AJCN Cancer Health Disparities Conference
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Related articles in AJCN
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rudnicka, A. R
Right arrow Articles by Strachan, D. P
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rudnicka, A. R
Right arrow Articles by Strachan, D. P
Agricola
Right arrow Articles by Rudnicka, A. R
Right arrow Articles by Strachan, D. P
American Journal of Clinical Nutrition, Vol. 87, No. 5, 1392-1399, May 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Effect of breastfeeding and sociodemographic factors on visual outcome in childhood and adolescence1,2,3

Alicja R Rudnicka, Christopher G Owen, Marcus Richards, Michael EJ Wadsworth and David P Strachan

1 From the Division of Community Health Sciences, St George's, University of London, London, United Kingdom (ARR, CGO, and DPS), and the Medical Research Council National Survey of Health and Development, Department of Epidemiology and Public Health, Royal Free Hospital, University College Medical School, London, United Kingdom (MR and MEJW)

See corresponding editorial on page 1120.

2 Supported by grant no. G0000934 from the Medical Research Council (ARR) and grant no. PG/04/072 from the British Heart Foundation (CGO).

3 Address reprint requests and correspondence to AR Rudnicka, Division of Community Health Sciences, St George's, University of London, Cranmer Terrace, London SW17 ORE, United Kingdom. E-mail: arudnick{at}sgul.ac.uk.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: It has been suggested that early life factors, including breastfeeding and birth weight, program childhood myopia.

Objective: We examined the relation of reduced unaided vision (indicative of myopia) in childhood and adolescence with infant feeding, parental education, maternal age at birth, birth weight, sex, birth order, and socioeconomic status.

Design: Three British cohorts recruited infants born in 1946 (n = 5362), 1958 (n = 18 558), and 1970 (n = 16 567). Adjusted odds ratios (ORs) for unaided vision of 6/12 or worse at ages 10–11 and 15–16 y from each cohort were pooled by using fixed-effects meta-analyses.

Results: The prevalence of reduced vision ranged from 4.4% to 6.5% at 10–11 y and from 9.4% to 11.4% at 16 y, with marginally higher levels in later cohorts. Breastfeeding declined across successive cohorts (65%, 43%, and 22% in those breastfed for >1 mo, respectively). Pooled ORs showed no associations between infant feeding and vision after adjustment at either age. Parental education (OR: 1.48, high versus low education; 95% CI: 1.23, 1.79), maternal age (OR: 1.10, per 5-y increase; 95% CI: 1.04, 1.17), birth weight (OR: 0.85, per 1-kg rise; 95% CI: 0.76, 0.95), number of older siblings (OR: 0.89, per older sibling; 95% CI: 0.83, 0.94), and sex (OR: 1.10, girls versus boys; 95% CI: 0.98, 1.23) were related to adverse visual outcome in childhood. Stronger associations were observed in adolescence, except that the association with birth weight was null.

Conclusions: Infant feeding does not appear to influence visual development. Consistent associations of reduced vision with parental education, sex, maternal age, and birth order suggest that other environmental factors are important for visual development and myopia in early life.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Myopia is a leading cause of correctable visual impairment in the developed world and a leading cause of blindness in developing countries (1). It has been estimated that 30.4 million Americans aged ≥40 y are myopic (2). Reduced vision in childhood is predominantly due to myopia (3-6) with a shift toward higher levels of myopia with increasing age (7). Marked geographic variations in the prevalence of myopia have been reported in both children and adults (8, 9). These variations coupled with the recent rapid increases in the prevalence of myopia, especially among children in Asian (10) and other industrialized societies (11), suggest that environmental factors are important determinants of visual outcome.

Vision is immature at birth and is programmed in response to visual stimuli in early life. There has been considerable interest in whether early nutrition might also influence visual development, particularly whether initial infant feeding may play a role. It has been suggested that those initially breastfed have better vision and are less likely to be myopic in later life than are those who are formula fed (12-15). The presence of long-chain polyunsaturated fatty acids (LCPUFAs) in breast milk, which are needed for rapid growth and neural development in early life (16, 17), has been proposed as a potential explanation for these findings, because formula milk does not universally contain LCPUFAs (15). Although this has been supported by experimental evidence showing the benefit on visual outcome of formula milk supplemented with LCPUFAs relative to nonsupplemented formula (13, 18), other experimental and observational evidence has been less supportive (19-23). Inconsistencies in results may reflect variation in statistical power and/or differences in the extent of adjustment for potentially confounding factors, such as socioeconomic status (14, 15, 24), maternal factors at delivery (14, 15, 25, 26), parental smoking and education (14, 15, 27), birth order (24, 27, 28), and differences in size at birth between feeding groups (14, 15, 20).

Some evidence links these potential confounding factors to visual outcomes in early life. Increasing maternal age at birth seems to be related to poorer visual outcomes in infancy (14) and is strongly associated with the degree of myopia in early childhood (29). In agreement with this latter finding, recent work has suggested a positive association between birth weight and eye size in young children (30, 31). A retrospective cohort study did not find birth weight to be related to visual function in the elderly (32). Higher levels of parental education (15, 27), higher socioeconomic status (15, 24, 33), and smaller family size (or birth order) (24, 27) have been reported to be positively related to the prevalence of myopia. Higher levels of myopia among girls than among boys observed among some studies (33-35) may also suggest environmental influences on refractive outcome, although not all studies have shown similar sex differences (15, 24). A comparable finding was observed in studies that examined the association between infant feeding and cognitive development, a neurological outcome related to vision, where failure to adjust systematically for similar confounders (36, 37) may explain the apparent inconsistencies in the findings (28, 37-43).

