Lactose Intolerance Rate Depends on Definition

How many people in the U.S. are lactose intolerant (LI)? The number that has been given for the past generation has been about 30 - 50,000,000. That the estimate hasn't changed in 30 or more years should give you an instant clue that we're not talking about real firm counts. It's at best a guess, although you can extrapolate it from census information, as I did in my book Milk Is Not for Every Body: Living with Lactose Intolerance.

The last estimate I've seen for the U.S. population is 307,000,000. That would make about 10 - 16% of the population LI.

If you want a better percentage, Nutrition Today published Prevalence of Self-reported Lactose Intolerance in a Multiethnic Sample of Adults by Theresa A. Nicklas et al., September/October 2009 - Volume 44 - Issue 5 - pp 222-227 doi: 10.1097/NT.0b013e3181b9caa6.

Abstract
According to the National Institute of Diabetes and Digestive and Kidney Diseases, between 30 and 50 million Americans have the potential for lactose-intolerance symptoms. However, lactose-intolerance prevalence rates in practical life settings may be lower than originally suggested. The goal of this study was to determine the prevalence of self-reported lactose intolerance among a national sample of European American (EA), African American (AA), and Hispanic American (HA) adults.

A nationally representative sample of randomly generated telephone numbers was purchased from a commercial sample provider. A nationally representative sample of randomly selected telephone numbers were called from the Survey Research Unit's Computer-Assisted Telephone Interviewing facility at the University of Alabama at Birmingham. Interviews were completed by a total of 1,084 respondents between the ages of 19 and 70 years with 486 EAs, 355 AAs, and 243 HAs. The response rate was 24.2%, and the cooperation rate was 34.2%.

The age-adjusted lactose-intolerance prevalence estimates were 7.72%, 19.50%, and 10.05% for EAs, AAs, and HAs, respectively. For all respondents in the sample, the crude and age-adjusted self-reported lactose-intolerance prevalence rates were 13.38% and 12.04%, respectively.

These results indicate that the prevalence of perceived lactose intolerance is significantly lower than what has been previously estimated. Health professionals need to be aware of the misrepresentation of currently estimated lactose-intolerance rates and should continue to encourage individuals with lactose intolerance to consume dairy foods first to help meet key nutrient recommendations with proper guidance and education.

Several items from this abstract need comment.

The term lactose intolerance has three overlapping but separate definitions. The first is that you get the standard symptoms - gas, bloating, flatulence, diarrhea - after eating or drinking dairy. The second is reacting positively to a lactose intolerance test. The third is having the gene that stops the production of lactase at some time in your life.

That third definition is the one that is normally used to estimate how many Americans are LI. It has to be. There has never been before this a good large-scale study of the issue. And only a small percentage of Americans have ever had a formal lactose tolerance test, certainly nothing like the tens of millions.

Although the full paper is not online so I haven't yet been able to read it, I have to assume that most respondents considered themselves LI because they get - or believe they get - symptoms from dairy products. That's almost certainly a different and much smaller population than those whose ethnic heritage makes them likely to stop producing lactase.

And that's probably why the numbers are being presented as "significantly lower." Here the abstract alone is misleading. A 13% average is obviously not lower than an estimate of 10 - 16%.

However, both Hispanics and African-Americans have always been considered to be groups with very high likelihoods of carrying the LI gene. Estimates range more in the 50 - 80% vicinity than the 10 - 20% actually found.

Genetically that's still likely true. Many earlier studies have already shown the percentage of people who actively get symptoms from dairy is much lower than the genetics would indicate. Those who have little or no dairy in their normal diets or who generally eat low-lactose products like cheese or butter are less likely to have symptoms. We simply don't know what the gap is between definition 1 and definition 3.

Unless we do now. Or do we? If I'm reading that abstract correctly it took over 13,000 phone calls to get the 1084 people to complete the survey. (One in 4 responded and only one in 3 of those cooperated. So you need 12 times as many calls to get that number of completions.) Without a better view of the demographics of the final sample it's hard to know how representative the final group actually was.

An interesting survey. It will need close checking, though, to see if there is less here than meets the eye.