IJE Advance Access originally published online on January 13, 2005
International Journal of Epidemiology 2005 34(3):664-671; doi:10.1093/ije/dyi006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Article |
Maternal smoking in pregnancy as a determinant of rheumatoid arthritis and other inflammatory polyarthropathies during the first 7 years of life
1 Institute of Occupational and Environmental Medicine, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
2 National Research and Development Centre for Welfare and Health, Helsinki, Finland
* Corresponding author. Institute of Occupational and Environmental Medicine, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. E-mail: j.jaakkola{at}bham.ac.uk
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionConclusionReferences |
|---|
Objectives Personal smoking and gender are determinants of adult rheumatoid arthritis. We assessed the independent and joint effects of maternal smoking in pregnancy and gender on the development of rheumatoid arthritis and other inflammatory polyarthropathies (RA&IP, ICD-9 code 714) in particular juvenile rheumatoid arthritis (JRA, ICD-9 code 714.3) in the first 7 years of life in a cohort of Finnish children born in 1987.
Methods We identified 58 841 singleton births from the Finnish Medical Birth Registry and followed-up on them through other nationwide registries for 7 years. The birth registry provided categorical information on the mother's smoking during pregnancy: no smoking as a reference, low exposure (<10 cigarettes per day) and high exposure (>10 cigarettes per day).
Results There were 75 cases of RA&IP yielding an incidence rate of 18.5 per 100 000 person-years. Of these, 31 were classified as JRA with an incidence rate of 7.6 per 100 000 person-years. In logistic regression, both the risks of RA&IP (adjusted odds ratio (OR) 2.10; 95% confidence interval (CI) 1.30–3.40) and JRA (3.03; 1.36–6.76) were increased in girls. High exposure to tobacco smoke increased the risks of RA&IP (2.57; 1.13–5.89) and JRA (2.98; 0.95–8.78) in girls, but not in boys. The adjusted ORs for girls with heavy exposure were 4.64 (1.94–11.07) for RA&IP and 6.76 (2.00–22.9) for JRA compared with unexposed boys.
Conclusion This is an original finding of a potential effect of foetal exposure to tobacco smoke on the risks of RA&IP and JRA in girls.
Keywords Smoking, pregnancy, juvenile rheumatoid arthritis, rheumatoid arthritis, cohort studies
Accepted 29 November 2004
The occurrence of juvenile arthritis, including juvenile rheumatoid arthritis (JRA), and juvenile chronic arthritis, varies substantially worldwide. A recent systematic review of 34 epidemiological studies conducted in different parts of the world, reported a range of prevalences from 0.007 to 0.401% and of incidence rates from 0.8 to 22.6 per 100 000 person-years.1 There is consistent evidence that JRA is more common in girls than boys.2–4 Rheumatoid arthritis has been described to have features of organ specific autoimmune diseases, where T helper 1 (Th1) dominates immune response.5 This immune response could be stimulated by antigens, but so far specific antigens have not been identified. Tobacco smoke could be a source of these immune system stimulating antigens. Consistent with this idea, the weight of evidence from the previous 13 epidemiological studies indicates that personal smoking increases the risk of rheumatoid arthritis in adults.6–18 In some studies the effect of smoking on the risk of rheumatoid arthritis has been present only in men, but other studies have shown the effect of smoking in both genders.
We hypothesized that the effect of tobacco smoke products on the immune system could begin during the foetal period, as a result of maternal smoking in pregnancy. We also considered the potential joint effect of gender and maternal smoking. Preterm delivery and reduction in foetal growth are established effects of maternal smoking in pregnancy.19 We were also interested, to what extent the possible effect of maternal smoking on the child's inflammatory polyarthropathies would be mediated by reduction in the duration of pregnancy and foetal growth. We elaborated the relations between maternal smoking in pregnancy and the risk of rheumatoid arthritis and other inflammatory polyarthropathies RA&IP (RA&IP, ICD-9 code 714) in particular juvenile rheumatoid arthritis (JRA, ICD-9 code 714.3) in the first 7 years of life in a cohort of Finnish children born in 1987.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
Data sources and study population
The source population comprised all children born in Finland in 1987 (n = 60 254). We focused on all 58 841 singleton births and followed-up on them through registries for 7 years.20–21 Childhood health data were received from five national administrative health registers for years 1987–94.20
Information on the child's birth weight, gestational age, and maternal smoking habits during pregnancy were obtained from the Finnish Medical Birth Registry established in 1987 and run by the National Research and Development Centre for Welfare and Health. Information on maternal smoking was categorical: no smoking; <10 cigarettes per day; and >10 cigarettes per day.
