Some disinfection by-products in drinking water may be fetotoxic and thus increase the risk of adverse reproductive outcomes among highly exposed pregnant women. Endpoints such as time-to-pregnancy,1 preterm birth,2 low birth weight or small-for-gestational-age (SGA),3–14 spontaneous abortions,15–22 or birth defects3,15,23–30 have been studied, but the empirical evidence for any of these potential hazards is inconclusive and potential biologic mechanisms are poorly understood.10 Chemical disinfection processes in swimming pools result in the formation of disinfection by-products through the reaction of chlorine with organic matter.31 Swimming pool water contains natural organic matter from the tap water itself and also from bathers' sweat, urine, skin particles, hair, microorganisms, cosmetics, and other personal care products.32 The specific types and levels of disinfection by-products depend on numerous factors, including the type and amount of disinfectant used, characteristics of the swimming pool and pool water, and swimmer hygiene.31 Exposure to disinfection by-products from swimming pools may occur through ingestion, inhalation, or dermal absorption when swimming.33–40 For several by-products including the most common trihalomethanes, inhalation and dermal absorption contribute more to the total uptake than does ingestion.41 Only one study has investigated the association of swimming in pools with birth weight; no association was apparent.42 Drinking water in Denmark is rarely chlorinated, which would make it easier to isolate possible health effects of chlorinated swimming pool water. Further, swimming in indoor pools is a recommended and common activity during pregnancy.43,44
We examined the association between swimming during pregnancy and several birth outcomes (preterm and postterm birth; fetal growth measures, including birth weight and SGA; and congenital malformations) in a large cohort of pregnant women in Denmark. We compared swimmers and bicyclists with physically inactive pregnant women. Bicyclists were included as a comparison group in the attempt to separate a possible swimming effect from that of exercise per se. Exercise, in general, has been evaluated in previous publications.43,45,46
The Danish National Birth Cohort is a nationwide population-based cohort with prospectively collected data from pregnant women and their offspring. The intention during recruitment was to invite as many eligible women in Denmark as possible until the goal of 100,000 pregnancies had been reached; inclusion criteria were residence in Denmark, no plans for an induced abortion, and fluency in Danish sufficient to participate in 4 telephone interviews during and after pregnancy. Recruitment took place between 1996 and 2002, using general practitioners to present the consent form together with written information at the first antenatal visit in early pregnancy. The initial data collection included telephone interviews, questionnaires, and blood samples. The 2 pregnancy interviews used in this study were scheduled to take place in pregnancy weeks 12–16 and 30. The median gestational age for the first pregnancy interview was 114 days (10th percentile, 84 days; 90th percentile, 160 days), corresponding to 16.3 completed weeks. For the second interview, the median gestational age was 218 days (10th percentile, 195 days; 90th percentile, 249 days), corresponding to 31.1 completed weeks. About half of the general practitioners agreed to take part in the recruitment process. We estimate that about 60% of the invited women accepted the invitation. More details about the cohort have been presented elsewhere.47
For this analysis, we included all pregnancies with available data from the first pregnancy interview (n = 90,165). This excludes women who had an induced or spontaneous abortion before the time of the first interview, or who could not be contacted by phone for this interview. We also excluded 1965 multiple pregnancies. We included only pregnancies for which the mother had reported swimming, bicycling, or no exercise. Main analyses were based on swimming data from the first pregnancy interview. The Figure describes the selection of women to this study. Specific exclusions were made for each of the 4 endpoints studied, and this resulted in minor variations in sample size (Fig.). To be included in additional analyses of swimming throughout pregnancy, the women needed to have reported swimming, bicycling, or no exercise in both interviews. Because many women changed their exercise habits from the first to the second interview, there is a difference of about 25,000 pregnancies between the 2 analyses. Altogether, 48,781 pregnancies were included in this study. (See eTable; http://links.lww.com/EDE/A365).
Data on Swimming
We used self-reported data from the first and the second pregnancy interview on engagement in physical exercise (yes/no), type of exercise, and duration of each exercise session. We generated 3 exposure categories: (1) any reported swimming, (2) any reported bicycling but no swimming, and (3) no exercise. We present data from early/mid-pregnancy and throughout pregnancy. The questionnaires are available at www.bsmb.dk.
