Use of Nonprescription Weight Loss Products:  Results From a Multistate Survey FREE

Successful weight loss and healthy weight management depend on long-term lifestyle changes such as reducing calorie consumption and increasing physical activity. However, because these changes are difficult, easily obtained nonprescription weight loss products and prescription diet pills are an appealing alternative to the increasingly overweight US population. It has been speculated that individuals may use nonprescription products and prescription pills in place of lifestyle changes.¹ No population-based studies have examined the relationship between use of overall nonprescription weight loss products and use of prescription weight loss pills or lifestyle changes for weight loss. Usage patterns of specific nonprescription products (eg, phenylpropanolamine [PPA] and ephedra) are also of particular interest because of safety concerns.

Ephedra products have stimulant properties and are purported to decrease weight when used in combination with caffeine through thermogenesis and reduced appetite.^2- 4 In June 1997, the Food and Drug Administration (FDA) proposed restrictions on dietary supplements containing ephedrine alkaloids.⁵ However, this proposal was withdrawn in April 2000 after the General Accounting Office concluded that additional evidence was needed to support these restrictions.⁶ Although the FDA withdrew certain provisions of the ephedrine alkaloids proposal, the agency remains concerned and is continuing to passively monitor adverse events associated with the use of these products.⁶ Because of potential adverse health effects among persons with diabetes, hypertension, heart disease, and other conditions, the FDA has recommended a labeling statement that instructs ephedra users to seek the advice of a health care provider before use.⁵

Phenylpropanolamine, the main ingredient in the over-the-counter (OTC) weight loss aids Dexatrim and Acutrim, is a synthetic ephedrine alkaloid with stimulant properties that may reduce appetite.⁷ Until recently, PPA was considered to be a safe short-term weight reduction agent⁸; however, case reports of adverse cerebrovascular and cardiac events^9- 11 and a study in which PPA increased the risk of stroke¹² resulted in the voluntary withdrawal of all OTC PPA products from the market in November 2000.¹³

To assess who uses nonprescription weight loss products in the United States, 5 states incorporated questions that asked about overall and specific nonprescription weight loss product use during the previous 2 years into their 1998 Behavioral Risk Factor Surveillance System (BRFSS) surveys. We used these data to examine the prevalence of overall and specific nonprescription product use by demographic characteristics, lifestyle choices, prescription pill use, and presence of diabetes.

The data come from adults 18 years or older who participated in the 1998 BRFSS in Florida, Iowa, Michigan, West Virginia, and Wisconsin. The BRFSS is a random-digit telephone survey, conducted by state health departments, that assesses individual health practices. (For a detailed description of the survey methods and quality control indexes, see Nelson et al.¹⁴) The average cooperation rate (completed interviews/refusals + terminations + completed interviews) for the 5 states was 67.5% (range, 45.4%-84.0%).

Respondents were first prompted by the following statement: "Now we would like you to tell us about any over-the-counter products such as pills, powders, or liquids, you have taken to lose weight. That is, products you do not need a prescription to purchase." Respondents were then asked, "In the past 2 years, have you taken any over-the-counter weight loss products?" If respondents replied positively, they were then asked, "Have you taken any of the following over-the-counter weight loss products in the past 2 years. Herbal fen-phen (also known as natural fen-phen, or fen-fuel)? Acutrim or Dexatrim? Ma-huang? St. John's wort? Ephedra? Or other?" Respondents were prompted for each category and responses recorded as "yes," "no," "don't know," or "refused." A positive response to herbal fen-phen, ma-huang, or ephedra was used to classify an individual as an "ephedra product" user in our analyses. One state (Michigan) asked the respondent to specify the product name or type when an "other" nonprescription product was taken.

Respondents were asked to report their current height and weight without shoes. Each respondent's body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. (The BMI was categorized as <25, normal weight; 25-29.9, overweight; ≥30, obesity.) Respondents were asked for information on age, race/ethnicity, education, current smoking status (current, former, never), current weight loss practices (whether they were currently trying to lose or maintain weight), diabetic status ("Have you ever been told by a doctor that you have diabetes?"), and usual daily fruit and vegetable consumption (how often they drank fruit juices and how often they ate fruit, green salad, potatoes, carrots, or other vegetables). Respondents were also asked about the frequency and duration in the previous month of their 2 most frequent leisure-time physical activities. Both physical activity and fruit and vegetable questions were used to determine whether respondents were meeting national recommendations of 5 or more servings of fruits and vegetables per day and 30 minutes or more of physical activity 5 or more times per week.¹⁵ Respondents were also asked about any prescription weight loss product use in the past 2 years by the following question, "In the past 2 years, have you taken any weight loss pill prescribed by a doctor? Do not include water pills or thyroid medications."

