Changes in Self-reported Quality of Life as Survivors of Extremely Preterm Birth Enter Adulthood

Advances in perinatal care of mothers and newborn infants have led to large increases in survival after extremely preterm birth. Currently, with proactive perinatal care, almost one-third of infants born at 22 weeks of gestation survive.¹ Interest continues to grow in longterm outcomes among survivors of extreme prematurity. Extreme prematurity is associated with high rates of perinatal brain injury, chronic lung disease, retinopathy, and systemic infections and contributes disproportionately to the burden of neurodevelopmental impairments, such as cognitive impairment and accompanying functional impairments. Specific impairments have been the focus of most longitudinal research on outcomes after extreme prematurity. An alternative and complementary approach is evaluation of more global outcomes as subjectively assessed by study participants (and their proxy reporters). One such outcome, health-related quality of life (HRQL), was the focus of a study reported in this volume of The Journal by Ni et al.²

HRQL is an aspect of quality of life that reflects the impact of illness from the perspective of the patient.³ Qualify of life encompasses complex multidimensional concepts, and no consensus has been reached on its definition. The World Health Organization defines quality of life as “the individual’s perception of their position in life in the context of culture and value systems in which they live and in relation to their goals, expectations, standards, and concerns.”⁴ Pioneering work by Saroj Saigal et al at McMaster University in Hamilton, Ontario provides a rich foundation for studies of HRQL after extreme prematurity. The Hamilton cohort is a geographically based sample of infants born with extremely low birth weights (ELBW) in 1977–1982. In this cohort, Saigal et al have evaluated quality of life at multiple ages, most recently in the fourth decade of life. Evaluations of HRQL were based on preference-based measures, the Health Utilities Index Mark 2 and 3 (HUI2, HUI3), elicited directly from patients by interviewers using a standard gamble approach anchored in economic and decision theory.⁵ HUI3 health states consist of 8 elements: vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain.⁶ The multi-attribute utility (MAU) score represents a single utility score derived from the health states elicited using the HUI3 and an algorithm based on a community survey of Hamilton, Ontario, Canada conducted in the 1990s. Over 20studies have used HUI2 or HUI3 internationally from midchildhood to adulthood.⁷ This groundbreaking work on HRQL opens a door to understanding the impact of extreme prematurity from the individuals affected.

The group studied by Saigal et al, infants with ELBW, are a subset of extremely preterm births with over-representation of fetal group restriction; hence, the need for studies of cohorts born extremely preterm. The extremely preterm cohort studies by Ni et al comprise the EPICure cohort, a population-based sample born in the United Kingdom in 1995. HRQL was measured as MAU scores and then compared young adults born extremely preterm (<26 weeks of gestation) with controls born at full term. In addition, parent-reported scores were compared with scores reported by the young adults, and scores reported by parents when their children were 11 years old were compared with scores reported when the children were 19 years old. The major findings were that (1) MAU scores were lower for the extremely preterm group; (2) adults born extremely preterm who had impairments, and their parents, rated their HRQL lower than the young adults born extremely preterm who did not have impairments; (3) MAU decreased between 11 and 19 years; and (4) parents’ ratings of their child’s HRQL were lower than their children’s ratings of their own HRQL.

Previous studies, including systematic reviews, have described mildly decreased HRQL in samples of low birth weight, extremely low birth weight, and preterm samples.^7–9 Many HRQL studies include infants born with ELBW in the presurfactant era,⁸ an intervention that could influence HRQL. An Australian cohort born in the postsurfactant era did not show any differences in HRQL between groups with extreme prematurity and controls, though many individuals with neurodevelopmental impairments were lost to follow-up.¹⁰ Perhaps because of less selective attrition, in their postsurfactant sample, Ni et al found lower HRQL among individuals born extremely preterm compared with term controls.

International comparison of HRQL measured with HUI3 is complex. Comparing ELBW samples from Canada, Bavaria, and the Netherlands, adolescents with ELBW in the Netherlands had higher quality of life in cognition and speech domains.⁸ Ni et al reported higher HRQL in individuals with ELBW compared with samples from Canada, Bavaria, and the Netherlands, but term controls also rated HRQL quite higher than Canadian term controls.^6,8,2 Notably, all studies using the HUI3 questionnaire utilize an algorithm based on a community survey of Hamilton, Ontario, Canada conducted in the 1990s, which may decrease generalizability.⁷

Ni et al describe lower HRQL scores for individuals with neurodevelopmental impairments, although scores were higher than many health care professionals would predict.² Median self-reported scores at 19 years of age were higher than those reported in early adulthood in Canadians with ELBW, which may be due to shifts in attitudes about disability over time or across cultures, or could be related to advances in care.⁶ Of note, clinicians rate health conditions more negatively than adolescents born preterm or full term and their parents.¹¹ This finding has been labeled the “disability paradox” whereby individuals with impairments or chronic illnesses report an unexpectedly high quality of life.¹² A possible explanation is that individuals with impairments find meaning and purpose in ways that able-bodied individuals might not understand.³

Another finding reported by Ni et al is that parent-reported mean HRQL was lower than self-reported mean HRQL at 19 years of age, consistent with research indicating that proxy respondents report lower quality of life than affected individuals themselves.¹³ Although most experts value self-report higher than proxy report, consideration of multiple perspectives seems appropriate, given the importance of family impact.^5,14 The youngest age at which self-report is appropriate is unclear. The International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Task Force report does not provide age cut-offs due to varying maturity but does posit that children under the age of 5 years are not able to provide reliable self-reports.¹⁵

Consistent with a prior Canadian study,⁶ Ni et al observed a decrease in parent-reported HRQL from 11- and 19-year old individuals, but only in children born extremely preterm and not controls born at term.² The authors postulate this finding could be due to difficulties in the transition to adulthood. A systematic review provides evidence that HRQL improves over time, beginning in preschool through adulthood.⁹ Thus, continuing study of individuals born extremely preterm is warranted across the lifespan, including assessments of HRQL in the interval between infancy and adolescence, when few studies of quality of life have been completed.^7,9,14

Among the strengths of the study by Ni et al are the use of a population-based cohort, the inclusion of controls born at term, and assessments of HRQL during both adolescence as well as adulthood. As pointed out by the authors, limitations include the lack of self-reported assessments of HRQL during adolescence, exclusion of individuals with severe impairments (for who parent proxy reports could have been collected), the lost to follow-up rate of 58%, and the use of communitybased utilities, which might not have reflected the same utilities as perceived by study participants and their parents.

The report by Ni et al adds to the rich body of information arising from the EPICure studies and provides a narrow window into the personal lives of infants born extremely preterm.¹² Larger studies are needed, as well as studies to identify factors, in addition to functional status, associated with HRQL. Also important is the exploration of methods for measuring HRQL with minimal participant burden to allow frequently repeated measurements on large samples. Of note, the National Institutes of Health-sponsored Environmental Influences on Child Health Outcomes is collecting data on a construct termed positive child health, which is based on several short questionnaires and accounts for biological, functional, behavioral, and/or experiential assets,¹⁶ thereby aligning with the Institute of Medicine’s definition of health as children’s ability and potential to (1) develop and realize their potential; (2) satisfy their needs; and (3) develop the capacities that allow them to interact successfully with their biological, physical, and social environments.”¹³ The promise of this field of investigation is a better understanding of the life course of those born prematurely from their own perspectives, providing guidance and hope, for clinicians, policymakers, and families.

Glossary

ELBW

Extremely low birth weights

HUI2

Health Utilities Index Mark 2

HUI3

Health Utilities Index Mark 3

HRQL

Health-related quality of life

MAU

Multi-attribute utility