Appetite, 1992, 19, 233-242
Taste and Food Preference Pregnancy
Changes Across the Course of
DEBORAH J. BOWEN Department of Public Health Sciences, Fred Hutchinson Cancer Research Center, and Department of Psychology, University of Washington
The present study investigates taste and specific food consumption changes across the course of pregnancy. These variables could potentially play a role in excess pregnancy-associated weight gains. Pregnant and postpartum women were asked to consume a series of everyday foods in the laboratory. Consumption and taste perception of each food were measured. In contrast to the self-report literature on cravings and aversions during pregnancy, which emphasizes changes in the first trimester, this study found that women in the second trimester consumed significantly more sweet food, but not salty or non-sweet/non-salty food, as compared with women at any other point in pregnancy. Subjects were restrained eaters, and so possibly refrained from daily consumption of excess sweet foods. This study suggests that psychological variables may interact with behavioral and physiological variables to control food preferences and eating in pregnancy.
Eating behavior and specific nutrient consumption during pregnancy have received a great deal of research and clinical attention. Most of the research has focused on a lack of needed nutrients in a pregnant woman’s diet (Dobbing, 1981; Hytten, 1981). Based on this research the American College of Obstetrics and Gynecology has published a set of specifications for minimum nutrient requirements of pregnant and lactating women. However, little is known about how a pregnant woman chooses her diet when food is plentiful and even in excess, as is the case in several segments of current industrialized societies. Taste is an important factor in food selection in non-pregnant people (Spiker & Rodin, 1985), so taste could play a role in food selection and consumption during pregnancy. The purpose of this paper is to investigate taste differences across pregnancy. For most women, general food consumption increases during pregnancy (Naismith, 1980; Strubbe & Gorissen, 1980; Dobbing, 1981) and by the third trimester, the mother has an increased layer of subcutaneous fat stored to meet the caloric drain incurred while nursing her infant (Newcombe, 1981; Holinka, 1980). Some This research project was conducted in partial fultillment of the requirements of a Ph.D at the Uniformed Services University of the Health Sciences. I was grateful for the support and advice of Neil E. Grunberg and Jerome E. Singer. Also valuable were the contributions of Emma Bockman, Sheryle Alagna, Richard Simmonds, and Donald Meek for serving as dissertation committee members. Address correspondence to Dr D. J. Bowen, Fred Hutchinson Cancer Research Center, 1124 Columbia St., Seattle WA 98104, U.S.A.. 0195-6663/92/060233
+ 10 %08.00/O
0
1992 Academic
Press Limited
234
D. J. BOWEN
women experience a change in consumption of specific foods. The incidence of specific cravings and preferences that develop during pregnancy has been’ reported in the literature (Brewin, 1980; Hook, 1978, 1980; Worthington-Roberts et al., 1988; Trethowan & Dickens, 1972). Most reported cravings are for a particular food (e.g., citrus fruit or for foods that provide energy or calcium) and frequently are reported during the first trimester of pregnancy. Investigations of cravings for a particular taste class are not present in the literature. Cravings could develop for a taste class of food that is high in calories. More broadly, enhanced sweet food preferences occur when an organism is in a state of increased caloric need (Cabanac, 1978). Pregnancy is surely a time when more calories are needed to support the development of a growing fetus (Picone et al., 1982). Sweet preferences could occur during pregnancy, particularly during the second and third trimesters when levels of steroid hormones involved in consumption regulation (e.g., estrogens and progestins, Wade, 1976) are increasing (Fuchs & Klopper, 1983). If sweet preferences and changes in daily consumption do occur during pregnancy, they could affect weight gains. During the course of normal pregnancy, a woman gains an average of 11 kg. Yet some pregnant women gain as much as 35 kg and remain obese after delivery of their infants (Kawakami et al., 1977; Niswander & Gordon, 1972; Heliovaara & Aromaa, 1981). Excessive weight gain has negative implications for both maternal and fetal health (Naeye, 1979). In addition, excessive weight increases can leave the mother weighing 5- 10 kg more after delivery than before her pregnancy (Woods et al., 1980; Beazley & Swinhoe, 1979). There is evidence from other literatures that preferences for sweet foods do lead to excess sweet food consumption and excess weight gains. For example, smokers in cessation gain excess weight as part of the smoking withdrawal syndrome (Rodin, 1987). One of the major reasons for this excess gain is increased consumption of sweet-tasting foods (Grunberg, 1982; Grunberg & Bowen, 1985; Rodin, 1987). If increased sweet preferences lead to excess gains after smoking cessation, then sweet preference changes during or after pregnancy could lead to excess gains as well. The present research was designed to determine patterns of taste preference and food consumption during and after pregnancy.
METHOD
Subjects Subjects for this study were 50 married women; 11, 15, and 12 from each of the three trimesters of pregnancy, respectively, and 12 postpartum subjects (average number of weeks postpartum-6.5; range=4-9). Women were of mixed parity and were equally distributed within each trimester period, except that no subjects were used from weeks l-7. Subjects were recruited through a newspaper advertisement asking for women interested in participating in a study of sensory sensitivity changes during and after pregnancy. They ranged in age from 19-34 years and were mentally and physically healthy, non-smoking for at least 2 years, and free from obstetrical complications at the time of the study. These women were generally financially stable (mean family income = US$43,000) and relatively well-educated (40 of 50 had received a college diploma, and the remaining had finished high school). Of the 50 subjects, 46 were currently employed or had been employed outside the home before
PREGNANCY,
235
FOOD AND TASTE
leaving for their pregnancies. There were no significant differences between groups for any demographic variable over the study periods when analysed with either analyses of variance or chi squares, where appropriate. Possible subjects were telephoned and were asked to participate in a study of sensation and perception during the course of pregnancy. If they agreed, they underwent a short initial telephone screening to ensure absence of obstetrical and health problems, and absence of food allergies and dietary restrictions. If the subject met all requirements, an appointment was scheduled for an evening laboratory session. The experimenter asked her to eat her regular lunch and to refrain from eating for 3 h before her appointment.
Setting When each subject arrived at the building the experimenter greeted her and escorted her to the experimental suite where several rooms were set up with equipment to test for perception of different stimuli. Subjects were told that the study involved tests of sensation and perception, including sight, sound, and taste. The experimenter consulted a chart on her clipboard and informed every subject that, “We don’t have time in one experimental session to test everyone on all of their senses. You are in the taste condition. I’d like you to taste some everyday foods and to fill out some taste judgment questionnaires”. This procedure was designed to avoid self-consciousness or suspicion about eating in relation to body weight.
