A model for making outcomes research standard practice in clinical dietetics.

In today's rapidly changing health care environment, it is imperative that dietitians demonstrate the importance of their role as health care providers by initiating and participating in outcomes research. Patient care should be based on empirical studies, and clinical dietitians should participate in or be at the helm of such investigations. Nutrition research is usually accomplished and reported by persons with MD or PhD degrees who do not have clinical training in nutrition and does not often address situations encountered by dietitians in clinical settings. This article examines the extent to which clinical dietitians are conducting and reporting outcomes research, their attitudes about such research, and how they think they could best acquire the skills needed to incorporate outcomes research into their practice. Results indicate that clinical dietitians are not writing a substantial percentage of articles and that reports of outcomes research are not commonly included in clinical nutrition journals. Clinical dietitians appear to value the inclusion of research in clinical practice, but they are not spending a great amount of time doing research. Dietitians report being comfortable about participating in research but not about designing, conducting, and reporting research. Results from several surveys were used to develop a model for integrating research in clinical practice. This model includes the research process as a basis for clinical practice. A model for collaborative efforts between clinical and academic dietitians is also proposed and emphasizes the responsibility of academic dietetics training programs in the integration of research and clinical practice.

Are weight concerns predictive of smoking cessation? A prospective analysis.

Participants in an 8-session, community based smoking cessation intervention rated whether they would stay quit if they experienced weight gain. The majority reported that they would not relapse to smoking, even after a 20-lb, (9.07-kg) weight gain. Those who were weight concerned were more likely to be female, to weight less and be normal or underweight, and to report chronic dieting. This group was also significantly less likely to be abstinent posttreatment, and at the 1-, 6- and 12-month follow-ups. Individuals presenting for formal smoking cessation interventions may be less weight concerned than the general population of smokers. However, weight-concerned smokers who do present for treatment are less likely to quit smoking. Implications for recruitment and intervention are discussed.

How much weight gain occurs following smoking cessation? A comparison of weight gain using both continuous and point prevalence abstinence.

Estimates of postcessation weight gain vary widely. This study determined the magnitude of weight gain in a cohort using both point prevalence and continuous abstinence criteria for cessation. Participants were 196 volunteers who participated in a smoking cessation program and who either continuously smoked (n = 118), were continuously abstinent (n = 51), or who were point prevalent abstinent (n = 27) (i.e., quit at the 1-year follow-up visit but not at others). Continuously abstinent participants gained over 13 lbs. (5.90 kg) at 1 year, significantly more than continuously smoking (M = 2.4 lb.) and point prevalent abstinent participants (M = 6.7 lbs., or 3.04 kg). Individual growth curve analysis confirmed that weight gain and the rate of weight gain (pounds per month) were greater among continuously smoking participants and that these effects were independent of gender, baseline weight, smoking and dieting history, age, and education. Results suggest that studies using point prevalence abstinence to estimate postcessation weight gain may be underestimating postcessation weight gain.

Differences in macronutrient selections in users and nonusers of an oral contraceptive.

One of the problems inherent in using women in clinical research is the effect that oral contraceptive (OC) use might have on physical indexes. Although weight gain is frequently reported as a side effect of OC use, there is little empirical evidence that such weight gain actually occurs. The current study investigated differences in energy balance [ie, dietary intake, resting energy expenditure (REE), and physical activity] between groups of users and nonusers of OCs. Each group completed a protocol that covered one menstrual cycle and consisted of daily recording of dietary intake, measurement of REE once during each phase of the menstrual cycle, and reporting of physical activity over the entire cycle. Comparisons indicate that there was a marginal interaction (P = 0.06) of OC use with total energy intake, indicating a different pattern of intake for the two groups. There were qualitative between-group differences such that the OC group consumed a greater percentage of energy as fat (P = 0.02) and a lesser percentage of energy as carbohydrate (P = 0.008). No group differences were found in the percentage of energy consumed as protein, but both groups consumed significantly less protein during menses (P = 0.008). There were no significant differences in REE. Both groups of women reported marginally more activity (P = 0.09) during menses than during the luteal phase.

Changes in food consumption and body weight associated with smoking cessation across menstrual cycle phase.

Forty women smokers were randomly assigned to smoking cessation for a 10-day period of time, either during the follicular or the luteal phase of their menstrual cycle. Measurements of dietary intake and body weight were collected during the same phase the previous (smoking as usual) month and during the cessation phase. Physical activity was controlled across the 2 months. Comparisons between the smoking and cessation months indicated that both groups increased their dietary intake during the cessation month, but no interaction occurred between phase and month. That is, women in both groups increased dietary intake to the same degree. All energy nutrients tested (fat, complex carbohydrates, and sugar) increased significantly from baseline to cessation. Conversely, body weight increased by 1.8 kg (4 lb) in the Luteal group, while weight remained stable (0.1 kg change) in the Follicular group.

