Science the Soft Science of Dietary Fat

8/13/2019 Science the Soft Science of Dietary Fat

1/10

The Soft Science of Dietary FatAuthor(s): Gary TaubesSource: Science, New Series, Vol. 291, No. 5513 (Mar. 30, 2001), pp. 2536-2541+2543-2545Published by: American Association for the Advancement of ScienceStable URL: http://www.jstor.org/stable/3082809

Accessed: 15/03/2009 22:09

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=aaas.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the

scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that

promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

American Association for the Advancement of Scienceis collaborating with JSTOR to digitize, preserve and

extend access to Science.
http://www.jstor.org/stable/3082809?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=aaashttp://www.jstor.org/action/showPublisher?publisherCode=aaashttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/3082809?origin=JSTOR-pdf


2/10

K y. *; i ffi z

Mainstream nutritional science has demonized dietary fat, yet 50 years and hundreds of millions ofdollars of research have failed to prove that eating a low-fat diet will help you live longer

T h e o f t Science oietary a t

When heU.S.SurgeonGeneral'sOffice setoff in 1988 to write he definitive eport nthedangers f dietaryat,the scientific askappearedtraightforward.ouryearsearlier,theNational nstitutes f Health NIH)hadbegunadvising veryAmerican ld enoughto walkto restrict at intake,and thepresi-dent of the AmericanHeartAssociation(AHA)had told Timemagazine hat f ev-eryone went along, "we will have[atherosclerosis] onquered"by the year2000. The SurgeonGeneral'sOffice itselfhadjust published ts 700-pagelandmark"Report n Nutrition ndHealth," eclaringfat the single most unwholesome ompo-nentof theAmerican iet.All of this was apparentlybased onsoundscience. So the taskbefore he pro-ject officer was merely o gather hat sci-ence together n one volume, have it re-viewedby a committeeof experts,whichhadbeenpromptlystablished,ndpublishit. Theprojectdidnotgo smoothly, owev-er.Fourproject fficerscameandwentoverthe next decade."Itconsumed roject ffi-cers," says MarionNestle, who helpedlaunch heprojectand now runsthe nutri-tion and food studiesdepartment t NewYorkUniversity NYU). Membersof theoversight ommittee awdraftsof an earlychapter r two, criticized hemvigorously,and hen aw ittleelse.Finally,n June1999, 11yearsafter heprojectbegan,the SurgeonGeneral'sOf-fice circulated letter,authored y the lastof theprojectofficers,explaininghat hereportwould be killed. There was noother public announcement and nopressrelease.Theletterexplainedhatthe relevantadministrators did notanticipate ully the magnitude f theadditional xternal xpertiseandstaffresources that would be needed." Inotherwords, says Nestle, the subjectmatter"was oo complicated."ill Har-lan, a member of the oversightcommittee nd associatedirectorof the Office of Disease Preven-tion at NIH, says "thereportwasinitiatedwith a preconceivedopinionof theconclusions," ut the science be-hind those opinionswas not holding up.

"Clearlyhethoughts f yesterdagoingto serveus verywell."During hepast30 years, heeatinghealthy n Americahas beonymous with avoiding dietarcreation ndmarketing f reduceproductshas become big busi15,000 have appeared on suIshelves.Indeed, nentireresearehas arisen to createpalatablesubstitutes, and the food indispendsbillions of dollarsyearlyless-fat-is-good-healthmessageernmentweighs in as well, wit

"They say,'You reneed a high levelproof to changerecommendatior

which is ironicbecause they ne,had a high levelproof to set ther

-Walter Willett

iywerenot Department of Agriculture's (USDA'sbookleton dietary uidelines, ublishedvconceptof ery 5 years,and its ubiquitous oodGuid:come yn- Pyramid,which recommends hat fats andy fat. The oils be eaten"sparingly."he ow-fatgospeed-fat ood spreadsarther y a kind of societalosmoness; over sis, continuously einforced y physicianpermarket nutritionists,ournalists,healthorganizach ndustry tions,and consumer dvocacygroups uchnonfat fat as the CenterorScience nthe Public nterustry now est, which refers to fat as this "greasysellingthe killer." InAmerica,we no longer earGodThe gov- or the communists,but we fearfat,"saysh the U.S. DavidKritchevskyf the Wistar nstitutenPhiladelphia,who in 1958 wrote the firstextbook n cholesterol.:ally As the SurgeonGeneral'sOffice dis-covered,however,hescienceof dietary aI of is notnearlyas simpleas it onceappearedTheproposition, ow50 yearsold,thatdi-the etary fat is a bane to health is based, chieflyon the fact thatfat,specifically helS, hard, aturatedat foundprimarilyn meaanddairyproducts, levatesblood choles:' terol levels. This in turn raises the likeli-fver hood that cholesterolwill clog arteries,aconditionknownas atherosclerosis, hichof then increasesrisk of coronary rterydis-ease, heartattack,anduntimelydeath.Byn." the 1970s, each individual step of thischainfromfat to cholesterol o heartdis-ease had been demonstrated eyondreasonable doubt, but the veracity of thechainas a wholehasneverbeenproven. notherwords,despitedecadesof research,tis still a debatable ropositionwhether heconsumptionof saturatedats above rec-ommended evels (step one in the chainby anyonewho'snotalready thighriskofheartdiseasewill increase he likelihoodof untimelydeath(outcome hree).Norhave hundredsof millions of dollarsintrialsmanaged o generatecompellingevidence that healthy individualsMH_i ^-. can extend their lives by

lIk^jw / : more than a few weeks, ifthat,by eatingless fat(see sidebar onp. 2538). To put itsimply, the data_ , ou as towhethemain ambigu-?:;A:&A:,.......: ~ ous as to whethe30 MARCH 001 VOL291 SCIENCEwww.sciencemag.org536