To examine the association of early nutrition on vision further, we examined the association between pattern of infant feeding and visual outcome in childhood and adolescence in 3 large British birth cohorts accounting for the potential confounding factors identified above. The cross-cohort comparison allows for the association between infant feeding and visual outcome to be gauged over a period when there has been stark changes in the rates of breastfeeding and social patterning of feeding in early life, together with considerable improvements in childhood nutrition (44, 45). In addition, parental educational attainment, especially maternal education, would also have changed over this time frame, and this would influence both diet and educational attainment in the offspring. The available data also allow the strength and consistency of the association between vision and other sociodemographic factors across cohorts to be gauged.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Participants were drawn from 3 British birth cohorts with information on infant feeding practices, vision outcome in childhood and adolescence, and information on potential confounding factors, including socioeconomic status in early childhood or infancy, birth weight, maternal age at delivery, parental education, and the number of younger and older siblings (birth order). Ethical approval for the examination of the cohort members was obtained from the relevant ethical committees at the time of each survey.

British birth cohort studies
From all those born in England, Scotland, and Wales during 1 wk in March 1946, a sample was selected for follow-up of single infants born to married women. All such births to wives of nonmanual and agricultural workers were selected together with 1 in 4 of all such infants born to wives of manual workers to form the Medical Research Council National Survey of Health and Development (NSHD; n = 5362), also known as the British 1946 birth cohort (46). The next cohort recruited all persons born in England, Scotland, and Wales during 1 wk in March 1958 for a Perinatal Mortality Survey (n = 17 638); this then became known as the National Child Development Study (NCDS), or the British 1958 birth cohort (47). The third cohort, the 1970 British Cohort Study (1970 BCS), recruited 98% of persons (n = 16567) born in Great Britain during 1 wk in April 1970 (48). Participants from all cohorts were followed up periodically from birth through childhood and into adulthood. The current analysis is based on data collected from birth up to age 16 y.

Definition of visual outcome
Vision was measured when the children were aged 11 and 15 y in the 1946 cohort, 11 and 16 y in the 1958 cohort, and 10 and 16 y in the 1970 cohort. In all 3 cohorts, unaided distance vision (ie, without optical correction) was measured with the use of conventional Snellen charts of block capitals under standardized conditions; each eye was tested separately. In the 1958 and 1970 cohorts, near vision was also tested at both ages using the Sheridan and Gardiner reduced Snellen near acuity cards at {approx}25 cm (10 inches) (49-51). Two dichotomous outcomes of presumed myopia were derived from these vision outcomes. The first used a cutoff of unaided distance vision in the better eye of 6/12 Snellen acuity or worse. The second, used for the 1958 and 1970 cohorts only, was based on an unaided distance vision of 6/12 Snellen acuity or worse and unaided near vision of 6/6 or 6/9 at 25 cm (in the better eye); those with both poor distance and near vision were excluded. This latter definition of presumed myopia was used previously in the 1958 NCDS (24) and is considered to be analogous with –1 to –5 diopters of myopia. Results from our analyses using this definition were not materially different from the visual outcome based on an unaided distance alone and, therefore, are not presented. This is not unexpected because it was shown in the 1958 cohort that those with reduced distance vision at 16 y (91%) had myopia in adulthood (6). Children in the 1946 cohort with distance vision 6/12 or worse at 16 y received an ophthalmic examination; the results showed that ≥61% were myopic, but 30% with reduced vision were not assessed, so the proportion of myopia in this group is an underestimate and is likely to be higher.

Definition of breastfeeding
In the 1946 NSHD, mothers were interviewed when the cohort member was 2 y old; infant feeding status and duration of breastfeeding was ascertained for 4784 (89% of the cohort at the outset) participants. In the other 2 cohorts, infant feeding was recalled by parental interview when the child was 5 (1970 BCS) or 7 (NCDS) y of age, and the number with data on breastfeeding status was 12 981 (74%) and 14498 (88%), respectively. Infant feeding status was classified as not breastfed at all, breastfed (partially or wholly) for <1 mo, and breastfed for ≥1 mo (in NSHD 1946, duration of breastfeeding in months up to 10 mo was available and in the 1970 BCS it was also possible to identify those breastfed for ≥3 mo). Those with unknown infant feeding status were excluded from the analyses throughout.

Definition of other risk factors
1946 NSHD
In the 1946 cohort, socioeconomic status was defined in the study sampling design according to the Registrar General system and was dichotomized to manual and nonmanual groups at birth; this dichotomization was preserved for the analyses. Birth weight, sex, maternal age, parental education, and birth order were recorded during the birth survey. Parental education at birth was classified as primary schooling only, primary or secondary education with formal qualifications after secondary school (but no diploma), and diplomas and professional degrees (highest group); the highest level achieved by either parent was used. From birth order the number of older siblings was defined, and the number of children at age 11 y was obtained by parental questionnaire, thereby allowing the number of younger siblings to be determined. In the analyses, the number of younger and older siblings was used to take account of birth order and family size.

