Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Association between perinatal depression in mothers and the risk of childhood infections in offspring: a population-based cohort study

Association between perinatal depression in mothers and the risk of childhood infections in... Background: Previous studies have suggested that children of mothers who experience depression during the perinatal period may have more infections, but such studies are few in number and none have been carried out in the United Kingdom (UK) population. The aim of this study was to investigate the association between perinatal depression in mothers and the risk of childhood infections in offspring in the UK general population. Methods: We used data from The Health Improvement Network (THIN), a large database of electronic primary care medical records to conduct a cohort study among all first-born singleton children born and enrolled in THIN between 1988 and 2004. We used Poisson regression to compare the incidence of gastrointestinal infections and lower respiratory tract infections reported between birth and age 4 years among children of mothers with a record of perinatal depression with those born to mothers with no such history. Results: Children of mothers with perinatal depression had a 40% increased risk of gastrointestinal infections and a 27% increased risk of lower respiratory tract infections compared with children of mothers without perinatal depression (incidence rate ratios = 1.40 and 1.27; 95% confidence intervals 1.37-1.42 and 1.22-1.32, respectively). On restricting to antibiotic-treated infections there was a slight increase in the magnitude of association with gastrointestinal infections but a decrease in that with lower respiratory tract infections (incidence rate ratios = 1.47 and 1.19; 95% confidence intervals 1.34-1.61 and 1.11-1.27, respectively). Conclusions: Maternal perinatal depression is associated with increased rates of childhood gastrointestinal infections, particularly more severe infections, and lower respiratory tract infections in the UK. Preventing maternal perinatal depression may avoid substantial morbidity among offspring, although further work is also needed to investigate the detailed reasons for these findings. Background is some evidence to suggest that a dysregulated hapotha- According to recent guidance from the National Insti- lamic pituitary adrenocortical axis may lead to immuno- tute for Health and Clinical Excellence [1], depression supression in mothers with perinatal depression [6,9], in mothers during and after pregnancy (maternal perina- which may directly impair their children’simmuneand tal depression) is one of the most important issues for neuroendocrine development. women’s health in the United Kingdom (UK). Depres- Researchers in Pakistan studying 265 mothers and sion is estimated to affect 10% to 20% of women during their children found an increased risk of diarrhoea the perinatal period in high-income countries [2-4] and among children born to mothers with postnatal depres- can have a substantial adverse impact on women ’s sion compared with those born to mothers without health as well as on maternal behaviour towards their postnatal depression (adjusted odds ratio [OR] = 3.1; children and childcare practices [5-8]. In addition, there 95% confidence interval [CI] 1.8-5.1) [10]. They also reported an unadjusted result for respiratory infections * Correspondence: mcxlb4@nottingham.ac.uk in children born to mothers with either antenatal or † Contributed equally postnatal depression (OR = 1.1; 95% CI 0.9-1.4) in the Department of Epidemiology & Public Health, University of Nottingham, same cohort [11]. Moreover, a cross-sectional study Nottingham, UK © 2010 Ban et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Ban et al. BMC Public Health 2010, 10:799 Page 2 of 8 http://www.biomedcentral.com/1471-2458/10/799 conducted in the United States of America reported an were excluded. Follow-up data were available for all association between postpartum depression and physical children for the first four years of their life unless they illness (including, but not limited to, infections) in chil- died or ceased to be registered at the practice. All ana- dren (p = 0.03) [9], suggesting that an association lyses were restricted to the first four-years of life since between maternal perinatal depression and childhood in the UK children generally start attending school at infections may also exist in high-income countries. between 4 and 5 years of age, leading to a considerable Since the prevalence of maternal perinatal depression change in their environmental exposure to infection. is high, any association with childhood infections could We extracted data including records of GI infections, LRTIs and antibiotic prescriptions for children and be an important public health concern in high-income countries. Although the relative impact of infections on records of maternal perinatal depression and antidepres- childhood survival and severe morbidity is lesser than in sant use for mothers. Based on previous research low income countries, infectious illnesses are still a [9,10,17,18], we also extracted data (where available) on major cause of mortality and morbidity among young maternal age at birth, body mass index [BMI] before children in higher income countries and are both pregnancy, smoking status before childbirth, caesarean socially and economically burdensome [12,13]. In section delivery, and household socioeconomic status European countries in 2004, more than 16% of deaths in (represented by quintiles of Townsend deprivation children under five-years of age were attributed to child- indices [19]), and the sex of the child, since these factors hood acute respiratory infections (ARIs) and about 14% may be related to maternal perinatal depression as well. were attributed to childhood diarrhoeal illness [12]. We therefore conducted a large population-based Measurement of maternal perinatal depression study in the UK with the primary aim of examining the Mothers were defined as having perinatal depression if, potential effect of maternal perinatal depression on at any time during pregnancy or in the first six months childhood gastrointestinal (GI) infections and lower after childbirth, they had a medical record of diagnosed respiratory tract infections (LRTIs) in the first four years depression (e.g. major depression, mild depression, or of life. We also investigated whether healthcare-seeking chronic depression), or if they received at least one pre- behaviour differed between mothers with and without scription for an antidepressant medication. perinatal depression, as this may lead to differential levels of diagnosis of minor self-limiting infections. Measurement of childhood infections and antibiotic prescriptions Methods Children were defined as having GI infections if they Dataset and study population had at least one clinical diagnosis of GI infection (e.g. We used data from The Health Improvement Network infectious colitis, enteritis or gastroenteritis) or infec- (THIN), a large database of anonymised computerised tion-related symptoms (e.g. vomiting or diarrhoea) primary health care records of patients throughout the recorded in their general practice records. The date of th UK. There is a high standard of validated recording of recovery from illness was defined as the 15 day after medical diagnoses, medical conditions and symptoms diagnosis. If a subsequent diagnosis was recorded prior and prescriptions in THIN [14,15]. The data in this to the recovery date, this was considered part of the study derive from 255 general practices across England same episode of illness and the recovery date was th and Wales with longitudinal records of 3.9 million extended to the 8 day after this recording (only if patients. Previous research [15,16] has shown that the later than the original recovery date, ensuring that an population in THIN is generally representative of the episode lasted at least 14 days). Further recordings UK population. Ethical approval for this study was prior to the recovery date were treated in the same granted by the South East Research Ethics Committee. way. Recordings after the date of recovery were treated For each woman who was of childbearing age (15 to as indicating the onset of new episodes of illness. We 50 years of age) at any time between January 1988 and used the same process to identify and classify episodes November 2004, birth-related entries in their general of LRTIs,wherethese were definedasatleastone practice records during this period were linked to regis- clinical diagnosis of an LRTI (e.g. pneumonia or bron- tered children born in the same household at the time chitis) in the primary care record. To obtain a measure of delivery. Women were included if they had one or of more severe episodes of illness, we also identified all more live born children recorded in the database. For antibiotic-treated GI and LRTI infectious episodes, this study, we only included the first recorded singleton which we defined as those where children had one or pregnancy ending in a live born child for each woman. more records of a prescription for an antibiotic dated Pregnancies ending in stillbirth, miscarriage, therapeutic on or after the date of onset and prior to the date of abortion and multiple pregnancies (e.g. twins delivered) recovery. Ban et al. BMC Public Health 2010, 10:799 Page 3 of 8 http://www.biomedcentral.com/1471-2458/10/799 Statistical analysis had perinatal depression. The mean age of the mothers We calculated crude incidence rates as the total number was 29 years (Standard Deviation 5.7). Complete follow- of episodes for each infectious outcome (episodes of GI up data to age 4 years was available for 80.5% of all chil- infections with and without antibiotic treatment and dren, and 96.0% of children were followed up to at least episodes of LRTIs with and without antibiotic treat- 2 years of age. ment) divided by children’s total follow-up period dur- Table 1 shows that compared with mothers without ing the first four years of life. We calculated the perinatal depression, mothers with perinatal depression absolute differences in crude incidence rates by sub- were slightly younger (p < 0.001). They were also more tracting the rates in children of mothers without perina- likely to be current smokers, obese, in a household with tal depression from those in children of mothers with higher material deprivation and to have had a caesarean perinatal depression. To assess whether the incidence section delivery than mothers without perinatal depres- rates changed over the first years of the child’s life and sion (p < 0.001 for all associations). whether the association of maternal perinatal depression with childhood infections changed with increasing time Maternal depression and childhood infections from birth, we also calculated the incidence rates for Proportionately more children born to mothers with each year of the child’s life separately. perinatal depression had at least one GI infection To estimate the effect of maternal perinatal depression (60.6%) than children born to mothers without perinatal on the risk of childhood infections, we used Poisson depression (49.2%). Similarly, children born to mothers regression to obtain incidence rate ratios (IRRs) for each with perinatal depression were morelikelytohaveone type of infectious outcome among children of mothers or more LRTIs than children born to mothers without with perinatal depression relative to children of mothers perinatal depression (20.9% vs. 17.0%). When only epi- without perinatal depression, overall and for each year sodes of childhood infections with at least one antibiotic of age up to 4 years. Findings were considered statisti- prescription were considered, the prevalence decreased cally significant where p was less than 0.05. to 5.2% and 3.6% for GI infections and 10.2% and 8.7% We also attempted to assess whether mothers with for LRTIs, respectively. perinatal depression might be more likely to report The overall rates of childhood GI infections were 3.41 childhood illness to their general practitioners (GPs), or per 10 person-years (95% CI 3.35-3.47) for children of to do so sooner than mothers without perinatal depres- mothers with perinatal depression and 2.44 per 10 per- sion, by examining whether time to first polio immuni- son-years (95% CI 2.43-2.46) for children of mothers sation in children differed between mothers with and without perinatal depression (Table 2). The overall rates without perinatal depression as a potential marker of for childhood LRTIs were 0.91 per 10 person-years (95% mothers’ healthcare-seeking behaviour. We determined CI 0.88-0.94) and 0.72 per 10 person-years (95% CI the date of each child ’s first polio immunisation as 0.71-0.73), respectively for mothers with and without recorded in the general practice records for children perinatal depression (Table 2). The overall rate differ- st st born between 1 January 1991 and 31 December 2003 ences were therefore 0.97 per 10 person-years (95% CI only, since the UK immunisation schedule and vaccine 0.90-1.03) for GI infections and 0.19 per 10 person- type did not change during this time period [20,21]. We years (95% CI 0.16-0.23) for LRTIs (Table 2). calculated a median time to first polio vaccination in In addition, the rates of childhood infections in chil- children of mothers with and without perinatal depres- dren of mothers with perinatal depression were consis- sion. All analyses were carried out using Stata SE 10.0 tently higher than those of mothers without perinatal software. depression through the first four years of life, generally decreasing gradually in both cohorts for both GI infec- Statistical Power tions and LRTIs as the children got older (Table 2). Based on previous studies [2-4,22], we estimated that In our study, children born to women with perinatal about 10% of mothers in our study population would depression were 40% more likely to have GI infections have perinatal depression. We calculated that our sam- and 27% more likely to have LRTIs compared with ple size would give us over 99% power to detect a 1.5- those born to women without perinatal depression (IRR fold increased rate in children of mothers with perinatal = 1.40; 95% CI 1.37-1.42 and IRR = 1.27; 95% CI 1.22- depression for both GI infections and LRTIs. 1.32, respectively). The observed effects were similar for each year within the first four years of the child’s life Results (Table 2). Furthermore, for more severe infectious epi- Study population sodes that were treated with antibiotics (Table 3), the We identified a cohort of 107,587 mothers and their effect of maternal perinatal depression on childhood first live born children. Of the mothers, 9,722 (9.1%) infections increased slightly for GI infections and Ban et al. BMC Public Health 2010, 10:799 Page 4 of 8 http://www.biomedcentral.com/1471-2458/10/799 Table 1 Characteristics of mothers with and without depression (N = 107587) Maternal perinatal depression* 2 † No (n = 97,815) Yes (n = 9,772) c (df)P N (%) N (%) Maternal age at birth of child (years) Mean (Standard deviation) 29.2 (5.6) 28.6 (5.9) t = 9.5 (107585) < 0.001 Maternal body mass index (kg/m ) Normal (18.5-24.9) 37971 (38.8) 4069 (41.6) 575.9 (4) < 0.001 Underweight (<18.5) 2660 (2.7) 395 (4.0) Overweight (25-29.9) 12093 (12.4) 1502 (15.4) Obese (>30) 5653 (5.8) 910 (9.3) Missing 39438 (40.3) 2896 (29.6) Maternal smoking Non-smoker 47266 (48.3) 3952 (40.4) 1300 (3) < 0.001 Ex-smoker 4698 (4.8) 567 (5.8) Current smoker 19985 (20.4) 3463 (35.4) Missing 25866 (26.4) 1790 (18.3) Household socioeconomic status 1 (Least deprivation) 16028 (16.4) 1112 (11.4) 419.1 (5) < 0.001 2 11714 (12.0) 972 (10.0) 3 11170 (11.4) 1106 (11.3) 4 9298 (9.5) 1214 (12.4) 5 (Most deprivation) 7063 (7.2) 1078 (11.0) Missing 42542 (43.5) 4290 (43.9) Caesarean section delivery 13604 (13.9) 1616 (16.5) 50.6 (1) < 0.001 Sex of the child Male 50192 (51.3) 4922 (50.4) 3.2 (1) 0.075 * Mothers with either depression diagnosis and/or at least one antidepressant prescription either during pregnancy and/or in the first six months after the child’s delivery. Degrees of freedom. Student t-test statistic. Table 2 Associations between maternal perinatal depression and childhood gastrointestinal infections and childhood lower respiratory infections Maternal perinatal depression No Yes Childhood infections Events Rate per 10 person-years Events Rate per 10 person-years Rate difference Unadjusted IRR (95% CI*) (95% CI*) (95% CI*) (95% CI*) GI infections Age (years) 0-4 84999 2.44 (2.43-2.46) 11634 3.41 (3.35-3.47) 0.97 (0.90-1.03) 1.40 (1.37-1.42) 0-1 34928 3.71 (3.67-3.75) 4950 5.31 (5.16-5.46) 1.60 (1.45-1.75) 1.43 (1.39-1.48) 1-2 26845 3.02 (2.98-3.06) 3533 4.06 (3.93-4.19) 1.04 (0.90-1.18) 1.34 (1.30-1.39) 2-3 14289 1.69 (1.66-1.72) 1981 2.40 (2.30-2.51) 0.71 (0.60-0.82) 1.42 (1.35-1.49) 3-4 8937 1.11 (1.09-1.13) 1170 1.49 (1.41-1.58) 0.38 (0.30-0.47) 1.35 (1.27-1.43) LRTIs Age (years) 0-4 25143 0.72 (0.71-0.73) 3139 0.91 (0.88-0.94) 0.19 (0.16-0.23) 1.27 (1.22-1.32) 0-1 12202 1.28 (1.26-1.31) 1632 1.72 (1.64-1.81) 0.44 (0.36-0.53) 1.35 (1.28-1.42) 1-2 6530 0.73 (0.71-0.75) 725 0.82 (0.76-0.88) 0.09 (0.03-0.16) 1.13 (1.05-1.22) 2-3 3783 0.45 (0.43-0.46) 454 0.55 (0.50-0.60) 0.10 (0.05-0.15) 1.23 (1.11-1.35) 3-4 2628 0.33 (0.31-0.34) 328 0.42 (0.37-0.46) 0.09 (0.04-0.14) 1.28 (1.14-1.44) * Confidence interval. Incidence rate ratio. Gastrointestinal infections. Lower respiratory tract infections. Ban et al. BMC Public Health 2010, 10:799 Page 5 of 8 http://www.biomedcentral.com/1471-2458/10/799 Table 3 Associations between maternal perinatal depression and childhood infections with antibiotic treatment Maternal perinatal depression No Yes Antibiotic-treated Events Rate per 10 Events Rate per 10 Rate Unadjusted childhood infections person-years person-years difference IRR (95% CI*) (95% CI*) (95% CI*) (95% CI*) GI infections Age (years) 0-4 3550 0.10 (0.10-0.10) 512 0.15 (0.14-0.16) 0.05 (0.03-0.06) 1.47 (1.34-1.61) 0-1 1144 0.12 (0.11-0.13) 176 0.18 (0.16-0.21) 0.07 (0.04-0.09) 1.54 (1.32-1.81) 1-2 1153 0.13 (0.12-0.14) 172 0.19 (0.17-0.23) 0.07 (0.04-0.10) 1.52 (1.29-1.78) 2-3 731 0.09 (0.08-0.09) 104 0.12 (0.10-0.15) 0.04 (0.01-0.06) 1.45 (1.18-1.78) 3-4 522 0.06 (0.06-0.07) 60 0.08 (0.06-0.10) 0.01 (0.00-0.03) 1.18 (0.90-1.54) LRTIs Age (years) 0-4 8868 0.25 (0.25-0.26) 1035 0.30 (0.28-0.32) 0.05 (0.03-0.07) 1.19 (1.11-1.27) 0-1 3462 0.36 (0.35-0.37) 418 0.44 (0.40-0.48) 0.07 (0.03-0.12) 1.21 (1.10-1.34) 1-2 2536 0.28 (0.27-0.29) 276 0.31 (0.28-0.35) 0.03 (0.00-0.07) 1.11 (0.98-1.25) 2-3 1661 0.20 (0.19-0.21) 195 0.23 (0.20-0.27) 0.04 (0.00-0.07) 1.20 (1.03-1.39) 3-4 1209 0.15 (0.14-0.16) 146 0.19 (0.16-0.22) 0.04 (0.00-0.07) 1.24 (1.04-1.47) * Confidence interval. Incidence rate ratio. Gastrointestinal infections. Lower respiratory tract infections. decreased slightly for LRTIs (IRR = 1.47; 95% CI 1.34- observed absolute differences in rates suggest that over 1.61 and IRR = 1.19; 95% CI 1.11-1.27, respectively). a one year period there were about 97 more GI infec- tions and 19 more LRTIs among every 1000 children Maternal perinatal depression and the time to first polio born to women with perinatal depression than among vaccination in children 1000 children born to women without perinatal depres- st We identified 92,218 children born between 1 January