggins M, Tersmette AC, Offerhaus GJ, et al.

Prospective risk of pancreatic cancer in familial pancreatic cancer kindreds. Cancer Res ; 64 : —8. Calle EE, Murphy TK, Rodriguez C, Thun MJ, Heath CW Jr. Diabetes mellitus and pancreatic cancer mortality in a prospective cohort of United States adults.

Cancer Causes Control ; 9 : —