Stress management training for parents and their children undergoing cardiac catheterization

Acknowledgement: Research was supported by the Pediatric Cardiology Medical Group and Children's Hospital and Health Center, San Diego.

There is increasing awareness by health professionals of the emotional impact of hospitalization. Preoperative anxiety and postoperative distress can affect the physical and emotional recuperation of both children and adults. Children, however, are more likely to be affected; without the awareness and understanding that comes with maturity, they must face separation from their parents, the strangeness of the hospital setting and routine, and unfamiliar social interactions, in addition to the physical concomitant of their illness.19

Children with congenital heart disease experience the stressors inherent in chronic illness as well as those related to the acute crisis of hospitalization for invasive procedures.12 High levels of anxiety have consistently been identified at the time of cardiac catherization.7,8,18 Children perceive catheterization as a noxious procedure and this experience may interact with previously existing conditions to produce longterm adverse effects.1

This investigation evaluated a method of preparing parents and their school age children for hospitalization and cardiac catheterization in order to reduce their stress and anxiety. It was hypothesized that, given preoperative counseling and instruction in stress management skills, children, with the support of their parents, could successfully cope with the traumatic experiences related to catheterization. Moreover, as the parents learned to function effectively as therapeutic allies, they could not only help their children and themselves to cope with a highly stressful situation, they could also help the medical staff provide optimum care for their children.

Participants were 26 children, with their parents, who were patients of a pediatric cardiology medical group, awaiting elective cardiac catheterization at a children's hospital. The children were all English speaking, between the ages of six and 17, and had at least average intelligence as determined by the clinical judgment of their pediatricians. After participants had signed a consent form outlining the procedures of the study and had been assured that participation was voluntary, the participants were stratified by age and then were randomly assigned to a treatment or control group in order to measure the effects of the stress management training. There were 12 children in the experimental group and 14 in the control group. The control group received a brochure that outlined the procedures related to catheterization and discussed hospitalization. This required one session with a member of the cardiology staff.

Prior to catheterization, three sessions were scheduled for participants in the experimental group for supportive counseling and stress management training. At the first session, one staff member worked with the child while, concurrently, a second staff member worked with the parents. Treatment included exploration of concerns related to medical issues and skills training, accompanied by homework assignments. During the second session, parents observed the child's therapist using the stress management techniques with the child. In the last session, parents and their child rehearsed their newly acquired skills within a structured, imagined hospital scene. The goal was to provide feelings of self-confidence and concrete skills which would help the participants cope with the impending hospitalization and catheterization.

Children and parents in the experimental group were taught stress management skills. This included providing detailed information about hospitalization and catheterization, supportive counseling, relaxation techniques, and coping skills. The skills taught were 1) controlled deep breathing as an adjunct to relaxation and an internal signal to initiate other methods of relaxing;5 2) progressive muscle relaxation, whereby the patient sequentially tenses and relaxes all of the major muscle groups of the body, resulting in a general feeling of relaxation while providing an awareness of body sensations;11 3) guided imagery, whereby the participant visualizes a comfortable and relaxing place and is, consequently, able to relax more completely;4 4) biofeedback awareness, which provides information via electromyographic readings concerning the degree of muscle tension experienced;10 5) self-hypnosis, whereby the participant learns to alter the state of consciousness in order to control feelings and perceptions;3 and 6) cognitive reframing, which incorporates the use of self-talk and cognitive restructuring to prepare the participant to cope with future stressors.

