- Email: [email protected]
Health-Related Quality of Life of HIV-Infected Children on Complex Antiretroviral Therapy at Home
Health-Related Quality of Life of HIV-Infected Children on Complex Antiretroviral Therapy at Home Mary Woods Byrne, PhD, FAAN Judy Honig, DrNP, EdD
The study describes quality of life as perceived by parents and measured by the Child Health Questionnaire (CHQ-28) for a convenience sample of 33 HIV-infected children age 5 to 18 years cared for at home on complex antiretroviral drug regimens. When compared with normative data on healthy U.S. children, parent reports for the HIV-infected children did not differ significantly on the following CHQ concepts: role/social related to emotional, behavioral or physical problems; bodily pain; behavior; mental health; self-esteem; impact on parental time; family activities; and family cohesion. Parents perceived their children to be functioning at a level significantly lower than the norm in general health (t ⫺6.47, p ⫽ .000), physical functioning (t ⫺2.37, p ⫽ .024), and physical summary (t ⫺3.80, p ⫽ .001). Parental impact-emotional was impressively higher than the norm (t ⫺3.74, p ⫽ .001). Differences were independent of gender, primary language, ethnicity, and Centers for Disease Control and Prevention clinical classifications for pediatric HIV disease. Key words: quality of life, health status, caregivers, children, HIV
During the past decade, the care of HIV-infected children has expanded from addressing the needs of acutely and terminally ill children to management of chronic illness (Clark & Byrne, 1993; Grubman et al., 1995). Protocols to prevent, recognize, and treat comorbidities as well as the availability of more antiretroviral pharmaceutical options have made pos-
sible longer and presumably more normalized lives spent primarily in family settings. Despite the many studies that have examined the pathology of HIV disease, the impact of medication therapies, developmental sequelae of HIV, and parent-child relationships in HIV-affected families, no studies have been located examining parent perception of the child’s health and well-being. This perception is an important aspect of family dynamics that can impact on the child’s treatment, adherence to a demanding medication protocol, and the child’s involvement in developmentally appropriate roles that enhance quality of life. Measuring the multiple dimensions of well-being can also provide an outcome perspective on the impact of treatment gains in terms of the child’s lifestyle. This study describes children’s quality of life as perceived by parents and expressed in multiple functional and role domains for a purposive sample of HIV-infected children observed long-term in a specialty clinic and cared for at home on complex antiretroviral drug regimens.
Background Children who are vertically infected with HIV make up a population that has experienced intense Mary Woods Byrne, PhD, FAAN, is professor of clinical nursing and Judy Honig, DrNP, EdD, is associate professor of clinical nursing, both at Columbia University School of Nursing.
JOURNAL OF THE ASSOCIATION OF NURSES IN AIDS CARE, Vol. 17, No. 2, March/April 2006, 27-35 doi:10.1016/j.jana.2006.01.003 Copyright © 2006 Association of Nurses in AIDS Care
28
JANAC Vol. 17, No. 2, March/April 2006
multidisciplinary health interventions during a time of important developments in treatment for the disease over the past decade. Children with HIV infection have complex needs across the disease spectrum. Caring for the child with HIV infection is demanding and very challenging to both clinicians and to the family caregivers who are primarily responsible for implementing care, often with little social support. Adherence to highly active antiretroviral therapy (HAART) drug protocols for viral suppression in HIV disease is a critical component of current disease management. Yet HAART regimens are composed of a very difficult plan of multiple doses, distasteful suspensions, and medications that must be taken precisely as prescribed and over a long period of time and therefore affect the child, the parent, and the family. Vertically transmitted HIV disease is a multigenerational family disease. Adherence to the prescribed antiretroviral medications not only poses challenges repeated daily, but it is influenced by family experiences with the disease process, decisions made about disclosure, the child’s developmental abilities, and family support (Byrne, Honig, Jurgrau, Heffernan, & Donahue, 2002).
Purpose The purpose of this study is to identify parents’ (or alternate primary caregivers’) perception of the wellbeing of their HIV-infected children who are on antiretroviral medication and to compare these perceptions with norms established for healthy children in the general population. Use of the Child Health Questionnaire (CHQ-28) in its American English and Spanish versions made it possible to determine data across ethnocultural groups.
History and Theoretical Foundations for Multidimensional Health Status Measurements The idea of multidimensional health status measurement is based on a view of health as composed of complementary dimensions including physical, mental, and social parameters. In concert with the World Health Organization definition of health, this per-
spective goes beyond a view of health as merely the absence of disease and moves toward a quality of life perspective. In early studies, rates of morbidity and mortality were used as indicators of health status. This limited perspective did not convey the quality of life or the person’s functioning in numerous ways as he or she undertook the roles considered appropriate for age and situation. The development of a health status instrument for children was launched with the work of Eisen, Ware, Donald, and Brook (1979). Their seminal research project attempted to construct a multiitem scale to measure health status in children 13 years and younger for the Health Insurance Study. The authors viewed health on three levels: physical, mental, and social. In addition, the integrative construct of general health perception was included as a cumulative subjective measure of health. Physical health was viewed as functional ability in activities that are common for a child in good health, such as self-care actions. Mental health was defined as positive and negative psychological states including mood and feelings. Social health was viewed as the quality of interpersonal relationships. General health was rated by proxy and included current health, resistance/ susceptibility, and prior health. More recently, three groups of researchers concurrently developed instruments that similarly aimed to quantify self-reported, multidimensional health in children (Starfield et al., 1993; Starfield et al., 1995; Wasson, Kairys, Nelson, Kalishman, & Baribeau, 1994). Continuing this genre of instrumentation, the CHQ survey instruments were developed using a process parallel to the short-form health survey (SF-36), a widely used selfreport assessment for adults (Ware & Sherbourne, 1992). The CHQ series was composed of several longer and shorter formats for parent and child report (Landgraf, Abetz, & Ware, 1996). The CHQ-28 was used to determine data in this study because of its theoretically sound development, a relatively short format to reduce subject burden, and availability in Spanish and English.
