Early Asthma Onset: Consideration of Parenting Issues

DAVID A. MRAZEK, M.D., M.R.C.PSYCH., MARY D. KLINNERT, PH.D., PATRICIA MRAZEK, PH.D., M.S. W., AND TERRI MACEY, mi.D., M.S. W.

Abstract. This report examines the relationship between early parental behavior and the later onset of asthma in a cohort of 150 children who were genetically at risk for developing asthma. Judgments of both parenting problems and maternal coping were made during a home visit when the infant was 3 weeks old. A clinical interview with the mother was developed and reliably coded. The sample was divided into two groups based on the presence or absence of concerns about coping and parenting. During the following 2 years, the respiratory status of the children was monitored. Four categories of respiratory status were defined: (1) asthma; (2) recurrent infectious wheezing; (3) a single isolated wheezing episode; or (4) no wheezing. Early problems in coping and parenting were associated with the later onset of asthma (p < 0.001). Furthermore, parents of children who developed asthma were more likely to have been having difficulties at the 3-week visit than those whose children developed infectious wheezing (p < 0.005). J . Am. Acad. Child Adolesc. Psychiatry, 1991, 30, 2:277-282. Key Words: asthma, early parenting, infancy.

Many studies have examined a wide range of psycholog- ical factors that influence the expression of asthmatic symp- toms (Mrazek, 1988). However, difficulties in the early parenting of children at genetic risk for developing asthma have not been linked to the initial onset of asthma. This is in large part because of the problems associated with col- lecting unbiased data designed to examine the association between a wide range of potential emotional stressors and the onset of the disease. To avoid the natural inclination to make retrospective judgments that provide an explanation for the onset of the illness, a prospective design is required. This report documents the results of such a prospective study of a sample of infants at increased genetic risk for devel- oping asthma. A highly significant statistical association was demonstrated between two ratings of parental behavior and the later onset of asthma.

It has become increasingly evident that some degree of genetic predisposition is a necessary condition for devel-

Accepted June 15, 1990. Dr. David Mrazek is a research scientist at the National Jewish

Center for Immunology and Respiratory Medicine and Professor of Psychiatry (Child) and Pediatrics at the University of Colorado Health Sciences Center. Dr. Klinnert is the Chief Pediatric Psychologist at the National Jewish Center for Immunology and Respiratory Medicine and Assistant Professor of Psychiatry (Psychology) at the University of Colorado Health Sciences Center. Dr. Patricia Mrazek is the Re- search Intervention Coordinator at the National Jewish Center for Immunology and Respiratory Medicine. Dr. Macey is an instructor in the University of Colorado Continuing Education Department.

This work was supported through grant No. 88-1013-8.5 from the W . T . Grant Foundation, grant No. 2KO2-MH00430 from the National Institute of Mental Health, and a grant from the Developmental Psy- chobiology Research Group.

The authors wish to thank Irene Anderson, Amy Brower. Florence Garyet, and David McCormick for their help in the preparation of the data, data analysis, and preparation of the manuscript.

Reprints may be requested from Dr. Mrazek, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson Street, Den- ver, Colorado 80206.

0890-8567/91/3002-0277$03 .OO/OO 1991 by the American Academy of Child and Adolescent Psychiatry.

oping asthma. However, simply having a genotypic vul- nerability does not result in phenotypic expression. Family studies have convincingly shown that there is an increased risk for the development of asthma for individuals who have affected relatives (Sibbald et al., 1980; Mrazek et al., 1990, submitted for publication). There is also an elevated risk for increased airway reactivity among these family members even in the absence of disease. A methacholine challenge is the current method of choice for the identification of hyperreactivity of the airways. Increasingly, large doses of methacholine are inhaled sequentially until a 20% drop in pulmonary function is produced. This technique provides an objective means of confirming a clinical diagnosis of asthma as well as identifying subclinically affected relatives. Using this technique, Longo et al. (1987) demonstrated that there was an elevated risk for increased methacholine sen- sitivity in nonsymptomatic relatives of asthmatic probands.

