Background: Findings from animal and human studies indicate that anxiety and stress have a negative influence on the child and mother. The aim of this study was to explore the risk for having an anxiety diagnosis and the impact of the diagnosis in a three generational perspective. Methods: The information was retrieved from Swedish population-based registries. All women who gave birth between 1973 and 1977 (n 169,782), their daughters (n 244,152), and subsequently also the offspring of the daughters (n 381,953) were followed until 2013. Results: We found that 4% of the mothers and 6% of the grandmothers had been diagnosed with anxiety. Women who had mothers with an anxiety disorder were more than twice as likely to have an anxiety disorder themselves compared to all other women (OR = 2.20, 95% CI = 2.04–2.30). In the third generation, the children born to mothers with an anxiety disorder, the odds ratio of being diagnosed with anxiety was more than twice as high than for the rest of the population (OR = 2.54, 95% CI = 2.01–3.20). If both the mother and the grandmother had had an anxiety disorder the odds ratio for the child having a diagnosis of anxiety was three times higher (OR = 3.11, 95% CI = 2.04–4.75). Anxiety diagnosis in the two previous generations also increased the likelihood of the child having either more than two inpatient visits or more than 10 outpatient visits (OR = 2.64, 95% CI = 2.40–2.91 and OR = 2.21, 95% CI = 2.01–2.43, respectively). Conclusions: The intergenerational effect on anxiety is high. In order to minimize the risk for further transmission of anxiety disorders, increased awareness and generous use of effective treatment regimes might be of importance. Keywords: Anxiety, Multi-generation, Women, Offspring Background individuals with anxiety disorders are genetically predis- Anxiety related diagnoses are prevalent among women posed to develop drug addiction . Comorbidity with in reproductive age groups . In the clinical setting other mental disorders is common, and the prevalence these conditions may be difficult to detect and therefore of individuals with both general anxiety disorders (GAD) may be difficult to diagnose. Consequently, women suf- and depression is high. In a recent Swedish prevalence fering from these conditions are often undertreated or study with 3000 participants it was found that among not treated at all. This may lead to a lower quality of life men and women with clinically significant depression or for these women, may have negative effects on family re- anxiety, nearly 50% had comorbid disorders. The point lations, and also have adverse effects on the children’s prevalence of major depression was 5.2, and 8.8% had mental health and wellbeing. Anxiety may also have an GAD . Women’s lifetime prevalence of being diag- impact on the woman’s behavior leading to, for example, nosed with an anxiety disorder is around 30% . For problems such as increased drug abuse or becoming women of reproductive age, pregnancy and childbirth prone to accidents, which have a deleterious effect on are sometimes triggers for worsening an already mani- the woman . It has also been suggested that some fested anxiety disorder or for deterioration of an existing anxiety disorder. Thus, pregnant women with mental * Correspondence: Gunilla.Sydsjo@liu.se; email@example.com health problems are common in the antenatal care set- Department of Obstetrics and Gynaecology and Department of Clinical and ting [5, 6]. Anxiety symptoms in pregnant women who Experimental Medicine, Linköping University, SE-581 85 Linköping, Sweden Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 2 of 11 may or may not have been given an anxiety diagnosis national population comprising three generations. More present themselves in a number of different ways. For specifically, the primary aim was to investigate the risk example, some of these women might have a severe fear for being diagnosed with anxiety given that previous of childbirth, be anxious about the child’s health, anxious generation(s) of mothers had been diagnosed with about the future, about their relationship with their part- anxiety. Additional aim was to examine the use of spe- ner, and even about their new role as a parent [7, 8]. For cialized health care among third generation children, if pregnant women in a population-based community sam- either or both previous generation of mothers had been ple, the prevalence of anxiety symptoms was found to be diagnosed with anxiety. almost 16% in early pregnancy . The study found that women under 25 years of age were at an increased risk Methods for anxiety symptoms during early pregnancy and also Data collection revealed that women who were more psychosocially dis- The information on all the participants in this study was advantaged were more often nicotine users before preg- retrieved from Swedish population-based registries. All nancy. A psychiatric history of depression increased the Swedish residents are given unique personal identifica- risk of anxiety symptoms, as did a prior history of anxiety tion numbers that allow us to individually link the infor- disorders. Women who showed symptoms of anxiety mation about each person from different registers. expressed a greater fear of childbirth than those who Background variables such as educational level, marital showed no such symptoms. The adverse effect of anxiety status, and parity were registered in the Swedish Medical on the fetus and the pregnant woman is not fully under- Birth Register (MBR) at the time of admission to antenatal stood. In a review, Alder and colleagues examined 35 care. From the other registers we collected information on studies published between 1990 and 2005 and found that parents’ country of birth, the women’s marital status, high- enhanced levels of anxiety symptoms during pregnancy est attained educational level as well as anxiety diagnosis contributed independently of other biomedical risk factors and use of specialized health care. More specifically, data to adverse obstetric, fetal, and neonatal outcomes . were collected from the following registers: Findings from animal studies have shown a link be- tween antenatal stress, measured as the occurrence of The Swedish Medical Birth Register (MBR): Medical major life events, and impact on behavior and emotional information on all births since 1973 and onwards adjustment as well as cognitive impairment of the off- has been stored in the MBR which is held by the spring both in childhood but also later in adulthood Swedish National Board of Health and Welfare . . The interpretation of these findings is that changes The Total Population Register (TPR): The TPR is in the function of the hypothalamic – pituitary - adrenal held by Statistics Sweden and was established in (HPA) axis account for these relationships . 