Patients and parents were assessed in the hospital and at a conference held one week after catheterization. Following hospital admittance, the child completed the Hospital Fears Rating Scale, on which the child indicated the degree of fear experienced using a numbered, color-coded thermometer with a five-point scale (1 = no fear, 5 = very much afraid). Melamed and Siegel14 developed this scale and indicated that it has been found to differentiate groups of 30 subjects at the .01 level and also delineate age groups (younger children scored highest). Ratings of the child's behavior were recorded by trained staff raters blind to experimental condition on The Manifest Upset Scale and The Cooperation Scale20,21 at three key stress points: venipuncture, administration of precatheterization medication, and separation from parents. Ratings on both scales range from one (complete cooperation) to five (extreme resistance). Interrater reliability was considered adequate by the authors who also claim validity for the measures since the five-point rating scales were anchored to operationalized observable behavior. After the catheterization, the parents completed a structured interview form developed by the investigators which uses a five-point rating scale to report their feelings at specific stress points (1 = no tension, 5 = very high tension).

One week after catheterization the child again completed the Hospital Fears Rating Scale. The parents completed the Post-Hospital Behavior Rating Questionnaire, which consists of 27 behavioral items most frequently found in children following hospitalization. Responses range from one (much less than before) to five (much more than before). This measure has received extensive use, provides good validity and reliability, and has identified six orthogonal factors of children's behavior.19 Parents also responded to a structured interview format, developed by the investigators, which assessed their degree of satisfaction with the information provided by the medical staff.

Demographic information. There were no significant pre-intervention differences found using chi square technique to compare the experimental and control groups on the following variables: age, sex, number of siblings, marital status of parents, employment of parents, religion, type of insurance coverage, and number of prior catheterizations and surgeries.

Children's self-report measure. Scores on the Hospital Fears Rating Scale from the time of admission and from one week after discharge were evaluated using an analysis of variance (ANOVA) for the mixed design with one between-groups factor (treatment) and one repeated measurement factor (time). The ANOVA is summarized in Table 1 and relevant means appear in Table 2 and in Figure 1. The treatment-by-time interaction was not significant in the ANOVA, indicating that the group difference in the change overtime was not significant; however, the means for the experimental group show a greater reduction over time than do the means for the control group. Additionally, the change over time for the subject sample as a whole and the overall difference between the treatment conditions were not significant.

Children's observed behavior. Scores for the Manifest Upset Scale obtained at venipuncture, precatheterization medication, and at the time of separation from parents were also evaluated in a two-factor ANOVA for mixed design. The analysis is summarized in Table 3 and the means for the two treatment groups appear in Table 2 and in Figure 2. The higher the score, the greater the degree of upset observed. Although the means for the control group were consistently higher than experimental group means, the overall difference between groups did not reach an acceptable level of significance. Irrespective of treatment condition, the mean score increased from the time of venipuncture to the time of precatheterization medication, and had decreased by the time of separation from parents. This effect was statistically significant, suggesting that the scores are sensitive to differences among specific stressors. Finally, the pattern of changes over time did not differ between the two treatment conditions.

Scores for the Cooperation Scale, collected at the same time that the Manifest Upset data were obtained, were evaluated in the same manner. The ANOVA of the Cooperation Scale is summarized in Table 3 and the means appear in Table 2 and in Figure 2. The higher the score, the greater the amount of uncooperative behavior observed. Mean scores in the control condition were consistently higher than mean scores in the experimental condition and the overall difference between treatment conditions was statistically significant. Like Manifest Upset mean scores, Cooperation Scale means increased from venipuncture to precatheterization and then decreased by the time of separation from parents. The time main effect, however, was not statistically significant for Cooperation scores, indicating that no significant change over time was observed for the subject sample as a whole. Finally, the treatment-by-time interaction again failed to achieve statistical significance. This indicates that there was no significant difference between treatment conditions in the pattern of changes overtime. Likewise, the lack of a significant interaction indicates that the observed overall difference between the two treatment groups, which was significant, did not vary significantly as a function of the particular stressor.

An ANOVA of the combined data from the Manifest Upset Scale and the Cooperation Scale was evaluated in the same manner. The ANOVA is summarized in Table 3 and the means are depicted in Figure 3. The overall difference between treatment conditions was statistically significant. Irrespective of treatment condition, the mean score significantly increased from the time of venipuncture to the time of precatheterization medication and had decreased by the time of separation from parents.