All water activities were included in the swimming group, including antenatal water-exercises or baby-swimming with older children. Although our data did not include information on exercise intensity, we expect both the bicycling and the swimming groups to include all levels from mild to vigorous training.
We obtained mother's age at conception and sex of the offspring from the National Discharge Registry. The first pregnancy interview provided data on prepregnancy body mass index, occupational status, gravidity, parity, previous spontaneous abortions, bleeding in early pregnancy, chronic diseases, uterine fibroids/malformations/cone biopsy, subfecundity, smoking, coffee and alcohol consumption, working hours, working position, physically strenuous work, and psychosocial jobstrain. The adjustment variables varied among the endpoint analyses and are displayed in the table footnotes. The decision on which factors to adjust for was made a priori, based upon existing knowledge regarding risk factors for a given end point.
We abstracted from the National Discharge Registry data on gestational age at birth (days), birth weight (g), length (cm), ponderal index (calculated as [weight in g × 100]/[length in cm3]),48 head circumference (cm), abdominal circumference (cm), placental weight (g), and congenital malformations. We defined preterm birth as a delivery after 153 gestational days (22 completed weeks) but before 259 days (37 completed weeks); postterm birth was a delivery at or after 294 gestational days (42 completed weeks). For SGA we used a 10% cutpoint of the sex- and gestation-specific birth weight within the Danish National Birth Cohort. Subgroups of congenital malformations were defined using ICD10 codes on malformations registered within the first year of life (the 10th revision of International Classification of Diseases). If a child had more than one malformation, the child was included in each subgroup.
We compared reproductive endpoints among swimmers, bicyclists, and nonexercisers. To evaluate the importance of time spent in swimming pools, we also calculated measures of association for 2 levels of time spent swimming and bicycling (<90 and 90+ minutes/week) relative to no exercise. We considered bicycling to be a similar exercise to swimming, in that both are nonweight-bearing sports that can be performed at a range of levels. We compared swimmers with nonexercisers to consider whether any possible effect of swimming on birth outcomes could be explained by physical activity in general, rather than to swimming specifically.
We estimated hazard ratios (HRs) and associated 95% confidence intervals (CIs) for preterm birth and SGA, using Cox regression analysis. If a second pregnancy interview was available, exercise data were updated by the time of the second interview. To adjust for the different entry times into follow-up, we stratified our models by gestational age at the time of the first and the second interview, respectively. For preterm birth, time at risk started from the day a woman completed the 22nd gestational week or on the day of her first pregnancy interview, whichever was later. Follow-up ended at birth or by the time a woman completed gestational week 37, whatever came first. For SGA analyses, the Cox model was constructed in the same way, except that time at risk was not restricted as described above. Hence, no time restrictions were applied to the analyses of SGA. We used linear regression models to calculate mean differences in birth weight, length, ponderal index, head and abdominal circumference, and placental weight. Logistic regression analysis was used to estimate odds ratios (ORs) for postterm birth, any congenital malformation, and subgroups of malformations.
Table 1 shows the distribution of each outcome according to swimming, bicycling, and no exercise. In this population, restricted to women who were in one of these exercise categories, 75% did not engage in any exercise, 14% reported swimming, and 11% reported bicycling (and no swimming). There were fewer preterm births among swimming mothers compared with the other groups. Apart from this, there were only minor differences between swimmers and bicyclists. There were fewer postterm deliveries, more SGA-babies, and a slightly higher occurrence of malformations in the offspring among nonexercising mothers compared with swimmers and bicyclists.
Table 2 shows the association of swimming and bicycling in early/mid-pregnancy with preterm and postterm birth, SGA, congenital malformations, and fetal growth measures, compared with no exercise. We found a slightly decreased risk of preterm birth and a modestly decreased prevalence of overall congenital malformations among swimmers compared with nonexercisers. No major differences were observed between swimming and bicycling. Risk estimates for respiratory and cleft lip/palate malformations indicated a decreased risk among swimmers compared with nonexercisers, but estimates are based on small numbers with substantial uncertainty. Except for cleft lip/palate malformations, malformation estimates were similar among swimmers and bicyclists.