The relationship between a history of depression and adherence to a multicomponent smoking-cessation program.

The purpose of this study was to determine whether a history of depression in female smokers (age 18-65) who did not self-report any current depression was associated with adherence to a multisession, multicomponent smoking-cessation program. Participants in a 13-week cognitive-behavioral group program plus random assignment to nicotine gum, appetite suppressant gum, or placebo chewing gum were grouped by depressive-history and compared on attendance, average expired carbon monoxide after planned cessation, and number of pieces of gum chewed. No significant differences between the depressive history (yes/no) subgroups were found on any of the three measures of adherence. The power to detect a significant difference (alpha = 0.05) was calculated to be 0.89. Group cognitive-behavioral treatment appears to be the basis of an effective smoking-cessation program for women with a history of depression who are not currently depressed.

Precision and estimated accuracy of two short-term food frequency questionnaires compared with recalls and records.

Two widely used food frequency questionnaires (Block FFQ, Willett FFQ) were modified to reflect intake over the past 7 days and compared to intake information gathered from diet records and 24-hr recalls covering the same 7-day period. The Block FFQ and the Willett FFQ were also gathered at the beginning of the assessment period to reflect the 7-day period of time before records and recalls were gathered. Fifty-six subjects were assigned to either recording diet for 3 days, recording diet for 6 days, or providing three 24-hr recalls. Results indicate similar levels of within-method test-retest reliabilities for 3-day RECORDS and 6-day RECORDS, and within the two Block FFQs and within the two Willett FFQs from each subject, while lower reliabilities were seen in 24-hr RECALLS. When the FFQs were compared to the 6-day RECORDS with between-method agreement coefficients, there was a moderate level of agreement, with most values between 0.5 and 0.8 for both FFQs. Significant differences between mean levels of nutrients estimated by the three methods indicated differences only in the estimates of carbohydrate and vitamin A. The use of FFQs to gather short-term intake information is discussed.

Predictors of waist circumference change in healthy young adults.

OBJECTIVE: To identify predictors of a change in waist circumference in a group of healthy young adults. SUBJECTS: Caucasian, 121 women and 109 men, participating in a longitudinal investigation of cardiovascular risk factors in parents and their young children. MEASUREMENTS: Evaluations of body weight, waist and hip circumferences, dietary intake, physical activity, cigarette smoking, and alcohol intake were performed annually. Age was recorded and family history of disease was assessed. RESULTS: For women, covariates and modifiable predictors accounted for 67% of the variance in waist circumference change from Year 1 to Year 3. Women with lower baseline waist girths, lower baseline hip girths, higher baseline body weight, and a greater change in body weight had larger increases in waist girth. For men, covariates and modifiable predictors accounted for 72% of the variance in waist circumference change. Men with lower baseline waist girth, a greater change in hip girth, higher baseline body weight, greater increases in body weight, and less percent of fat in the diet at baseline had larger increases in waist girth. Other non-modifiable variables did not predict change in either gender. CONCLUSION: Reducing excess body weight and decreasing weight gain appear to be the most important factors in preventing the accumulation of upper body fat.

Who underreports dietary intake in a dietary recall? Evidence from the Second National Health and Nutrition Examination Survey.

The present study sought to identify the presence and degree of apparent underreporting of dietary intake in 11,663 participants in the Second National Health and Nutrition Examination Survey (NHANES II). Self-reported dietary intake was compared with estimated basal metabolic rate. Underreporting was based on cutoff limits that identified plausible levels of energy expenditure for adult individuals. Results indicated that up to 31% of adults in this sample may have underreported dietary intake. Those individuals at greatest risk of underreporting were less well educated and heavier. The Sex x Race interaction indicated that for both ethnic categories, women were more likely to underreport than men, but the difference between men and women was greater among Caucasian participants. It is concluded that such factors as reduced energy needs, deliberate falsification, and measurement error inherent in dietary assessment contribute to apparent underreporting, and this occurs in a large percentage of dietary data.

A longitudinal analysis of accelerated weight gain in preschool children.