1968 [15, 16]. The register contains information on In a review by Monk et al., evidence for the impact of variables such as births, deaths, migrations, and human maternal distress on fetal and infant outcomes marital status. through epigenetic mechanisms was outlined. Prenatal ex- The Causes of Death Register: The Causes of Death posure to maternal anxiety and depression can have last- Register, which is held by the Swedish National ing effects on infant development with consequences for Board of Health and Welfare, contains information risk of psychopathology . The recurrence rate and on the cause of death and was established in chronicity of anxiety conditions are high. In addition, 1961 [17, 18]. many women with anxiety are not clinically recognized The Education Register and the Population and which means that many children are exposed to maternal Housing Census: Since 1985, Statistics Sweden anxiety not just temporarily but during a substantial part has continuously collected information on the of their childhood. educational level of the population in the Education Hence, the findings from both animal and human Register [19–21]. studies indicate that anxiety have a negative influence on The National Patient Register (NPR): The NPR was both child and mother. In a long-term perspective these originally established in 1964, with its main focus findings provide a solid foundation on which to form hy- being on psychiatric diagnoses. From 1987 all potheses concerning how anxiety may shape risks for inpatient visits are included and in 2001 outpatient coming generations. To our knowledge there are no visits were added to the register. studies based on a national population investigating whether there is an impact beyond the two generations These registers have all been evaluated [14, 22–26]; i.e. mother and child. The Inpatient Register was most recently evaluated by The aim of the present study was therefore to investi- Ludvigsson et al. in 2011, who concluded that the regis- gate the intergenerational transmission of anxiety in a ter is of good quality with a high validation rate . Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 3 of 11 Similarly, The Education Register, The Medical Birth education (Elementary, High school and Graduate/Post- Register, The Total Population Register and The Cause graduate), region of origin (Nordic/Non-Nordic), marital of Death Register have been evaluated and deemed to be status (Married, Unmarried and Divorced/widowed), age of a high quality. when giving birth (< 20 years, 20–26 years, 27–33 years and > 33 years). The same set of socio-demographic fac- Study population tors was collected for the mothers, except for origin For the purpose of this study we selected all women since all mothers were born in Sweden. who gave birth between 1973 and 1977 (n = 169,782, all In addition to anxiety diagnosis according to ICD-10, of these women were born between 1924 and 1963), patient data regarding the total number of outpatient their daughters (n = 244,152 who were born between and inpatient visits were collected for the third- 1973 and 1977), and subsequently also the offspring of generation children, and arbitrarily cut off incidences the daughters (n = 381,953 who were born between 1987 of 0–1visits/≥ 2 visits for inpatient data and 0–10 and 2012); all three groups were followed until the 31 visits/≥11 visits for out-patient data were chosen. December, 2012. At time of data collection no further These served as proxies for overall morbidity among data were available by the registers holder for a longer the third-generation children. follow-up, since data have to be validated before being released to researchers. The data collection included Statistical analysis identifying if the two first generations of women had be- To examine the risk for anxiety disorder in the studied come mothers, and their socio-demographic characteris- generation of women and their children we analyzed the tics, whether the individuals in the study (all generations) data by using Pearson’s chi-square to analyze bivariate had been diagnosed with an anxiety diagnosis during the differences. Data were also analyzed by unadjusted as entire follow-up time, as well as how many times they had well as adjusted logistic regression in order to estimate used specialized hospital care, either as an outpatient or the odds ratio of being diagnosed with anxiety, each gen- inpatient resulting in an anxiety diagnosis. eration modeled separately. The dependent variable was the presence of anxiety diagnoses and the independent Diagnoses variables were educational level, marital status, and par- To study these women and third-generation children, ity. For the third-generation children the same set of we used The National Patient Register (NPR), which independent variables were used to estimate the odds ra- contains all psychiatric inpatient care diagnoses and tios for having a larger amount of inpatient or outpatient from 2001 all outpatient-diagnoses in a hospital setting. visits to the hospital, or being diagnosed with anxiety. The anxiety diagnoses are based on the Swedish These models also included an additional independent version of The International Classification of Diseases variable; a three level indicator on the presence of (ICD) from the World Health Organization . Be- anxiety in the two previous generations (only 1st gener- tween 1969 and 1986 ICD version 8 was in use. In 1987 ation woman diagnosed with anxiety, only 2nd generation a new version of ICD was released and was in use until woman diagnosed