QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Thornquist, M. D. Articles by Cheskin, L. J. Search for Related Content PubMed PubMed Citation Articles by Thornquist, M. D. Articles by Cheskin, L. J.

---

Article

Olestra Consumption Does Not Predict Serum Concentrations of Carotenoids and Fat-Soluble Vitamins in Free-Living Humans: Early Results from the Sentinel Site of the Olestra Post-Marketing Surveillance Study^{1 ,2}

Mark D. Thornquist³, Alan R. Kristal, Ruth E. Patterson, Marian L. Neuhouser, Cheryl L. Rock^*, Dianne Neumark-Sztainer and Lawrence J. Cheskin^**

Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA 98109; ^* Department of Family and Preventive Medicine, University of California, San Diego, La Jolla, CA 92093; Division of Epidemiology, School of Public Health, University of Minnesota, Minneapolis, MN 55454; and ^** Division of Gastroenterology, Johns Hopkins University School of Medicine, Baltimore, MD 21224

³To whom correspondence should be addressed.

	ABSTRACT

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
In 1996, the U.S. Food and Drug Administration approved olestra, a fat substitute, for use in snack foods. Previous studies had shown that olestra consumption could reduce absorption of carotenoids and fat-soluble vitamins. To determine the association between consumption of olestra-containing snack foods and serum concentrations of carotenoids and fat-soluble vitamins in a free-living population, we interviewed independent population-based cross-sectional samples of 1043 adults before olestra was available and 933 adults 9 mo after olestra snacks were introduced into the marketplace in Marion County, IN, the first major test market for olestra. A cohort composed of 403 adults from the first survey, oversampling those most frequently reporting olestra consumption during follow-up telephone interviews, completed a second survey. We assessed diet, lifestyle factors and olestra consumption, and collected blood for assays for the serum concentrations of six carotenoids, four fat-soluble vitamins and lipids. Nine months after the introduction of olestra into the marketplace, 15.5% of Marion County residents reported consuming an olestra-containing snack in the previous month, with a median frequency among consumers of 3.0 times per month. There were no significant associations or consistent trends for decreased serum carotenoids or fat-soluble vitamins associated with olestra consumption, although cohort members consuming 2 g/d of olestra had adjusted total serum carotenoids 15% lower compared with baseline. There were increases in serum vitamin K concentrations associated with olestra consumption (P = 0.03 in the cross section and P = 0.06 in the cohort). In summary, there was no statistically significant evidence in this free-living population of associations between olestra consumption and decreased serum concentrations of carotenoids and fat-soluble vitamins.

KEY WORDS: • olestra • carotenoids • fat-soluble vitamins • serum concentrations • humans

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
In 1996, the U.S. Food and Drug Administration (FDA)⁴ approved a new fat substitute, olestra (Olean®), for use in "savory snacks" (potato, tortilla and corn chips; crackers; and extruded snacks) (Federal Register 1996 ). Olestra has the taste and mouth feel of regular fat, but because it cannot be absorbed, it provides no energy (Daher et al. 1997 ). Despite its potential public health benefits as a nonenergy fat substitute, there is controversy concerning the effects of olestra (Blume 1995 ). Experimental studies in both humans and animals find that olestra can sequester fat-soluble nutrients in the gut and thereby reduce their absorption (Westrate and van het Hof 1995 ). The FDA therefore requires the addition of vitamins A, D, E and K to foods made with olestra (Federal Register 1996 ). Addition of carotenoids was not required because health effects of carotenoids, beyond their provitamin A activity, are not well understood (Rock 1997 ).

In approving olestra, the FDA made assumptions about the level of its consumption and its effects on serum micronutrient concentrations. As a condition for approval, the FDA mandated a program of active postmarketing surveillance to assess the accuracy of these assumptions. The Olestra Post-Marketing Surveillance Study (OPMSS) was designed to investigate whether consumption of olestra-containing foods, as part of self-selected mixed diets, will affect nutritional status (Kristal et al. 1998 ). The specific aim was to examine the association of olestra consumption with serum concentrations of six carotenoids (- and ß-carotene, lycopene, lutein, zeaxanthin, and ß-cryptoxanthin) and four fat-soluble vitamins [retinol (vitamin A), 25-hydroxyvitamin D, -tocopherol (vitamin E) and vitamin K]. The full study, with baseline data and 0 y of follow-up at four sites, will conclude data collection in October 2000.