3/10

lzv,, . i .low-fat diets will benefit healthyAmeri-cans. Worse, the ubiquitous admonish-ments to reduce total fat intake haveencourageda shift to high-carbohydratediets, which may be no better-and mayeven be worse-than high-fatdiets.Since the early 1970s, for instance,Americans' verage at intakehas droppedfromover40% of totalcalories o 34%;av-erage erum holesterolevelshavedroppedas well. But no compellingevidencesug-gests that these decreaseshaveimprovedhealth.Although eartdiseasedeath ateshavedropped-and publichealthoffi-cials insist low-fat diets are partlyresponsible-the ncidence f heartdis-ease does not seem to be declining,aswouldbeexpectedf lower at dietsmadea difference. his was theconclusion,orinstance, f a 10-year tudyof heartdis-easemortality ublishedn TheNew En-gland Journal of Medicine in 1998, whichsuggestedthatdeathratesaredeclininglargelybecause doctors are treating hediseasemoresuccessfully.AHA statisticsagree:Between1979and1996, henumberof medicalproceduresorheartdisease n-creased rom 1.2 millionto 5.4 million ayear."Idon'tconsider hat his diseasecate-goryhasdisappearedranythinglose to it,"saysoneAHAstatistician.Meanwhile,obesityin America,whichremained constant from the early 1960sthrough 1980, has surged upwardsincethen-from 14%of thepopulationo over22%. Diabetes has increasedapace.Bothobesityand diabetes ncreaseheartdiseaserisk,whichcouldexplainwhyheartdiseaseincidence s notdecreasing. hat his obesi-tyepidemic ccurredustas thegovernmentbeganbombardingmericanswiththelow-fatmessage uggests hepossibility, owev-erdistant,hat ow-fatdietsmighthaveun-intended consequences-among them,weight gain."Mostof us wouldhavepre-dicted hat if we canget the populationochange ts fatintake,with ts densecalories,we wouldsee a reductionnweight," dmitsHarlan. Instead, e see the exactopposite."In the face of this uncertainty,kepticsandapostateshave comealong repeatedly,onlyto see theirworkalmost eligiously g-noredas themainstream edical ommuni-ty soughtconsensuson the evils of dietaryfat.For20 years, or instance, he HarvardSchoolof PublicHealthhas run heNurses'Health Study and its two sequelae-theHealthProfessionals ollow-UpStudyanda theNurses'HealthStudyII-accumulating| overa decadeof dataon the diet andhealth| of almost300,000Americans.The results- suggest hat otalfatconsumed asno rela-| tion to heartdiseaserisk; hatmonounsatu--rated ats like olive oil lowerrisk;and thatu saturatedats arelittleworse, f at all, than

the pastaand othercarbohydrateshat theFood GuidePyramid uggestsbe eatenco-piously. Thestudiesalsosuggest hat ransfatty acids are unhealthful.These are thefatsinmargarine,orinstance, ndarewhatmanyAmericans tarted atingwhentheywere told thatthe saturated ats in buttermightkill them.)Harvard pidemiologistWalterWillett, pokespersonortheNurses'HealthStudy,pointsoutthatNIHhasspentover$100 million on the threestudiesand

"InAmerica, we nolonger fear God or the

communists, butwe fear fat."

-David Kritchevskyyetnot onegovernmentgencyhaschangeditsprimary uidelineso fit theseparticulardata."Scandalous,"aysWillett."They ay,'Youreallyneed a high level of prooftochange the recommendations,'which isironic,because heyneverhad a highlevelof proof o setthem."Indeed, he historyof the national on-viction hatdietaryat is deadly, nd ts evo-lution romhypothesiso dogma, s one inwhichpoliticians,bureaucrats,he media,and hepublichaveplayedas largea role asthe scientists ndthe science.It'sa storyofwhatcanhappenwhen he demands f pub-lic healthpolicy-and the demandsof thepublic orsimpleadvice-run up againstheconfusing mbiguityf realscience.Fearof fatDuringhe firsthalf of the 20thcentury, u-tritionistswere more concerned boutmal-nutrition han about he sins of dietary x-cess. AfterWorldWarII, however, coro-

nary heart disease epidemic seemed tosweep hecountry see sidebar n p. 2540)."Middle-aged en,seemingly ealthy,weredroppingdead,"wrote biochemistAncelKeysof theUniversity f Minnesota, winCities,who was among he firstto suggestthat dietaryfats might be the cause. By1952, Keys was arguingthat Americansshouldreduce heir fat intake o less than30% of totalcalories,although e simulta-neouslyrecognizedhat"direct videnceonthe effectof the dieton human rteriosclero-sis isvery ittleand ikely oremain o forsome ime."nthefamous ndverycon-| troversial evenCountries tudy,orin-stance,Keysandhiscolleagues eportedthat he amount f fatconsumedeemedto be thesalientdifference etween op-ulations uchas those nJapan ndCreteH thathad ittleheartdiseaseand hose,asin Finland, hat wereplaguedby it. In1961, the FraminghamHeart Studylinked holesterolevels o heartdisease,Keysmade he coverof Timemagazine,and heAHA,underhisinfluence, eganadvocatingow-fatdietsasapalliativeormen with high cholesterol evels. Keyshad alsobecomeone of the firstAmeri-cans to consciouslyadopta heart-healthydiet:He andhiswife,Time eported,donoteat 'carvingmeat'-steaks, chops,roasts-more han hree imesa week."Nonetheless, y 1969 he stateof the sci-ence could still be summarizedy a singlesentence romareport f theDiet-Heart e-view Panel of the NationalHeartInstitute(nowthe NationalHeart,Lung,and BloodInstitute, or NHLBI): "It is not knownwhether ietarymanipulationasanyeffectwhatsoever n coronary eartdisease." hechairof thepanelwas E. H. "Pete"Ahrens,whoselaboratoryt RockefellerUniversityin New YorkCitydidmuch of the seminalresearch n fatandcholesterolmetabolism.Whereasproponentsof low-fat dietswereconcerned rimarily bout he effectsof dietaryat on cholesterolevelsandheartdisease,Ahrensand his panel-10 expertsin clinical medicine, epidemiology,bio-statistics, umannutrition, ndmetabolism-were equallyconcerned hateatinglessfat couldhaveprofound ffectsthroughoutthebody,manyof whichcouldbe harmful.Thebrain, or instance, s 70%fat, whichchiefly erves o insulate eurons. at s alsotheprimary omponent f cell membranes.Changingheproportionf saturatedo un-saturated ats in the diet changesthe fatcomposition in these membranes. Thiscould conceivablychangethe membranepermeability,which controls he transportof everythingromglucose, signalingpro-teins,andhormonesobacteria, iruses, ndtumor-causing gents into and out of thecell.Therelative aturationf fats nthediet

www.sciencemag.org CIENCEVOL 91 30 MARCH001 2537


4/10

NEWS FOCUS

What IfAmericansAte LessSaturated Fat?Eat less saturated fat, live longer.For 30 years, this has stood asone cornerstoneof nutritionaladvicegivento Americans see maintext). But how much longer?Between 1987 and 1992, three inde-pendent researchgroups used computer models to work out theanswer.All three analyses agreed,but their conclusions have beenburied n the literature, arely f ever cited.

