Mitigating the stress of transition: An exploration of the effects and effectiveness of a preparatory course for junior obstetric trainees transitioning to senior training roles.

BACKGROUND: Attrition amongst obstetrics trainees is high worldwide and attributed to sources of stress and burnout. The role of formal education and simulation as a means to prepare trainees for stressful periods such as transition into senior roles is underexplored. OBJECTIVE: This study set out to explore whether the creation of a dedicated educational intervention might positively influence burnout and self-estimated preparedness for practice among obstetric trainees transitioning into more senior roles. STUDY DESIGN: A six-week preparatory training programme for year 2 trainees was created specifically for this study. The intervention used the flipped classroom design incorporating online learning that prepared participants for six simulation-based workshops. Participants were randomised by training cluster into an intervention group (n = 4) who participated in the educational intervention and a control group (n = 7) who received standard online and workplace training. The effects on trainee well-being was assessed using the Maslach burnout inventory (MBI) and a self-report questionnaire estimating preparedness for practice. Technical and non-technical skills were assessed using standardised OSAT and NOTSS assessment tools. The primary outcomes were MBI and preparedness for practice scores. Secondary outcomes included OSAT and NOTSS scores. Group comparisons were made using by t-test or Pearson Chi2 analysis where appropriate. RESULTS: The study indicated a positive, non-significant trend in pre-post burnout scores in the intervention group. The following improving trends were noted in all subscales: emotional exhaustion 21.5 ± 2.6 (pre-intervention 23 ± 6.2); depersonalisation 9.8 ± 4.0 (pre-intervention 12.3 ± 2.8); personal accomplishment 35.5 ± 6.51 (pre-intervention 33 ± 5.5). The educational intervention engendered an increase in self estimated preparedness for practice amongst the intervention group (p = 0.006). From a training perspective, increased preparedness was noted for the following practical skills: forceps delivery (p = 0.0001), rotational forceps delivery (p = 0.02), delivery of twins vaginally (p = 0.0007) and performing a pudendal block (p = 0.001). CONCLUSION: This is one of the first studies to investigate whether the provision of a targeted training module can improve burnout scores and preparedness for practice amongst obstetrics trainees at an important time of transition. The positive but largely non-significant findings of this study should be examined in larger longitudinal and adequately powered studies.

The emergence of a change in the prevalence of preeclampsia in a tertiary maternity unit (2004-2016).

BACKGROUND: Pre-eclampsia (PET) affects 2-3% of all pregnancies, rising to 5-7% in nulliparous women. This study aimed to investigate the prevalence of PET over a 13-year period. METHODS: A retrospective review was performed over a 13-year period (2004-2016) via interrogation of the annual clinical reports of The Rotunda Hospital, Dublin. RESULTS: There was a fall in the overall incidence of PET (nulliparous and multiparous), from a peak of 3.8% in 2007 to 1.5% in 2015. Comparing the first and second halves of the study time-period this decrease was statistically significant (p < .0001). In nulliparous women, the thirteen-year mean was 4.4% for the study period, with a similar observed reduction from a peak of 5.3% in 2005 to a trough of 2.4% in 2015. DISCUSSION: In our institution, we have shown a decrease in preeclampsia rates over a 13-year period. While the reason for this trend remains unclear, a similar trend has been observed in another tertiary unit and additional research is required to explain the etiology behind these observations.

Alterations in fibrin formation and fibrinolysis in early onset-preeclampsia: Association with disease severity.