Health outcomes
The outcomes of interest were RA&IP in general and JRA in particular. RA&IP was defined on the basis of at least one hospitalization, and/or at least one entitlement to free medication due to rheumatoid arthritis and/or at least one entitlement to special care support (which can be granted for families with a disabled child, or with a child who has a long-term illness needing continuous help or surveillance) diagnosed as ICD-9 code 714 before the age of 7 years. JRA was a subcategory of RA&IP with an ICD-9 code 714.3.
Covariates
The basic adjustment was made using the following core covariates: gender, birth order, maternal age, marital status, and maternal occupation. Foetal growth and preterm delivery were considered both as potential confounders and intermediate variables for the relations between maternal smoking and the two outcomes, RA&IP and JRA. We used three measures of foetal growth: low birth weight (<2500 g), high birth weight (
4000 g) and small-for-gestational-age (SGA). SGA was defined according to the Finnish population-based growth curves.22 The preterm delivery was defined as the length of gestation <37 weeks. Gestational age is practically always verified at the maternity care clinics by ultrasound examination during the 18th week of gestation.
Statistical methods
We compared the risks (cumulative incidence) of the two health outcomes, RA&IP and JRA, according to foetal exposure to tobacco smoke because of maternal smoking. Odds ratio (OR) was the measure of effect. We used logistic regression analysis to estimate adjusted ORs for the relations of interest. The basic adjustment was made using the core covariates listed above. Additional adjustment was made for low birth weight and preterm delivery when assessing to what extent the effect of maternal smoking on the two outcomes was mediated through intrauterine growth.
Second, we studied independent and joint effects of female gender and foetal exposure to tobacco smoke products on an additive scale.23,24 We compared the risks of RA&IP and JRA in six exposure categories: (i) male and no exposure (R00, reference category); (ii) female and no exposure (R10); (iii) male and low exposure (R01); (iv) male and high exposure (R02); (v) female and low exposure (R11); and (vi) female and high exposure (R12). On an additive scale, the interaction (IA) of joint effect of maternal smoking and gender was quantified by calculating the risk that is more than expected based on the independent effects of these factors. The IA for heavy smoking and gender was:
 |
We also assessed how gender alone predicts the risk of developing RA&IP and JRA. To assess the joint effect of female gender and exposure to tobacco smoke products, we calculated ORs contrasting each of the five exposure categories to the reference category. Estimates for the independent effects of female gender and maternal smoking in pregnancy and their joint effect were derived from the same logistic regression model adjusting for the covariates.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
Study population
Smoking in pregnancy was related to young age, not being married, and low education, as shown in Table 1. The follow-up rate was 99.9%, which eliminated the possibility of selection bias.
|
Determinants of RA&IP
There were 75 cases in the broad category of RA&IP (ICD-9 code 714) yielding an estimated incidence rate of 18.5 per 100 000 person-years. The subcategory of JRA (ICD-9 code 714.3) included 31 cases with an incidence rate of 7.6 per 100 000 person-years. The rest of the cases (n = 44) were diagnosed as rheumatoid arthritis (ICD-9 code 714.0). One of the cases was first diagnosed as 714.3 but later as 714.0. Girls had an increased risk of both RA&IP (25.2 per 100 000 person-years) and JRA (11.7 per 100 000 person-years) compared with boys (12.0 and 3.8 per 100 000 person-years, respectively).