Allthree models estimated how much longer people might ex-pect to live, on average, if only 10% of their calories came fromsaturatedfat as recommended.Inthe processtheir total fat intakewould dropto the recommended30% of calories.All three modelsassumed that LDL holesterol-the "bad cholesterol"-levelswould drop accordinglyand that this diet would have no adverseeffects, although that was optimistic at the time and has becomeconsiderablymore so since then. Allthree combined national vitalstatistics data with cholesterol risk factor data from the Framing-ham HeartStudy.The first study came out of HarvardMedical School and waspublishedin the Annalsof InternalMedicine n April1987. LedbyWilliam Taylor, t concluded that individuals with a high risk ofheart disease-smokers, for instance,with high blood pressure-could expect to gain,on average,one extrayear by shunningsatu-rated fat. Healthy nonsmokers, however, might add 3 days to3 months. "Although here are undoubtedly persons who wouldchoose to participate in a lifelong regimen of dietary change toachieve results of this magnitude, we suspect that some might

not,"wroteTaylor nd his colleagues.The following year, the U.S.SurgeonGeneral'sOffice funded astudy at the Universityof California, an Francisco,with the expec-tation that its results would counterbalance hose of the Harvardanalysis.Ledby epidemiologistWarrenBrowner,his study conclud-ed that cutting fat consumption in America would delay 42,000deaths each year,but the net increasein life expectancywould av-erage out to only 3 to 4 months. The key word was "delay," ordeath, like diet, is a trade-off: Everyonehas to die of something."Deaths are not prevented, they are merely delayed," Brownerlater wrote. "The saved'people mainlydie of the same things ev-eryone else dies of;they do so a little later in life."To be precise,awoman who might otherwise die at 65 could expect to live twoextra weeks after a lifetime of avoidingsaturatedfat. If she lived tobe 90, she could expect 10 additional weeks.The thirdstudy,fromresearchers at McGillUniversity in Montreal, came to virtuallyidentical conclusions.Brownerreported his results to the Surgeon General'sOffice,then submitted a paper to TheJournalof the AmericanMedicalAssociationU(AMA).Meanwhile, he SurgeonGeneral'sOffice-hissource of funding-contacted JAMA nd tried to prevent publica-tion, claimingthat the analysiswas deeply flawed.JAMA eviewersdisagreedand publishedhis article,entitled"WhatIfAmericansAteLess Fat?" n June 1991. As for Browner,he was left protectinghiswork from his own funding agents. "Shootingthe messenger,"hewrote to the Surgeon General's Office, "or creating a smokescreen-does not changethose estimates." -G.T.

could also influence cellular aging as wellas the clottingabilityof blood cells.Whether the potentialbenefits of low-fatdiets would exceed the potentialrisks couldbe settled by testing whether low-fat dietsactually prolong life, but such a test wouldhave to be enormous. The effect of dieton cholesterol levels is subtle for mostindividuals-especially those living in therealworld rather han the metabolicwards ofnutrition researchers-and the effect ofcholesterol levels on heart disease is alsosubtle. As a result,tens of thousandsof indi-vidualswould have to switch to low-fat dietsand theirsubsequenthealthcompared o thatof equal numbers who continued eating fatto alleged excess. And all these peoplewould have to be followed for years untilenoughdeathsaccumulated o providestatis-

tically significantresults.Ahrensand his col-leagues were pessimistic about whether sucha massive and expensive trial could ever bedone. In 1971, an NIH task force estimatedsuch a trial would cost $1 billion, consider-ably more than NIH was willing to spend.Instead, NIH administrators opted for ahandful of smaller studies, two of whichalone would cost $255 million. Perhapsmore important, hese studies would take adecade. Neither the public,the press,nor theU.S. Congresswas willing to wait that ong.Science by committeeLike the flourishing American affinity foralternative medicine, an antifat movementevolved independently of science inthe 1960s. It was fed by distrust of theestablishment-in this case, both the medi-

cal establishment and the food industry-and by countercultureattacks on excessiveconsumption, whether manifested in gas-guzzling cars or the classic American cui-sine of bacon and eggs and marbled steaksAnd while the data on fat and health re-mained ambiguous and the scientific com-munity polarized, the deadlock was brokennot by any new science, but by politiciansIt was Senator George McGovern'sbipartisan, nonlegislative Select Committee onNutrition and Human Needs-and, to beprecise, a handful of McGovern's staffmembers-that almost single-handedlychanged nutritional policy in this countryand initiated the process of turning the di-etary fat hypothesis into dogma.McGovern's committee was founded in1968 with a mandateto eradicatemalnutrition in America, and: ?' "There comes a point when, if you

don't make a decision, the conse-quences can be great as well. Ifyoujust allow Americans to keep on

consuming 40% of calories from fat,there's an outcome to that as well."-Basil Rifkind

it instituted a seriesof landmark federalfood assistance pro-grams.As the malnu-trition work began topeter out in the mid-1970s, however, thecommittee didn't dis-band. Rather, ts gen-eral counsel, Mar-shall Matz, and staffdirector,Alan Stone,both young lawyers,decidedthatthe com-mittee would address

30 MARCH 001 VOL291 SCIENCEwww.sciencemag.org538


5/10

"overnutrition,"he dietary excesses ofAmericans. t was a "casual ndeavor,"aysMatz."We eallywere otallynaive,a bunchof kids,whojust thought,Hell,we shouldsay something n thissubjectbeforewe goout of business.'"McGovernndhis fellowsenators-all middle-aged men worriedabouttheirgirthand theirhealth-signedon; McGovern ndhis wife had bothgonethrough iet-guru athanPritikin'sery owfat diet and exerciseprogram.McGovernquit heprogramarly, utPritikin emainedamajornfluence n histhinking.McGovern's committee listened to2 daysof testimonyon diet and disease nJuly 1976.ThenresidentwordsmithNickMottern,a former aborreporter or TheProvidence Journal, was assigned the taskof researchingand writingthe first "Di-etaryGoals for the United States."Mot-tern,who hadno scientificbackgroundndno experiencewritingaboutscience,nutri-tion, or health,believedhis DietaryGoalswould launch a "revolution in diet andagriculturen thiscountry." e avoided hescientificand medicalcontroversy y rely-ing almostexclusivelyon Harvard choolof PublicHealthnutritionistMarkHegstedfor inputon dietary at.Hegstedhad stud-ied fat andcholesterolmetabolism n theearly 1960s,and he believeduncondition-ally in the benefitsof restrictingat intake,although e sayshe wasaware hathis wasan extremeopinion.WithHegstedas hismuse,Mottern awdietary at as the nutri-tionalequivalent f cigarettes, nd he foodindustry s akin to the tobacco ndustrynits willingnessto suppress cientifictruthinthe interests f profits.ToMottern,hosescientistswho spokeout againstfat werethose willing to take on the industry."Ittook a certainamount f guts,"he says,"tospeakabout this because of the financialinterestsnvolved."Mottern's eport uggested hatAmeri-cans cut their otalfat intake o 30%of thecalories heyconsumeand saturatedat in-take to 10%, n accordwith AHA recom-mendations or men at high risk of heartdisease.Thereport cknowledgedhe exis-tence of controversybut insisted Ameri-cans hadnothingto lose by followingitsadvice. "Thequestionto be asked is notwhy should we change our diet but whynot?"wrote Hegstedin the introduction."There re[no risks] hatcan be identifiedand important enefits can be expected."This was an optimisticbut still debatable

m position,and whenDietaryGoals was re-| leased in January1977, "all hell brokeloose," recalls Hegsted. "Practicallyno-bodywasin favorof the McGover recom-mendations. amn ewpeople."McGovernespondedwith three ollow-u up hearings,whichaptlyforeshadowedhe

