Mothers who have given birth at an advanced age - health status before and after childbirth

Women postpone childbirth to an age when morbidity is higher and fertility has decreased and yet the knowledge of mothers’ morbidity related to age remains scarce. Swedish national register data from the Medical Birth Register and National Patient Register was used to investigate the incidence of diseases listed in the International Classification of Diseases, version 10 (ICD-10) in women who gave birth 2007–8. The index group consisted of women 40 years of age or older (n = 8 203) were compared to a control group of women, younger than 40 years (n = 15 569) at childbirth. The period studied was five years before childbirth to five years after. The main outcome measures were incidence of disease diagnosed in specialized hospital care. Demographical data and use of assisted reproduction (ART) were adjusted for. The results showed that older women were more likely to be single; less frequently used tobacco; were educated on a higher level; had a higher BMI and more often had used ART to become pregnant. The older women showed a higher morbidity rate. In the diagnostic groups: Neoplasms, Blood and immune system, Eye and adnexa, Ear and mastoid, Circulatory, Digestive, Skin and subcutaneous tissue, Musculoskeletal and connective tissue, and Genitourinary. The results add to the body of knowledge of a number of specific risks faced by older mothers and may be used to identify preventive actions concerning fertility and morbidity both before and after childbirth.

Definitions. Maternal age when giving birth was categorized into two levels, "≤39 years of age", "≥40 years of age", civil status was categorized into "Married/cohabiting" and "Single" while educational level was defined as "Elementary", "High school", and "College" as registered at the time of childbirth. Further, tobacco use during pregnancy, twinning, chronic disease, and ART were each categorized into "No" and "Yes", mother's BMI at registration was divided into "Normal", "Underweight", and "Overweight/obese" while parity was categorized into "Primiparous" and "Multiparous".
Morbidity up to 5 years prior to childbirth as well as up to 5 years after childbirth was investigated by the number of incidence of diagnoses within each of the first 14 main chapters defined in the International Classification of Diseases, version 10 (ICD-10). For each chapter, indicator variables were created where the levels indicated either "No chapter specific diagnosis during follow-up" or "Having at least one chapter specific diagnosis during follow-up", these were retrieved by type of visit (in-or outpatient) and summarized into an overall indicator variable for each of the ICD-10 chapters included in the study. In addition, all somatic diagnoses (chapters I-XIV, excluding chapter V) were summarized into a total indicator; Any type of somatic disease, where the two levels were "No somatic diagnosis during the follow-up" and "At least one somatic diagnosis during the follow-up".
Statistical analysis. The relationship between age when becoming a mother and socio-demographic and medical data, as well as morbidity, was initially analyzed using Pearson's Chi-square statistic. Multivariable analyses included multiple logistic regression models to calculate the odds ratio of being diagnosed with chapter-specific diagnoses, each modeled separately. Dependent variables were the indicator variables for having a chapter-specific diagnosis and independent variables included were: Age when giving birth, Civil status, ART treatment, Educational level, Tobacco use during pregnancy, Mother's BMI at registration, Parity, Twin birth, and Chronic disease, and for the analyses of morbidity up to 5 years after delivery the indicator variable "Any somatic disease" regarding morbidity up to 5 years prior to delivery was also included. To further investigate the relationship between age and morbidity, the same set of multiple regression models were performed on age-stratified data to evaluate age specific risk factors. A p-value < 0.05 (two-sided) was considered statistically significant. All analyses were performed using SPSS, version 22.0 (IBM Inc., Armonk, NY). ethics approval and consent to participate. This study was approved by the Regional Ethical Review Board, Linköping, Sweden and was performed according to the Declaration of Helsinki. No 2014/111-31. Date 26-03-2016. Informed consent is not applicable. Informed consent was not required, according to national

Results
Socio-demographic and medical background. By stratifying the data according to type of pregnancy (ART or spontaneous pregnancy) it was found that among mothers who had had a spontaneous pregnancy, women age 40 or older were single to a greater extent, had a higher level of education, and a higher BMI than the younger mothers, Table 1.
As for spontaneously pregnant women, women 40 years or older who became pregnant by using ART were more often single, and had a higher level of education, and were less likely to have delivered twins. But in contrast to spontaneously pregnant women, no difference was detected in their BMI, Table 1.
In addition, by stratifying the data according to age it was found that mothers in both age groups who had become pregnant using ART were more often married, had a higher level of education, used tobacco to a lesser extent, were more often primipara, and more often had twins compared to mothers who had become spontaneously pregnant. It was also noted that mothers age 40 years or older and who had been treated with ART more often had a chronic disease, Table 1.