At a public meeting in June 1998, the FDA reviewed the information about the potential health effects of olestra that had accumulated since the FDA’s approval of olestra in December 1996. This review included the early results from the OPMSS arising from the first major test market for olestra-containing snacks in Marion County (Indianapolis), IN. This paper presents the OPMSS data considered by the FDA during the deliberations leading to its continued approval of olestra. These data represent the first findings of the associations between olestra consumption and serum levels of micronutrients in a free-living population.

	SUBJECTS AND METHODS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Design.

The OPMSS has the following three principal aims: 1) to monitor the adoption and patterns of use of olestra-containing food products in representative samples of the U.S. population; 2) to assess the associations between introduction of olestra-containing foods into the marketplace and serum concentrations of carotenoids and fat-soluble vitamins in representative samples of the U.S. population; and 3) to assess the associations between consumption of olestra-containing foods and serum concentrations of carotenoids and fat-soluble vitamins in a cohort of olestra consumers. The study was designed to have at least 80% power to test the following two null hypotheses: 1) changes in population-level total serum carotenoid concentrations that are attributable to olestra will be <10%, and 2) changes in serum concentrations of total carotenoids attributable to olestra, among olestra consumers, will be <10%.

Details of the design of the OPMSS, dietary assessment methods and baseline descriptive results have been reported previously (Kristal et al. 1998 ). Briefly, the OPMSS is a series of three related studies, each designed to address one of the three principal aims given above. The first study, called the population cross section, uses annual, independent, random-digit dial (RDD) telephone surveys to collect population-level data on the prevalence and patterns of olestra consumption. The second study, called the clinic cross section, collects serum and extensive dietary and health-behavior information from volunteers drawn from each RDD survey. The third study, called the cohort, collects serum samples and dietary information annually from a subsample of the baseline clinic cross-section participants, chosen to target olestra consumers. In Marion County, the baseline cross-sectional survey was conducted from September 1996 to January 1997; marketing of olestra began February 1997, and the first follow-up survey and follow-up cohort visits were conducted from September 1997 to January 1998. The study populations include children aged 7–17 y. However, because of the small number of children in this preliminary analysis, we report data only for adults 18 y old.

The OPMSS is funded by The Procter & Gamble Company (Cincinnati, OH), the manufacturer of olestra. By contractual agreements, scientists at the coordinating center and the field sites are solely and independently responsible for data management, analysis and publication. All study activities were approved by the institutional review boards, and participants provided oral consent for the telephone surveys and written consent for clinic visit activities. To avoid biasing participants’ reports of olestra consumption, we did not tell them that the primary focus of the study was olestra.

Participant selection.

Participants in the two independent population cross sections were selected and interviewed using a list-assisted RDD sampling methodology, which screens out nonworking numbers but does not exclude nonresidential numbers. One adult (18 y old) was selected at random from each household. Of the 9170 numbers called in y 0 (baseline, before the marketing of olestra), interviews were completed on 2173, with an efficacy rate (completed interviews divided by the number of known eligibles plus estimated eligibles from the residential numbers on the list) of 64.2%. Of the 7300 calls made in y 1 (after introduction of olestra in the marketplace), interviews were completed on 1538, with an efficacy rate of 60.9%. These rates are similar to or better than those found in other RDD surveys (Kristal et al. 1993 ).

Clinic cross-section participants were a random sample of persons who completed the telephone survey. We excluded persons with medical conditions such as hyperparathyroidism or short bowel syndrome that would confound interpretation of serum measures. In y 0, of those we invited, 72% agreed to participate and 58% completed the clinic visit. In y 1, 75% agreed to participate and 63% completed the visit. Participants received $100 as compensation for travel and their time.

Cohort participants were selected from the baseline clinic cross section for higher level of olestra consumption. On the basis of dietary information collected from telephone interviews that were conducted ~3, 6 and 9 mo after the clinic visit and that assessed diet in the previous month, we recruited all adults who reported at least two occasions of eating olestra-containing foods, a random sample of 67% of those adults who reported a single eating occasion and a sample of 16% of those adults who did not report eating olestra. Of the 478 adults invited into the cohort, 88% completed the y-1 follow-up visit.

Measures.