Food Consumption Subjects were asked to taste nine different foods and to rate each of them on a series of basic taste dimensions. To do this, the experimenter escorted the subject into the taste room and explained to her the procedure for tasting the foods and filling out the taste judgment forms by saying “As you can see, in front of you are nine different foods for you to taste. We’d like you to taste each food in order and then describe how it tastes to you by rating each food on this set of basic taste dimensions. What I mean is, start out with Food # 1 and taste it thoroughly. Then on the sheet marked # 1, rate the food on these taste dimensions by circling a number from 1 to 7, depending on how the food tastes on each dimension. For example, if Food # 1 tastes extremely sour, circle # 1, and if it tastes not at all sour, circle #7. If you think it’s somewhere in between, circle the number that most closely corresponds to your judgment. Only circle numbersdon’t circle words. Rate Food # 1 on all the dimensions. After you completely finish rating Food # 1, take a drink of water to clean out your mouth, then move on to Food #2 and Sheet #2. Continue in order for each of the nine foods”. After explaining
the taste procedure
the experimenter
said,
“It’s important that your ratings are as accurate as possible, so eat as much of each food as you need in making your judgments. After you’ve finished rating all foods, feel free to go back and eat as much as you want of any of them. Just make sure you taste all foods before nibbling”. Almost
as an afterthought
the experimenter
said in a joking
voice
“Hey, after you leave, I’m going to throw the foods out anyway, so you might as well eat what you want”.
236
D. J. BOWEN
The experimenter left the subject in the room for 20min to taste the foods. Extensive pilot testing and previous research found that subjects took approximately 7-8 min to complete their initial tasting, and then had approximately 12 min to go back and nibble on foods of their choice. Subjects took approximately 2-3 bites of each food when tasting and rating it. All foods were presented in bite-size portions in beige bowls. The order of foods was counterbalanced for food taste class. The goal of the session was to provide a naturalistic snacking period where subjects would feel comfortable to consume freely, in contrast to more rigorous taste perception procedures, such as rinsing between mouthfuls or foods or tasting several concentrations of the same taste or substance. Subjects believed that they were in a taste testing session. In actuality, the foods were weighed before and after the taste testing session to obtain measures of general and specific food consumption in the laboratory. This deceptive manipulation was necessary to prevent self-consciousness about eating freely and to obtain an accurate measure of laboratory food consumption. Amount consumed in the laboratory provided data on taste preferences during and after pregnancy, since amount of food consumed was selected freely. This method of measuring preferences for food taste classes was based on the work of Grunberg (1982), who used it to identify sweet taste preferences for smokers in cessation. Three sweet (Sara Lee pecan frosted coffee cake, Hershey’s milk chocolate bars, Vernell’s fruit gum drops), three salty (deli-style boiled ham, Planter’s salted peanuts, Mr Salty salted pretzel sticks), and three non-sweet, non-salty foods (Kraft low-fat mozzarella cheese, Nabisco unsalted saltine crackers, Planter’s unsalted peanuts) were chosen for the study. Sweet foods were chosen because of the importance of sweet taste in the regulation of food consumption in other literatures (Cabanac, 1978; Grunberg, 1982; Grunberg & Bowen, 1985; Roils, 1985). Salty foods were chosen to represent a second common taste that is non-sweet. In the present study non-sweet and non-salty foods were chosen to indicate changes in general food consumption. These foods were chosen on the basis of taste class only, and were not balanced for nutritional content, because this study addressed taste and not macronutrient consumption. In particular, carbohydrate and fat content were not manipulated independent of the two taste dimensions. The individual foods were chosen after extensive pilot testing for the following reasons: (1) they represented both snack and meal choices; (2) they represented the taste of each taste class (sweet, salty, and non-sweet/non-salty); (3) they were easy for subjects to handle; and (4) subjects liked them. Taste judgments For each food that the subject tasted, she was asked to complete one taste judgment sheet. On each sheet was a list of taste dimensions. The subject was asked to circle a number between 1 (extremely) and 7 (not at all) for each of the following taste dimensions: sweet, sour, salty, bitter, spicy, flavorful/bland, smooth/grainy, light/heavy, strong/mild, and fresh/stale. The judgments for each type of food were used as manipulation checks (i.e., did the “sweet” foods really taste sweet?) and as an evaluation of taste perception in the present study. The first four taste dimensions (sweet, sour, salty, bitter) represented standard taste qualities used in taste research. The remaining dimensions were inserted to give subjects plenty of opportunity to consume their preferred foods.
PREGNANCY,
FOOD AND
TASTE
231
Other Assessments
Height and weight were measured at the end of the laboratory session. A body mass index (weight/height2) was calculated for each woman. The body mass index has a high correlation with obesity as measured with more accurate but involved methods (Bray, 1980). In addition, body mass indices were computed for each woman using a weight taken at her first obstetrical visit. Two questionnaires were used to assess dietary restraint. One, developed by Herman & Polivy (1980), is a lo-item scale designed to measure behavioral restraint of eating. The other restraint scale is longer and contains three subscales, including conscious restraint in eating, behavioral and weight liability, and susceptibility to hunger (Stunkard & Messick, 1985). Although both scales are used in eating and obesity research, the relationship between these two scales is not well characterized in the literature and not at all in a pregnant population. The relationship between these two scales regarding changes in eating and weight associated with pregnancy was investigated by giving subjects both scales.