Morbidity up to 5 years prior to childbirth. In analyzing morbidity up to 5 years prior to childbirth it was found that mothers age 40 years or older who had become spontaneously pregnant were more likely to have been given diagnoses listed in the following chapters ("Neoplasms", "Blood and Immune System", "Endocrine, Nutritional and Metabolic", "Mental and Behavioral Disorders", "Nervous System", "Eye and Adnexa", "Ear and Mastoid", "Circulatory", "Musculoskeletal and Connective Tissue", and "Genitourinary") compared to the mothers in the control group (≤39 years of age), Table 2. On the other hand, spontaneously pregnant mothers in the control group were more likely to have been diagnosed with "Infections" and "Respiratory disease" compared to mothers age 40 years or older, Table 2.
The only statistically significant differences regarding morbidity in relation to the mother's age among mothers who had undergone ART treatment to achieve pregnancy were for the diagnoses "Neoplasms" and "Ear and Mastoid" where older mothers were more likely to be given these diagnoses compared to younger mothers. Younger mothers were more likely to have been diagnosed with genitourinary disease, Table 2. By stratifying by age, it was found that women in both groups who had been treated with ART were more likely to have been diagnosed with "Neoplasms" and "Endocrine, nutritional and metabolic disease" compared to mothers who had become spontaneously pregnant. Mothers of all ages who had been treated with ART were more likely to have been diagnosed with a "Genitourinary disease" and "Any type of somatic disease" compared to women with a spontaneous pregnancy, Table 2. www.nature.com/scientificreports www.nature.com/scientificreports/ Morbidity up to 5 years after childbirth. In Table 3, data on mothers' morbidity up to five years after childbirth are displayed. Older mothers who had become spontaneously pregnant were more likely to have been diagnosed with these conditions: ("Neoplasms", Nervous System", "Eye and Adnexa", "Ear and Mastoid", "Circulatory", "Skin and Subcutaneous Tissue", "Musculoskeletal and Connective Tissue", "Genitourinary") as well as "Any type of somatic disease" compared to the younger age group (≤39 years of age). Amongst the mothers who had used ART to become pregnant, the older mothers had more often been given the diagnosis "Neoplasms" and "Eye and Adnexa" than the younger mothers had. Younger mothers, pregnant with ART, on the other hand, had more often been diagnosed with "Genitourinary disorder" as well as "Any type of somatic disease", Table 3. By stratifying the data by age it was found that mothers 40 years or older who had been treated with ART were more likely to have been diagnosed with "Infections", Neoplasms", "Endocrine, Nutritional and Metabolic", "Eye and Adnexa, "Respiratory", "Genitourinary", and "Any type of somatic disease" after childbirth, compared to mothers age 40 years or older who had become spontaneously pregnant, Table 3. Moreover, mothers in the control group who had been treated with ART were more likely to have been given the diagnoses "Endocrine, Nutritional and Metabolic", "Genitourinary" and "Any type of somatic disease" compared to the control-group mothers who were under 40 years of age and had become spontaneously pregnant, Table 3.
Descriptive analysis of morbidity up to 5 years prior to and after childbirth. To further investigate the relationship between morbidity and single status, usage of ART, and age when giving birth, multiple logistic regression was performed for each of the ICD-10 chapter diagnoses. Dependent variables were the indicator for having a chapter-specific diagnosis, and all the variables presented in Table 1 were included as independent variables (Civil status, Educational level, Tobacco use during pregnancy, Mother's BMI at registration, Parity, Twin birth, and Chronic disease), and for the analyses of morbidity up to 5 years prior and 5 years after delivery the indicator variable "Any type of somatic disease" was included.
As for morbidity prior to childbirth, dividing maternal age at birth in to the same six categories, a dose-response relationship was found between maternal age and morbidity due to "Neoplasms", "Blood and Immune system", "Nervous system", "Eye and adnexa", "Circulatory, and "Somatic disease" among all women as well as in the subgroup of primiparous women only (data not shown).

Discussion
In this case-control /observational epidemiological study of women's morbidity five years before and up to five years after childbirth, we found that spontaneously pregnant mothers who were 40 years of age or older at time of childbirth were more likely to have been diagnosed for several diagnosis chapters in ICD-10, in specialized care, both during the five years prior to and the five years following childbirth compared to their younger counterparts. Diagnosis within "Neoplasms", "Blood and Immune system", "Endocrine, nutritional and metabolic", "Mental and behavioral disorders", "Ear and Mastoid", "Circulatory", "Musculoskeletal and connective tissue" and "Genitourinary" were more common in the older age group.
