Adults demonstrate considerable variability in their use of health care services. Some persons delay in the lace of life-threatening complaints. Others appear in physicians' offices or emergency rooms for what professionals consider to be trivial complaints.2 As the costs associated with unnecessary use of ambulatory services increase, it may be important to understand the determinants of individuals' health-related behaviors. There is an enormous literature on adults' utilization of health care. In contrast, there are only a handful of inquiries into children's health-related beliefs and behaviors. This disparity is interesting, particularly if one assumes that some portion of adult utilization behaviors may originate in childhood.
Initial studies of adults examined the effects of demographic variables such as ethnicity, sex and socioeconomic status on symptom sensitivity and illness behavior. ~` Subsequently the health belief model has been developed, studied and embellished) Most of the theoretical approaches used to explain adults' use of services have acknowledged the positive association between cues and/or psychological distress and the frequency of visits to the doctor.6
A paraallel but related stream of literature reports the relations between stressftil events and the occurrence of disease-not just the use of services. Studies of adults have suggested that the frequency of occurrence of certain life events is associated with mental health status and a variety of acute and chronic medical disorders.7 As with utilization behavior studies, there have been few investigations of this type with children as subjects.
In the fowing sections we briefly and selectively review the literature on psychological distress in adults/thmilies and children's use of health services. We shall then describe our work conducted over the past 15 years on the determinants of children's healthrelated behaviors, their use of health services and sources of psychological distress.
The literature that reports the stresses of adults has been criticized for failure to provide an operational definition of stress. In this article we differentiate between "stress" and "psychological distress" as follows: both involve psychological reactions to unpleasant, or unexpected phenomena. Stress, per the original work of Cannon and the subsequent studies of Selve was associated with physiological changes and the biological concomitants of reactions to stimuli. We use "psychological" distress to indicate our primary concern with emotional or mental health, rather than the somatic consequences of stress.
"Worried-well" adults represent only 10% to 15% of a defined population. However, they make over 50% to 60% of all visits to physicians. The worried or distressed parent has received less attention. In an early study, Roghmann and Haggerty examined the relation of adult stress to children's illness.8 The children of parents who reported more "upsetting" events had more upper respiratory infections. Tessler and Mechanic studied mothers' psychological perception of stress and their children's use of physicians' services in a prepaid group practice.9 The families who reported higher levels of psychological distress took their children to physicians more frequently.
Are there worried-well children? Until quite recently, there seemed to be little interest in this question. In fact, the health beliefs and behaviors of children have received little attention. In 1971, we created an "adult-free system" in an elementary school. The school's health care policy allowed children ages 5 to 12 to leave the classroom without the teacher's permission to initiate visits to a school nursepractitioner. l0 In the first year of the study, we identified a sub-group of children (about 12% of the population) who accounted for over 50% of all visits to the nurse. These children had certain characteristics. More of them were girls. If they were under 7 years, their most frequent complaint was stomachache, beyond 7 years their most common complaint was headache. Teachers reported that most of these children were having academic or behavior problems in 'class, and difficulties in getting along with their peers. None of them had chronic medical problems.
We designed the study to involve children in the processes of their own care. After leaving the classroom to visit the nurse, the children provided a history and actively participated in identification of their problems. The nurse examined the children and shared the physical findings. At each visit children were asked, when did your problem start, what was going on, does anybody else at home have that, etc. I During the study, some children began to state spontaneously, "I have another New Math headache." Despite the ability to attribute their medical complaints to stressful academic situations, these children continued to seek services. After all, the health care system afforded the most effective (temporary) treatment for their discomfort. Unfortunately, the sick-role offers one of the few legitimate reasons to be excused from personal responsibilities in our society.
Children Define Distress
Most studies of stress on adults focus on life events. A large literature has developed linking the frequency of stressful events or those causing a change in daily routine with the onset of a variety of serious illnesses. A more recent analysis ofthat literature suggests that it is not the severity of the event, loss of a spouse, bankruptcy, etc. , but the frequency of or chronicity of "daily hassles" that seem to be associated with medical problems. n The measurement of stress in adults has revolved around the life event analysis approach, but a variety of questions have been raised regarding an actual definition of "stress," since the problems that seem injurious to some do not affect others.
As we reflected upon the worried-well children in our studies who were high users of services, we believed that they were "distressed. " A review of the literature on stress in childhood revealed only a few studies. AU used instruments containing items that aduks presumed would be stressful to children. Our experience suggested that children would be a better source of information about things that concerned them. Therefore, we developed a "feel bad" scale by asking children, "what happens that makes you feel bad, get upset or nervous?"12 This inventory of items consists primarily of events that might be considered minor, everyday problems related to relations with parents and peers. The events/conditions cited by children tended to be chronic, or continuing. Almost all would not be classified by adults as major life events.
Most items indicate the child's concern with a lack of respect or loss of self-esteem, ie, being the last one chosen on a team, or being left out of a group. The items related to parent-child relations include, not spending enough time with your parents, having to follow house-rules, etc. Factor analysis indicated the things that children cited as making them feel bad fell into three groups - relations with peers, relations with parents and "dislocations," ie, moving or changing schools. We were able to demonstrate an association between the "feel-bad" scores, and items reflecting the mental health status of these children (depression and anxiety). Children with the highest wfeel bad" scores were significantly more likely to indicate they felt sad, tired, worried and didn't like themselves.