sion. Moreover, the near identical timing of the first st 1991 and 31 December 2003 who accounted for 85.7% polio vaccination among both groups provides no evi- of the whole cohort and who were representative of the dence to support the possibility that higher rates of GI whole cohort as they had almost identical distributions infections and LRTIs in children of mothers with peri- of maternal characteristics (data not shown). Overall, natal depression may be explained by differential levels 98% of children born to women with perinatal depres- of diagnosis due to increased healthcare-seeking beha- sion had their first polio vaccination, as did the same vior among this group. proportion of children born to women without perinatal depression. The median time of receipt of the first polio Strengths and limitations vaccination was 61 days after birth for children of We have conducted a large population-based cohort mothers with perinatal depression (interquartile range study of the association between maternal perinatal [IQR] 58-68) and this was almost identical for children depression and childhood infections using primary care of mothers without perinatal depression (median = 61 data in theUK. Ourconsiderablesample sizemeans days; IQR 57-67). that our findings are unlikely to be due to chance. The data we used were obtained from a national database Discussion and prospectively recorded by GPs, thus excluding the Principal findings possibility of recording or recall bias in both our expo- Our results show that children born to women with sure and outcome. perinatal depression had a 40% increased rate of GI Full 4-year follow-up data were not available for 19.5% infections and a 27% increased rate of LRTIs compared of children. Such drop-outs occur due to death or with children born to women without perinatal depres- patients deregistering from GP practices. There was no sion. For more severe, antibiotic-treated infections, the evidence of an increased risk of death in either cohort. observed effects increased slightly for GI infections to In the UK registration with a general practitioner is con- 47% but decreased slightly for LRTIs to 19%. Our tinuous unless a patient actively requests removal from Ban et al. BMC Public Health 2010, 10:799 Page 6 of 8 http://www.biomedcentral.com/1471-2458/10/799 the practice list, or makes an application to join another mothers with perinatal depression therefore could be an practice (typically due to a change of address, or the overestimate due to biased ascertainment. The first opening of a new practice). Drop-out rates were similar childhood polio vaccination is a routine appointment in both cohorts and we feel they are unlikely to have common to both cohorts of children and mothers and had a substantial impact on our results. one might expect the timing to be earliest and uptake Although the inclusion of postpartum depression in the greatest amongst children of mothers with high levels of healthcare-seeking behaviour. Although timing the exposure definition meant that there could be and uptake of polio vaccination were all but identical in potential for reverse causation due to some overlap both cohortsweacceptthatthisisonlyanindirect, between postpartum depression and childhood infec- tions in the first six months of a child’s life, such over- measure of mothers’ health seeking behaviour and we lap could not affect the observations at age 1, 2, and cannot conclusively exclude the possibility of biased 3 years. As we found that the increased risk of child- ascertainment. hood infections in mothers with perinatal depression Although we considered other factors related to was consistent throughout the first four years of the maternal perinatal depression in our analysis, the child’s life, we believe it is unlikely that our results were records of a substantial number of individuals did not due to reverse causality. provide information about factors such as household There was a legitimate weakness in our exposure mea- socioeconomic status (44% missing), maternal smoking surement as a certain proportion of mothers with peri- (26.4%) and maternal BMI (40% missing). We were also natal depression are neither diagnosed nor treated in unable to adjust for some other factors, such as the per- primary care. The prevalence of diagnosed maternal iod of breastfeeding and the child’s birth weight which perinatal depression found in our study (about 9.1%) have previously been linked to childhood infections was in general 4-6% lower than some previous studies [13,24,25] in low-income countries, because birth weight conducted in high-income countries [2-4,23]. However, and breastfeeding status are seldom recorded in THIN. these studies were based in research settings where It could be argued, however, that these factors may selected populations were screened for depression form part of a causal pathway linking maternal perinatal mostly using self-administered questionnaires and their depression to childhood infections. After adjusting for samples were much smaller than ours. On restricting to these factors, one previous study [10] (carried out in diagnosed depression using a standardized interviewing Pakistan) still found a statistically significant association between maternal postpartum depression and childhood schedule, two previous studies (including a GI infections, suggesting that incomplete adjustment for meta-analysis) [4,22] found fairly similar prevalence figures to ours. Although we acknowledge that existing confounding is unlikely to completely explain the obser- depression does remain undiagnosed in the population, vations of our study. during pregnancy and in the first few months after delivery, women typically havemorefrequentcontact Interpretation in context of previous studies with health professionals than they would usually do Although to our knowledge there are only three pre- because of antenatal/postnatal check-ups. This would vious studies [9-11] examining the potential association tend to reduce under-diagnosis of depression this per- between maternal perinatal depression and childhood iod. Nevertheless, the presence of under-diagnosis in infections, all have found an increased risk of childhood this population would only result in an underestimation infections in children born to women with perinatal of the risk of childhood GI infections and LRTIs in chil- depression. For example, in a recent American cross- dren associated with perinatal depression, since such sectional study of 194 mothers and their children at 4-6 misclassification would produce a bias towards the null- weeks postpartum, Groer and Morgan found that chil- hypothesis (of no association between perinatal depres- dren of mothers with postpartum depression experi- sion in mothers and childhood infections in offspring). enced more physical illness (including diarrhoeal illness) We defined childhood infections as children having since birth compared with children of mothers without been diagnosed with GI infections or LRTIs in primary postpartum depression (p = 0.03) [9]. This study, how- care, which primarily relied on mothers bringing their ever, did not provide any measure of effect and did not children to see their GPs, so they most likely repre- adjust for potential confounding factors. sented the more severe end of the spectrum of these In our study, we combined the period of antenatal and infections. We supposed that, if women with perinatal postnatal depression together and therefore we did not depression felt more anxious or less capable to take care examine whether the increased risk of childhood infec- of their children, they may take children to see their tions was independently associated with antenatal GPs more often than women without perinatal depres- depression or early postnatal depression. Rahman et al sion and the increased risk of childhood infections in carried out a similar cohort study [11] of 265 mothers Ban et al. BMC Public Health 2010, 10:799 Page 7 of 8 http://www.biomedcentral.com/1471-2458/10/799 (130 mothers with perinatal depression) in Pakistan and more acute respiratory infections than those who are found an increased risk of diarrhoea and a slightly not [13]. Likewise, children of mothers with perinatal increased, but not statistically significant, risk of ARIs in depression have been found to be less well nourished children of mothers with antenatal or postnatal depres- than those of mothers without depression [27,28], which sion compared with children of mothers without depres- will increase susceptibility to infection. sion (OR = 2.4; 95% CI 1.7-3.3 and OR = 1.1; 95% CI In addition, postpartum depression in mothers can lead to child neglect and poorer childcare practices, 0.9-1.4, respectively), although they did not adjust for especially in high-income countries [5,29,30]. Children any confounding factors. A subsequent study by the may thus be more exposed to unsanitary and dangerous same authors [10] which focused only on the relation- ship between postnatal depression and childhood diar- environments, increasing the potential for exposure to rhoeal illness showed a similar result to the first, sources of infection. although the risk of diarrhoea in children of mothers with postnatal depression compared with children of Conclusions mothers without postnatal depression was much stron- Our findings show that depression in mothers during ger once adjusted for important maternal and child fac- and after pregnancy is associated with an increased risk tors (unadjusted OR = 2.3; 95% CI 1.6-3.1 and adjusted of childhood infections in offspring. If causally related, OR = 3.1; 95% CI 1.8-5.6). The latter study used data our findings suggest that a substantial social and