We excluded all respondents for whom certain data were missing: weight, height, or weight loss status (n = 411) and sociodemographic factors (n = 95). We also excluded all pregnant women (n = 177). Three respondents were excluded because they reported weight, height, or BMI outside the minimum and maximum reference values of measured weight, height, and BMI by sex from the Third National Health and Nutrition Examination Survey (NHANES III), 1988-1994.¹⁶ We believe these outliers were due to either erroneous reporting or data entry errors. The final analytical sample was 14 679.

The BRFSS uses a stratified random sample approach and the data are weighted for age, race, and sex prior to data analysis. This weighting allows for inference to the state population. To account for the complex sampling design, we used SUDAAN for the primary analysis.¹⁷ We used logistic regression to assess the association between use of nonprescription weight loss products (both overall and specific use) and demographic characteristics (sex, age [18-34 years, 35-54 years, ≥55 years], race/ethnicity), current BMI (normal weight, overweight, obese), prescription weight loss pill use in the past 2 years (yes or no), and lifestyle characteristics including current smoking status (current vs former or never), usual daily fruit and vegetable consumption (<1, 1-2, 3-4, ≥5 times per day), and physical activity (inactive, somewhat active, met the physical activity recommendation). Biologically relevant 2-way interaction terms were evaluated, eg, BMI × age, BMI × sex. None of the interaction terms assessed were significantly associated with overall nonprescription product use or specific nonprescription product use at the α = .05 level. No collinearity was observed. Odds ratios and accompanying 95% confidence intervals were obtained from the RLOGISTIC procedure in SUDAAN.

More than half of the respondents were women (Table 1). The majority of all respondents were non-Hispanic white. Slightly more than half had at least some college education and most were older than 35 years. Less than half of the participants were normal weight, one third were overweight, and one fifth were obese. Approximately one third reported they were currently trying to lose weight, and one third reported they were currently trying to maintain weight.

Seven percent of the respondents reported using at least 1 nonprescription weight loss product during the previous 2 years (Table 2). Women and younger adults were significantly more likely to be users, whereas there was no difference in use by ethnicity. People with at least a high school diploma were also more likely to report using nonprescription products than those with less education. Nonprescription product use increased significantly with increasing BMI. Nonprescription product use was common among obese women of all ethnic groups 18 to 34 years of age (28.4%): non-Hispanic white, 30.3%; non-Hispanic black, 26.1%; and Hispanic, 27.1%. Nonprescription product use was also common among those who reported they were trying to lose weight (14.3%; SE, 0.6%) and less common among those trying to maintain their current weight (3.6%; SE, 0.4%). We found that some respondents who at the time of the survey were not overweight or obese also reported taking nonprescription products in the past 2 years (overall, 5.1%; women, 7.9%; men, 0.8%).

Of those who took any prescription weight loss product in the previous 2 years, over one third reported also using nonprescription products. In relation to lifestyle choices, there was no consistent difference in nonprescription product use by daily fruit and vegetable consumption; however, those who reported at least some physical activity were more likely than inactive respondents to report using nonprescription products.

We also assessed the prevalence of specific types of nonprescription weight loss products, specifically ephedra and PPA products, by select demographic characteristics and lifestyle choices (Table 3). We found that 1% of respondents used ephedra products and 2% used PPA. Multivariate logistic regression results for specific nonprescription product use were generally similar to those for overall nonprescription product use, but the magnitude of the association measure differed for some relationships. For example, women were almost 9 times more likely than men to report use of a PPA weight loss product and prescription pill users were 9 times more likely than nonusers to have also taken ephedra products.

Because of possible safety issues, we also assessed use of nonprescription products among persons who reported physician-diagnosed diabetes. Among the people with diabetes, 5.9% (SE, 1.2%) reported having used any nonprescription weight loss product, 1.2% (SE, 0.5%) used PPA, and 0.6% (SE, 0.4%) used an ephedra product compared with 7.0% (SE, 0.3%), 2.1% (SE, 0.2%), and 1.0% (SE, 0.1%) of people without diabetes, respectively.

Of the 183 respondents in Michigan who reported "other" nonprescription product use, 58% reported using liquid meal-replacement products (eg, Slim Fast and Sweet Success), 33% reported using name-brand products that claim to contain both ephedra products and chromium picolinate, and 6% reported using products claiming to contain chromium picolinate without ephedra. An additional 3% of respondents could not remember the name of the product(s).

In this population-based study of US adults in 5 states, 7% reported using nonprescription weight loss products, 2% reported using PPA, and 1% reported using ephedra products from 1996 to 1998. Extrapolated nationally, we estimate that during 1996 through 1998, approximately 17.2 million Americans used nonprescription weight loss products, 5.0 million used PPA, and 2.5 million used products containing ephedra. Overall use was common among women, especially young obese women, over one quarter of whom reported use. Moreover, 8% of normal-weight women reported nonprescription product use.