OBJECTIVE: ; Early-onset preeclampsia is a rare pregnancy-specific disorder associated with significantly increased maternal and fetal morbidity and mortality. Whilst it is known that even normotensive pregnancies are associated with changes in clot formation and dissolution, the nature of how these changes differ in those with early onset preeclampsia has not been well established. We sought to evaluate parameters of fibrin formation and fibrinolysis in individuals with early onset preeclampsia in comparison to both pregnant and non-pregnant controls. Furthermore, such parameters were correlated with markers of disease severity in this patient cohort, including the presence of multiorgan involvement, the rate of disease progression and the extent of the anti-angiogenic state in this condition. STUDY DESIGN: ; Patients with early onset preeclampsia (N = 20) and both pregnant (N = 16) and non -pregnant (N = 16) controls were recruited from the cohort at a large urban maternity hospital which saw over 15,000 deliveries during the study period. Platelet poor plasma was prepared from collected whole blood and analysed for parameters of fibrin formation and fibrinolysis (lagtime to and rate of fibrin formation; PAI-1; PAI-2; D-dimer; plasmin-antiplasmin; tPA) in addition to markers of angiogenesis (sFLT-1; Endoglin) using commercially available specific immunoassays. RESULTS: ; The maximum rate of fibrin formation as well as PAI-1, PAI-2 and D-dimer levels were all significantly increased in those with early onset preeclampsia and pregnant controls when compared to non-pregnant controls without significant differences between the 2 former groups. Plasmin-antiplasmin levels were significantly reduced in a similar manner. tPA levels were significantly elevated in EOP compared to both pregnant and non-pregnant controls. EOP was associated with significantly increased anti-angiogenic factors (sFLT-1; Endoglin) when compared to both pregnant and non-pregnant controls. CONCLUSION: ; Markers of fibrin formation and fibrinolysis are significantly alerted in early onset preeclampsia; furthermore, certain markers correlate with disease severity in this patient cohort.

An abnormal cerebroplacental ratio (CPR) is predictive of early childhood delayed neurodevelopment in the setting of fetal growth restriction.

BACKGROUND: Fetal growth restriction accounts for a significant proportion of perinatal morbidity and death. The cerebroplacental ratio is gaining much interest as a useful tool in differentiating the "at-risk" fetus in both fetal growth restriction and appropriate-for-gestational-age pregnancies. The Prospective Observational Trial to Optimize Pediatric Health in Fetal Growth Restriction group has demonstrated previously that the presence of this "brain-sparing" effect is associated significantly with adverse perinatal outcomes in the fetal growth restriction cohort. However, data about neurodevelopment in children from pregnancies that are complicated by fetal growth restriction are sparse and conflicting. OBJECTIVE: The aim of the Prospective Observational Trial to Optimize Pediatric Health in Fetal Growth Restriction NeuroDevelopmental Assessment Study was to determine whether children born after fetal growth-restricted pregnancies are at additional risk of adverse early childhood developmental outcomes compared with children born small for gestational age. The objective of this secondary analysis was to describe the role of cerebroplacental ratio in the prediction of adverse early childhood neurodevelopmental outcome. STUDY DESIGN: Participants were recruited prospectively from the Perinatal Ireland multicenter observational Prospective Observational Trial to Optimize Pediatric Health in Fetal Growth Restriction study cohort. Fetal growth restriction was defined as birthweight <10th percentile with abnormal antenatal umbilical artery Doppler indices. Small for gestational age was defined similarly in the absence of abnormal Doppler indices. Cerebroplacental ratio was calculated with the pulsatility indices of the middle cerebral artery and divided by umbilical artery with an abnormal value <1. Children (n=375) were assessed at 3 years with the use of the Ages and Stages Questionnaire and the Bayley Scales of Infant and Toddler Development, 3rd edition. Small-for-gestational-age pregnancies with normal Doppler indices were compared with (1) fetal growth-restricted cases with abnormal umbilical artery Doppler and normal cerebroplacental ratio or (2) fetal growth restriction cases with both abnormal umbilical artery and cerebroplacental ratio. Statistical analysis was performed with statistical software via 2-sample t-test with Bonferroni adjustment, and a probability value of .00625 was considered significant. RESULTS: Assessments were performed on 198 small-for-gestational-age children, 136 fetal growth-restricted children with abnormal umbilical artery Doppler images and normal cerebroplacental ratio, and 41 fetal growth-restricted children with both abnormal umbilical artery Doppler and cerebroplacental ratio. At 3 years of age, although there were no differences in head circumference, children who also had an abnormal cerebroplacental ratio had persistently shorter stature (P=.005) and lower weight (P=.18). Children from fetal growth restriction-affected pregnancies demonstrated poorer neurodevelopmental outcome than their small-for-gestational-age counterparts. Fetal growth-restricted pregnancies with an abnormal cerebroplacental ratio had significantly poorer neurologic outcome at 3 years of age across all measured variables. CONCLUSION: We have demonstrated that growth-restricted pregnancies with a cerebroplacental ratio <1 have a significantly increased risk of delayed neurodevelopment at 3 years of age when compared with pregnancies with abnormal umbilical artery Doppler evidence alone. This study further substantiates the benefit of routine assessment of cerebroplacental ratio in fetal growth-restricted pregnancies and for counseling parents regarding the long-term outcome of affected infants.