&XJLTS
Laboratory Consumption
Table 1 presents the result of the laboratory food consumption measurements combined into four pregnancy study periods: first trimester (weeks l-l 5), second trimester (weeks 16-27), third trimester (weeks 28-birth), and pospartum (all postpartum weeks). Total consumption of each food and food taste class (grams of sweet, salty, and non-sweet/non-salty food) and of each individual food is presented separately. Sweetness and saltiness ratings for each food and each food taste class are included in Table 1. Total number of subjects varied slightly due to measurement error. The quantity of food consumed varies across pregnancy periods, as shown in Table 1. Subjects in the second trimester consumed more grams of chocolate, coffee cake, and gumdrops than did the subjects in any other period. A one-way analysis of variance statistically confirmed that total amount of sweet foods consumed differed during the pregnancy periods [F(3,46) = 9.72, p c O.OOOl].Using a Newman-Keuls range test to compare individual groups, women in the second trimester consumed significantly more sweet food than did women from any other group. No other individual group comparisons were significantly different. Consumption of other foods also differed throughout pregnancy. Salty peanuts were consumed more in the third trimester, compared with other pregnancy periods. When grams of all salty foods were summed women in their third trimester of pregnancy consumed marginally more grams of salty food than did subjects in either of the other trimesters or during the post-partum period [F(3,45) = 2.24, p < 0.0981. There were no significant differences in total non-sweet/non-salty food consumption. Table 1 also presents the taste perception data for all foods. Subjects rated all three sweet foods as being very sweet. There were no significant differences in the subjects’ taste ratings of the sweetness or saltiness of the sweet foods across the pregnancy periods. Subjects in the third trimester and during the postpartum period rated their salty foods as less salty overall than did subjects in either of the other two periods. The overall differences in saltiness ratings were statistically significant using
Non-sweet/Non-salty Mozarella cheese Unsalted saltines Unsalted peanuts
Salted peanuts Salted pretzels Boiled ham
Salty taste
Chocolate Coffee cake Gumdrops
Sweet taste
Total
taste
Total
Total
18.5
10.1 2.6 5.8
33.7
11.8 18.0 3.2
44.0
17.7 18.3 8.0
g
5.4 5.5 6.1
5.8 6.1 5.5
1.4 2.6 1.9
Rating
Trimester
1
5.8 6.2 6.8
1.1 1.8 2.5
6.1 5.7 6.3
Rating
19.5
11.4 1.4 6.7
31.8
11.0 16.7 4.1
72.0
32.8 26.7 12.5
g
Food consumption
TABLE 1
5.9 6.0 6.2
5.5 6.0 5.9
1.4 2.7 1.2
Rating
Trimester
2
5.8 5.3 6.2
1.3 1.6 2.4
5.9 5.7 6.2
Rating
data across pregnancy
24.2
11.3 1.9 11.3
52.1
28.9 17.6 4.5
53.1
27.8 19.2 6.1
g
4.9 5.6 6.0
5.9 6.0 5.7
1.3 0.7 1.5
Rating
Trimester
study periods 3
5.5 5.4 6.2
2.6 2.8 3.1
5.4 5.9 6.3
Rating
21.3
7.9 3.3 9.8
223
13.8 13.0 5.2
60.0
21.1 21.0 7.9
g
5.1 6.0 5.9
6.2 6.1 5.4
1.7 2.9 1.8
Rating
Postpartum
5.8 5.7 6.5
2.5 2.2 2.1
5.9 5.8 6.4
Rating
PREGNANCY,
239
FOOD AND TASTE
a one-way analysis of variance [F(3,45) = 4.80, p < O*Ol]and a Neuman-Keul’s test confirmed that saltiness ratings were lower in the third trimester and the postpartum period. The salty foods were rated as being not sweet. Assumption about the taste of the non-sweet/non-salty foods were confirmed, as those foods were rated not sweet and not salty. Total laboratory consumption shown as kilocalories consumed is presented in Figure 1. Subjects in the second trimester consumed more calories in the laboratory than did subjects in any other period. The group differences were statistically significant using an analysis of variance [F(3,45) = 2.8 1, p < 0.051. A Newman-Keuls range test confirmed that women in the second trimester consumed significantly more kilocalories than did any other group. No other group differed significantly. This difference in caloric consumption differed from the similarity of consumption values when expressed as grams of food consumed. Over the four periods, laboratory food consumption computed in grams of food was not significantly different, although subjects in the second and third trimester consumed slightly more grams of food than did subjects in the other two pregnancy periods. As expected, weight gain was higher from first through third trimester for subjects in the present study. The weight gains were approximately equal to guidelines set by the American College of Obstetrics and Gynecology. Data for the postpartum subjects showed a slight weight gain (roughly 71b) from pre-pregnancy levels that remained after the birth of the child. Body mass index (weight/height2) values for subjects at time of study participation paralleled weight gain data. In contrast to BMI’s at the time of study, BMI’s calculated using body weight from the first obstetrical visit were virtually identical for women in each of the study periods. Subjects were generally restrained eaters, as shown by their scores on the two restraint questionnaires, the Stunkard and the Herman and Polivy (mean for the Stunkard scale = 31.32, range =0-51; mean score in this range is high for normal population. Scores on the HP restraint scale were generally high for normal individuals). There were no differences in scores for any of the scales among the
50C
400 %
Y
$g I 5
300
f
200
3 100
-
0
-L
2nd
L
3rd Tnmesters
FIGURE1.
Total calorie consumption
across pregnancy.
PP
D. J. BOWEN
measurement periods. The Stunkard restraint scale was positively, significantly, and highly correlated with scores on the Herman and Polivy restraint scale (r=0.76; p c O-01). This relationship was expected, because the Stunkard scale was developed, in part, using items from the Herman and Polivy scale.
DISCUSSION
The results of the present study suggested that during the course of pregnancy, preference for certain foods varied. Women in the second trimester consumed more sweet-tasting food than did women in any other pregnancy study period. There were no significant differences in the sweetness ratings of the sweet foods for any group of subjects. Therefore, the consumption difference was not due to changes in the perception of sweet taste as measured in the present study. Some of the sweet foods, however, were also high in fat as well as sweet taste. It is not possible from the present selection of foods to compare fat consumption, fat preference, or fat perception taste changes during pregnancy. This problem invites further investigation, because of the importance of fat consumption and fat preference in human obesity (Drewnowski & Greenwood, 1982). Previous research reported no differences in sweet preferences during pregnancy (Brown & Toma, 1986); however, most of the subjects in the previous study were in the third trimester. Changes in sweet preference were found in the present study only in the second trimester. The second trimester is the time of most dramatic changes in levels of hormones, particularly estrogens and progestins, that are responsible for body weight and food consumption regulation in the nonpregnant state (Fuchs & Klopper, 1983; Czaja, 1983). These hormones could be responsible for changes in sweet food consumption in the second trimester period. Consumption of salty foods also differed for women across pregnancy periods. Women in the third trimester consumed marginally more salty foods than did any other group of women in the present study, a finding supported by previous research (Brown & Toma, 1986). This behavioral change corresponded to a change in the taste perception of the salty foods for women in the third trimester, who rated those foods as less salty compared with the ratings of other women. This is a noteworthy point, because in the later trimesters of pregnancy, women retain more fluids and are vulnerable to eclampsia, or pregnancy-induced hypertension. Whether or not consumption of salty foods is involved in this serious problem is unknown from the present study, since all subjects were healthy. Women with eclampsia could experience a change in taste perception for salty foods, consume more of these foods, and so unknowingly assist in the development of this obstetrical problem. There were significant differences in the number of calories consumed across pregnancy. The increase in caloric consumption in the second trimester resulted from the increase in consumption of certain foods used in the study, such as coffee cake and candy. This increased consumption of sweet-tasting and/or high-fat foods might contribute to excess weight gains if continued in daily food consumption. Subjects in the present study were restrained eaters, as shown from their scores on both restraint scales. Although women in the second trimester exhibited an increase in preference for and consumption of sweet foods in the present laboratory study they did not gain excessive weight during their pregnancies. The answer to this puzzle may lie in the relatively high levels of restraint reported by these women.