We also used a measure of swimming, bicycling, and no exercise, respectively, throughout pregnancy, ie reported at both the first and the second pregnancy interview (eTable, http://links.lww.com/EDE/A365). The results were very similar to those from early/mid-pregnancy.
In Table 3 the amount of swimming is evaluated. The table shows only minor differences between high and low levels of swimming, and, further, the relation between high levels of swimming and high levels of bicycling reflects the results presented in Table 2.
This study does not indicate any adverse effects of swimming, or other water activities, in early/mid-pregnancy on preterm and postterm birth, SGA, birth weight and other fetal growth measures, or congenital malformations. Similarly reassuring results were found for those who swam throughout pregnancy and those who reported higher amounts of swimming, ie, at least 1.5 hours per week. These findings do not rule out the possibility that chemicals used in indoor pools are fetotoxic at higher exposure levels. Furthermore, although this is a large mother-child cohort, the study has limited power to detect an increased risk for rare outcomes such as specific congenital malformations. Our findings are in line with those of the only previously published study with a specific focus on swimming during pregnancy, by Nieuwenhuijsen et al,42 which indicated no association between swimming and birth weight. We did not observe marked differences in outcomes between swimmers and bicyclists, and the use of a secondary exercise group for comparison (bicyclists) strengthens our conclusions and further supports previous findings. Our data, however, suggest that physical exercise may be beneficial, or alternatively that the modestly elevated risk for preterm birth and perhaps congenital malformations seen among nonexercisers compared with swimmers could be related to medical reasons for not doing exercise (confounding by the indication for not being physically active). Reverse causation may also explain these differences if an affected fetus impairs maternal health during pregnancy, although this seems unlikely particularly in the early phase of pregnancy. Even though both swimming and bicycling have been found to increase uteroplacental vascular resistance, and thereby reduce blood flow to the uterus, the flow in the umbilical artery has been found to remain unchanged.49 This is consistent with the lack of substantial birth weight differences in offspring between the exercise groups and nonexercise group in our study.
Bicycling is a common means of commuting in Denmark, and many of the bicyclists in the study most likely used their bike for commuting as part of a daily routine. Since data on physical activity were both self-reported and did not include information on intensity, we expect the bicycling group to include all levels of intensity. Information on swimming faces a similar problem, because swimming and all other water activities were included. Subdividing the swimming and bicycling groups into 2 levels of weekly time spent on the activity did not change the results.
Even though we have considered bicycling a fair comparison with swimming in this paper, there may be arguments against this assumption. Bicycling is done in a sitting position, which might increase abdominal pressure on the pelvis; bicycling is perhaps more aerobic, because breathing is not as limited as it is during swimming; bicycling uses the lower body, which has larger musculature, while swimming may use more of the upper body; and swimming may allow for better heat dissipation from the mother's body. We have, however, not been able to identify another exercise group that would be a more appropriate comparison group.
Our study addresses exercise as commonly practiced in the population, and most of the participating women are not expected to be heavily exposed to swimming pool chemicals. There is almost no publicly available information on levels of disinfection by-products in swimming pools in Denmark. Studies of swimming pools in other countries, such as Spain, indicate that levels of trihalomethanes in swimming pools are not higher than those found in drinking water.50 Our findings suggest that the current practice of swimming in pools for pregnant women is safe for the endpoints we studied.
The Danish National Research Foundation has established the Danish Epidemiology Science Centre that initiated and created the Danish National Birth Cohort. The cohort is furthermore a result of a major grant from this foundation. Additional support for the Danish National Birth Cohort is obtained from the Pharmacy Foundation, the Egmont Foundation, the March of Dimes Birth Defects Foundation, and the Augustinus Foundation.