Parents' ratings of their children were obtained one week after discharge via the Post-Hospital Behavior Rating Questionnaire. The mean scores for the two treatment conditions are included in Table 2 and graphically presented in Figure 3. A one-factor completely randomized ANOVA indicated that the mean scores for the two groups differed significantly (F(1,19) = 5.00, p < 04). Thus, scores for the control group were significantly greater than scores for the experimental group.

Parent self-report measures. Parents' ratings of their own negative affect were obtained during the structured interview completed after the catheterization. The mean score for the experimental group was lower than the mean for the control group (Table 2; Figure 4). A one-factor complete randomized ANOVA indicated that this difference was statistically significant (F(1,24)=22.96, p<.0001). Parental satisfaction with information was rated one week after discharge. The mean score for the control group was greater than the mean for the experimental group (Table 2; Figure 4); however, a one-factor completely randomized ANOVA indicated that this difference was not statistically significant (F(1, 18)=.71, p<.41).

Relationships among variables. The amount of fear expressed by all of the children on the Hospital Fears Rating Scale the day of catheterization and the upset and cooperative ratings were further evaluated utilizing Pearson's Product-Moment Correlation (Table 4). Upset and cooperative behavior at venipuncture and precatheterization medication yielded significant positive correlations with scores from the Hospital Fears Rating Scale. This effect was not observed for upset and uncooperative behavior ratings after separation from parents.

The correlational analysis was also completed separately for each of the treatment conditions (Table 5). Of note, the relationship between Hospital Fears Rating Scale and Manifest Upset Scale yielded negative correlations in the experimental group and significantly positive correlations in the control group at all three stress points. Tests of the difference between independent correlations indicated that the association between hospital fears and upset behavior differed significantly between treatment conditions at venipuncture (z = 2.387, p < .05) and precatheterization medication (z =2.65, p < .01), but not at separation (z = 1.83). Hospital fears and cooperation did not differ significantly between the two treatment conditions although the same pattern of responses was observed.

The relationship between the parent responses on the structured interview of negative affect and upset and cooperative behavior ratings also yielded directionally opposite patterns of correlations in the experimental and control groups (Table 5). Significant negative correlations were found at all three points in the control group. Tests of the difference between independent correlations indicated that the association between parent affect and child upset differed significantly between the two treatment conditions only at the time of precatheterization medication (z = 1.96, p < .05). Parent affect and cooperation did not differ significantly between the two treatment conditions.

The Manifest Upset Scale and Cooperation Scale scores for all of the children were significantly positively correlated at all three stress points (Table 4). The correlational analysis was also completed separately for each of the treatment conditions (Table 5). Positive correlations between upset and cooperative behavior were found for both the experimental and control group. However, tests of the difference between independent correlations indicated that the direction of change differed significantly between the two treatment conditions at the time of venipuncture (z = 2.79, p < .01) and precatheterization medication (z = 2.65, p < .01). That is, scores for the experimental group were in the direction of less upset and more cooperation while scores for the control group were in the direction of more upset and less cooperation. This effect was not found for upset and cooperative behavior ratings at separation from parents.

The only significant correlation among demographic variables and outcome variables was between the age of the child and the amount of fear expressed (Hospital Fears Rating Scale pretest, r = .354, p < .10; post-test, r = − .450, p < .05). This would indicate that the younger the child, the more fear stated and the older the child, the less fear stated. This finding is consistent with prior research.

The behavioral observations reported by the medical staff of the children in the experimental group showed a reduction in manifest upset and a significant increase in cooperative behavior at key stress points in the hospital. In addition, these children made a more satisfactory post-hospital adjustment as reported by parents. However, since these latter scores were reported by the parents, the results could be a reflection of the parental perceptions biased by personal feelings rather than an objective evaluation of the children's behavior. That is, those parents who felt less anxious themselves, may have been less inclined to report disturbed behavior for their children following hospitalization. Although the mean scores of fear expressed by children in the experimental group indicated a trend toward lower reports of anxiety than those of control children, there were no significant differences between groups. Thus, the children's observed behavior appeared to be inconsistent with the degree of fear they expressed on the self-report measure. This may have been due to the small sample size, in that the Hospital Fears Rating Scale was not sensitive enough to identify differences or that the perception recorded by the parents was not an accurate measure of the child's response.

