Early Childhood Education Journal

, Volume 39, Issue 2, pp 95–102

Academic Outcomes for Children Born Preterm: A Summary and Call for Research

Authors

    • Department of Educational PsychologyUniversity of Houston
  • Allison Dempsey
    • Department of Educational PsychologyUniversity of Houston
  • Ashlie Llorens
    • Department of Educational PsychologyUniversity of Houston
Article

DOI: 10.1007/s10643-011-0446-0

Cite this article as:
Keller-Margulis, M., Dempsey, A. & Llorens, A. Early Childhood Educ J (2011) 39: 95. doi:10.1007/s10643-011-0446-0

Abstract

The developmental outcomes for children born preterm have been examined by many, with results unequivocally indicating that children born preterm tend to have poorer cognitive outcomes and more developmental difficulties. Less attention has been paid to academic outcomes. The purpose of this paper is to review the academic skills assessment of children born preterm, examine the methodologies used to ascertain skill deficits, and identify essential directions for future research. Overwhelmingly the results of studies of academic skills indicate that children born preterm function lower than their full term peers. The methodological flaws with existing studies that impede broad conclusions about specific skill deficits will be discussed. It is critical that future research examine the academic skill deficits contributing to disability status so that effective early intervention strategies may be developed and implemented.

Keywords

Children born pretermAcademic skillsAcademic outcomesSpecial education

Introduction

Estimates suggest that preterm births are an important health issue worldwide. According to 2005 estimates, Africa experienced the highest rate of preterm births at 11.9%, followed by North America 10.6%, Australia/New Zealand (6.4%) and Europe had the lowest rate (6.2%; Beck et al. 2010). Collectively, Africa and Asia account for 85% of preterm births worldwide. This statistic should be interpreted with caution as researchers partially attribute the high prevalence to greater number of births that occur in these regions of the world. The increase in premature births over the past 20–30 years in developed countries may be partially accounted for by variables such as, increased use of reproductive technologies, changes in labor management (i.e., elective inductions and cesarean sections), increasing rate of multiple births, and higher rate of mothers giving birth at an older age (Beck et al. 2010). Currently, the effect of preterm births on education and social service systems may not be seen in developing countries due to the variation in survival rate among developed and developing countries. In developing countries, infants born at less than 32 weeks gestation or 2,000 g have a rare chance of survival (Beck et al. 2010), compared to approximately half of infants born at 22–25 weeks in the US surviving (Tucker and McGuire 2004).

Children born preterm and at low birth weight (LBW) frequently experience a variety of developmental deficits associated with their premature status, some of which are evident at birth and others that become apparent as the children mature (Aylward, 2002a, b, 2009; Salt and Redshaw 2006). Lower birth weight and gestational age at birth significantly increases the likelihood of developmental and medical difficulties (Saigal and Doyle 2008; Taylor, Klein, and Hack 2000).

Learning difficulties and deficits in academic achievement in students born prematurely have been noted in samples of children from various regions of the world such as, New Zealand (Pritchard et al. 2009), Canada (Downie et al. 2005; Grunau et al. 2002; Saigal et al. 2003), Australia (Anderson et al. 2003; Bowen et al. 2002; Wocadlo and Rieger 2006), Germany (Saigal et al. 2003), Holland (Saigal et al. 2003), and the United Kingdom and Ireland (Johnson et al. 2009).

Children born preterm are an increasingly relevant issue for the educational system, as the number of children born preterm in the United States has increased to 12.8% of births–up 36% from the 1980s (Aylward 2002a, b; Martin et al. 2008; Saigal and Doyle 2008). Although survival rates for those children born the most preterm have improved, there has been no reduction in the rate of disability experienced by these children (Saigal and Doyle 2008; Tyson and Saigal 2005). This discussion will include an overview of the recent research on academic skills outcomes for children born preterm, a review and critique of the methodology employed, and the identification of important directions for future work in this area.

Developmental Outcomes and Disability Rates

The difficulties that develop as premature infants become older can include cognitive, visual-motor skill deficits, language delays, behavioral difficulties, and eventually school and learning problems (Arpino et al. 2010; Chyi et al. 2008; Roberts et al. 2008). Collectively, the results of outcome studies indicate that children born preterm often experience deficits in cognitive functioning with increased prematurity resulting in an increased likelihood of functioning deficits (Baron and Rey-Casserly 2010).