1958 NCDS
Maternal age, birth weight, and sex were obtained from the 1958 Perinatal Mortality Survey. Socioeconomic status in childhood was based on Registrar General classification of the father's occupation in 1958 or at age 7 y if data were unavailable at birth. The age at which the father left full-time education was ascertained when the cohort member was 7 and 11 y of age. Data at age 7 y were preferred; however, data from age 11 y were used if data at age 7 y were missing. The age at which the mother left full-time education was also ascertained when the cohort member was 11 y. Parental education was classified into 3 groups: left school at the minimum age of 15 y (or younger), remained in full-time education to the age of 18 y, and remained in full-time education beyond the age of 18 y; the highest level achieved by either parent was used in the analyses. At age 11 y, the number of younger siblings and the child's position in relation to all siblings was obtained by parental questionnaire.

1970 BCS
Birth weight and sex were recorded in the original birth survey. Age of the mother at delivery was ascertained by parental interview when the child was 5 y of age. Socioeconomic status in childhood was defined by paternal occupation (using the Registrar General classification) if available; if not available, it was by maternal occupation when the child was 10 y old. Family size, including the number of older and younger siblings, and parental education were determined at the same age. Parental education was classified into 3 groups using qualifications indicative of leaving full-time education at ≤16 y, 16–18 y, or >18 y of age. Again, the highest level achieved by either parent was used in the analyses.

Statistical analysis
Statistical analyses were performed by using STATA version 9 (STATA Corporation, College Station, TX). The prevalences of unaided distance vision 6/12 or worse at ages 10–11 and 15–16 y were compared by a chi-square test in those breastfed for >1 mo and those never breastfed. Chi-square tests were also used to compare the prevalences of reduced vision by social class (manual versus nonmanual), sex, level of parental education, maternal age (<25 y, 25 to <35 y, and ≥35 y), quintiles of birth weight, and number of younger (0, 1–2, and ≥3) and older (0, 1–2, and ≥3) siblings. On the basis of the number of individuals in each cohort with vision outcome data, we used logistic regression of grouped data to examine trends in the log odds of poor visual outcome over the 3 cohorts. Trends were examined for child and adolescent age groups separately. Within each cohort, multiple variable logistic regression was used to obtain mutually adjusted odds ratios for the vision outcomes for infant feeding group, socioeconomic position in childhood, sex, and parental education as categorical variables and maternal age at delivery, birth weight, and number of younger and older siblings as continuous variables. Adjusted odds ratios from each cohort were pooled by using a fixed-effects (inverse variance) meta-analysis throughout. Heterogeneity across cohorts was examined by using chi-square tests (52), and tests for trend of the odds ratios across the 3 cohorts were examined by using meta-regression analysis (metareg command in STATA). All analyses were restricted to those with complete data on the relevant vision outcome and all risk factors considered above.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The overall prevalence of reduced vision at 10–11 and 15–16 y in the 3 cohorts is summarized by all risk factors considered in the analyses in Table 1Go. Overall, the prevalence of distance vision of 6/12 or worse at 10–11 and 15–16 y was 6.0% and 9.4% in the 1946 NSHD, 6.5% and 10% in the 1958 NCDS, and 4.4% and 11.4% in the 1970 BCS. The lower prevalence of poor vision in childhood for the 1970 BCS may in part be explained by the fact that vision was assessed 1 y earlier in this cohort. It is known that myopia increases with age, and rapid changes in prevalence occur at these ages (11). Logistic regression of the log odds of poor vision at 15–16 y showed some evidence of an increase from 1948 to 1970 (P = 0.004), but vision was assessed 1 y earlier in the 1946 NSHD. Hence, this trend may have been partly or wholly confounded by age.


View this table:
[in this window]
[in a new window]

  		TABLE 1 Prevalence of unaided distance vision of 6/12 or worse by all risk factors included in the analyses in the participants with complete data on vision outcome

 
The prevalence of breastfeeding fell markedly across the 3 successive cohorts. Of those with complete data for these analyses, the proportions breastfed for >1 mo were 65% in the 1946 NSHD, 43% in the 1958 NCDS, and 22% in the 1970 BCS. The prevalence of breastfeeding for >1 mo was more common in nonmanual social classes and with increasing parental education, especially among the latter 2 cohorts (See Table S1 under "Supplemental data" in the online issue.). There is some evidence that infants with lower birth weights and older siblings were less likely to have been breastfed. No clear association of infant feeding to maternal age was observed (See Table S1 under "Supplemental data" in the online issue.). The patterns observed among those with complete data on all risk factors did not materially change when those without complete data were included.

Infant feeding and visual outcome in later life
In the 1946 NSHD and 1958 NCDS, the prevalence of distance vision of 6/12 or worse at 10–11 and 15–16 y of age was similar in those breastfed and formula fed. In the 1970 BCS, the prevalence of reduced vision was marginally lower in those who were formula fed, but these crude differences were not statistically significant (Table 1Go). None of the corresponding odds ratios showed any evidence of an association of breastfeeding with reduced vision after adjustment for potential confounders, including socioeconomic status in childhood, sex, parental education, maternal age, birth weight, and number of younger and older siblings (Table 2Go). The pooled fixed-effects estimate of reduced vision comparing those breastfed for >1 mo with those never breastfed was 1.03 (95% CI: 0.90, 1.18) at 10–11 y and 1.02 (95% CI: 0.89, 1.16) at 15–16 y.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Adjusted odds ratios (and 95% CIs) for the association of each risk factor listed with unaided distance vision of 6/12 or worse at 10–11 y of age1

 
The duration of breastfeeding defined in each cohort was ≥1 mo, and this potentially diluted associations with longer durations of breastfeeding. It was possible to examine the association of breastfeeding for ≥3 mo in the 1970 BCS, and, in the 1946 NSHD, the duration of exclusive breastfeeding in months up to 12 mo was available. There was still no association between infant feeding on visual outcome, either in childhood or adolescence, when this prolonged duration and exclusivity of breastfeeding (1946 birth cohort only) was considered (data not presented).

