Socio-economic and psychosocial factors in the management and prevention of preterm labour
Correspondence: Professor J.-M. Moutquin, CHUS, Fleurimont, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, QC, J1H 5N4, Canada.
Socio-economic factors associated with preterm labour include social class, (usually assessed by earnings and education), working conditions (professional status, ergonomic environment, working hours), physical and travelling activities, daily life activities, lifestyle, family status and psychosocial state as related to past and current pregnancy history together with current stress factors. A review of the association of these factors with preterm birth will be reported with an emphasis of the biological plausibility linking mostly emotional, and at a lesser degree, physical and psychological stress to the occurrence of preterm labour. A case control study, carried out in Quebec City among 101 women in preterm labour and 202 matched pregnancies for parity and gestational age, identified 7 risk factors in an explanatory multivariate model among 117 variables: Body mass index (BMI) <20 (OR; 95% CI: 3.96; 2.61–7.09), previous preterm labour (OR; 95% CI: 3.61; 1.12–11.65) previous low birth weight (OR; 95% CI: 2.24; 1.05–7.71), standing at work >2 hours (OR; 95% CI: 3.90; 1.53–9.91), Abruptio placentae (OR; 95% CI: 5.88; 1.20–28.76), urinary tract infection (UTI) (OR; 95% CI: 4.4.3; 1.47–13.34), and stress score >5 (OR; 95% CI: 2.56; 1.20–5.54). The most stressful events were related to family illness, mortality, disruption, violence or financial distress. Some risk factors cannot be modified (previous preterm labour, low birth weight and UTI), while preventive efforts should be directed towards attaining BMI >20 before conception, modifying working conditions during current pregnancy and appropriate management of acute emotional stress.
This paper is dedicated to the prediction and prevention of preterm labour. Predictors or markers for preterm labour include screening tools, which allow the early identification of risk factors. Therefore, we are mainly concerned about the issue of secondary prevention i.e. attempting to modify an existing risk of preterm labour with its usual consequence: preterm birth. This may also have some relevance for primary (population-based) prevention when health promotion interventions or programmes may prevent the occurrence of a risk factor (i.e. nutrition or working conditions).
Among the reported risk factors, a number of socio-economic and psychosocial factors will be assessed in relation to preterm labour, with particular emphasis on the influence of unhealthy behaviour and lifestyle. A brief review of current strategies to prevent preterm birth will be presented with newer insights to pursue our efforts to better understand these factors.
The preterm birth rate is greatly increased among the socially disadvantaged population. In the USA, a population-based study showed that the risks of preterm birth were directly related to education, income and occupation in both black and white pregnant women1. This was also reported in several European countries including the UK2. Among the strongest determinants, the level of education has been reported as a general predictor of health outcomes; minimal education (i.e. primary school) increases the preterm birth rate at least two-fold2. Extremes of age have also been reported: teenage pregnancy is associated with increased spontaneous preterm labour2,3, while older gravidas (>35 years old) have increased rates of medically induced preterm births3. Low income, single marital status and isolation during pregnancy are also associated with an increased preterm birth rate1–3.
Among other socio-economic disparities, determinants, such as low body mass index (BMI), inadequate weight gain during pregnancy, poor nutrition and short stature are all associated with an increased preterm birth rate. However, these risk factors could be attributed to the consequences of a lower socio-economic status3. Ethnicity is a more complex issue. It has been demonstrated that in African women, the duration of pregnancy is minimally shorter, however this does not explain the excess of prematurity encountered among Afro-American women3,4.
Working conditions including long hours (>35 hours/week), standing >2 hours, physical strenuous working conditions5, together with high stress jobs with high demands and low control6, are all associated with an increased preterm birth rate. However, this has not been observed in a Scandinavian country, where a highly developed social support system, with few work-related hazards, was already in place7.
Lifestyle is also related to socio-economic status. Lower social class is directly related to poor housing facilities, more people living in the household, unplanned pregnancies, unemployment and unhealthy habits, such as illicit drug use, smoking and alcohol abuse, together with social exclusion and violence in the home1,2,8.