eco- from the same cohort of women and only about 5% of nomic burden of childhood morbidity could be avoided women with antenatal depression did not also have through effective prevention of antenatal and postnatal postnatal depression [10]. depression [1], although further work is required to Compared with our study, Rahman et al found a lar- investigate the detailed reasons for these findings. In pri- ger effect on GI infections but a slightly smaller one on mary care, physicians and other healthcare professionals respiratory infections. Their study population was from should be aware of the potential increased presentation a rural area of Pakistan where childhood infections were of childhood infections in children of mothers with peri- much more common than in our study population. natal depression. Furthermore, they used only binary outcomes of more or fewer than five childhood diarrhoeal episodes and Abbreviations more or fewer than six ARI episodes per year, while our 95% CI: 95% confidence interval; ARIs: acute respiratory infections; BMI: body outcomes were incidence rates of individual episodes of mass index; GI infections: gastrointestinal infections; GP: general practitioner; HR: hazard ratio; IQR: interquartile range; IRR: incidence rate ratio; LRTIs: GI infections and LRTIs. Consequently, our results are lower respiratory tract infections; OR: odds ratio; THIN: The Health not directly comparable. In spite of different study Improvement Network; UK: United Kingdom. populations and different measures of outcome, we both Acknowledgements found an increased risk of GI infections (and also a This study was founded through the Division of Epidemiology & Public similar magnitude of respiratory infection risk) in chil- Health of the University of Nottingham. dren born to women with perinatal depression com- Authors’ contributions pared with those born to women without perinatal LB carried out the literature review and participated in the development of depression. the study design, data management, data analysis, interpretation of the data, There could be several potential explanations for the writing the article and revising article drafts for publication. JEG had the initial idea for the study and its design and participated in the development association between perinatal depression in mothers and of the study design, data management, interpretation of the data, and childhood infections in offspring. First of all, previous revising article drafts for publication. JW participated in the development of research [6,9] has found that mothers with depression the study design, interpretation of the data, and revising article drafts for publication. LJT had the initial idea for the study and its design and during the perinatal period have a dysregulated hypotha- participated in the development of the study design, data management, lamic pituitary adrenocotical axis resulting in decreased interpretation of the data, and revising article drafts for publication. All levels of salivary cortisol and potentially depressed cellu- authors read and approved the final manuscript. lar immunity. This may have a direct impact on the Competing interests child’s immune and neuroendocrine development. Sec- The authors declare that they have no competing interests. ondly, perinatal depression can have a considerable Received: 13 April 2010 Accepted: 31 December 2010 impact on mothers’ childcare abilities. Previous studies Published: 31 December 2010 [9,26] reported that mothers with postpartum depres- sion were less likely to breastfeed their children and References more likely to give up exclusive breastfeeding than 1. National Institute for Health and Clinical Excellence: Antenatal and postnatal mental health (Clinical management and service guidance) London; 2007. mothers without depression. Breastfeeding can 2. Gavin NI, Gaynes BN, Lohr KN, Meltzer-Brody S, Gartlehner G, Swinson T: strengthen the immune system of the child and it has Perinatal depression: a systematic review of prevalence and incidence. been shown that children who are not breastfed have Obstet Gynecol 2005, 106(5 Pt 1):1071-1083. Ban et al. BMC Public Health 2010, 10:799 Page 8 of 8 http://www.biomedcentral.com/1471-2458/10/799 3. Cooper PJ, Murray L, Hooper R, West A: The development and validation 29. Beck CT: The effects of postpartum depression on maternal-infant of a predictive index for postpartum depression. Psychol Med 1996, interaction: a meta-analysis. Nurs Res 1995, 44(5):298-304. 26(3):627-634. 30. Windham AM, Rosenberg L, Fuddy L, McFarlane E, Sia C, Duggan AK: Risk 4. O’hara MW, Swain AM: Rates and risk of postpartum depression–a meta- of mother-reported child abuse in the first 3 years of life. Child Abuse & analysis. International Review of Psychiatry 1996, 8(1):37-54. Neglect 2004, 28(6):647-669. 5. McLennan JD, Kotelchuck M: Parental prevention practices for young children in the context of maternal depression. Pediatrics 2000, Pre-publication history 105(5):1090-1095. The pre-publication history for this paper can be accessed here: 6. Groër M, Davis M, Casey K, Short B, Smith K, Groër S: Neuroendocrine and http://www.biomedcentral.com/1471-2458/10/799/prepub immune relationships in postpartum fatigue. MCN Am J Matern Child Nurs doi:10.1186/1471-2458-10-799 2005, 30(2):133-138. Cite this article as: Ban et al.: Association between perinatal depression 7. Turner C, Boyle F, O’Rourke P: Mothers’ health post-partum and their in mothers and the risk of childhood infections in offspring: a patterns of seeking vaccination for their infants. Int J Nurs Pract 2003, population-based cohort study. BMC Public Health 2010 10:799. 9(2):120-126. 8. Bartlett SJ, Krishnan JA, Riekert KA, Butz AM, Malveaux FJ, Rand CS: Maternal depressive symptoms and adherence to therapy in inner-city children with asthma. Pediatrics 2004, 113(2):229-237. 9. Groer MW, Morgan K: Immune, health and endocrine characteristics of depressed postpartum mothers. Psychoneuroendocrinology 2007, 32(2):133-139. 10. Rahman A, Bunn J, Lovel H, Creed F: Maternal depression increases infant risk of diarrhoeal illness: –a cohort study. Arch Dis Child 2007, 92(1):24-28. 11. Rahman A, Iqbal Z, Bunn J, Lovel H, Harrington R: Impact of maternal depression on infant nutritional status and illness: a cohort study. Arch Gen Psychiatry 2004, 61(9):946-952. 12. World Health Organisation: The global burden of disease: 2004 update Geneva; 2008. 13. Graham NM: The epidemiology of acute respiratory infections in children and adults: a global perspective. Epidemiol Rev 1990, 12:149-178. 14. Lewis JD, Schinnar R, Bilker WB, Wang X, Strom BL: Validation studies of the health improvement network (THIN) database for pharmacoepidemiology research. Pharmacoepidemiol Drug Saf 2007, 16(4):393-401. 15. Bourke A, Dattani H, Robinson M: Feasibility study and methodology to create a quality-evaluated database of primary care data. Inform Prim Care 2004, 12(3):171-177. 16. Tata L, Hubbard R, McKeever T, Smith C, Doyle P, Smeeth L, West J, Lewis SA: Fertility Rates in Women with Asthma, Eczema, and Hay Fever: A General Population-based Cohort Study. Am J Epidemiol 2007, 165(9):1023-1030. 17. Rahman A, Lovel H, Bunn J, Iqbal Z, Harrington R: Mothers’ mental health and infant growth: a case-control study from Rawalpindi, Pakistan. Child Care Health Dev 2004, 30(1):21-27. 18. Rahman A, Harrington R, Bunn J: Can maternal depression increase infant risk of illness and growth impairment in developing countries? Child Care Health Dev 2002, 28(1):51-56. 19. Morris R, Carstairs V: Which deprivation? A comparison of selected deprivation indexes. J Public Health Med 1991, 13(4):318-326. 20. Dobson R: UK health officials launch “five in one” vaccine for babies. BMJ 2004, 329(7462):365. 21. Department of Health: Poliomyelitis. Immunisation Against Infectious Disease -’The Green Book’ 2006 edition Norwich: The Stationery Office; 2007, 313-328. 22. Cox JL, Murray D, Chapman G: A controlled study of the onset, duration and prevalence of postnatal depression. Br J Psychiatry 1993, 163:27-31. 23. Kumar R, Robson KM: A prospective study of emotional disorders in childbearing women. Br J Psychiatry 1984, 144:35-47. 24. Pelletier DL, Frongillo EA, Schroeder DG, Habicht JP: The effects of malnutrition on child mortality in developing countries. Bull World Health Organ 1995, 73(4):443-448. Submit your next manuscript to BioMed Central 25. Tupasi TE, Mangubat NV, Sunico ME, Magdangal DM, Navarro EE, and take full advantage of: Leonor ZA, Lupisan S, Medalla F, Lucero MG: Malnutrition and acute respiratory tract infections in Filipino children. Rev Infect Dis 1990, • Convenient online submission 12(Suppl 8):S1047-1054. 26. Hasselmann MH, Werneck GL, Silva CVCD: Symptoms of postpartum • Thorough peer review depression and early interruption of exclusive breastfeeding in the first • No space constraints or color figure charges two months of life. Cad Saude Publica 2008, 24(Suppl 2):S341-352. • Immediate publication on acceptance 27. de Miranda CT, Turecki G, Mari JDJ, Andreoli SB, Marcolim MA, Goihman S, Puccini R, Strom BL, Berlin JA: Mental health of the mothers of • Inclusion in PubMed, CAS, Scopus and Google Scholar malnourished children. Int J Epidemiol 1996, 25(1):128-133. • Research which is freely available for redistribution 28. Stewart RC: Maternal depression and infant growth: a review of recent evidence. Matern Child Nutr 2007, 3(2):94-107. Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Public Health Springer Journals

Association between perinatal depression in mothers and the risk of childhood infections in offspring: a population-based cohort study

BMC Public Health , Volume 10 (1) – Dec 31, 2010

Loading next page...