All of the children initially demonstrated efforts toward cooperating with hospital procedures. This is consistent with the conclusions of prior researchers that children do the best they can, but do not always have the inherent resources to adapt to the hospital experience. In examining the correlations between the Hospital Fears Rating Scale and manifest behavior, the degree of upset and cooperative behavior observed at venipuncture and premedication was commensurate with the level of fear reported by the children. However, after being sedated prior to the time the children separated from their parents and entered the catheterization laboratory, the fear expressed was no longer significantly correlated with the observed behavior.

Further, children in the experimental group cooperated in spite of the degree of fear expressed; conversely, the control group's manifest behavior was consistent with the amount of fear they expressed. This difference appears to be due to the effect of the intervention which allowed the experimental children to manage the stressors significantly more effectively than the control children, even though they felt frightened. The effects of the intervention can again be seen when analyzing the correlations between manifest upset and cooperative behavior. The data suggest that, although all the children became less cooperative as they became more upset, the children in the experimental group were initially able to use their coping skills to cooperate to a greater degree than those in the control group. This ability was no longer found after they had been sedated.

The statistical results consistently indicate treatment effects at venipuncture and premedication, but not after the sedative had taken effect at the time of the observation at parent separation. The fact that the control group no longer exhibited the upset and uncooperative behavior observed prior to the premedication most likely demonstrated the effectiveness of the sedative in masking anxiety.

The parents reported highly significant group differences in their feelings. The experimental group parents expressed considerably less stressful responses than the control group parents. Both groups felt that they were adequately informed and were very satisfied with the care their children received. Although prior research has shown positive effects of the dissemination of information to patients before their hospitalization, the current findings suggest that more benefits may be derived when additional interventions are provided. The emphasis in this study was placed on the parent assuming an active role as therapeutic ally, rather than the more passive role as recipient of information only. This focus, aimed at providing the parents with new coping strategies, may have given them a sense of control beyond what might be expected from just learning new information about the catheterization.

The correlations between parent self-reported negative affect and children's manifest behavior indicated that the less negativity the control group parent expressed, the more upset the child, and the less negative the experimental group parent, the less anxious behavior exhibited by the child. The stress management training may have provided the experimental group parents with ways to deal with stress and effectively work it through for themselves. Thus, the experimental group children were not overloaded with parental concerns. Rather, they were provided with an appropriate model to cope with their own fears, while receiving congruent messages from their parents. The implication of this stress management model is that it offers a team approach utilizing the parent-child relationship, psychological support services, and medical expertise to reduce anxiety and to increase cooperation. This can decrease the trauma associated with medical procedures and contribute to successful physical and emotional recuperation.

The results of this exploratory study suggest that the stress management program designed for pediatric cardiology patients and their families had a positive effect on the way parents experienced the hospitalization and the manner in which the children reacted to catheterization, both in the hospital and following discharge. The parents felt better and the children behaved more positively when they had the opportunity to learn to manage their own fears and anxieties.

Although the results must be viewed as preliminary due to the small sample size, these findings are consistent with other research studies that have demonstrated significant benefits from various prehospitalization preparation models. While most studies have examined normal healthy children hospitalized for routine procedures such as tonsillectomy, this study examines a chronically ill pediatric population and emphasizes the role of both patients and their parents in assuming greater responsibility for their own medical care. Not only do participants receive information which will help them understand the medical procedures better, they also learn skills that can be used to cope successfully with these procedures. It appears that this stress management model reinforces the parent-child relationship in that the child continues to look toward the parent as provider of nurturance, stability, and protection, instead of unexpectedly having to cope single-handedly with strangers (i.e., doctors, nurses, hospital staff) as caretakers. Parents do not abdicate their role by stepping passively into the background in the face of medical staff and hospital procedures. Instead, they reinforce and strengthen their role as the primary support persons for the well-being of their children by using the skills they have learned; therefore, they become active allies to the medical team, which contributes to optimum patient care.

