Actively try to lose weight may cause people to eat more fatty food

[7rmr]

Here is a very interesting study I came across while doing some research for a health psychology project. Don't worry below I included a summary write up of the study that I had to write for my project. So even if you don't want to read the whole study, there is a quick summary at the bottom. Just wondering what you guys thought about it. I think its some very intersting stuff and can help explain how difficult it can be for some individuals to lose weight. At the same time its also helpful information for those trying to lose weight, because you can now watch out for this in your own shopping habits.

Copyright ? 2008 Elsevier Inc. All rights reserved.

Dieting and food choice in grocery shopping

S?bastien Paradisa and Michel CabanacCorresponding Author Contact Information, a, E-mail The Corresponding Author

aCentre de recherche sur le m?tabolisme ?nerg?tique (CREME), D?partement d'anatomie et physiologie, Universit? Laval, Qu?bec, Canada

Received 28 September 2007;
revised 9 January 2008;
accepted 16 January 2008.
Available online 26 January 2008.

Abstract

As obesity becomes increasingly prevalent, many people are trying to control their body weight through dieting, with mitigated results. We analyzed the impact of weight loss on food choice by recording the grocery basket composition of 100 participants, using their grocery receipts. Participants also anonymously completed a questionnaire about age, sex, diet, recent body weight change, reasons for recent body weight changes, and perceived difficulties in losing weight. Participants who had deliberately lost weight chose more dairy products, meat, and sweets and fewer fruits and vegetables than did controls. Passive weight-losers were similar to controls in their food choices. Active weight-losers show a stronger desire for high-caloric intake, probably because of a behavioral mechanism that seeks to maintain their original body weight set-point.

Keywords: Food choice; Set-point; Regulation; Body weight loss; Diet
Article Outline

1. Introduction
2. Methods
3. Results
4. Discussion
Acknowledgements
References

1. Introduction
Dieting is often considered the best way to lose weight, but even with encouraging short-term results [1] most dieters do not maintain their new weight for long [2]. From 50% to 80% of dieters experience a frustrating ?rebound? after weight loss [3] and [4]. The long-term effectiveness of dieting is doubtful. Even in structured weight-loss programs, most participants regain almost all of their initial weight-loss after 5 yr, with only 40% maintaining a 5% reduction in body weight for 5 yr [5]. Outcomes are probably worse with non-structured, individually managed diets. There is also a report of more successful long-term weight-loss, but most implies important energy expenditure through exercise [6].

It may be that people with recent deliberate weight loss tend to choose food with higher caloric content, this being a result of their falling below a ?set-point? that regulates their body weight [7], [8] and [9]. The difference between this set-point and current body weight (called ?error-signal?) is internally sensed as a ?lack? or ?need? and the individual is thus driven to consume food with higher caloric content.

In this study, we investigated how dieting and intentional weight-loss may influence food choice in a real-life context, i.e., by analyzing grocery receipts that consumers provided after making their purchases. This method eliminated observational bias: when the consumers were choosing their food items, they were unaware that they would be participating in a study.
2. Methods

In September 2005, we collected grocery receipts from 100 consumers in a medium-sized supermarket in Quebec City. The methodology was approved by Laval University Ethics Committee and the supermarket manager, and had the cooperation of store personnel. After payment at the cashier, each consumer was asked to provide a copy of the receipt and to complete a short questionnaire about diet, food shopping habits, recent weight loss, age, and sex. No personal identification was requested, full anonymity and confidentiality were observed throughout the study. The participants were selected arbitrarily as the first 100 people who accepted to participate to the study and who had spend more than 50 CA$ (approximately 50 US$). This last criterion was determined to ensure that the receipt would present a sufficient number of items to allow a significant analysis of its composition. From memory, refusal rate was very low. Because the purpose of this study was not epidemiology but intended to investigate how a physiological factor (intentional weight-loss) might influence food choice, the number of participants did not have to be representative of a large population. Out of the 100 participants, 57 were women and 43 were men. Mean age of participants was 40 ? 1.5 yr. Most of participants (52%) were purchasing for their single need, 32% for two persons and 16% for more than two persons. This repartition was similar in each subgroup described below (MANOVA, P = 0.32).

Participants were subsequently categorized according to their answers to questions about weight-loss in the last month. Those who reported no weight loss over the last month were classified as controls. Other participants were categorized according to their answers to the question: ?Did you make deliberate efforts to loose weight?? Participants who answered yes were considered ?active weight-losers,? those who answered ?no? were considered ?passive weight-losers.? This classification based on self-reported body weight loss and effort to achieve it, is coherent with the key variable of our classification: participants' self-perception of effort rather than actual body weight.

The receipts furnished enough information for us to classify all food items. The items were assigned to one of the following categories: vegetable; fruit; cereal; dairy product; ready-to-eat; meat; sweet; drink.

The number of items in each category provided a portrait of each consumer's food choices.

Intergroup differences in relative importance of each food category were compared through multivariate analysis of variance (MANOVA); independent post hoc Student's t-tests were performed when necessary (Sig: P < 0.05).
3. Results

Out of 100 participants, 41 stated they had lost weight in the last month. Weight loss ranged from 0.5 kg to 12.0 kg with a mean (? se) of 4.6 ? 0.4 kg. The self-reported weight-loss was similar in both subgroups: 4.7 ? 0.5 kg for active weight-losers and 4.9 ? 0.6 kg for passive weight-losers. Mean weight loss did not significantly differ between men and women (P = 0.55). Nor did MANOVA reveal any significant difference between men and women in the relative composition of their grocery receipts, i.e., their categorized food choices were similar. Similar analysis did not show significant difference between the six age groups presented in the questionnaire (under 25; 26?35; 36?45; 46?55; 56?65 and over 65 yr old). Therefore, the results for both sexes and different age groups were pooled for subsequent analysis. ANOVA neither revealed a difference in the food choice according to the answer to the question ?For how many persons are you buying that food??

Of the 41 participants who had reported recent weight loss, 11 were passive weight-losers and 30 active weight-losers, i.e., were still dieting. The remaining 59 were considered to be controls.

Fig. 1 presents relative grocery composition based on the receipts. It can be seen that the grocery composition for passive weight-losers was similar to that for controls, except for the ?sweet? category. Fig. 1 also shows that, compared to controls and passive weight-losers, active weight-losers chose fewer vegetables and fruits and more dairy products and meat. They also chose more sweets than did the controls, while choosing about the same amount as the passive weight-losers chose.


Full-size image (39K) - Opens new window Full-size image (39K)

Fig. 1. Relative composition of grocery baskets of three groups of consumers, based on the number of items in each category. Different lowercases indicate significant differences between groups (two-tailed Student's T-test, P < 0.05).

View Within Article


4. Discussion

We wished to see whether active weight loss influences decisions when shopping for food. Passive weight-losers and controls made similar food choices, as would be expected if a regulation was maintaining their body weights. Since passive weight-losers lose weight unintentionally, without effort and frustration, we may assume that their weight is tracking their body weight set-point. They thus resembled the controls in that they were not experiencing any error-signals from their body weight regulation. Such a conclusion is consistent with results from a previous study where active weight-losers, passive weight-losers, and controls were compared with respect to acceptance of aggressive responses in several situations [10]. In that study too, passive weight-losers had results similar to those of controls, but different from those of active weight-losers.

At first glance, the active weight-losers seem to have been acting paradoxically. They were actively trying to lose weight and yet chose fewer vegetables and fruits and more meat and dairy products than did the controls and passive weight-losers. They also chose more sweets than did the controls. Overall, they were choosing food with presumably higher caloric content. This paradox may be resolved by examining it within the context of body weight-regulation. By ?fighting? their body weight set-point, they were generating an ?error-signal,? i.e., a difference between their set-point and current body weight [7], [9] and [11]. Their bodies interpreted the error-signal as an energy imbalance that must be rectified, the result being a feeling of discomfort. Food choice provides the body with a means to defend its set-point. When body weight deviates from this set-point because of deliberate dieting, an error-signal is generated that leads to apparently paradoxical food choices. The dieters themselves were probably unaware of this change in their food choices.

This conclusion may partially explain why dieting is ineffective in lowering body weight over the long-term. Despite much effort and frustration, successful weight loss generates an error-signal that unconsciously promotes behavior to oppose the loss.

The food choice of self-reported dieters may also be explained from the restriction behavior point of view, used notably by Stunkard in his three-factor eating questionnaire [12]. Our methodology, based on self-reported dieting, has probably selected restrained eaters among active weight-losers. Evidence has shown that self-reported restrained eaters do not eat any less in the natural environment than unrestrained eaters [13] E. Stice, M. Fisher and M.R. Lowe, Are dietary restraint scales valid measures of acute dietary restriction? Unobtrusive observational data suggest not, Psychol Assess 16 (1) (2004), pp. 51?59. Abstract | Article | PDF (68 K) | Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (70)[13], a conclusion compatible with the one presented here. Finally, restrained eating is known to prospectively predict future weight gain. [14] and [15], which is consistent with our observation that self-reported dieters tended to choose food susceptible to make them heavier.
Acknowledgments

[7rmr]

Here is the summary

Sebastien Paradis and Michel Cabanac?s 2007 study Dieting and food choice in grocery shopping explores possible reasons for the rebound affect that occurs so often when people try and lose weight. Even in structured weight loss programs upwards of ?50% to 80%?(Paradis & Cabanac, 2008) of dieters experience some sort of rebound after successfully losing weight. In order to explore possible reasons for this effect, Paradis and Cabanac looked at the grocery habits of 100 shoppers in a Quebec City grocery store. The shoppers were not aware that they would be participatin in the study until after purchasing their groceries. They were then approached and asked if they would take part by letting the researchers see their grocery bills, as well as fill out a quick survey. The first 100 shoppers to agree made up the sample. The questionnaires the participants filled out asked them a number of questions regarding age, sex, weight and whether they had lost any weight in the past month. Those who had lost weight but were not trying to lose weight were labeled passive weight losers, while those who had lost weight and were trying to lose weight were labeled active weight losers. Those who had not lost any weight in the previous month were used as a control. It was found that those who were active weight losers bought fewer fruits and vegetables and more meat, sweets and dairy then the controls. Paradis and Cabanac hypothesized that the reason for this counter intuitive buying pattern of buying higher calorie foods while trying to lose weight was the response of their body sending error signal,s as their current body weight dropped below their set point weight. These error signals subconsciously influenced the participant?s behaviour to help bring their body weight back up to their body?s set point.

[gautm1]

I typically don't like these studies. Too many variables to consider and based on small sample size.

[determined4000]

I typically don't like these studies. Too many variables to consider and based on small sample size.

this is really not that complex
it says people who aren't eating enough get hungry
it would be like a study that found people who got 4 hours of sleep a night felt more tired during the day and were more likely to want a nap.