Www.bournemouth.ac.uk 1 DCP Annual Conference, 3 rd December 2015. London. Dr. Carol Wilkins, Lead Midwife for Education. Faculty of Health and Social.

www.bournemouth.ac.uk 1

DCP Annual Conference, 3rd December 2015. London.

Dr. Carol Wilkins, Lead Midwife for Education.Faculty of Health and Social Sciences, Bournemouth University. Professor Roger Baker, Bournemouth University

Professor Debra Bick, Kings College, London

Professor Peter Thomas, Bournemouth University

Does poor emotional processing predict the development of postnatal depression?Findings from the Emotional Processing in Childbirth Study

The EPiC Study


Emotions in pregnancy

• Childbirth continuum -engenders complex range of positive and negative emotions

• Emotions triggered by • changes in role/lifestyle • physical and psychological pregnancy specific stimuli• ‘normal’ life stressors

• Yet no studies have explored how the management of these emotions impacts on perinatal mental health


Predicting PND

Postnatal depression (PND) – serious public health concern impacting on whole family

3 meta-analyses comprising 100 international studies (approx 24,00 women) (O'Hara and Swain 1996; Beck 2001; Robertson et al.

2004). have identified the strongest predictors of postnatal depression as being :

• Strong - depressed mood antenatally, history of depression, perceived low levels of social support, life stresses

• Medium - low self-esteem and poor marital relationship • Low - socioeconomic status and obstetric factors


What is Emotional Processing?

• Emotional Processing (EP) describes the way people deal with the feelings/emotions caused by stressful events in their lives.

• Effective EP is achieved when emotions are processed in such a way that they do not impact on a person’s ability to continue with their everyday lives (Rachman 2001).


Aims of study

•To investigate the possibility of predicting postnatal depression from scores on the Emotional Processing Scale (in conjunction with other identified risk factors for postnatal depression).

•To examine the relationship between emotional processing in pregnancy and the development of postnatal depression (in conjunction with other identified risk factors for postnatal depression)


Methods

• Approach - Prospective longitudinal cohort study• Setting/ recruitment– Hospital Trust in the South of England• Participants - Cohort of 974 pregnant women, aged 16 to 44 – recruited

at first antenatal screening appointment at 13 weeks (between Nov. 2007 and Feb. 2009)

• Data collection – Questionnaires given personally at 13 weeks and postal questionnaires sent at 34 weeks gestation and 6 weeks postpartum.

• Outcome measures – validated tools - EPS, EPDS, SF-36, RSE• Data analysis - SPSS version 16 – independent samples t-test, one way

ANOVA, repeated measures ANOVA, multiple and binary logistic regression modelling

• Ethics – Approval from Local Research Ethics Committee and clinical governance department of hospital Trust


Questionnaires

• EPS

• Edinburgh Postnatal Depression Scale (EPDS)¹

• Short Form-36 (SF 36)²

• Rosenberg Self-Esteem Scale (RSE)³

• Practical and emotional support

• Life stresses

25-item self-report scale

10-item self-report scale .

36-item generic measure of 8 domains of positive and negative physical and mental health.

10-item self-report scale

….perceived from partners, family, friends

…..during the last year

¹Cox et al.1987, ²Ware and Sherbourne 1992, ³Rosenberg 1989


Questionnaires

• Questionnaire 1:• Demographics - age, occupation, parity, marital status, ethnicity,

past/current psychiatric history, family mental health history, current medical history

• Questionnaire 2: • Health during pregnancy, GP or hospital in-patient treatment

• Questionnaire 3:• Birth experiences and care, feeding choices, postnatal health


The EPiC Study

FINDINGS


Response rates

Questionnaire 1:•1333 women agreed to participate•974 women completed and returned Q1- sample

Questionnaire 2:•75% (n=713) responded•23% non-return (n=243) (remained in study)•2% withdrawn

Questionnaire 3:•57% (n=554) of original cohort responded•876 distributed•63% returned

•53% (n=520) returned all three questionnaires completed


Demographics Variable Number % Total responses

Parity Primiparous 460 47.6

966Multiparous 506 52.3

Age

19 and under 49 5.1

965

20-24 117 12.1

25-29 260 26.9

30-34 304 31.5

35 and over 235 24.3

Partner Has partner 947 98.0

966No partner 19 2.0

Relationship with

partner

Good 937 99.5

942Not good 5 0.5

Lives with…..