Other factors associated with visual outcome in childhood and adolescence
In general, the adjusted odds ratios were consistent across the cohorts, and tests for heterogeneity were not statistically significant for any risk factor, except for mid versus low level of parental education in adolescence (Table 2Go and Table 3Go). Females were more likely to have reduced vision than were males (Tables 2Go and 3Go), especially in adolescence, with a pooled odds ratio of 1.23 (95% CI: 1.11, 1.37). Individuals from a manual social class were less likely to have poor vision at both ages than were those from a nonmanual social class. Although socioeconomic status did not show a strong association within each cohort, the pooled estimate at 15–16 y suggests a 12% reduction in the odds of reduced vision in the manual versus nonmanual social classes (95% CI: 0.78, 0.99). In all 3 cohorts, unaided distance vision of 6/12 or worse was more likely in those with higher levels of parental education than in those with lower levels; this was highly statistically significant at both ages (Tables 2Go and 3Go) with stronger associations in adolescence. The pooled odds ratio for the comparison of the highest with the lowest level of parental education in adolescence showed a 69% increase in the odds of poor unaided distance vision (95% CI: from 43% to 100% increase). Parental education was related to social class in the cohorts, and exclusion of social class from the model resulted in stronger associations for parental education and vice versa (data not shown). A higher maternal age at birth was related to reduced vision both in childhood (odds ratio: 1.10 per 5-y increase in maternal age; 95% CI: 1.04, 1.17) and in adolescence (odds ratio: 1.15; 95% CI: 1.09, 1.22); this association possibly strengthened across successive cohorts (P value for trend in odds ratios: 0.06). In childhood, heavier birth weight was marginally associated with a reduced risk of poor visual outcome, especially in the 1970 BCS. However, no association was observed by adolescence. The number of younger siblings showed no consistent association with visual outcome at either age. However, an increase in the numbers of older siblings was consistently associated with a decreased risk of reduced vision, both at 10–11 and at 15–16 y of age and across cohorts (Tables 2Go and 3Go). In childhood, the risk of poor vision decreased by 11% (95% CI: 6% to 17% reduction) and by 16% (95% CI: 11% to 20% reduction) in adolescence per number of older siblings.


View this table:
[in this window]
[in a new window]

  		TABLE 3 Adjusted odds ratios (and 95% CIs) for the association of each risk factor listed with unaided distance vision of 6/12 or worse at 15–16 y of age1

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Analyses of data from these 3 birth cohorts showed that there were no differences in visual outcome in childhood or adolescence among those initially breastfed for >1 mo compared with those formula fed. Rates of breastfeeding fell across successive cohorts considerably, but there was no evidence of a subsequent increase in adverse visual outcome in childhood, although, there was a small increase in adverse visual outcome in adolescence from 1946 to 1970. However, this latter observation does not take into consideration important confounding factors, especially the age at which vision was assessed. Notably, within each individual cohort there was no relation between mode of infant feeding and visual outcome. The strongest determinants of unaided distance vision of 6/12 or worse at both ages were educational attainment of the parents, maternal age, and number of older siblings; birth weight appeared to be important at age 10–11 y, with no influence at 15–16 y. A higher prevalence of adverse visual outcome in girls than in boys emerged in childhood and was more apparent by adolescence.

Infant feeding and adverse visual outcomes
Previous studies have suggested better visual outcome in infancy (12, 13) and early childhood (14) among those who were breastfed than among those who were formula fed. A cross-sectional study in Singaporean children of an age (10–12 y) similar to that of the participants in the present study suggested that this association may persist into later childhood, when the risk of myopia was found to be 42% lower in those initially breastfed than in those who were formula fed (adjusted odds ratio: 0.58; 95% CI: 0.39, 0.84) (15). However, the refractive status of this South Asian population was markedly different from that of UK children (ie, Singaporean children had myopia levels 10 times those of UK children), and the findings may not be comparable. Although, we do not have direct data on refractive status during childhood for these UK cohorts, impaired unaided distance vision was predominantly due to refractive error (mostly myopia) in this age group. Reassuringly, the prevalence of reduced vision in these cohorts was comparable with estimates from other white populations of similar age (3, 53) and agrees well with estimates of myopia in contemporary studies of white children.(54, 55) Unlike the South Asian study in this age group (15), these UK cohorts collected data on exposures prospectively without knowledge of visual outcome, eliminating the potential for bias in recalling initial feeding status. The detailed information ascertained on early feeding practices, particularly in the 1946 and 1970 birth cohorts, enabled us to show that the duration and exclusivity of breastfeeding (1946 birth cohort only) exerted little influence on visual outcome. In addition, childhood vision has remained relatively stable over a time period when there has been a stark decline in rates of initial breastfeeding (65%, 43%, and 22% in those breastfed for >1 mo in the 1946, 1958, and 1970 cohorts, respectively). Hence, our findings provide strong evidence that infant feeding is unrelated to visual outcome in childhood and adolescence in a European setting. This result may not be applicable to populations in developing countries with poorer nutrition in early life. It remains possible that the benefits of breastfeeding on visual outcome in later life might be masked, particularly in later cohorts, by overall improvements in early nutrition (44, 45). Comparison of heights, as a marker of early nutrition, across the cohorts at the same age showed that at 11 y of age, children born in 1958 were {approx}3.5 cm taller than those born in 1948. Similarly, at age 16 y, those born in 1970 were {approx}2.5 cm taller than those born in 1958. However, there was no evidence of any differential association of breastfeeding versus formula feeding, at either age, on vision by tertiles of height in childhood or adolescence (data not shown).