NEWS FOCUSnext 7 yearsof controversy. mongthosetestifying,orinstance,was NHLBIdirectorRobertLevy, who explainedthat no oneknew if eating less fat or loweringbloodcholesterol evels wouldpreventheart at-tacks,whichwaswhyNHLBIwasspending$300 million to studythe question.Levy'spositionwas awkward, e recalls,because"thegoodsenators ameoutwith heguide-lines and hencalledus intogetadvice."Hewasjoined by prominentcientists, nclud-ing Ahrens, who testified that advisingAmericans o eat less fat on the strength fsuchmarginal vidence was equivalent oconducting nutritionalxperiment ith he

American public as subjects. Even theAmericanMedical Associationprotested,suggesting that the diet proposedby theguidelines aised he"potentialor harmfuleffects."But as these scientistsestified, odidrepresentativesrom hedairy, gg, andcattleindustries,who also vigorouslyop-posed the guidelinesfor obvious reasons.This uxtapositionerved o taint he scien-tific criticisms: Any scientists arguingagainst he committee'suidelines ppearedto be eitherhopelessly ehind heparadigm,whichwasHegsted's iew,orindustry pol-ogists,whichwasMottern's,f not both.Although he committee ublished re-vised editionof the DietaryGoals later ntheyear, he thrust f the recommendationsremained nchanged.t didgiveinto indus-trypressure y softeninghesuggestionhatAmericans at less meat. Mottern ays heconsideredeven that a "disservice o thepublic,"efused o do therevisions, ndquitthe committee.Mottern ecamea vegetari-an whilewritingheDietaryGoalsandnowrunsa foodco-op nPeekskill,NewYork.)

The guidelinesmighthave then died aquiet death when McGovern's ommitteecame to an end in late 1977if two federalagencies had not felt it imperative o re-spond. Althoughthey took contradictorypointsof view, one message-with mediaassistance-wonout.The first was the USDA, whereconsumer-activistCarol TuckerForemanhad recentlybeen appointedan assistantsecretary. oreman elieved t was incum-bent on USDA to turnMcGovern'secom-mendations nto official policy, and,likeMottern, he was not deterred y the exis-tenceof scientific ontroversy.Tellus whatyou know and tell us it's not the final an-swer," he wouldtell scientists."I have toeat and feedmy childrenhree imes a day,and I wantyou to tell me whatyourbestsenseof the data s rightnow."Of course, given the controversy, he"best sense of the data"woulddependonwhich scientistswere asked.The Food andNutrition oardof the NationalAcademy fSciences NAS),whichdecides he Recom-mendedDietaryAllowances,wouldhavebeen a natural hoice, but NAS presidentPhilipHandler,an experton metabolism,had told Foreman hat Mottern'sDietaryGoals were "nonsense." Foreman thenturned o McGovern's taffers for adviceandthey recommended he hire Hegsted,which shedid.Hegsted,n turn,reliedon astate-of-the-scienceeportpublishedby anexpertbutvery divergent ommittee f theAmericanSociety for ClinicalNutrition."Theywere nowherenear unanimousonanything,"ays Hegsted,"but he majoritysupportedsomethinglike the McGoverncommitteeeport."Theresulting ocument ecame he firsteditionof "Using heDietaryGuidelinesorAmericans." lthought acknowledgedheexistence f controversyndsuggestedhatasingledietaryecommendationightnotsuitan entirediversepopulation,he advice toavoid at andsaturatedatwas, ndeed, irtu-ally denticaloMcGover'sDietaryGoals.Threemonths ater, he NAS FoodandNutrition oard eleasedts ownguidelines:"TowardHealthfulDiets."Theboard, on-sistingof adozennutritionxperts, onclud-ed thatthe onlyreliableadvice forhealthyAmericanswas to watchtheirweight;ev-erythingelse, dietaryfat included,wouldtakecareof itself.The advicewas not takenkindly,however, t least not by the media.The first reports-"rather ncredulously,"said Handler at the time-criticized theNAS advice orconflictingwith heUSDA'sandMcGovern's nd thus somehowbeingirresponsible. ollow-up eports uggestedthat the boardmembers, n the words ofJaneBrody,who covered he storyfor TheNew York imes,were "allin thepocketof

www.sciencemag.org SCIENCE VOL291 30 MARCH 001 2539


6/10

NEWS FOCUS

TheEpidemicThatWasn't?Forhalf a century,nutritionistshave pointedto soaringdeath ratesas the genesis of their research nto dietary fat and heart diseaseand as reason to adviseAmericans o eat less fat (see main text)."Wehad an epidemic of heart disease after WorldWarII," besityexpertJules Hirschof RockefellerUniversity n New YorkCity saidjust 3 months ago in The New YorkTimes."Therates were growinghigherand higher,and people became suddenlyawareof that, andthat diet was a factor."To proponents of the antifat message, this heart disease epi-demic has always been an indisputablereality.Yet,to the statisti-cians at the mortality branchof the National Center for HealthStatistics (NCHS), he source of all the relevantstatistics, the epi-demic was illusory.In their view, heart disease deaths have beensteadily decliningsince the late 1940s.According o HarryRosenberg,directorof the NCHSmortalitybranch ince 1977, the keyfactor n the apparent pidemic,paradoxi-cally,was a healthierAmericanpopulation.Bythe 1950s, prematuredeathsfrom infectiousdiseasesand nutritional eficiencieshad beenallbuteliminated,which eft moreAmericansiving ongenoughto dieof chronicdiseasessuch as heartdisease. notherwords, he actualrisk

of dying roma heartattackat any particularge remained nchanged:Rather,he risingnumberof 50-year-oldsdroppingdead of heart at-tackswasprimarilyueto the risingnumber f 50-year-olds.Thesecondaryfactor was an increase from 1948 to 1968 in theprobability hat a death would be classifiedon a death certificateas arterioscleroticdisease or coronaryheart disease. This increase,however,was a figment of new diagnostictechnologies-the wideruse of electrocardiograms,or instance-and the changingtermi-nology of death certificates. In 1949, the InternationalClassifica-tion of Diseases (ICD)added a new category,"arteriosclerotic eartdisease,"underthe more generalrubric"diseasesof the heart."Theresult,as a 1958 reportto the AmericanHeart Association noted,was dramatic: Inone year,1948 to 1949, the effect of this revisionwas to raisecoronarydisease death rates by about 20% for whitemales and about 35% for white females."In1965, the ICDaddedacategory for coronaryheart disease, which addedyet more deathsand cappedoff the apparentepidemic.To Rosenbergand others at NCHS, he most likely explanationfor the postwar upsurge in coronaryheart disease deaths is thatphysiciansslowly caught on to the new terminology and changedthe wordingon death certificates."There s absolutely no evidencethat there was an epidemic," ays Rosenberg. -G.T.