The Impact of Maternal Gestational Hypertension and the Use of Anti-Hypertensives on Neonatal Myocardial Performance.

BACKGROUND: Assessment of myocardial performance in neonates using advanced techniques such as deformation imaging and rotational mechanics has gained considerable interest. The applicability of these techniques for elucidating abnormal myocardial performance in various clinical scenarios is becoming established. We hypothesise that term infants born to mothers with gestational hypertension (GH) may experience impaired performance of the left and right ventricles during the early neonatal period. OBJECTIVES: We aimed to assess left and right ventricular (LV and RV) function using echocardiography in infants born to mothers with GH and compare them to a control group. METHODS: Term infants (>36+6 weeks) born to mothers with GH underwent assessment to measure biventricular function using ejection fraction (EF), deformation imaging, left-ventricle rotational mechanics (apical rotation, basal rotation, twist, twist rate, and untwist rate), and right ventricle-specific functional parameters (tricuspid annular plane systolic excursion and fractional area change) in the first 48 h after birth. A control group comprising infants born to healthy mothers was used for comparison. RESULTS: Fifteen infants with maternal GH and 30 age-matched controls were enrolled. The GH infants exhibited no differences in birthweight or LV or RV length, but they had lower EF (54 vs. 61%; p < 0.01), LV global longitudinal strain (-20 vs. -25%; p < 0.01), and LV twist (11 vs. 16°; p = 0.04). There were no differences in any of the RV functional parameters. CONCLUSION: Infants born to mothers with GH exhibited lower LV function than healthy controls, while RV function appeared to be preserved. This relationship warrants further exploration in a larger cohort.

Early onset preeclampsia is associated with an elevated mean platelet volume (MPV) and a greater rise in MPV from time of booking compared with pregnant controls: results of the CAPE study.

OBJECTIVE: To characterise Mean platelet volume (MPV) in patients with early onset preeclampsia (EOPE) and unaffected controls from time of first antenatal visit until the postpartum. MATERIALS AND METHODS: Retrospective secondary analysis of an observational study in an Irish tertiary referral centre with 9000 deliveries annually. The MPV of 27 women with EOPE was compared to 19 unaffected controls. The inclusion criteria for the disease state was the development of EOPE defined by the National Institute for Health and Care Excellence (NICE) guideline, as new onset hypertension presenting after 20 weeks and prior to 34 weeks with significant proteinuria. Between October 2013 and July 2015 we recruited 27 women with EOPE and 19 pregnant controls. Statistical analysis was performed using paired T-test of Mann-Whitney test where appropriate and a P-value <0.05 was deemed significant. RESULTS: At time of diagnosis and late in the third trimester MPV was significantly increased to 9.0 (±0.3) fL in cases of EOPE in comparison to 8.5 (±0.6) fL in normotensive controls (P<0.05). There was no significant difference during the first trimester or postpartum when comparing the MPV in EOPE to controls. CONCLUSION: Despite an increased MPV at time of diagnosis of EOPE this study did not demonstrate a potential use for increased MPV as a first trimester screening tool.

Non-invasive cardiac output monitoring (NICOM®) can predict the evolution of uteroplacental disease-Results of the prospective HANDLE study.