PREGNANCY,
FOOD AND TASTE
241
Possibly, these women were so restrained that they would hold back from eating foods so that they would not consume increased amounts for fear of gaining weight. For other women who were not so restrained, an increase in preference might lead to an increase in daily consumption, and so to increased weight. An attempt was made in the laboratory study to allow the women to feel as comfortable and relaxed as possible. They might have eaten more sweet foods in the present study because of a second-trimester preference change, but would restrain themselves from eating extra amounts of food when in a more familiar environment. This possibility needs to be investigated both using naturalistic methods of actual daily food consumption collection and in other populations that might not be as restrained as were women in the present study. This change in food preference and consumption may have implications for proper nutritional intake during pregnancy. Women who are not educated as to proper prenatal nutrition may be allowing their preferences to guide their food choices during pregnancy. If laboratory sweet preferences do guide daily food consumption, some women may not be selecting their diets to consist mainly of nonsweet foods that are very nutritious. Low income and minority populations that might not have access to adequate prenatal or nutritional care and advice are the same groups that gain the most excess pregnancy-associated weight. These populations also have the highest incidence of low-birthweight infants. The role of taste preference changes, dietary intake, and nutritional advice in the incidence of lowbirthweight infants remains to be investigated. The present study has documented taste preference changes during the second trimester of pregnancy. More research relating normal and excess weight gains in pregnant populations to food preferences and daily eating habits must be conducted. An understanding of the role of taste in daily consumption changes and weight gains during pregnancy could provide a model for psychobiological determinants of weight regulation, in addition to improving the health of women who are at risk of obesity due to pregnancy.
Beazley, J. M. & Swinhoe, J. R. (1979) Body weight in parous women: Is there any alteration between successive pregnancies? Acta Obstetrica Gynecologia Scandinavia, 58, 45-47. Bray, G. A. (Ed.) (1980) Obesity in America. USDEW (PHS) NIH. Brewin, T. B. (1980) Can a tumor cause the same appetite perversion or taste change as a pregnancy? L..ancet, 907-908. Brown, J. E. & Toma, R. B. (1986) Taste changes during pregnancy. American Journal of Clinical Nutrition, 43, 414-418.
Cabanac, M. (1978) Sensory pleasure. Quarterly Review of Biology, 54, l-29. Czaja, J. A. (1983) Body weight and growth rates throughout the guinea pig pregnancy: Evidence for modulation by endogenous estrogens. Physiology and Behavior, 30, 225-231.
Dobbing, J. (Ed.) (1981) Maternal nutrition: Eating for two? New York: Academic Press. Drewnowski, A. & Greenwood, M. R. C. (1983) Cream and sugar: Human preferences for high-fat foods. Physiology and Behavior, 30, 629-633. Fuchs, F. & Klopper, A. (Eds) (1983) Endocrinology of Pregnancy. Philadelphia: Harper & Row. Grunberg, N. E. (1982) The effects of nicotine and cigarette smoking on food consumption and taste preference. Addictive Behaviors, 7, 317-331.
D. J. BOWEN
242 Grunberg,
N. E. & Bowen, D. J. (1985) Coping with the sequelae of smoking cessation.
American Journal of Cardiac Rehabilitation, 5, 285-289.
Heliovaara,
M. & Aromaa,
A. (1981) Parity and obesity. Journal of Epidemiology and
Community Health, 35, 197-199.
Herman, C. P. & Polivy, J. (1980) Experimental and clinical aspects of restrained eating. In A. Stunkard (Ed.), Obesity. Philadelphia: W. B. Saunders. Holinka, C. (1980) Age-related weight changes in female G57BL/6J mice: nonpregnant body weight and extra-fetoplacental maternal weight throughout pregnancy. Mount Sinai Journal of Medicine, 47, 104-l 10. Hook, E. B. (1980) Influence of pregnancy on dietary selection. International Journal of Obesity, 4, 338-340.
Hook, E. B. (1978) Dietary cravings and aversions during pregnancy. American Journal of Clinical Nutrition, 31, 1355- 1362.
Hytten, F. E. (1981). Weight gain and pregnancy-30
years of research. South Africa Mediese
Tydskrif 15- 19.
Kawakami, S., Ishiwata, C., Hayashi, K., Kawaguchi, Y., Kondo, M. & Iizuk, R. (1977) Alteration of maternal body weight in pregnancy and postpartum. Keio Journal of Medicine, 26, 53-62.
Naeye, R. L. (1979) Weight gain and the outcome
of pregnancy.
American Journal of
Obstetrics and Gynecology, 13.5, 5-9.
Naismith, D. J. (1980) Maternal nutrition and the outcome of pregnancy-a critical appraisal. Proceedings of the Nutritional Society, 39, l-l 1. Newcombe, R. G. (1982) Development of obesity in parous women. Journal of EpidemioZogy and Community Health, 36, 306-309.
Niswander, E. L. & Gordon, J. J. (1972) The women and their pregnancies. New York: Academic Press. Picone, T. J., Allen, L. H., Schramm, M. M. & Olseny, P. N. (1982) Pregnancy outcome in North American women. I. Effects of diet, cigarette smoking, and psychological stress on maternal weight gain. American Journal of Clinical Nutrition, 36, 1205-1213. Rodin, J. (1987) Weight change following smoking cessation: the role of food intake and exercise. Addictive Behaviors, 12, 303-319. Rolls, B. J. (1985) Experimental analyses of the effects of variety in a meal on human feeding. American Journal of Clinical Nutrition, 42, 932-939.
Spitzer, L. & Rodin, J. (1981). Human eating behavior: A critical review of studies in normal weight and overweight individuals. Appetite, 2, 293-329. Strubbe, J. H. & Gorissen, J. (1980) Meal patterning in the lactating rat. Physiology and Behavior, 25, 775-777.
Stunkard, A. J. & Messick, S. (1985) The three-factor eating questionnaire to measure dietary restraint, disinhibition, and hunger. Journal of Psychosomatic Research, 29, 71-84. Trethowen, W. H. & Dickens, G. (1972) Cravings, aversion, and pica of pregnancy. In J. G. Howells (Ed.), Modern perspectives in psychiatry (Vol. 5): Modern perspectives in psychoobstetrics, Edinburgh: Oliver 8z Boyd, pp. 251-268. Wade, G. (1976) Sex hormones, regulatory behavior, and body weight. In J. Rosenblatt (Ed.), Advances in the study of behavior. New York: Academic. Woods, D. L., Dewar, R., Heese, A. F. & Rush, R. W. (1980) Maternal post-delivery weight in the assessment of obesity. South Africa Mediese Tydskrtf, 500-501. Worthington-Roberts, B., Little, R. E., Lambert, M. D. & Wu, R. (1989) Dietary cravings and aversions in the postpartum period. Journal of the American Dietetic Association, 89, 647-651.