 
/lp/springer-journals/association-between-perinatal-depression-in-mothers-and-the-risk-of-TP2F02l8dy

References (36)

Publisher
Springer Journals
Copyright
Copyright © 2010 by Ban et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; Public Health; Medicine/Public Health, general; Epidemiology; Environmental Health; Biostatistics; Vaccine
eISSN
1471-2458
DOI
10.1186/1471-2458-10-799
pmid
21194453
Publisher site
See Article on Publisher Site

Abstract

Background: Previous studies have suggested that children of mothers who experience depression during the perinatal period may have more infections, but such studies are few in number and none have been carried out in the United Kingdom (UK) population. The aim of this study was to investigate the association between perinatal depression in mothers and the risk of childhood infections in offspring in the UK general population. Methods: We used data from The Health Improvement Network (THIN), a large database of electronic primary care medical records to conduct a cohort study among all first-born singleton children born and enrolled in THIN between 1988 and 2004. We used Poisson regression to compare the incidence of gastrointestinal infections and lower respiratory tract infections reported between birth and age 4 years among children of mothers with a record of perinatal depression with those born to mothers with no such history. Results: Children of mothers with perinatal depression had a 40% increased risk of gastrointestinal infections and a 27% increased risk of lower respiratory tract infections compared with children of mothers without perinatal depression (incidence rate ratios = 1.40 and 1.27; 95% confidence intervals 1.37-1.42 and 1.22-1.32, respectively). On restricting to antibiotic-treated infections there was a slight increase in the magnitude of association with gastrointestinal infections but a decrease in that with lower respiratory tract infections (incidence rate ratios = 1.47 and 1.19; 95% confidence intervals 1.34-1.61 and 1.11-1.27, respectively). Conclusions: Maternal perinatal depression is associated with increased rates of childhood gastrointestinal infections, particularly more severe infections, and lower respiratory tract infections in the UK. Preventing maternal perinatal depression may avoid substantial morbidity among offspring, although further work is also needed to investigate the detailed reasons for these findings. Background is some evidence to suggest that a dysregulated hapotha- According to recent guidance from the National Insti- lamic pituitary adrenocortical axis may lead to immuno- tute for Health and Clinical Excellence [1], depression supression in mothers with perinatal depression [6,9], in mothers during and after pregnancy (maternal perina- which may directly impair their children’simmuneand tal depression) is one of the most important issues for neuroendocrine development. women’s health in the United Kingdom (UK). Depres- Researchers in Pakistan studying 265 mothers and sion is estimated to affect 10% to 20% of women during their children found an increased risk of diarrhoea the perinatal period in high-income countries [2-4] and among children born to mothers with postnatal depres- can have a substantial adverse impact on women ’s sion compared with those born to mothers without health as well as on maternal behaviour towards their postnatal depression (adjusted odds ratio [OR] = 3.1; children and childcare practices [5-8]. In addition, there 95% confidence interval [CI] 1.8-5.1) [10]. They also reported an unadjusted result for respiratory infections * Correspondence: mcxlb4@nottingham.ac.uk in children born to mothers with either antenatal or † Contributed equally postnatal depression (OR = 1.1; 95% CI 0.9-1.4) in the Department of Epidemiology & Public Health, University of Nottingham, same cohort [11]. Moreover, a cross-sectional study Nottingham, UK © 2010 Ban et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Ban et al. BMC Public Health 2010, 10:799 Page 2 of 8 http://www.biomedcentral.com/1471-2458/10/799 conducted in the United States of America reported an were excluded. Follow-up data were available for all association between postpartum depression and physical children for the first four years of their life unless they illness (including, but not limited to, infections) in chil- died or ceased to be registered at the practice. All ana- dren (p = 0.03) [9], suggesting that an association lyses were restricted to the first four-years of life since between maternal perinatal depression and childhood in the UK children generally start attending school at infections may also exist in high-income countries. between 4 and 5 years of age, leading to a considerable Since the prevalence of maternal perinatal depression change in their environmental exposure to infection. is high, any association with childhood infections could We extracted data including records of GI infections, LRTIs and antibiotic prescriptions for children and be an important public health concern in high-income countries. Although the relative impact of infections on records of maternal perinatal depression and antidepres- childhood survival and severe morbidity is lesser than in sant use for mothers. Based on previous research low income countries, infectious illnesses are still a [9,10,17,18], we also extracted data (where available) on major cause of mortality and morbidity among young maternal age at birth, body mass index [BMI] before children in higher income countries and are both pregnancy, smoking status before childbirth, caesarean socially and economically burdensome [12,13]. In section delivery, and household socioeconomic status European countries in 2004, more than 16% of deaths in (represented by quintiles of Townsend deprivation children under five-years of age were attributed to child- indices [19]), and the sex of the child, since these factors hood acute respiratory infections (ARIs) and about 14% may be related to maternal perinatal depression as well. were attributed to childhood diarrhoeal illness [12]. We therefore conducted a large population-based Measurement of maternal perinatal depression study in the UK with the primary aim of examining the Mothers were defined as having perinatal depression if, potential effect of maternal perinatal depression on at any time during pregnancy or in the first six months childhood gastrointestinal (GI) infections and lower after childbirth, they had a medical record of diagnosed respiratory tract infections (LRTIs) in the first four years depression (e.g. major depression, mild depression, or of life. We also investigated whether healthcare-seeking chronic depression), or if they received at least one pre- behaviour differed between mothers with and without scription for an antidepressant medication. perinatal depression, as this may lead to differential levels of diagnosis of minor self-limiting infections. Measurement of childhood infections and antibiotic prescriptions Methods Children were defined as having GI infections if they Dataset and study population had at least one clinical diagnosis of GI infection (e.g. We used data from The Health Improvement Network infectious colitis, enteritis or gastroenteritis) or infec- (THIN), a large database of anonymised computerised tion-related symptoms (e.g. vomiting or diarrhoea) primary health care records of patients throughout the recorded in their general practice records. The date of th UK. There is a high standard of validated recording of recovery from illness was defined as the 15 day after medical diagnoses, medical conditions and symptoms diagnosis. If a subsequent diagnosis was recorded prior and prescriptions in THIN [14,15]. The data in this to the recovery date, this was considered part of the study derive from 255 general practices across England same episode of illness and the recovery date was th and Wales with longitudinal records of 3.9 million extended to the 8 day after this recording (only if patients. Previous research [15,16] has shown that the later than the original recovery date, ensuring that an population in THIN is generally representative of the episode lasted at least 14 days). Further recordings UK population. Ethical approval for this study was prior to the recovery date were treated in the same granted by the South East Research Ethics Committee. way. Recordings after the date of recovery were treated For each woman who was of childbearing age (15 to as indicating the onset of new episodes of illness. We 50 years of age) at any time between January 1988 and used the same process to identify and classify episodes November 2004, birth-related entries in their general of LRTIs,wherethese were definedasatleastone practice records during this period were linked to regis- clinical diagnosis of an LRTI (e.g. pneumonia or bron- tered children born in the same household at the time chitis) in the primary care record. To obtain a measure of delivery. Women were included if they had one or of more severe episodes of illness, we also identified all more live born children recorded in the database. For antibiotic-treated GI and LRTI infectious episodes, this study, we only included the first recorded singleton which we defined as those where children had one or pregnancy ending in a live born child for each woman. more records of a prescription for an antibiotic dated Pregnancies ending in stillbirth, miscarriage, therapeutic on or after the date of onset and prior to the date of abortion and multiple pregnancies (e.g. twins delivered) recovery. Ban et al. BMC Public Health 2010, 10:799 Page 3 of 8 http://www.biomedcentral.com/1471-2458/10/799 Statistical analysis had perinatal depression. The mean age of the