Byrne, Honig / QOL HIV-Infected Children
The Child Health Questionnaire Parent Form, Short Version (CHQ-28) The CHQ-28 was designed to measure multidimensional health for children 5 years and older and to be completed by the parent or guardian on paper or in interview format. Consistent with all CHQ versions, it incorporates 14 concepts, each of which comprises a scale: physical functioning; role/social physical; role/social-emotional; role/social-behavioral; bodily pain; general behavior; mental health; self-esteem; general health perceptions; change in health; parental impact-time; parental impact-emotional; family activities; and family cohesion. In the CHQ-28, each scale contains one to four items, and the concepts of social role limitations related to emotional and to behavioral problems are combined into one scale. Each item has between four and six rankordered, forced-choice responses indicating frequency of a functional limitation, quality of a health characteristic, or level of agreement with a statement about child health. Combinations of items are used to create two summary scores: physical summary and psychosocial summary. Instrument development for the CHQ-28 included testing for reliability and validity with healthy child samples (Landgraf et al., 1996). All items achieved or exceeded the .40 standard for the corrected itemtotal correlations method of internal consistency except the two parental impact-emotional items (.38 for each) and two of the four general health perception items (.19 and .34). Reliability by Cronbach’s alpha ranged from r ⫽ .69 to .89 for all scales except general health perceptions (r ⫽ .59) and parent impact/emotional (r ⫽ .54). Four scales were singleitem scales for which internal consistency could not be computed. Content validity was determined by comparing the items and concepts with those in six other published child and adolescent health assessment measures. Item discriminant validity was supported for all scales with more than one item. Factor analysis confirmed the conceptual framework of the CHQ and provided construct validity. Cross-cultural validity was determined through input of representatives from 11 countries/languages and application of international guidelines to rigorous translation processes.
29
Finally, because of the lack of a gold standard among the numerous health assessment instruments, norms were obtained based on healthy children in the national population to provide a frame of reference for interpretation (Landgraf et al., 1996). At the time the CHQ manual was published, there were no reports of clinical populations studied using the CHQ-28. The longer CHQ-50 (in which the CHQ-28 is embedded) had been used to create clinical profiles for children with attention deficit hyperactivity disorder, asthma, psychiatric disorders, juvenile rheumatoid arthritis, and epilepsy, and each of these was considered valid because the scores were logically representative of the respective disorder. The authors of this study established psychometrics for use of the CHQ-28 with HIV-infected children in this sample (Byrne & Honig, 2005). Internal consistency using Cronbach’s alpha was acceptable for a new scale for all but one of the eight multiple item scales (r ⫽ .62 to .83 for all but general health perceptions, r ⫽ .39) and for both the physical (r ⫽ .63) and psychosocial (r ⫽ .82) summary scales. Construct validity was established using item discriminant validity and hypothesis testing.
Setting and Sample The study setting was a comprehensive specialty clinic located in a major academic health sciences center in the northeastern United States. It offered a full array of patient services in a one-stop, family-centered context within a research environment. All HIV-seropositive children enrolled in this clinic who were on HAART regimens at the time of the study were eligible for inclusion in this study, which was part of a larger study on medication adherence (Byrne et al., 2002). Approval was received following full review of human subjects protection by the institutional review board of the medical center where the clinic is located. Of the 67 families who were eligible for the larger study, 43 caregivers were reached, and following informed consent, 33 who had children 5 years or older completed the CHQ-28. The average age of the children in this subset was 8.8 years, with a range of 5 to 18 years. Males made up 51% of this sample. Child ethnicity as reported by the caregiver was 33% African American, 37% Latino, 3% White, and 27%
30
JANAC Vol. 17, No. 2, March/April 2006
Table 1.
Sample Characteristics: Children With Perinatal HIV Infection on HAART (N ⴝ 33)
Age Gender Ethnicity (self-report)
Primary language in home Centers for Disease Control and Prevention classification (by clinical symptom) Medication burden: Daily antivirals Number of medications Number of daily doses Daily additional medications Number of medications Number of daily doses Time on any HIV antiviral Time on HAART
Mean (range) 8.8 years (5-18 years) Male 51% Latino 37% African American 33% Biracial, other 27% White 3% Spanish 31.4% English 63.6% N (asymptomatic) 12% A (mild) 9% B (moderate) 42% C (severe) 36% Mean (Range) 2.6 (1-4) 5.7 (2-10) 2.5 (0-8) 3.7 (0-14) 32 mo (1 mo-10 yr) 23 mo (1 mo-5 yr)
NOTE: HAART ⫽ highly active antiretroviral therapy.
biracial or other. Caregiver choice of instrument was Spanish for 12 families (36.4%) and English for 21 families (63.6%), which reflects the preferred language used at home. All children were infected with HIV through maternal vertical transmission during pregnancy. Three quarters of the children in the sample were moderately (Category B, 42%) or severely (Category C, 36%) symptomatic; 9% were mildly symptomatic (Category A); and 12% were asymptomatic (Category N), using the Centers for Disease Control and Prevention (CDC) classification system for pediatric HIV disease (Centers for Disease Control and Prevention [CDC], 1994). The prescribed medication regimens for these children included up to 12 daily medications, of which up to 4 were antiretrovirals (Table 1). Half of the families (49%) were single-parent families, and 70% of all participating families were composed of only two generations (parent and child) in one household. Twenty of the primary caregivers completed high school and six had some college education. Most of the caregivers (69%) were unemployed, and nearly all had publicly funded insurance
(New York State Medicaid). The birth mother was the primary caregiver for 30% of the children, whereas the grandmother was the primary caregiver for 24%. The other caregivers included kinship guardian (12%), foster mother (9%), adoptive mother (15%) and biological father (6%). Parent is used as a generic term throughout this paper to describe any of the primary caregivers. Clinicians identified 21 children (64%) as developmentally appropriate for age and 12 children (36%) with developmental delay or dysfunction. These categorizations were made initially by primary clinician’s screening and impression and corroborated by referral for developmental testing in 62% of cases. The majority of identified delays were described as neurological, speech, or behavioral.
Methods As the families came in for routine care, the primary care providers informed them of the study and referred those interested to a research assistant previously unknown to them. For a period of 10 months, this research assistant was available 4 of the 5 days that the clinic was in session each week. The research assistant was not available on the day the clinic session designed for adolescents took place, so this sample has proportionately less adolescents than were enrolled in the clinic. Availability of bilingual versions of CHQ-28 made it possible to determine data from both English-speaking and Spanish-speaking caregivers. In addition, the research assistant was bilingual (primarily Spanish speaking) and facilitated reading of questions and recording of answers for all families who needed this assistance.
Data Analysis Questionnaire responses were entered into SPSS-PC version 11.0 (SPSS, Inc, Chicago, IL), and a syntax file was created to simplify scoring them in accord with the directions in the CHQ manual (Landgraf et al., 1996). Reverse scoring, recalibrating, or both is required for many items, followed by transformation of all raw scores to standard scores with a mean of 50 and a standard deviation of 10. Descrip-
Byrne, Honig / QOL HIV-Infected Children Table 2.