Zuckerman, M. (1960) The development of an Affect Adjective Check List for the measurement of anxiety. Journal of Consulting Psychology, 24, 57-462. Received 9 September 1988, revision IS April 1989
Taste and Food Preference Pregnancy
Changes Across the Course of
DEBORAH J. BOWEN Department of Public Health Sciences, Fred Hutchinson Cancer Research Center, and Department of Psychology, University of Washington
The present study investigates taste and specific food consumption changes across the course of pregnancy. These variables could potentially play a role in excess pregnancy-associated weight gains. Pregnant and postpartum women were asked to consume a series of everyday foods in the laboratory. Consumption and taste perception of each food were measured. In contrast to the self-report literature on cravings and aversions during pregnancy, which emphasizes changes in the first trimester, this study found that women in the second trimester consumed significantly more sweet food, but not salty or non-sweet/non-salty food, as compared with women at any other point in pregnancy. Subjects were restrained eaters, and so possibly refrained from daily consumption of excess sweet foods. This study suggests that psychological variables may interact with behavioral and physiological variables to control food preferences and eating in pregnancy.
Eating behavior and specific nutrient consumption during pregnancy have received a great deal of research and clinical attention. Most of the research has focused on a lack of needed nutrients in a pregnant woman’s diet (Dobbing, 1981; Hytten, 1981). Based on this research the American College of Obstetrics and Gynecology has published a set of specifications for minimum nutrient requirements of pregnant and lactating women. However, little is known about how a pregnant woman chooses her diet when food is plentiful and even in excess, as is the case in several segments of current industrialized societies. Taste is an important factor in food selection in non-pregnant people (Spiker & Rodin, 1985), so taste could play a role in food selection and consumption during pregnancy. The purpose of this paper is to investigate taste differences across pregnancy. For most women, general food consumption increases during pregnancy (Naismith, 1980; Strubbe & Gorissen, 1980; Dobbing, 1981) and by the third trimester, the mother has an increased layer of subcutaneous fat stored to meet the caloric drain incurred while nursing her infant (Newcombe, 1981; Holinka, 1980). Some This research project was conducted in partial fultillment of the requirements of a Ph.D at the Uniformed Services University of the Health Sciences. I was grateful for the support and advice of Neil E. Grunberg and Jerome E. Singer. Also valuable were the contributions of Emma Bockman, Sheryle Alagna, Richard Simmonds, and Donald Meek for serving as dissertation committee members. Address correspondence to Dr D. J. Bowen, Fred Hutchinson Cancer Research Center, 1124 Columbia St., Seattle WA 98104, U.S.A.. 0195-6663/92/060233
+ 10 %08.00/O
0
1992 Academic
Press Limited
234
D. J. BOWEN
women experience a change in consumption of specific foods. The incidence of specific cravings and preferences that develop during pregnancy has been’ reported in the literature (Brewin, 1980; Hook, 1978, 1980; Worthington-Roberts et al., 1988; Trethowan & Dickens, 1972). Most reported cravings are for a particular food (e.g., citrus fruit or for foods that provide energy or calcium) and frequently are reported during the first trimester of pregnancy. Investigations of cravings for a particular taste class are not present in the literature. Cravings could develop for a taste class of food that is high in calories. More broadly, enhanced sweet food preferences occur when an organism is in a state of increased caloric need (Cabanac, 1978). Pregnancy is surely a time when more calories are needed to support the development of a growing fetus (Picone et al., 1982). Sweet preferences could occur during pregnancy, particularly during the second and third trimesters when levels of steroid hormones involved in consumption regulation (e.g., estrogens and progestins, Wade, 1976) are increasing (Fuchs & Klopper, 1983). If sweet preferences and changes in daily consumption do occur during pregnancy, they could affect weight gains. During the course of normal pregnancy, a woman gains an average of 11 kg. Yet some pregnant women gain as much as 35 kg and remain obese after delivery of their infants (Kawakami et al., 1977; Niswander & Gordon, 1972; Heliovaara & Aromaa, 1981). Excessive weight gain has negative implications for both maternal and fetal health (Naeye, 1979). In addition, excessive weight increases can leave the mother weighing 5- 10 kg more after delivery than before her pregnancy (Woods et al., 1980; Beazley & Swinhoe, 1979). There is evidence from other literatures that preferences for sweet foods do lead to excess sweet food consumption and excess weight gains. For example, smokers in cessation gain excess weight as part of the smoking withdrawal syndrome (Rodin, 1987). One of the major reasons for this excess gain is increased consumption of sweet-tasting foods (Grunberg, 1982; Grunberg & Bowen, 1985; Rodin, 1987). If increased sweet preferences lead to excess gains after smoking cessation, then sweet preference changes during or after pregnancy could lead to excess gains as well. The present research was designed to determine patterns of taste preference and food consumption during and after pregnancy.
METHOD
Subjects Subjects for this study were 50 married women; 11, 15, and 12 from each of the three trimesters of pregnancy, respectively, and 12 postpartum subjects (average number of weeks postpartum-6.5; range=4-9). Women were of mixed parity and were equally distributed within each trimester period, except that no subjects were used from weeks l-7. Subjects were recruited through a newspaper advertisement asking for women interested in participating in a study of sensory sensitivity changes during and after pregnancy. They ranged in age from 19-34 years and were mentally and physically healthy, non-smoking for at least 2 years, and free from obstetrical complications at the time of the study. These women were generally financially stable (mean family income = US$43,000) and relatively well-educated (40 of 50 had received a college diploma, and the remaining had finished high school). Of the 50 subjects, 46 were currently employed or had been employed outside the home before
PREGNANCY,
235
FOOD AND TASTE
leaving for their pregnancies. There were no significant differences between groups for any demographic variable over the study periods when analysed with either analyses of variance or chi squares, where appropriate. Possible subjects were telephoned and were asked to participate in a study of sensation and perception during the course of pregnancy. If they agreed, they underwent a short initial telephone screening to ensure absence of obstetrical and health problems, and absence of food allergies and dietary restrictions. If the subject met all requirements, an appointment was scheduled for an evening laboratory session. The experimenter asked her to eat her regular lunch and to refrain from eating for 3 h before her appointment.
Setting When each subject arrived at the building the experimenter greeted her and escorted her to the experimental suite where several rooms were set up with equipment to test for perception of different stimuli. Subjects were told that the study involved tests of sensation and perception, including sight, sound, and taste. The experimenter consulted a chart on her clipboard and informed every subject that, “We don’t have time in one experimental session to test everyone on all of their senses. You are in the taste condition. I’d like you to taste some everyday foods and to fill out some taste judgment questionnaires”. This procedure was designed to avoid self-consciousness or suspicion about eating in relation to body weight.