mothers We calculated crude incidence rates as the total number was 29 years (Standard Deviation 5.7). Complete follow- of episodes for each infectious outcome (episodes of GI up data to age 4 years was available for 80.5% of all chil- infections with and without antibiotic treatment and dren, and 96.0% of children were followed up to at least episodes of LRTIs with and without antibiotic treat- 2 years of age. ment) divided by children’s total follow-up period dur- Table 1 shows that compared with mothers without ing the first four years of life. We calculated the perinatal depression, mothers with perinatal depression absolute differences in crude incidence rates by sub- were slightly younger (p < 0.001). They were also more tracting the rates in children of mothers without perina- likely to be current smokers, obese, in a household with tal depression from those in children of mothers with higher material deprivation and to have had a caesarean perinatal depression. To assess whether the incidence section delivery than mothers without perinatal depres- rates changed over the first years of the child’s life and sion (p < 0.001 for all associations). whether the association of maternal perinatal depression with childhood infections changed with increasing time Maternal depression and childhood infections from birth, we also calculated the incidence rates for Proportionately more children born to mothers with each year of the child’s life separately. perinatal depression had at least one GI infection To estimate the effect of maternal perinatal depression (60.6%) than children born to mothers without perinatal on the risk of childhood infections, we used Poisson depression (49.2%). Similarly, children born to mothers regression to obtain incidence rate ratios (IRRs) for each with perinatal depression were morelikelytohaveone type of infectious outcome among children of mothers or more LRTIs than children born to mothers without with perinatal depression relative to children of mothers perinatal depression (20.9% vs. 17.0%). When only epi- without perinatal depression, overall and for each year sodes of childhood infections with at least one antibiotic of age up to 4 years. Findings were considered statisti- prescription were considered, the prevalence decreased cally significant where p was less than 0.05. to 5.2% and 3.6% for GI infections and 10.2% and 8.7% We also attempted to assess whether mothers with for LRTIs, respectively. perinatal depression might be more likely to report The overall rates of childhood GI infections were 3.41 childhood illness to their general practitioners (GPs), or per 10 person-years (95% CI 3.35-3.47) for children of to do so sooner than mothers without perinatal depres- mothers with perinatal depression and 2.44 per 10 per- sion, by examining whether time to first polio immuni- son-years (95% CI 2.43-2.46) for children of mothers sation in children differed between mothers with and without perinatal depression (Table 2). The overall rates without perinatal depression as a potential marker of for childhood LRTIs were 0.91 per 10 person-years (95% mothers’ healthcare-seeking behaviour. We determined CI 0.88-0.94) and 0.72 per 10 person-years (95% CI the date of each child ’s first polio immunisation as 0.71-0.73), respectively for mothers with and without recorded in the general practice records for children perinatal depression (Table 2). The overall rate differ- st st born between 1 January 1991 and 31 December 2003 ences were therefore 0.97 per 10 person-years (95% CI only, since the UK immunisation schedule and vaccine 0.90-1.03) for GI infections and 0.19 per 10 person- type did not change during this time period [20,21]. We years (95% CI 0.16-0.23) for LRTIs (Table 2). calculated a median time to first polio vaccination in In addition, the rates of childhood infections in chil- children of mothers with and without perinatal depres- dren of mothers with perinatal depression were consis- sion. All analyses were carried out using Stata SE 10.0 tently higher than those of mothers without perinatal software. depression through the first four years of life, generally decreasing gradually in both cohorts for both GI infec- Statistical Power tions and LRTIs as the children got older (Table 2). Based on previous studies [2-4,22], we estimated that In our study, children born to women with perinatal about 10% of mothers in our study population would depression were 40% more likely to have GI infections have perinatal depression. We calculated that our sam- and 27% more likely to have LRTIs compared with ple size would give us over 99% power to detect a 1.5- those born to women without perinatal depression (IRR fold increased rate in children of mothers with perinatal = 1.40; 95% CI 1.37-1.42 and IRR = 1.27; 95% CI 1.22- depression for both GI infections and LRTIs. 1.32, respectively). The observed effects were similar for each year within the first four years of the child’s life Results (Table 2). Furthermore, for more severe infectious epi- Study population sodes that were treated with antibiotics (Table 3), the We identified a cohort of 107,587 mothers and their effect of maternal perinatal depression on childhood first live born children. Of the mothers, 9,722 (9.1%) infections increased slightly for GI infections and Ban et al. BMC Public Health 2010, 10:799 Page 4 of 8 http://www.biomedcentral.com/1471-2458/10/799 Table 1 Characteristics of mothers with and without depression (N = 107587) Maternal perinatal depression* 2 † No (n = 97,815) Yes (n = 9,772) c (df)P N (%) N (%) Maternal age at birth of child (years) Mean (Standard deviation) 29.2 (5.6) 28.6 (5.9) t = 9.5 (107585) < 0.001 Maternal body mass index (kg/m ) Normal (18.5-24.9) 37971 (38.8) 4069 (41.6) 575.9 (4) < 0.001 Underweight (<18.5) 2660 (2.7) 395 (4.0) Overweight (25-29.9) 12093 (12.4) 1502 (15.4) Obese (>30) 5653 (5.8) 910 (9.3) Missing 39438 (40.3) 2896 (29.6) Maternal smoking Non-smoker 47266 (48.3) 3952 (40.4) 1300 (3) < 0.001 Ex-smoker 4698 (4.8) 567 (5.8) Current smoker 19985 (20.4) 3463 (35.4) Missing 25866 (26.4) 1790 (18.3) Household socioeconomic status 1 (Least deprivation) 16028 (16.4) 1112 (11.4) 419.1 (5) < 0.001 2 11714 (12.0) 972 (10.0) 3 11170 (11.4) 1106 (11.3) 4 9298 (9.5) 1214 (12.4) 5 (Most deprivation) 7063 (7.2) 1078 (11.0) Missing 42542 (43.5) 4290 (43.9) Caesarean section delivery 13604 (13.9) 1616 (16.5) 50.6 (1) < 0.001 Sex of the child Male 50192 (51.3) 4922 (50.4) 3.2 (1) 0.075 * Mothers with either depression diagnosis and/or at least one antidepressant prescription either during pregnancy and/or in the first six months after the child’s delivery. Degrees of freedom. Student t-test statistic. Table 2 Associations between maternal perinatal depression and childhood gastrointestinal infections and childhood lower respiratory infections Maternal perinatal depression No Yes Childhood infections Events Rate per 10 person-years Events Rate per 10 person-years Rate difference Unadjusted IRR (95% CI*) (95% CI*) (95% CI*) (95% CI*) GI infections Age (years) 0-4 84999 2.44 (2.43-2.46) 11634 3.41 (3.35-3.47) 0.97 (0.90-1.03) 1.40 (1.37-1.42) 0-1 34928 3.71 (3.67-3.75) 4950 5.31 (5.16-5.46) 1.60 (1.45-1.75) 1.43 (1.39-1.48) 1-2 26845 3.02 (2.98-3.06) 3533 4.06 (3.93-4.19) 1.04 (0.90-1.18) 1.34 (1.30-1.39) 2-3 14289 1.69 (1.66-1.72) 1981 2.40 (2.30-2.51) 0.71 (0.60-0.82) 1.42 (1.35-1.49) 3-4 8937 1.11 (1.09-1.13) 1170 1.49 (1.41-1.58) 0.38 (0.30-0.47) 1.35 (1.27-1.43) LRTIs Age (years) 0-4 25143 0.72 (0.71-0.73) 3139 0.91 (0.88-0.94) 0.19 (0.16-0.23) 1.27 (1.22-1.32) 0-1 12202 1.28 (1.26-1.31) 1632 1.72 (1.64-1.81) 0.44 (0.36-0.53) 1.35 (1.28-1.42) 1-2 6530 0.73 (0.71-0.75) 725 0.82 (0.76-0.88) 0.09 (0.03-0.16) 1.13 (1.05-1.22) 2-3 3783 0.45 (0.43-0.46) 454 0.55 (0.50-0.60) 0.10 (0.05-0.15) 1.23 (1.11-1.35) 3-4 2628 0.33 (0.31-0.34) 328 0.42 (0.37-0.46) 0.09 (0.04-0.14) 1.28 (1.14-1.44) * Confidence interval. Incidence rate ratio. Gastrointestinal infections. Lower respiratory tract infections. Ban et al. BMC Public Health 2010, 10:799 Page 5 of 8 http://www.biomedcentral.com/1471-2458/10/799 Table 3 Associations between maternal perinatal depression and childhood infections with antibiotic treatment Maternal perinatal depression No Yes Antibiotic-treated Events Rate per 10 Events Rate per 10 Rate Unadjusted childhood infections person-years person-years difference IRR (95% CI*) (95% CI*) (95% CI*) (95% CI*) GI infections Age (years) 0-4 3550 0.10 (0.10-0.10) 512 0.15 (0.14-0.16) 0.05 (0.03-0.06) 1.47 (1.34-1.61) 0-1 1144 0.12 (0.11-0.13) 176 0.18 (0.16-0.21) 0.07 (0.04-0.09) 1.54 (1.32-1.81) 1-2 1153 0.13 (0.12-0.14) 172 0.19 (0.17-0.23) 0.07 (0.04-0.10) 1.52 (1.29-1.78) 2-3 731 0.09 (0.08-0.09) 104 0.12 (0.10-0.15) 0.04 (0.01-0.06) 1.45 (1.18-1.78) 3-4 522 0.06 (0.06-0.07) 60 0.08 (0.06-0.10) 0.01 (0.00-0.03) 1.18 (0.90-1.54) LRTIs Age (years) 0-4 8868 0.25 (0.25-0.26) 1035 0.30 (0.28-0.32) 0.05 (0.03-0.07) 1.19 (1.11-1.27) 0-1 3462 0.36 (0.35-0.37) 418 0.44 (0.40-0.48) 0.07 (0.03-0.12) 1.21 (1.10-1.34) 1-2 2536 0.28 (0.27-0.29) 276 0.31 (0.28-0.35) 0.03 (0.00-0.07) 1.11 (0.98-1.25) 2-3 1661 0.20 (0.19-0.21) 195 0.23 (0.20-0.27) 0.04 (0.00-0.07) 1.20 (1.03-1.39) 3-4 1209 0.15 (0.14-0.16) 146 0.19 (0.16-0.22) 0.04 (0.00-0.07) 1.24 (1.04-1.47) * Confidence interval. Incidence rate ratio. Gastrointestinal infections. Lower respiratory tract infections. decreased slightly for LRTIs (IRR = 1.47; 95% CI 1.34- observed absolute differences in rates suggest that over 1.61 and IRR = 1.19; 95% CI 1.11-1.27, respectively). a one year period there were about 97 more GI infec- tions and 19 more LRTIs among every 1000 children Maternal perinatal depression and the time to first polio born to women with perinatal depression than among vaccination in children 1000 children born to women without perinatal depres- st We identified 92,218 children born between 1 January sion. Moreover, the near identical timing of the first st 1991 and 31 December 2003 who accounted for 85.7% polio vaccination among both groups provides no evi- of the whole cohort and who were representative of the dence to support the possibility that higher rates of GI whole cohort as they had almost identical distributions infections and LRTIs in children of mothers with peri- of maternal characteristics (data not shown). Overall, natal depression may be explained by differential levels 98% of children born to women with perinatal depres- of diagnosis due to increased healthcare-seeking beha- sion had their first polio vaccination, as did the same vior among this group. proportion of children born to women without perinatal depression. The median time of receipt of the first polio Strengths and limitations vaccination was 61 days after birth for children of We have conducted a large population-based cohort mothers with perinatal depression (interquartile range study of the association between maternal perinatal [IQR] 58-68) and this was almost identical for children depression and childhood infections using primary care of mothers without perinatal depression (median = 61 data in theUK. Ourconsiderablesample sizemeans days; IQR 57-67). that our findings are unlikely to be due to chance. The data we used were obtained from a national database Discussion and prospectively recorded by GPs, thus excluding the Principal findings possibility of recording or recall bias in both our expo- Our results show that children born to women with sure and outcome. perinatal depression had a 40% increased rate of GI Full 4-year follow-up data were not available for 19.5% infections and a 27% increased rate of LRTIs compared of children. Such drop-outs occur due to death or with children born to women without perinatal depres- patients deregistering from GP practices. There was no sion. For more severe, antibiotic-treated infections, the evidence of an increased risk of death in either cohort. observed effects increased slightly for GI infections to In the UK registration with a general practitioner is con- 47% but decreased slightly for LRTIs to 19%. Our tinuous unless a patient actively requests removal from Ban et al. BMC Public Health 2010, 10:799 Page 6 of 8 http://www.biomedcentral.com/1471-2458/10/799 the practice list, or makes an application to join another mothers with perinatal depression therefore could be an practice (typically due to a change of address, or the overestimate due to biased ascertainment. The first opening of a new practice). Drop-out rates were similar childhood polio vaccination is a routine appointment in both cohorts and we feel they are unlikely to have common to both cohorts of children and mothers and had a substantial impact on our results. one might expect the timing to be earliest and uptake Although the inclusion of postpartum depression in the greatest amongst children of mothers with high levels of healthcare-seeking behaviour. Although timing the exposure definition meant that there could be and uptake of polio vaccination were all but identical in potential for reverse causation due to some overlap both cohortsweacceptthatthisisonlyanindirect, between postpartum depression and childhood infec- tions in the first six months of a child’s life, such over- measure of mothers’ health seeking behaviour and we lap could not affect the observations at age 1, 2, and cannot conclusively exclude the possibility of biased 3 years. As we found that the increased risk of child- ascertainment. hood infections in mothers with perinatal depression Although we considered other factors related to was consistent throughout the first four years of the maternal perinatal depression in our analysis, the child’s life, we believe it is unlikely that our results were records of a substantial number of individuals did not due to reverse causality. provide information about factors such as household There was a legitimate weakness in our exposure mea- socioeconomic status (44% missing), maternal smoking surement as a certain proportion of mothers with peri- (26.4%) and maternal BMI (40% missing). We were also natal depression are neither diagnosed nor treated in unable to adjust for some other factors, such as the per- primary care. The prevalence of diagnosed maternal iod of breastfeeding and the child’s birth weight which perinatal depression found in our study (about 9.1%) have previously been linked to childhood infections was in general 4-6% lower than some previous studies [13,24,25] in low-income countries, because birth weight conducted in high-income countries [2-4,23]. However, and breastfeeding status are seldom recorded in THIN. these studies were based in research settings where It could be argued, however, that these factors may selected populations were screened for depression form part of a causal pathway linking maternal perinatal mostly using self-administered questionnaires and their depression to childhood infections. After adjusting for samples were much smaller than ours. On restricting to these factors, one previous study [10] (carried out in diagnosed depression using a standardized interviewing Pakistan) still found a statistically significant association between maternal postpartum depression and childhood schedule, two previous studies (including a GI infections, suggesting that incomplete adjustment for meta-analysis) [4,22] found fairly similar prevalence figures to ours. Although we acknowledge that existing confounding is unlikely to completely explain the obser- depression does remain undiagnosed in the population, vations of our study. during pregnancy and in the first few months after delivery, women typically havemorefrequentcontact Interpretation in context of previous studies with health professionals than they would usually do Although to our knowledge there are only three pre- because of antenatal/postnatal check-ups. This would vious studies [9-11] examining the potential association tend to reduce under-diagnosis of depression this per- between maternal perinatal depression and childhood iod. Nevertheless, the presence of under-diagnosis in infections, all have found an increased risk of childhood this population would only result in an underestimation infections in children born to women with perinatal of the risk of childhood GI infections and LRTIs in chil- depression. For example, in a recent American cross- dren associated with perinatal depression, since such sectional study of 194 mothers and their children at 4-6 misclassification would produce a bias towards the null- weeks postpartum, Groer and Morgan found that chil- hypothesis (of no association between perinatal depres- dren of mothers with postpartum depression experi- sion in mothers and childhood infections in offspring). enced more physical illness (including diarrhoeal illness) We defined childhood infections as children having since birth compared with children of mothers without been diagnosed with GI infections or LRTIs in primary postpartum depression (p = 0.03) [9]. This study, how- care, which primarily relied on mothers bringing their ever, did not provide any measure of effect and did not children to see their GPs, so they most likely repre- adjust for potential confounding factors. sented the more severe end of the spectrum of these In our study, we combined the period of antenatal and infections. We supposed that, if women with perinatal postnatal depression together and therefore we did not depression felt more anxious or less capable to take care examine whether the increased risk of childhood infec- of their children, they may take children to see their tions was independently associated with antenatal GPs more often than women without perinatal depres- depression or early postnatal depression. Rahman et al sion and the increased risk of childhood infections in carried out a similar cohort study [11] of 265 mothers Ban et al. BMC Public Health 2010, 10:799 Page 7 of 8 http://www.biomedcentral.com/1471-2458/10/799 (130 mothers with perinatal depression) in Pakistan and more acute respiratory infections than those who are found an increased risk of diarrhoea and a slightly not [13]. Likewise, children of mothers with perinatal increased, but not statistically significant, risk of ARIs in depression have been found to be less well nourished children of mothers with antenatal or postnatal depres- than those of mothers without depression [27,28], which sion compared with children of mothers without depres- will increase susceptibility to infection. sion (OR = 2.4; 95% CI 1.7-3.3 and OR = 1.1; 95% CI In addition, postpartum depression in mothers can lead to child neglect and poorer childcare practices, 0.9-1.4, respectively), although they did not adjust for especially in high-income countries [5,29,30]. Children any confounding factors. A subsequent study by the may thus be more exposed to unsanitary and dangerous same authors [10] which focused only on the relation- ship between postnatal depression and childhood diar- environments, increasing the potential for exposure to rhoeal illness showed a similar result to the first, sources of infection. although the risk of diarrhoea in children of mothers with postnatal depression compared with children of Conclusions mothers without postnatal depression was much stron- Our findings show that depression in mothers during ger once adjusted for important maternal and child fac- and after pregnancy is associated with an increased risk tors (unadjusted OR = 2.3; 95% CI 1.6-3.1 and adjusted of childhood infections in offspring. If causally related, OR = 3.1; 95% CI 1.8-5.6). The latter study used data our findings suggest that a substantial social and eco- from the same cohort of women and only about 5% of nomic burden of childhood morbidity could be avoided women with antenatal depression did not also have through effective prevention of antenatal and postnatal postnatal depression [10]. depression [1], although further work is required to Compared with our study, Rahman et al found a lar- investigate the detailed reasons for these findings. In pri- ger effect on GI infections but a slightly smaller one on mary care, physicians and other healthcare professionals respiratory infections. Their study population was from should be aware of the potential increased presentation a rural area of Pakistan where childhood infections were of childhood infections in children of mothers with peri- much more common than in our study population. natal depression. Furthermore, they used only binary outcomes of more or fewer than five childhood diarrhoeal episodes and Abbreviations more or fewer than six ARI episodes per year, while our 95% CI: 95% confidence interval; ARIs: acute respiratory infections; BMI: body outcomes were incidence rates of individual episodes of mass index; GI infections: gastrointestinal infections; GP: general practitioner; HR: hazard ratio; IQR: interquartile range; IRR: incidence rate ratio; LRTIs: GI infections and LRTIs. Consequently, our results are lower respiratory tract infections; OR: odds ratio; THIN: The Health not directly comparable. In spite of different study Improvement Network; UK: United Kingdom. populations and different measures of outcome, we both Acknowledgements found an increased risk of GI infections (and also a This study was founded through the Division of Epidemiology & Public similar magnitude of respiratory infection risk) in chil- Health of the University of Nottingham. dren born to women with perinatal depression com- Authors’ contributions pared with those born to women without perinatal LB carried out the literature review and participated in the development of depression. the study design, data management, data analysis, interpretation of the data, There could be several potential explanations for the writing the article and revising article drafts for publication. JEG had the initial idea for the study and its design and participated in the development association between perinatal depression in mothers and of the study design, data management, interpretation of the data, and childhood infections in offspring. First of all, previous revising article drafts for publication. JW participated in the development of research [6,9] has found that mothers with depression the study design, interpretation of the data, and revising article drafts for publication. LJT had the initial idea for the study and its design and during the perinatal period have a dysregulated hypotha- participated in the development of the study design, data management, lamic pituitary adrenocotical axis resulting in decreased interpretation of the data, and revising article drafts for publication. All levels of salivary cortisol and potentially depressed cellu- authors read and approved the final manuscript. lar immunity. This may have a direct impact on the Competing interests child’s immune and neuroendocrine development. Sec- The authors declare that they have no competing interests. ondly, perinatal depression can have a considerable Received: 13 April 2010 Accepted: 31 December 2010 impact on mothers’ childcare abilities. Previous studies Published: 31 December 2010 [9,26] reported that mothers with postpartum depres- sion were less likely to breastfeed their children and References more likely to give up exclusive breastfeeding than 1. National Institute for Health and Clinical Excellence: Antenatal and postnatal mental health (Clinical management and service guidance) London; 2007. mothers without depression. Breastfeeding can 2. Gavin NI, Gaynes BN, Lohr KN, Meltzer-Brody S, Gartlehner G, Swinson T: strengthen the immune system of the child and it has Perinatal depression: a systematic review of prevalence and incidence. been shown that children who are not breastfed have Obstet Gynecol 2005, 106(5 Pt 1):1071-1083. Ban et al. BMC Public Health 2010, 10:799 Page 8 of 8 http://www.biomedcentral.com/1471-2458/10/799 3. Cooper PJ, Murray L, Hooper R, West A: The development and validation 29. Beck CT: The effects of postpartum depression on maternal-infant of a predictive index for postpartum depression. Psychol Med 1996, interaction: a meta-analysis. Nurs Res 1995, 44(5):298-304. 26(3):627-634. 30. Windham AM, Rosenberg L, Fuddy L, McFarlane E, Sia C, Duggan AK: Risk 4. O’hara MW, Swain AM: Rates and risk of postpartum depression–a meta- of mother-reported child abuse in the first 3 years of life. Child Abuse & analysis. International Review of Psychiatry 1996, 8(1):37-54. Neglect 2004, 28(6):647-669. 5. McLennan JD, Kotelchuck M: Parental prevention practices for young children in the context of maternal depression. Pediatrics 2000, Pre-publication history 105(5):1090-1095. The pre-publication history for this paper can be accessed here: 6. Groër M, Davis M, Casey K, Short B, Smith K, Groër S: Neuroendocrine and http://www.biomedcentral.com/1471-2458/10/799/prepub immune relationships in postpartum fatigue. MCN Am J Matern Child Nurs doi:10.1186/1471-2458-10-799 2005, 30(2):133-138. Cite this article as: Ban et al.: Association between perinatal depression 7. Turner C, Boyle F, O’Rourke P: Mothers’ health post-partum and their in mothers and the risk of childhood infections in offspring: a patterns of seeking vaccination for their infants. Int J Nurs Pract 2003, population-based cohort study. BMC Public Health 2010 10:799. 9(2):120-126. 8. Bartlett SJ, Krishnan JA, Riekert KA, Butz AM, Malveaux FJ, Rand CS: Maternal depressive symptoms and adherence to therapy in inner-city children with asthma. Pediatrics 2004, 113(2):229-237. 9. Groer MW, Morgan K: Immune, health and endocrine characteristics of depressed postpartum mothers. Psychoneuroendocrinology 2007, 32(2):133-139. 10. Rahman A, Bunn J, Lovel H, Creed F: Maternal depression increases infant risk of diarrhoeal illness: –a cohort study. Arch Dis Child 2007, 92(1):24-28. 11. Rahman A, Iqbal Z, Bunn J, Lovel H, Harrington R: Impact of maternal depression on infant nutritional status and illness: a cohort study. Arch Gen Psychiatry 2004, 61(9):946-952. 12. World Health Organisation: The global burden of disease: 2004 update Geneva; 2008. 13. Graham NM: The epidemiology of acute respiratory infections in children and adults: a global perspective. Epidemiol Rev 1990, 12:149-178. 14. Lewis JD, Schinnar R, Bilker WB, Wang X, Strom BL: Validation studies of the health improvement network (THIN) database for pharmacoepidemiology research. Pharmacoepidemiol Drug Saf 2007, 16(4):393-401. 15. Bourke A, Dattani H, Robinson M: Feasibility study and methodology to create a quality-evaluated database of primary care data. Inform Prim Care 2004, 12(3):171-177. 16. Tata L, Hubbard R, McKeever T, Smith C, Doyle P, Smeeth L, West J, Lewis SA: Fertility Rates in Women with Asthma, Eczema, and Hay Fever: A General Population-based Cohort Study. Am J Epidemiol 2007, 165(9):1023-1030. 17. Rahman A, Lovel H, Bunn J, Iqbal Z, Harrington R: Mothers’ mental health and infant growth: a case-control study from Rawalpindi, Pakistan. Child Care Health Dev 2004, 30(1):21-27. 18. Rahman A, Harrington R, Bunn J: Can maternal depression increase infant risk of illness and growth impairment in developing countries? Child Care Health Dev 2002, 28(1):51-56. 19. Morris R, Carstairs V: Which deprivation? A comparison of selected deprivation indexes. J Public Health Med 1991, 13(4):318-326. 20. Dobson R: UK health officials launch “five in one” vaccine for babies. BMJ 2004, 329(7462):365. 21. Department of Health: Poliomyelitis. Immunisation Against Infectious Disease -’The Green Book’ 2006 edition Norwich: The Stationery Office; 2007, 313-328. 22. Cox JL, Murray D, Chapman G: A controlled study of the onset, duration and prevalence of postnatal depression. Br J Psychiatry 1993, 163:27-31. 23. Kumar R, Robson KM: A prospective study of emotional disorders in childbearing women. Br J Psychiatry 1984, 144:35-47. 24. Pelletier DL, Frongillo EA, Schroeder DG, Habicht JP: The effects of malnutrition on child mortality in developing countries. Bull World Health Organ 1995, 73(4):443-448. Submit your next manuscript to BioMed Central 25. Tupasi TE, Mangubat NV, Sunico ME, Magdangal DM, Navarro EE, and take full advantage of: Leonor ZA, Lupisan S, Medalla F, Lucero MG: Malnutrition and acute respiratory tract infections in Filipino children. Rev Infect Dis 1990, • Convenient online submission 12(Suppl 8):S1047-1054. 26. Hasselmann MH, Werneck GL, Silva CVCD: Symptoms of postpartum • Thorough peer review depression and early interruption of exclusive breastfeeding in the first • No space constraints or color figure charges two months of life. Cad Saude Publica 2008, 24(Suppl 2):S341-352. • Immediate publication on acceptance 27. de Miranda CT, Turecki G, Mari JDJ, Andreoli SB, Marcolim MA, Goihman S, Puccini R, Strom BL, Berlin JA: Mental health of the mothers of • Inclusion in PubMed, CAS, Scopus and Google Scholar malnourished children. Int J Epidemiol 1996, 25(1):128-133. • Research which is freely available for redistribution 28. Stewart RC: Maternal depression and infant growth: a review of recent evidence. Matern Child Nutr 2007, 3(2):94-107. Submit your manuscript at www.biomedcentral.com/submit

Journal

BMC Public HealthSpringer Journals

Published: Dec 31, 2010

There are no references for this article.