31
Comparison of Study Sample With U.S. Population Norms on CHQ-28 Dimensions U.S. Population Norms (Landgraf, et al., 1996)
Study Sample N ⴝ 33
CHQ-28 Concept
Mean
SD
Mean
SD
T Value
p Value
Physical functioning Role/social-physical General health perceptions Bodily pain Family activities Role/social-emotional, behavior Parental impact-time Parental impact-emotional Self-esteem Mental health Behavior Family cohesion
95 93.7 74.0 81.3 91.1 92.5 88.4 81.3 80.1 79.7 70.8 72.4
16.2 19.7 19.8 19.7 18.9 19.1 20.9 18.4 19.1 15.5 18.7 21.6
84.2 90.5 50.5 80.63 83.33 86.73 83.85 59.09 81.99 83.60 64.38 71.06
26.20 22.89 20.53 21.24 28.07 25.06 24.50 34.13 21.53 17.01 27.33 29.12
⫺2.37 ⫺0.78 ⫺6.47 ⫺0.18 ⫺1.59 ⫺1.32 ⫺1.05 ⫺3.74 .49 1.28 ⫺1.33 ⫺0.264
.024 .440 .000a .859 .122 .195 .302 .001a .629 .211 .193 .793
Summary scores: Physical summary Psychosocial summary
53.2 51.2
9.5 9.1
47.10 48.50
8.49 10.22
⫺3.80 ⫺1.39
.001a .174
NOTE: CHQ ⫽ Child Health Questionnaire. a. Significant at alpha ⫽ .0038 based on Bonferroni correction for alpha of .05 with multiple Student’s t-tests.
tive statistics were used to identify the sample composition and demographics. Demographic variables, item scores, and scale scores were explored to identify data patterns, outliers, and relationship to normality. Spearman’s rho correlations were calculated for all scales. One-sample Student’s t-tests were used to make comparisons between this sample’s transformed scale and summary scores and those reported as norms for healthy children in the general U.S. population. Because multiple Student’s t-tests (14) were run, the Bonferroni correction was applied, and the originally set alpha of .05 was changed to the more stringent .0038 to determine statistical significance.
Results: Physical and Psychosocial WellBeing of a Sample of HIV-Infected Children This study’s main question was, “How well are HIV-infected children, who are cared for at home and implementing a complex HAART regimen, able to function in terms of multi-dimensional health?” For the most part the answer was surprisingly positive, regardless of the child’s age. Compared by
one-sample Student’s t-tests with the normative data in the CHQ database for children in the general U.S. population (Landgraf et al., 1996), parents’ perceptions of this sample of HIV-infected children did not differ significantly from the healthy norms on the following concepts: role/social related to emotional/ behavioral or physical limitations, bodily pain, behavior, mental health, self-esteem, parental time-impact, family activities, and family cohesion (Table 2). Parents did score these children somewhat lower than the norm on all these concepts except mental health and self-esteem, which were scored somewhat higher, but none of these differences were statistically significant. Concepts on which this sample of HIV-infected children were perceived by their parents to be significantly different from the healthy norm children included physical functioning (t ⫺2.37, p ⫽ .024), general health (t ⫺6.47, p ⫽.000), parental impactemotional (t ⫺3.74, p ⫽ .001), and the physical summary score (t ⫺3.80, p ⫽.001) (Table 2). After the Bonferroni correction for multiple Student’s ttests, the differences in physical functioning were no longer accepted as significant. In all these areas, the
32
JANAC Vol. 17, No. 2, March/April 2006
100 US Population Norms
Study Sample
90 80 70 60 50 40 30 20 10 0
Psychosocial Summary
Physical Summary**
Family Cohesion
Behavior
Mental Health
Self Esteem
Parental Impact/ Emotional*
Parental Impact Time
Role Social/Emotional
Family Activities
Bodily Pain
General Health Perceptions***
Role Social/Physical
Physical Function
* t = -3.47, p = 0.001 ** t = -3.80, p = 0.001 *** t = -6.47, p = 0.000
Figure 1. Comparison of study sample with U.S. population norms on Child Health Questionnaire (CHQ-28) scales.
direction of difference was that these children (or their parents in the case of emotional impact) did less well than the healthy norm children and parents (Figure 1). CHQ scores for individual and summary physical and psychosocial dimensions were also compared by gender, primary language, ethnicity, and CDC clinical classification for pediatric HIV disease, and no significant differences were found related to these factors.
Discussion The results of this study should be considered in terms of the physical and psychosocial well-being of these HIV-infected surviving children and the resources of their families. Considering the seriousness of HIV infection and the breadth of care required, these parent perceptions are heartening in terms of the normalization of life activities that both the children and their families seem to have achieved. Compared with perceptions of parents with healthy children, these parents perceived their children to be
experiencing no more social role limitations related either to emotional, behavioral, or physical limitations (although they acknowledged physical limitations to be significantly greater than the norm group). Parents also perceived that these children had no more bodily pain and no difference in behavior, mental health, or self-esteem than the general population. Considering the burden of social stigmatization associated with HIV disease from the earliest years of its emergence, the scores related to social roles and to self-esteem concepts are encouraging. Remarkably, these parents also reported no differences in impact on parent time, family activities, or family cohesion than did parents of children in the norm group of the general U.S. population. The positive outcomes perceived by parents on this wide range of concepts may reflect the comprehensive specialty clinic care they have been receiving over several years. Other studies have shown that if children infected perinatally with HIV do survive the early years of life, then they stay relatively well; that is, they can
Byrne, Honig / QOL HIV-Infected Children
live at home with minimal disruption to school, play, and other activities and roles that are part of normal childhood development. This does not mean that the children who survive are among the least ill children. With advances in disease management including the availability of more antiretroviral drug combinations for the pediatric population, the survivors include many children classified to the more serious clinical categories for HIV disease. For example, Thorne et al. (2002) reported in the European Collaborative Study the status of 34 vertically HIV-infected children who had survived 10 or more years. Their profile was much like this sample: median age was 11.4 years, and the group included 3 children classified by CDC criteria as N (nonsymptomatic), 6 as A (mildly symptomatic), 17 as B (moderately symptomatic), and 8 as C (severely symptomatic). Nevertheless, in 73 % of scheduled clinic visits, these children were asymptomatic and 29 were on three or more antiretrovirals. While the trend seems to be toward higher quality of functioning for these children, they still must face an astounding array of issues including the high risk of becoming orphaned, changing and multiple caregivers, the probable need to remain on complex medication regimens for a lifetime, and ongoing issues of disclosure, treatment, sexual health, transition to self-care, and life-choice decisions! The areas in which our study sample parents perceived deficits for their children or themselves are reflective of the seriousness of HIV disease and the burden on the primary caregiver as well as the child. This group of caregivers identified the disease-related threats to their children’s general health and their contribution to their children’s physical functioning limitations. Compared with perceptions of healthy children’s parents, they reported a major impact on their own emotional status with average scores 27.2 points lower (p ⫽ .001) than the norm group. This is not an unexpected toll, because these parents, many of whom are ill themselves, seem to be facilitating their children to achieve so much in the way of social roles, pain-free status, mental health, and self esteem while maintaining family cohesion and family activities as well. These caregivers have an overall optimistic perception of their children’s multidimensional functioning, but they do not seem to be in denial about the underlying seriousness of
33
the physical health threats. The difference in average scores for this sample and healthy norms was 23.5 points lower (p ⫽ .000) for general health perception. Low scores for this CHQ concept mean that the parent believes the child’s health is poor and likely to get worse (Landgraf et al., 1996). The next largest difference between sample and norm scores was 10.8 points lower (p ⫽ .024) for physical functioning. Low scores on this CHQ concept mean the child is much limited because of health in performing all physical activities, including self-care (Landgraf et al., 1996). However, the conservative interpretation previously noted is that this difference would not be considered statistically significant when the Bonferroni correction is applied to account for the effect on alpha when multiple Student’s t-tests are done. Average differences for all other concepts were much smaller and not statistically significant, ranging from ⫺7.7 to ⫹3.9 points discrepancy (p ⫽ .122 to .859) between the sample and norm children (Table 1). The parents’ focus on their children’s physical limitations may accurately reflect the toll of HIV disease. However, it is not entirely consistent with the fact that one third of this sample had been identified by the primary care practitioners as having some degree of developmental delay, primarily operationalized to psychosocial domains. It is not clear to what extent denial or overconfidence may be reflected in the parents’ positive perceptions of their children’s psychological and social status and the low impact on family functioning. There is no strong reason to doubt their perceptions, however. These are families who have coped with HIV disease across generations, have experienced hopeful treatment changes, and are receiving comprehensive services at a research-based specialty clinic in a major academic health sciences medical center. The caregivers’ resilience and potential fragility as they foster quality of life for their children in all domains merits continued support.