Nowhere is the critical interaction between nature and nurture more obvious than in the results of twin studies. These reports show both a high heritability of airway reac- tivity and a considerable degree of discordance between identical twins (Hopp et al., 1984). Given that the genomes of monozygotic twins are by definition identical, variation in the nonshared environment of monozygotic twins must necessarily be responsible for differential expression of a gene or set of genes that is ultimately involved in the onset of reactive airway disease.

Four classes of environmental risk factors have been hy- pothesized to affect gene activation. An enduring specula- tion has been that respiratory viral infections play a role in the pathogenesis of asthma (Busse, 1989). Evidence for this view includes clinical studies demonstrating that asthmatic children frequently experience attacks exacerbated by res- piratory syncytial virus (RSV) and parainfluenza viral in- fections (McIntosh et al., 1973), and that RSV infection has been demonstrated to precede the onset of asthma in a small sample of children at genetic risk for the development of asthma (Frick et al., 1979). Furthermore, an intriguing model for the “switching on” of a putative set of genes exists as

277 J . Am. Acad. Child Adolesc. Psychiatry, 30:2, March 1991

MRAZEK ET AL.

it is plausible that viral DNA could become incorporated into host cell nuclei and subsequently activate genes linked to expression of airway reactivity.

A second class of potential environmental activators is the wide range of specific antigens that can come in contact with the immunoregulatory components of the immune sys- tem. These allergens would include antigens derived from a small number of foods (e.g., milk, eggs, peanuts), dust mites, molds, animal danders, and animal saliva. A possible mechanism for the pathogenic effects of environmental an- tigens is that they may stimulate a classic antibody response involving immunoglobulin E antibodies that could result in a persistent increase in the reactivity of this system. Acti- vated antibody molecules attach to the surfaces of mast cells and basophiles, which leads ultimately to their degranula- tion. This results in the release of a wide array of mediators from these cells that have been demonstrated to have an influence on the regulation of airway reactivity.

A third class of hypothesized environmental activators are nonantigenic irritants. The most widely considered sub- stance in this category is smoke, although a wide range of air pollutants are plausible pathogenic activators. Irritants could directly affect the bronchial receptor systems or act through potentiating allergic mechanisms, resulting in greater reactivity to the antigens.

The fourth class of potential environmental risk factors is emotional stressors. The mechanism for the impact of stressors on gene expression has not been demonstrated. One possibility is that stressors could result in a shift in the balance of autonomic tone leading to increased bronchial sensitivity. Another hypothesis is that stressors could have a regulatory effect on the modulation of the immune re- sponse. A wide range of potential neuropeptides including substance P and vaso-intestinal protein have been shown to influence the immune response.

No prospective longitudinal study has conclusively dem- onstrated the etiological role of any of these factors for asthma onset. However, associations between the first three classes of risk factors and the development of asthma have been demonstrated. This report presents empirical evidence of a statistical association between early parental behavior and a later increase in the incidence of asthma in a cohort of genetically vulnerable infants. The implications of this link have not been determined, and this relationship should not be considered to be a causal one.

The Asthma Risk Study The W.T. GranVNational Jewish Center Asthma Risk

Study was designed to identify possible associations be- tween early environmental factors and subsequent asthma expression. Pregnant asthmatic women were recruited with the intention of measuring and monitoring possible risk fac- tors for the onset of asthma in their infants. The development of their infants was subsequently carefully documented. A comparison sample of pregnant nonasthmatic women was also recruited with the expectation that few of their infants would develop asthma. The study is truly prospective as the index infants were still in utero at the time that the data collection with their parents was initiated.

2 78

Method

SAMPLE

The index risk sample is the focus of this report and consists of 150 families who were living in the Metropolitan Denver area within 2 hours of the National Jewish Center for Immunology and Respiratory Medicine. All 150 mothers in the index risk sample were pregnant. Fourteen of the mothers (9.3%) had a documented history of asthma but had been symptom free for at least 2 years at the time of their entry into the study. Of the 136 mothers who were experiencing symptoms, 68 (45.3%) of the mothers required only intermittent medication to control their wheezing, while 68 (45.3%) were taking medication on a regular basis to control symptoms. Although the mothers in the study had a full range of severity of asthmatic symptoms, this sample does have a larger percentage of more severely asthmatic women than one would expect from an epidemiological study designed to include a representative sample of all the individuals within a catchment area who had asthma. Fur- thermore, at the 3-week visit, 82% of the mothers reported that their asthma was either in control or that their symptoms were easily resolved with a small modification of their med- ication dosage. In summary, this was not a severely impaired cohort of asthmatic women. Thirty-two of the fathers in the index sample had a first-degree relative with asthma. An additional 28 of the fathers in the index sample had asthma themselves.