the industrieseinghurt." obeprecise, heboard hairandoneof its members onsult-ed for food industries, ndfunding or theboard tself camefrom ndustrydonations.Theseindustry onnectionswere leaked othepress rom he USDA.Hegstednow defends heNASboard, l-thoughhe didn'tat the time,and calls thiskind of conflictof interest ahell of an is-sue.""Everybodysed to complainhat n-dustrydidn'tdo anything n nutrition,"etoldScience,"yetanybodywhogotinvolvedwasblackballedecauseheirpositionswerepresumablynfluenced y the industry."In1981,Hegstedreturned o Harvard,wherehis researchwas fundedby Frito-Lay.)hepresshad mixedfeelings,claiming hat heconnections soiled"he academy'seputa-tion "for enderingareful cientificadvice"2.1 -1.9-1.7-1.5-1.3-1.1 -0.9 -0.7-0.5

TotalMortality

F

I I I 0 I240Bloodtotal cholesterol (mg/dl)

The big picture. Pooled risk ratios of death from


7/10

had failed-but nowtheyhad establishedfundamentalink in the causalchain,fromlowercholesterol evels to cardiovascularhealth.With hat, heycould ake heleapoffaith fromcholesterol-lowering rugsandhealth to cholesterol-lowering diet andhealth.And afterall theireffort, heywereeager-not to mention rgedby Congress-to renderhelpfuladvice."Therecomes apointwhen,if you don't make a decision,theconsequencesan be greataswell," aysRifkind. "If you just allow Americans okeep on consuming40% of calories fromfat, here's noutcomeo thataswell."With heLRC esultsnpress, heNHLBIlaunched hatLevycalled"amassivepublichealth ampaign."hemediaobliginglywentalong.Time, orinstance, eportedhe LRCfindingsunder he headline Sorry,t'sTrue.Cholesterolreally is a killer."The articleabout drug rialbegan: Nowholemilk.Nobutter.No fattymeats ..." Time followed up3 months later with a cover story: "AndCholesterolndNow the BadNews. .." Thecoverphotowas a frowningace:a breakfastplatewith two friedeggs as theeyes and abacon trip or the mouth.Riflindwasquot-ed saying hat heir esults"stronglyndicatethat hemoreyou ower holesterol nd at nyourdiet,the moreyoureduceyourrisk ofheart isease," statementhat till ackeddi-rect cientificupport.ThefollowingDecember,NIHeffective-ly ended hedebatewitha "Consensus on-ference."The idea of such a conference sthatanexpertpanel, deallyunbiased,istensto 2 daysof testimony nd arrives t a con-clusionwith whicheveryone grees. nthiscase,Rifkind hairedheplanning ommit-tee, which chose his LRCco-investigatorSteinbergo lead the expertpanel.The 20speakersdid includea handfulof skeptics-including Ahrens, or nstance, ndcardi-ologistMichaelOliverof ImperialCollegein London-who argued hat t was unsci-entific to equate he effects of a drugwiththe effectsof a diet.Steinberg's anelmem-bers,however, s Oliver ater omplainednTheLancet,"were elected o includeonlyexpertswhowould,predictably,aythatalllevels of blood cholesterol n the UnitedStatesare too highandshouldbe lowered.And, of course, this is exactly whatwassaid." ndeed,heconferenceeport,writtenby Steinberg ndhispanel,revealed o evi-dence of discord.Therewas "nodoubt,"tconcluded, hat low-fat diets "will affordsignificant protection against coronaryheartdisease" to everyAmericanover 2yearsold. TheConsensusConference ffi-

N cially gave the appearanceof unanimitywherenone existed.Afterall, if therehad0 beena trueconsensus, s Steinberg imselftold Science, "youwouldn't have had tos havea consensus onference."

NEWS FOCUSThe test of timeTo the outsideobserver, he challengeinmaking ense of anysuch ong-runningci-entific controversys to establishwhethertheskeptics resimplyon thewrong ideofthe new paradigm, r whether heirskepti-cism is well founded. notherwords, s thescience at issue basedon sound scientificthinking ndunambiguousata,or is it whatSirFrancisBacon, orinstance,wouldhavecalled "wishful cience,"based on fancies,opinions, ndthe exclusionof contraryvi-dence?Baconofferedone viablesuggestionfordifferentiatinghe two: the test of time.Goodscience s rootednreality,o itgrowsanddevelopsandthe evidencegetsincreas-

inglymorecompelling,whereaswishful ci-ence flourishesmost under ts firstauthorsbefore"goingdownhill."Such is the case, for instance,with thepropositionhatdietary at causescancer,which was an integralpartof dietaryfatanxiety n the late 1970s.By 1982,the evi-dencesupportinghis idea was thought obe so undeniable hat a landmark AS re-porton nutrition nd cancerequated hoseresearcherswho remainedskepticalwith"certain nterestedparties[who] formerlyargued hat the associationbetweenlungcancerandsmokingwas not causational."Fifteenyearsand hundreds f millionsofresearchdollars ater,a similarlymassiveexpert eport ytheWorldCancerResearchFundand he American nstitute or CancerResearchcould find neither"convincing"nor even"probable"eason o believethatdietaryat caused ancer.Thehypothesishat ow-fatdiets are therequisiteoute oweightosshas aken sim-ilardownward ath.This was the ultimatefallback ositionn all low-fat ecommenda-