Factors associated with unaided distance vision of 6/12 or worse
Although early feeding showed little association with visual outcome, other sociodemographic factors showed stronger influences. A 10% higher prevalence of poor unaided vision in girls than in boys was consistently observed at 10–11 y of age across the 3 cohorts. Similar sex differences have been reported in visual outcome with the use of the same cutoffs used in this study, in children aged as young as 6 y (35), and in levels of myopia among Singaporean children aged 7–9 y (10). The magnitude of difference between girls and boys was more marked in adolescence (23%), and a higher prevalence of myopia in females has been shown to persist into adulthood in contemporary cohorts (2). The reason for this sex difference remains unclear, but the development of the difference over time may reflect that this is environmentally programmed from an early age. There is a consistent association across cohorts of parental education, maternal age, and number of older siblings (indicating a higher risk for first births) to unaided distance vision, with trends emerging in childhood and becoming stronger by adolescence. These associations were observed previously, but were not quantified collectively, with adjustment for potential confounders (24, 27, 49, 50). This study has enabled us to show that parental education was more strongly related to poor vision than was social class in this age group. This pattern appears to develop in childhood and appears stronger in adolescence, which may explain why associations were not observed in other subjects at an earlier age (35). The positive association between risk of adverse visual outcome and maternal age may partly be explained by maternal education, because highly educated mothers tend to postpone their first birth. The average age at first birth was lowest in the 1970 BCS and highest in the 1946 NSHD. This agrees with population values from England and Wales, which showed a decline in the average age of first births during this time period (56). Despite the strong association observed between maternal age and visual outcome, these time trends in maternal age do not appear to have materially influenced levels of vision in these different cohorts.

In the current analyses, a 1-kg increase in birth weight reduced the risk of poor vision in childhood, with no effect being observed by adolescence. Measures of growth in early life, including birth weight (31), length, and head circumference at birth (30, 31) and height in childhood have been found to be associated with refractive error and ocular biometric measures in young children (57, 58). A study from Denmark showed that the association between birth weight and reduced vision might persist into adulthood (59). However, in the current study, it is not clear why the relation observed in childhood was not replicated in adolescence.

The degree to which myopia is determined by genes or environmental influences (or both) is much debated. Although family studies have reported correlations in refractive error across generations and within twins (60-63), this may reflect shared environments rather than shared genes. We have shown that first births are at highest risk of poor vision, and this risk declines as the number of older siblings increases. It is likely that environmental factors related to prolonged near vision and emphasis on academic achievements are also likely to be correlated with sibship and birth order. In the 1946 cohort, those from smaller families received better care and had higher cognitive performance than did those from relatively larger families (64). There is a large body of evidence of an association between the individual level of educational attainment and prevalence of myopia (11), but the causal pathways remain unclear. Certainly, the worldwide urban versus rural comparisons of the prevalence of myopia are consistent with a near work hypothesis that increased reading and computer use may be a risk factor for myopia (65). Gene loci for high levels of myopia have been identified (66-70), but not for low to mid levels. Hence, these lower levels of myopia, which are more common, may be more environmentally determined.

Conclusions
Eyes are part of the wider base of neurodevelopment, and our findings are contrary to the results of studies linking improved cognitive development with breast feeding rather than formula feeding (37, 39-42). It has been has suggested that these effects are subject to residual confounding (36), especially from maternal intelligence (43). Breastfeeding is associated with early life social environment, and this influences both childhood behavior and lifestyle. It is important that such confounders operating across the life course are identified and taken into consideration, as we have done in the current study, to examine the association between early nutrition and vision outcomes in later life. Breastfeeding is beneficial for many reasons, but does not appear to offer any protection against poor unaided vision (as a marker of myopia) in childhood or adolescence in a Westernized setting. In countries with poorer nutrition in early life, the associations between breastfeeding and visual outcome may be different. Whether early myopia can be regarded as a neurological outcome is debatable, because it may represent a physiologic response to the wider environment associated with educational levels and near work that accompanies modern economic development (11). We have shown that more important predictors of poor unaided vision include parental education, maternal age, birth order, and possibly birth weight. However, unlike infant feeding practices, these latter factors are less modifiable. Hence, the burden of early myopia seems likely to increase because of increasing economic development.


   ACKNOWLEDGMENTS
 
We acknowledge the original data creators, depositors, copyright holders, and funders of the Data Collections and the UK Data Archive (University of Essex, Colchester, United Kingdom) for use of data from the 1970 BCS and the National Child and Development Survey. They bear no responsibility for the analysis or interpretation of this data.