The purpose of the current investigation was to determine the dietary, physical activity, family history, and demographic predictors of relative weight change in a cohort of 146 children over a 3-year period. Results indicated that boys of normal-weight parents or who had only one parent overweight showed decreases in their body mass index (BMI) while those with two parents overweight showed increases. Girls with an overweight father showed BMI increases while others experienced decreases in BMI. Additionally, baseline intake of kilocalories from fat as well as decreases in fat intake were related to decreases in BMI. At higher levels of baseline aerobic activity, subsequent changes in BMI decreased. There was also a trend for changes in leisure activity--increases in children's leisure activity was associated with decreases in subsequent weight gain. Modifiable variables (ie, dietary intake, physical activity) accounted for more of the variance in changes in child BMI change than nonmodifiable variables (eg, number of parents obese). These results strongly suggest that encouragement of heart healthy dietary intake patterns and participation in physical activity can decrease accelerated weight gain and obesity, even in preschool children.

Physicians' diagnosis of obesity status in NHANES II.

The aim of this work was to assess the accuracy of physicians' subjective assessments of obesity status. The subjects were participants in The Second National Health and Nutrition Examination (NHANES II) Survey. The physicians' subjective judgments of obesity were compared to BMI, an objective measure of actual body mass. Subjects with a body mass index (BMI = weight in kg/(height in cm/100)2) less than or equal to 27.5 were classified as normal weight and those with a BMI greater than or equal to 30.4 were considered to be obese. Physicians were accurate in their diagnosis of the normal weight group with only 4.03% being misdiagnosed as obese. However, 12.6% of the obese group was misdiagnosed as normal weight. The odds of an incorrect normal weight diagnosis increased with age. Similarly, as the fat distribution ratio increased, i.e., a more central pattern, the odds of being actually obese but incorrectly diagnosed as normal weight increased. Men were more likely than women to be incorrectly diagnosed as normal weight. Non-Caucasian normal weight persons appear to have been diagnosed more stringently than Caucasians as they were more likely to be misdiagnosed as obese regardless of their gender. There appear to be several variables affecting the physicians' subjective assessment of obesity status in this data set.

Relationship between psychosocial functioning and body fat in preschool children: a longitudinal investigation.

This study examined whether preschool children differed on measures of psychosocial functioning both cross-sectionally and longitudinally. One hundred and thirty-two children who varied in levels of body fat participated in the study along with their natural parents. Results indicated that the children did not differ in levels of self-esteem and family functioning as a function of their body fat. Prospectively, physical self-esteem weakly (but significantly) correlated with body fat at 1 and 2 years, and father's perception of family functioning predicted body fat at 1 year only. Results suggested that childhood obesity may not develop as a result of psychosocial factors.

A prospective study of familial aggregation of blood pressure in young children.

In a prospective study, we evaluated familial aggregation of blood pressure in a sample of 175 normotensive families with children 3 to 6 yr old. Systolic and diastolic blood pressure measurements of parents and children were correlated at 1, 2, and 3 yr intervals. Spearman rank-order correlation coefficients between parents and children were significant for mother-son pairs. In particular, mother-son blood pressure correlation coefficients were significant for systolic blood pressure across all 3 yr and for diastolic blood pressure during yr 2. Further analyses were performed adjusting for body mass index (BMI), age, physical activity, sodium intake, potassium intake, and parental smoking status, and alcohol use. The Spearman correlation coefficients for mother-son pairs remained significant for yr 1 and 3 after adjusting for these blood pressure correlates. These results are consistent with cross-sectional studies and suggest that both genetic and environmental contributions to blood pressure status are important in young children.

Impact of diabetic status, dietary intake, physical activity, and smoking status on body mass index in NHANES II.

Individuals with non-insulin-dependent diabetes mellitus (NIDDM) have a higher body mass index (BMI) than nondiabetic individuals and it has been reported that individuals with NIDDM have more difficulty losing excess weight. To assess an independent effect of NIDDM on weight status, analyses were conducted to predict BMI with variables previously associated with weight status. Subjects from the Second National Health and Nutrition Examination Survey were categorized into three groups: individuals with NIDDM and on oral medication, individuals with diet-controlled NIDDM, and nondiabetic adults. Additionally, planned orthogonal comparisons between these groups were conducted after other predictors were controlled for; 9.18% of the variability was explained for men (P less than 0.001) whereas 12.11% was explained for women (P less than 0.0001), with several variables significantly related in each model. An independent effect for diabetes status was found in both sexes and there were no differences in BMI between NIDDM groups in either sex.

Calcium intake in youth: sex, age, and racial differences in NHANES II.