Partner 876 90.7

965Alone 37 3.8

Parents or relation 52 5.3

Stressful life events

in preceding 12

months

Divorce 23 2.3

963

Death of loved one 163 16.9

Moved house 286 29.6

New job 221 22.9

Chronic illness 90 9.3


Socioeconomic statusOccupation Number %

Higher managerial/professional (e.g. doctors, lawyers, dentists) 22 2.3Lower managerial (e.g. teachers, nurses, journalists) 149 15.6

Intermediate occupations (e.g. health care assistants, secretaries) 219 22.9Small employers (e.g. hairdressers) 22 2.3Lower supervisory and technical (e.g. supervisors, foreman) 130 13.5Semi-routine (e.g. shop assistant, call centre workers, care assistants) 142 14.8Routine (e.g. waitresses, cleaners, bus drivers) 49 5.1Never worked, long term unemployed 21 2.2Not classified ( incl. students, housewives, insufficient information) 204 21.3

Total 958 100

Missing information 16


EP in pregnancy and postpartum

• Mean EPS scores improved over time – statistically significant. (2.72; 2.62; 2.38)

• Greatest increase in scores (worsening of EP) between early and late pregnancy (22.7%). Greatest decrease (improvement in EP) between early pregnancy and postpartum (24.7%).

• Significantly higher EPS scores in pregnancy found in:• Younger maternal age groups (19 years and under, 20-24 years)

• Those with past mental health history

• Those with current mental health problems

• Those without a partner

• Higher pregnancy and postpartum EPS scores found in:• Multiparous women with a history of postnatal depression

• Family history of depression

• Parity, physical health and SES made no significant difference to EP


Relationship between poor EP and likelihood of postnatal depression

• Significantly high positive correlations between EPS and EPDS at each time point (p<0.001)

• Strong positive correlations between EPS 1 scores and EPDS 3 and between EPS 2 and EPDS 3

• Significant difference of 1.8 in mean EPS 1 scores between women who scored above (n= 76) and below (n=468) threshold in EPDS 3. (95% CI 1.4 to 2.2, t-9.5, p<0.001)

EPDS and EPS scores dichotomised into high and low:• Significant difference of 2.2 in mean EPS 2 scores between those scoring high (n=

72) and low (n= 453) on the EPDS 3. (95% CI 1.8 to 2.6, t -10.6, p<0.001)

• 40% of women(n=30) with high EPS 1 scores had correspondingly high EPDS 3 scores, compared with 10% (n = 46) of women with low EPS 1 scores who had correspondingly high EPDS 3 scores

• 50% of women with high EPS 2 scores (n=80) had correspondingly high EPDS 3 scores.


Prediction of PND in early pregnancy

Multiple regression modelling performed:

Model 1: Four modifiable early pregnancy variables made a contribution to prediction of depression:•EPDS 1 strongest (β = 0.21, t=3.08, p = 0.002, 95% CI 0.36 to 2.34)

•EPS 1 next strongest (β = 0.19, t = 3.13, p = 0.002, 95% CI 0.07 to 0.33)

•Past history of depression•Physical wellbeing in early pregnancy

Model 2: adding variables associated with birth experience…..•Significant contributions to prediction of PND in order of strength – EPDS 1, EPS 1, satisfaction with birth experience, feeding difficulties, past history of depression, physical wellbeing.


Prediction of PND in early pregnancy

After adjusting for all other risk factors for PND regression modelling predicted that:•for every 1 unit increase in EPS 1 scores there would be an average increase of 0.2 in mean EPDS 3 scores (p = 0.002, B = 0.6)

With variables associated with the birth experience added regression modelling predicted that:•for every increase of 1 unit in EPS 1 scores there would be an average increase of 0.6 in mean EPDS 3 scores (p = 0.001, B= 0.58)


Prediction of PND in late pregnancy

6 late pregnancy variables made a contribution to the prediction of PND:•EPS 2 strongest predictor (β = 0.29, t = 5.08, p <0.001, 95% CI 0.5 to 1.13)

•Poor self esteem•Poor practical support from partner•Poor mental wellbeing (MCS 2)•New job•Moving house

For every 1 unit increase in EPS 2 scores there would be a predicted average increase of 0.8 in mean EPDS 3 scores (p <0.001, B = 0.82)

BUT – if depression added to model only partner support remained significant with late pregnancy depression the strongest predictor