The authors’ responsibilities were as follows—ARR and CGO: drafted the manuscript, carried out the statistical analysis, and as guarantors accepted full responsibility for the integrity of the work as a whole. All authors contributed substantially to the conception and design of this study, had access to the data, contributed to the revision of the manuscript, and approved the final version to be published. None of the authors had any conflicts of interest to declare.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Congdon NG, Friedman DS, Lietman T. Important causes of visual impairment in the world today. JAMA 2003;290:2057–60.[Free Full Text]
  2. Kempen JH, Mitchell P, Lee KE, et al. The prevalence of refractive errors among adults in the United States, Western Europe, and Australia. Arch Ophthalmol 2004;122:495–505.[Abstract/Free Full Text]
  3. Junghans B, Kiely PM, Crewther DP, Crewther SG. Referral rates for a functional vision screening among a large cosmopolitan sample of Australian children. Ophthalmic Physiol Opt 2002;22:10–25.[Medline]
  4. Kleinstein RN, Jones LA, Hullett S, et al. Refractive error and ethnicity in children. Arch Ophthalmol 2003;121:1141–7.[Abstract/Free Full Text]
  5. Lam CS, Goldschmidt E, Edwards MH. Prevalence of myopia in local and international schools in Hong Kong. Optom Vis Sci 2004;81:317–22.[Medline]
  6. Cumberland PM, Peckham CS, Rahi JS. Inferring myopia over the lifecourse from uncorrected distance visual acuity in childhood. Br J Ophthalmol 2007;91:151–3.[Abstract/Free Full Text]
  7. Weale RA. Epidemiology of refractive errors and presbyopia. Surv Ophthalmol 2003;48:515–43.[Medline]
  8. Chow YC, Dhillon B, Chew PT, Chew SJ. Refractive errors in Singapore medical students. Singapore Med J 1990;31:472–3.[Medline]
  9. Saw SM. A synopsis of the prevalence rates and environmental risk factors for myopia. Clin Exp Optom 2003;86:289–94.[Medline]
  10. Saw SM, Tong L, Chua WH, et al. Incidence and progression of myopia in Singaporean school children. Invest Ophthalmol Vis Sci 2005;46:51–7.[Abstract/Free Full Text]
  11. Morgan I, Rose K. How genetic is school myopia? Prog Retin Eye Res 2005;24:1–38.[Medline]
  12. Birch E, Birch D, Hoffman D, Hale L, Everett M, Uauy R. Breast-feeding and optimal visual development. J Pediatr Ophthalmol Strabismus 1993;30:33–8.[Medline]
  13. Carlson SE, Ford AJ, Werkman SH, Peeples JM, Koo WW. Visual acuity and fatty acid status of term infants fed human milk and formulas with and without docosahexaenoate and arachidonate from egg yolk lecithin. Pediatr Res 1996;39:882–8.[Medline]
  14. Williams C, Birch EE, Emmett PM, Northstone K. Stereoacuity at age 3.5 y in children born full-term is associated with prenatal and postnatal dietary factors: a report from a population-based cohort study. Am J Clin Nutr 2001;73:316–22.[Abstract/Free Full Text]
  15. Chong YS, Liang Y, Tan D, Gazzard G, Stone RA, Saw SM. Association between breastfeeding and likelihood of myopia in children. JAMA 2005;293:3001–2.[Free Full Text]
  16. Huttenlocher PR, de Courten C. The development of synapses in striate cortex of man. Hum Neurobiol 1987;6:1–9.[Medline]
  17. Blakemore C. Mysteries in the making of the cerebral cortex. Ciba Found Symp 1995;193:1–20.[Medline]
  18. Morale SE, Hoffman DR, Castaneda YS, Wheaton DH, Burns RA, Birch EE. Duration of long-chain polyunsaturated fatty acids availability in the diet and visual acuity. Early Hum Dev 2005;81:197–203.[Medline]
  19. Innis SM, Nelson CM, Lwanga D, Rioux FM, Waslen P. Feeding formula without arachidonic acid and docosahexaenoic acid has no effect on preferential looking acuity or recognition memory in healthy full-term infants at 9 mo of age. Am J Clin Nutr 1996;64:40–6.[Abstract/Free Full Text]
  20. Makrides M, Neumann MA, Jeffrey B, Lien EL, Gibson RA. A randomized trial of different ratios of linoleic to alpha-linolenic acid in the diet of term infants: effects on visual function and growth. Am J Clin Nutr 2000;71:120–9.[Abstract/Free Full Text]
  21. Auestad N, Halter R, Hall RT, et al. Growth and development in term infants fed long-chain polyunsaturated fatty acids: a double-masked, randomized, parallel, prospective, multivariate study. Pediatrics 2001;108:372–81.[Abstract/Free Full Text]
  22. Innis SM, Akrabawi SS, Diersen-Schade DA, Dobson MV, Guy DG. Visual acuity and blood lipids in term infants fed human milk or formulae. Lipids 1997;32:63–72.[Medline]
  23. Krasevec JM, Jones PJ, Cabrera-Hernandez A, Mayer DL, Connor WE. Maternal and infant essential fatty acid status in Havana, Cuba. Am J Clin Nutr 2002;76:834–44.[Abstract/Free Full Text]
  24. Peckham CS, Gardiner PA, Goldstein H. Acquired myopia in 11-year-old children. Br Med J 1977;1:542–5.[Medline]