tions:Fat has nine caloriesper gramcom-pared o fourcalories orcarbohydratesndprotein, nd ocuttingatfrom hedietsurelywouldcutpounds. This s heldalmost o bea religioustruth," ays Harvard'sWillett.Considerableata,however,owsuggest th-erwise.Theresults f well-controlledlinicaltrialsare consistent: eopleon low-fatdietsinitiallyose a coupleof kilograms, s theywouldonanydiet,and hen heweight endsto return.After1 to 2 years, ittlehas beenachieved.Consider,orinstance,he50,000women nrolledn theongoing$100millionWomen'sHealthInitiative WHI).Half ofthese women have been extensivelycoun-seled to consume nly20%of theircaloriesfrom at.After3 yearsonthisnear-draconianregime,sayWHIsources, he women hadlost,onaverage, kilogramach.The link betweendietary at and heartdisease is morecomplicated,because thehypothesishas diverged nto two distinctpropositions:irst,that owering holesterolpreventsheartdisease;second, hateatingless fat notonlylowers holesterol ndpre-vents heartdiseasebutprolongs ife. Since1984, he evidence hatcholesterol-loweringdrugsarebeneficial-propositionnumberone-has indeed blossomed, at least forthose at high risk of heartattack. Thesedrugsreduceserumcholesterolevels dra-matically,andthey preventheartattacks,perhaps y othermeansas well. Theirmar-ket has now reached 4 billiona year ntheUnited Statesalone, and every new trialseems o confirmheirbenefits.The evidence supportingthe secondproposition,hateatingless fat makes fora healthierand longer life, however,hasremainedstubbornlyambiguous.If any-thing,it has only becomeless compellingovertime. Indeed, inceAncel Keysstart-ed advocating low-fat diets almost50 yearsago,thescienceof fat andcholes-terol has evolvedfrom a simple storyintoa very complicated one. The catch hasbeen that few involved in this businesswerepreparedo deal with a complicatedstory.Researchersnitiallypreferredo be-lieve it was simple-that a single un-wholesome nutrient,in effect, could beisolatedfrom the diverse richnessof hu-mandiets;publichealthadministratorse-quireda simplestoryto give to Congressand the public; and the press needed asimple story-at least on any particularday-to give to editorsandreaders n 30column nches.But as contrarian atacon-tinued to accumulate,the complicationsbecameincreasinglymore difficultto ig-nore or exclude,and thepressbeganwaf-flingoraddingcaveats.The scientists hengot the blamefor not sticking o the origi-nal simple story,which had,regrettably,neverexisted.

www.sciencemag.org SCIENCE VOL291 30 MARCH001 254


8/10

NEWS FOCUSMore ats, fewer answersTheoriginal imple story n the 1950swasthathigh cholesterol evels increaseheartdisease risk. The seminal FraminghamHeartStudy, or instance,which revealedthe association between cholesterol andheartdisease,originallymeasured nlyto-tal serumcholesterol.But cholesterol hut-tles throughhe blood in an arrayof pack-ages. Low-density lipoprotein particles(LDL, the "bad"cholesterol) deliver fatand cholesterol from the liver to tissuesthat need it, includingthe arterialcells,where it can lead to atheroscleroticplaques.High-density ipoproteinsHDLs,the "good"cholesterol)returncholesterolto the liver.Thehigher heHDL,the lowerthe heart disease risk. Then there aretriglycerides,which contain fatty acids,and very low density lipoproteins(VLDLs),which ransportriglycerides.All of theseparticleshave some effecton heartdiseaserisk,while thefats,carbo-hydrates, ndprotein n the diet havevary-ing effects on all these particles. The1950s story was that saturatedfats in-crease total cholesterol, polyunsaturatedfats decrease t, and monounsaturatedatsare neutral.By the late 1970s-when re-searchers ccepted he benefits of HDL-theyrealized hatmonounsaturatedats arenot neutral. Rather,they raise HDL, atleastcomparedo carbohydrates,nd low-er LDL.This makes hem an ideal nutrientas far as cholesterolgoes. Furthermore,saturated ats cannot be quite so evil be-cause, while they elevate LDL, which isbad, heyalso elevateHDL,which s good.And some saturated ats-stearic acid, inparticular,the fat in chocolate-are atworstneutral.Stearicacid raisesHDL lev-els butdoes little or nothing o LDL. Andthen hereare rans attyacids,whichraiseLDL, ust like saturatedat,but also lowerHDL.Today,none of this is controversial,although t has yet to be reflected n anyFoodGuidePyramid.To understandwhere this complexitycan lead in a simple example,considerasteak-to be precise,a porterhouse,electcut,witha half-centimeterayerof fat,thenutritionalconstituentsof which can befound in the NutrientDatabase or Stan-dardReference t the USDA Website. Af-terbroiling, his porterhouseeducesto aservingof almostequal partsfat andpro-tein.Fifty-onepercentof the fat is mono-- unsaturated,f whichvirtually ll (90%) sg oleic acid, the same healthyfat that's in

2 olive oil. Saturatedat constitutes 5% ofS the total fat, but a thirdof thatis stearic, cid,whichis, at the very least,harmless.Theremaining % of the fat is polyunsatu-| rated,which also improves holesterol ev-o els. Insum,well overhalf-and perhaps s

The antifat movementwas founded on thePuritannotion that

"something bad had tohave an evil cause, andyou got a heart attackbecause you did some-

thing wrong, whichwas eating too muchof a bad thing, ratherthan not having enoughof a good thing."

-John Powles

much as 70%-of the fat content of aporterhouse ill improve holesterolevelscomparedo whattheywould be if bread,potatoes,or pastawereconsumed nstead.The remaining30% will raise LDL butwill also raise HDL.All of this suggeststhateatinga porterhouseteak rather hancarbohydrates ightactually mprove eartdisease risk, althoughno nutritionalau-thoritywho hasn't writtena high-fatdietbookwill saythispublicly.As for the scientific tudies,n theyearssince the 1984 consensusconference, heonething heyhavenotdone s pile upevi-dencein supportof the low-fat-for-all p-proacho thepublicgood.If anything,heyhave addedweight to Ahrens'sfears thattheremaybe a downside opopulationwidelow-fatrecommendations.n 1986, for in-stance, ust 1 yearafterNIH launched he

National CholesterolEducationProgram,alsoadvisingow-fatdiets oreveryone ver2 yearsold,epidemiologist avidJacobsofthe Universityof Minnesota,TwinCities,visitedJapan. herehe learnedhatJapanesephysicianswereadvisingpatients o raisetheircholesterolevels,because ow choles-terol levels were linked to hemorrhagicstroke.At thetime,Japanesemen weredy-ing from stroke almost as frequentlyasAmericanmen were succumbing o heartdisease.Backin Minnesota, acobs ookedfor this low-cholesterol-strokeelationshipin the MRFITdataand found t there, oo.And the relationship ranscended troke:Menwithvery ow cholesterolevelsseemedproneto premature eath;below 160 mil-ligramsperdeciliter mg/dl), he lower hecholesterolevel, heshorterhe ife.Jacobsreportedhis results to NHLBI,whichin 1990hosted a conference o dis-cuss the issue, bringing together re-searchers from 19 studies around theworld.The datawere consistent:When n-vestigatorstrackedall deaths, insteadofjust heart disease deaths,the cholesterolcurveswereU-shaped or men and flat forwomen.In otherwords,men with choles-terol levels above 240 mg/dltended o dieprematurelyromheartdisease.But below160 mg/dl, the men tendedto die prema-turelyfromcancer,respiratory nddiges-tivediseases,and rauma.As forwomen, fanything, he highertheircholesterol, helonger hey ived(see graphonp. 2540).Thesemortalitydatacan be interpretedin twoways.One,preferred y low-fatad-vocates, s thattheycannotbe meaningful.Rifkind, or instance, old Sciencethattheexcess deathsat low cholesterolevels mustbe due to preexisting onditions. n otherwords,chronic llnessleads to low choles-terol evels,notvice versa.Hepointedo the1990 conferencereportas the definitivedocument nthe ssueandassupportorhisargument,althoughthe reportstates un-equivocally hatthis interpretations notsupportedythe data.The otherinterpretations that what alow-fatdiet does to serumcholesterol ev-els, and whatthat n turndoes to arteries,may be only one componentof the diet'seffect on health. In other words, whilelow-fatdietsmight help preventheartdis-ease,they mightalso raisesusceptibilityootherconditions.This is whatalwayswor-ried Ahrens. It's also one reasonwhy theAmericanCollege of Physicians,for in-stance,now suggests that cholesterolre-duction s certainlyworthwhile orthose athigh, short-term iskof dyingof coronaryheart disease but of "much smaller or ...uncertain" enefit foreveryone lse.Thisinterpretation-thathe connectionbetween diet and health far transcends