Nutrient intakes, with the exception of vitamin K, were measured with the use of a food-frequency questionnaire (FFQ) (Patterson et al. 1999 ) with a supplementary questionnaire on savory snack consumption (Neuhouser et al. 2000 ). The reference period for dietary assessment was "in the past month," because serum carotenoid concentrations are sensitive to dietary changes in recent weeks (Rock et al. 1992 ). The nutrient database for these questionnaires was derived from the University of Minnesota Nutrition Coordinating Center nutrient database (Minneapolis, MN) (Schakel et al. 1998 ) and included the USDA-National Cancer Institute carotenoid database for foods (Mangels et al. 1993 ). Because serum vitamin K concentrations are most strongly influenced by recent diet (Shearer et al. 1974 ), vitamin K intake was estimated from a 24-h dietary recall collected at each clinic visit using the Nutrition Data System version 2.9 (Nutrition Coordinating Center, University of Minnesota) (Feskanich et al. 1989 ), with supplementary vitamin K food composition data provided by Tufts University (Boston, MA) (Booth et al. 1995 ). Data on vitamin supplement use over the past month were obtained using a validated inventory procedure (Patterson et al. 2000 ), which was modified to collect precise dose information on vitamins A, D, E and K and ß-carotene (the only carotenoid available in supplements at the time). Interviewers collected extensive information on demographic characteristics, health-related behavior and medical history. Clinic staff collected anthropometric measures (weight, height and waist circumference) using a standardized protocol (ARIC Investigators 1989 , Friedman et al. 1988 ), and body mass index (BMI) was calculated as weight (kg)/height squared (m²).

Blood collection and processing.

Details on blood collection, processing and analysis are given elsewhere (Rock et al. 1999 ). In brief, serum was protected from light, stored at -20°C for no longer than 4 d, shipped to the study’s coordinating center on dry ice and then stored at -70°C until analysis. Sera from y 0 and 1 were analyzed at different times. Carotenoids (- and ß-carotene, lycopene, lutein, zeaxanthin and ß-cryptoxanthin), retinol and -tocopherol were analyzed using HPLC methods. Interassay CV for individual analytes ranged from 1.9 to 9.8%. Minimal detectable concentrations were as follows (µmol/L): -tocopherol, 1.163; -carotene, 0.005; ß-carotene, 0.011; lycopene, 0.011; lutein, 0.004; zeaxanthin, 0.01; and ß-cryptoxanthin, 0.011. 25-Hydroxyvitamin D was analyzed using the INCSTAR (Stillwater, MN) ¹²⁵I RIA kit. The interassay CV ranged between 5.7 and 9.2%. Serum vitamin K (phylloquinone) was analyzed by HPLC with vitamin K₁₍₂₅₎ as an internal standard using the method of Davidson and Sadowski (1997) . One low normal and one high normal concentration quality control sample was analyzed with each batch of samples with between-day variability of 9.7 and 5.3%, respectively. Total serum cholesterol and triacylglycerols were analyzed using enzymatic methods (Rock et al. 1999 ). Precision was evaluated using packaged reagents, pooled human serum and control sera; both interassay precision and bias were 3%.

Statistical analysis.

We assigned individual weights to each observation in the clinic cross-sectional surveys, poststratified on age and sex, such that statistics calculated from these samples were representative of the Marion County adult population. We excluded from all analyses participants for whom sex (<1%) or age (1%) was unavailable. Because pregnancy has profound effects on serum lipid and nutrient concentrations (Lockitch 1997 ), we excluded from analyses women who were pregnant (<1%). After these exclusions, there were 1042 y-0 cross-section, 932 y-1 cross-section and 398 cohort participants available for our analyses. For participants whose FFQ or BMI was missing or unreliable, we completed the data set by assigning to them the mean of the variable among all participants with valid values and included dummy variables to indicate that values had been imputed. Unreliable values were defined as computed energy intake <3347 kJ/d or >20,920 kJ/d for men (11%) or <2510 kJ/d or >16,736 kJ/d for women (9%), and BMI <15 or >60 kg/m² (3%). For a small number of participants who had serum concentrations of carotenoids or vitamins that were undetectable by laboratory methods (4% undetectable for vitamin D, 3% for vitamin K, 7% for -carotene, <1% for all other analytes), we replaced the missing values with the midpoint between zero and the laboratory’s minimum detectable value. Adults not fasting at both y 0 and 1 were excluded from the clinic cohort analyses; we used dummy variables to indicate nonfasting status at either y 0 or 1. Sensitivity analyses found that our conclusions were not changed by these analytic decisions.

We categorized olestra intake into the following four levels on the basis of daily intake (g) in the previous month: no intake (0 g/d), low intake (>0 to <0.4 g/d, approximately the median intake among users in the y-1 clinic cross section), moderate intake (0.4 to <2.0 g/d, from the median up to the 90th percentile of intake among users), and heavy intake (2.0 g/d, the 90th percentile of intake and higher). For regression modeling, we transformed serum and dietary distributions (except percentage of energy from fat, alcohol consumption and servings per day of fruits and vegetables) by the natural logarithm.