Limitations This study is limited by a convenience sample from one setting. Participants were limited to those available when the study team was present (notably, a small percentage of adolescents enrolled in the
34
JANAC Vol. 17, No. 2, March/April 2006
clinic was reached), and who opted to participate. Because this was part of a larger study, inclusion was limited to those children on HAART regimens. Nevertheless, the sample is a purposive one deliberately chosen from a site where a representative group of surviving children would be likely to be found in an urban epicenter of HIV disease. In 2003, New York State continued to have the most reported AIDS cases in children under 13 years old at diagnosis and the largest cumulative number of children under 13 years old living with HIV or AIDS in the United States (CDC, 2003). The demographics of our sample reflect disproportionately more Hispanic and fewer Black and White children than are reported to be living in the United States with HIV/AIDS. Direct demographic comparisons are difficult to make because our study sample self-reported ethnicity using categories that differ from those traditionally used for national statistics. Likewise, age range of this sample reflects age of surviving children, not age at diagnosis. Most children in this study were born at the height of and survived through the decline of perinatally transmitted AIDS cases in the United States. The study is also limited by reliance on a single data source, that of parent self-report about their children. This has the validating advantage of providing the parent perspective, but no check could be made on either unconscious or intended biases in reporting. The CHQ questions were completed independently by some and read to others, but the script recommended by the instrument developers was used in all cases. The instrument used has documented acceptable psychometrics overall and for most of its subscales for healthy children and for several clinical samples but has not been used before for HIV-infected children. The authors documented acceptable psychometrics for this clinical sample for the summary scores and for all scales with the exception of inadequate internal consistency of the general health perceptions scale (Byrne & Honig, 2005). This scale also had borderline reliabilities reported by its developers for healthy norm samples (Landgraf et al., 1996). The significant differences this study identified in this area for this clinical sample, therefore, might be interpreted with caution, although they are logically consistent with the clinical situation.
Summary This study of health-related quality of life in a purposive sample of HIV-infected children on HAART medication regimens offers hope concerning the ability of these children to reach for optimum development in their homes and communities. The multiple functional dimensions of the CHQ-28 showed, from the perspective of parents, many similarities between these seriously ill children and the normative group of healthy children in the general U.S. population. As could be expected, there were significant differences between the HIV-infected sample and the normative group in the domains of physical functioning, general health, and the physical summary score. It could be interpreted as encouraging that these children were scored so well from the parent perspective on a range of psychosocial dimensions and that the perceived impacts on parental time, family activities, and family cohesion were not perceived to be significantly different from the norm for healthy children. These observations seem to reflect the commitment and sacrifice being made by these parents for their children and also might suggest some heightened, albeit protective, parental optimism. It is worrisome that the parental emotional impact was reported to be so profoundly different from the norm for families with healthy children. These parents merit support to bear the burdens they take on to facilitate optimum development of their chronically and seriously ill children.
Acknowledgments This study was supported in part by the Columbia University School of Nursing. The first author was also supported by National Institutes of Health grant award KO1 NR00107 (Home Care Monitoring of HIV Exposed Children, M. Byrne, principal investigator.).
References Byrne, M. W., & Honig, J. (2005). Psychometrics of the Child Health Questionnaire (CHQ-28) used to measure quality of
Byrne, Honig / QOL HIV-Infected Children life in HIV-infected children. Quality of Life Research, 14, 1769-1774. Byrne, M. W., Honig, J., Jurgrau, A., Heffernan, S. M., & Donahue, M. C. (2002). Achieving adherence with antiretroviral medications for pediatric HIV disease. The AIDS Reader, 12, 151-154; 161–154. Centers for Disease Control and Prevention (1994) 1994 revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morbidity and Mortality Weekly Report, 43, 1-19. Centers for Disease Control and Prevention. (2003). HIV/AIDS surveillance report. Cases of HIV infection and AIDS in the United States, 2003. Vol.15. Atlanta, GA: Author. Clark, P. J., & Byrne, M. W. (1993). Clinical issues in long-term pediatric HIV disease. MCN: The American Journal of Maternal Child Nursing, 18, 164-167. Eisen, M., Ware, J. E. Jr, Donald, C. A., & Brook, R. H. (1979). Measuring components of children’s health status. Medical Care, 17, 902-921. Grubman, S., Gross, E., Lerner-Weiss, N., Hernandez, M., McSherry, G. D., Hoyt, L. G., et al. (1995). Older children and adolescents living with perinatally acquired human immunodeficiency virus infection. Pediatrics, 95, 657-663. Landgraf, J. L., Abetz, L., & Ware, J. (1996). The Child Health Questionnaire: User’s manual. Boston: The Health Institute, New England Medical Center. Starfield, B., Bergner, M., Ensminger, M., Riley, A., Ryan, S., Green, B., et al. (1993). Adolescent health status measurement: Development of the Child Health and Illness Profile. Pediatrics, 91, 430-435. Starfield, B., Riley, A. W., Green, B. F., Ensminger, M. E., Ryan, S. A., Kelleher, K., et al. (1995). The adolescent child health and illness profile. A population-based measure of health. Medical Care, 33, 553-566. Thorne, C., Newell, M. L., Botet, F. A., Bohlin, A. B., Ferrazin, A., Giaquinto, C., et al. (2002). Older children and adolescents surviving with vertically acquired HIV infection. Journal of Acquired Immune Deficiency Syndromes, 29, 396-401. Ware, J. E., Jr, & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Medical Care, 30, 473-483. Wasson, J. H., Kairys, S. W., Nelson, E. C., Kalishman, N., & Baribeau, P. (1994). A short survey for assessing health and social problems of adolescents. Dartmouth Primary Care Cooperative Information Project (The COOP). The Journal of Family Practice, 38, 489-494.