The determinants of RA&IP and JRA are shown in Table 2. Female gender increased the risk of both RA&IP and JRA with adjusted ORs of 2.10 (95% confidence interval (CI) 1.30–3.40) and 3.03 (1.36–6.76), respectively. The risks of RA&IP and JRA were related to low birth weight and SGA, although the CIs were wide and included unity (Table 2). There were no RA&IP cases among premature children (Table 2). Children of single mothers had a slightly increased risk for RA&IP as well as JRA compared with those of married mothers, although the estimates were imprecise.
|
The risk of developing RA&IP during the first 7 years of life was related to heavy maternal smoking with an adjusted OR of 1.98 (0.93–4.23), as shown in Table 3. However, the effect estimate was higher in girls (2.57, 1.13–5.89), whereas no effect was seen in boys (0.74, 0.10–5.55). The corresponding effect estimate for JRA was of the same magnitude, 2.17 (0.74–6.35), and similarly it was higher in girls (2.98, 0.95–8.78).
|
Joint effect of female gender and foetal tobacco smoke product exposure
Table 4 shows risks of RA&IP and JRA in the six exposure categories. On additive scale, the excess risk of rheumatoid arthritis related to female gender was 0.0006. There was no effect related to maternal smoking in boys indicated by a negative risk difference for both levels of smoking. The effect due to interaction of heavy smoking and female gender was 0.0036. In terms of relative risks, female gender in unexposed increased the risk of RA&IP, with an adjusted OR of 1.64 (0.97–2.77). There was no effect of maternal smoking in boys, with an adjusted OR of 0.64 (0.19–2.13). The estimates for joint effects of female gender and low and high exposure were 1.65 (0.62–4.39) and 4.64 (1.94–11.07), correspondingly. The study of interaction for JRA was restricted by a small number of cases. However, the pattern was similar to the pattern for the broader outcome category, RA&IP. Female gender in unexposed increased the risk of JRA with an adjusted OR of 2.02 (0.87–4.66). The joint effect of female gender and high exposure was 6.76 (95% 2.00–22.9).
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
In our large cohort study representing approximately 409 500 person-years of disease experience, we tested an original hypothesis that maternal smoking in pregnancy increases the risk of RA&IP, in particular JRA in childhood. Consistent with our hypothesis, the risk of developing both RA&IP and JRA during the first 7 years of life was 2-fold higher in children of mothers who smoked >10 cigarettes per day compared with children of non-smoking mothers. Interestingly, the effect of maternal smoking was limited to girls. The effect of maternal smoking was slightly reduced when taking into account foetal growth and preterm delivery, indicating that only a small fraction of the effect is mediated through these pregnancy outcomes. The joint effect of female gender and heavy maternal smoking resulted in >4-fold risk increase for the broader outcome category, RA&IP, and >6-fold increase for JRA compared with unexposed boys.
Validity of results
The source population included all children registered to be born in Finland in 1987. The coverage of the Finnish Birth Registry is close to 100%.21 For the purposes of the study we focused on all the singleton births. The registry-based follow-up of RA&IP was expected to identify almost all the diagnosed cases. The National Social Insurance Institute covers all residents of Finland and provides 75% reimbursement of rheumatoid arthritis medications for those with rheumatoid arthritis fulfilling their diagnostic criteria. The reimbursement right for RA&IP (ICD-9 code 714) is indicated with a number in the Social Insurance Card. Registry information on special support and hospital discharge registration served as complementary information on cases not receiving drug treatment.
A potential source of selection bias was the lack of information on smoking in pregnancy. This information was missing only for 3.8% of the mothers, and therefore the magnitude of bias even in the worst scenario would be small.
Information on smoking in pregnancy and other relevant information were collected before the onset of the outcomes minimizing the possibility of information bias. Information on maternal smoking was unlikely to influence the diagnosis made by the physicians, because there was no knowledge about the potential hazards of maternal smoking on the child's RA&IP.