OBJECTIVES: We aimed to firstly identify the different haemodynamic profiles amongst nulliparous women who develop either gestational hypertension (GH), pre-eclampsia (PE), normotensive fetal growth restriction (FGR) versus unaffected pregnancies using non-invasive cardiac output monitoring (NICOM®). Our second primary objective was to assess the ability of NICOM® derived variables to predict the evolution of PE, GH and FGR. STUDY DESIGN: Low risk nulliparous women were enrolled in a single center prospective observational study. NICOM® assessments were performed at 14, 20 and 28 weeks' gestation and data was obtained on cardiac output (CO), total peripheral resistance (TPR), indexed TPR (adjusted for maternal body surface area; TPRi), stroke volume (SV), indexed SV (adjusted for maternal body surface area; SVi) and heart rate (HR). Logistic regression was used to model GH, PE and FGR with NICOM® measurements as predictors. Linear, non-linear and interaction terms were assessed using the Akaike Information Criterion. RESULTS: The haemodynamic profile of pregnancies complicated by uteroplacental disease- GH (n=18), PE (n=6) and FGR (n=24) were compared to 318 healthy unaffected pregnant controls. Women with evolving PE have a different haemodynamic profile to those developing either GH or FGR. The best independent predictors for the evolution of uteroplacental disease at 14 weeks' gestation were CO in the prediction of FGR (AUC=0.61; p 0.002), TPR in the prediction of GH (AUC=0.63; p<0.02) and SVi in the prediction of PE (AUC=0.62; p<0.05). The performance of haemodynamic variables was enhanced when combined in a multivariate logistic model. We demonstrated that TPR, CO and SV when combined with BP were significant predictors of pregnancies complicated by FGR (AUC=0.64, p=0.004; AUC=0.65, p=0.004; and AUC=0.65, p=0.007 respectively). Whereas in pregnancies complicated by PE, HR and SVi in combination with BP were also statistically significant predictors (AUC=0.75, p=0.017 and AUC=0.77, p=0.007 respectively). CONCLUSIONS: NICOM® derived maternal haemodynamic profile at 14 weeks' gestation has the novel potential to identify pregnancies which will ultimately develop uteroplacental disease.

Elevated plasma TFPI activity causes attenuated TF-dependent thrombin generation in early onset preeclampsia.

Early onset preeclampsia (EOP) is a pregnancy-specific proinflammatory disorder that is characterised by competing thrombotic and bleeding risks. It was the aim of this study to characterise thrombin generation, a major determinant of thrombotic and bleeding risk, in order to better understand the haemostatic balance in patients with EOP. Patients with EOP were recruited at the Rotunda Hospital, Dublin. Twenty-six cases of EOP were recruited over a 21-month period, out of 15,299 deliveries at the Rotunda. Blood samples were collected into sodium citrate plus corn trypsin inhibitor anticoagulated vacutainers, platelet-poor plasma was prepared, and calibrated automated thrombography was used to assess thrombin generation. Results were compared to age and sex-matched non-pregnant controls (n=13) and age- and gestation-matched pregnant controls (n=20). The rate and extent of thrombin generation triggered by low-dose tissue factor (TF) was significantly reduced in patients with EOP compared to pregnant controls, most significantly in cases of severe EOP. EOP patients displayed a trend towards an increased response to endogenous activated protein C and thrombomodulin relative to pregnant controls. Plasma tissue factor pathway inhibitor (TFPI) activity was increased in EOP patients. Inhibition of TFPI abolished the attenuation of thrombin generation stimulated by low-dose TF. In conclusion, patients with EOP are characterised by an attenuated coagulation response characterised by reduced thrombin generation stimulated by low-dose TF and elevated plasma TFPI activity. These changes in coagulation may modulate thrombotic risk and bleeding risk in patients with EOP.

Left ventricular rotational mechanics in infants with hypoxic ischemic encephalopathy and preterm infants at 36 weeks postmenstrual age: A comparison with healthy term controls.

BACKGROUND AND AIMS: There is a paucity of data on left ventricle (LV) rotational physiology in neonates. We aimed to assess rotational mechanics in infants with hypoxic ischemic encephalopathy (HIE) and premature infants (<32 weeks) at 36 weeks postmenstrual age (PMA) (preterm group) and compare them with healthy term controls (term controls). We also compared the parameters in preterm infants with and without chronic lung disease (CLD). METHODS: Echocardiography was performed within 48 hours of birth or at 36 weeks PMA. LV basal and apical rotation, twist (and torsion=twist/LV length), twist rate (LVTR), and untwist rate (LVUTR) were measured. One-way ANOVA was used to compare values. RESULTS: There was no difference in gestation (40.0 [39.1-40.3] vs 39.9 [39.0-40.9], P>.05) or birthweight (3.7 [3.4-4.1] vs 3.5 [3.2-3.9], P>.05) between the HIE group (n=16) and term controls (n=30). The preterm group (n=35) had a gestation and weight of 36.0 [34.6-36.3] weeks and 2.3 [2.0-2.4] kg. The HIE group had lower twist, torsion, LVTR, and LVUTR than the other two groups. The preterm group had a more negative (clockwise) basal rotation while the term group had a more positive (counterclockwise) apical rotation. Preterm infants with CLD had higher apical rotation, twist, and torsion when compared to infants without CLD. CONCLUSION: Infants with HIE have reduced rotational mechanics. Preterm infants at 36 weeks PMA have comparable measurements of twist to term infants. This is achieved by predominant basal rather than apical rotation. Infants with CLD have increased apical rotation.