In 1975, we replicated our child-initiated care study in four public schools, serving primarily a disadvantaged population. The results were similar, but ethnic differences in utilization were observed.. Children of Hispanic descent used the "adult-free" system significantly less often than Asian, Black or Caucasian children. However, the same portion (10% to 15%) of "high users" were identified.
An unanticipated event occurred during the replication of the child-initiated care study. With some unexpected funds, and without our knowledge, the school district began to provide counseling services in the schools. During the several months that the childinitiated care system operated, the nurses identified the high users in each school. The nurses referred their high utilizers to the counselors, who treated the children in group and individual sessions. A dramatic reduction in the number of visits to the nurses alerted us to the additional treatment of psychological services. While the provision of mental health services for adults has failed to decrease the utilization of health services by worried-well adults, it is tempting to believe that children, given a proper diagnosis, may be more responsive to specific treatments.
The data presented are consistent with the studies of Rogers et al, on "frequent" visitors to the health room. In his investigations, associations were found between high users, poor academic performance, high absenteeism, initiation of cigarette smoking and dropping out of school.13,14
We are continuing to examine the associations among children's independent use of health services, levels of psychological distress and their mental health. The data available to date suggest that the associations known to exist among adults are also present among children over the ages of 7 to 8 years. However, evidence from longitudinal studies of populations of children in varied settings would be necessary to generalize these phenomena.
We suggest that after physicians have assured themselves that a child has no underlying medical illness, and that the basic diagnosis is a child in "distress," that a referral to psychological services is warranted. Physicians who provide pediatric services might ask themselves the following questions:
1. How many children do I see who come (are brought) frequently or unnecessarily, ie, without underlying medical problems?
2. What do I know about the family situations or potential sources of distress in these children and their families?
3. Should I refer these children and their parents for counseling help?
We emphasize that children and their parents should be referred for help. Although the child is the patient, our work with children has documented the importance of dealing with the child within the family setting. 15 If there are to be any changes in children's and parents' behaviors, psychological assistance must be provided for both.
Why Should 1 Bother with Trivial Problems?
Why should anybody be concerned about worriedwell children? Perhaps administrators in pre-paid health organizations, or HMOs should be concerned about reducing unnecessary utilization. However, we believe a more substantive reason is to reduce the child's sense of dependency on the physician that is reinforced by repeated visits for socially distressing problems. Most importantly, if the basic lesion or underlying pathology in worried-well children is lack of self-esteem or self-concept, and we fail to provide appropriate treatment, then these children may live a long, unfulfilled life, without achieving their full potential. Our experience suggests that a "fractured" self-concept may heal with the appropriate treatment.
1. Hackett TP, Casaem MH, Raker JW: Patient delay in cancer. N Engl Med 1973; 289:14.
2. Frogart P, Merritt JD: Consultation in general practice - Analysis of individual frequencies. British Journal of Preventive Social Medicine 1969; 23:1-9.
3. Hetherington RW1 Hopkiru OE: Symptom sensitivity: Its social and cukuml correlates. Heakh Serv Res Spring, 1969; 63-75.
4. Stoeckle JD, Zola IK, Davidson CE: On going to see the doctor: The contributions of the patient to the decision to seek medical aid. J Chron Du 1963; 16:981.
5. Becker MH, Green LW: A family approach to compliance with medical treatment: A selective review of the literature. International Journal of Heakh Education 1975; 18:2-11.
6. Gortmaker SL, Eckenrode J, Gore S: Stress and the utilization of health services; A time series and cross-cultural analysis. J Health Soc Behau 1982; 23:25-38.
7. PaykelE, Myers JK, Dienert MN, et al: Life events and depression: A controlled study. AnAGeii Psychiatry 1969; 21:753-760.
8. Roghmann KJ. Haggerty RJ: Daily stress, illness and use of health services in young families, ftdiotr Res 1973; 7:353-364.
9. TesslerRD, Mechanic D, Dimond M: Theeffectofpsychologicaldistrcsson physician utilization: A prospective study. J Heakh Soc Behav 1976; 17:353-374.
10. Lewis CE, Lewis MA, Lorimer AA, et al: Child-initiated care: The use of school nursing services in an adult-free system. Pediatrics 1977; 60:499-507.
11. Pearlin Ll, Lieberman MA: Social sources of emotional distress. Re search Communications m Mental Heakh 1979; 1:217-248.
12. Lewis CE, Siegel JM, Lewis MA: Feeling bad; Exploring sources of distress among preadolescent children. Am J PiMc HeM 1984; 74(2). Il 7- 1 22.
13. Rogers KD, Reese G: Health studies - presumably normal high school students, II. Absence from school. Am i Dis Child 1965; 109:9-27.
14. Rogers KD, Reese O: Health studies - presumably normal high school students, Ul, Health room visits. Am J Dis Child 1965; 109:28-42.
15. Lewis CE. Rachelefsky G, Lewis MA, et al: A randomized trial of ACT (asthma cate training) for kids. Pediorrics October, 1984; 74(4):478-486.