Our data are generally supported by a 1991 nationally representative study of persons trying to lose weight. Leavy and Heaton¹⁸ found that 20% of women and 11% of men reported using a weight control product (including weight loss pills, diet supplements, and laxatives). In our study, 18% of women and 8% of men who were currently trying to lose weight reported using nonprescription weight loss products. Although our data included liquid or powder meal-replacement products, we did not have specific information on the use of diuretics or laxatives.

Recently, questions were raised regarding the safety of PPA and ephedra. Between June 1997 and March 1999, the FDA received 140 reports of adverse events among users of ephedra products.^19- 20 Ephedra products are regulated as dietary supplements under the 1994 Dietary Supplement Health and Education Act.²¹ Dietary supplements are generally regarded as safe and are regulated as foods rather than drugs. Under the Dietary Supplement Health and Education Act, the burden of proof for establishing that dietary supplements are unsafe falls to the FDA rather than to the manufacturer.^21- 22 In addition, the FDA is not responsible for quality control, which means that there can be a discrepancy between the actual composition or potency of a product and the specifications on the label. For example, 11 of 20 ephedra supplements tested failed to list the ephedrine alkaloid content on the label or had more than a 20% difference between the actual amount and the amount listed on the label.²³

As a synthetic ephedrine alkaloid, PPA is not regulated as a dietary supplement but as an OTC drug. In November 2000, the FDA's Nonprescription Drugs Advisory Committee concluded that PPA was associated with hemorrhagic stroke and recommended that PPA not be considered safe for OTC use. The committee recommended removal of all OTC PPA products from the market.¹³ This withdrawal left future use of PPA products uncertain and may increase sales of other weight loss products such as prescription drugs, ephedra products, and other dietary supplements.

Health care professionals need to know about their patients' use of both prescription and nonprescription weight loss products. In our study, over one third of women users of prescription pills and one tenth of men users also reported taking nonprescription products at some time during the 2-year time period. In fact, prescription pill users were 9 times more likely than nonusers to have also taken ephedra products in the 2-year period and twice as likely to have taken PPA products. Our survey did not collect information as to whether the products were taken serially or simultaneously; the dose, duration, or frequency; prior use of these products; or whether users of nonprescription weight loss products told their physicians. It is important for physicians to know if multiple weight loss products are being taken at the same time, as there is a possibility for herb-drug and drug-drug interactions.

Dietary supplements and alternative therapies are a particular challenge for physicians. Many patients do not inform their physicians about their use of these products.²⁴ Of particular concern was our finding that ephedra and PPA were used by people with diabetes. In this group of individuals, use of these products may result in adverse effects,^5,9,25 especially if uncontrolled hypertension is present. We did not have data for nonprescription weight loss product use in persons who have other weight-related health conditions such as hypertension and heart disease. Use of ephedra and PPA products may put these individuals at risk for adverse health events such as myocardial infarction and stroke.^{10- 11,19- 20}

We found little evidence to support the speculation that nonprescription product users are less likely to change their lifestyle compared with nonusers. There was no difference in nonprescription product use by fruit and vegetable consumption, but nonprescription product users were less likely to be sedentary than non-nonprescription product users. However, the proportion meeting the national recommendations for physical activity was similar for both groups. Our analysis is limited in that respondents were asked about any use of nonprescription and prescription weight loss products in the past 2 years, whereas they were asked about current weight and height, usual fruit and vegetable consumption, and previous month leisure-time physical activity.

We were not able to verify the actual product(s) taken from the BRFSS. There are 2 potential effects on our prevalence estimates that cannot be confirmed. Some respondents may have not been aware that they were taking ephedra products and thus underreported their intakes. To the contrary, it is also possible that some respondents took an herbal fen-phenlike product that did not contain ephedra and thus overreported ephedra use.

Since obesity is a chronic disease it is possible that individuals may use nonprescription products to maintain weight loss; however, use of these products by normal-weight individuals could expose them to risks for which there are no counterbalancing benefits. Our survey did not collect information on whether the respondent experienced adverse effects from the nonprescription product or whether weight loss or weight maintenance was achieved. Although respondents were asked about whether they were currently trying to lose or maintain weight, they were not asked about the current type of diet or current weight loss product they were taking.

Providing appropriate science-based advice will be a challenge for health care professionals because of the increasing variety of nonprescription products on the market and the lack of methodologically sound efficacy studies.^1,26 The continuing increase in the rate of obesity in the United States²⁷ and the attractiveness and ease of obtaining weight loss products will probably increase the use of both prescription and nonprescription products. With this increase comes a greater need for health care professionals to take an active role in educating themselves to help their patients make appropriate choices.