METHODS: Data from the NHANES II survey were used to determine whether the calcium intake of children ages 3-18 met the minimum recommended daily allowance. In addition, the impact of age, race, sex, and socioeconomic status on calcium intake and the density of calcium intake was explored. RESULTS: The results indicated significant differences between groups of children based on age, sex, and race (P less than 0.05). Males across all age groups (3-5, 6-7, 8-10, 11-12, 13-18) had a higher calcium intake than females. In addition, calcium intake levels among males always met or exceeded the minimum RDA. Levels among all blacks started below the minimum RDA in the age group 3-5. During the adolescent years, calcium intake again fell below the RDA for black males and for females of both races, with the calcium intake of the adolescent females resembling that of the group ages 3-5. In a multiple regression model sex, age, race, and total energy intake predicted 49% (P less than 0.0001) of the variance in calcium intake. Socioeconomic status (using parents' reported income) was not a significant predictor. Another regression model was used to examine the relationship among age, race, sex, and socioeconomic status and its effect on the density of calcium intake (mg/kcal). While the overall model was significant (P less than 0.0001) it accounted for only 3.5% of the variance. CONCLUSION: Age, sex, and race of the child significantly predicted the density of calcium (P less than 0.05); however, parents' gross income did not (P less than 0.81), indicating that income level does not contribute to explaining these differences.

A longitudinal analysis of the impact of dietary intake and physical activity on weight change in adults.

The current investigation is a longitudinal analysis of the relationship between dietary intake, physical activity, and body weight change in adult men (n = 142) and women (n = 152). Measures of dietary intake, physical activity, and cigarette and alcohol consumption were obtained for 3 y. Results indicated a different pattern of predictors of weight change for men vs women. For women a high dietary energy and fat intake as well as increases in total energy intake were related to higher weight gain and increases in work activity levels were related to decreased weight gain. For men weight gain was predicted by increases in dietary fat intake. Sex differences are discussed as a possible moderator variable in the energy balance equation.

Children at familial risk for obesity: an examination of dietary intake, physical activity and weight status.

A large cohort of children was divided into two groups based on their parents' weight status. The high risk group had one or two overweight parent(s) (n = 92, 4.4 +/- 0.5 years), while the low risk group had no parent overweight (n = 95, 4.5 +/- 0.5 years). Weight of the two groups was similar at the start of the study but the high risk group gained marginally more weight over a one year period (P = 0.05). Although total energy intake was similar, the high risk group was consuming a larger percentage of energy from fat (P = 0.0004) and a smaller percentage from carbohydrate (P = 0.0002). Observed physical activity as similar for most levels but marginally higher for the stationary level and marginally lower in total activity in the high risk group. These results suggest a pattern that may lead to increased weight gain in a high risk group as they grow older.

Relationships among smoking status, body composition, energy intake, and physical activity in adult males: a longitudinal analysis.

This study evaluates the longitudinal relationships among smoking and adiposity, dietary intake, and physical activity in a group of adult males. Subjects were 101 nonsmokers and 19 regular cigarette smokers. Adiposity, dietary intake, and physical activity were assessed annually for three consecutive years. Results indicated that nonsmokers had larger tricep skinfold measurements than smokers over the 3-year period. However, dietary intake and physical activity did not differ between groups. Due to the absence of differences on these two variables for the smoking and nonsmoking groups, it was concluded that the lower adiposity in smokers was largely metabolically determined, which may make difficult the treatment of postcessation weight gain.

Measuring short-term dietary intake: development and testing of a 1-week food frequency questionnaire.

Measures typically used to assess short-term intake are 24-hour recalls or records of food intake. However, these methods require a large time commitment from both researchers and subjects. Our study sought to develop a more time-efficient method of assessing short-term intake by developing a 7-day food frequency questionnaire (7DFFQ). We modified Willett's 1-year food frequency questionnaire to assess intake for a 7-day period. Content validity has previously been exhibited in Willett's work. We assessed criterion validity by comparing the 7DFFQ with 24-hour recalls. Twenty-one female and 20 male college students were asked to recall their previous day's intake three times during a 1-week period. After the last recall, subjects completed a 7DFFQ covering the same week. Data were analyzed by comparing each recall and the mean of the three recalls with values obtained on the 7DFFQ. Analyses revealed high correlations between the mean of the three recalls and the 7DFFQ. Further analysis ranked subjects into quartiles for each nutrient on both measures. Most subjects changed either no quartiles or only one quartile for total energy and each nutrient. Test/retest reliability was assessed by administering the 7DFFQ to a group of women for two consecutive weeks and by administering the 7DFFQ to a group twice 3 hours apart. Pearson's r and intraclass correlations for the two administrations show moderately high reliability over the 1-week period and considerable improvement with the 3-hour administration. The 7DFFQ appears to meet several of the criteria for a measure of dietary intake.