Further studies are warranted to compare the effectiveness of this training model, which uses the parent as a therapeutic ally, to other preparatory models shown to reduce hospital stress in children.

AISENBERG, R., WOLFF, P. and ROSENTHAL, A.1973. Psychological impact of cardiac catheterization. Pediatrics51(6):52–55.

AZARNOFF, P. and HARDGROVE, C.1981. The Family in Child Health Care. John Wiley, New York.

BARBER, T.1970. Hypnosis and pain. In LSD, Marijuana, Yoga, and Hypnosis, T.Barber,ed.Aldine, Chicago.

BENSON, H., BEAR, J., ANDCAROL, M.1974. The relaxation response. Psychiatry37:37–46.

BRESLER, D. and TRUBO, R.1979. Free Yourself from Pain. Simon and Schuster, New York.

FERGUSON, B.1979. Preparing young children for hospitalization: a comparison of two methods. Pediatrics65(5):656–664.

FINESILVER, C.1980. Reducing stress in patients having cardiac catheterization. Amer. J. Nurs.80(10)1805–1807.

FREEMAN, W., PICHARD, A. and SMITH, H.1981. Effect of informed consent and educational background on patient knowledge, anxiety, and subjective responses to cardiac catheterization. Cath. C-V Diagnos. 7(2):119–134.

GARFIELD, C.1979. Stress and Survival: The Emotional Realities of Life-Threatening Illness. Mosby, St. Louis.

HAYNES, S., MOSELY, D. and MCGOWAN, W.1975. Relaxation training and biofeedback in the reduction of frontalis muscle tension. Psychophysiology12(5):546–552.

JACOBSON, E.1938. Progressive Relaxation. University of Chicago Press, Chicago.

KUPST, M., BLATTERBAUER, S. and WEST-MAN, J.1977. Helping parents cope with the diagnosis of congenital heart defect: an experimental study. Pediatrics. 59:266–272.

MEICHENBAUM, D., TURK, D. and BERNSTEIN, S.1975. The nature of coping with stress. In Stress and Anxiety, C.Spielberger and I.Sarason,eds.Hemisphere, Washington, D.C.

MELAMED, B. and SIEGEL, L.1975. Reduction of anxiety in children facing hospitalization and surgery by use of filmed modeling. J. Consult. Clin. Psychol.43:511–521.

MUMFORD, E., SCHLESIGNER, H. and GLASS, G.1982. The effects of psychological intervention on recovery from surgery and heart attacks: an analysis of the literature. Amer. J. Pub. Hlth72:141–151.

PETERSON, L. and SHIGETOMI, C.1981. The use of coping techniques to minimize anxiety in hospitalized children. Behav. Ther.12:1–14.

SKIPPER, J. and LEONARD, R.1968. Children, stress, and hospitalization: afield experiment. J. Hlth Soc. Behav.9:275–287.

TEASLEY, D.1982. Don't let cardiac catheterization strike fear in your patient's heart. Nursing12(3):52–55.

VERNON, D., FOLEY, J. and SPIWICZ, J.1965. The psychological responses of children to hospitalization and illness. Charles C Thomas, Springfield, Ill.

WOLFER, J. and VISINTAINER, M.1975. Pediatric surgical patients' and parents' stress responses and adjustment. Nurs. Res.24:244–255.

WOLFER, J. and VISINTAINER, M.1979. Prehospital psychological preparation for tonsillectomy patients: effects on children's and parents' adjustment. Pediatrics64(5):646–655.

Revised: November 1984