Families who were planning to leave the Denver Met- ropolitan Area within 2 years of the birth of their infant were not included in this study. Additionally, families in which either parent had a severe psychiatric disturbance were excluded from the study. Only one family was actually not included in the study because of parental psychopath- ology. The mean scores of the standard clinical scales of the Minnesota Multiphasic Personality Inventory (MMPI) for the mothers and fathers in the sample varied from 49.72 (social introversion) to 57.60 (mania), which is a range that is typical of a nonclinical sample.

The index sample was 92% Caucasian and predominantly middle class (Hollingshead: I = 26%, I1 = 43.3%, I11 = 22%, IV = 8.7%; Hollingshead, 1975). At the time of the birth of the index child, the mean age of the mothers was 29.3, and the mean age of the fathers was 3 1.1. Determi- nation of the respiratory status of the children at 2 years of age was possible for all 150 index children as no families were lost to follow-up.

The comparison sample consisted of 30 families in which neither parent had asthma. No significant differences in ethnic background, socioeconomic status (SES), or parental age existed between the comparison and index samples. Similarly, the means of the MMPI scales did not vary be- tween comparison and index samples on any of the clinical scales except for the social introversion scale of the mothers. While the mean score of asthmatic mothers was slightly higher than the controls (52.4 versus 48.60), both mean scores were in the normal range.

J.Am. Acud. Child Adolesc. Psychiutly,30:2,March1991

EARLY ASTHMA ONSET

METHOD OF ASSESSMENT

Perinatal Variables The parental relationship is a critical aspect of the early

experience of the infant. In order to quantify maternal coping and parenting during the first weeks of life, two scales were developed to measure these variables. A guiding principle in the development of these scales was that emotional stres- sors for the primary caregiver could have a potential effect on the infant. However, the degree of impact of the stressors on the child would be largely determined by the ability of the caregiver to cope with them and to concurrently mod- ulate the emotional experience of the infant.

An interview-based format was chosen for the purpose of assessing parenting strategies and maternal coping. The mother and child were observed together during a home visit while the mother was interviewed using a newly de- veloped semistructured interview. One objective of this new interview was to provide an opportunity for the clinician to observe the efforts of the parent to modulate the child’s experience and to collect data reflecting parental sensitivity and competence. Another objective was to assess how the mother was coping with family stressors that were not re- lated to parenting. The parenting data in this report were derived from this interview with the mother and her infant in the family home 3 weeks after the birth of the baby. Given that this sample is not a psychiatrically disturbed cohort, measures were needed to focus on the variability of parenting and coping in relation to the issues of adapting to the new infant. The authors were aware of no standardized measure that was available to capture these aspects of early parenting. Thus, while standardized measures of parental personality and the quality of parental marital relationship would be expected to correlate with early measures of par- enting, these more general measures of psychopathology and adjustment would be expected to be less adequate pre- dictors of problems in infant development.

Parenting Interview Rating. The first variable was a global rating of concern regarding parenting based on the maternal interview. Essentially, six characteristics of both parents were taken into account in making this judgment. These included: (1) the attitude of the parents toward the new infant including their enthusiasm for the parenting role; (2) their sensitivity to the needs of the infant; (3) their effectiveness in responding to the infant; (4) the nature of the parents’ strategy for sharing parenting responsibilities; ( 5 ) any evi- dence of disturbed emotional adjustment that would impact upon caring for the infant including the presence of post- partum depression; and (6) adequacy of the plan of the parents to continue with their employment while providing adequate child care. Parenting was scored on a three-point scale. The sample was subsequently dichotomized into two categories. The better performing group experienced only minimal subjective problems and was doing well in all spheres. The problematical group was viewed as experi- encing either moderate or severe difficulties. Interrater re- liability for this categorical rating was greater than 76%.