In a retrospective study carried out in the Sherbrooke area, Quebec Province, Canada, which has an annual preterm birth rate of 7.0–8.0%, social risk factors were recorded from hospital charts. The study was unable to collect data in relation to family status, education, and earnings. In 22% of cases presenting with preterm premature rupture of the membranes (PPROM), and in 33% of cases with spontaneous preterm labour, the only finding was a social risk that could be attributed to the event leading to preterm birth (Table 1)9. Extremes of maternal age were equally distributed among pregnancies presenting with PPROM or spontaneous preterm labour. Smoking, as the only encountered risk factor, was also observed in 7% and 18% of cases with PPROM and spontaneous preterm labour, respectively. With spontaneous preterm labour, two cases occurred whilst travelling in a car and two others were associated with family disruption or domestic violence. Two cases of a pregnancy with a BMI <20 were reported with PPROM (Table 1)9. Importantly, this study was unable to identify an aetiological risk factor in approximately one out of eight women, both with PPROM and spontaneous preterm labour subtypes of preterm birth, implying that other social factors may have been overlooked9. These observations are in accordance with Berkowitz3, suggesting that subcategories of preterm birth may not represent aetiologically different entities.
Table 1. Social factors as main risk factors explaining preterm birth among women presenting with PPROM and spontaneous preterm labour, Sherbrooke Region, Qc, Canada (1998–1999)9.
|Maternal age <18||Maternal age <18|
|Maternal age >35||Maternal age >35|
|Strenuous work||Strenuous work|
|BMI <19.7||Family disruption|
| ||Marital violence|
However, disconcerting observations have been reported. In a French study, the highest risk of preterm labour was encountered among female physicians or nurses. Chinese women used to endure strenuous working conditions, but their reported preterm birth rate is among the lowest in the world. Another report from France showed that the socially disadvantaged Maghrebian migrant population living in Paris displayed a very low preterm birth rate. Finally, a psychiatrist in Saint-Justine hospital, Montreal, Canada, was able to stop preterm labour in a substantial fraction of women from all social backgrounds after a 40 minute interview.
These observations suggest that socio-economic factors may not be an independent determinant of preterm birth8, but rather preterm birth is a direct consequence of psychosocial stress initiated by socio-economic disadvantages.
Psychosocial Risk Factors
Psychosocial risk factors can be encountered in disadvantaged populations, although these may be present in any socio-economic class. Recent evidence, based on neuroendocrine and epidemiological studies, supports a role for maternal psychosocial stress as a significant and independent risk factor for preterm birth10.
Empirical reports linking maternal psychosocial stress and preterm birth were published in the late 1970s11. Since then, numerous reports have highlighted the role of stressful life events12,13, anxiety14, nervousness or depression2, but also psychic functioning15, as significant mediators to the onset of preterm labour.
Herrera16 studied the psychosocial environment and cellular immunity of 38 patients of whom 32 delivered preterm compared to 34 term normal pregnancies. Anxiety and stress, family functioning and the level of social support were assessed. The relative risk of preterm delivery increased up to 10.2 (95% CI 1.3–79.0) when pregnant women displayed a high stress level without social support (Table 2)16.
Table 2. Psychosocial stress level in pregnancy and adverse pregnancy outcomes (preterm birth: n= 32, preterm induced hypertension: n= 6) (modified from Herrera et al.)16.
|Low stress level||0.2||–|
|High stress level||5.1||1.7–15.6|
|High stress level, no support||10.2||1.3–79.0|
A case–control study, carried out in Quebec City, Canada, among 101 singleton pregnancies with spontaneous preterm labour and 202 normal pregnancies without preterm labour, matched for parity, maternal and gestational ages, identified seven significant risk factors in an explanatory multivariate model: BMI <20, previous preterm labour, previous birth less than 2500 g, standing up more than 2 hours at work, abruptio placentae, urinary tract infection (UTI) and anxiety or stress17. In relation to anxiety and stress (OR: 2.56, 95% CI: 1.20–5.54), women were invited by the interviewer to state whether they were stressed and why. Spontaneous answers were collected in cases and controls. Among the cases, 65% reported at least one stressful life event compared with 51% among pregnancies without preterm labour, within the last three months before the questionnaire. It is surprising that a fair proportion of the control group had also faced a stressful life event. Among significant psychosocial factors in women with preterm labour, mortality or illness of a close family member or children increased by 3–4 fold the risk of preterm delivery17. The occurrence of a pregnancy complication scored highest (OR 5.2; 95% CI: 2.3–11.8). Among the 20 self-reported stressful life events, there were three groups of commonly cited situations: partner abuse or family disruption; a partner who is regularly absent, and financial insecurity due to job loss, transfer or no money at all (Table 3)17.