with anxiety, or both generations diag- 1997 when the health care system changed the ICD- nosed with anxiety). version used from ICD-9 to ICD-10. During this year All analyses were performed using SPSS, version 22.0 ICD-9 and ICD-10 were used interchangeably. Anxiety (IBM SPSS Inc., Armonk, NY). diagnoses in ICD-8 were identified as codes 300–301, 305–308, ICD-9 codes were identified as codes 290–319 Results and in ICD-10 codes F40-F42 (24), this includes diagno- The study population is shown in Table 1. Approxi- ses such as phobic anxiety, panic disorders, generalized mately 4% of the second-generation mothers and 6% of anxiety disorder, and obsessive compulsive disorders. the first-generation mothers had, at some point, been di- Therefore we searched for both ICD-9 and 10 codes in agnosed with anxiety. Out of 381,953 children a total of 1997 and only ICD-10 in 1998 to 2004. ICD-8 and 65,838 of the children (i.e. the third-generation) had had ICD-9 codes were translated to ICD-10 using a conver- more than 10 outpatient visits for at some kind of medical sion table . or psychiatric disorder (median = 5, range = 1–261) and 53,649 had had two or more inpatient visits for some kind Definitions of disorder (median = 10, range = 2–504) as presented in Anxiety disorders in all three generations were divided Table 1. into two categories according to ICD-10 (24), diagnosis The background characteristics of the first- and present and diagnosis not present. Socio-demographic second-generation mothers are shown in Table 2. variables on the grandmothers included parity (previous We found that second-generation women diagnosed children/no previous children), highest attained level of with anxiety were more likely to have had mothers with Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 4 of 11 Table 1 The study population encompassing two generations visits were highest among those children for whom both of women and their offspring the first- and second-generation mothers had been diag- Study population nosed with anxiety (OR = 2.64, 95% CI = 2.40–2.91 and OR = 2.21, 95% CI = 2.01–2.43, respectively) compared n (%) to children where none of the previous generations had Total no. of children (third generation) 381,953 an anxiety diagnosis, Table 4. Stratifying by gender re- No. of children with anxiety diagnosis 749 (0.2) vealed that the increased likelihood of having an anxiety No. of children with ≥10 outpatient visits 65,838 (17.2) diagnosis, having more than two inpatient visits and/or No. of children with ≥2 inpatient visits 53,649 (14.0) more than 10 inpatient visits were approximately the Total no of mothers (second generation) 244,153 same among both boys and girls whose mothers and No of mothers with anxiety diagnosis 10,285 (4.2) grandmothers had been diagnosed with anxiety, in com- parison to boys and girls where none of the previous Total no. of grandmothers (first generation) 169,782 generations had had an anxiety diagnosis, Table 4. No of grandmothers with anxiety diagnosis 10,301 (6.1) After adjusting for socio-demographic factors (educa- tional level, marital status and parity) the odds ratios a lower level of education, with higher rates of divorce decreased. However, having a mother and/or a grand- or widowhood, and higher rates of having been diag- mother diagnosed with anxiety still remained an im- nosed with anxiety than the mothers of second- portant factor in determining whether the child had a generation women who had not been diagnosed with relatively higher number of visits for specialized med- anxiety. ical care and an anxiety diagnosis, Table 5. If both Moreover, second-generation women diagnosed with mother and grandmother were diagnosed with anxiety anxiety were less likely to have given birth during the the odds ratio was almost twice as high for in- and out- study period, and those who had given birth to at least patient care (OR = 1.94, 95% CI = 1.75–2.14 and OR = 1.74, one child, had reproduced at a younger age than women 95% CI = 1.58–1.92, respectively) while having been diag- who had not been diagnosed with anxiety. Limiting the nosed with anxiety the odds ratio was threefold (OR = 3.11, study population to second-generation women who had 95% CI = 2.04–4.75) compared to children where none of had at least one child it was found that second-generation the previous generations had been diagnosed with anxiety, mothers diagnosed with anxiety were more likely to have Table 5. Moreover, in the gender stratified analysis, both had mothers with a lower level of education, more often boys and girls where the two previous generations had been had mothers who had been divorced or become a widow, diagnosed with anxiety, had an increased likelihood of be- and who had had their child at a younger age, Table 3. ing diagnosed with anxiety, having two or more inpatient They themselves also had a lower level of education, were visits and/or 10 or more outpatient visits compared to more often divorced or widowed, and had had children at children where none of the previous generations had been a younger age compared to mothers not diagnosed with diagnosed with anxiety. This increased likelihood was ap- anxiety. Furthermore, second-generation mothers diag- proximately of the same magnitude among both boys and nosed with anxiety were also more prone to have children girls, Table 5. To further elucidate the impact of previous who had been diagnosed with anxiety and children who generation’s anxiety diagnosis, the 3rd generation was di- more frequently received specialized medical care; both as vided into two strata, 0–12 year olds, and 13 or older. This inpatients and outpatients. analysis validates the increased risk for an anxiety diagnosis The unadjusted odds ratio of having been diagnosed among the third generation children if any or both of the with anxiety was more than twice as high among previous generations have been diagnosed with anxiety, second-generation mothers who themselves had mothers Table 6. This was especially evident among girls, where the who at some point had been diagnosed with anxiety ORs, were higher and with a narrower confidence interval, (OR = 2.58, 95% CI = 2.43–2.74), Table 4. Among