9/10

cholesterol-is alsosupported ythesinglemost dramatic iet-heartrialeverconduct-ed: the Lyon Diet Heart Study, led byMichelde Lorgeril f the FrenchNationalInstituteof Health andMedical Research(INSERM) ndpublishedn CirculationnFebruary 999.The investigatorsandom-ized 605 heart attack survivors, all oncholesterol-lowering drugs, into twogroups.Theycounseled ne to eat an AHA"prudent iet,"very similar o thatrecom-mended or all Americans.Theycounseledthe other o eat a Mediterranean-typeiet,with morebread,cereals, egumes,beans,vegetables, ruits,and fish and less meat.Total at andtypesof fatdifferedmarkedlyinthe twodiets,buttheHDL,LDL,and o-tal cholesterolevels in the two groupsre-mainedvirtually identical. Nonetheless,over 4 years of follow-up, the Mediter-ranean-diet group had only 14 cardiacdeathsandnonfatalheartattacks omparedto 44 for the "Western-type"iet group.The likely explanation,wrote de Lorgeriland his colleagues, s that the "protectiveeffects[of the Mediterraneaniet]werenotrelatedto serumconcentrationsof total,LDLorHDLcholesterol."Manyresearchersind the Lyondataso perplexinghatthey're eft question- cing the methodology of the trial.Nonetheless,says NIH'sHarlan, hedata"arevery provocative.They dobringup the issue of whether f welook only at cholesterol evels wearen'tgoing to miss somethingvery important." e Lorgerilbe-lieves the diet'sprotectiveeffectcomes primarilyfrom omega-3fatty acids, found in seed oils,meat, cereals,greenleafy vegeta-bles, andfish, and fromantioxidantcompounds,ncluding itamins,raceel-ements, nd lavonoids.He toldScience hatmost researchersnd ournalistsn the fieldareprisoners f the "cholesterolaradigm."Althoughdietary at and serumcholesterol"areobviouslyconnected,"he says, "theconnection is not a robustone" when itcomes o heartdisease.Dietarytrade-offsOne inescapablerealityis that death is atrade-off, nd so is diet."Youhave to eatsomething,"says epidemiologistHughTunstall Pedoe of the University ofDundee, U.K., spokesperson or the 21-nationMonitoringCardiovascular iseaseProject unby the WorldHealthOrganiza-tion. "Ifyoueat moreof onething,youeata lot less of somethingelse. So for everytheorysayingthis disease is causedby anexcessinx, youcanproducean alternativetheory saying it's a deficiency in y." Itwould be simple if, say, saturatedfats

NEWS Focuscould be cut fromthe diet andthe calorieswith it, butthat'snot the case. Despiteallexpectationso thecontrary, eopletend oconsume he samenumberof caloriesde-spitewhateverdietthey try.If theyeat lesstotal fat, for instance,they will eat morecarbohydrates nd probably ess protein,becausemost proteincomes in foods like

"Most of us wouldhave predicted that ifwe can get the popula-tion to change its fatintake, with its densecalories, we would seea reduction in weight.

Instead, we see theexact opposite."-William Harlan

Lnrpretathatlsohave con-siderableamounts f fat.This plus-minus problemsuggests a differentinterpretationorvirtually v-erydietstudyeverdone,including, orin-stance, he kind of metabolic-wardtudiesthatoriginallydemonstratedhe abilityofsaturated ats to raise cholesterol. If re-searchersreduce the amountof saturatedfat in the test diet, they have to makeupthe calories elsewhere. Do they addpolyunsaturatedats, for instance,or addcarbohydrates? single carbohydrate r

mixed carbohydrates? o they add greenleafy vegetables, or do they add pasta?And so it goes. "Thesky'sthe limit," aysnutritionistAlice Lichtenstein of TuftsUniversityn Boston."There re a millionperturbations."These trade-offs lsoconfound hekindof epidemiologicaltudies hat demonizesaturated at from the 1950s onward.Inparticular, ndividuals who eat copiousamountsof meatanddairy products,andplentyof saturatedatsin theprocess, endnot to eat copiousamountsof vegetableand fruits.The sameholds forentirepopulations. The easternFinns, for instancewhose lofty heartdiseaserates convincedAncelKeysanda generationf researcheof the evils of fat, live within500 kilometersof theArcticCircleandrarely ee freshproduceor a greenvegetable.The Scotsinfamous for eating perhaps the leastwholesomediet n thedevelopedworld, rein a similar ix. BasilRifkind ecallsbeinglaughedat onceon thispointwhenhe lec-tured to Scottish physicians on healthydiets:"One aid, You alkabout ncreasinfruitsandvegetable onsumption,ut n theareaI work n there'snot a singlegrocerstore.'" In bothcases,researchersokethatheonly green eafyvegetablehesepopulations consumeregularlys tobacco.As forthepurportedenefitsof thewidelyhailedMediterraneaniet, is it the fish, the oliveoil, or thefreshvegetables? fterall, saysj Harvard epidemiologist DimitriosTrichopoulos,a nativeof Greece,theolive oil is used either o cookvegetaf bles or as dressingover salads."Thequantity f vegetables onsumeds al-most a pound[halfa kilogram]a day,he says, "andyou cannoteat it withouoliveoil. Andwe eat a lot of legumes,andwe cannot at egumeswithout liveoil."Indeed, recent data on heart diseasetrends n Europesuggestthata likely ex-planationor the differences etween ountries and over time is the availabilityoffreshproduce ear-roundatherhandifferences in fat intake.While the pressoftenplays up the Frenchparadox-the Frenchhavelittleheartdiseasedespiteseeminglyhigh saturated at consumption-the reaparadox s throughoutSouthernEuropewhereheartdiseasedeath ateshavesteadily droppedwhile animalfat consumptiohassteadily isen,saysUniversity f Cambridge epidemiologistJohnPowles, whostudies nationaldiseasetrends.The sametrendappearsn Japan."Wehave this ideathat it's the Arcadianpast, the life in thevillage,theutopia hatwe've lost,"Powlessays; "that he really protectiveMediterranean diet is what people ate in the1950s."But that notion sn't supported ythe data:As these Mediterraneanation