Coping Interview Rating. The second rating was of “ma- ternal coping.’’ The interview explored three primary areas

of coping: (1) the mother’s current management of respon- sibilities within the family; (2) her ability to plan for the continuation of her own career objectives; and (3) a judg- ment of her degree of satisfaction with her current adjust- ment. The interview was first coded on a hierarchical five- point scale, and, subsequently, the sample was dichoto- mized into two groups. The group designated as coping well was functioning adequately in all three areas and experi- encing no chronic difficulties. The mothers designated as demonstrating problematical coping were having difficulties in one or more of the three domains of adaptation. The interrater reliability for this dichotomy was greater than 82%. It should be emphasized that this global rating is of a construct of behavior that is an important component of a traditional family assessment. Consequently, the adequate levels of interrater reliability achieved by clinicians after a training period of approximately 10 hours is possible be- cause of the fairly straightforward nature of these clinical judgments.

Temperament. A third global rating was designed to quan- tify the degree of difficulty of temperament of each of the infants at the 3-week interview. This judgment was based on both the maternal report of the child’s rhythmicity and ability to be soothed during the first 3 weeks of life and the direct observation of the infant. Ratings on a five-point hierarchical scale ranged from very difficult to very easy. The reliability of the differentiation of a dichotomy of “rel- atively difficult” versus “relatively easy” infants was 86%.

Standardized measures. The MMPI (Hathaway and McKinley, 1970) and the Dyadic Adjustment Scale (DAS) (Spanier, 1976) were administered to the parents during the final trimester of the pregnancy with the index infant. These standardized instruments measure some aspects of the per- sonality and adjustment of the parents that should be as- sociated with parenting. Consequently, examining the as- sociation between these scale scores and the newly developed global scales provided an opportunity to demonstrate aspects of the validity of the global scales. The correlation between these standardized instruments and the rating scales are re- viewed in the results section.

Subsequent Health Status The health status of the children was carefully monitored

over the first 2 years of their lives with the specific objective of documenting the occurrence of wheezing and respiratory illnesses. Three classes of reactive airway disease were de- fined for the purpose of characterizing respiratory illnesses with bronchoconstriction. Class I reactive airway disease included only children diagnosed as having “asthma” based on conservative diagnostic criteria, which included docu- mentation of recurrent wheezing episodes by the child’s pediatrician. Although many of these attacks occurred with a concurrent viral infection, at least one of these wheezing episodes must have been precipitated by an environmental trigger other than a respiratory infection to be classified as asthma. Class I1 reactive airway disease was labelled “in- fectious wheezing” and defined as multiple wheezing ep- isodes with every episode having been associated with a respiratory infection. This category would not be differ-

2 79 J . Am. Acad. Child Adolesc. Psychiatry, 30:2, March I991

MRAZEK ET AL.

entiated from ‘‘asthma’ ’ by many American practitioners, but it is widely used as a distinct diagnostic category in Britain. In essence, “infectious wheezing,” as used in this study, is distinguished from ‘‘asthma” by the requirement that all wheezing episodes must have been associated with documented infections. Class I11 reactive airway disease was a category reserved for children who had only experienced a single isolated episode of wheezing that was identified by the child’s physician. This single documented episode could have occurred with or without a respiratory infection. These children were considered to be at an elevated risk for sub- sequent attacks but would not be classified as having either “infectious wheezing” or “asthma” until a second episode was documented.

DATA ANALYSIS The relationships between early parental behavior and

subsequent infant illness were demonstrated by chi-square analyses. Comparisons of characteristics of the sample were done using chi-square and t-test analyses for categorical and continuous data, respectively.

RESULTS Asthma Status

Twenty-one of the 150 children at genetic risk for de- veloping reactive airway disease had been diagnosed with asthma (Class I) when they had reached 2 years of age. Fourteen additional children had experienced multiple ep- isodes of bronchoconstriction that had all been associated with viral infections and were consequently diagnosed as having infectious wheezing (Class 11). Twenty-one children had experienced only a single wheezing episode (Class 111), and 94 children had never wheezed.

Parenting The clinical judgment of “early parenting difficulties”

was found to be a predictor of asthma. This rating of the presence of early parenting difficulties was made based on concerns elicited during interview of some aspect of the ability of the parents to deal with the demands of their young infant. Fifty-two of the infants had parents who were rated as having problematical parenting, while 98 infants had parents who were felt to be adjusting well to the parenting role. Thirteen of the 52 infants whose parents were judged to be having some problems with parenting subsequently developed asthma (25%) as compared with only eight of the infants whose parents were perceived as parenting their infant well (8%) (p < 0.005).

Coping The maternal coping variable was designed to quantify

the ability of the mother to cope with a broader range of family stressors that extended beyond parenting. It included a judgment of how well the mother had managed a wide range of family relationships, her family’s economic real- ities, and her changing life circumstances. Sixty-seven of the mothers were having some difficulties in coping, and 13 of their infants subsequently developed asthma (19%).

280

Only eight children of the 83 mothers who were judged to be coping well developed asthma (10%). A trend reflecting an association between better coping and less infant disease was demonstrated 07 = 0.087).

Parental Dificulty The sample was subsequently divided into two categories

based on whether either problematical parenting or prob- lematical coping had been coded. Using this method, 76 mothers were classified as having early parenting difficul- ties. Figure 1 illustrates that 18 (24%) of these 76 infants whose parents were having difficulties developed asthma. Of those 74 infants whose parents were both coping and parenting well, only three (4%) children had developed asthma by 2 years of age. This difference is highly signif- icant (p < 0.001).

In contrast to this association between the presence of parenting or coping difficulties and the later onset of asthma, there was no association between parental difficulties and infectious wheezing. Specifically, only 29% of the parents of children who had experienced infectious wheezing, but in whom the problem had not progressed to asthma, were having parenting difficulties during the initial 3 weeks of life of the infant. In contrast, as illustrated in Figure 2, 86% of the parents of children who had developed asthma were having difficulties (p < 0.005).

Parenting Difficulties in the Comparison Group The incidence of parenting difficulties as determined by

the clinical interview was virtually identical between the control group families and the index families. Fifty-one percent of the asthmatic families were having some par- enting difficulties as compared with 53% of the control families.

Comparison of Parents with and without Dij5culties A set of analyses was conducted to examine the possible

differences in maternal age and SES between infants with parents having difficulties and those doing well based on the parenting and maternal coping codings that had been made when the infants were only 3 weeks old. No differ- ences in maternal age was evident between the two groups, as the mothers having difficulties had a mean age of 28.6 as opposed to the mothers who were doing well who had a

n = 74 n = 7 6 Panntal Difficultie~ Good Palontal Adaptation

FIG. 1. Four percent of the children from families who were coded as having good parental adaptation developed asthma in contrast with 24% of the children whose parents were coded as having parental difficulties. This relationship between the development of asthma and parental adaptation is highly significant (p < 0.001).

J . Am. Acad. Child Adolesc. Psychiatry, 30:2, March 1991

EARLY ASTHMA ONSET

onset in the infants of these women (p < 0.05). Given the large number of possible comparisons and the relatively modest level of significance of this association, little inter- ference from the finding of greater maternal dependency and asthma onset should be implied.

Discussion An association between parenting behavior and later ill-

ness expression in genetically at-risk children is in many ways intriguing, but it cannot be considered to be evidence of a causal link. The mechanism by which a particular pattern of parent-children interaction might eventually lead to an increased likelihood for the expression of airway reac- tivity cannot be elucidated from these analyses. From one perspective, it is quite surprising that a statistical association exists at a single point in time with the later onset of asthma. The fact that such an association does exist suggests, but does not definitively prove, that these judgments capture a more stable aspect of parental behavior. Further exploration of possible mechanisms would be a particularly interesting direction for future research inquiry.

One limitation of this report is that the findings are rel- evant only to the early onset of asthma. The future clinical course of the respiratory status of these children is as yet unknown. Some children may have a very circumscribed form of the illness and essentially be symptom free for the remainder of their childhood, while other children currently unaffected are likely to develop respiratory symptoms. Fol- lowing this cohort into the future to examine variations in the pattern of illness expression is yet another direction for future investigation.

The decision to make a home visit when the infant was 3 weeks of age was based on a variety of considerations. One advantage was that a postnatal visit would provide an opportunity to make an assessment of the early adaptation of the mother to the infant at a specific developmental period across the entire sample. All of the mothers were still at home with their infants, and there had been relatively few external factors to complicate the assessment of the mother's parenting. Additionally, a 3-week assessment ruled out the possibility that any of the children would be already dem- onstrating early airway reactivity, as it is an extremely un- usual occurrence for bronchoconstriction to occur within the first month of life. In that regard, this assessment of par- enting behavior would, in all likelihood, predate any com- plications resulting from the impact of the onset of asthma on the behavior of the parents. Finally, doing an assessment in the home had the advantage of maximizing the ability of the interviewer to assess the actual living circumstances of the mother and infant. Despite these advantages for choosing an early postnatal assessment, there is also the clear limi- tation that this single point does not capture later devel- opments.

Generalizations drawn from these data must be circum- spect. It should be emphasized that the characteristics of this sample are not that of a general population cohort. Specifically, the sample is primarily a middle- to upper middle-class white sample, and these results should not be assumed to be valid for disadvantaged cohorts. Addition-

281

mean age of 29.9. Furthermore, no significant differences in SES were shown between the two groups.

A further analysis was conducted to examine whether a possible difference in genetic risk existed between the two groups. Thirty (39%) of the children whose parents were having difficulty had high genetic loading based on the criteria of having asthma documented on the paternal side of the family and having a mother with asthma. Similarly, 30 (41%) of the infants whose parents were doing well had a high genetic loading for asthma.

Correlates of the Clinical Global Ratings A set of analyses to examine correlates with the clinical

global ratings judgments was conducted. Temperament. An association between parenting diffi-

culties and difficult temperament was also noted. Thirty- three infants (43%) of families where parenting difficulties were identified were coded as having difficult temperaments as opposed to 12 (16%) infants of the 74 families in which the parents were doing well (p < 0.001). While more chil- dren with difficult temperaments did develop asthma, the association between difficult temperament and illness expression was not significant.

Dyadic Adjustment Scale. Families without parenting dif- ficulties were more likely to have higher levels of marital satisfaction as measured by the DAS (p = 0.05). However, lower marital satisfaction was not significantly associated with later asthma onset in the children.

MMPI. Mothers who were having parenting difficulties were more likely to have elevated scores on the F scale (p < O.OOOl), depression scale (p < 0.002), and the psy- chopathic deviation scale (p < 0.02). Two derived scales were also associated with parenting difficulties. The mothers having problematical parenting and/or coping had elevated scores on the dependency scale (p < 0.0005) and lower scores on the ego strength scale (p < 0.02). They also have a lower score on the K scale (p < 0.005). While the means scores of all of the scales of the mothers with parental difficulties were within the normal range, variability within this range was significantly associated with the clinical rat- ings. These associations provide evidence of the concurrent validity for the clinical interview judgments. Interestingly, no clinical scale and only one derived scale, the dependency scale, of the MMPI was significantly associated with asthma

FIG. 2. Twenty-nine percent of the children who developed infectious wheezing were from families who were coded as having parental difficulties; whereas 86% of the parents of children who developed asthma were coded as having difficulties. This difference is highly significant (p < 0.005).

J . Am. Acad. Child Adolesc. Psychiatry, 30:2, March 1991

MRAZEK ET AL.

ally, the range of asthmatic symptoms represented in this sample was extensive, ranging from an asymptomatic his- tory of asthma in 9.3% of the index mothers to a requirement of continuous medication in 45.3% of the index mothers. Given the range of medication requirements in epidemio- logical samples, more seriously asthmatic women were overrepresented. One reason for the somewhat more seri- ously asthmatic women being referred to the study is that they may well have a greater motivation to participate in a study designed to better understand the expression of an illness that has affected their own lives. It is possible that the frequency of expression of asthma in the infants of more severely asthmatic women may be somewhat greater be- cause the genes associated with more severe illness may be more penetrant. However, this is strictly a speculation, as genetic risk studies of asthma have not demonstrated an association between the severity of the parental disease and a higher incidence of affected offspring. Similarly, in this study, there was no association between the severity of asthma in the parents and an increased risk for asthma expression in the infants.

Interestingly, the association between parenting difficul- ties and the subsequent development of asthma by 2 years of age in these infants would have been considerably ob- scured if a less rigorous definition of asthma had been cho- sen. It was striking that the mothers of children with infec- tious wheezing were actually doing very well. There is good evidence to suggest that all 35 children with a diagnosis of either asthma or infectious wheezing have a genetic com- ponent to their illness that puts them at increased risk for developing asthma. This is supported by the observation that all of these children responded to viral infections with the characteristic pattern of bronchoconstriction and wheez- ing that is the hallmark of asthma. However, it may be that in families in which parents are functioning at a highly adaptive level, these early viral infections can be resolved without a change in the subsequent airway reactivity of the child. In contrast, in a caregiving environment where par- enting difficulties exist, bronchoconstriction may be con- ditioned to the emotional distress associated with the attack.

While a variety of hypotheses can be put forward to explain the associations between early parenting difficulties and later airway reactivity, definitive conclusions regarding the true nature of underlying mechanisms are not possible without additional research and analyses.

This report provides empirical evidence documenting a link between early parenting behavior and the subsequent expression of asthma. If these findings are confirmed, the implications will be considerable as they would suggest that interventions designed to support the parenting of geneti- cally at-risk children may well result in a decrease in the expression of asthma.

References Busse, W. W. (1989). The relationship between viral infections and

onset of allergic diseases in asthma. Clin. Exp. Allergy, 19:l-9. Frick, 0. L., German, D. F., & Mills, J. (1979), Development of

allergy in children. I . Association with virus infections. J. Allergy Clin. Immunol. 63:228-241.

Hathaway, S. R. & McKinley, J. C. (1970), Minnesota Multiphasic Personality Inventory. Minneapolis: The University of Minnesota Press.

Hollingshead, A. B. (1975), Four Factor Index of Social Status. New Haven, CT: Available from Yale University Sociology Department.

Hopp, D. O., Bewtra, A. K. , Watt, G. D., Nair, N. M. & Townley, R. G. (1984), Genetic analysis of allergic disease in twins. J . Allergy Clin. Immunol., 731265-270.

Longo, G., Strinati, R., Poli, F. & Fumi, F. (1987), Genetic factors in nonspecific bronchial hyperreactivity. Am. J . Dis. Child., 41:331- 334.

McIntosh, K., Ellis, E. F., Hoffman, L. S. , Lybass, T. G., Eller, J. J. & Fulginiti, V. A. (1973), The association of viral and bacterial respiratory infections with exacerbations of wheezing in young asth- matic children. Pediatrics, 82:578-590.

Mrazek, D. A. (1988), Asthma: psychiatric considerations, evaluation, and management. In: Allergy: Principles and Practice, 3rd edition, eds. E. Middleton, C. E. Reed & E. F. Ellis. St. Louis: C. V. Mosby, pp. 1 1 7 6 1 196.

Sibbald, B., Horn, M. E. C. & Gregg, I . (1980), A family study of the genetic basis of asthma and wheezy bronchitis. Arch. Dis. Child., 55:354-357.

Spank , G. B. (1976), Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38:15-28.

From Early Issues of the Journal

In other words, as the constitutional bases for personality are better defined and the role of individual constitutional differences comes to be more commonly included in the child psychiatrist’s formulations about children, there will be an increasing use in the training of the child psychiatrist of the body of information available from pediatrics. . . . Pediatric specialties which can contribute to the training of the child psychiatrist include biochemistry, genetics, neurology, and neurophysiology. The sounder the grounding of the child psychiatrist in information about the constitution, the firmer will be his foundations.

JAACP, Vol. 1, No. 3 July, 1962

282 J.Am.Acad. ChildAdolesc. Psychiatry,30:2,March1991