Evaluation of normalization of cerebro-placental ratio as a potential predictor for adverse outcome in SGA fetuses.

BACKGROUND: Intrauterine growth restriction accounts for a significant proportion of perinatal morbidity and mortality currently encountered in obstetric practice. The primary goal of antenatal care is the early recognition of such conditions to allow treatment and optimization of both maternal and fetal outcomes. Management of pregnancies complicated by intrauterine growth restriction remains one of the greatest challenges in obstetrics. Frequently, however, clinical evidence of underlying uteroplacental dysfunction may only emerge at a late stage in the disease process. With advanced disease the only therapeutic intervention is delivery of the fetus and placenta. The cerebroplacental ratio is gaining much interest as a useful tool in differentiating the at-risk fetus in both intrauterine growth restriction and the appropriate-for-gestational-age setting. The cerebroplacental ratio quantifies the redistribution of the cardiac output resulting in a brain-sparing effect. The Prospective Observational Trial to Optimize Pediatric Health in Intrauterine Growth Restriction group previously demonstrated that the presence of a brain-sparing effect is significantly associated with an adverse perinatal outcome in the intrauterine growth restriction cohort. OBJECTIVE: The aim of the Prospective Observational Trial to Optimize Pediatric Health in Intrauterine Growth Restriction study was to evaluate the optimal management of fetuses with an estimated fetal weight <10th centile. The objective of this secondary analysis was to evaluate if normalizing cerebroplacental ratio predicts adverse perinatal outcome. STUDY DESIGN: In all, 1116 consecutive singleton pregnancies with intrauterine growth restriction completed the study protocol over 2 years at 7 centers, undergoing serial sonographic evaluation and multivessel Doppler measurement. Cerebroplacental ratio was calculated using the pulsatility and resistance indices of the middle cerebral and umbilical artery. Abnormal cerebroplacental ratio was defined as <1.0. Adverse perinatal outcome was defined as a composite of intraventricular hemorrhage, periventricular leukomalacia, hypoxic ischemic encephalopathy, necrotizing enterocolitis, bronchopulmonary dysplasia, sepsis, and death. RESULTS: Data for cerebroplacental ratio calculation were available in 881 cases, with a mean gestational age of 33 (interquartile range, 28.7-35.9) weeks. Of the 87 cases of abnormal serial cerebroplacental ratio with an initial value <1.0, 52% (n = 45) of cases remained abnormal and 22% of these (n = 10) had an adverse perinatal outcome. The remaining 48% (n = 42) demonstrated normalizing cerebroplacental ratio on serial sonography, and 5% of these (n = 2) had an adverse perinatal outcome. Mean gestation at delivery was 33.4 weeks (n = 45) in the continuing abnormal cerebroplacental ratio group and 36.5 weeks (n = 42) in the normalizing cerebroplacental ratio group (P value <.001). CONCLUSION: The Prospective Observational Trial to Optimize Pediatric Health in Intrauterine Growth Restriction group previously demonstrated that the presence of a brain-sparing effect was significantly associated with an adverse perinatal outcome in our intrauterine growth restriction cohort. It was hypothesized that a normalizing cerebroplacental ratio would be a further predictor of an adverse outcome due to the loss of this compensatory mechanism. However, in this subanalysis we did not demonstrate an additional poor prognostic effect when the cerebroplacental ratio value returned to a value >1.0. Overall, this secondary analysis demonstrated the importance of a serial abnormal cerebroplacental ratio value of <1 within the <34 weeks' gestation population. Contrary to our proposed hypothesis, we recognize that reversion of an abnormal cerebroplacental ratio to a normal ratio is not associated with a heightened degree of adverse perinatal outcome.

Prenatal detection of major congenital heart disease - optimising resources to improve outcomes.