30 MARCH001 VOL291 SCIENCEwww.sciencemag.org544


10/10

NEWS FOCUSbecamemore affluent,says Powles, theybegan o eatproportionally oremeatandwithit moreanimal at. Theirheartdiseaserates,however, ontinuedo improve om-pared to populations that consumed asmuch animal fat but had less access tofresh vegetables throughout he year.ToPowles, he antifatmovementwas foundedon the Puritannotionthat"something adhad to have an evil cause, andyou got aheart attackbecause you did somethingwrong, which was eating too much of abadthing,ratherhannothavingenoughofa goodthing."The other salient trade-offintheplus-minus roblem f human AI:diets is carbohydrates.When the ,federalgovernmentbegan pushing \?low-fat diets, the scientists and |:administrators,ndvirtually veryoneelse involved,hopedthatAmericanswouldreplacefat calorieswith fruitsand vegetables and legumes, but it 'didn'thappen. f nothing lse, economicsworkedagainst t. The food industryhaslittle incentive o advertisenonproprietaryitems:broccoli,for instance. nstead, aysNYU's Nestle, the greatbulkof the $30-billion-plus pent yearlyon food advertis-ing goes to selling carbohydratesn theguiseof fastfood,sodas,snacks,andcandybars.And carbohydratesre all too oftenwhatAmericansat.CarbohydratesrewhatHarvard'sWil-lett calls the flip side of the calorietrade-off problem.Because t is exceedinglydif-ficult to addpureprotein o a diet in anyquantity, low-fat diet is, by definition,ahigh-carbohydrateiet-just as a low-fatcookie or low-fatyogurtare,by definition,high in carbohydrates. umerousstudiesnow suggest thathigh-carbohydrateietscan raisetriglycerideevels, createsmall,denseLDLparticles,andreduceHDL-acombination, along with a conditionknownas "insulinresistance,"hat Stan-fordendocrinologist eraldReavenhas la-beled "syndromeX." Thirty percent ofadult males and 10% to 15% of post-menopausalwomen have this particularsyndromeX profile, which is associatedwith a several-fold ncreasein heartdis-ease risk,says Reaven,even amongthosepatientswhose LDL levels appearother-wise normal. Reaven and Ron Krauss,who studies fats and lipids at LawrenceBerkeleyNationalLaboratoryn Califor-nia, have shownthat whenmen eat high-1 carbohydrate iets their cholesterolpro-| files may shift fromnormal o syndromeI X. Inotherwords, hemorecarbohydratesy replacesaturatedats, the morelikely theI end resultwill be syndromeX andan in-- creasedheartdiseaserisk."Theproblemsu so clearrightnowit'salmostajoke,"says

Reaven.How this balancesout is the un-known. "It's a bitch of a question," aysMarcHellerstein,a nutritional iochemistat the Universityof California,Berkeley,"maybethe greatpublic healthnutritionquestionof ourera."The otherworrisomeaspectof the car-bohydraterade-off s the possibilitythat,for some individuals, at least, it mightactually be easier to gainweighton low-fat/high-carbohydrateegi-mens than on

"Whenyou don't haveany real good answersin this business, youhave to accept a few

not so good ones as thenext best thing."-Ron Krauss

higherfat diets. One of the manyfactorsthat influencehunger s the glycemic in-dex, which measureshow fast carbohy-dratesare brokendown ntosimple sugarsand moved into the bloodstream.Foodswith the highestglycemic index are sim-ple sugarsand processedgrain productslike pastaand white rice, which cause arapid rise in blood sugar after a meal.Fruits,vegetables, egumes,andeven un-processed starches-pasta al dente, forinstance-cause a much slower rise inblood sugar.Researchershave hypothe-sized that eating high-glycemic indexfoods increaseshunger aterbecause in-sulinoverreactso the spikein bloodsug-ar."Thehigh insulinlevels causethe nu-trients romthe meal to get absorbedandvery avidly stored away,and once theyare, the body can't access them," says

David Ludwig, director of the obesityclinic at Children'sHospitalBoston."Thebody appears o run out of fuel."A fewhoursaftereating,hunger eturns.If the theoryis correct,calories fromthe kind of processedcarbohydrateshathave become the staple of the Americandiet are not the sameas caloriesfromfat,protein,or complexcarbohydrateshenitcomes to controllingweight. "Theymaycause a hormonalchangethat stimulateshungerand eads to overeating,"aysLud-wig, "especially in environmentswherefood is abundant..."In 1979,2 yearsafterMcGovern'som-mittee released ts DietaryGoals,Ahrenswrote o TheLancetdescribingwhathehadlearnedover30 yearsof studyingfat and cholesterolmetabolism: It|^ : is absolutelycertain hatno onecanreliablypredictwhetherachange in dietary regimenswill haveanyeffectwhat-

-,p /~ soever on the incidence of:o"y ~new events of [coronaryheartdisease],nor nwhom."To-day,manynutrition esearchers, cknowl-edging hecomplexity f thesituation,indthemselves idingwithAhrens.Krauss, orinstance, who chairs the AHA DietaryGuidelinesCommittee, owcallsit "scien-tificallynaive"to expectthat a single di-etaryregimecan be beneficialfor every-body:"The goodness'or 'badness' f any-thingas complexas dietary at and ts sub-typeswill ultimately ependon thecontextof theindividual."Given heproven uccessand owcostofcholesterol-loweringrugs,mostphysiciansnowprescribe rug reatmentorpatients thighriskof heartdisease.Thedrugs educeLDLcholesterolevelsby as muchas 30%.Diet rarelydropsLDLby morethan10%,which is effectively rivial orhealthy ndi-viduals,althought maybe worth heeffortfor thoseathighriskof heartdiseasewhosecholesterolevelsrespondwellto it.The logic underlyingpopulationwiderecommendationsuchas the latestUSDADietaryGuideliness thatlimitingsaturat-ed fatintake-even if it does little ornoth-ing to extend helives of healthy ndividu-als and even if not all saturated ats areequallybad-might stilldelay ens of thou-sandsof deathseach yearthroughoutheentirecountry.Limiting otalfat consump-tion is consideredreasonableadvice be-cause it's simple andeasy to understand,and t may imitcalorie ntake.Whethert'sscientificallyustifiablemay simplynotberelevant."Whenyou don't have any realgood answers in this business," saysKrauss,"youhaveto accepta few not sogoodonesasthe nextbestthing."-GARYTAUBES


Science the Soft Science of Dietary Fat

Documents

Transcript of Science the Soft Science of Dietary Fat