the 3rd indicating a more reliable estimate. Also, among both boys generation children, the lowest increased odds ratios and girls, there were increased risks for having been diag- were seen for those for whom only the first-generation nosed with anxiety in the younger groups (0–12 years of mothers had been diagnosed with anxiety (OR = 2.14 age), but the estimates did not always reach statistical sig- 95% CI 1.68–2.73), Table 4, while children where both nificance or had very wide confidence intervals. This was 1st and 2nd generation women had been diagnosed with probably due to the limited number of children having re- anxiety exhibited the highest odds ratio for being diag- ceived specialized hospital care due to anxiety. nosed with anxiety (OR = 7.74, 95% CI = 5.18–11.59). Among the third generation children, the odds ratios Discussion of being diagnosed with anxiety, having more than two In this nationwide population-based study we have been inpatient visits and/or having more than 10 outpatient able to shed light on the intergenerational transmission Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 5 of 11 Table 2 Socio-demographic characteristics of the study participants – limited to study persons born between 1973 and 1977, indifferent on future child Anxiety Anxiety p-value No Yes n (%) n (%) First-generation mothers Educational level Elementary 50,015 (22.1) 2539 (25.2) < 0.001 High school 110,512 (48.9) 4937 (49.1) Graduate/post-graduate 65,699 (29.0) 2589 (25.7 Civil status Married 190,869 (84.2) 7830 (78.6) < 0.001 Unmarried 17,493 (7.7) 990 (9.9) Divorced/widowed 18,261 (8.1) 1144 (11.5) Parity No previous children 102,328 (43.8) 4444(43.2) 0.275 Previous children 131,540 (56.2) 5841 (56.8) Age when giving birth < 20 16,099 (6.9) 1018 (9.9) < 0.001 20–26 111,212 (47.6) 4792 (46.6) 27–33 88,450 (37.8) 3626 (35.3) > 33 18,107 (7.7) 849 (8.3) Origin Nordic 9905 (4.2) 408 (4.0) 0.185 Non-Nordic 223,963 (95.8) 9877 (96.0) Second-generation women Educational level Elementary 9954 (4.4) 1513 (15.0) < 0.001 High school 89,281 (39.2) 4466 (44.2) Graduate/post-graduate 128,453 (56.4) 4129 (40.8) Civil status Married 123,020 (52.6) 4596 (44.7) < 0.001 Unmarried 107,549 (46.0) 5387 (52.4) Divorced/widowed 3298 (1.4) 302 (2.9) Parity No previous children 94,215 (54.9) 3891 (56.8) 0.003 Previous children 77,348 (45.1) 2964 (43.2) Age when giving birth < 20 3220 (1.9) 422 (6.2) < 0.001 20–26 37,956 (22.1) 2119 (30.9) 27–33 95,051 (55.4) 3093 (45.1) > 33 35,336 (20.6) 1221 (17.8) Childbirth No 48,825 (20.9) 2893 (28.1) < 0.001 Yes 185,043 (79.1) 7392 (71.9) First-generation mother diagnosed with anxiety Yes 13,480 (5.8) 8884 (86.4) < 0.001 No 220,388 (94.2) 1401 (13.6) First-generation mother, no. of visits to hospital 0–3 visits 104,551 (44.7) 3162 (30.7) < 0.001 due to anxiety 4- visits 129,317 (55.3) 7123 (69.3) < 0.001 Second-generation women, no. of visits to hospital 0–3 visits 133,878 (57.2) 652 (6.3) due to anxiety 4- visits 99,990 (42.8) 9633 (93.7) Chi2-test The category Unmarried includes women cohabiting with the child’s father though they are not legally married and women who are considered single (i.e. not married or cohabiting with the child’s father) Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 6 of 11 Table 3 Socio-demographic characteristics of the study participants, limited to women born between 1973 and 1977 who had become mothers Anxiety Anxiety p-value No Yes n (%) n (%) First-generation mothers Educational level Elementary 87,638 (22.6) 4197 (26.7) < 0.001 High school 191,237 (49.3) 7869 (50.0) Graduate/post-graduate 108,767 (28.1) 3661 (23.3) Civil status Married 322,053 (84.8) 12,314 (79.0) < 0.001 Unmarried 27,341 (7.2) 1458 (9.4) Divorced/widowed 30,443 (8.0) 1811 (11.6) Parity No previous children 107,573 (43.5) 7009 (43.6) 0.720 Previous children 221,502 (56.5) 9049 (56.4) Age when giving birth < 20 30,254 (7.7) 2020 (12.6) < 0.001 20–26 191,791 (48.9) 7857 (48.9) 27–33 142,058 (36.2) 5013 (31.2) > 33 27,972 (7.1) 1168 (7.3) Origin Nordic 379,255 (96.7) 15,459 (96.3) 0.001 Non-Nordic 12,820 (3.3) 599 (3.7) Second-generation mothers Educational level Elementary 17,193 (4.4) 2452 (15.3) < 0.001 High school 156,213 (39.9) 7297 (45.5) Graduate/post-graduate 218,399 (55.7) 6297 (39.2) Civil status Married 254,795 (65.0) 9559 (59.5) < 0.001 Unmarried 129,781 (33.1) 5766 (35.9) Divorced/widowed 7498 (1.9) 733 (4.6) Parity No previous children 174,958 (47.7) 6979 (46.4) 0.002 Previous children 191,962 (52.3) 8054(53.6) Age when giving birth < 20 8067 (2.2) 1003 (6.7) < 0.001 20–26 86,576 (23.6) 4862 (32.3) 27–33 203,016 (55.3) 6706 (44.6) > 33 69,261 (18.9) 2461 (16.4) Second-generation mother diagnosed Yes 23,073 (5.9) 2172 (13.5) < 0.001 with anxiety No 369,002 (94.1) 13,886 (86.5) First-generation mother, no. of visits to 0–3 visits 169,608 (43.3) 4721 (29.4) < 0.001 hospital due to anxiety 4- visits 222,467 (56.7) 11,337 (70.6) Second-generation mother, no. of visits 0–3 visits 171,508 (43.7) 346 (2.2) < 0.001 to hospital due to anxiety 4- visits 220,567 (56.3) 15,712 (97.8) Children (3rd generation) Gender Boy 188,199 (51.3) 7753 (51.6) 0.498 Outpatient visits Girl 178,721 (48.7) 7280 (48.2) 0–10 305,181 (83.2) 10,934 (72.7) < 0.001 ≥11 61,739 (16.8) 4099 (27.3) Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 7 of 11 Table 3 Socio-demographic characteristics of the study participants, limited to women born between 1973 and 1977 who had become mothers (Continued) Anxiety Anxiety p-value No Yes n (%) n (%) Inpatient visits 0–1 316,924 (86.4) 11,380 (75.7) < 0.001 ≥2 49,996 (13.6) 3653 (24.3) Anxiety diagnosis (child) Yes 391,447 (99.8) 15,397 (99.2) < 0.001 No 628 (0.2) 121 (0.8) The category Unmarried includes women cohabiting with the child’s father though they are not legally married and women who are considered single (i.e. not married or cohabiting with the child’s father) of anxiety in three generations. It was found that the had been diagnosed with anxiety disorders. If both the transmission of anxiety from one generation to the next mother and the grandmother had had an anxiety dis- is very high. More specifically, we found that women order the unadjusted odds ratio for the