Advanced maternal age as an independent variable for a range of pregnancy and delivery-related complications and as a risk factor for worse health postpartum as has been reported in previous studies [10][11][12][13] . Our results confirm that age is a risk factor for higher morbidity. The current study also found a dose-response relationship between maternal age and morbidity prior and after childbirth.
We also found that mothers in both age groups (≤39 and ≥40 years of age), who had achieved pregnancy by using assisted reproductive technology (ART), were more than twice as likely to have been diagnosed with "Any type of somatic disease" during the five years following childbirth compared to spontaneously pregnant women. Also, up to five years before childbirth, the mothers using ART had been significantly more often diagnosed with "Neoplasms", "Endocrine, nutritional and metabolic disorders" and more than three times more likely to have been diagnosed with "Any type of somatic disease" compared to mothers spontaneously pregnant.
Previous studies have concluded that there are higher overall risks for negative maternal and neonatal outcomes for conception resulting from use of ART compared to spontaneous conceptions 3,31-34 and our resulthigher morbidity both before and after childbirth, in women using ART-further indicates the need for giving greater attention to this group and doing further research on the use of ART for women with comorbidity to infertility.
The older mothers were more often single, regardless of type of pregnancy, compared to the younger mothers and ART-treated mothers were more than twice as often single. Divorce, separation or partner's death are probable explanations for the mothers' single status. However, single mothers pursuing parenthood on their own, both by usage of ART and by natural conception are probably also present in the group of single mothers. This is in line with previous studies describing women who become planned single mothers (also referred to as solo mothers) tending to be of higher age than partnered women 20,35 .
Substantially worse birth outcomes for single women than for cohabiting or married women have been reported 26,36 . Our results, showing that older single mothers giving birth have a significantly higher morbidity compared to both married/cohabiting mothers of the same age and compared to younger mothers whether single or married respectively probably indicate that being single is a factor contributing to increased risk of adverse birth outcomes.
This as well as the overall worse health amongst ART-using single and married women, calls for giving more attention to the ever-increasing number of single women pursuing parenthood by planning to become single mothers.
Of specific interest is our result that mothers in both age groups and regardless of type of pregnancy, who are single at time of childbirth, had an increased likelihood of being diagnosed with mental and behavioral disorders both up to five years before as well as up to five years after childbirth compared to cohabiting or married mothers. According to previous studies, the lack of a partner is one of many factors associated with psychiatric illness during and after pregnancy [37][38][39][40] . Parental mental health problems are highly negatively correlated with children's assessment of their own health and children growing up in single mother families are at larger risk for emotional problems 41 . The Fragile Family Study in the United States found that children growing up with single mothers are at elevated risk of poorer health 23 Table 4. Odds ratios and corresponding 95% confidence intervals (CI) for being diagnosed within each of the presented ICD-10 chapters, up to 5 years before childbirth.  Table 5. Odds ratios and corresponding 95% confidence intervals (CI) for being diagnosed within each of the presented ICD-10 chapters up to 5 years after childbirth.
economic disadvantage compared to the health and financial situation of partnered parents 42 . Single status is considered to be a psychosocial risk factor within antenatal care today. The women who chose to become single mothers were not specifically identified in this study, so we do not know if the poorer health seen in single mothers in this study would also be seen in mothers who had become single unintentionally. The active choice of becoming a single mother is sometimes seen as being protective against the psychological distress often considered to be generated from unplanned pregnancy or single status due to divorce or separation. However, a recent study by Munk-Olsen et al. 14 showed that even the women with a planned IVF pregnancy had a significantly heightened risk of experiencing psychiatric episodes requiring treatment at a psychiatric facility after delivery compared to women not giving birth. This is of interest in our discussion since the planned single mothers who used ART (IVF with donated sperm) sometimes are excluded from researchers' discussions of the risks of single motherhood. Perhaps pregnancy intention does not reduce the risk of psychiatric illness and, as recently noted, single mothers with psychiatric illness are far more vulnerable and their psychiatric illness increases the risk that the child's health will be adversely affected.
Women who plan to be single mothers, are sometimes described as being primarily professional women with higher education and more financially secure and therefore better protected against physical and mental health problems faced by mothers to be who are less well off. These women are also more likely to be able to afford using ART. In Sweden and a few other countries, ART is mainly financed by the government, a fact that consequently might not call for these otherwise protective characteristics in the single mother by choice, as when ART is privately financed. It is generally easier to become a single parent in Sweden and some other countries today than it was a few decades ago. If this leads to a growth in the size of the single-parent-by choice group, further research on this group is needed. However, the results of our study indicate that health care needs to give more attention to the ever-growing population of single mothers, not just to any special subgroup (21,39,41).