Food Consumption Subjects were asked to taste nine different foods and to rate each of them on a series of basic taste dimensions. To do this, the experimenter escorted the subject into the taste room and explained to her the procedure for tasting the foods and filling out the taste judgment forms by saying “As you can see, in front of you are nine different foods for you to taste. We’d like you to taste each food in order and then describe how it tastes to you by rating each food on this set of basic taste dimensions. What I mean is, start out with Food # 1 and taste it thoroughly. Then on the sheet marked # 1, rate the food on these taste dimensions by circling a number from 1 to 7, depending on how the food tastes on each dimension. For example, if Food # 1 tastes extremely sour, circle # 1, and if it tastes not at all sour, circle #7. If you think it’s somewhere in between, circle the number that most closely corresponds to your judgment. Only circle numbersdon’t circle words. Rate Food # 1 on all the dimensions. After you completely finish rating Food # 1, take a drink of water to clean out your mouth, then move on to Food #2 and Sheet #2. Continue in order for each of the nine foods”. After explaining
the taste procedure
the experimenter
said,
“It’s important that your ratings are as accurate as possible, so eat as much of each food as you need in making your judgments. After you’ve finished rating all foods, feel free to go back and eat as much as you want of any of them. Just make sure you taste all foods before nibbling”. Almost
as an afterthought
the experimenter
said in a joking
voice
“Hey, after you leave, I’m going to throw the foods out anyway, so you might as well eat what you want”.
236
D. J. BOWEN
The experimenter left the subject in the room for 20min to taste the foods. Extensive pilot testing and previous research found that subjects took approximately 7-8 min to complete their initial tasting, and then had approximately 12 min to go back and nibble on foods of their choice. Subjects took approximately 2-3 bites of each food when tasting and rating it. All foods were presented in bite-size portions in beige bowls. The order of foods was counterbalanced for food taste class. The goal of the session was to provide a naturalistic snacking period where subjects would feel comfortable to consume freely, in contrast to more rigorous taste perception procedures, such as rinsing between mouthfuls or foods or tasting several concentrations of the same taste or substance. Subjects believed that they were in a taste testing session. In actuality, the foods were weighed before and after the taste testing session to obtain measures of general and specific food consumption in the laboratory. This deceptive manipulation was necessary to prevent self-consciousness about eating freely and to obtain an accurate measure of laboratory food consumption. Amount consumed in the laboratory provided data on taste preferences during and after pregnancy, since amount of food consumed was selected freely. This method of measuring preferences for food taste classes was based on the work of Grunberg (1982), who used it to identify sweet taste preferences for smokers in cessation. Three sweet (Sara Lee pecan frosted coffee cake, Hershey’s milk chocolate bars, Vernell’s fruit gum drops), three salty (deli-style boiled ham, Planter’s salted peanuts, Mr Salty salted pretzel sticks), and three non-sweet, non-salty foods (Kraft low-fat mozzarella cheese, Nabisco unsalted saltine crackers, Planter’s unsalted peanuts) were chosen for the study. Sweet foods were chosen because of the importance of sweet taste in the regulation of food consumption in other literatures (Cabanac, 1978; Grunberg, 1982; Grunberg & Bowen, 1985; Roils, 1985). Salty foods were chosen to represent a second common taste that is non-sweet. In the present study non-sweet and non-salty foods were chosen to indicate changes in general food consumption. These foods were chosen on the basis of taste class only, and were not balanced for nutritional content, because this study addressed taste and not macronutrient consumption. In particular, carbohydrate and fat content were not manipulated independent of the two taste dimensions. The individual foods were chosen after extensive pilot testing for the following reasons: (1) they represented both snack and meal choices; (2) they represented the taste of each taste class (sweet, salty, and non-sweet/non-salty); (3) they were easy for subjects to handle; and (4) subjects liked them. Taste judgments For each food that the subject tasted, she was asked to complete one taste judgment sheet. On each sheet was a list of taste dimensions. The subject was asked to circle a number between 1 (extremely) and 7 (not at all) for each of the following taste dimensions: sweet, sour, salty, bitter, spicy, flavorful/bland, smooth/grainy, light/heavy, strong/mild, and fresh/stale. The judgments for each type of food were used as manipulation checks (i.e., did the “sweet” foods really taste sweet?) and as an evaluation of taste perception in the present study. The first four taste dimensions (sweet, sour, salty, bitter) represented standard taste qualities used in taste research. The remaining dimensions were inserted to give subjects plenty of opportunity to consume their preferred foods.
PREGNANCY,
FOOD AND
TASTE
231
Other Assessments
Height and weight were measured at the end of the laboratory session. A body mass index (weight/height2) was calculated for each woman. The body mass index has a high correlation with obesity as measured with more accurate but involved methods (Bray, 1980). In addition, body mass indices were computed for each woman using a weight taken at her first obstetrical visit. Two questionnaires were used to assess dietary restraint. One, developed by Herman & Polivy (1980), is a lo-item scale designed to measure behavioral restraint of eating. The other restraint scale is longer and contains three subscales, including conscious restraint in eating, behavioral and weight liability, and susceptibility to hunger (Stunkard & Messick, 1985). Although both scales are used in eating and obesity research, the relationship between these two scales is not well characterized in the literature and not at all in a pregnant population. The relationship between these two scales regarding changes in eating and weight associated with pregnancy was investigated by giving subjects both scales.