child having a (2nd generation) who had mothers (1st generation) with diagnosis of anxiety close to eight times higher. Adjust- an anxiety disorder were more than twice as likely to ing confounding factors such as marital status, educa- have an anxiety disorder themselves in comparison to all tional level and parity, the odds ratios decreased but other women. Moreover, in the third generation, among still remained elevated at three, and two and a half children born to mothers with an anxiety disorder, the times higher, respectively, compared to children where unadjusted odds ratio of being diagnosed with anxiety none of the previous generations had an anxiety diag- was more than four times higher compared to children nosis. An explanation for this might be that the special- (3rd generation) where none of the previous generations istcareismoreproneto investigateachild’s problem Table 4 Unadjusted odds ratios (OR) and corresponding 95% confidence intervals (CI) on the intergenerational effect of anxiety a,b,c disorder in three generations Total Boys Girls OR (95% CI) OR (95% CI) OR (95% CI) OR Second-generation mother diagnosed with anxiety First-generation mother diagnosed with anxiety 2.58 (2.43–2.74) 2.61 (2.36–2.88) 2.56 (2.31–2.84) First-generation mother not diagnosed with anxiety Reference Reference Reference OR child (3rd generation) has inpatient care First- and second-generation mother diagnosed with anxiety 2.64 (2.40–2.91) 2.36 (2.06–2.69) 3.02 (2.63–3.47) Only second-generation mother diagnosed with anxiety 1.98 (1.90–2.07) 1.92 (1.81–2.03) 2.08 (1.95–2.12) Only first-generation mother diagnosed with anxiety 1.30 (1.25–1.35) 1.25 (1.19–1.32) 1.36 (1.29–1.44) None diagnosed with anxiety Reference Reference Reference OR child (3rd generation) has outpatient care First- and second-generation mother diagnosed with anxiety 2.21 (2.01–2.43) 2.28 (2.01–2.59) 2.14 (1.86–2.46) Only second-generation mother diagnosed with anxiety 1.83 (1.76–1.90) 1.79 (1.70–1.89) 1.87 (1.77–1.99) Only first-generation mother diagnosed with anxiety 1.23 (1.19–1.27) 1.22 (1.16–1.28) 1.24 (1.81–1.31) None diagnosed with anxiety Reference Reference Reference OR child (3rd generation) diagnosed with anxiety First- and second-generation mother diagnosed with anxiety 7.74 (5.18–11.59) 6.73 (3.15–14.35) 8.40 (5.21–13.53) Only second-generation mother diagnosed with anxiety 4.63 (3.72–5.75) 4.40 (2.97–6.52) 4.92 (3.78–6.40) Only first-generation mother diagnosed with anxiety 2.14 (1.68–2.73) 2.36 (1.57–3.57) 2.06 (1.52–2.80) None diagnosed with anxiety Reference Reference Reference All second-generation mothers are born between 1973 and 1977 All children (third-generation) are born between 1987 and 2012 All first-generation mothers are born between 1924 and 1963 Inpatient care categorized into 0–1 visits and ≥ 2 visits Outpatient care categorized into 0–10 visits and ≥ 11 visits Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 8 of 11 Table 5 Adjusted odds ratios (OR) and corresponding 95% confidence intervals (CI) on the intergenerational effect of anxiety a,b,c disorder in three generations Total Boys Girls OR (95% CI) OR (95% CI) OR (95% CI) OR second generation mother diagnoses with anxiety First-generation mother diagnosed with anxiety 2.20 (2.04–2.38) 2.20 (1.98–2.44) 2.22 (1.99–2.47) First-generation mother not diagnosed with a anxiety Reference Reference Reference OR child (3rd generation) has inpatient care First- and second-generation mother diagnosed with anxiety 1.94 (1.75–2.14) 1.76 (1.53–2.02) 2.17 (1.88–2.52) Only second-generation mother diagnosed with anxiety 1.66 (1.59–1.74) 1.63 (1.53–1.73) 1.72 (1.60–1.83) Only first-generation mother diagnosed with anxiety 1.14 (1.10–1.19) 1.11 (1.05–1.07) 1.19 (1.12–1.26) None diagnosed with anxiety Reference Reference Reference OR child (3rd generation) has outpatient care First- and second-generation mother diagnosed with anxiety 1.74 (1.58–1.92) 1.84 (1.61–2.10) 1.63 (1.41–1.89) Only second-generation mother diagnosed with anxiety 1.62 (1.55–1.69) 1.62 (1.52–1.71) 1.63 (1.53–1.74) Only first-generation mother diagnosed with anxiety 1.11 (1.07–1.15) 1.12 (1.06–1.18) 1.11 (1.05–1.17) None diagnosed with anxiety Reference Reference Reference OR child (3rd generation) diagnosed with anxiety First- and second-generation mother diagnosed with anxiety 3.11 (2.04–4.75) 2.97 (1.37–6.45) 3.14 (1.88–5.22) Only second-generation mother diagnosed with anxiety 2.54 (2.01–3.20) 2.38 (1.57–3.62) 2.63 (1.99–3.49) Only first-generation mother diagnosed with anxiety 1.46 (1.14–1.88) 1.57 (1.02–2.42) 1.42 (1.04–1.93) None diagnosed with anxiety Reference Reference Reference All second-generation mothers are born between 1973 and 1977 All children (third-generation) are born between 1987 and 2012 All first-generation mothers are born between 1924 and 1963 Inpatient care categorized into 0–1 visits and ≥ 2 visits Outpatient care categorized into 0–10 visits and ≥ 11 visits and to be more precise in collecting medical history in present his or hers problems more systematically when order to diagnose appropriately. Another possible ex- seeking help. planation might be that the parents who seek help for The children in the third generation of mothers with anxiety related disorders are able to present a child’s anxiety also had a higher consumption of relatively ad- problem or help the child to present his or her prob- vanced medical care such as inpatient, and outpatient lems in an accurate way since they themselves have specialist care. With the exception of anxiety disorders, knowledge of anxiety disorders and how anxiety has af- the children’s exact medical diagnoses and interventions fected them. Lastly, the child might also be able to are outside the scope of this study and were therefore Table 6 Unadjusted odds ratios (OR) and corresponding 95% confidence intervals (CI) on the intergenerational effect of anxiety in a,b,c three generations, stratified by age and gender of the child (3rd generation) Total Boys Girls 0–12 years 13- years 0–12 years 13- years 0–12 years 13- years OR child (3rd generation) diagnosed with anxiety First- and second-generation diagnosed 6.41(2.03–20.23) 3.74 (2.42–5.77) 3.83 (0.53–27.23) 9.07 (2.35–40.14) 3.58 (1.57–8.19) 3.96 (2.36–6.62) with anxiety Second-generation mother diagnosed 2.13 (0.99–4.58) 3.14 (2.49–3.96) 1.06 (0.26–4.35) 3.54 (1.40–8.95) 3.48 (2.28–5.30) 3.06 (2.32–4.05) with anxiety Only first-generation mother diagnosed 2.07 (1.14–3.77) 1.51 (1.16–1.98) 2.73 (1.35–5.52) 1.20 (0.37–3.87) 1.55 (0.93–2.58) 1.51 (1.10–2.07) with anxiety None diagnosed with anxiety Reference Reference Reference Reference Reference Reference All second-generation mothers are born between 1973 and 1977 All children (third-generation) are born between 1987 and 2012 All first-generation mothers are born between 1924 and 1963 Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 9 of 11 not investigated. However, a study of the full medical also an increase in morbidity and mortality indicating a record might be expected to show a higher level of greater vulnerability in general . somatic ill-health that in turn might be a signal of In a prevalence study on mental health, the results additional problems within the family, the personality showed that among individuals with depression or of the child, and even childhood experiences outside anxiety, around one third did not receive treatment. Co- the family . morbidity was associated with higher symptom severity The transmission of anxiety from parents to offspring and lower health-related quality of life and that these is well known, but the underlying processes are poorly mental disorders form a unit and thus depression and understood [29–32]. Most studies have evaluated the anxiety can be seen together as a rule rather than an ex- risks from a two-generation perspective and there is no ception.  Overall, comorbidity is of great importance evidence that allows us to determine the relative import- to acknowledge and to investigate. Mainly in order to ance of genetic factors and environmental factors, re- understand the effect of comorbidity, but also to better spectively, on the transmission of anxiety disorders for screen the patient and design individual treatment and the 3rd generation. In this study we only have infor- care for the patients. A high recurrence rate of anxiety mation on anxiety disorders in the maternal probands conditions has also been found . [13, 29–32]. Little is known about the effect of pater- Reproductive events and particularly childbirth are risk nal anxiety disorders on children’s psychopathology factors for acquiring mental disorders for women. In a but in a study by Cooper et al. 2006 it was evident register study from Denmark it was concluded that prim- that there is a strong familiality of anxiety disorders iparous women had an increased risk of incident-related in general but the impact of maternal anxiety was more hospital admission to a psychiatric hospital for a mental evident . In a Swedish twin study, the authors argue disorder through the first 3 months after childbirth but that environmental transmission from parents is stronger among fathers there was no increase of severe mental dis- than genetic factors because the children learn an anxious orders that required their admission to a hospital . behavior from their parents through modeling . Differ- Animal as well as human studies have shown associa- ent paths of environmental transmission are plausible. For tions between antenatal stress and or anxiety develop- example, maternal anxiety has been shown to be associ- ment and behavioral/emotional disturbance in the child. ated with reduced tolerance to negative emotions in chil- Yet, the strength of this link was unclear since these dren . Anxious mothers were also shown to have studies did not examine the covariance of antenatal risks lower expectations of their children’s performance com- and did not distinguish between ante- and postnatal pared to non-anxious mothers . Another study found stress. The large Avon Longitudinal Study of Parents anxiety in children to be associated with overinvolved and and Children (ALSPAC) cohort showed a strong rela- critical parenting . Whether these parenting behaviors tionship between maternal anxiety in late pregnancy and are influenced by anxiety in the offspring, or whether the behavioral/emotional problems in their children at age rearing environment increases the risk of the development four . of anxiety in children is not fully understood. While par- Animal studies have shown that offspring of mothers enting styles were not investigated in this study, family who have suffered antenatal stress are over-reactive to structure and socioeconomic factors such as education stressors and hypersecrete cortisol compared with con- was found to impact the risk for anxiety. These findings trols. Both the behavioral and physiological disturbances underline the multifactorial etiology of anxiety conditions, last into adulthood in rodents and for several years, sug- and the importance to address the psychosocial environ- gesting that the HPA axis can be ‘programmed’ during ment. Moreover, there are studies suggesting that different the fetal period . genetic factors impact the development of anxiety disor- In most prevalence studies there are, generally, sig- ders during childhood, adolescence and early adulthood nificant gender differences with mental health disorders respectively , indicating genetic innovation and at- being more common in women than in men. The effect tenuation or possibly epigenetic mechanisms. Kendler et on the families and generations might therefore be of al. (2008) found support for a developmentally dynamic significance as our results show. Since mental health hypothesis of genetic effects on anxiety and depression, problems seem to have an impact on subsequent gener- which could explain the low level of homotypic con- ations it is of the utmost importance to detect and treat tinuity of anxiety disorders from childhood to adult- anxiety. hood . In a recent study by Weissman and A limitation of this study is that only diagnoses set at colleagues , it was found that offspring to depressed hospitals or specialist clinics such as psychiatric out- parents had a risk for major depression and that the patient clinics are used. Therefore one can suspect that period of peak