There has been a global increase in the percentage of children who grow up in single-parent households which suggests that more attention needs to be given to this group in many countries 43 . However, in Sweden, it has actually become less common for children to grow up in single-parent households 44 . The risk associated to single parenthood for both mother and child 23,45,46 calls for interventions aiming to promote the well-being of the mothers and by extension improving child adjustment. In a recent study, Taylor and Conger discuss how to promote strength and resilience in single mothers by considering factors amenable to change via behavioral interventions, for example by providing peer-group support combined with cognitive-behavioral training 47 .
Through multivariable statistical analysis we found that single status, pregnancy by use of ART and being over 40 years of age at time of childbirth, respectively, increased the likelihood of having been diagnosed with "Any type of somatic disease" (i.e. diagnosis in chapter 1-14 in ICD-10) during the five years prior to childbirth, while after childbirth only the use of ART and being 40 years of age or older entailed an increased risk of having been diagnosed with "Any type of somatic disease". The older mothers were more often single compared to the younger mothers; they had more often used ART to become pregnant and had more often a chronic disease and a higher morbidity. It appears that more and more women will choose postponement of parenthood suggesting that we are most likely to see an increase in the percentage of older single women with health problems who choose to use ART. This suggests that healthcare providers must look forward to a growing need for care for this population.
Maternal single status is associated with an increased risk for giving birth to a child with low birth weight, small for gestational age and pre-term 26 . The present study shows that the older mothers had a higher number of diagnoses and were more often single than their younger counterparts. Women (and men) postpone having children to a later time in life due to improved social conditions, better contraceptives and time-consuming academic and work-related achievements. Also, an increased societal acceptance of postponing parenthood to a later time in life, meeting a partner later in life, and a wish to establish financial security before starting a family 3,10 contribute to the postponement of parenthood. The underestimation of the impact of age-related factors on female fertility and the overestimation of the use of ART in overcoming age-related infertility [48][49][50][51][52][53] probably contribute to the postponement of childbearing. The substantial risks for both the woman and the child associated with giving birth at advanced maternal age 21,39,54 call for appropriate physician counseling of older women who are considering ART 55 . It appears that education about fertility and infertility should be given a more prominent place in health programs focused on how to prevent sexually transmitted diseases and unwanted pregnancies in schools and health settings.
Previous studies of mother's mental and physical health and morbidity, before and after childbirth, are to a large extent based on self-reported questionnaires or interviews analyzed with qualitative methods. Although problems have been reported after childbirth, mother's self-rated quality of life tend to differ due to mode of delivery and civil status and most of them perceive their health as good 16,17,56 . The current study contributes to building a body of knowledge about older mothers in a more objective manner than is possible with self-report questionnaire studies. The main results in our study are based on the diagnoses reported by physicians at in-and outpatient specialized care units who have clinically examined older mothers. Primary care data and findings from visits not involving a physician are not entered in the registers from which our data were collected, which further strengthen the validity of this study. The symptoms leading to a specialized care visit are also assumingly more severe and the diagnoses registered are thus more serious.
The large number of mothers included in this study is a major strength because it reduces the importance of a range of error sources such as skewness in recruitment and participation. The underreporting of inpatient data to the National patient register is estimated to be less than 1%. This validates our study's findings even further.
Definite conclusions about causality cannot be made due to the study's observational design but the study does contribute to the knowledge of the role of maternal age, civil status and use of ART in affecting a woman's morbidity. In accordance with previous knowledge of advanced age-related and single status-related risks for worse outcome for mothers, our findings further enhance the need for careful considerations in planning and providing reproductive guidance for older women. Being fully informed about risks that may affect the future mental and (2020) 10:9739 | https://doi.org/10.1038/s41598-020-66774-4 www.nature.com/scientificreports www.nature.com/scientificreports/ physical health of both the woman and child is advisable. Age is obviously a factor contributing to higher morbidity in the group of mothers in our study, and probably provides part of the explanation for higher morbidity after childbirth as well. However, the vulnerability to both physical and mental strains that characterizes the mere pregnancy and postpartum-period, in combination with pre-existing medical conditions, could, in part explain our results and this is in line with a recent study by Morris et al. 57 .

conclusions
The increased risk of several somatic and mental and behavioral disorders amongst older, single, or ART treated mothers points to the importance of developing new policy outlining a long-term commitment to developing systems to inform the women and to carefully consider interventions and treatments needed by an ever-increasing group of women who become pregnant at what is considered to be an advanced age. More resources need to be provided to support preventive actions and the provision of evidence-based information on fertility and reproduction for younger women at all societal levels with one goal being to reduce the number of pregnancies in women of advanced age.