INTRODUCTION: Congenital heart disease (CHD) is the most common major structural fetal abnormality and the benefits of prenatal detection are well described. The objective of this study was to evaluate the precision of prenatal diagnosis at a single tertiary referral unit over two three year periods (2006, 2007, 2008 and 2010, 2011, 2012), before and after a prenatal screening protocol for CHD was developed to include extended cardiac views, mandatory recall for suboptimal views, and a multidisciplinary Fetal Cardiac clinic was established. There exists a single national centre for paediatric cardiothoracic surgery in Ireland, a situation which facilitates near complete case ascertainment. MATERIALS AND METHODS: Surgery records of the National Children's Cardiac Centre were interrogated for all cases of major congenital heart defects requiring surgical intervention in the first six months of life. Minor procedures such as ligation of a patent ductus arteriosus and isolated atrial septal defect repairs were excluded. Analyses of the Fetal Medicine database at the Rotunda Hospital (a stand-alone tertiary level perinatology centre with 8500 deliveries per year) and the mortality data at the Perinatal Pathology department were conducted. The Cochrane-Armitage trend test was used to determine statistical significance in prenatal detection rates over time. RESULTS: 51,822 women delivered during the study period, and the incidence of major congenital heart disease either that underwent surgical intervention or that resulted in perinatal mortality, was 238/51,822 (0.5%). Prenatal detection of major CHD increased from 31% to 91% (p<0.001). Detection of critical duct-dependant lesions rose from 19% to 100%. CONCLUSION: We attribute the dramatic improvement in prenatal detection rates to the multifaceted changes introduced during the study period. Improved prenatal detection for births that are geographically remote from the National Paediatric Cardiac Centre will require local replication of this prenatal programme.

Hepatitis C virus-associated thrombocytopenia in pregnancy: impact upon multidisciplinary care provision.

OBJECTIVE: Recent studies have implicated hepatitis C virus (HCV) in the pathogenesis of immune thrombocytopenia. In pregnancy-associated immune thrombocytopenia, multidisciplinary management is required due to a potential for bleeding complications. We performed a retrospective review of HCV-infected pregnant women and age-matched controls who were not infected with HCV. METHODS: One hundred and six women with a HCV viral load were identified from 2009 to 2011. RESULTS: Thrombocytopenia was identified in 10.3% of HCV-infected pregnant women and 1.6% of age-matched controls (P<0.001). Mean platelet count during pregnancy was 120 ± 23 × 109/L in HCV-infected women and at delivery was significantly lower in HCV-infected women than in controls (P=0.01). Despite the significant difference in platelet counts, there was no significant difference in estimated blood loss (EBL) at delivery. Regional anaesthesia was performed in 73% of thrombocytopenic HCV-infected women and no complications were recorded. There were no fetal bleeding complications. CONCLUSION: In the first study to date to investigate the impact of HCV on maternal platelet count we demonstrated a significantly higher frequency of thrombocytopenia and a significantly lower platelet count in HCV-infected pregnant women compared with controls. Interestingly, thrombocytopenia had no detectable impact on EBL at delivery.

Real increasing incidence of hysterectomy for placenta accreta following previous caesarean section.

OBJECTIVE: Placenta accreta, morbid adherence to the uterus to the myometrium, is commonest in association with placenta previa in women previously delivered by caesarean section (CS). It has become proportionally a greater cause of major maternal morbidity and mortality as the frequency of other serious obstetric complications has declined. The aim of this study was to examine the incidence of placenta accreta in the context of a rising caesarean delivery rate. STUDY DESIGN: Retrospective review of the incidence of placenta accreta in parous women during the 36 years 1975-2010. Cases were identified from hospital records and then correlated with pathological reports. The incidence of placenta accreta was analysed in the context of women previously delivered by CS. RESULTS: During the 36-year period in our unit, 157,162 multiparous women delivered, of whom 15,151 (9.6%) had a previous CS scar. The institutional incidence of CS rose from 4.1% in 1975 to 20.7% in 2010. Twenty-five parous women, all with a previous CS, had placenta accreta requiring hysterectomy. The overall incidence of placenta accreta was 1.65 per 1000 parous women with a previous CS, but was low (1.06/1000) until 2002. From 2003 to 2010 the incidence rose to 2.37/1000 previous CS deliveries (OR 2.2; 95% CI 1.05-5.1). CONCLUSION: The frequency of placenta accreta correlated steadily with the CS rate until 2000. Since then, the incidence has nearly doubled in women with previous CS scars, suggesting an additional causative influence on risk.