1. MacLehose RF, Savitz DA, Herring AH, Hartmann KE, Singer PC, Weinberg HS. Drinking water disinfection by-products and time to pregnancy. Epidemiology
2. Hoffman CS, Mendola P, Savitz DA, et al. Drinking water disinfection by-product exposure and duration of gestation. Epidemiology
3. Bove FJ, Fulcomer MC, Klotz JB, Esmart J, Dufficy EM, Savrin JE. Public drinking water contamination and birth outcomes. Am J Epidemiol
4. Gallagher MD, Nuckols JR, Stallones L, Savitz DA. Exposure to trihalomethanes and adverse pregnancy outcomes. Epidemiology
5. Hinckley AF, Bachand AM, Reif JS. Late pregnancy exposures to disinfection by-products and growth-related birth outcomes. Environ Health Perspect
6. Kallen BA, Robert E. Drinking water chlorination and delivery outcome-a registry-based study in Sweden. Reprod Toxicol
7. Kanitz S, Franco Y, Patrone V, et al. Association between drinking water disinfection and somatic parameters at birth. Environ Health Perspect
8. Kramer MD, Lynch CF, Isacson P, Hanson JW. The association of waterborne chloroform with intrauterine growth retardation. Epidemiology
9. Lewis C, Suffet IH, Ritz B. Estimated effects of disinfection by-products on birth weight in a population served by a single water utility. Am J Epidemiol
10. Nieuwenhuijsen MJ, Toledano MB, Eaton NE, Fawell J, Elliott P. Chlorination disinfection byproducts in water and their association with adverse reproductive outcomes: a review. Occup Environ Med
11. Porter CK, Putnam SD, Hunting KL, Riddle MR. The effect of trihalomethane and haloacetic acid exposure on fetal growth in a Maryland county. Am J Epidemiol
12. Tardiff RG, Carson ML, Ginevan ME. Updated weight of evidence for an association between adverse reproductive and developmental effects and exposure to disinfection by-products. Regul Toxicol Pharmacol
13. Toledano MB, Nieuwenhuijsen MJ, Best N, et al. Relation of trihalomethane concentrations in public water supplies to stillbirth and birth weight in three water regions in England. Environ Health Perspect
14. Wright JM, Schwartz J, Dockery DW. The effect of disinfection by-products and mutagenic activity on birth weight and gestational duration. Environ Health Perspect
15. Aschengrau A, Zierler S, Cohen A. Quality of community drinking water and the occurrence of spontaneous abortion. Arch Environ Health
16. Deane M, Swan SH, Harris JA, Epstein DM, Neutra RR. Adverse pregnancy outcomes in relation to water consumption: a re-analysis of data from the original Santa Clara County Study, California, 1980–1981. Epidemiology
17. Savitz DA, Andrews KW, Pastore LM. Drinking water and pregnancy outcome in central North Carolina: source, amount, and trihalomethane levels. Environ Health Perspect
18. Savitz DA, Singer PC, Herring AH, Hartmann KE, Weinberg HS, Makarushka C. Exposure to drinking water disinfection by-products and pregnancy loss. Am J Epidemiol
19. Swan SH, Neutra RR, Wrensch M, et al. Is drinking water related to spontaneous abortion? Reviewing the evidence from the California Department of Health Services studies. Epidemiology
20. Swan SH, Waller K, Hopkins B, et al. A prospective study of spontaneous abortion: relation to amount and source of drinking water consumed in early pregnancy [comments]. Epidemiology
21. Waller K, Swan SH, DeLorenze G, Hopkins B. Trihalomethanes in drinking water and spontaneous abortion. Epidemiology
22. Wrensch M, Swan SH, Lipscomb J, Epstein DM, Neutra RR, Fenster L. Spontaneous abortions and birth defects related to tap and bottled water use, San Jose, California, 1980–1985. Epidemiology
23. Cedergren MI, Selbing AJ, Lofman O, Kallen BA. Chlorination byproducts and nitrate in drinking water and risk for congenital cardiac defects. Environ Res
24. Dodds L, King WD. Relation between trihalomethane compounds and birth defects. Occup Environ Med
25. Hwang BF, Magnus P, Jaakkola JJ. Risk of specific birth defects in relation to chlorination and the amount of natural organic matter in the water supply. Am J Epidemiol
26. Hwang BF, Jaakkola JJ, Guo HR. Water disinfection by-products and the risk of specific birth defects: A population-based cross-sectional study in Taiwan. Environ Health
27. Klotz JB, Pyrch LA. Neural tube defects and drinking water disinfection by-products. Epidemiology
28. Magnus P, Jaakkola JJ, Skrondal A, et al. Water chlorination and birth defects. Epidemiology
29. Nieuwenhuijsen MJ, Toledano MB, Bennett J, et al. Chlorination disinfection by-products and risk of congenital anomalies in England and Wales. Environ Health Perspect
30. Shaw GM, Ranatunga D, Quach T, Neri E, Correa A, Neutra RR. Trihalomethane exposures from municipal water supplies and selected congenital malformations. Epidemiology
31. Zwiener C, Richardson SD, DeMarini DM, Grummt T, Glauner T, Frimmel FH. Drowning in disinfection byproducts? Assessing swimming pool water. Environ Sci Technol
32. Weisel CP, Richardson SD, Nemery B, Aggazzotti G, Baraldi E. Childhood asthma and environmental exposures at swimming pools: state of the science and research recommendations. Environ Health Perspect
33. Beech JA, Diaz R, Ordaz C, Palomeque B. Nitrates, chlorates and trihalomethanes in swimming pool water. Am J Public Health
34. Levesque B, Ayotte P, LeBlanc A, et al. Evaluation of dermal and respiratory chloroform exposure in humans. Environ Health Perspect
35. Levesque B, Ayotte P, Tardif R, et al. Evaluation of the health risk associated with exposure to chloroform in indoor swimming pools. J Toxicol Environ Health A
36. Lindstrom AB, Pleil JD, Berkoff DC. Alveolar breath sampling and analysis to assess trihalomethane exposures during competitive swimming training. Environ Health Perspect
37. McKone TE. Linking a PBPK model for chloroform with measured breath concentrations in showers: implications for dermal exposure models. J Expo Anal Environ Epidemiol
38. Weisel CP, Chen WJ. Exposure to chlorination by-products from hot water uses. Risk Anal
39. Weisel CP, Jo WK. Ingestion, inhalation, and dermal exposures to chloroform and trichloroethene from tap water. Environ Health Perspect
40. Weisel CP, Kim H, Haltmeier P, Klotz JB. Exposure estimates to disinfection by-products of chlorinated drinking water. Environ Health Perspect
41. Leavens TL, Blount BC, DeMarini DM, et al. Disposition of bromodichloromethane in humans following oral and dermal exposure. Toxicol Sci
42. Nieuwenhuijsen MJ, Northstone K, Golding J. Swimming and birth weight. Epidemiology
43. Juhl M, Andersen PK, Olsen J, et al. Physical exercise during pregnancy and the risk of preterm birth: a study within the Danish National Birth Cohort. Am J Epidemiol
44. Owe KM, Nystad W, Bo K. Correlates of regular exercise during pregnancy: the Norwegian mother and child cohort study. Scand J Med Sci Sports
45. Juhl M, Olsen J, Andersen PK, Nohr EA, Andersen AN. Physical exercise during pregnancy and fetal growth measures: a study within the Danish National Birth Cohort. Am J Obstet Gynecol
. 2009 Oct 2. [Epub ahead of print.]
46. Madsen M, Jørgensen T, Jensen ML, et al. Exercise during pregnancy and the risk of spontaneous abortion. Br J Obstet Gynaecol
47. Olsen J, Melbye M, Olsen SF, et al. The Danish National Birth Cohort—its background, structure and aim. Scand J Public Health
48. Nguyen RH, Wilcox AJ. Terms in reproductive and perinatal epidemiology: 2. Perinatal terms. J Epidemiol Community Health
49. Watson WJ, Katz VL, Hackney AC, Gall MM, McMurray RG. Fetal responses to maximal swimming and cycling exercise during pregnancy. Obstet Gynecol
50. Villanueva CM, Cantor KP, Grimalt JO, et al. Bladder cancer and exposure to water disinfection by-products through ingestion, bathing, showering, and swimming in pools. Am J Epidemiol