  25. Carlson SE, Cooke RJ, Rhodes PG, Peeples JM, Werkman SH. Effect of vegetable and marine oils in preterm infant formulas on blood arachidonic and docosahexaenoic acids. J Pediatr 1992;120(suppl):S159–67.[Medline]
  26. Guesnet P, Pugo-Gunsam P, Maurage C, et al. Blood lipid concentrations of docosahexaenoic and arachidonic acids at birth determine their relative postnatal changes in term infants fed breast milk or formula. Am J Clin Nutr 1999;70:292–8.[Abstract/Free Full Text]
  27. Douglas JWB, Ross JM, Simpson HR. The ability and attainment of short-sighted pupils. J Royal Stat Soc (series A) 1967;130:479–503.
  28. Rodgers B. Feeding in infancy and later ability and attainment: a longitudinal study. Dev Med Child Neurol 1978;20:421–6.[Medline]
  29. Huynh SC, Wang XY, Ip J, et al. Prevalence and associations of anisometropia and aniso-astigmatism in a population based sample of 6 year old children. Br J Ophthalmol 2006;90:597–601.[Abstract/Free Full Text]
  30. Ojaimi E, Robaei D, Rochtchina E, Rose KA, Morgan IG, Mitchell P. Impact of birth parameters on eye size in a population-based study of 6-year-old Australian children. Am J Ophthalmol 2005;140:535–7.[Medline]
  31. Saw SM, Tong L, Chia KS, et al. The relation between birth size and the results of refractive error and biometry measurements in children. Br J Ophthalmol 2004;88:538–42.[Abstract/Free Full Text]
  32. Hall N, Sayer AA, Evans J, Rauf A, Cooper C, Wormald R. Visual acuity and its relationship to early growth, eye disease, and aging in north Hertfordshire. Ophthalmic Epidemiol 1998;5:83–90.[Medline]
  33. Sperduto RD, Seigel D, Roberts J, Rowland M. Prevalence of myopia in the United States. Arch Ophthalmol 1983;101:405–7.[Abstract]
  34. Goldschmidt E. On the etiology of myopia. An epidemiological study. Acta Ophthalmol (Copenh) 1968;suppl 98:1.
  35. Robaei D, Rose K, Ojaimi E, Kifley A, Huynh S, Mitchell P. Visual acuity and the causes of visual loss in a population-based sample of 6-year-old Australian children. Ophthalmology 2005;112:1275–82.[Medline]
  36. Drane DL, Logemann JA. A critical evaluation of the evidence on the association between type of infant feeding and cognitive development. Paediatr Perinat Epidemiol 2000;14:349–56.[Medline]
  37. Jain A, Concato J, Leventhal JM. How good is the evidence linking breastfeeding and intelligence? Pediatrics 2002;109:1044–53.[Abstract/Free Full Text]
  38. Rogers IS, Emmett PM, Golding J. The incidence and duration of breast feeding. Early Hum. Dev 1997;49(suppl):S45–74.[Medline]
  39. Lucas A, Morley R, Cole TJ, Lister G, Leeson-Payne C. Breast milk and subsequent intelligence quotient in children born preterm. Lancet 1992;339:261–4.[Medline]
  40. Golding J, Rogers IS, Emmett PM. Association between breast feeding, child development and behaviour. Early Hum Dev 1997;49(suppl):S175–84.[Medline]
  41. Anderson JW, Johnstone BM, Remley DT. Breast-feeding and cognitive development: a meta-analysis. Am J Clin Nutr 1999;70:525–35.[Abstract/Free Full Text]
  42. Richards M, Hardy R, Wadsworth ME. Long-term effects of breast-feeding in a national birth cohort: educational attainment and midlife cognitive function. Public Health Nutr 2002;5:631–5.[Medline]
  43. Der G, Batty GD, Deary IJ. Effect of breast feeding on intelligence in children: prospective study, sibling pairs analysis, and meta-analysis. BMJ 2006;333:945.[Abstract/Free Full Text]
  44. Prynne CJ, Paul AA, Mishra GD, Hardy RJ, Bolton-Smith C, Wadsworth ME. Sociodemographic inequalities in the diet of young children in the 1946 British birth cohort. Public Health Nutr 2002;5:733–45.[Medline]
  45. Wadsworth MEJ, Butterworth S, Montgomery S, Ehlin A, Bartley M. Changing Britain, changing lives. In: Ferri E, Bynner J, Wadsworth MEJ, eds. London, United Kingdom: Institute of Education Press, 2003:221.
  46. Wadsworth ME, Butterworth SL, Hardy RJ et al. The life course prospective design: an example of benefits and problems associated with study longevity. Soc Sci Med 2003;57:2193–205.[Medline]
  47. Power C, Elliott J. Cohort profile: 1958 British birth cohort (National Child Development Study). Int J Epidemiol 2006;35:34–41.[Free Full Text]
  48. Bynner D, Butler N, Ferri E, Shepherd P, Smith K. The design and conduct of the 1999-2000 surveys of the national child development study and the 1970 British birth cohort study. UK data archive. London: Centre for Longitudinal Studies, Institute of Education, 2002. (CLS Cohort Studies Working Paper 1.)
  49. Peckham C, Adams B. Vision screening in a national sample of 11-year-old children. Child Care Health Dev 1975;1:93–106.[Medline]
  50. Peckham C, Pearson R. Preliminary findings at the age of 16 years on children in the National Child Development Study (1958 cohort). Public Health 1976;90:271–80.[Medline]