&XJLTS
Laboratory Consumption
Table 1 presents the result of the laboratory food consumption measurements combined into four pregnancy study periods: first trimester (weeks l-l 5), second trimester (weeks 16-27), third trimester (weeks 28-birth), and pospartum (all postpartum weeks). Total consumption of each food and food taste class (grams of sweet, salty, and non-sweet/non-salty food) and of each individual food is presented separately. Sweetness and saltiness ratings for each food and each food taste class are included in Table 1. Total number of subjects varied slightly due to measurement error. The quantity of food consumed varies across pregnancy periods, as shown in Table 1. Subjects in the second trimester consumed more grams of chocolate, coffee cake, and gumdrops than did the subjects in any other period. A one-way analysis of variance statistically confirmed that total amount of sweet foods consumed differed during the pregnancy periods [F(3,46) = 9.72, p c O.OOOl].Using a Newman-Keuls range test to compare individual groups, women in the second trimester consumed significantly more sweet food than did women from any other group. No other individual group comparisons were significantly different. Consumption of other foods also differed throughout pregnancy. Salty peanuts were consumed more in the third trimester, compared with other pregnancy periods. When grams of all salty foods were summed women in their third trimester of pregnancy consumed marginally more grams of salty food than did subjects in either of the other trimesters or during the post-partum period [F(3,45) = 2.24, p < 0.0981. There were no significant differences in total non-sweet/non-salty food consumption. Table 1 also presents the taste perception data for all foods. Subjects rated all three sweet foods as being very sweet. There were no significant differences in the subjects’ taste ratings of the sweetness or saltiness of the sweet foods across the pregnancy periods. Subjects in the third trimester and during the postpartum period rated their salty foods as less salty overall than did subjects in either of the other two periods. The overall differences in saltiness ratings were statistically significant using
Non-sweet/Non-salty Mozarella cheese Unsalted saltines Unsalted peanuts
Salted peanuts Salted pretzels Boiled ham
Salty taste
Chocolate Coffee cake Gumdrops
Sweet taste
Total
taste
Total
Total
18.5
10.1 2.6 5.8
33.7
11.8 18.0 3.2
44.0
17.7 18.3 8.0
g
5.4 5.5 6.1
5.8 6.1 5.5
1.4 2.6 1.9
Rating
Trimester
1
5.8 6.2 6.8
1.1 1.8 2.5
6.1 5.7 6.3
Rating
19.5
11.4 1.4 6.7
31.8
11.0 16.7 4.1
72.0
32.8 26.7 12.5
g
Food consumption
TABLE 1
5.9 6.0 6.2
5.5 6.0 5.9
1.4 2.7 1.2
Rating
Trimester
2
5.8 5.3 6.2
1.3 1.6 2.4
5.9 5.7 6.2
Rating
data across pregnancy
24.2
11.3 1.9 11.3
52.1
28.9 17.6 4.5
53.1
27.8 19.2 6.1
g
4.9 5.6 6.0
5.9 6.0 5.7
1.3 0.7 1.5
Rating
Trimester
study periods 3
5.5 5.4 6.2
2.6 2.8 3.1
5.4 5.9 6.3
Rating
21.3
7.9 3.3 9.8
223
13.8 13.0 5.2
60.0
21.1 21.0 7.9
g
5.1 6.0 5.9
6.2 6.1 5.4
1.7 2.9 1.8
Rating
Postpartum
5.8 5.7 6.5
2.5 2.2 2.1
5.9 5.8 6.4
Rating
PREGNANCY,
239
FOOD AND TASTE
a one-way analysis of variance [F(3,45) = 4.80, p < O*Ol]and a Neuman-Keul’s test confirmed that saltiness ratings were lower in the third trimester and the postpartum period. The salty foods were rated as being not sweet. Assumption about the taste of the non-sweet/non-salty foods were confirmed, as those foods were rated not sweet and not salty. Total laboratory consumption shown as kilocalories consumed is presented in Figure 1. Subjects in the second trimester consumed more calories in the laboratory than did subjects in any other period. The group differences were statistically significant using an analysis of variance [F(3,45) = 2.8 1, p < 0.051. A Newman-Keuls range test confirmed that women in the second trimester consumed significantly more kilocalories than did any other group. No other group differed significantly. This difference in caloric consumption differed from the similarity of consumption values when expressed as grams of food consumed. Over the four periods, laboratory food consumption computed in grams of food was not significantly different, although subjects in the second and third trimester consumed slightly more grams of food than did subjects in the other two pregnancy periods. As expected, weight gain was higher from first through third trimester for subjects in the present study. The weight gains were approximately equal to guidelines set by the American College of Obstetrics and Gynecology. Data for the postpartum subjects showed a slight weight gain (roughly 71b) from pre-pregnancy levels that remained after the birth of the child. Body mass index (weight/height2) values for subjects at time of study participation paralleled weight gain data. In contrast to BMI’s at the time of study, BMI’s calculated using body weight from the first obstetrical visit were virtually identical for women in each of the study periods. Subjects were generally restrained eaters, as shown by their scores on the two restraint questionnaires, the Stunkard and the Herman and Polivy (mean for the Stunkard scale = 31.32, range =0-51; mean score in this range is high for normal population. Scores on the HP restraint scale were generally high for normal individuals). There were no differences in scores for any of the scales among the
50C
400 %
Y
$g I 5
300
f
200
3 100
-
0
-L
2nd
L
3rd Tnmesters
FIGURE1.
Total calorie consumption
across pregnancy.
PP
D. J. BOWEN
measurement periods. The Stunkard restraint scale was positively, significantly, and highly correlated with scores on the Herman and Polivy restraint scale (r=0.76; p c O-01). This relationship was expected, because the Stunkard scale was developed, in part, using items from the Herman and Polivy scale.
DISCUSSION
The results of the present study suggested that during the course of pregnancy, preference for certain foods varied. Women in the second trimester consumed more sweet-tasting food than did women in any other pregnancy study period. There were no significant differences in the sweetness ratings of the sweet foods for any group of subjects. Therefore, the consumption difference was not due to changes in the perception of sweet taste as measured in the present study. Some of the sweet foods, however, were also high in fat as well as sweet taste. It is not possible from the present selection of foods to compare fat consumption, fat preference, or fat perception taste changes during pregnancy. This problem invites further investigation, because of the importance of fat consumption and fat preference in human obesity (Drewnowski & Greenwood, 1982). Previous research reported no differences in sweet preferences during pregnancy (Brown & Toma, 1986); however, most of the subjects in the previous study were in the third trimester. Changes in sweet preference were found in the present study only in the second trimester. The second trimester is the time of most dramatic changes in levels of hormones, particularly estrogens and progestins, that are responsible for body weight and food consumption regulation in the nonpregnant state (Fuchs & Klopper, 1983; Czaja, 1983). These hormones could be responsible for changes in sweet food consumption in the second trimester period. Consumption of salty foods also differed for women across pregnancy periods. Women in the third trimester consumed marginally more salty foods than did any other group of women in the present study, a finding supported by previous research (Brown & Toma, 1986). This behavioral change corresponded to a change in the taste perception of the salty foods for women in the third trimester, who rated those foods as less salty compared with the ratings of other women. This is a noteworthy point, because in the later trimesters of pregnancy, women retain more fluids and are vulnerable to eclampsia, or pregnancy-induced hypertension. Whether or not consumption of salty foods is involved in this serious problem is unknown from the present study, since all subjects were healthy. Women with eclampsia could experience a change in taste perception for salty foods, consume more of these foods, and so unknowingly assist in the development of this obstetrical problem. There were significant differences in the number of calories consumed across pregnancy. The increase in caloric consumption in the second trimester resulted from the increase in consumption of certain foods used in the study, such as coffee cake and candy. This increased consumption of sweet-tasting and/or high-fat foods might contribute to excess weight gains if continued in daily food consumption. Subjects in the present study were restrained eaters, as shown from their scores on both restraint scales. Although women in the second trimester exhibited an increase in preference for and consumption of sweet foods in the present laboratory study they did not gain excessive weight during their pregnancies. The answer to this puzzle may lie in the relatively high levels of restraint reported by these women.