For the clinic cross-sectional analyses, we first used data from the y-0 clinic visit to develop models predicting serum nutrient concentrations, weighted to be representative of the Marion County population. All models included age, sex, race, total energy intake (except for the model for vitamin K) and dietary plus supplemental intake of the nutrient. Additional variables, such as serum cholesterol concentration, were added if they were significant at the 0.05 level. The variables considered for the models have been reported elsewhere (Rock et al. 1999 ). We then modeled serum nutrient concentrations at y 1 using the control variables identified from the y-0 data and added the olestra intake variable. This modeling strategy ensured that associations between olestra intake and other control variables would not confound the y-1 analysis.

For the cohort analyses, we developed models to predict changes in serum nutrient concentrations without regard to olestra consumption. All models included age, sex, race, y-0 serum concentration, y-0 and change in total intake of the nutrient, and y-0 and change in energy intake, with additional variables included if they were significant at the 0.05 level. If a change variable (such as change in fat consumption) was included in the final model, we also included the baseline value regardless of statistical significance. Once the most parsimonious model was selected, we added olestra to the model to assess its association with the change in serum concentration. Details on the most parsimonious models found for the cross section and the cohort are available upon request.

In comparisons of the distributions of demographic characteristics, we used t tests for continuous measures and Pearson’s ² test for unordered categorical measures. Geometric mean serum concentrations and 95% confidence limits (weighted to be representative of the Marion County population in the cross section, unweighted in the cohort) by level of olestra consumption were computed from the least-squares means for the log-transformed serum concentrations, back-transformed to the original units. Tests for olestra effects in the regression models used Type III sums of squares. The P-values reported are not adjusted for multiple tests. All analyses were performed using SAS version 6.12 (SAS Institute 1989 ).

Power.

When completed, the national data from the OPMSS clinic cross section is projected to have 80% power to detect a 8.0% difference in serum total carotenoid concentration between olestra consumers and nonconsumers, assuming that 25% of participants report olestra consumption in the previous month. The cohort should be able to detect a 5% difference, assuming 80% reporting consumption. Under the same assumptions, the y-1 data from Marion County were projected to be able to detect differences of 13% in the clinic cross-sectional analyses and 20% in the cohort. However, the rate of olestra intake and the variance in serum measures differed from our assumptions. Post-hoc, the detectable differences actually achieved were slightly better (8.7% in the clinic cross section and 11% in the cohort) than the original projections.

	RESULTS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Demographic characteristics of clinic cross-section participants were similar in y 0 and 1 (Table 1 ). We recruited relatively high percentages of minorities (23–25%) and less well-educated people (40–42% with no more than high school education), groups often underrepresented in research studies. Cohort participants (Table 2 ), who were selected to include the highest consumers of olestra-containing foods, were more likely to be Caucasian and overweight than the y-0 clinic cross section from which they were selected. The composition of the cohort is consistent with our previous report that Caucasian race, higher education, being overweight, lower fat intake and concerns about diet and health were predictors of early adoption of olestra-containing foods (Neumark-Sztainer et al. 2000 ).

Table 3 gives the mean serum carotenoid and fat-soluble vitamin concentrations for the clinic cross section at y 1, for each category of olestra consumption. These results are adjusted for covariates and weighted to be representative of the Marion County population. There was no significant variation across levels of olestra consumption in the serum concentration of total carotenoids (P = 0.83 for heterogeneity, P = 0.88 for trend), any individual carotenoid or vitamins A, D or -tocopherol. There was a significant trend for increased serum vitamin K concentration with increasing consumption of olestra, with an estimated 10.5% increase in serum concentration for each category of increasing olestra consumption. The test for heterogeneity for serum -tocopherol had a P-value of 0.08 due to an anomalously high mean concentration among the highest consumers of olestra. There was, however, no consistent or significant trend for increasing serum -tocopherol with increased olestra consumption.

	DISCUSSION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

Although the intakes of olestra in this study may appear to be low, they are consistent with the premarket estimates of olestra intake (Webb et al. 1997 ). The highest level of intake is similar to that used in a controlled olestra feeding study that found effects of olestra on serum carotenoids. A previous controlled trial found that a daily dose of 3 g of sucrose polyester eaten as margarine with the main meal of the day reduced serum concentrations of ß-carotene and lycopene by 20 and 38%, respectively (Westrate and van het Hof 1995 ). In the OPMSS clinic cross section at y 1, participants in the highest category of olestra consumption (2 g/d) had reductions in serum concentrations of 11% for ß-carotene and 8% for lycopene, compared with nonconsumers. In the cohort, reductions between baseline and follow-up in the highest category of olestra consumption were 21% for ß-carotene (compared with 7% for nonconsumers) and 7% for lycopene (compared with no change for nonconsumers). In contrast to the Westrate and van het Hof data, none of the effects of olestra in this analysis was significant and there were no ordered, dose-response relationships between olestra and serum carotenoids in either sample. For example, adults in the cohort with moderate consumption levels of olestra (0.4–2 g/d) had increased serum concentrations for all carotenoids except lycopene. Therefore, we interpret the results for analyses of olestra consumption and serum carotenoid concentrations as being consistent with random variability. The data are also consistent with lower serum carotenoid concentrations among the small subgroup of individuals (~3% of Marion County residents in Fall 1997) consuming 2 g/d of olestra, but not with a linear trend for decreasing serum carotenoid concentrations with increasing olestra consumption. Data from pigs (Cooper et al. 1997a ) and humans (Schlagheck et al. 1997 ) do not suggest a threshold for effect of olestra; in fact, the greatest incremental effects are seen at the lowest consumption levels studied.

There are a number of reasons why results from an observational study of olestra consumption would be expected to differ from feeding studies. First, olestra intake is estimated from self-report and necessarily contains error. Second, from a biological perspective, there are many determinants of serum carotenoids in free-living persons eating a varied diet. For example, fiber reduces carotenoid absorption, whereas fat facilitates absorption (Rock 1997 ). This type of variation in nutrient absorption is a well-known aspect of a mixed diet and generally does not pose health risks. People who choose to consume olestra are likely to make different dietary choices than those who do not choose to consume olestra, and these differences can reduce the observed association between olestra consumption and serum levels. In feeding studies in which participants are unaware of whether they are consuming olestra, dietary choices are unlikely to confound the association. Third, from a population-level perspective, the effect of olestra on carotenoids will be determined by the following factors, among others: 1) the fraction of the population consuming olestra; 2) the amount of olestra generally consumed; 3) the percentage of total carotenoids consumed with olestra-containing foods (co-consumption); and 4) the degree to which co-consumption decreases carotenoid absorption. For an example of the effect of co-consumption, in a feeding study in which olestra was provided with every meal, 20 g/d olestra reduced serum ß-carotene by 50% (Schlagheck et al. 1997 ), whereas in another study in which the participants were able to choose when they consumed olestra, 18 g/d olestra reduced serum ß-carotene by 27% (Koonsvitsky et al. 1997 ). In a separate analysis of 489 24-h dietary recalls from study participants, we found that co-consumption of high carotenoid foods with savory snacks on any single day was low, i.e., on average, only 13–15% of total dietary carotenoids were consumed within 2 h of eating any type of savory snack; co-consumption with olestra was too small to be estimated reliably. Thus, it is not surprising that the effects of olestra consumption on serum micronutrient and carotenoid concentrations in free-living persons will be less than those in controlled studies.

The finding of increased serum vitamin K concentration with increased olestra consumption may be causal. The associations between olestra consumption and increased serum vitamin K were consistent in both the cross-section and cohort samples, and could be due to the level of vitamin K added to olestra. The FDA requires the addition of 8 µg vitamin K/g olestra (Federal Register 1996 ). At the time of approval, the manufacturer estimated, on the basis of animal and human studies, that the amount of vitamin K necessary to offset any olestra effect on absorption was 3.3 µg/g (Peters et al. 1997 ). Thus, those consuming 2.0 g/d olestra are consuming 9.4 µg/d vitamin K beyond that estimated to account for an olestra effect. This is 13% of the mean dietary intake of vitamin K in the y-0 cross section (70.1 µg/d). In its approval of olestra, the FDA considered the possibility that the mandated level of vitamin K supplementation would be greater than that required to correct for absorption effects of olestra and concluded that any excess did not pose a health risk (Federal Register 1996 ).

The lack of an association between olestra consumption and serum concentrations of vitamins A, D and E suggests that the level of addition to olestra by these vitamins is sufficient to offset any olestra effects on absorption. The reductions in serum 25-hydroxyvitamin D concentration in y 1 compared with y 0, regardless of olestra intake, is likely due to decreased exposure to solar radiation, a major source of vitamin D in humans (Need et al. 1993 ). Data provided by the Midwestern Climate Center (Champaign, IL) show that the mean solar radiation in Indianapolis, the primary city in Marion County, was 15.7 MJ/m² June 1996–January 1997 and 10.1 MJ/m² June 1997–January 1998, a 36% decrease.

