OBJECTIVE: The goal was to examine the impact of illness on families and the long-term effects on the health of parents of young adults (YAs) who were born with extremely low birth weight (ELBW), compared with normal birth weight (NBW) control subjects.
METHODS: A longitudinal cohort study was performed. Participants were mothers of eligible ELBW and NBW YAs. Information was obtained with well-validated questionnaires.
RESULTS: At young adulthood, 130 (81%) of 161 ELBW group and 126 (89%) of 141 NBW group mothers participated. There were no significant differences in scores between groups with respect to marital disharmony, family dysfunction, maternal mood, state anxiety, social support, depression, and maternal physical and mental health. The finding of no differences was unchanged when 27 YAs with neurosensory impairment (NSI) were excluded, except for family dysfunction scores, which were paradoxically lower for families with YAs with NSI. Although the impact scores revealed that significantly more parents of ELBW YAs were negatively affected with respect to their jobs and educational or training opportunities, mothers of ELBW YAs reported that the experience of caring for their child brought their family closer together and that relatives and friends were more helpful and understanding, compared with mothers of NBW YAs. Significantly more mothers of ELBW YAs with NSI, compared with those without NSI, felt better about themselves for having managed their child's health.
CONCLUSION: It seems that, by young adulthood, there is a minimally negative long-term impact of having an ELBW child in the family, regardless of the presence of NSI.
WHAT'S KNOWN ON THIS SUBJECT:
A common theme in medical and social literature is that coping with prematurity and disabilities may be highly stressful and may have a significant adverse impact on the entire family, particularly the mother.
WHAT THIS STUDY ADDS:
By young adulthood, the few differences between the ELBW and NBW groups that were reported at adolescence seemed to have been ameliorated, except for a continuing negative impact on parents' jobs.
The birth of a premature infant is recognized as a stressful event for the family. Several investigators reported on the early adverse impact of prematurity on the psychological and emotional health of families,1 which persisted over the first 3 years of life.2 Further follow-up monitoring of the latter cohort to school age revealed that the significant stress among mothers of very low birth weight infants continued and there was less advancement in educational attainment than for mothers of term infants.3 Low socioeconomic status (SES) and low IQ of the child added to maternal stress across all domains. However, the mothers did not differ in psychological distress symptoms, divorce rates, perceptions of family cohesion, or sense of mastery in parenting. In fact, mothers of children at high risk with greater caregiving demands reported more personal growth and mastery.3,4
There are a limited number of studies on the impact of prematurity at school age. Taylor et al5 measured family functioning among parents of very low birth weight children at 11 years of age and reported more long-term burden and parental psychological distress, in comparison with parents of children born at term. In addition, there seemed to be a gradient of family effects, with greater effects for the <750-g birth weight group; the outcomes of the 750- to 1499-g group fell between those of <750-g and children born at term. Consistent with other studies, these negative consequences were more frequent among families of low SES and those with a child with neurosensory impairment (NSI) or behavioral problems. Follow-up monitoring to adolescence showed that the family burden and parental psychological distress were higher in the <750-g group than in the term-born group but varied with the child's age and family environment.6 It seemed that, at the older ages of 13 and 14 years, there was a moderating effect and the differences were smaller, except in low-income families. The authors reiterated that, although the negative impact was greater in the <750-g group, ∼29% of those families reported no burden or psychological distress. In a more-recent cohort of extremely low birth weight (ELBW), school-aged children born in Cleveland in the 1990s, the total family impact was found to be greater in the premature group, compared with the control group.7 Again, however, both low SES and neurodevelopmental problems predicted negative family impact.
We showed previously that, at adolescence, parents of ELBW children seemed to have adjusted fairly well to their work and family life.8 Although there were some stresses, parents of ELBW children reported more-positive feelings about their family and themselves than did parents in the control group.8 Little is known, however, about the impact of prematurity on families during the transition to young adulthood or the long-term effects on the health of parents. In this study, we compared the impact of illness on families of young adults (YAs) who were born with ELBW with findings for families in a normal birth weight (NBW) reference group. We also investigated whether the negative impact was greater specifically for parents of YAs with NSI and whether there were changes in family burden between adolescence and young adulthood. On the basis of our findings at adolescence, we hypothesized that there would be little if any residual differences in impact at adulthood between ELBW and NBW families.
Parents of ELBW survivors who were born between 1977 and 1982 in a geographically defined region in central-west Ontario and were monitored longitudinally from birth were approached again to participate in the study.9 Parents of 8-year-old children born at term, who were matched to the index children with respect to age, gender, and social class, were recruited to serve as control families.10 For the current study, parents of both groups were recalled when the YAs were between 22 and 25 years of age.
Parent participants completed the following self-administered questionnaires. The sociodemographic information questionnaire inquired about parents' age and marital status, family composition, and mother's education, employment status, and occupation. Social class was classified according to the Hollingshead 4-factor Social Class Index.11
Maternal mood was measured with the Bradburn Affect Balance Scale, which is a 10-item rating scale consisting of 5 statements reflecting negative feelings and 5 statements reflecting positive feelings and focusing on the experiences of the previous 6 months of the person's life. Scores range from 0 (lowest affect balance) to 10 (highest affect balance).12 Marital disharmony was measured with 4 items from the Ontario Child Health Study Questionnaire, which inquired about the quality of relationship as indicated by the frequency of positive (sign of affection) versus negative (quarreling) affect, with higher scores indicating greater disharmony. Scores range from 4 to 22.13
The Spielberger State-Trait Anxiety Inventory measures state anxiety by using a 20-item questionnaire, with scores ranging from 20 to 80.14 Higher scores indicate increased anxiety. The general functioning subscale of the Family Assessment Device measures the overall health/pathologic condition of the family, with scores ranging from 12 to 48. Higher scores indicate greater family pathologic conditions (scores of ≥26 indicate pathologic conditions).15 The Center for Epidemiologic Studies Depression scale measures anxiety and depression, with scores ranging from 0 to 60 (scores of =16 indicate depression).16
The Social Support Index is a 17-item instrument that uses a 5-point Likert scale (from strongly disagree to strongly agree). Scores range from 0 to 68. This instrument records the degree to which families integrate into the community, view the community as a support, and feel the community can provide emotional, self-esteem, and network support.17 Higher scores indicate more support. The Health Survey Short Form 36 (SF-36) is a multipurpose survey of general health status, which yields a profile of 8 concepts and a summary of physical and mental health measures.18 Raw scores are transformed into scores ranging from 0 to 100 for each dimension (concept), from the least-favorable to most-favorable outcome.
The Impact of Child's Illness on the Family is a 17-item questionnaire that includes work-, marriage-, and family-related issues and items derived from the Child General Health Survey,19 the National Health Survey,20 the Survey of Disabled Children,21 and the Impact on the Family questionnaire, which measures the impact of the child's illness on the parents' emotional suffering, health, meeting own needs, job/work, and relationships with others (including siblings).22 This questionnaire also was used at adolescence.8 The Spielberger State-Trait Anxiety Inventory,14 Center for Epidemiologic Studies Depression scale,16 Social Support Index,17 and SF-3618 were not administered at adolescence; therefore, comparisons at adulthood were restricted to the measures that were administered at both ages.
The study was approved by the research ethics board of McMaster University and Hamilton Health Sciences, and written informed consent was obtained from all parents.
The χ2 test of significance was used to assess differences in categorical variables between the ELBW and NBW mothers, and analysis of variance was used to compare mean differences between groups. Regression analyses were used to determine the effect of SES on marital disharmony, family dysfunction, and social support. Data presented are inclusive of parents of children with NSI. Although exact P values are provided where applicable, a value of ≤.01 was used as a more-conservative estimate of significance, to account for multiple testing.
Demographic Characteristics of Study Participants
Data collection was conducted between January 2002 and April 2004. At young adulthood, parents of 142 (88%) of 161 ELBW subjects and 130 (92%) of 141 NBW subjects from the original groups participated. For consistency with our previous assessments, we report on the results for mothers only, that is, 130 (81%) of 161 ELBW subjects and 126 (89%) of 141 NBW subjects. No significant differences between groups were noted for the maternal sociodemographic variables listed in Table 1. Most mothers in both groups were white, were married, spoke English at home, and were of upper or middle SES. There were no statistically significant differences between groups in current marital status or in the proportions of mothers who were ever widowed, separated, or divorced.
Table 1 also shows the characteristics of both groups of YAs. At young adulthood, 35 (27%) of 126 subjects in the ELBW group had ≥1 NSI, compared with 3 (3%) of 126 subjects in the NBW group (P < .0001); however, there were no differences in terms of general health. Approximately 60% of YAs in both groups were living at home; only 11% of subjects in the ELBW group and 12% of subjects in the NBW group had less than a high school education.
Impact on Marital Disharmony, Family Dysfunction, Social Support, Maternal Mood, Anxiety, and Depression
With respect to marital disharmony, family dysfunction, and social support scores, there were no significant differences between the groups (Table 2). However, paradoxically, mothers of ELBW YAs with NSI reported significantly less family dysfunction than did mothers of ELBW YAs without NSI (P = .006). Comparisons of the current maternal mood, anxiety, and depression scores according to group and according to NSI of the YAs showed no statistically significant differences between groups for any of the measures.
On the basis of data from all mothers, there was a significant positive association between SES and social support (P = .005) but not between SES and marital disharmony or family dysfunction. These findings were similar (P = .01) when data were analyzed for mothers of ELBW YAs with and without NSI.
Impact on Family Relationships and Work-Related Effects
Mothers in the ELBW group reported that their child's health brought the family closer together and that relatives and friends were more understanding and helpful, compared with mothers in the NBW group (P ≤ .0001) (Table 3). Within the ELBW group, significantly more mothers of ELBW YAs with NSI felt better about themselves for having managed their child's health, compared with mothers of ELBW YAs without NSI (P = .001). Mothers also were asked to rate the work-related impact attributable to the health of the YAs. Mothers in the ELBW group reported more negative effects on their work (P = .01 to <.0001) and on their spouses' work (P = .01) than did mothers in the NBW group. However, there were no significant differences between mothers of ELBW YAs with and without NSI.
Impact on Maternal Physical and Mental Health
There were no reported differences between maternal groups with respect to measures of current physical and emotional health (SF-36) (Table 4). Within the ELBW groups, there were no differences regardless of the presence of NSI in the YAs.
Mothers were asked to rate whether the physical health and emotional well-being of the YAs in the past 6 months had an impact on their own emotional suffering or worry and on their personal time. Results in Fig 1 demonstrate that there were no significant differences according to group or according to the presence of NSI in the YAs.
In this study of the impact of prematurity, our findings were consistent with our hypothesis that, at young adulthood, there would be no differences in scores between the ELBW and NBW parental groups with respect to marital disharmony, family dysfunction, family impact, maternal mood, state anxiety, social support, maternal depression, or maternal physical and mental health. The only differences in family function between the groups in our study were within the ELBW cohort. Ironically, mothers of children with NSI reported less family dysfunction, compared with mothers of children without NSI. One possible explanation may be that families of children with NSI experienced more crises, which might have brought them closer together.
In an earlier study of this cohort at adolescence, there were mixed reports of the effects of the child's health on the parents' marriage; in the ELBW group, more parents reported additional stresses and strains, compared with the NBW group, and reported more negative impact on the siblings as a result of less parental attention.8 Also at adolescence, more parents in the ELBW group reported that the experience of having an ELBW child brought the partners closer together than did parents in the NBW group.8 These findings were consistent with the observations at young adulthood.
At young adulthood, there were no group differences between the ELBW and NBW groups with respect to improved feelings of self for the parents. However, there was a statistically significant difference within the ELBW group; mothers of YAs with NSI felt better about themselves, compared with mothers of YAs without NSI. At adolescence, however, there were group differences in how the parents felt about themselves; the ELBW group parents felt better about themselves than did the NBW group parents.8
At young adulthood, the jobs of both mothers and their spouses were more negatively affected in the ELBW group. Mothers also reported that this negative effect was greater in the past. This finding was unchanged from the adolescent assessment, at which more ELBW group parents reported that they were unable to take a job because of their child's health, compared with NBW group parents.8 These findings support those of Parish et al,23 in which families of YAs with intellectual disabilities were found to have lower average household incomes and lower family savings and the mothers were less likely to have ever been employed full-time or to have been employed for >5 consecutive years.
Both at young adulthood and at adolescence, this cohort reported no differences in the proportions of parents in the ELBW and NBW groups who had emotional suffering or worry because of their child's physical or emotional health or in the time available for their personal needs.8 At adolescence, however, significantly larger proportions of ELBW group parents felt that their own emotional health was negatively affected by their child's illness and also reported other negative effects on the family.8 At young adulthood, these differences between the ELBW and NBW groups were no longer apparent.
In the interest of completeness of the literature review, a few studies on the impact of disabilities at late adolescence and adulthood on the health of the primary caregivers are relevant for discussion, although they involve nonpremature populations. Brehaut et al24 investigated the physical and psychological health of a large cohort of parents of children with cerebral palsy in Ontario, Canada, through self-completed questionnaires and face-to-face interviews, and compared the data with findings from 2 large-scale, Canadian surveys, the National Population Health Survey and the National Longitudinal Study of Children and Youth. The findings were consistent in demonstrating that caregivers of adolescents with disabilities reported greater psychological distress and emotional problems and were more likely to report a variety of chronic health problems. They also were less likely to be engaged in full-time work and had lower family incomes. In a series of articles using National Longitudinal Study of Children and Youth data, the Ontario investigators reported that caregivers of children 4 to 11 years of age with health problems25 and neurologic and behavioral problems26 reported significantly greater chronic conditions, greater activity limitations, greater depressive symptoms, and poorer general health than did caregivers of healthy children. The poorer physical and psychological health of caregivers, primarily mothers, was strongly influenced by the child's behavior, demands of caregiving, and family function.27 A study of families with YAs with intellectual disabilities found that behavior and mental health problems at young adulthood were associated with negative impacts on the family, as perceived by the mothers.28
The current study is unique in many respects. First, to our knowledge, no other studies assessed the impact of prematurity on families beyond the middle of childhood and adolescence. Second, our participants represented the most-vulnerable group of infants, with birth weights of <1000 g. Third, this was a population-based study sample, and the responses were obtained consistently from the same respondents (ie, mothers). Fourth, the longitudinal nature of the study provides information on whether the impact is long-lasting. It seems that parental coping strategies and adaptations continue to play a role, so that by young adulthood any significant impact noted earlier has been ameliorated.
Some limitations need to be acknowledged. Our comparison group of term-born children was recruited at school age; therefore, we do not have information on this cohort from birth. Although a standardized approach with validated, self-completed questionnaires was used for both time periods, we cannot discount the possibility that mothers might have underreported negative feelings because of a social desirability bias.4,29 As in most other studies, another bias was the lack of perceptions of the fathers, the children themselves, and their siblings. However, we did study the perceptions of the premature YAs and showed no differences in terms of their self-reported quality of life, in comparison with the NBW group.30
The literature reveals contradictory findings on the impact of disability on families, with some studies showing significant adverse effects and others showing mixed effects, both positive and negative. Our findings are in contrast to studies that showed a continued significant negative family impact of having a child with a disability.23,24,–,28 However, the latter cohorts were composed entirely of children with disabilities, whereas, in studies of premature populations, only approximately one-fourth to one-third of children had impairments. Might it be that parents of premature infants adapted differently because they were unique in having experienced a long period of uncertainty regarding the survival of their infants, having developed bonding and camaraderie with other parents in similar situations during the prolonged period of neonatal hospitalization, and having received significantly greater support from follow-up programs and the community during infancy and childhood? These speculations have not been tested.
We cannot entirely explain what factors might have contributed to this positive long-term adaptation. We were able to demonstrate that there was a significant positive association between SES and social support among both cohorts of mothers, as well as among mothers of ELBW YAs with or without NSI, but not between SES and marital disharmony or family dysfunction. Our parents were predominantly white and more advantaged in their education and social class than were parents in other studies of premature infants,5 and they also were beneficiaries of universal health care and treatment centers for children with disabilities. Therefore, our findings are generalizable only to populations with similar demographic characteristics and health care systems. We must acknowledge the remarkable resilience of the parents and the sacrifices they have made in supporting their premature infants. It seems that by adulthood there is a minimally negative, long-term impact of having an ELBW child in the family, regardless of the presence of NSI.
The study was supported by the Canadian Institutes of Health Research (grant MOP42536) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant 1 RO1 HD40219).
We are extremely grateful to the ELBW and NBW YAs and their parents for their generosity in participating in our ongoing studies for the past 25 years. We thank Liz Merz for her diligence in tracing the subjects, Lorraine Hoult and Mary Lou Schmuck for statistical analyses, Alaina Wilkins and Melissa Johnson for data entry and accuracy checks, and Diane Turcotte for typing of the manuscript. We appreciate the support of the Department of Pediatrics and Children's Hospital, McMaster University.
- Accepted February 25, 2010.
- Address correspondence to Saroj Saigal, MD, FRCPC, McMaster University, Department of Pediatrics, Room 4G40, 1200 Main St W., Hamilton, ON L8N 3Z5, Canada. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
Funded by the National Institutes of Health (NIH) and the Canadian Institutes of Health Research.
- YA =
- young adult •
- ELBW =
- extremely low birth weight •
- NBW =
- normal birth weight •
- NSI =
- neurosensory impairment •
- SES =
- socioeconomic status •
- SF-36 =
- Short Form 36
- Moore M,
- Taylor HG,
- Klein N,
- Minich N,
- Hack M
- Drotar D,
- Hack M,
- Taylor G,
- Schluchter M,
- Andreias L,
- Klein N
- Hollingshead AS
- Bradburn NM
- Spielberger CD
- Keyser D,
- Sweatland R
- Devins G,
- Orme C
- McCubbin H,
- Thompson A,
- McCubbin M
- Spiker B
- Ware JE Jr.
- Landgraf JM,
- Abetz L,
- Ware JE Jr.
- Hamill PV,
- Drizd TA,
- Johnson CL,
- Reed RB,
- Roche AF
- Bone M
- Lee SK,
- Penner PL,
- Cox M
- Brehaut JC,
- Kohen DE,
- Raina P,
- et al
- Raina P,
- O'Donnell M,
- Rosenbaum P,
- et al
- Saigal S,
- Stoskopf B,
- Pinelli J,
- et al
- Copyright © 2010 by the American Academy of Pediatrics