35
The study describes quality of life as perceived by parents and measured by the Child Health Questionnaire (CHQ-28) for a convenience sample of 33 HIV-infected children age 5 to 18 years cared for at home on complex antiretroviral drug regimens. When compared with normative data on healthy U.S. children, parent reports for the HIV-infected children did not differ significantly on the following CHQ concepts: role/social related to emotional, behavioral or physical problems; bodily pain; behavior; mental health; self-esteem; impact on parental time; family activities; and family cohesion. Parents perceived their children to be functioning at a level significantly lower than the norm in general health (t ⫺6.47, p ⫽ .000), physical functioning (t ⫺2.37, p ⫽ .024), and physical summary (t ⫺3.80, p ⫽ .001). Parental impact-emotional was impressively higher than the norm (t ⫺3.74, p ⫽ .001). Differences were independent of gender, primary language, ethnicity, and Centers for Disease Control and Prevention clinical classifications for pediatric HIV disease. Key words: quality of life, health status, caregivers, children, HIV
During the past decade, the care of HIV-infected children has expanded from addressing the needs of acutely and terminally ill children to management of chronic illness (Clark & Byrne, 1993; Grubman et al., 1995). Protocols to prevent, recognize, and treat comorbidities as well as the availability of more antiretroviral pharmaceutical options have made pos-
sible longer and presumably more normalized lives spent primarily in family settings. Despite the many studies that have examined the pathology of HIV disease, the impact of medication therapies, developmental sequelae of HIV, and parent-child relationships in HIV-affected families, no studies have been located examining parent perception of the child’s health and well-being. This perception is an important aspect of family dynamics that can impact on the child’s treatment, adherence to a demanding medication protocol, and the child’s involvement in developmentally appropriate roles that enhance quality of life. Measuring the multiple dimensions of well-being can also provide an outcome perspective on the impact of treatment gains in terms of the child’s lifestyle. This study describes children’s quality of life as perceived by parents and expressed in multiple functional and role domains for a purposive sample of HIV-infected children observed long-term in a specialty clinic and cared for at home on complex antiretroviral drug regimens.
Background Children who are vertically infected with HIV make up a population that has experienced intense Mary Woods Byrne, PhD, FAAN, is professor of clinical nursing and Judy Honig, DrNP, EdD, is associate professor of clinical nursing, both at Columbia University School of Nursing.
JOURNAL OF THE ASSOCIATION OF NURSES IN AIDS CARE, Vol. 17, No. 2, March/April 2006, 27-35 doi:10.1016/j.jana.2006.01.003 Copyright © 2006 Association of Nurses in AIDS Care
28
JANAC Vol. 17, No. 2, March/April 2006
multidisciplinary health interventions during a time of important developments in treatment for the disease over the past decade. Children with HIV infection have complex needs across the disease spectrum. Caring for the child with HIV infection is demanding and very challenging to both clinicians and to the family caregivers who are primarily responsible for implementing care, often with little social support. Adherence to highly active antiretroviral therapy (HAART) drug protocols for viral suppression in HIV disease is a critical component of current disease management. Yet HAART regimens are composed of a very difficult plan of multiple doses, distasteful suspensions, and medications that must be taken precisely as prescribed and over a long period of time and therefore affect the child, the parent, and the family. Vertically transmitted HIV disease is a multigenerational family disease. Adherence to the prescribed antiretroviral medications not only poses challenges repeated daily, but it is influenced by family experiences with the disease process, decisions made about disclosure, the child’s developmental abilities, and family support (Byrne, Honig, Jurgrau, Heffernan, & Donahue, 2002).
Purpose The purpose of this study is to identify parents’ (or alternate primary caregivers’) perception of the wellbeing of their HIV-infected children who are on antiretroviral medication and to compare these perceptions with norms established for healthy children in the general population. Use of the Child Health Questionnaire (CHQ-28) in its American English and Spanish versions made it possible to determine data across ethnocultural groups.
History and Theoretical Foundations for Multidimensional Health Status Measurements The idea of multidimensional health status measurement is based on a view of health as composed of complementary dimensions including physical, mental, and social parameters. In concert with the World Health Organization definition of health, this per-
spective goes beyond a view of health as merely the absence of disease and moves toward a quality of life perspective. In early studies, rates of morbidity and mortality were used as indicators of health status. This limited perspective did not convey the quality of life or the person’s functioning in numerous ways as he or she undertook the roles considered appropriate for age and situation. The development of a health status instrument for children was launched with the work of Eisen, Ware, Donald, and Brook (1979). Their seminal research project attempted to construct a multiitem scale to measure health status in children 13 years and younger for the Health Insurance Study. The authors viewed health on three levels: physical, mental, and social. In addition, the integrative construct of general health perception was included as a cumulative subjective measure of health. Physical health was viewed as functional ability in activities that are common for a child in good health, such as self-care actions. Mental health was defined as positive and negative psychological states including mood and feelings. Social health was viewed as the quality of interpersonal relationships. General health was rated by proxy and included current health, resistance/ susceptibility, and prior health. More recently, three groups of researchers concurrently developed instruments that similarly aimed to quantify self-reported, multidimensional health in children (Starfield et al., 1993; Starfield et al., 1995; Wasson, Kairys, Nelson, Kalishman, & Baribeau, 1994). Continuing this genre of instrumentation, the CHQ survey instruments were developed using a process parallel to the short-form health survey (SF-36), a widely used selfreport assessment for adults (Ware & Sherbourne, 1992). The CHQ series was composed of several longer and shorter formats for parent and child report (Landgraf, Abetz, & Ware, 1996). The CHQ-28 was used to determine data in this study because of its theoretically sound development, a relatively short format to reduce subject burden, and availability in Spanish and English.
Byrne, Honig / QOL HIV-Infected Children
The Child Health Questionnaire Parent Form, Short Version (CHQ-28) The CHQ-28 was designed to measure multidimensional health for children 5 years and older and to be completed by the parent or guardian on paper or in interview format. Consistent with all CHQ versions, it incorporates 14 concepts, each of which comprises a scale: physical functioning; role/social physical; role/social-emotional; role/social-behavioral; bodily pain; general behavior; mental health; self-esteem; general health perceptions; change in health; parental impact-time; parental impact-emotional; family activities; and family cohesion. In the CHQ-28, each scale contains one to four items, and the concepts of social role limitations related to emotional and to behavioral problems are combined into one scale. Each item has between four and six rankordered, forced-choice responses indicating frequency of a functional limitation, quality of a health characteristic, or level of agreement with a statement about child health. Combinations of items are used to create two summary scores: physical summary and psychosocial summary. Instrument development for the CHQ-28 included testing for reliability and validity with healthy child samples (Landgraf et al., 1996). All items achieved or exceeded the .40 standard for the corrected itemtotal correlations method of internal consistency except the two parental impact-emotional items (.38 for each) and two of the four general health perception items (.19 and .34). Reliability by Cronbach’s alpha ranged from r ⫽ .69 to .89 for all scales except general health perceptions (r ⫽ .59) and parent impact/emotional (r ⫽ .54). Four scales were singleitem scales for which internal consistency could not be computed. Content validity was determined by comparing the items and concepts with those in six other published child and adolescent health assessment measures. Item discriminant validity was supported for all scales with more than one item. Factor analysis confirmed the conceptual framework of the CHQ and provided construct validity. Cross-cultural validity was determined through input of representatives from 11 countries/languages and application of international guidelines to rigorous translation processes.