for first onset was between ages 15–25. the true percentage of individuals with an anxiety dis- Onsets before adolescence were uncommon; there was order is higher as a number of individuals may have Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 10 of 11 been diagnosed and treated by their general practi- Ethics approval and consent to participate The study was approved by the regional Ethical Review Board, Linköping, tioners. Thus, only including diagnoses from a hospital Sweden, no. 03–556, 03–557, 07-M66 08–08-M 233–8 and 2014/112–31. or specialist clinic setting implies that only the most se- Informed consent not applicable. vere forms of anxiety are included. This may cause an overestimation of the intergenerational transmission. Competing interests The material contained in the manuscript has not been published or submitted Also, misclassification problems caused by unrecorded elsewhere for publication. The authors declare that they have no competing cases and/or incorrect registration of diagnostic codes interests. are known limitations in register studies. If so, the incor- rect registration is random and not systematic. More- Publisher’sNote over, the children (third-generation) in the present study Springer Nature remains neutral with regard to jurisdictional claims in were born between 1987 and 2012. While anxiety disor- published maps and institutional affiliations. ders develop relatively early in life with a mean onset Author details age of 11 years , the results are strengthened by the 1 Department of Obstetrics and Gynaecology and Department of Clinical and presence of an intergenerational effect already evident at Experimental Medicine, Linköping University, SE-581 85 Linköping, Sweden. Department of Child and Adolescent Psychiatry and Department of Clinical an early age. Another limitation is the lack of informa- and Experimental Medicine, Linköping University, Linköping, Sweden. tion of the timing of anxiety in relation to childbirth. Anxiety symptoms during pregnancy have been shown Received: 14 August 2017 Accepted: 2 May 2018 to increase the risk for adverse obstetric, fetal, and neo- natal outcomes , while maternal anxiety during early References childhood might impact the children through negative 1. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime or over-controlling parenting behaviors . Moreover, prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:593–602. recurrent or chronic anxiety episodes would have a 2. Brady K, Tolliver B, Verdiun M. Alcohol use and anxiety: diagnostic and greater impact on both the woman and her children, management issues. Am J Psychiatry. 2007;164:217–21. compared to a single anxiety episode. 3. Pandey SC, Zhang H, Roy A, Xu T. Deficits in amygdaloid cAMP-responsive element-binding protein signaling play a role in genetic predisposition to anxiety and alcoholism. J Clin Invest. 2005;115:2762–73. Conclusion 4. Johansson R, Carlbring P, Heedman Å, Paxling B, Andersson G. Depression, anxiety and their comorbidity in the Swedish general population: point Second-generation mothers diagnosed with anxiety had prevalence and the effect on health-related quality of life. Peer J. 2013;1:e98. children who had been diagnosed with anxiety and chil- https://doi.org/10.7717/peerj.98. dren who more frequently received medical care; both as 5. Andersson L, Sundström-Poromaa I, Bixo M, Wulff M, Bondestam K, Åström M. Point prevalence of psychiatric disorders during the second trimester of inpatients and outpatients. Anxiety disorders have a pregnancy: a population-baser study. Am J Obstet Gynecol. 2003;189:148–54. three-generational effect. Since intergenerational effect 6. Josefsson A, Berg G, Nordin C, Sydsjö G. Prevalence of depressive symptoms in on anxiety is high and in order to minimize the risk for late pregnancy and postpartum. Acta Obstet Gynecol Scand. 2001;80:251–5. 7. Feinberg ME, Jones DE, Roettger ME, Hostetler ML, Sakuma KL, Paul IM, further transmission of anxiety disorders, increased Ehrenthal DB. Preventive effects on birth outcomes: buffering impact of awareness and generous use of effective treatment re- maternal stress, depression and anxiety. Matern Child Health J. 2016;20:55–65. gimes might be important. 8. Storksen HT, Eberhard-Grahn M, Garthus-Niegel S, Eskild A. Fear of childbirth: the relation to anxiety and depression. Acta Obstet Gynecol Acknowledgements Scand. 2012;91:237–42. Professor Lawrence Lundgren for language corrections. 9. Rubertsson C, Hellström J, Cross M, Sydsjö G. Anxiety in early pregnancy: prevalence and contributing factors. Arch Womens Ment Health. 2014;17: Availability of data and materials 221–8. The Ethical Review Board approval was obtained for public sharing and 10. Alder J, Fink N, Bitzer J, Hösli I, Holzgreve W. Depression and anxiety during presentation of data on group level only. This means that the data used in pregnancy: a risk factor for obstetric, fetal and neonatal outcome? A critical this study can only be used for the approved research and cannot be shared review of the literature. J Matern Fetal Neonatal Med. 2007;20:189–209. by the authors. 11. Weinstock M. Alterations induced by gestational stress in brain morphology and behaviour of the offspring. Prog Neurobiol. 2001;65:427–51. Authors’ contributions 12. Schneider ML, Moore CF, Roberts AD, Dejesus O. Prenatal stress alters early All the authors fulfill the authorship criteria. GS and AJ have been responsible neurobehaviour, stress reactivity and learning in non-human primates: a for the idea and the design of the study. GS, SA, MB and AJ have been brief review. Stress. 2001;4:183–93. responsible of the writing and the critical discussion of the manuscript. MB has 13. Monk C, Spicer J, Champagne FA. Linking prenatal maternal adversity to done the statistical analysis together with GS and AJ. All authors have approved developmental outcomes in infants: the role of epigenetic pathways. Dev the final version of the manuscript. Psychopatol. 2012;24:1361–76. 14. Centre for Epidemiology, National Board of Health and Welfare. The Swedish Authors’ information Medical Birth Register; A summary of content and quality National Board of Gunilla Sydsjö is professor in psychosocial obstetrics and gynecology with Health and Welfare. Stockholm, Sweden. 