Children who are born preterm and at the lowest birth weights can experience low-incidence, high severity conditions such as mental retardation and autism Taylor, Klein, and Hack (2000) and 20–25% are diagnosed with cerebral palsy (Aylward 2002a). These conditions are often identified early in life, even at birth, and children are more likely to receive early educational intervention to address the impact of these conditions.

Although autism may be considered a low-incidence, high severity condition, Attention Deficit/Hyperactivity Disorder (AD/HD) (Johnson et al. 2010) and executive functioning problems (Anderson and Doyle 2004) are low-severity, high incidence conditions that also impact the population of children born preterm. Low-severity but high incidence conditions are not as easy to detect early in life but often impact functioning once children enter the school setting (Marlow et al. 2005; Salt and Redshaw 2006). Although less severe, these conditions have been found to affect 50–70% of children born at low birth weight (Goyen et al. 1998; Msall et al. 1991; Taylor, Klein, and Hack 2000).

Learning disabilities (LDs) are also considered high-incidence, low-severity conditions and are often experienced by children born preterm. Johnson and Breslau (2000) examined the rates of learning disabilities (LD) using both a discrepancy and a low achievement definition of LD. Their results indicated that boys born at LBW are more likely to have a reading or math disability than children born at term. Just over fifteen percent (15.2%) of LBW boys were identified with a reading disability and 7.0% of LBW girls. In contrast, children born at normal weights had rates of 6.4% for boys and 7.4% for girls. In addition, 10.9% of LBW boys and 4.1% of LBW girls were identified with math disability compared to 2.3% and 5.9% of same sex controls, respectively. This study, like many studies of cognitive and developmental outcomes for children born preterm, did not report the mean achievement scores across academic areas and did not include an assessment of functional skills in the school setting.

Service Utilization

The range and frequency of deficits observed in children born preterm qualify many for early intervention services. As many as 30% of late preterm infants (34–36 weeks gestation) and 70% of very preterm infants (<32 weeks gestation) may meet eligibility for early intervention services (Kalia et al. 2009), though the number of families accessing these services is much lower (Roberts et al. 2008). In one study of service utilization, rates of special education services ranged from 6% among late preterm infants to 21% for very preterm infants (Kalia et al. 2009). In another sample of preterm infants, approximately 50% of children with moderate to severe disabilities and 72% of children with mild disabilities were not receiving any early intervention services by 2 years of age (Roberts et al. 2008). The need for service increases as children enter elementary school, with nearly 60–70% of children born preterm demonstrating a need for special education services (Aylward 2002a, b; Wocadlo and Rieger, 2006).

Identification of Academic Difficulties

In addition to the specific disability conditions mentioned above, children born preterm or at LBW are known to experience increased difficulty with academic skills and overall school functioning (Aarnoudse-Moens et al. 2009). In fact, academic achievement problems have been found to persist when controlling for cognitive standard scores above 70 and above 85, with those children with the lowest birth weight demonstrating the lowest scores on measures of academic achievement regardless of cognitive functioning (Klebanov et al. 1994). Even late preterm children, those born at 34 weeks gestation or more, can evidence academic achievement difficulties in kindergarten including a higher likelihood (36%) of being identified as having difficulty in school (Morse et al. 2009).

Review of Recent Academic Skills Studies

Children are typically identified as having learning difficulties or disabilities while in elementary school. This review focuses on studies conducted on children of school age because of the implications it has for early intervention and preschool programming for this at-risk population. The most recent outcomes of studies examining school-age outcomes for children born preterm will be reviewed and the methodology used including direct assessment measures and other forms of data collection at school-age follow-up will be discussed.