29
Finally, because of the lack of a gold standard among the numerous health assessment instruments, norms were obtained based on healthy children in the national population to provide a frame of reference for interpretation (Landgraf et al., 1996). At the time the CHQ manual was published, there were no reports of clinical populations studied using the CHQ-28. The longer CHQ-50 (in which the CHQ-28 is embedded) had been used to create clinical profiles for children with attention deficit hyperactivity disorder, asthma, psychiatric disorders, juvenile rheumatoid arthritis, and epilepsy, and each of these was considered valid because the scores were logically representative of the respective disorder. The authors of this study established psychometrics for use of the CHQ-28 with HIV-infected children in this sample (Byrne & Honig, 2005). Internal consistency using Cronbach’s alpha was acceptable for a new scale for all but one of the eight multiple item scales (r ⫽ .62 to .83 for all but general health perceptions, r ⫽ .39) and for both the physical (r ⫽ .63) and psychosocial (r ⫽ .82) summary scales. Construct validity was established using item discriminant validity and hypothesis testing.
Setting and Sample The study setting was a comprehensive specialty clinic located in a major academic health sciences center in the northeastern United States. It offered a full array of patient services in a one-stop, family-centered context within a research environment. All HIV-seropositive children enrolled in this clinic who were on HAART regimens at the time of the study were eligible for inclusion in this study, which was part of a larger study on medication adherence (Byrne et al., 2002). Approval was received following full review of human subjects protection by the institutional review board of the medical center where the clinic is located. Of the 67 families who were eligible for the larger study, 43 caregivers were reached, and following informed consent, 33 who had children 5 years or older completed the CHQ-28. The average age of the children in this subset was 8.8 years, with a range of 5 to 18 years. Males made up 51% of this sample. Child ethnicity as reported by the caregiver was 33% African American, 37% Latino, 3% White, and 27%
30
JANAC Vol. 17, No. 2, March/April 2006
Table 1.
Sample Characteristics: Children With Perinatal HIV Infection on HAART (N ⴝ 33)
Age Gender Ethnicity (self-report)
Primary language in home Centers for Disease Control and Prevention classification (by clinical symptom) Medication burden: Daily antivirals Number of medications Number of daily doses Daily additional medications Number of medications Number of daily doses Time on any HIV antiviral Time on HAART
Mean (range) 8.8 years (5-18 years) Male 51% Latino 37% African American 33% Biracial, other 27% White 3% Spanish 31.4% English 63.6% N (asymptomatic) 12% A (mild) 9% B (moderate) 42% C (severe) 36% Mean (Range) 2.6 (1-4) 5.7 (2-10) 2.5 (0-8) 3.7 (0-14) 32 mo (1 mo-10 yr) 23 mo (1 mo-5 yr)
NOTE: HAART ⫽ highly active antiretroviral therapy.
biracial or other. Caregiver choice of instrument was Spanish for 12 families (36.4%) and English for 21 families (63.6%), which reflects the preferred language used at home. All children were infected with HIV through maternal vertical transmission during pregnancy. Three quarters of the children in the sample were moderately (Category B, 42%) or severely (Category C, 36%) symptomatic; 9% were mildly symptomatic (Category A); and 12% were asymptomatic (Category N), using the Centers for Disease Control and Prevention (CDC) classification system for pediatric HIV disease (Centers for Disease Control and Prevention [CDC], 1994). The prescribed medication regimens for these children included up to 12 daily medications, of which up to 4 were antiretrovirals (Table 1). Half of the families (49%) were single-parent families, and 70% of all participating families were composed of only two generations (parent and child) in one household. Twenty of the primary caregivers completed high school and six had some college education. Most of the caregivers (69%) were unemployed, and nearly all had publicly funded insurance
(New York State Medicaid). The birth mother was the primary caregiver for 30% of the children, whereas the grandmother was the primary caregiver for 24%. The other caregivers included kinship guardian (12%), foster mother (9%), adoptive mother (15%) and biological father (6%). Parent is used as a generic term throughout this paper to describe any of the primary caregivers. Clinicians identified 21 children (64%) as developmentally appropriate for age and 12 children (36%) with developmental delay or dysfunction. These categorizations were made initially by primary clinician’s screening and impression and corroborated by referral for developmental testing in 62% of cases. The majority of identified delays were described as neurological, speech, or behavioral.
Methods As the families came in for routine care, the primary care providers informed them of the study and referred those interested to a research assistant previously unknown to them. For a period of 10 months, this research assistant was available 4 of the 5 days that the clinic was in session each week. The research assistant was not available on the day the clinic session designed for adolescents took place, so this sample has proportionately less adolescents than were enrolled in the clinic. Availability of bilingual versions of CHQ-28 made it possible to determine data from both English-speaking and Spanish-speaking caregivers. In addition, the research assistant was bilingual (primarily Spanish speaking) and facilitated reading of questions and recording of answers for all families who needed this assistance.
Data Analysis Questionnaire responses were entered into SPSS-PC version 11.0 (SPSS, Inc, Chicago, IL), and a syntax file was created to simplify scoring them in accord with the directions in the CHQ manual (Landgraf et al., 1996). Reverse scoring, recalibrating, or both is required for many items, followed by transformation of all raw scores to standard scores with a mean of 50 and a standard deviation of 10. Descrip-
Byrne, Honig / QOL HIV-Infected Children Table 2.