The major limitation of this study is that dietary data, including consumption of olestra-containing savory snacks, are self-reported and therefore likely to contain significant measurement error (Willett 1998 ). The FFQ used in this study is the same questionnaire developed for the Women’s Health Initiative, a multicenter clinical trial and observational study of ~164,500 women. In a validity study comparing nutrient estimates from this FFQ to 8 d of food records and recalls, correlation coefficients for some of the nutrients examined in this study were as follows: 0.62 for percentage of energy from fat, 0.57 for ß-carotene, 0.56 for retinol, 0.73 for vitamin D and 0.83 for vitamin E (Patterson et al. 1999 ). The reliability of this instrument was good, with intraclass correlation coefficients ranging from 0.67 to 0.79 for these same nutrients. In a separate analysis, we compared fruit and vegetable consumption measured by this FFQ with serum concentrations of carotenoids in this sample. The Pearson correlation of fruit and vegetable intake with total carotenoids was 0.23, which is similar to estimates found in similar studies (Tucker et al. 1999 ).

This study has several strengths. Our primary end point (serum concentration of carotenoids) is a sensitive indicator of the effect of olestra consumption on nutrient absorption because olestra exclusively affects the absorption of lipophilic substances (Cooper et al. 1997b ), and carotenoids are considerably more lipophilic than fat-soluble vitamins or other potentially beneficial food components, such as phenols or folate (Lawson et al. 1997 ). Therefore, serum carotenoids can serve as a type of "worst-case scenario" marker for any compounds that could be sequestered by olestra. Another strength is the study design. A RDD survey combined with community-based, cross-sectional samples and a cohort allows for considerable confidence in our findings. In addition, we successfully recruited a heterogeneous population, including substantial numbers of minorities and those of low socioeconomic status, persons typically difficult to enroll in research studies. However, this first analysis is limited in its ability to assess the effect of olestra consumption among frequent consumers because only 3–5% of individuals report an average consumption 2 g/d (equivalent to one 2-oz serving every 8 d). Further follow-up of the full OPMSS study population will be important to clarify the effects in the heaviest consumers.

In summary, this first analysis found no significant or otherwise compelling evidence of a marked effect of olestra consumption on serum concentrations of carotenoids and fat-soluble vitamins in free-living individuals. However, patterns of consumption of olestra may change over time, particularly as new food products containing olestra become available. Continued monitoring of olestra consumption and its effects on nutritional status is warranted. At the conclusion of the Olestra Post Marketing Surveillance Study in October 2000, we will have analyzed serum from ~4800 individuals before olestra availability and 2500 individuals in three cross-sectional surveys after the introduction of olestra to the market. In addition, we will have followed a cohort of over 2000 individuals from baseline for up to 2.5 y after the introduction of olestra. The total sample size gives the OPMSS good power to detect even small changes in serum concentrations of carotenoids and fat-soluble vitamins attributable to the consumption of olestra-containing foods.

	ACKNOWLEDGMENTS

 
The authors gratefully acknowledge Martha Berlin and her staff at Westat, Incorporated, for providing input to the design of the OPMSS, designing and conducting the RDD surveys and operating the OPMSS study site in Marion County, IN.

	FOOTNOTES

 
¹ Presented at a meeting of the Food Advisory Committee, U.S. Food and Drug Administration, June 16, 1998, Reston, VA; and in part at the 12th International Carotenoid Symposium, July 18–23, 1999, Cairns, Australia [Thornquist, M. D., Kristal, A. R., Patterson, R. E., Rock, C. L., Neumark-Sztainer, D., Cheskin, L. J. & Neuhouser, M. L. (1999) Associations between consumption of olestra and serum concentrations of carotenoids and fat-soluble vitamins].

² Funded by The Procter & Gamble Company, Cincinnati, OH.

⁴ Abbreviations used: BMI, body mass index; FDA, Food and Drug Administration; FFQ, food-frequency questionnaire; OPMSS, Olestra Post-Marketing Surveillance Study; RDD, random-digit dialing.