Study Inclusion and Exclusion Criteria

Studies were included in the review if the participants were school-age children, between 6 and 11 years of age and a direct assessment of academic skills was conducted in English using a standardized achievement measure. Only studies published from 2000 through June, 2010 were examined. Studies that did not include a direct assessment of academic skills or included only teacher or parent report of academic achievement were excluded. Reported use of special services in the school setting was not a requirement for inclusion in the review but was noted if included in study outcomes. A literature review was conducted using scholarly online databases including PubMed, Medline, PsychINFO, ERIC, and CINAHL Plus. The search terms that were used to find articles of academic outcomes for preterm infants included preterm infants, children born preterm, academic assessment, academic outcomes, school outcomes, achievement. When articles that met the criteria for inclusion were located, reference lists were used to further identify articles for inclusion in the review that met criteria.

Studies were excluded if they did not use a standardized measure of either reading or math and if they did not report outcomes for reading and math separately. In addition, studies were excluded if they used a sample that was tested in a language other than English. The resulting sample consisted of exactly 20 studies from January 2000 through July 2010 (demarcated with an asterisk in the reference list) and included studies primarily from the United States but also from Canada, the United Kingdom, Australia, and New Zealand.

Themes in the Literature

Academic Skills Outcomes for Children Born Preterm

Over half of the 20 studies indicated that children born preterm or at low birth weight demonstrate academic skills in reading, math, spelling, and writing that are significantly below that of their full term peers. Of the 20 studies that were reviewed, 14 reported mean performance on the standardized measure, 12 found significant group differences, while two studies did not perform any statistical comparisons of performance across groups. Some studies yielded outcomes indicating that children born preterm are more likely to experience difficulties in the academic area of math, even when cognitive functioning is considered normal (Taylor et al. 2009).

Measures and Methods Used to Assess Academic Skills

Methods for gathering information regarding academic functioning in follow-up studies included multiple traditional, standardized measures of academic outcomes and parent and teacher report of skill functioning and service utilization (Sullivan and McGrath 2003). The achievement measures used in the studies reviewed are shown in Table 1. Nearly half (n = 9) of the studies used a combination of tests to assess academic skills, most often combining portions of the Woodcock-Johnson Tests of Achievement (WJTA; Woodcock et al. 2001) with the Reading Comprehension Subtest of the Wechsler Individual Achievement Test (WIAT; Wechsler 2001) (e.g., Litt et al. 2005). When a single test of academic skills was used, it was most often the Wide Range Achievement Test (WRAT; Jastak and Wilkinson 1984).
Table 1

Measures used to assess academic skills

Test

Total

Academic area

  

Math

Reading

WRAT

6

6

6

WIAT

6

3

6

WJTA

10

10

10

PIAT

1

1

1

PPVT-R

1

TERA

1

1

TEMA-2

3

3

TOWL-2

1

WRMT

2

2

GORT-R

1

1

WRAT Wide Range Achievement Test, WIAT Weschler Individual Achievement Test, WJTA Woodcock Johnson Tests of Achievement, PIAT Peabody Individual Achievement Test, PPVT Peabody Picture Vocabulary Test, TERA Test of Early Reading Ability, TEMA-2 Test of Early Mathematics Ability, TOWL-2 Test of Written Language-2, WRMT Woodcock Reading Mastery Test, GORT-R Gray Oral Reading Test Revised

None of the studies reviewed used enough of the subtests of the WJ-III to obtain the Broad Reading score and only one study (Taylor, Klein, Minich, et al. 2000) used enough subtests to obtain the Broad Math composite score. A more common approach was to use two subtests in a skill area to measure functioning in an academic area (Breslau et al. 2001; Downie et al. 2005; Short et al. 2003; Taylor et al. 2006). Several studies used a diagnostic measure of academic skills instead of a broad measure of academic skills that includes multiple domains. A total of four studies used this approach to academic skills assessment. Lastly, some studies used regional or local assessments to measure academic outcomes in a specific area (Bowen et al. 2002; Wocadlo and Rieger 2006).

Table 2 shows the methods used to assess academic skills for all reviewed studies. Approximately half of the studies also used parent or teacher report and sometimes both. Eleven studies procured student data related to school performance, either directly from school records or via parent and/or teacher report of educational history, in order to support conclusions about preterm children’s skill deficits compared to those of fullterm peers.
Table 2

Summary of methods used to assess academic skills in all reviewed studies

 

Number of studies

Type of data

 

 Direct assessment only

8

 Included parent and/or teacher report

12

 Included data from school records

11

  Directly from school record

5

  Parent/teacher report of educational history

6

 Only used one measure to assess skill area

11

  Reading

8

  Math

8

  Spellinga

11

  Writinga

1

  Languagea

1

aNot assessed by all studies

School Functioning and Disability Status

Statements about students’ educational placement, services received related to disability status, and academic performance history were included in 13 of the 20 studies reviewed. Approximately 40% of studies that measured reading and math used a single measure to assess the skill. All studies that assessed spelling (n = 11), language (n = 1) and writing (n = 1) used a single measure to assess the skill. Two studies asserted conclusions about disability status solely based on the data they collected, which did not include information from school records and only used a single measure to classify LD in math (Grunau et al. 2002; Pritchard et al. 2009) and reading (Pritchard et al. 2009). While some studies (e.g., Bowen et al. 2002; Gross et al. 2001; Johnson et al. 2009; Saigal et al. 2003; Short et al. 2003; Wocadlo and Rieger 2006) utilized teacher or parent report of services used in the school setting, only five studies accessed school records to objectively determine rates of LD identification and service utilization for participants.

Methodology Critique

A review of the most recent studies of academic skills functioning of children born preterm clearly indicates that this group is at risk for experiencing academic difficulties early in elementary school. Although it is possible to make preliminary conclusions about the academic skills of children born preterm, many methodological differences make a complete synthesis of the outcomes for this population difficult.

Measures and Methods Used

The use of single subtest to measure academic skills is insufficient for conclusions about deficits or relative weaknesses. The tools and the data collection methods used across these studies often were abbreviated in nature and not supportive of the conclusions derived by the authors. Many studies used one subtest from an achievement test or a brief achievement measure to assess academic skills (Anderson et al. 2003; Downie et al. 2005; Grunau et al. 2002; Hack et al. 2005; Kan et al. 2008; McGrath and Sullivan 2002; Saigal et al. 2003; Sullivan and McGrath 2003) making conclusions based on a very small and limited sample of student academic behavior. The reading subtest of the WRAT requires students to simply read aloud a list of words presented in isolation. The arithmetic task involves having students solve simple computation while the spelling subtest requires the student to spell a list of dictated words. Arguably none of these single measures of academic skill can be generalized to make conclusions about functioning in the academic areas of reading, math, or spelling.

Conclusions about overall academic skills based on a measure like the WRAT are tenuous at best. Where more lengthy and appropriate measures are used in studies of academic outcomes, they represent a small segment of the literature base and not enough on which to establish broad conclusions about academic skills.

Conclusions About School Functioning and Disability Status

The use of brief assessment approaches does not provide the quality or quantity of data required to substantiate conclusions about either specific skill deficits or the presence of disability conditions. Out of all studies that examined whether participants were demonstrating a disability, it is promising that only two relied on singular, direct assessment of skills (Grunau et al. 2002; Sullivan and McGrath 2003). When disability decisions were made in some cases, however, arbitrary designations such as cut scores without attention to functional skill deficits, were often used.

Patterns of academic skill difficulty in the population of children born preterm differ from those of typically developing peers. In the general population, children identified as having academic difficulty most often have trouble with reading and writing (Fletcher et al. 2007). Children born preterm have also been found to demonstrate deficits in reading and writing (Frye et al. 2009). An emerging theme in the research on academic outcomes among children born preterm, however, suggests that this group experiences more deficits in the area of math achievement than other academic skill areas (Litt et al. 2005; McGrath and Sullivan 2002; Pritchard et al. 2009; Short et al. 2003; Sullivan and McGrath 2003; Taylor et al. 2002). Deficits in math skills appear to persist even when controlling for neurodevelopmental differences and socioeconomic status among children born preterm and their peers (Anderson et al. 2003; Pritchard et al. 2009; Taylor et al. 2009). The mean effect size for the difference in math achievement scores among children born at low birth weight or premature and controls is .73 (Taylor et al. 2006).

General Conclusions and Practical Implications

Despite the methodological concerns that make synthesis of the literature on academic skills for children born preterm challenging, the current consensus is that these children are at increased risk for developing academic skill problems once they reach school age. Children born the most preterm are more likely to be receiving special education services by kindergarten and the cost to educate these children with special needs is significantly higher than the education of those children born less preterm or full term (Roth et al. 2004). Nearly all existing work with this population points to a need for early screening and intervention to identify the academic needs of this population before the children reach school age and the skill difficulties that they are experiencing often manifest (Johnson et al. 2008).

Although data on academic skills are collected in a substantial number of studies, follow-up studies on the developmental outcomes for children born preterm are typically initiated by the medical community in research follow-up settings. Most often, the focus of the follow-up assessment is gathering data for research examining brain functioning as well as for disability screening and not for clinical diagnosis or instructional programming recommendations. Consequently, examination of academic achievement is approached broadly, with the goal of identifying general deficits and the degree to which children born preterm meet disability criteria. Knowledge of the specific academic skills that underlie broad achievement deficits, poor grades, and the need for support services during early childhood education is lacking. This information would directly inform what skills require focus in early childhood education and early intervention at school-age.

The definitions of LD and approaches to LD determination used in the academic outcomes literature for children born preterm are not consistent with common practices in early childhood education or school settings, which typically include an examination of the child in context as well as multi-method and multi-informant strategies. In the existing literature, LDs often were defined as academic achievement performance that is one standard deviation below the mean of the control group while in other cases, a low achievement model was used. Newer approaches to LD determination have emerged with the reauthorization of the Individuals with Disabilities Education Act (IDEIA 2004) that allow for the consideration of a child’s response to intensive, research-based academic interventions, and consideration of the learning environment, when determining LD status. Ultimately, LD remains a thorny construct to define, that is only complicated when applied to a population of children who routinely experience a heterogeneous collection of cognitive and neurological complications like those experienced by children born preterm. Although it is clear that children born preterm are likely to experience LDs, the specific pattern of academic skill deficits that lead to this conclusion is not yet known.

Although there are still some questions regarding the academic skills patterns of children born preterm, it is well established that survival rates of premature births are increasing and the number of children entering the school system with special needs is also increasing. The existing evidence suggests a need for the systematic delivery of early childhood intervention services to decrease the severity of skill deficits at school entry and improve the likelihood for future academic success. Developing a more detailed understanding of the school-age academic skills for children born preterm has significant implications for early intervention and preschool programming.

Future Directions: a Call for Additional Research

Evidence from existing literature indicates that the academic skills of children born preterm are significantly lower than the skills of their peers (Downie et al. 2005; Short et al. 2003; Taylor et al. 2009; Taylor, Klein, Minich et al. 2000). Studies published in the last 10 years have methodological flaws that make conclusions about specific academic skill patterns difficult to derive. Future research must address two main issues in the literature. First, the methodological issues associated with the descriptive literature on academic skills outcomes for preterm infants must be addressed.

The use of shortened or piecemeal versions of cognitive and neuropsychological tests must be replaced with comprehensive assessment batteries that focus on specific skills accompanied by multi-method, multi-informant assessment. Skills assessments must be supplemented with parent and teacher report, educational file reviews, and direct observations of school performance (Sattler 2008). A complete assessment battery that includes screening, broad achievement, and diagnostic measures of academic skills would substantiate more confident conclusions about academic skills outcomes for children born preterm.

In addition to addressing methodological issues, future research on academic skills must be designed such that the specific pattern of academic skill deficits experienced by children born preterm can be identified. Although the assessment approach described above may be unrealistic in research or hospital-based clinic follow-up settings, researchers must begin to use a more comprehensive approach to academic outcomes assessment. Follow-up assessment must be designed to inform instructional planning that can improve academic outcomes for children born preterm before the skill problems become disabilities (Xu and Filler 2005). It is not enough to merely suggest that learning difficulties may be experienced by children born preterm. Future studies should focus on the identification of the specific skills that are lacking for children born preterm at the time of school entry into kindergarten. Identification of these critical skill deficits would lead to subsequent studies of early educational interventions to improve skill deficits and overall achievement. The literature surrounding the academic skill deficits of children born preterm will remain stagnant and disconnected from practice unless a shift is made to understanding specific skill deficits that lead to the provision of targeted and effective early intervention programming.

Copyright information

© Springer Science+Business Media, LLC 2011