31
Comparison of Study Sample With U.S. Population Norms on CHQ-28 Dimensions U.S. Population Norms (Landgraf, et al., 1996)
Study Sample N ⴝ 33
CHQ-28 Concept
Mean
SD
Mean
SD
T Value
p Value
Physical functioning Role/social-physical General health perceptions Bodily pain Family activities Role/social-emotional, behavior Parental impact-time Parental impact-emotional Self-esteem Mental health Behavior Family cohesion
95 93.7 74.0 81.3 91.1 92.5 88.4 81.3 80.1 79.7 70.8 72.4
16.2 19.7 19.8 19.7 18.9 19.1 20.9 18.4 19.1 15.5 18.7 21.6
84.2 90.5 50.5 80.63 83.33 86.73 83.85 59.09 81.99 83.60 64.38 71.06
26.20 22.89 20.53 21.24 28.07 25.06 24.50 34.13 21.53 17.01 27.33 29.12
⫺2.37 ⫺0.78 ⫺6.47 ⫺0.18 ⫺1.59 ⫺1.32 ⫺1.05 ⫺3.74 .49 1.28 ⫺1.33 ⫺0.264
.024 .440 .000a .859 .122 .195 .302 .001a .629 .211 .193 .793
Summary scores: Physical summary Psychosocial summary
53.2 51.2
9.5 9.1
47.10 48.50
8.49 10.22
⫺3.80 ⫺1.39
.001a .174
NOTE: CHQ ⫽ Child Health Questionnaire. a. Significant at alpha ⫽ .0038 based on Bonferroni correction for alpha of .05 with multiple Student’s t-tests.
tive statistics were used to identify the sample composition and demographics. Demographic variables, item scores, and scale scores were explored to identify data patterns, outliers, and relationship to normality. Spearman’s rho correlations were calculated for all scales. One-sample Student’s t-tests were used to make comparisons between this sample’s transformed scale and summary scores and those reported as norms for healthy children in the general U.S. population. Because multiple Student’s t-tests (14) were run, the Bonferroni correction was applied, and the originally set alpha of .05 was changed to the more stringent .0038 to determine statistical significance.
Results: Physical and Psychosocial WellBeing of a Sample of HIV-Infected Children This study’s main question was, “How well are HIV-infected children, who are cared for at home and implementing a complex HAART regimen, able to function in terms of multi-dimensional health?” For the most part the answer was surprisingly positive, regardless of the child’s age. Compared by
one-sample Student’s t-tests with the normative data in the CHQ database for children in the general U.S. population (Landgraf et al., 1996), parents’ perceptions of this sample of HIV-infected children did not differ significantly from the healthy norms on the following concepts: role/social related to emotional/ behavioral or physical limitations, bodily pain, behavior, mental health, self-esteem, parental time-impact, family activities, and family cohesion (Table 2). Parents did score these children somewhat lower than the norm on all these concepts except mental health and self-esteem, which were scored somewhat higher, but none of these differences were statistically significant. Concepts on which this sample of HIV-infected children were perceived by their parents to be significantly different from the healthy norm children included physical functioning (t ⫺2.37, p ⫽ .024), general health (t ⫺6.47, p ⫽.000), parental impactemotional (t ⫺3.74, p ⫽ .001), and the physical summary score (t ⫺3.80, p ⫽.001) (Table 2). After the Bonferroni correction for multiple Student’s ttests, the differences in physical functioning were no longer accepted as significant. In all these areas, the
32
JANAC Vol. 17, No. 2, March/April 2006
100 US Population Norms
Study Sample
90 80 70 60 50 40 30 20 10 0
Psychosocial Summary
Physical Summary**
Family Cohesion
Behavior
Mental Health
Self Esteem
Parental Impact/ Emotional*
Parental Impact Time
Role Social/Emotional
Family Activities
Bodily Pain
General Health Perceptions***
Role Social/Physical
Physical Function
* t = -3.47, p = 0.001 ** t = -3.80, p = 0.001 *** t = -6.47, p = 0.000
Figure 1. Comparison of study sample with U.S. population norms on Child Health Questionnaire (CHQ-28) scales.
direction of difference was that these children (or their parents in the case of emotional impact) did less well than the healthy norm children and parents (Figure 1). CHQ scores for individual and summary physical and psychosocial dimensions were also compared by gender, primary language, ethnicity, and CDC clinical classification for pediatric HIV disease, and no significant differences were found related to these factors.
Discussion The results of this study should be considered in terms of the physical and psychosocial well-being of these HIV-infected surviving children and the resources of their families. Considering the seriousness of HIV infection and the breadth of care required, these parent perceptions are heartening in terms of the normalization of life activities that both the children and their families seem to have achieved. Compared with perceptions of parents with healthy children, these parents perceived their children to be
experiencing no more social role limitations related either to emotional, behavioral, or physical limitations (although they acknowledged physical limitations to be significantly greater than the norm group). Parents also perceived that these children had no more bodily pain and no difference in behavior, mental health, or self-esteem than the general population. Considering the burden of social stigmatization associated with HIV disease from the earliest years of its emergence, the scores related to social roles and to self-esteem concepts are encouraging. Remarkably, these parents also reported no differences in impact on parent time, family activities, or family cohesion than did parents of children in the norm group of the general U.S. population. The positive outcomes perceived by parents on this wide range of concepts may reflect the comprehensive specialty clinic care they have been receiving over several years. Other studies have shown that if children infected perinatally with HIV do survive the early years of life, then they stay relatively well; that is, they can
Byrne, Honig / QOL HIV-Infected Children
live at home with minimal disruption to school, play, and other activities and roles that are part of normal childhood development. This does not mean that the children who survive are among the least ill children. With advances in disease management including the availability of more antiretroviral drug combinations for the pediatric population, the survivors include many children classified to the more serious clinical categories for HIV disease. For example, Thorne et al. (2002) reported in the European Collaborative Study the status of 34 vertically HIV-infected children who had survived 10 or more years. Their profile was much like this sample: median age was 11.4 years, and the group included 3 children classified by CDC criteria as N (nonsymptomatic), 6 as A (mildly symptomatic), 17 as B (moderately symptomatic), and 8 as C (severely symptomatic). Nevertheless, in 73 % of scheduled clinic visits, these children were asymptomatic and 29 were on three or more antiretrovirals. While the trend seems to be toward higher quality of functioning for these children, they still must face an astounding array of issues including the high risk of becoming orphaned, changing and multiple caregivers, the probable need to remain on complex medication regimens for a lifetime, and ongoing issues of disclosure, treatment, sexual health, transition to self-care, and life-choice decisions! The areas in which our study sample parents perceived deficits for their children or themselves are reflective of the seriousness of HIV disease and the burden on the primary caregiver as well as the child. This group of caregivers identified the disease-related threats to their children’s general health and their contribution to their children’s physical functioning limitations. Compared with perceptions of healthy children’s parents, they reported a major impact on their own emotional status with average scores 27.2 points lower (p ⫽ .001) than the norm group. This is not an unexpected toll, because these parents, many of whom are ill themselves, seem to be facilitating their children to achieve so much in the way of social roles, pain-free status, mental health, and self esteem while maintaining family cohesion and family activities as well. These caregivers have an overall optimistic perception of their children’s multidimensional functioning, but they do not seem to be in denial about the underlying seriousness of
33
the physical health threats. The difference in average scores for this sample and healthy norms was 23.5 points lower (p ⫽ .000) for general health perception. Low scores for this CHQ concept mean that the parent believes the child’s health is poor and likely to get worse (Landgraf et al., 1996). The next largest difference between sample and norm scores was 10.8 points lower (p ⫽ .024) for physical functioning. Low scores on this CHQ concept mean the child is much limited because of health in performing all physical activities, including self-care (Landgraf et al., 1996). However, the conservative interpretation previously noted is that this difference would not be considered statistically significant when the Bonferroni correction is applied to account for the effect on alpha when multiple Student’s t-tests are done. Average differences for all other concepts were much smaller and not statistically significant, ranging from ⫺7.7 to ⫹3.9 points discrepancy (p ⫽ .122 to .859) between the sample and norm children (Table 1). The parents’ focus on their children’s physical limitations may accurately reflect the toll of HIV disease. However, it is not entirely consistent with the fact that one third of this sample had been identified by the primary care practitioners as having some degree of developmental delay, primarily operationalized to psychosocial domains. It is not clear to what extent denial or overconfidence may be reflected in the parents’ positive perceptions of their children’s psychological and social status and the low impact on family functioning. There is no strong reason to doubt their perceptions, however. These are families who have coped with HIV disease across generations, have experienced hopeful treatment changes, and are receiving comprehensive services at a research-based specialty clinic in a major academic health sciences medical center. The caregivers’ resilience and potential fragility as they foster quality of life for their children in all domains merits continued support.
Limitations This study is limited by a convenience sample from one setting. Participants were limited to those available when the study team was present (notably, a small percentage of adolescents enrolled in the
34
JANAC Vol. 17, No. 2, March/April 2006
clinic was reached), and who opted to participate. Because this was part of a larger study, inclusion was limited to those children on HAART regimens. Nevertheless, the sample is a purposive one deliberately chosen from a site where a representative group of surviving children would be likely to be found in an urban epicenter of HIV disease. In 2003, New York State continued to have the most reported AIDS cases in children under 13 years old at diagnosis and the largest cumulative number of children under 13 years old living with HIV or AIDS in the United States (CDC, 2003). The demographics of our sample reflect disproportionately more Hispanic and fewer Black and White children than are reported to be living in the United States with HIV/AIDS. Direct demographic comparisons are difficult to make because our study sample self-reported ethnicity using categories that differ from those traditionally used for national statistics. Likewise, age range of this sample reflects age of surviving children, not age at diagnosis. Most children in this study were born at the height of and survived through the decline of perinatally transmitted AIDS cases in the United States. The study is also limited by reliance on a single data source, that of parent self-report about their children. This has the validating advantage of providing the parent perspective, but no check could be made on either unconscious or intended biases in reporting. The CHQ questions were completed independently by some and read to others, but the script recommended by the instrument developers was used in all cases. The instrument used has documented acceptable psychometrics overall and for most of its subscales for healthy children and for several clinical samples but has not been used before for HIV-infected children. The authors documented acceptable psychometrics for this clinical sample for the summary scores and for all scales with the exception of inadequate internal consistency of the general health perceptions scale (Byrne & Honig, 2005). This scale also had borderline reliabilities reported by its developers for healthy norm samples (Landgraf et al., 1996). The significant differences this study identified in this area for this clinical sample, therefore, might be interpreted with caution, although they are logically consistent with the clinical situation.
Summary This study of health-related quality of life in a purposive sample of HIV-infected children on HAART medication regimens offers hope concerning the ability of these children to reach for optimum development in their homes and communities. The multiple functional dimensions of the CHQ-28 showed, from the perspective of parents, many similarities between these seriously ill children and the normative group of healthy children in the general U.S. population. As could be expected, there were significant differences between the HIV-infected sample and the normative group in the domains of physical functioning, general health, and the physical summary score. It could be interpreted as encouraging that these children were scored so well from the parent perspective on a range of psychosocial dimensions and that the perceived impacts on parental time, family activities, and family cohesion were not perceived to be significantly different from the norm for healthy children. These observations seem to reflect the commitment and sacrifice being made by these parents for their children and also might suggest some heightened, albeit protective, parental optimism. It is worrisome that the parental emotional impact was reported to be so profoundly different from the norm for families with healthy children. These parents merit support to bear the burdens they take on to facilitate optimum development of their chronically and seriously ill children.
Acknowledgments This study was supported in part by the Columbia University School of Nursing. The first author was also supported by National Institutes of Health grant award KO1 NR00107 (Home Care Monitoring of HIV Exposed Children, M. Byrne, principal investigator.).
References Byrne, M. W., & Honig, J. (2005). Psychometrics of the Child Health Questionnaire (CHQ-28) used to measure quality of
Byrne, Honig / QOL HIV-Infected Children life in HIV-infected children. Quality of Life Research, 14, 1769-1774. Byrne, M. W., Honig, J., Jurgrau, A., Heffernan, S. M., & Donahue, M. C. (2002). Achieving adherence with antiretroviral medications for pediatric HIV disease. The AIDS Reader, 12, 151-154; 161–154. Centers for Disease Control and Prevention (1994) 1994 revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morbidity and Mortality Weekly Report, 43, 1-19. Centers for Disease Control and Prevention. (2003). HIV/AIDS surveillance report. Cases of HIV infection and AIDS in the United States, 2003. Vol.15. Atlanta, GA: Author. Clark, P. J., & Byrne, M. W. (1993). Clinical issues in long-term pediatric HIV disease. MCN: The American Journal of Maternal Child Nursing, 18, 164-167. Eisen, M., Ware, J. E. Jr, Donald, C. A., & Brook, R. H. (1979). Measuring components of children’s health status. Medical Care, 17, 902-921. Grubman, S., Gross, E., Lerner-Weiss, N., Hernandez, M., McSherry, G. D., Hoyt, L. G., et al. (1995). Older children and adolescents living with perinatally acquired human immunodeficiency virus infection. Pediatrics, 95, 657-663. Landgraf, J. L., Abetz, L., & Ware, J. (1996). The Child Health Questionnaire: User’s manual. Boston: The Health Institute, New England Medical Center. Starfield, B., Bergner, M., Ensminger, M., Riley, A., Ryan, S., Green, B., et al. (1993). Adolescent health status measurement: Development of the Child Health and Illness Profile. Pediatrics, 91, 430-435. Starfield, B., Riley, A. W., Green, B. F., Ensminger, M. E., Ryan, S. A., Kelleher, K., et al. (1995). The adolescent child health and illness profile. A population-based measure of health. Medical Care, 33, 553-566. Thorne, C., Newell, M. L., Botet, F. A., Bohlin, A. B., Ferrazin, A., Giaquinto, C., et al. (2002). Older children and adolescents surviving with vertically acquired HIV infection. Journal of Acquired Immune Deficiency Syndromes, 29, 396-401. Ware, J. E., Jr, & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Medical Care, 30, 473-483. Wasson, J. H., Kairys, S. W., Nelson, E. C., Kalishman, N., & Baribeau, P. (1994). A short survey for assessing health and social problems of adolescents. Dartmouth Primary Care Cooperative Information Project (The COOP). The Journal of Family Practice, 38, 489-494.
35