Manuscript received December 17, 1999. Initial review completed January 24, 2000. Revision accepted March 9, 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. The ARIC Investigators The atherosclerosis risk in communities study: design and objectives. Am. J. Epidemiol. 1989;129:687-702[Abstract/Free Full Text]

2. Blume E. The trials of olestra: is a fat-free lunch too costly?. J. Natl. Cancer Inst. 1995;87:1741-1743[Free Full Text]

3. Booth S., Sadowski J. A., Pennington J.A.T. The phylloquinone (vitamin K₁) content of foods in the US-FDA Total Diet Study. J. Agric. Food Chem. 1995;43:1574-1579

4. Cooper D. A., Berry D. A., Spendel V. A., King D., Kiorpes A. L., Peters J. C. Olestra dose response on fat-soluble and water-soluble nutrients in the pig. J. Nutr. 1997a;127:1573S-1588S

5. Cooper D. A., Webb D. R., Peters J. C. Evaluation of the potential for olestra to affect the availability of dietary phytochemicals. J. Nutr. 1997b;127:1699S-1709S

6. Daher G. C., Cooper D. A., Peters J. C. Physical or temporal separation of olestra and vitamins A, E and D intake decreases the effect of olestra on the status of the vitamins in the pig. J. Nutr. 1997;127:1566S-1572S

7. Davidson K. W., Sadowski J. A. Determination of vitamin K compounds in plasma or serum by high performance liquid chromatography (HPLC) using post-column chemical reduction and fluorimetric detection. Methods Enzymol 1997;282:408-421[Medline]

8. Federal Register, Part III, U.S. Department of Health and Human Services, Food and Drug Administration, 21 CFR part 172. Food additives permitted for direct addition to food for human consumption: olestra; final rule. January 30 1996;62:3118-3173

9. Feskanich D., Sielaff B. H., Chong K., Buzzard I. M. Computerized collection and analysis of dietary intake information. Comput. Methods Programs Biomed. 1989;30:47-57[Medline]

10. Friedman G. D., Cutter G. R., Donahue R. P., Hughes G. H., Hulley S. B., Jacobs D. R., Jr, Liu K., Savage P. J. CARDIA: study design, recruitment, and some characteristics of examined subjects. J. Clin. Epidemiol. 1988;41:1105-1116[Medline]

11. Koonsvitsky B. P., Berry D. A., Jones M. B., Lin P. Y., Cooper D. A., Jones D. Y., Jackson J. E. Olestra affects serum concentrations of -tocopherol and carotenoids but not vitamin D or vitamin K status in free-living subjects. J. Nutr. 1997;127:1636S-1645S

12. Kristal A. R., Patterson R. E., Neuhouser M. L., Thornquist M., Neumark-Sztainer D., Rock C., Berlin M., Cheskin L., Schreiner P. The Olestra Post-Marketing Surveillance Study: design and baseline results from the sentinel site. J. Am. Diet. Assoc. 1998;98:1290-1296[Medline]

13. Kristal A. R., White E., Davis J. R., Corycell G., Raghunathan T., Kinne S., Lin T.-K. The effects of enhanced calling efforts in random digit dial surveys on response rates, population-level estimates of health behavior, and costs. Public Health Rep 1993;108:372-380[Medline]

14. Lawson K. D., Middleton S. J., Hassall C. D. Olestra, a nonabsorbed, noncaloric replacement for dietary fat: a review. Drug Metab. Rev 1997;29:651-703[Medline]

15. Lockitch G. Clinical biochemistry of pregnancy. Crit. Rev. Clin. Lab. Sci. 1997;34:67-139[Medline]

16. Mangels A. R., Holden J. M., Beecher G. R., Forman M. R., Lanza E. Carotenoid content of fruits and vegetables: an evaluation of analytic data. J. Am. Diet. Assoc. 1993;93:284-296[Medline]

17. Need A. G., Morris H. A., Horowitz M., Nordin B.E.C. Effects of skin thicknes, age, body fat, and sunlight on serum 25-hydroxyvitamin D. Am. J. Clin. Nutr. 1993;58:882-885[Abstract/Free Full Text]

18. Neuhouser, M. L., Patterson, R. E., Kristal, A. R., Rock, C. L., Neumark-Sztainer, D., Thornquist, M. D. & Cheskin, L. J. (2000) Do consumers of savory snacks have poor quality diets? J. Am. Diet. Assoc. (in press).

19. Neumark-Sztainer D., Kristal A. R., Thornquist M. D., Patterson R. E., Neuhouser M. L., Barnett M. J., Rock C. L., Cheskin L. J., Schreiner P., Miller D. L. Early adopters of olestra-containing foods: who are they?. J. Am. Diet. Assoc. 2000;100:198-204[Medline]

20. Patterson R. E., Kristal A. R., Carter R. A., Fels-Tinker L., Bolton M. P., Agurs-Collins T. Measurement characteristics of the Women’s Health Initiative food frequency questionnaire. Ann. Epidemiol. 1999;9:178-187[Medline]

21. Patterson R. E., Levy L., Fels-Tinker L., Kristal A. R. Validity of a simplified vitamin supplement inventory procedure developed for the Women’s Health Initiative. Public Health Nutr. 2000;2:273-276