PREGNANCY,
FOOD AND TASTE
241
Possibly, these women were so restrained that they would hold back from eating foods so that they would not consume increased amounts for fear of gaining weight. For other women who were not so restrained, an increase in preference might lead to an increase in daily consumption, and so to increased weight. An attempt was made in the laboratory study to allow the women to feel as comfortable and relaxed as possible. They might have eaten more sweet foods in the present study because of a second-trimester preference change, but would restrain themselves from eating extra amounts of food when in a more familiar environment. This possibility needs to be investigated both using naturalistic methods of actual daily food consumption collection and in other populations that might not be as restrained as were women in the present study. This change in food preference and consumption may have implications for proper nutritional intake during pregnancy. Women who are not educated as to proper prenatal nutrition may be allowing their preferences to guide their food choices during pregnancy. If laboratory sweet preferences do guide daily food consumption, some women may not be selecting their diets to consist mainly of nonsweet foods that are very nutritious. Low income and minority populations that might not have access to adequate prenatal or nutritional care and advice are the same groups that gain the most excess pregnancy-associated weight. These populations also have the highest incidence of low-birthweight infants. The role of taste preference changes, dietary intake, and nutritional advice in the incidence of lowbirthweight infants remains to be investigated. The present study has documented taste preference changes during the second trimester of pregnancy. More research relating normal and excess weight gains in pregnant populations to food preferences and daily eating habits must be conducted. An understanding of the role of taste in daily consumption changes and weight gains during pregnancy could provide a model for psychobiological determinants of weight regulation, in addition to improving the health of women who are at risk of obesity due to pregnancy.
Beazley, J. M. & Swinhoe, J. R. (1979) Body weight in parous women: Is there any alteration between successive pregnancies? Acta Obstetrica Gynecologia Scandinavia, 58, 45-47. Bray, G. A. (Ed.) (1980) Obesity in America. USDEW (PHS) NIH. Brewin, T. B. (1980) Can a tumor cause the same appetite perversion or taste change as a pregnancy? L..ancet, 907-908. Brown, J. E. & Toma, R. B. (1986) Taste changes during pregnancy. American Journal of Clinical Nutrition, 43, 414-418.
Cabanac, M. (1978) Sensory pleasure. Quarterly Review of Biology, 54, l-29. Czaja, J. A. (1983) Body weight and growth rates throughout the guinea pig pregnancy: Evidence for modulation by endogenous estrogens. Physiology and Behavior, 30, 225-231.
Dobbing, J. (Ed.) (1981) Maternal nutrition: Eating for two? New York: Academic Press. Drewnowski, A. & Greenwood, M. R. C. (1983) Cream and sugar: Human preferences for high-fat foods. Physiology and Behavior, 30, 629-633. Fuchs, F. & Klopper, A. (Eds) (1983) Endocrinology of Pregnancy. Philadelphia: Harper & Row. Grunberg, N. E. (1982) The effects of nicotine and cigarette smoking on food consumption and taste preference. Addictive Behaviors, 7, 317-331.
D. J. BOWEN
242 Grunberg,
N. E. & Bowen, D. J. (1985) Coping with the sequelae of smoking cessation.
American Journal of Cardiac Rehabilitation, 5, 285-289.
Heliovaara,
M. & Aromaa,
A. (1981) Parity and obesity. Journal of Epidemiology and
Community Health, 35, 197-199.
Herman, C. P. & Polivy, J. (1980) Experimental and clinical aspects of restrained eating. In A. Stunkard (Ed.), Obesity. Philadelphia: W. B. Saunders. Holinka, C. (1980) Age-related weight changes in female G57BL/6J mice: nonpregnant body weight and extra-fetoplacental maternal weight throughout pregnancy. Mount Sinai Journal of Medicine, 47, 104-l 10. Hook, E. B. (1980) Influence of pregnancy on dietary selection. International Journal of Obesity, 4, 338-340.
Hook, E. B. (1978) Dietary cravings and aversions during pregnancy. American Journal of Clinical Nutrition, 31, 1355- 1362.
Hytten, F. E. (1981). Weight gain and pregnancy-30
years of research. South Africa Mediese
Tydskrif 15- 19.
Kawakami, S., Ishiwata, C., Hayashi, K., Kawaguchi, Y., Kondo, M. & Iizuk, R. (1977) Alteration of maternal body weight in pregnancy and postpartum. Keio Journal of Medicine, 26, 53-62.
Naeye, R. L. (1979) Weight gain and the outcome
of pregnancy.
American Journal of
Obstetrics and Gynecology, 13.5, 5-9.
Naismith, D. J. (1980) Maternal nutrition and the outcome of pregnancy-a critical appraisal. Proceedings of the Nutritional Society, 39, l-l 1. Newcombe, R. G. (1982) Development of obesity in parous women. Journal of EpidemioZogy and Community Health, 36, 306-309.
Niswander, E. L. & Gordon, J. J. (1972) The women and their pregnancies. New York: Academic Press. Picone, T. J., Allen, L. H., Schramm, M. M. & Olseny, P. N. (1982) Pregnancy outcome in North American women. I. Effects of diet, cigarette smoking, and psychological stress on maternal weight gain. American Journal of Clinical Nutrition, 36, 1205-1213. Rodin, J. (1987) Weight change following smoking cessation: the role of food intake and exercise. Addictive Behaviors, 12, 303-319. Rolls, B. J. (1985) Experimental analyses of the effects of variety in a meal on human feeding. American Journal of Clinical Nutrition, 42, 932-939.
Spitzer, L. & Rodin, J. (1981). Human eating behavior: A critical review of studies in normal weight and overweight individuals. Appetite, 2, 293-329. Strubbe, J. H. & Gorissen, J. (1980) Meal patterning in the lactating rat. Physiology and Behavior, 25, 775-777.
Stunkard, A. J. & Messick, S. (1985) The three-factor eating questionnaire to measure dietary restraint, disinhibition, and hunger. Journal of Psychosomatic Research, 29, 71-84. Trethowen, W. H. & Dickens, G. (1972) Cravings, aversion, and pica of pregnancy. In J. G. Howells (Ed.), Modern perspectives in psychiatry (Vol. 5): Modern perspectives in psychoobstetrics, Edinburgh: Oliver 8z Boyd, pp. 251-268. Wade, G. (1976) Sex hormones, regulatory behavior, and body weight. In J. Rosenblatt (Ed.), Advances in the study of behavior. New York: Academic. Woods, D. L., Dewar, R., Heese, A. F. & Rush, R. W. (1980) Maternal post-delivery weight in the assessment of obesity. South Africa Mediese Tydskrtf, 500-501. Worthington-Roberts, B., Little, R. E., Lambert, M. D. & Wu, R. (1989) Dietary cravings and aversions in the postpartum period. Journal of the American Dietetic Association, 89, 647-651.
Zuckerman, M. (1960) The development of an Affect Adjective Check List for the measurement of anxiety. Journal of Consulting Psychology, 24, 57-462. Received 9 September 1988, revision IS April 1989