Table 3. Stressful life events within the last three months reported in women in preterm labour (n= 101) and matched normal pregnancies (n= 202) matched for parity, maternal and gestational ages17.
|Personal health problem||0.5||0.1–2.2|
|Personal psychosocial problem||3.6||0.9–13.9|
|Problem with current partner||1.6||0.7–3.7|
|Violence current pregnancy||0.5||0.1–2.2|
|Problem with previous partner||1.9||0.3–11.8|
|Mortality within close family||4.8||2.1–11.5*|
|Serious illness close family||2.0||0.8–5.0|
|Anxiety for a close family member||1.8||0.8–4.3|
|Problems within close family||1.1||0.4–3.1|
|Problems with in-laws||1.9||0.3–11.8|
|Problems with own children||3.3||1.1–10.5*|
|Loss of job||1.3||0.4–4.3|
|Moving in new environment||1.2||0.3–5.0|
|Partner being away||2.1||0.5–10.00|
This suggests that stressful life events are either different in cases of preterm labour or that individual personality traits, life experience or coping resources have modified significantly the emotional perceptions and/or responses among women with preterm labour.
A stressor agent or a stressful event requires an immediate individual response to such a physical or psychological challenge. Any stress is perceived as a potential threat to the stability of the organism10. Human beings have the capacity to adapt to this situation (stress without distress) in recovering the previous steady state (homeostasis). This involves complex neuroendocrine mechanisms to recover a steady state. Firstly, there is a release of corticotrophin releasing hormone (CRH) and stimulation of the locus ceruleus noradrenaline autonomic (LC/NA) sympathetic neurons in the hypothalamus and brainstem, which modulate the peripheral activities of the hypothalamic–pituitary–adrenal (HPA) axis and the adrenomedullary sympathetic nervous system10. Activation of the HPA axis and LC/NA results in the systemic elevation of circulating corticosteroids and catecholamines, respectively, which interact to maintain or initiate a return to homeostasis18. Hypercorticism, if sustained, may lead to earlier onset of parturition by premature placental–fetal activation together with being immunosuppressive. Stress has been demonstrated to be immunosuppressive in women who delivered preterm with decreased lymphocyte activity16.
Stressful events may be acute where adaptative mechanisms attempt to re-establish homeostasis. However, stress, when chronic in duration, leads to an exhaustion of adaptive mechanisms, where anxiety and depression can be encountered. Among other forms of stress, there are experiences of racism and poverty, which elicit social isolation and loss of self esteem19.
The response to a stressor agent is very much dependent on host susceptibility or vulnerability19. Pregnancy per se can be stressful and require comprehensive physiological and psychosocial adaptations for the duration of pregnancy. In addition, inherited or acquired personality traits through life experiences, together with individual coping resources (sense of mastery or control), will play a determinant role in the way an individual copes with stressful events19.
In our case–control study, the majority of women in preterm and control pregnancies experienced stressful events. Emotional integration and reaction might have been different. This is in accordance with the different perceptions of pregnant women with preterm birth reported recently by Mamelle et al.15. Further evidence was reported among Danish pregnant women, where scales of stressful life events were not independently associated with preterm birth (OR: 1.05, 95% CI: 0.77–1.42), although the maternal perceptions of the individual importance of a specific life event was highly correlated with preterm birth (OR 1.76, 95% CI: 1.15–2.71), specifically between 16 and 30 weeks of gestational age20.
Thus, although the occurrence of stressful life events constitutes a challenge for pregnant women, the dimension of individual integration is of paramount importance to regain physiological and/or psychosocial homeostasis. The presence of a medical condition, personality traits, past personal experience, actual vulnerability with the current pregnancy state together with individual or cultural susceptibility, are all factors, which influence the outcome.