22. Peters J. C., Lawson K. D., Middleton S. J., Triebwasser K. C. Assessment of the nutritional effects of olestra, a nonbasorbed fat replacement: summary. J. Nutr 1997;127:1719S-1728S

23. Rock C. L. Carotenoids: biology and treatment. Pharmacol. Ther. 1997;75:185-197[Medline]

24. Rock C. L., Swendseid M. E., Jacob R. A., McKee R. W. Plasma carotenoid levels in human subjects fed a low carotenoid diet. J. Nutr. 1992;122:96-100

25. Rock C. L., Thornquist M. D., Kristal A. R., Patterson R. E., Cooper D. A., Neuhouser M. L., Neumark-Sztainer D., Cheskin L. J. Demographic, dietary and lifestyle factors differentially explain variability in serum carotenoids and fat-soluble vitamins: baseline results from the sentinel site of the Olestra Post-Marketing Surveillance Study. J. Nutr 1999;129:855-864[Abstract/Free Full Text]

26. SAS Institute Inc. SAS Language and Procedures 6.12 ed. 1989 SAS Institute Cary, NC.

27. Schakel S. F., Sievert Y. A., Buzzard I. M. Sources of data for developing and maintaining a nutrient database. J. Am. Diet. Assoc. 1998;88:1268-1271

28. Schlagheck T. G., Riccardi K. A., Zorich N. L., Torri S. A., Dugan L. D., Peters J. C. Olestra dose response on fat-soluble and water-soluble nutrients in humans. J. Nutr. 1997;127:1646S-1665S

29. Shearer M. J., McBurney A., Barkhan P. Studies on the absorption and metabolism of phylloquinone (vitamin K₁) in man. Harris R. S. Diczfalusy E. Munson P. L. Glover J. eds. Vitamins and Hormones Advances in Research and Applications 1974:513-542 Academic Press New York, NY.

30. Tucker K. L., Chen H., Vogel S., Wilson P.W.F., Schaefer E. J., Lammi-Keefe C. J. Carotenoid intakes, assessed by dietary questionnaire, are associated with plasma carotenoid concentrations in an elderly population. J. Nutr. 1999;129:438-445[Abstract/Free Full Text]

31. Webb D. R., Harrison G. G., Lee M.-J., Huang M.-H. Estimated consumption and eating frequency of olestra from savory snacks using menu census data. J. Nutr. 1997;127:1547S-1554S

32. Westrate J. A., van het Hof K. H. Sucrose polyester and plasma carotenoid concentrations in healthy subjects. Am. J. Clin. Nutr. 1995;62:591-597[Abstract/Free Full Text]

33. Willett W. Nutritional Epidemiology 2nd ed. 1998 Oxford University Press New York, NY.

This article has been cited by other articles:

				M. L Neuhouser, C. L Rock, A. R Kristal, R. E Patterson, D. Neumark-Sztainer, L. J Cheskin, and M. D Thornquist Olestra is associated with slight reductions in serum carotenoids but does not markedly influence serum fat-soluble vitamin concentrations Am. J. Clinical Nutrition, March 1, 2006; 83(3): 624 - 631. [Abstract] [Full Text] [PDF]

				W. M. R. Broekmans, I. A. A. Klopping-Ketelaars, J. A. Weststrate, L. B. M. Tijburg, G. van Poppel, A. A. Vink, T. T.J.M Berendschot, M. L. Bots, W. A. M. Castenmiller, and A. F. M. Kardinaal Decreased Carotenoid Concentrations Due to Dietary Sucrose Polyesters Do Not Affect Possible Markers of Disease Risk in Humans J. Nutr., March 1, 2003; 133(3): 720 - 726. [Abstract] [Full Text] [PDF]

				J. Wylie-Rosett Fat Substitutes and Health: An Advisory From the Nutrition Committee of the American Heart Association Circulation, June 11, 2002; 105(23): 2800 - 2804. [Full Text] [PDF]

				R. E. Patterson, A. R. Kristal, J. C. Peters, M. L. Neuhouser, C. L. Rock, L. J. Cheskin, D. Neumark-Sztainer, and M. D. Thornquist Changes in Diet, Weight, and Serum Lipid Levels Associated With Olestra Consumption Arch Intern Med, September 25, 2000; 160(17): 2600 - 2604. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Thornquist, M. D. Articles by Cheskin, L. J. Search for Related Content PubMed PubMed Citation Articles by Thornquist, M. D. Articles by Cheskin, L. J.