Tag Archive for: HR

BRI: It Just Doesn’t Matter

/in /by

Continuing our look at the BRI, the mathematician demonstrated that BRI is associated with body fat distribution. It makes sense; the waist measurement would provide an indication of fat around the waist. The next question is: would the BRI be a better predictor for cardiometabolic disorders than BMI?

BRI and Mortality

A group of researchers decided to use open-source data from the National Health and Nutrition Examination Survey database to examine the relationship between BRI and all-cause mortality. They coincided with the years that physical assessments were done including height and waist circumference; body weight was collected but not used in this instance. The time period began in 1999 and continued every two years through 2018.

There were two observations that were significant. First, in every demographic group, regardless of age, gender, or race/ethnicity, the BRI has increased during every examination period. As a country, the U.S. has gotten fatter. That matches every other measure such as body weight or BMI as well.

The second observation was that the hazard ratio (HR) increased as the BRI dropped below normal, then normalized when the normal BRI was reached, and the HR rose again as the BRI increased. Simply stated, there was an increased risk of mortality when people were too lean or too fat.

You may be wondering why I don’t give you a formula to do calculations for yourself. It’s very complicated and there are BRI calculators available on the website below. The main reason is that it just doesn’t matter—the BRI is no better at predicting mortality than the BMI. The researchers had the body weight data they needed to compare the BRI with the BMI directly. They just didn’t do it. However, looking at the mathematicians’ validation study, the categories of adiposity associated with BMI matches up quite well with the BRI and thus with body fatness. There’s no need for any more precision than is achieved with BMI.

The Bottom Line

It’s really the clinical use that seems to bother everyone, but with rare exceptions, the BMI gives an indication of body fatness. If physicians or other health care professionals cannot see the patient before them and realize they are too lean or too muscular to fit the typical interpretation of BMI, the fault lies with them, not the tool they are using.   

What are you prepared to do today?

        Dr. Chet

References:
1. JAMA. 2024; 332(16):1317-1318. 10.1001/jama.2024.20115
2. JAMA Netw Open. 2024; 7(6):e2415051. 10.1001/jamanetworkopen.2024.15051.
3. https://doi.org/10.1002/oby.20408
4. https://bri-calculator.com/#calculator

The FFQ: Still Too Vague

/in /by

I spent a long time examining validation and reliability studies on the Food Frequency Questionnaires (FFQ). It was interesting to compare the original validation studies with a new FFQ that was published in early 2024; researchers asked subjects in those studies that began decades ago to participate in this recent validation study.

The Stats

I learned more about a variety of statistics that I don’t typically encounter: coefficient of correlation, and then attenuated and deattenuated coefficient of correlations, and more. The researchers concluded that the “study showed that the FFQ used in prior studies has reasonably high reproducibility and validity in measuring food and food groups intakes among both women and men.” I disagree.

The coefficient of correlation is important (COC) because it gives an indication of the association of the variable with a standard, in this case a 7 Day Dietary Recall. The best COC is 1.0 or -1.0, which means it’s perfectly correlated or not correlated with the standard. A COC greater than 0.8 is considered a strong relationship, but a relationship of 0.6 – 0.79 is considered moderate.

The COC for most categories of food was well below 0.6. How can that in any way be valid? It may be reproduceable, but you’re reproducing the same mistake over and over again.

How Dangerous Is Meat?

High level analytics like this aren’t my area of expertise, but logic dictates that you can’t get precision even with large numbers of subjects. This is especially true when using FFQ data to correlate nutrition with disease. Remember the study on red meat intake and type 2 diabetes? The Hazard Ratio was only 10% per 100-gram serving of red meat. If the meat intake is moderately correlated, how much does any error of intake impact the HR?

Whether researchers are trying to estimate how much of each type of meat a person eats or trying to calculate the heme-iron content of that meat, the FFQ doesn’t have enough precision to be used in determining those values. Remember, the increase in HR was 10% per 100 grams—that’s 3.3 ounces—of unprocessed red meat per day. If a patty were 100 grams, a reasonable size, and you ate six patties every day, that would be 600 grams or over 1.5 pounds of hamburger patties per day. Would that raise the HR to 60% based on that single answer? What about a vegan who gets no heme iron? Would they never get type 2 diabetes? We know that’s not true either.

One more thing: People under-report what they eat. It can be 100 to 200 calories per day, or even up to 500 calories per day. No after-the-fact adjustment of the food intake can make up for that kind of imprecision.

The Bottom Line

What we’re left with is this: There may be a relationship between red meat, and subsequently, heme iron intake, and the risk of type 2 diabetes, but we don’t know how much. That’s about it. We’re going to need much better studies to nail that down before we make a pronouncement. For now, you’re probably safe eating red meat, especially if you keep this in mind: eat better, eat less, and move more.

What are you prepared to do today?

        Dr. Chet

References:
1. Am J Epidemiol. 1985;122(1):51–65.
2. Am J Epidemiol. 2024;193(1):170–179

Nutritional Epidemiology: Specification Curve Analysis

/in /by

Did you look up quadrillion? It’s a 1 with a whole lot of 0s—15 to be exact.

When I finished Saturday’s Memo, the researchers had chosen an area of nutritional epidemiology to focus on: the analytics used to analyze the data. They began with the premise that there are many ways to analyze any data set. They then identified published research studies that examined the consumption of red meat and mortality. They identified 15 publications reporting on 24 studies that examined the effect of red meat on all-cause mortality.

They weren’t doing a meta-analysis to see all the results of all the studies combined; they used a newer technique called specification curve analysis. They identified the type of data used in the analysis, the number of variables, the number of covariates, as well as demographic data. From that information, they then calculated the total number of ways each data set could be analyzed—the specification curve analysis. Turns out that number is 10 quadrillion! That exceeds the capacity of the computing power of a small country, and I can’t even imagine how much electricity that would consume.

They decided to take a randomized sample of the possible ways to analyze the data with specific variables and covariates in each and came up with 1,440 different approaches to analyzing the data. They ran additional tests on the data and eliminated 232 approaches because the data exceeded norms.

Then they ran the remaining analytic approaches on data from several years of the NHANES study. What did they find?

  • The median hazard ratio (HR) was 0.94 for the effect of red meat on all-cause mortality. That means the mortality risk was decreased 6% if the subject ate red meat.
  • HRs ranged from 0.51 to 1.75; 435 approaches yielded HRs more than 1.0 (increased risk) and 773 less than 1.0 (decreased risk). Most analyses showed that eating red meat reduced the HR, and thus reduced the risk of dying.
  • Of all the results, 48 (almost 4%) were statistically significant; of those, 40 indicated that red meat reduced all-cause mortality and 8 that red meat increased all-cause mortality.

Does this mean that eating red meat decreases your risk of dying early? We’re not done yet. We’ll put it all in perspective on Saturday.

What are you prepared to do today?

        Dr. Chet

References:
1. https://www.sensible-med.com/p/the-definitive-analysis-of-observational
2. Journal of Clinical Epidemiology 168 (2024) 111278