Prevention of Preterm Labour
Secondary preventive interventions or programmes aimed at reducing or modifying the risks of preterm labour in high-risk pregnancies have been carried out. Among these, educational programmes aimed at early recognition of preterm labour either by risk scoring systems or antenatal visits, failed to show any benefit in reducing preterm birth rates21. Similar findings were observed for the education of health-care providers21.
Restrictions to activity, such as, bed rest and/or hospitalisation were ineffective21. Social support, especially in disadvantaged populations, was tested in at least 10 randomised controlled trials (n= 8085) and was found to be ineffective (OR: 0.9; 95% CI: 0.8–1.1)22.
The lack of effective prevention programmes directed at high-risk populations is related to several factors. There is some evidence that risk factors for preterm birth, such as previous preterm births, multiple pregnancies and uterine anomalies, cannot be modified. In addition, the underlying cause of preterm birth is still unknown for a fair proportion of preterm births. Finally, only a small fraction of all preterm births occur within the high-risk group, making it impractical to identify a high-risk group, while missing the majority of preterm births with unrecognised risks21.
Primary prevention with population-based strategies have been reported with some success21. However, none of these strategies have been tested with a randomised-controlled trial design, thus limiting their validity. Despite this, community-based strategies, used by Papiernik, with women empowerment together with social advantages and political enforcements, reduced the preterm birth rate by 30% in France and was repeated, with a similar reduction, within two years in Minnesota, USA, by Barbara Yawn21.
Thus, as of yet, no study has addressed specific programmes, apart from education or social support by health-care providers, to socially disadvantaged populations. However, designing a randomised-controlled trial for this population may be more harmful than beneficial; tagging a high-risk population increases anxiety, stress and isolation that we want to avoid.
Management Options for Preterm Labour
When preterm labour is established, a diagnostic work-up should include the assessment of any risk factors that could be modified together with a complete medical history, a thorough physical examination, ultrasound assessment of fetal growth, morphology and well-being, together with infectious screening and blood screening.
In spontaneous labour, without an evident cause, several management options should be considered according to the gestational age, a cervical examination, the state of the membranes together with an assessment of the physical well being of both the mother and her infant.
In some cases, expediting the delivery is indicated to minimise fetal complications if unstable or because the maternal condition is critical. A second option is to allow delivery to occur when cervical dilation is too advanced (>4 cm) although in rare instances, especially in very early gestational ages, tocolysis has demonstrated a prolongation in pregnancy for up to 48 hours, despite a cervical dilation of up to 6 cm, which allows enough time for the administration of corticosteriods23. A third option involves tocolysis to delay delivery in order to take advantage of corticosteroids for 48 hours; in this situation, further prolongation may be considered at earlier gestational ages. This also allows maternal transfer, if appropriate, to a tertiary perinatal centre. Finally, observation for several hours should be considered when no cervical changes are observed or when uterine activity has stopped by itself.
Corticosteroids, for acceleration of pulmonary maturity, should be given for every pregnancy below 34 weeks and it is usual to prescribe antibiotics to mothers in preterm labour to prevent neonatal early-onset Group B streptococcal sepsis.
Towards a Comprehensive Approach to Preterm Labour
Recent advances in potential bio-psychosocial causal pathways to understand the underlying mechanisms responsible for early cervical modifications and preterm uterine activity have renewed interest and enthusiasm to study these mechanisms. A transdisciplinary approach has already brought newer insights, such as inheritance susceptibility, the role of inflammatory processes rather than infection, and the maternal or fetal activation of the hypothalamic–pituitary–adrenal axis, which antedate the initiation of preterm parturition.
Socio-economic and psychosocial factors are now better defined and recent studies have included measures of objective stressor agents and/or subjective ratings of individual perception and impact. Psychologists together with sociologists may help basic and clinical scientists to assess the real impact of these factors and to understand their physiological and psychological response.
Newer directions in ascertaining socio-economic and psychosocial risk factors substantiate the need to pursue research in screening in order to better understand the causal pathways of preterm birth, with the hope that appropriate, and innovative preventive interventions, customised to individuals needs, may prevent preterm births.