We were able to control for some potential confounders such as maternal age, parity, marital status, and occupation as well as birth weight and gestational age. Following the primary hypothesis that maternal smoking in pregnancy increases the risk of RA&IP in the child, also maternal exposure to environmental tobacco smoke (ETS) and the child's exposure to ETS could be determinants of these outcomes and thus potential confounders. The birth registry data did not include information on exposure to ETS during pregnancy and the registry-based follow-up data focusing on the health outcomes included no information on family smoking habits after the birth. Maternal smoking in pregnancy and post-natal exposure to ETS from maternal smoking are strongly correlated. Therefore the effect estimates calculated for smoking in pregnancy also include partially the potential effect of ETS during childhood. However, there is no evidence of the effects of ETS on the risk of rheumatoid arthritis and other polyarthropathies. In general there is little evidence that inhalation exposures could increase the risk of rheumatoid arthritis either in children or adults.
Synthesis with previous knowledge
This is to our knowledge the first study to show the association between foetal exposure to tobacco smoke products and the risk of rheumatoid arthritis and other polyarthropathies in general or JRA in particular in childhood. There is accumulating evidence that personal smoking increases the risk of rheumatoid arthritis in adults.6–18 There is also evidence of exposure–response relation between cumulative smoking expressed in cigarette-years and the risk of rheumatoid arthritis.10,13,15–17 A large Swedish incident case–control study showed that the effect of cigarette smoking on rheumatoid arthritis is present both in men and women, but the effect is focused on rheumatoid factor positive illness.18 It was also shown that development of disease required long duration, but merely a moderate intensity of smoking.
The estimated incidence rate of JRA was 7.6 cases per 100 000 person-years. In a previous Finnish study conducted in five central hospital districts the corresponding incidence rate was 13.5 cases per 100 000 person-years (95% CI 9.5–18.7) among 0- to 15-year olds and the median age of onset varied from 7 to 9 years.25 In the present study, girls had a 2-fold risk of RA&IP and 3-fold risk of JRA compared with boys. This is consistent with a previous estimate of 2.4 for the Finnish population aged 0–15 years.2 Similar gender difference has been reported in adult rheumatoid arthritis. The gender difference in the risk of rheumatoid arthritis has given raise to the hypothesis that sex hormones may influence the pathogenesis.13
The effect of maternal smoking was found only in girls, although the small numbers of cases is a serious limitation for the study. We used the absolute effects on an additive scale in the definition of interaction. This is justified by the idea that the public health impact of independent effects and their possible interaction follows additive rather than for example multiplicative scale.23,24
Children with low birth weight and SGA had an increased risk of rheumatoid arthritis even after adjusting for maternal smoking which is a well-established determinant of these pregnancy outcomes.19 In a recent Swedish case–control study, the risk of rheumatoid arthritis in adults was associated with both low birth weight (<3000 g) with an adjusted OR of 1.8 (0.8–4.2), but especially with high birth weight (4000 g) with an adjusted OR of 3.6 (1.4–9.1).26 In the present study, the effect estimate for high birth weight was only 1.14 (0.87–1.49).
We found that children of blue-collar workers had an increased risk of in general, as well as of JRA in particular, although the small numbers of cases resulted in wide CIs. Previous epidemiological studies have reported an increased risk of rheumatoid arthritis among farmers and farm workers,27,28 concrete and construction workers,28 machine and engine repairers,27 textile workers,29 and workers exposed to mineral dust.30 Occupation could also indicate a complex influence of socioeconomic factors as suggested by results of the recent Swedish case–control, which reported an increased risk of rheumatoid arthritis among children whose fathers were manual worker compared with children of non-manual workers.26 Our finding raises interesting questions about the role of environmental exposures during foetal period as causes of rheumatoid arthritis in childhood.
|
Conclusion |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
This is an original finding of the potential effect of foetal exposure to tobacco smoke on the risk of RA&IP and JRA in girls. Work environment in pregnancy may also influence the risk. Polyarthropathies in children are relatively rare. Although the cohort represents over 400 000 person-years of disease experience, the effect estimates are compromised in precision.
KEY MESSAGES
|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionConclusionReferences |
|---|
1 Manners PJ, Bower C. Worldwide prevalence of juvenile arthritis—why does it vary so much? J Rheumatol 2002; 29:1520–30.
2 Kaipiainen-Seppanen O, Savolainen A. Incidence of chronic juvenile rheumatic diseases in Finland during 1980–90. Clin Exp Rheumatol 1996; 14:441–44.[Web of Science][Medline]
3 Symmons DPM, Jones M, Osborne J, Sills J, Southwood TR, Woo P. Pediatric rheumatology in the United Kingdom: data from the British Pediatric Rheumatology Group National Diagnostic Register. J Rheumatol 1996; 23:1975–80.[Web of Science][Medline]
4 Von Koskull S, Truckenbrodt H, Holle R, Hormann A. Incidence and prevalence of juvenile arthritis in an urban population of southern Germany: a prospective study. Ann Rheum Dis 2001; 60:940–45.
5 Miossec P, van der Berg W. Th1/Th2cytokine balance in arthritis. Arthritis Rheum 1997; 40:2105–15.[Web of Science][Medline]
6 Vessey MP, Villard-Mackintosh L, Yeates D. Oral contraceptives, cigarette smoking and other factors in relation to arthritis. Contraception 1987; 35:457–64.[CrossRef][Web of Science][Medline]
7 Hazes JMW, Dijkmans BA, Vandenbroucke JP, de Vriea RR, Cats A. Life style and the risk of rheumatoid arthritis: cigarette smoking and alcohol consumption. Ann Rheum Dis 1990; 49:980–82.
8 Hernandez Avila M, Liang MH, Willett WC et al. Reproductive factors, smoking, and the risk for rheumatoid arthritis. Epidemiology 1990; 1:285–91.[Medline]
9 Heliovaara M, Aho K, Aromaa A, Knekt P, Reunanen A. Smoking and risk of rheumatoid arthritis. J Rheumatol 1993; 20:1830–35.[Web of Science][Medline]
10 Voigt LF, Koepsell TD, Nelson JL, Dugowson CE, Daling JR. Smoking, obesity, alcohol consumption, and the risk of rheumatoid arthritis. Epidemiology 1994; 5:525–32.[Web of Science][Medline]
11 Silman AJ, Newman J, MacGregor AJ. Cigarette smoking increases the risk of rheumatoid arthritis: results from a nationwide study of disease-discordant twins. Arthritis Rheum 1996; 39:732–35.[Web of Science][Medline]
12 Symmons DPM, Bankhead CR, Harrison BJ et al. Blood transfusion, smoking, and obesity as risk factors for the development of rheumatoid arthritis: results from a primary care-based incident case-control study in Norfolk, England. Arthritis Rheum 1997; 40:1955–61.[Web of Science][Medline]
13 Karlson EW, Lee IM, Cook NR, Manson JE, Buring JE, Hennekns CH. A retrospective cohort study of cigarette smoking and risk of rheumatoid arthritis in female health professionals. Arthritis Rheum 1999; 42:910–17.[CrossRef][Web of Science][Medline]
14 Uhlig T, Hagen KB, Kvien TK. Current tobacco smoking, formal education, and the risk of rheumatoid arthritis. J Rheumatol 1999; 42:910–17
15 Hutchinson D, Shepstone L, Moots R, Lear JT, Lynch JT, Lynch MP. Heavy cigarette smoking is strongly associated with rheumatoid arthritis (RA), particularly in patients without a family history of RA. Ann Rheum Dis 2001; 60:223–27.
16 Reckner Olsson A, Skogh T, Wingren G. Comorbidity and lifestyle, reproductive factors, and environmental exposures associated with rheumatoid arthritis. Ann Rheum Dis 2001; 60:934–39.
17 Criswell LA, Merlino LA, Cerhan JR et al. Cigarette smoking and the risk of rheumatoid arthritis among post menopausal women: results from the Iowa Women's Health Study. Am J Med 2002; 112:465–71.[CrossRef][Web of Science][Medline]
18 Stolt P, Bengtsson C, Nordmark B et al. Quantification of the influence of cigarette smoking on rheumatoid arthritis: results from a population based case-control study, using incident cases. Ann Rheum Dis 2003; 62:835–41.
19 Kramer MS. Determinants of low birth weight: methodological assessment and meta-analysis. Bull World Health Organ 1987; 65:663–737.[Web of Science][Medline]
20 Gissler M, Hemminki E, Louhiala P, Järvelin M. Health registers are feasible for measuring health status in childhood—a seven-year follow-up of the 1987 Finnish birth cohort. The use of register information in measuring health status in childhood. Paediatr Perinat Epidemiol 1998; 12:437–55.[CrossRef][Web of Science][Medline]
21 Gissler M, Teperi J, Hemminki E, Meriläinen J. Data quality after restructuring a nationwide medical birth registry. Scand J Soc Med 1995; 23:75–80.[Web of Science][Medline]
22 Pihkala J, Hakala T, Voutilainen P, Raivio K. Uudet suomalaiset sikiön kasvukäyrät [Characteristics of recent fetal growth curves in Finland] (in Finnish). Duodecim 1989; 105:1540–46.[Medline]
23 Rothman KJ. Modern epidemiology. Boston/Toronto: Little, Brown and Company, 1985, pp. 311–26.
24 Greenland S, Rothman KJ. Concepts of interaction. In: Rothman KJ, Greenland S (eds.). Modern Epidemiology. 2nd edn. Philadelphia: Lippincott-Raven, 1998, pp. 329–42.
25 Kaipiainen-Seppanen O, Savolainen A. Changes in the incidence of juvenile rheumatoid arthritis in Finland. Rheumatology 2001; 40:928–32.
26 Jacobsson LTH, Jacobsson ME, Askling J, Knowler WC. Perinatal characteristics and risk of rheumatoid arthritis. BMJ 2003; 326:1068–69.
27 Hellgren L. The prevalence of rheumatoid arthritis in occupational groups. Acta Rheumatol Scand 1970; 16:106–13.[Medline]
28 Lundberg I, Alfredsson L, Plato N, Svedrup B, Klareskog L, Kleinau S. Occupation, occupational exposure to chemicals and rheumatological disease. Scand J Rheumatol 1994; 23:305–10.[Web of Science][Medline]
29 Koskela RS, Klockars M, Jarvinen E. Mortality and disability among cotton mill workers. Br J Ind Med 1990; 47:384–91.[Web of Science][Medline]
30 Klockars M, Koskela RS, Jarvinen E, Kolari PJ, Rossi A. Silica exposure and rheumatoid arthritis: a follow-up study of granite workers 1940–81. BMJ 1987; 294:997–1000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C Carlens, L Jacobsson, L Brandt, S Cnattingius, O Stephansson, and J Askling Perinatal characteristics, early life infections and later risk of rheumatoid arthritis and juvenile idiopathic arthritis Ann Rheum Dis, July 1, 2009; 68(7): 1159 - 1164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Simard, K. Costenbader, M. Liang, E. Karlson, and M. Mittleman Exposure to maternal smoking and incident SLE in a prospective cohort study Lupus, April 1, 2009; 18(5): 431 - 435. [Abstract] [PDF]
|
|
|
|
|
|
K. A Young, L. A Parrish, G. O Zerbe, M. Rewers, K. D Deane, V Michael Holers, and J. M Norris Perinatal and early childhood risk factors associated with rheumatoid factor positivity in a healthy paediatric population Ann Rheum Dis, February 1, 2007; 66(2): 179 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Lucina Arch. Dis. Child., November 1, 2005; 90(11): 1210 - 1210. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Symmons Commentary: Juvenile idiopathic arthritis--issues of definition and causation Int. J. Epidemiol., June 1, 2005; 34(3): 671 - 672. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||