  51. Butler NR, Golding J, Haslum M, Stewart-Brown S. Recent findings from the 1970 child health and education study: preliminary communication. J R Soc Med 1982;75:781–4.[Abstract]
  52. Deeks JJ, Altman DG, Bradburn MJ. Statistical methods for examining heterogeneity and combining results from several studies in meta-analysis. In: Egger M, Davey Smith G, Altman DG, eds. Systematic reviews in health care: meta-analysis in context. London, United Kingdom: BMJ Books, 2001:285–312.
  53. Lee DJ, Gomez-Marin O, Lam BL, Ma F, Vilar NF. Prevalence of usual-corrected distance visual acuity impairment in Hispanic and non-Hispanic children and adolescents. Paediatr Perinat Epidemiol 2000;14:357–62.[Medline]
  54. Laatikainen L, Erkkila H. Refractive errors and other ocular findings in school children. Acta Ophthalmol (Copenh) 1980;58:129–36.[Medline]
  55. Blum HL, Peter HB, Bettman JW. Vision screening for elementary schools. Los Angeles, CA: University of California Press, 1959.
  56. Office of National Statistics. Births: 1938-2004. Mean age of women at marriage and at live birth, England and Wales. London, United Kingdom: ONS, 2006. [Historical Series FM1, no, 13 (27, 28, 29, and 33)].
  57. Ojaimi E, Morgan IG, Robaei D, et al. Effect of stature and other anthropometric parameters on eye size and refraction in a population-based study of Australian children. Invest Ophthalmol Vis Sci 2005;46:4424–9.[Abstract/Free Full Text]
  58. Saw SM, Chua WH, Hong CY, et al. Height and its relationship to refraction and biometry parameters in Singapore Chinese children. Invest Ophthalmol Vis Sci 2002;43:1408–13.[Abstract/Free Full Text]
  59. Olsen J, Sorensen HT, Steffensen FH, et al. The association of indicators of fetal growth with visual acuity and hearing among conscripts. Epidemiology 2001;12:235–8.[Medline]
  60. Wu MM, Edwards MH. The effect of having myopic parents: an analysis of myopia in three generations. Optom Vis Sci 1999;76:387–92.[Medline]
  61. Lee KE, Klein BE, Klein R, Fine JP. Aggregation of refractive error and 5-year changes in refractive error among families in the Beaver Dam Eye Study. Arch Ophthalmol 2001;119:1679–85.[Abstract/Free Full Text]
  62. Lyhne N, Sjolie AK, Kyvik KO, Green A. The importance of genes and environment for ocular refraction and its determiners: a population based study among 20–45 year old twins. Br J Ophthalmol 2001;85:1470–6.[Abstract/Free Full Text]
  63. Wojciechowski R, Congdon N, Bowie H, Munoz B, Gilbert D, West SK. Heritability of refractive error and familial aggregation of myopia in an elderly American population. Invest Ophthalmol Vis Sci 2005;46:1588–92.[Abstract/Free Full Text]
  64. Douglas JWB. The home and the school. London, United Kingdom: McGibbon & Kee, 1964:129–32.
  65. Mutti DO, Mitchell GL, Moeschberger ML, Jones LA, Zadnik K. Parental myopia, near work, school achievement, and children's refractive error. Invest Ophthalmol Vis Sci 2002;43:3633–40.[Abstract/Free Full Text]
  66. Young TL, Ronan SM, Alvear AB, et al. A second locus for familial high myopia maps to chromosome 12q. Am J Hum Genet 1998;63:1419–24.[Medline]
  67. Young TL, Atwood LD, Ronan SM, et al. Further refinement of the MYP2 locus for autosomal dominant high myopia by linkage disequilibrium analysis. Ophthalmic Genet 2001;22:69–75.[Medline]
  68. Naiglin L, Gazagne C, Dallongeville F, et al. A genome wide scan for familial high myopia suggests a novel locus on chromosome 7q36. J Med Genet 2002;39:118–24.[Free Full Text]
  69. Lam DS, Tam PO, Fan DS, Baum L, Leung YF, Pang CP. Familial high myopia linkage to chromosome 18p. Ophthalmologica 2003;217:115–8.[Medline]
  70. Paluru P, Ronan SM, Heon E, et al. New locus for autosomal dominant high myopia maps to the long arm of chromosome 17. Invest Ophthalmol Vis Sci 2003;44:1830–6.[Abstract/Free Full Text]
Received for publication August 1, 2007. Accepted for publication November 1, 2007.


Related articles in AJCN:

Infant feeding and vision
William C Heird
AJCN 2008 87: 1120. [Full Text]  



This article has been cited by other articles:


Home page
 Am. J. Clin. Nutr.Home page
W. C Heird
Infant feeding and vision
Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1120 - 1120.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Related articles in AJCN
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rudnicka, A. R
Right arrow Articles by Strachan, D. P
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rudnicka, A. R
Right arrow Articles by Strachan, D. P
Agricola
Right arrow Articles by Rudnicka, A. R
Right arrow Articles by Strachan, D. P

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS