Stark, Lori J.; Mulvihill, Mary M.*; Powers, Scott W.; Jelalian, Elissa; Keating, Kristin*; Creveling, Susan*; Byrnes-Collins, Barbara; Harwood, Ivan*; Passero, Mary Anne†; Light, Michael*; Miller, Deborah L.; Hovell, Melbourne F.‡
Despite treatment advances in cystic fibrosis (CF), inadequate weight gain continues to be problematic for most patients; ≈40% of patients fell below the fifth percentile of weight for age in 1991 (1). Nutritional status has been found to be highly correlated with pulmonary functioning and to affect recovery from acute infections (2,3). Inadequate weight gain in CF is hypothesized to be attributable to an energy imbalance (4) due to the increased calorie demands caused by malabsorption, chronic respiratory disease, and possibly an elevated metabolic rate (5).
Treatment recommendations for CF patients include consumption of 120-150% of the recommended daily allowance (RDA) of calories for healthy individuals (6), with a normal to high fat (40%) intake (7) to offset increased energy requirements. When malnourished patients have been provided with this increased level of calorie intake via medical interventions, such as artificial diets (8) and parenteral (9,10) and enteral supplementation (11,12), they have shown improved weight gain pre- to posttreatment. The positive results of these studies have prompted increased attention to improving the nutritional status of all CF patients via prevention and early intervention (13).
Unfortunately, while patients receiving medical intervention have shown improved nutritional status, these interventions are not applicable to the population at large because of their intrusiveness and expense. However, without treatment it appears that many children with CF are unable to meet their recommended calorie intake. In recent studies examining the intake of preschool (14) and school-age (15) children with CF, most were found to be consuming only 100% of the RDA for their age. Furthermore, compliance studies have consistently found that diet is one of the most problematic aspects of CF treatment (16,17), and many patients and their parents do not understand that diet is a formal treatment recommendation (17).
While nutritional education appears to be important (18), it has not been effective in increasing CF children's calorie consumption to the recommended level (18,19). Investigations of mealtime interactions of children with CF and their parents indicate that certain behavioral and environmental factors may be contributing to the inadequacy of nutritional education. Parents of children with CF have reported that their children engage in behaviors incompatible with eating, such as feeling full early in the meal, complaining of abdominal pain, eating slowly, and talking instead of eating (i.e., dawdling) (20,21). Children with CF take twice as long to eat dinner as their healthy peers (14), and a significant negative correlation has been found between the number of problems reported by parents at meal-times and the calorie intake of children with CF (21). Quittner et al. (22) reported that mealtime problems were the issue most commonly cited by parents of preschool children with CF.
Behavioral interventions targeting these problematic mealtime interactions have been advocated in the treatment (23) and prevention (13) of inadequate calorie intake and malnutrition in CF patients. Results of studies conducted thus far are encouraging. Stark and colleagues (20,24) have used a 6-week behavioral intervention in which children with CF and their parents are seen in independent parent/child treatment groups. The behavioral intervention provided nutrition education and behavioral management training that specifically taught parents how to motivate their children to eat more. In each study, subjects acted as their own controls and improved calorie intake by 25-43% (20) and 32-60% (24), respectively. The children also showed significant increases in weight over the 6 weeks of treatment, with an average gain of 1.48 and 0.66 kg, respectively (20,24). Of equal importance was the maintenance of these treatment gains at 9 months (20) and 2 years (24) posttreatment. Most medical interventions are effective only as long as the patient is receiving treatment: a return to baseline typically occurs 6 months to 1 year after the withdrawal of treatment (10).
To date, most studies have used subjects as their own controls pre- to posttreatment (20,24-26) or compared treatment subjects to patients identified as being similar in age, sex, height and pulmonary function via retrospective chart review (12,27). The purpose of the present study was to replicate the behavioral treatment protocol of Stark and colleagues (20,24) using a wait list control group of children with CF as a comparison to the children receiving treatment. The control group was prospectively identified and assessed on all treatment measures at the same points in time as the treatment group, allowing for control of the experimental demands of diet diary recording and increased health care contact with regard to assessment of weight gain.
MATERIALS AND METHODS
Nine children with CF and their parents were recruited to participate in a treatment study to improve nutritional status in CF patients. Subjects were recruited from two CF centers, Rhode Island Hospital (RIH) and the University of California, San Diego (UCSD) Medical Center. Eight of the nine families had participated in an assessment study on eating behaviors (28) and were interested in receiving treatment. Four subjects were recruited from RIH, and five subjects were recruited from UCSD Medical Center. The study protocol was approved by the Human Subjects Protection Committee from RIH and UCSD, and signed consent forms were obtained from the parents of all participants in accordance with the human subjects committee guidelines.
The nine subjects were randomly assigned to either a behavioral intervention or a wait list control group. All families were told they would be offered intervention and that half would receive intervention first. The other half of the families would be offered intervention 3 months later but would be asked to participate in data collection (as described herein) at two points in time before treatment. After random assignment to experimental conditions, four families' assignment had to be changed because of conflicts with scheduled vacations.
Five children were assigned to the treatment group, two from RIH and three from UCSD Medical Center. The children ranged in age from 5 years, 3 months, to 10 years, 1 month, with a mean age of 7 years, 3 months (SD = 1.7). Their weight percentile for age ranged from the fifth to the 30th, with a mean of the 18th weight percentile for age (SD = 9.7). Using the Hollingshead two-factor index of socioeconomic status (SES), the average SES ranking for the five families was category III—skilled laborers or clerical and sales workers (29).