2003. (http://www.socialstyrelsen.se/ main research focus on mental health issues for women and children, Lists/Artikelkatalog/Attachments/10655/2003-112-3_20031123.pdf). Accessed 24 Sara Agnafors is MD, PhD presently working and doing research in the area May 2018. of child and adolescent psychiatry, Marie Bladh is PhD and a statistician, with 15. Population and welfare department, Statistics Sweden. The total population primary interest in medical epidemiology, Ann Josefsson is professor and register. The population as at 31 December 2011. Statistics Sweden, Örebro, MD in obstetrics and gynecology, with main research focus in women’s Sweden. 2012. (http://ec.europa.eu/eurostat/cache/metadata/Annexes/ mental health issues. demo_gind_esms_se_an1.docx). Sydsjö et al. BMC Psychiatry (2018) 18:168 Page 11 of 11 16. Statistics Sweden. Registret över totalbefolkningen (RTB) (https://www.scb. 38. Scholten WD, Batelaan NM, van Balkom AJ, Wjh Penninx B, Smit JH, van se/vara-tjanster/bestalla-mikrodata/vilka-mikrodata-finns/individregister/ Oppen P. Recurrence of anxiety disorders and its predictors. J Affect Disord. registret-overtotalbefolkningen-rtb/. Accessed 24 May 2018. 2013;147(1–3):180–5. 17. Centre for Epidemiology, National Board of Health and Welfare. 39. Munk-Olsen T, Laursen TM, Pedersen CB, Mors O, Mortensen PB. New Dödsorsaksstatistik Historik, produktionsmetoder och tillförlitlighet. parents and mental disorders: a population-based register study. JAMA. Komplement till rapporten Dödsorsaker 2008. Stockholm, Sweden: National 2006;296(21):2582–9. Board of Health and Welfare; 2008. http://www.socialstyrelsen.se/Lists/ 40. O'Connor TG, Heron J, Glover V. Alspac study Team.Antenatal anxiety Artikelkatalog/Attachments/18019/2010-4-33.pdf. Accessed 24 May 2018. predicts child behavioral/emotional problems independently of postnatal depression. J Am Acad Child Adolesc Psychiatry. 2002;41(12):1470–7. 18. Centre for Epidemiology, National Board of Health and Welfare. Dödsorsaksregistret (http://www.socialstyrelsen.se/register/dodsorsaksregistret). Accessed 24 May 2018. 19. Statistics Sweden. Educational attainment of the population 2002. (publication no. UF0506). Örebro: Statistics Sweden; 2003. (https://www.scb.se/ statistik/UF/UF0506/2003M00/UF37SM0301.pdf). Accessed 24 May 2018. 20. Statistics Sweden. Population and housing census 1970 (SOS). Part 12. Stockholm, Sweden: National Central Bureau of Statistics. 21. Statistics Sweden. The Swedish register of education. Örebro: Statistics Sweden; 2004. (http://www.scb.se/statistik/UF/UF0506/Produktbeskrivning_ short_English_UF0506_20040101r.doc). Accessed 24 May 2018. 22. Ludvigsson JF, Andersson E, Ekbom A, Feychting M, Kim JL, Reuterwall C, Heurgren M, Otterbladh Olausson P. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450. 23. Cnattingius S, Ericson A, Gunnarskog J, Källén B. A quality study of a medical birth registry. Scand J Soc Med. 1990;18:143–8. 24. Statistics Sweden. Evalvering av utbildningsregistret. Örebro, Sweden: Author. 2006. Retrieved from https://www.scb.se/statistik/_publikationer/ be9999_2006a01_br_be96st0604.pdf. Accessed 24 May 2018. 25. National Board of Health and Welfare, Centre for Epidemiol- ogy. Kodningskvalitet i patientregistret, slutenvård 2008. Stockholm, Sweden: 2010. Author. Retrieved from http://www.socialstyrelsen.se/Lists/Artikelkatalog/ Attachments/18082/2010-6-27.pdf. Accessed 24 May 2018. 26. D’Amico M, Agozzino E, Biagino A, Simonetti A, Marinelli P. Ill-defined and multiple causes on death certificates: a study of misclassification in mortality statistics. Eur J Epidemiol. 1999;15:141–8. 27. Internationell statistisk klassifikation av sjukdomar och relaterade hälsoproblem – Systematisk förteckning – Svensk version 2018. http://www. socialstyrelsen.se/Lists/Artikelkatalog/Attachments/20928/2018-4-12.pdf. http://www.socialstyrelsen.se/Lists/Artikelkatalog/Attachments/20929/2018- 4-13.pdf. http://www.socialstyrelsen.se/Lists/Artikelkatalog/Attachments/ 20930/2018-4-14.pdf. 28. Jenkins JM, McGowan P, Knafo-Noam A. Parent-offspring transaction: mechanisms and the value of within family designs. Horm Behav. 2016;77:53–61. 29. Biederman J, Faraone SV, Hirshfeld-Becker DR, Friedman D, Robin JA, Rosenbaum JF. Patterns of psychopatology and dysfunction in high-risk children of parents with panic disorder and major depression. Am J Psychiatry. 2001;158:49–57. 30. Cooper PJ, Fearn V, Willetts L, Seabrook H, Parkinson N. Affective disorders in a clinical sample of children with anxiety disorders. J Affect Disord. 2006; 93:205–12. 31. Micco JA, Henin A, Mick E, Kim S, Hopkins CA, Biederman J, Hirshfeld-Becker DR. Anxiety and depressive disorders in offspring at high risk for anxiety: a meta-analysis. J Anxiety Disord. 2009;23:1158–64. 32. Hirshfeld-Becker DR, Micco JA, Henin A, Petty C, Faraone SV, Mazursky H, Bruett L, Rosenbaum JF, Biederman J. Psychopatology in adolescent offspring of parents with panic disorder, major depression, or both: a 10-year follow-up. Am J Psychiatry. 2012;169:1175–84. 33. Eley TC, McAdams TA, Rijsdik FV, Lichtenstein P, Narusyte J, Reiss D, Spotts EL, Ganiban JM, Neiderhiser JM. The intergenerational transmission of anxiety: a children-of-twins study. Am J Psychiatry. 2015;172:630–7. 34. Creswell C, Apetroaia A, Murray L, Cooper P. Cognitive, affective, and behavioral characteristics of mothers with anxiety disorders in the context of child anxiety disorder. J Abnorm Psychol. 2013;122(1):26–38. 35. Gar NS, Hudson JL. An examination of the interactions between mothers and children with anxiety disorders. Behav Res Ther. 2008; 46(12):1266–74. 36. Kendler KS, Gardner CO, Lichtenstein P. A developmental twin study of symptoms of anxiety and depression: evidence for genetic innovation and attenuation. Psychol Med. 2008;38(11):1567–75. 37. Weissman MM, Wickramaratne P, Gameroff MJ, Warner V, Pilowsky D, Kohad RG, Verdeli H, Skipper J, Talati A. Offspring of depressed parents: 30 years later. Am J Psychiatry. 2016;173:1024–32.
BMC Psychiatry – Springer Journals
Published: Jun 4, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera