PEDIATRICS Vol. 107 No. 2 February 2001, pp. 357-362
Maternal Prenatal Lifestyle Factors and Infectious Disease in Early Childhood: A Follow-Up Study of Hospitalization Within a Danish Birth Cohort
, and
From the * Danish Epidemiology Science Centre, University of
Aarhus, Aarhus, Denmark; Department of Clinical Epidemiology, Aarhus
University Hospital and Aalborg Hospital, Aarhus, Denmark; and
§ Shanghai Institute of Planned Parenthood Research, Shanghai, China.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objectives. To examine whether maternal prenatal lifestyle factors were associated with the risk of hospitalization with infectious disease during early childhood and whether a possible association was modified by fetal growth reduction.
Methods. The study was based on a birth cohort of 10 400 newborns whose mothers attended the midwife centers in Odense and Aalborg, Denmark, from April 1984 to April 1987 at approximately the 36th week of gestation. Information on hospitalization with infectious disease was extracted from the National Hospital Discharge Registry, and newborns were followed up to the end of 1996.
Results. Among 6022 children who were hospitalized at least once, 31.4% (n = 1892) were hospitalized with infectious diseases. The cumulative incidence rate of hospitalization with infections in children from the age of 6 months to 12 years was 18.9%. The incidence rate ratio in children born to mothers smoking during pregnancy was 1.24 (95% confidence interval: 1.13-1.36) compared with those of nonsmoking mothers. Mothers whose body mass index was <18 kg/m2 were more likely to give birth to infants who were hospitalized with infectious disease (incidence rate ratio: 1.29; 95% confidence interval: 1.05-1.59). The increased risk was present in children only up to the age of 5 years.
Conclusions. The study shows that maternal smoking during pregnancy and a low prepregnancy body mass index are associated with a higher risk of hospitalization with infectious disease during early childhood. These associations are independent of fetal growth indicators. Key words: lifestyle factors, maternal, infectious disease, child.
Fetal growth impairment has been associated with chronic
diseases, such as cardiovascular disease and diabetes.1
Although there are a number of possible reasons for these associations, particular attention has been given to the programming hypothesis, which postulates that impaired fetal growth may permanently alter organ
function.2 Most studies have focused on the functions of
the liver and the pancreas, but the immune system may be subjected to
programming as well.3 Several factors may influence fetal
growth, eg, nutrition, smoking, alcohol intake, maternal diseases,
genetic factors, and social conditions. However, it is not known
whether reduced fetal growth itself is responsible for the observed
associations, regardless of the underlying determinants of fetal
growth.
We observed in previous study that prematurity and reduced birth length
were associated with hospitalization with infectious disease in
children up to the age of 12 years. Other studies have shown an
association between low birth weight and infectious disease mortality.4,5 Some fetal growth determinants, such as
maternal smoking, may affect the levels of immune components in the
blood of newborns,6,7 but fetal growth may be nothing but
an epiphenomenon to a causal link between external exposures and
susceptibility to infections.
The aim of this study was to examine whether maternal prenatal
lifestyle factors, including fetal growth determinants, were associated
with the risk of hospitalization with infectious disease during early
childhood. We present our findings before and after adjustment for
fetal growth measures to evaluate whether a possible association was
modified by fetal growth reduction.
Study Population and Data Linkage
The study cohort comprised children born to mothers attending
the midwife centers at approximately the 36th week of gestation in 2 study areas, Odense and Aalborg (in Denmark), from April 1984 to April
1987. A total of 11 980 women, >80% of all pregnant women in these
areas, participated in the study and completed a questionnaire on their
sociodemographic characteristics and lifestyle during pregnancy. After
delivery, obstetric information, including detailed fetal growth
measurements, was extracted from the medical records and linked with
the questionnaire data. This cohort has previously been
described.8 Children born to these mothers during the
above period were identified through the Civil Registration System
using the mothers' Central Personal Registration (CPR) number, given
to all Danish residents since 1968. From the National Hospital
Discharge Registry (NHDR),9 we extracted information on
the children's hospitalizations from birth to the end of 1996 (up to
12 years of age) and linked this with the background information by
means of the CPR number.
After excluding stillbirths, multiple births, and children with
congenital malformations (identified from the NHDR data), 10 440
newborns were eligible for this study.
Exposure Data
Maternal smoking information was available as ever having smoked
during pregnancy, current cigarette consumption per day (in the 36th
week of gestation), and nicotine level of the cigarettes currently
smoked. Paternal smoking information was also obtained from mothers as
smoking or not at the time of the interview. The amount of alcohol
consumption per week during pregnancy was calculated in grams according
to normal quantities of alcoholic beverages in Denmark (12 g per unit).
Coffee and tea consumption was recorded as the average number of cups
per day. Food frequency data included intake of vegetables, fruit,
offal, fish, and poultry as the number of meals during the latest
month. Maternal prepregnancy body mass index (BMI) was calculated as
weight in kilograms divided by height squared in meters. Gestational
weight gain was calculated by subtracting the prepregnancy weight from
the prelabor weight, but the prelabor weight was missing for 27% of
cohort members because of lack of staff, time, or opportunities to
weigh the women before delivery.
Confounding Variables
The questionnaire data provided information on sociodemographic
factors. These factors may confound the association between prenatal
factors and hospitalization with infectious disease by being associated
with prenatal exposures and by being alternative causes of
susceptibility to infections (or being associated with the threshold
for hospitalization). The social status of the couples was divided into
3 levels based on the one with the highest occupational status;
maternal and paternal education was divided into 3 levels according to
the highest school grade attained; housing density was measured as the
number of rooms per person in the dwelling. Other information included
parental age, maternal marital and cohabitation status, parity, and
number of siblings.
Outcome Data
All inpatients admitted for an infectious disease in Denmark are
treated at public hospitals, and since 1977 all patients discharged
from hospitals have been registered in the NHDR.9 The
recorded information includes the CPR number, hospital, department, date of admission and discharge, and up to 20 diagnoses according to
the International Classification of Diseases (ICD). During the study, a Danish version of ICD-8 was used before 1994 and ICD-10
since then. However, the registration does not specify the reason for
admission. The main outcome variable was the first hospitalization with
a discharge diagnosis of any selected infectious diseases presented in
Table 1, regardless of its order in the
recording, according to the NHDR. To corroborate our findings, the main
outcome was restricted to some subgroups of discharge diagnoses such as
the first diagnosis (supposed to be the primary or main one), the first
diagnosis of a serious infection (meningitis, pneumonia, appendicitis,
and kidney infections), and respiratory tract infections or any other
infections.
TABLE 1
METHODS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
ICD Codes for the Infectious Diseases Used as Endpoints and the Number
of Children Hospitalized for the First Time With Infectious Disease
During Follow-Up
Follow-Up
The Civil Registration System, which contains information on the vital status of the study population, was used to identify children who died or emigrated during the follow-up period. Because hospitalizations during the first few months after birth often are birth-related, and because breastfeeding, to some extent, protects against infections, we started the follow-up of the cohort at the age of 6 months. The time at risk was calculated from the age of 6 months to the first hospitalization with an infectious disease, death, emigration, or the end of the study, whichever came first. Children who died (n = 24) or emigrated (n = 16) during the first 6 months after birth were excluded from the analyses. Among the 10 400 children alive and living in Denmark at the age of 6 months, all had a complete follow-up except for the 1% who died (n = 20) or emigrated (n = 91) during the follow-up and who thus were censored in the analyses.
Statistical Analysis
We used the life table method to estimate the cumulative incidence rate (CIR) of hospitalization with infectious disease from the age of 6 months to 12 years. The relative risk (or hazard ratio) was presented as the incidence rate ratio (IRR) estimated with the Cox regression model.10 By examining the associations between these potential confounders and the cumulative incidence rate of all hospitalizations, parental age, family social group, and maternal cohabitation status were identified as factors that may influence the threshold for hospitalization. The number of siblings at birth was also included because it was associated with hospitalization with infectious disease. Univariate analyses were performed before multivariate analyses. In additional analyses, we adjusted for fetal growth measures such as gestational age, birth weight, birth length, and head circumference. Data analyses were performed in SPSS, Version 9.0 (SPSS, Chicago, IL).
| |
RESULTS |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Among the 10 400 children, 6022 were hospitalized at least once, and 1892 (31.4%) of them with 1 or more infectious diseases (CIR: 18.9%; 95% confidence interval [CI]: 18.2-19.7%). Table 1 shows the number of children hospitalized for the first time with an infectious disease. For 80.4% of these hospitalizations, the first diagnosis listed in the discharge record was an infection, most often a respiratory tract infection.
Descriptive information on the prenatal risk factors of interest and
their associations with hospitalization with infectious disease are
presented in Table 2. A higher risk was
found in children whose parents smoked or whose mothers had a high
consumption of coffee or tea (
8 cups/day), low consumption of fruit
(<5 meals/month), or had low maternal prepregnancy BMI (<18
kg/m2).
|
Table 3 shows that, after adjustment for sociodemographic factors, children born to mothers who smoked during pregnancy were more likely to be hospitalized with infectious disease than those born to nonsmoking mothers (adjusted IRR: 1.24; 95% CI: 1.13-1.36). Adjustment for other possible determinants, such as intake of alcohol, coffee, tea, and fruit during pregnancy did not alter the estimate of the association. The risk increased with the number of cigarettes smoked daily and the nicotine level (which is closely correlated to the content of tar). The association with the nicotine level was also independent of the number of cigarettes smoked daily (adjusted IRR per level increase of nicotine: 1.12; 95% CI: 1.01-1.25 among children of smoking mothers). Adjustment for fetal growth measurements in the above analyses gave similar estimates (adjusted IRR: 1.27; 95% CI: 1.15-1.40 for children of mothers who smoked during pregnancy).
|
Table 4 shows the association with paternal smoking in relation to maternal smoking. Paternal smoking did not increase the risk of hospitalization with infections in children of smoking or nonsmoking mothers. The highest risk was seen in children of mothers who smoked during pregnancy, regardless of paternal smoking habit.
|
To examine whether the association with maternal smoking was time-dependent, we divided the follow-up into 4 periods. The association was strongest for the period closest to the time of birth (adjusted IRR: 1.45; 95% CI: 1.18-1.78 for the second half year of life), and little, if any, was seen after the age of 5 years (adjusted IRR: 1.11; 95% CI: .93-1.33).
Similar associations with parental smoking during pregnancy were found when hospitalization was restricted to respiratory infectious diseases or any other infections (data not shown).
The associations between maternal intake of alcohol, coffee, tea, and fruit during pregnancy and hospitalization with infectious disease are shown separately for the children exposed and the ones not exposed in utero to maternal smoking (Table 5). The table shows that a high level of alcohol intake was associated with the risk of hospitalization with infectious disease, but only in the children of nonsmoking mothers. A high level of tea intake was associated with hospitalization with infectious disease in both groups of children. A potential protective effect of fruit consumption was shown only in children of smoking mothers.
|
Maternal nutritional status in relation to the risk of hospitalization
with infectious disease was measured using 2 indicators
maternal prepregnancy BMI and gestational weight gain. There was an increased risk of hospitalization with infections in children of mothers whose
BMI was <18 kg/m2 compared with those born to
mothers with a BMI of 18 kg/m2 or more (adjusted
IRR: 1.29; 95% CI: 1.05-1.59). The risk was not associated with
maternal weight gain during pregnancy (Table
6), and the associations remained
unchanged after adjustment for gestational age and fetal growth
measures. As with the influence of maternal smoking, a stronger
association with maternal BMI was seen in the period close to the time
of birth (adjusted IRR: 1.37; 95% CI: 1.12-1.68 for the period
between the age of 6 months and 4 years vs adjusted IRR: 1.05; 95% CI:
.72-1.55 for the period between 5 and 12 years). Stratification of the
analyses by either maternal prepregnancy BMI or weight gain during
pregnancy did not show any difference in the associations between
maternal smoking, alcohol, tea, and fruit intake and hospitalization
with infectious disease.
|
We checked our findings by looking at proportional hospitalization rates. By analyzing children hospitalized with infectious disease as a proportion of all hospitalized children, we adjusted for differential hospitalization thresholds because they were all hospitalized. Among the children who experienced at least 1 hospitalization before the age of 12 years, the proportion of children hospitalized with infectious disease was higher in children of mothers who smoked during pregnancy (IRR: 1.14; 95% CI: 1.05-1.22) or whose BMI was <18 kg/m2 before pregnancy (IRR: 1.25; 95% CI: 1.08-1.43). This indicates that the excessive risk of infections is not only related to a lower threshold for hospitalization.
The supplementary analyses restricting the outcome to the first diagnosis or to the first diagnosis of a serious infection produced similar results to those from the main outcome presented above (data not shown).
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
This 9- to 12-year follow-up study found an association between maternal smoking during pregnancy and hospitalization with infectious disease during early childhood. The study also showed that children born to mothers with a low prepregnancy BMI had a high risk of hospitalization with infectious disease, regardless of the gestational weight gain. The associations, however, seemed to disappear after the age of 5 years. These associations were independent of indicators of fetal growth.
Our design has its strengths and limitations. Because of the prospective nature of the data, differential misclassification is unlikely. We had rather detailed information on prenatal exposures and a long and complete follow-up. We have data only on hospital discharge diagnoses, and we expect that only the most severe cases involved hospitalization. The data quality of the discharge registry is considered good for severe infections11 but probably less so for common infections, which makes the analyses vulnerable to differential threshold for hospitalization. Because social factors may influence the likelihood of hospitalization, we adjusted for these factors in the analyses, but residual confounding of social status may still bias the results to some extent. Supplementary analyses using various outcome measures showed results that, to some extent, distracted from the hypothesis that an association is caused by differential thresholds for hospitalization according to the risk factors under study.
The effects of maternal use of tobacco, alcohol, and caffeine on a fetus have been examined before, but the evidence for long-term effects in children is limited.12,13 A maternal-placental-fetal effect on infectious disease was not supported by our previous study in which no association was detected between placental size and hospitalization with infectious disease. This suggests direct effects of some prenatal factors on the functional and organic development of a fetus by virtue of their ability to cross the placental barrier.12,13 Our data showed that the associations were independent of gestational age and fetal growth indicators.
Taylor and Wadsworth14 found that rates of admissions to hospitals for lower respiratory tract diseases in children whose mothers started smoking postnatally were not higher than in those whose mothers remained nonsmokers. This suggests that the influence of maternal smoking is mainly through a prenatal effect. The lack of effect of paternal smoking in our study suggests that environmental tobacco smoking is not the most likely explanation for the association we found. Our findings of a protective role of fruit consumption during pregnancy only in smoking mothers point to a programming effect rather than to a postnatal effect.
Studies on adults have shown that smoking reduces serum levels of immunoglobulin G, immunoglobulin M, and immunoglobulin A,15 neutrophils,16 and natural killer cytotoxic activity,17,18 but these effects are temporary and reversible.19,20 The increased incidence of urinary tract infections and viral diseases during pregnancy in smokers further suggested a smoking effect on the depression of immune functions.21 Because nicotine and alcohol are able to pass the placental barrier12,13 and hypoxia resulting from maternal smoking reduces the oxygen supply to the fetus, a direct influence on fetal development is possible. Studies on the cord blood of neonates have shown that the neutrophils and the natural killer cell activities are reduced in infants born to smoking mothers.7,22 In contrast, the structural development of some target organs could be affected, which may lead to a higher susceptibility to bacterial and viral attacks. Tager et al23 showed that impaired lung growth during childhood and adolescence was associated with maternal smoking.
Studies have shown that maternal and infant nutritional status might serve as either a mediating or a moderating factor in the unfavorable effects of smoking on fetal growth and development.24,25 The study by Zhou and Olsen26 showed the potential benefit of a high gestational weight gain to birth weight, especially in thinner women. However, our data did not show the intermediary effect of maternal nutrition on the association between maternal smoking and hospitalization with infectious disease, and the association with maternal prepregnancy BMI was independent of maternal smoking. Furthermore, a high gestational weight gain did not eliminate the increased risk of hospitalization with infections in children of thin mothers.
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
The study shows that maternal smoking during pregnancy and a low prepregnancy BMI are associated with a higher risk of hospitalization with infectious disease during early childhood. These associations are independent of fetal growth indicators.
| |
ACKNOWLEDGMENTS |
|---|
The study was funded by the Medical Research Council (Grant 12-1663-1). The activities of the Danish Epidemiology Science Center are financed by a grant from the Danish National Research Foundation.
| |
FOOTNOTES |
|---|
Received for publication Jan 24, 2000; accepted Jun 26, 2000.
Reprint requests to (W.Y.) Danish Epidemiology Science Centre, University of Aarhus, Vennelyst Blvd 6, DK-8000 Aarhus C, Denmark. E-mail: yw{at}soci.au.dk
| |
ABBREVIATIONS |
|---|
CPR, Central Personal Registration; NHDR, National Hospital Discharge Registry; BMI, body mass index; ICD, International Classification of Diseases; CIR, cumulative incidence rate; IRR, incidence rate ratio; CI, confidence interval.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
- Leon DA Fetal growth and adult disease. Eur J Clin Nutr 1998; 52:S72-S78
- Godfrey KM, Barker DJ Maternal nutrition in relation to fetal and placental growth. Eur J Obstet Gynecol Reprod Biol 1995; 61:15-22 [CrossRef][Medline]
- Chandra RK Interactions between early nutrition and the immune system. Ciba Found Symp 1991; 156:77-89 [Medline]
-
Read JS,
Clemens JD,
Klebanoff MA
Moderate low birth weight and
infectious disease mortality during infancy and childhood.
Am J Epidemiol
1994;
140:721-733
[Abstract/Free Full Text] -
Victora CG,
Smith PG,
Vaughan JP,
Influence of birth weight on
mortality from infectious diseases: a case-control study.
Pediatrics
1988;
81:807-811
[Abstract/Free Full Text] - Cervi PL, Feighery C IgG subclasses in foetal cord and maternal serum: associations with infections in infancy and smoking in pregnancy. J Clin Lab Immunol 1991; 34:23-30 [Medline]
- Castellazzi AM, Maccario R, Moretta A, Effect of active and passive smoking during pregnancy on natural killer-cell activity in infants. J Allergy Clin Immunol 1999; 103:172-173 [CrossRef][Medline]
- Olsen J, Frische G, Poulsen AO, Kirchheiner H Changing smoking, drinking, and eating behaviour among pregnant women in Denmark: evaluation of a health campaign in a local region. Scand J Soc Med 1989; 17:277-280 [Medline]
- Andersen TF, Madsen M, Jorgensen J, Mellemkjaer L, Olsen JH The Danish National Hospital Register: a valuable source of data for modern health sciences. Dan Med Bull 1999; 46:263-268 [Medline]
- Clayton D, Hills M. Cox's regression analysis. In: Clayton D, Hills M, eds. Statistical Models in Epidemiology. New York, NY: Oxford University Press; 1993:298-306
- Sorensen HT, Hansen I, Ejlersen E, Schonheyder HC, Hamburger H, Sabroe S Identification of cases of meningococcal disease: data quality in two Danish population-based information systems during a 14-year period. Int J Risk Safety Med 1995; 7:179-189
- Abel EL Smoking during pregnancy: a review of effects on growth and development of offspring. Hum Biol 1980; 52:593-625 [Medline]
- Aaronson LS, Macnee CL Tobacco, alcohol, and caffeine use during pregnancy. J Obstet Gynecol Neonatal Nurs 1989; 18:279-287 [CrossRef][Medline]
-
Taylor B,
Wadsworth J
Maternal smoking during pregnancy and lower
respiratory tract illness in early life.
Arch Dis Child
1987;
62:786-791
[Abstract/Free Full Text] -
McMillan SA,
Douglas JP,
Archbold GP,
McCrum EE,
Evans AE
Effect of
low to moderate levels of smoking and alcohol consumption on serum
immunoglobulin concentrations.
J Clin Pathol
1997;
50:819-822
[Abstract/Free Full Text] - Jedrychowski WA, Maugeri U, Adamczyk B Effect of smoking on serum immunoglobulins and cellular blood constituents in healthy individuals. G Ital Med Lav 1986; 8:53-56 [Medline]
- Phillips B, Marshall ME, Brown S, Thompson JS Effect of smoking on human natural killer cell activity. Cancer 1985; 56:2789-2792 [CrossRef][Medline]
- Ferson M, Edwards A, Lind A, Milton GW, Hersey P Low natural killer-cell activity and immunoglobulin levels associated with smoking in human subjects. Int J Cancer 1979; 23:603-609 [Medline]
- Meliska CJ, Stunkard ME, Gilbert DG, Jensen RA, Martinko JM Immune function in cigarette smokers who quit smoking for 31 days. J Allergy Clin Immunol 1995; 95:901-910 [CrossRef][Medline]
-
Miller LG,
Goldstein G,
Murphy M,
Ginns LC
Reversible alterations in
immunoregulatory T cells in smoking: analysis by monoclonal antibodies
and flow cytometry.
Chest
1982;
82:526-529
[Abstract/Free Full Text] - Nymand G Maternal smoking and immunity. Lancet 1974; 2:1379-1380 [Medline]
- Mercelina-Roumans PE, Breukers RB, Ubachs JM, van Wersch JW Hematological variables in cord blood of neonates of smoking and nonsmoking mothers. J Clin Epidemiol 1996; 49:449-454 [CrossRef][Medline]
- Tager IB, Weiss ST, Munoz A, Rosner B, Speizer FE Longitudinal study of the effects of maternal smoking on pulmonary function in children. N Engl J Med 1983; 309:699-703 [Abstract]
- Nafstad P, Jaakkola JJ, Hagen JA, Botten G, Kongerud J Breastfeeding, maternal smoking and lower respiratory tract infections. Eur Respir J 1996; 9:2623-2629 [Abstract]
- Rush D, Andrews J, Kristal A Maternal cigarette smoking during pregnancy, adiposity, social class, and perinatal outcome in Cardiff, Wales, 1965-1977. Am J Perinatol 1990; 7:319-326 [Medline]
- Zhou W, Olsen J Gestational weight gain as a predictor of birth and placenta weight according to pre-pregnancy body mass index. Acta Obstet Gynecol Scand 1997; 76:300-307 [Medline]
Pediatrics (ISSN 0031 4005). Copyright ©2001 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  B. E. Mahon, V. Ehrenstein, M. Norgaard, L. Pedersen, K. J. Rothman, and H. T. Sorensen Perinatal Risk Factors for Hospitalization for Pneumococcal Disease in Childhood: A Population-Based Cohort Study Pediatrics, April 1, 2007; 119(4): e804 - e812. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. T. Sorensen, R. Labouriau, E. S. Jensen, P. B. Mortensen, and H. C. Schonheyder Fetal growth, maternal prenatal smoking, and risk of invasive meningococcal disease: a nationwide case-control study Int. J. Epidemiol., August 1, 2004; 33(4): 816 - 820. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