Four children were assigned to the control group, two from RIH and two from UCSD Medical Center. The children ranged in age from 3 years, 8 months, to 10 years, 1 month, with a mean age of 6 years, 3 months (SD = 2.7). There was no significant difference between treatment and control groups in age. Their weight percentile for age ranged from below the fifth to the 15th, with a mean of the ninth percentile weight for age (SD = 4.8). Since it is difficult to compare subjects on weight percentiles across ages, the children's weights were standardized for age and sex using Z scores for weight to determine their equivalence before treatment. The mean and the SD for Z scores are always 0 and 1, respectively (30). The mean Z score was - 1.188 for the treatment group and -1.715 for the control group. The groups were not significantly different on their Z scores for weight before treatment [t(8) = 1.722, p = 0.13]. The average SES category for the control group was II—owners of medium-size businesses or technical workers (29).
RIH families were seen at RIH. The families at UCSD were seen on the campus of San Diego State University (SDSU) owing to space constraints at the UCSD Medical Center. At each site, the parents and children were seen simultaneously once a week in separate groups. The parents' group met in a conference room with a clinical psychologist and a dietitian. The children were seen in a similar conference room or classroom. The children were seen by a postdoctoral fellow in pediatric psychology or a graduate student in public health. Parent and child group leaders at RIH had previous experience with the treatment protocol. To ensure similarity between the two sites, group leaders at SDSU received 2 weeks of training in the treatment protocol from the RIH group leaders. In addition, all treatment sessions (parent and child groups) were videotaped and reviewed by the first author (L.J.S.) each week. Feedback was provided to the SDSU group leaders by phone on a weekly basis.
Children's calorie intake was assessed using 7-day weighted food diaries. Parents in the behavioral intervention group were given standard food scales and taught to accurately measure and record all foods and liquids consumed by their children during Session 1. Parents of the control children were also seen as a group and taught the same procedures for recording the food intake of their children. Diet diaries were kept for 7 days at baseline and at the end of treatment for both treatment and control groups. Diaries were analyzed for nutritional content using the Nutripractor 4000, which assessed the percentage of calories derived from protein, fat, and carbohydrates. The Nutripractor is a professional microcomputer software system designed for nutrient analysis and diet planning. Nutripractor results were reviewed by the dietitian for accuracy. The treatment group also kept food diaries throughout the 6 weeks of treatment, Sessions 2-7. Treatment diaries were reviewed weekly by the dietitian for thoroughness and accuracy. Corrective feedback was provided to parents during treatment sessions when necessary.
Weight, height, and skinfolds were measured by children's group leaders according to standardized protocols. Weights were obtained pre- and posttreatment for both groups as well as weekly throughout intervention for the treatment group using a SECA Integra 815 digital scale. This model is reported to be accurate within 0.2 kg. Heights were obtained at pre- and posttreatment using the Shorr Infant/Child Height Measuring Board, which is calibrated in centimeters. Skinfold measures using Lange skinfold calipers were obtained in order to derive percentage of body fat. Skinfold measures were taken at four sites (i.e., biceps, triceps, subscapula, and suprailiac), and percentage body fat was calculated using equations derived specifically for children (31).
Pulmonary function tests (PFTs) were conducted pre- and posttreatment by staff from the pulmonary departments at each hospital (RIH and UCSD). Standard spirometric techniques were used to assess pulmonary functioning as reflected by forced vital capacity, forced expiratory volume at 1 s, and mean rate of airflow during the middle half of forced expiration.
Resting Energy Expenditure (REE)
All children were nonfebrile and had no recent history of acute lung infection. REE was measured in a thermoneutral environment by indirect calorimetry using open-circuit spirometry. Expired air was analyzed for oxygen and carbon dioxide content by Applied Electrochemistry gas analyzers with an online system connected to an IBM computer. Children rested supine for 15 min in a fasted state (12 h overnight). Following a rest period, a mask was placed over the nose and mouth for the purpose of collecting expired air. REE was measured for 20 min. The mean VO2 and VCO2 from the equilibrated period was used to derive the respiratory quotient. REE was calculated via the abbreviated formula of Weir (32). The percentage of predicted REE was calculated by dividing the measured REE by the values developed by the World Health Organization for predicted REE based on age, height, weight, and gender (33).
Pre- and posttreatment activity was measured objectively by the Caltrac electronic accelerometer, which assesses the quantity and intensity of movement in the vertical plane. Interinstrument reliability is 0.96, and the Caltrac has been found to significantly correlate with heart-rate monitoring, direct observation, and self-reports of physical activity in laboratory and field settings (34,35). The Caltrac accelerometer was programmed to function solely as a physical activity monitor in the current study. Some activities, such as bicycle riding, are not detected well by the Caltrac since it assesses acceleration in the vertical plane (36). Swimming, a common activity for some children, could not be measured since the Caltrac is not waterproof.
All subjects (treatment and control) wore the Caltrac accelerometer for 3 days—2 weekdays and 1 weekend day—during the baseline and posttreatment assessment weeks. The instruments were worn in a “fanny pack” placed snugly around the waist, which was affixed by the parent in the early morning and taken off in the evening, just before bedtime. To prevent tampering, plastic ties were placed so that the pouches could not be opened by the child. Each child wore the same instrument throughout the study. The Caltracs were worn continuously during the day, except for bathing and swimming. The mean monitoring time was 11.7 h per day. At the end of the day, the parents detached the accelerometers and immediately recorded the cumulative number of “Caltrac counts” displayed on the small screen. A simultaneous log of the child's physical activity was also kept by the parent during the monitoring period.
Intervention followed the behavioral treatment protocol developed by Stark et al. (20,24). Treatment calorie goals were determined individually based on each child's average calorie intake during baseline. Children's treatment goals were set at 25-50% above their baseline calorie consumption. These goals are consistent with the goals used in calorie-supplement research with CF patients (9,12,27). The overall calorie goal for each child was divided by the number of meals targeted (four) so that calories were gradually increased sequentially across the four targeted meals, with an average increase of 100-200 calories per meal.
The parents' treatment group included nutritional education and child behavior management training (37). Nutritional education provided information on each of the major food groups, recipes for nutritious, high-calorie meals and snacks, and methods of increasing calories without increasing bulk (e.g., adding butter, powdered milk). Training in behavior management focused on teaching parents how to encourage their children to eat the foods recommended in an adaptive manner (38). Child management skills taught included differential attention, contigency management, and implementation of mealtime rules and consequences. Baseline assessment of the dependent measures was obtained during Session 1. Sessions 2-7 targeted snack, breakfast, relaxation skills training, lunch, dinner, and maintenance strategies, respectively.
Similarly, the children's group consisted of presenting nutritional information (identification of high- versus low-energy foods and methods of boosting calories at meals and snacks), reviewing behavior management techniques to be implemented at home by their parents, eating a sample meal that met new calorie goals, and practicing new eating behaviors (e.g., eating without dawdling). A behavioral program involving the exchange of stars earned on a chart for achieving daily calorie goals at each meal and following mealtime rules was also implemented. Children earned trophies in group each week contingent on earning 75% of total possible stars. During Session 4 the children learned a relaxation technique (39) as a method of coping with feelings of fullness that accompany increased caloric intake. Relaxation has been shown to be an effective coping skill for children experiencing abdominal pain (40) and has been used effectively with children with CF (24). The specific sequence and content of the behavioral management techniques are described herein.
Session 1: Baseline
Parents were instructed on how to monitor and record their child's oral intake using food scales, measuring cups, and measuring spoons. Instruction included oral explanation, demonstration, and practice supervised by the dietitian. Children were taught to help their parents keep the food diaries by telling their parents everything they ate. The children were told that they could earn a star on their star charts for each meal where they assisted with record-keeping.
Session 2: Snack Intervention
Parents were instructed in the use of praising and describing appropriate eating behavior. Appropriate eating was generally defined as accepting and eating snacks at defined times. In addition, parents were taught to identify and praise specific behaviors that promoted the intake of calories, such as eating quickly, taking bites, swallowing, etc. For example, to promote compliance with snack time, a parent would praise their child for coming to the table (e.g., “I like the way you came to snack right away. Good job listening to mom.”). Parents were taught to praise specific behaviors that promote higher calorie intake (e.g., “I like the way you take one bite after another. That is a great way to eat your snack.”). Parents were also taught to use a behavioral star chart and award a star to their child each time he or she was compliant with coming to snack and consuming the food offered. Conversely, if a child refused snack, the parent was taught to pay minimal attention to this behavior and to say to the child, “Since you did not eat your snack, I cannot give you a star on your chart.” Parents were also taught to ignore behaviors incompatible with eating, such as excessive talking, dawdling, and complaining about food during snack time. Children were provided with information on high- and low-energy snack foods and told they would be able to earn a star on their star charts each time they ate a snack when their mom or dad told them to do so.
Session 3: Breakfast Intervention
Parents were instructed on the initial use of contingency management: providing rules and consequences. Parents were taught how to define and set basic rules to promote eating. The rules set during this session were that the child must sit at the table until excused by mom or dad and that the child must meet his or her energy goal for the breakfast meal within 20 min to earn a star. The use of praising, describing, and ignoring in combination with rule setting and contingency management was explained. The consequence of the child's leaving the table before being excused was that the parent would calmly walk the child back to the table without talking or making eye contact. The importance of praising the child for engaging in positive eating behaviors was emphasized through instruction, role-playing, and feedback. The consequence for not meeting the energy goal within 20 min was removal of the child's plate and no access to food until the next scheduled meal or snack. The consequences for completing the meal within 20 min was release from the table to engage in more desirable activities (i.e., playing), a star on the star chart, parental praise, and access to additional food upon request. Children were taught their energy goals for breakfast and the new rules about staying at the table and eating within 20 min.
Session 4: Relaxation
No additional calorie goals were introduced during this session. Parents learned Contingency Management Level II: Awarding and Withholding Home-Based Privileges. Parents were taught to identify common privileges that were typically provided to their child noncontingently at home. They were then instructed on how to award these privileges after snack and breakfast if their child followed the mealtime rules and met their energy goals for each meal. For example, some families routinely allowed their children to watch television during breakfast. Parents were instructed that television viewing was a privilege and that they could use it to motivate their children to eat their breakfast in a timely manner. Parents who routinely allowed television viewing were taught to use it contingently by allowing the child to watch television only if they met their breakfast calorie goal and ate breakfast within 20 min. Privileges were individualized for each child.
Children were taught the new rules for home-based privileges. In addition, they were taught a progressive muscle relaxation (PMR) exercise as a way to cope with sensations of fullness. After the rationale for using PMR was reviewed, the children practiced by following a PMR exercise from an audiotape. The group leaders assisted the children in their acquisition of the skills, and each child was provided a copy of the audiotape for use at home. The children were given a star chart for achieving calorie goals and for practicing relaxation skills once a day.
Session 5: Lunch Intervention
Parents were provided with a review of the behavioral child management skills taught to date. Each family was encouraged to share their experience of implementing the strategies at home, and the group leaders facilitated problem-solving for difficulties that parents were experiencing. Privileges that could be applied to the lunch meal were identified for each family, and the mealtime rules from Session 3 were reviewed and applied to lunch. Parents were also taught how to use the behavioral strategies to increase the variety of foods their children accepted. The continued importance of praising and describing appropriate behaviors and ignoring behaviors incompatible with eating was emphasized. Role-playing of the behavioral strategies was conducted with parents. Children were given a review of mealtime rules and consequences. They helped prepare and then eat a lunch meal. As usual, the group leaders used behavioral management strategies to facilitate the children's consumption of the practice lunch and compliance with mealtime rules.
Session 6: Dinner Intervention
Parents were taught to implement behavioral child management strategies at dinner. Several strategies aimed at increasing parents' independence in using behavioral management skills were introduced. First, parents were taught to provide primary rewards for children's behavior changes at meals in preparation for fading of group rewards. Second, parents were encouraged to provide rewards at the end of each day contingent upon meeting calorie goals across all meals rather than after each meal. Finally, group leaders facilitated parent problem-solving efforts by having parents describe behavior problems at dinner and role-play with other parents and group leaders ways in which to manage these problems using strategies taught previously. Children were given a practice dinner meal. Meal energy goals and rules were reviewed. Children were prepared for the shift in the privilege system from meal by meal to end of day and from the group leaders to their parents.
Session 7: Maintenance and Sick Days
The last group was held 2 weeks after Session 6. This session provided a gradual fading of the treatment group and allowed parents to problem-solve on their own for a longer period. Session 7 focused on reviewing parents' and children's progress and problem-solving and on emphasizing how to continue the use of the behavioral strategies to maintain treatment gains. Parents were instructed to continue using star charts at home, with stars exchangeable for home-based daily and weekly privileges. Parents were instructed on how to fade from daily to weekly rewards over the next month. Parents were advised to manage sick days with high-calorie drinks. They were also instructed to gradually increase their child's calorie intake after recovery from an illness that affected their child's appetite. Children received a posttreatment party in their group to celebrate their progress. They were given a variety of fun high-calorie drinks to sample and instructed to use these drinks whenever they were sick.
Seven-day food diaries, physiological measures, and anthropometric measures of the children in the control group were taken at times corresponding to baseline and the last week of intervention for the treatment group. Similarly to the parents participating in treatment, the parents in the control group were seen together at the start of the study to explain how to keep an accurate food diary.
All children receiving treatment were followed for calorie intake and weight gain 3 months and 6 months posttreatment.
T tests for independent samples were conducted on change scores (pre- to posttreatment) between the groups (treatment and control) on the primary dependent measures of calorie intake and weight gain (absolute weight gain and weight Z score). Change scores were reported because they are a more rigorous test of change on the dependent measures, since subjects served as their own controls and change scores may thus be averaged and compared between groups (41). Because of the small sample size, only the primary dependent measures were used in planned analyses. Because of the short duration of treatment (6 weeks), changes in height (including Z scores for height), PFTs, REEs, or physical activity were not anticipated. These measures are presented pre- and posttreatment as exploratory and are considered to be important to assess because little is understood about the effect of early intervention on disease and nutritional status in CF. Therefore, while unexpected, change on these measures within the 6 weeks of treatment would be significant.
Owing to multiple family stressors, one of the control subjects was unable to complete the postintervention measure of calorie intake; therefore calorie results are reported for five treatment subjects and three control subjects. Average daily calorie intake, percentage RDA of energy, and percentage of daily calories derived from protein, carbohydrate, and fat pre- to postintervention for the treatment and control groups are presented in Table 1. During the intervention period, the children in the behavioral intervention group increased their average daily calorie intake by ≈1,000 cal/day. During this same period the children in the control group remained relatively stable and increased their intake by only 244 cal/day. The change in calorie intake was significantly greater in the behavioral intervention group compared with the control group [t(6) = 2.826, p = 0.03].
The increased calorie consumption is also reflected by the increase in the percentage RDA of energy consumed by the treatment group during intervention. The children in the treatment group exceeded the CF nutrition recommendations of 120-150% of the RDA for energy for healthy children postintervention. By contrast, the control group did not change their percentage RDA for energy during this same period. The percentage of calorie intake derived from protein, carbohydrates, and fat remained relatively stable for both groups.
Mean daily calorie consumption for each subject in the treatment and control groups is graphically presented in Fig. 1. As can be seen in Fig. 1, the calorie increase in the treatment group was achieved because all subjects increased their calorie consumption pre- to postintervention, whereas only one subject in the control group increased calorie consumption. Thus, intervention was effective for all children who participated in treatment.
The effects of the increased calorie intake on the anthropometric measures of weight, height, and percentage body fat are presented in Table 2.
The children in the treatment group gained an average of 1.7 kg (range of 0.5-3.2 kg) during treatment. During this same time period the mean weight gain by children in the control group was 0 kg (range -0.4-1 kg). The change score for weight gain was significant between the two groups [t(7) = 2.588, p = 0.03]. As can be seen in Table 2, the improvement in weight gain resulted in improved Z scores for the children who were enrolled in the treatment group. The Z scores for the children in the control group did not change during this time period. An unpaired t test of the change scores pre- to post-treatment was significant between groups [t(7) = 2.625, p = 0.03], indicating that the treatment group showed improvement in weight compared with the controls. The weight Z scores are presented graphically in Fig. 2 for each individual subject in the treatment and control groups pre- and postintervention. As can be seen in Fig. 2, all but one subject in the treatment group showed an increase in their Z scores for weight, indicating “catchup” growth for weight. Two of the five subjects achieved a Z score of ≈0, indicating that they were at the 50th percentile weight for age at the end of treatment. In contrast, none of the control subjects showed an increase in their Z scores during this same time period.
While assessment of height needs to be conducted over a period of 1 year to accurately reflect change (42), height data are presented in Table 2 for comparison purposes between the two groups. The children in both the treatment and control groups showed an increase in absolute height. Neither group showed an increase in their Z scores, indicating that neither group experienced catchup growth for height during the time of the intervention.
Body fat for the two groups remained relatively stable during the time of intervention.
The effects of the increased calorie intake on the physiological measures of pulmonary functioning, resting energy expenditure, and physical activity are presented by group pre- and posttreatment in Table 3. No changes were found on any of the physiological measures.
One month after completion of the behavioral intervention, the families in the control group were offered treatment. Two of the four families, Subjects 6 and 8, received treatment and provided a replication of the behavioral intervention. Both subjects were boys. Subject 6 was 10 years, 3 months of age, and Subject 8 was 6 years, 6 months of age. The results of the replication of the intervention with Subjects 6 and 8 for calorie intake can be seen in Fig. 3. The increased calorie intake for Subject 6 was comparable to that of the subjects in the initial intervention at 1,134 calories per day above baseline. Subject 8 increased his calories only slightly, 100 calories a day above baseline. Subject 8 exhibited extreme noncompliance with all aspects of dietary treatment, including enzyme replacement therapy before and during treatment. Consequently, treatment was modified to teach his mother to apply the behavioral child management strategies to enzyme compliance first, then to dietary intake. Because of the additional focus, treatment was extended 1 month posttreatment via phone contact with the family. Phone contact focused on providing continued support and instruction on the use of the behavioral child management strategies in improving the child's compliance with taking enzymes and increasing calorie intake. At the end of the month, Subject 8's calorie intake was 346 calories per day above baseline.
Both subjects gained more weight during intervention than they had during the 6-week period in which they served as controls. During intervention, Subject 6 gained 2.4 kg and Subject 8 gained 1.1 kg. During the previous 6-week period in which they were assessed for weight change, but did not receive intervention, Subject 6 had lost 0.4 kg and Subject 8 had gained only 0.4 kg. The effect of intervention on weight Z score is shown in Fig. 4. Both subjects showed an improvement in their weight Z score, thus demonstrating catchup weight gain.
All children receiving treatment were assessed for calorie intake and weight at 3 months and 6 months posttreatment. Calorie intake at pretreatment, posttreatment, and 3-month and 6-month follow-up for Subjects 1, 2, 3, 4, 5, 6, and 8 is shown in Fig. 5. As can be seen in Fig. 5, all subjects maintained their calorie intakes at a level above their baseline intakes. For Subjects 1, 2, 4, and 6 this level was comparable to their posttreatment intake. For Subjects 3 and 5, the calorie level at the 3- and 6-month follow-up was less than their posttreatment intake, but continued to be greater than their pretreatment intake.
Absolute weight in kilograms for all subjects receiving treatment at pretreatment, posttreatment, and 3- and 6-month follow-up is shown in Fig. 6. Subjects 1, 3, 4, 5, and 8 maintained their improved weight and showed continued weight gain at the 3- and 6-month follow-ups. Subjects 2 and 6 showed a slight decline in weight at the 3- and 6-month follow-ups, but continued to be above their pretreatment weight.
The present study supports our previous research on the efficacy of behavioral treatment to increase calorie consumption in children with cystic fibrosis. The children in the current study who received the behavioral intervention increased their caloric intake an average of 1,000 calories/day pre- to post-treatment. This increase is similar to that found in our previous treatment studies (20,24) and comparable to changes produced by medical intervention (9,11,12). Furthermore, in the current study, the change in the calorie intake of the children receiving treatment was significantly greater than that of the control group of children who did not receive treatment. The calorie intake of the children in the control group was monitored by their parents at times corresponding to the beginning and end of treatment for the behavioral intervention group. The similar demand for monitoring provides support that experimental expectation and calorie monitoring alone are not sufficient to produce significant increases in the oral intake of children with CF.
The present study is the only investigation of nutrition and CF to assess the calorie intake of a control group of patients. It is interesting to note that although not significant, the children in the control group increased their calorie intakes an average of 244 calories/day. This increase may indicate that self-monitoring was associated with a small increase in intake, or, more likely, it represents the normal fluctuation in caloric intake across weeks (43).
Another point worth discussing is that the groups were not equal on measures of calorie intake and weight percentile at baseline. While not significantly different on either of these measures, the treatment group was consuming more calories and was at a greater weight percentile than the control group before intervention. These differences may indicate behavioral differences between the two groups at meals that led the treatment group to be more responsive to the intervention. This hypothesis, however, is not supported by our assessment research on parent and child behaviors at meal-times. In this research we found behavior problems at meals regardless of the weight percentile of the child with CF (28). That is, children with CF who were at the appropriate weight for age had behavior problems at a rate similar to underweight children with CF (28,38). Furthermore, the present study actually represents three replications of the treatment effects because the children reported in the initial treatment group were seen at two sites, 3,000 miles apart (Providence, RI, and San Diego, CA). The third replication was the application of behavioral intervention to two of the control subjects.
The replication of treatment effects across all three groups of children support the efficacy of the intervention in producing the observed changes in calorie intake and weight gain. When the two control children received the behavioral intervention, one child increased his calorie intake by 1,000 calories per day above his pretreatment intake. The second child increased his intake by 100 calories pre- to posttreatment. However, this child presented extreme behavior problems at meals, was refusing to take pancreatic enzymes, and had shown a decrease in his calorie intake of 298 cal/day before treatment. When treatment was modified to include applying the behavioral management strategies to increasing the child's compliance with his enzyme intake as well as calorie intake, improved calorie intake was observed for this subject as well. Thus, it is unlikely that the reported changes in calorie intake would happen by chance across three treatment groups.
Increased calorie intake was associated with improved weight gain for all of the children in the behavioral intervention group over the 6 weeks of treatment, with two of the children achieving at or near the 50th percentile weight for age. During this same time, the children in the control group did not significantly increase their weight, and two of the children lost weight, thus showing that the weight gain achieved by the children who received treatment was greater than would be expected during the same time period without intervention. This increase is especially important to establish because children are expected to gain weight as part of their normal growth and development (44). This finding was replicated with the two children from the control group after they received the behavioral intervention. Both gained more weight during the 6 weeks of treatment than they did during the 6 weeks they served as controls. Subject 8's weight gain, however, is probably due to the impact of treatment on his enzyme compliance and points to the need to assess compliance to enzyme therapy before treatment. It also raises the issue that enzyme compliance may need to be a routine part of any intervention targeting nutrition.
As in our previous studies (20,24,38), the present study found maintenance of the treatment gains at 3 and 6 months posttreatment. Four of the seven subjects maintained their posttreatment calorie intake, and two subjects lessened their calorie intake from their posttreatment levels but continued to be above their pretreatment calorie intake. Subject 8 showed an increase in calorie intake at the 3-month follow-up, which most likely reflects the delayed effects of treatment implemented over a longer timeframe, as noted earlier. Subject 8 then showed maintenance of an improved calorie intake above his baseline consumption at 6 months posttreatment.
The critical components of treatment success cannot be readily identified from the present study because the intervention included both child behavior management training and nutrition education. However, previous interventions targeting nutrition education alone have not shown substantial increase in calorie intake (19). The importance of the behavioral component is also supported by previous studies that found adaptive changes in children's eating behavior and parent-child mealtime behaviors. For example, Stark et al. (24) reported that behavioral intervention led to increased eating (bites per minute), increased calories consumed per minute, and decreased length of mealtime for children with CF. In a subsequent study targeting parent-child mealtime interaction in two children with CF whose weights were appropriate for age, Stark et al. (38) found that providing behavioral intervention without nutrition education resulted in improved parental control at dinner time, increased parental attention to appropriate child eating behavior, and decreased parental attention to inappropriate child eating behavior. These changes in parent behavior were associated with increases in appropriate eating behavior and improved intake by the children. This finding supports the importance of modifying parent-child interaction at meals in improving oral intake (45).
While the impact of the intervention on the children's health status, as measured by pulmonary functioning and activity level, was assessed, changes were not anticipated on these measures for several reasons. The present intervention is designed for early intervention; therefore the children targeted were not in pulmonary decline and thus not likely to show improvements. Second, the length of the intervention was only 6 weeks, and improved nutritional status may need to be of a longer duration to positively affect other health measures. Studies showing pulmonary improvement have typically been of longer duration (≤2 years) and involved intervention with patients with more severe malnutrition (46). Furthermore, while some medical interventions have reported improved pulmonary functioning posttreatment (12), it is more common for studies to report maintenance of baseline functioning (10) or a slowed rate of decline in treatment subjects when pulmonary functioning is compared with pretreatment medical records (47,48). The benefits of nutritional intervention on pulmonary function may be most apparent in long-term follow-up when subjects are compared with patients who never received treatment. In such a comparison, Dalzell et al. (46) found that subject groups receiving enteral supplementation had a slower rate of decline on pulmonary functioning, were less under-weight, and had fewer deaths when compared with a group of patients matched on age, sex, and disease severity 5 years posttreatment.
Previous research has indicated that children with CF have an elevated REE compared with norms (49) and healthy controls (50). REE was measured in the present study because previous interventions have reported increased REE after refeeding (51). Such an increase in REE would have implications for maintenance of weight gain and indicate a need for further increases in intake to compensate. While elevated REE was found at baseline for both groups in the present study, the REE was not found to increase in the behavioral intervention group after the increase in calorie intake. It may be that increased oral intake does not affect REE in the same manner as energy supplied via enteral supplementation. Conversely, it may be that the subjects in the present study were not followed long enough to detect changes in REE post-treatment. The increase in REE reported by Vaisman et al. (51) was found 6 months posttreatment. Longer-term follow-up needs to be conducted with treatments that increase the oral intake of CF patients, such as behavioral interventions, to assess whether increases in REE are associated with increased energy consumption in CF patients.
In summary, the behavioral intervention was successful in increasing the caloric intake of five underweight children with CF compared with a group of control children who did not receive intervention but whose calorie intake was monitored at the same points pre- and posttreatment. These treatment gains were then replicated with two of the four controls. Furthermore, the increased calorie consumption was associated with increased weight gain in the treatment group. Again, this finding was replicated with two of the control subjects. Thus, all subjects who received treatment showed increases in calorie consumption and weight. Subjects receiving treatment in the present study were followed for calorie intake and weight at 3 and 6 months post-treatment, and the findings support our previous behavioral studies with CF patients that such changes may be long-lasting (20,24). The long-term implications of these treatment gains remain to be investigated. It is hypothesized that by improving the parent-child interaction around meals, the behavioral intervention would lead to more adaptive eating behaviors. Adaptive eating behaviors are hypothesized to result in improved calorie intake and weight gain that would delay or possibly prevent the need for artificial feedings at later stages of disease progression.
The current intervention was comprehensive and included behavioral child management training as well as nutrition education. It was hypothesized, however, that the behavioral component of treatment was crucial to the success of the intervention. The current study has been an important addition to the literature on the application of behavior therapy to nutrition in CF; however, further studies need to provide a stronger test of the intervention by comparing the behavioral intervention with nutrition education alone. Larger sample sizes and longer follow-ups are also needed to evaluate the effect of behavioral treatment on disease status as indicated by pulmonary functioning. The current study shows that the behavioral intervention can be successfully exported to other CF centers, thereby allowing for larger sample sizes in future studies; it also highlights the potential for widespread application of an early intervention strategy to improve nutritional status of children with CF.
Acknowledgment: This research was supported by a grant from the National Cystic Fibrosis Foundation (no. 2117) to Lori J. Stark.
1. FitzSimmons SC. The changing epidemiology of cystic fibrosis. J Pediatr 1993;122:1-9.
2. Gurwitz D, Corey M, Francis PS, Crozier D, Levison H. Perspectives in cystic fibrosis. Pediatr Clin North Am 1979;26:603-15.
3. Kraemer R, Rudeberg A, Hadorn B, Rossi E. Relative underweight in cystic fibrosis and its prognostic value. Acta Paediatr Scand 1978;67:33-7.
4. Durie PR, Pencharz PB. A rational approach to the nutritional care of patients with cystic fibrosis. J Roy Soc Med 1989;82(suppl 17):11-20.
5. O'Rawe A, McIntosh I, Dodge JA, et al. Increased energy expenditure in cystic fibrosis is associated with specific mutation. Clin Sci 1992;82:71-6.
6. Dodge JA. Nutritional requirements in cystic fibrosis: a review. J Pediatr Gastroenterol Nutr 1988;7:S8-11.
7. MacDonald A, Holden C, Harris G. Nutritional strategies in cystic fibrosis: current issues. J Roy Soc Med 1991;84(suppl 18):28-35.
8. Allan JD, Mason A, Moss AD. Nutritional supplementation in treatment of cystic fibrosis of cystic fibrosis of the pancreas. Am J Dis Child 1973;126:22-6.
9. Lester LA, Rothberg RM, Dawson G, Lopez AL, Corpuz Z. Supplemental parenteral nutrition in cystic fibrosis. J Parenter Enter Nutr 1986;10:289-95.
10. Mansell AL, Anderson JC, Muttart CR, et al. Short-term pulmonary effects of total parenteral nutrition in children with cystic fibrosis. J Pediatr 1984;104:700-5.
11. Levy LD, Durie PR, Pencharz PB, Corey ML. Effects of long term nutritional rehabilitation on body composition and clinical weight status in malnourished children and adolescents with cystic fibrosis. J Pediatr 1985;107:225-30.
12. Shepherd R, Holt PT, Thomas BJ, et al. Nutritional rehabilitation in cystic fibrosis: controlled studies of effects on nutritional growth retardation, body protein turnover, and course of pulmonary disease. J Pediatr 1986;109:788-94.
13. Ramsey B, Farrell P, Pencharz P. Nutritional assessment and management in cystic fibrosis: consensus conference. Am J Clin Nutr 1992;55:108-16.
14. Stark LJ, Jelalian E, Mulvihill MM, et al. Eating in preschool children with cystic fibrosis and healthy peers: a behavioral analysis. J Pediatr 1995;95:210-15.
15. Tomezsko JL, Stallings VA, Scanlin TF. Dietary intake of healthy children with cystic fibrosis compared with normal control children. Pediatrics 1992;90:547-53.
16. Carter BD, Kronenberger WG, Conradsen E. Marital quality, parenting stress and compliance in families of children with cystic fibrosis. Presented at the Florida Conference on Child Health Psychology, Gainsville, Florida, April 1993.
17. Gudas LJ, Koocher GP, Wypij D. Perceptions of medical compliance in children and adolescents with cystic fibrosis. Devel Behav Pediatr 1991;12:236-42.
18. MacDonald A, Kelleher J, Littlewood JM. A normal fat diet for cystic fibrosis: Is a dietitian still needed? Scand J Gastroenterol 1988;23:157-9.
19. Bell L, Durie P, Forstner GG. What do children with cystic fibrosis eat? J Pediatr Gastroenterol Nutr 1984;3(suppl 1):S137-46.
20. Stark LJ, Bowen AM, Tyc VL, Evans S, Passero MA. A behavioral approach to increasing calorie consumption in children with cystic fibrosis. J Pediatr Psychol 1990;15:309-26.
21. Crist W, McDonnell P, Beck M, Gillespie CT, Barrett P, Mathews J. Behavior at mealtimes and nutritional intake in the young child with cystic fibrosis. Devel Behav Pediatr 1994;15:157-61.
22. Quittner AL, DiGirolamo AM, Winslow EB. Problems in parenting a child with cystic fibrosis: a contextual analysis. Presented at the Florida Conference on Child Health Care Psychology, Gainsville, Florida, April 1991.
23. Bowen AM, Stark LJ. Malnutrition in cystic fibrosis: a behavioral conceptualization of cause and treatment. Clin Psychol Rev 1991;11:315-31.
24. Stark LJ, Knapp L, Bowen AM, et al. Behavioral treatment of calorie consumption in children with cystic fibrosis: replication with two year follow-up. J Appl Behav Anal 1993;26:435-50.
25. Bertrand JM, Morin CL, Lasalle R, Patrick J, Coates AL. Short-term clinical nutritional and functional effects of continuous elemental enteral alimentation in children with cystic fibrosis. J Pediatr 1984;104:41-6.
26. Pencharz P, Hill R, Archibald E, Levy L, Newth C. Energy needs and nutritional rehabilitation in undernourished adolescents and young adult patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 1984;(suppl 1):S147-53.
27. Shepherd RW, Thomas BJ, Bennett D, Cooksley WGE, Ward LC. Changes in body composition and muscle protein degradation during nutritional supplementation in nutritionally growth-retarded children with cystic fibrosis. J Pediatr Gastroenterol Nutr 1983;2:439-46.
28. Stark LJ, Bowen AM, Mulvihill MM, et al. Behavioral and environmental factors affecting nutrition in CF [Summary]. Pediatr Pulm 1991;10(suppl 6):207.
29. Hollingshead AB. Two factor index of social status. New Haven: Yale University, 1975.
30. Hays WL. Statistics, 3rd ed. New York: Holt, Rinehart, & Winston, 1981.
31. Brook CGD. Determination of body composition of children from skinfold measurements. Arch Dis Child 1971;46:182-4.
32. Weir JB. New methods for calculating metabolic rate with reference to protein metabolism. J Physiol 1949;109:1-9.
33. Joint FAO/WHO/UNU Expert Consultation. Energy and protein requirements. Technical Report No. 724. Geneva: World Health Organization, 1988.
34. Sallis JF, Buono MJ, Roby JJ, Carlson D, Nelson JA. The caltrac accelerometer as a physical activity monitor for school-age children. Med Sci Sports Exerc 1990;22:698-703.
35. Klesges LM, Klesges RL. The assessment of children's physical activity: a comparison of methods. Med Sci Sports Exerc 1987;19:511-7.
36. Montoye HJ, Washburn R, Servais S, Ertt A, Webster J, Nagle F. Estimation of energy expenditure by a portable accelerometer. Med Sci Sports Exerc 1983;15:403-7.
37. Forehand RL, McMahon RJ. Helping the noncompliant child. New York: Guilford Press, 1981.
38. Stark LJ, Powers SW, Jelalian E, Rape R, Miller DL. Modifying problematic mealtime interactions of children with cystic fibrosis and their parents via behavioral parent training. J Pediatr Psychol 1994;19:751-68.
39. Koeppen AS. Relaxation training for children. Elem School Guid Counsel 1974;9:14-21.
40. Sanders MR, Rebgetz M, Morrison M, et al. Cognitive-behavioral treatment of recurrent nonspecific abdominal pain in children: an analysis of generalization, maintenance, and side effects. J Consult Clin Psychol 1989;57:294-300.
41. Willett JB, ed. Questions and answers in the measurement of change. In: Rothkopf EZ, ed. Review of education research. Washington, D.C.: American Education Research Association, 1988:345-422.
42. Marshall WA. Evaluation of growth rate in height over periods of less than one year. Arch Dis Child 1971;46:414-20.
43. St. Jeor ST, Guthrie HA, Jones MB. Variability in nutritient intake in a 28-day period. J Am Diet Assoc 1983;83:155-62.
44. Fomon SJ, Haschke F, Ziegler EE, Nelson SE. Body composition of reference children from birth to age ten years. Am J Clin Nutr 1982;35:1169-75.
45. Palmer S, Thompson RJ, Linscheid TR. Applied behavior analysis in the treatment of childhood feeding problems. Devel Med Child Neurol 1975;17:333-9.
46. Dalzell AM, Shepherd RW, Dean B, Cleghorn GJ, Holt TL, Francis PJ. Nutritional rehabilitation in cystic fibrosis: a five year follow-up study. J Pediatr Gastroenterol Nutr 1992;15:141-5.
47. Boland MP, Patrick J, Stoski DS, Soucy P. Permanent enteral feeding in cystic fibrosis: advantages of a replaceable jejunostomy tube. J Pediatr Surg 1987;22:843-7.
48. O'Loughlin E, Forbes D, Parsons H, Scott B, Cooper D, Gall G. Nutritional rehabilitation of malnourished patients with cystic fibrosis. Am J Clin Nutr 1986;43:732-8.
49. Vaisman N, Pencharz PB, Corey M, Canny GJ, Hahn E. Energy expenditure of patients with cystic fibrosis. J Pediatr 1987;111:496-500.
50. Buchdahl RM, Cox M, Fullylove C, et al. Increased resting energy expenditure in cystic fibrosis. J Appl Physiol 1988;64:1810-6.
51. Vaisman N, Clarke R, Pencharz PB. Nutritional rehabilitation increases resting energy expenditure without affecting protein turnover in patients with cystic fibrosis. J Gastroenterol Nutr 1991;13:383-90.