Prediction of PND from EPS sub-scales

In early pregnancy -2 sub-scale variables made statistically significant contribution to prediction of EPDS 3 scores:•Unregulated emotions - strongest (β = 0.17, t=2.7, p=0.0007, 95% CI 0.1to 0.8)

•Suppression (β=0.13, t=2.16, p = 0.31, 95%CI0.03 to 0.61)

Late pregnancy – 2 sub-scale variables made statistically significant contribution to prediction of EPDS 3 scores•Unprocessed emotions - strongest (β=0.22, t=2.9, p=0.003, 95% CI 0.2 to 0.8)

•Unregulated emotions (β =0.18, t=2.6, p=0.009, 95% CI 0.1 to 8.2)


Odds of high EPS 1 scores predicting high EPDS 3 scores

3 early pregnancy variables made a significant contribution to prediction of PND:•High EPS 1 scores•Poor physical wellbeing•Low self-esteem

With birth experiences added – dissatisfaction with birth experience became strongest predictor and feeding problems was also significant predictor

After adjusting for other significant variables the odds of women with high EPS scores in early pregnancy developing PND were 2.7 times greater than women with low EPS scores. (Exp(B) = 2.7, 95% CI 1.4 to 5.3, p = 0.004). Sensitivity 20%; specificity 99%; ppv 68%.

BUT – EPS 1 no longer predictive when EPDS 1 added to model


Odds of high EPS 2 scores predicting high EPDS 3 scores

Only 2 late pregnancy variables made a significant contribution to prediction of PND: •High mean EPS 2 •High mean EPDS 2 scores

With birth experiences added EP became strongest predictor followed by dissatisfaction with birth experience and depression in late pregnancy

After adjusting for effects of variables in late pregnancy the odds of women with high EPS scores in late pregnancy developing PND were 6 times greater than women with low EPS scores (Exp (B) = 6.1, 95% CI 2.9 to 12.9, p <0.001).


Summary of findings

• After adjusting for other variables identified as risk factors for PND, poor EP in early and late pregnancy significantly predicted the likelihood of PND

• The odds of developing PND were 2.7 times greater in women with high EPS 1 scores than in those with low EPS 1 scores (in the absence of antenatal depression in early pregnancy)

• The odds of developing PND were 6 times greater in women with high EPS 2 scores than in women with low EPS 2 scores (even with antenatal depression in late pregnancy)


Implications for practice

Understanding EP in pregnancy and its interaction with other recognised risk factors is valuable in planning appropriate support for perinatal emotional health needs

•Reduction in postnatal care in UK – less opportunity to support emotional difficulties. Pregnancy is ideal time to assess women’s EP and initiate timely support which might subsequently reduce the risks of postnatal depression.

•Need to explore resource effective ways to integrate a supportive structure of emotion management into existing and proposed framework of antenatal care


Further research

•RCTs to explore whether intervention strategies to manage EP antenatally can be successful in reducing the incidence of depression.

•Exploration of whether EPS as a screening intervention would prove socially, psychologically and economically effective and safe for the population of pregnant women in the UK – necessary to meet the rigorous criteria laid down by the UK National Screening Committee


Thank you

[email protected]

Tel: 01202 968317


References and additional reading

Baker, R., Thomas, S., Thomas, P. W., and Owens, M., 2007. Development of an emotional processing scale. Journal of Psychosomatic Research, 62 (2), 167-178.Baker, R., Thomas, S., Thomas, P. W., Gower, P., Santonastaso, M., and Whittlesea, A., 2010. The emotional processing scale: Scale refinement and abridgement (EPS-25). Journal of Psychosomatic Research, 68 (1), 83-88.Beck, C., 2001. Predictors of postpartum depression. Nursing Research, 50 (5), 275-284.Cox, J. L., Holden, J. M., and Sagovsky, R., 1987. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. British Journal of Psychiatry, 150, 782-786.Da Costa, D., Larouche, J., Dritsa, M., and Brender, W., 1999. Variations in stress levels over the course of pregnancy: Factors associated with elevated hassles, state anxiety and pregnancy-specific stress. Journal of Psychosomatic Research, 47 (6), 609-621.DiPietro, J. A., Ghera, M. M., Costigan, K., and Hawkins, M., 2004. Measuring the ups and downs of pregnancy stress. Journal of Psychosomatic Obstetrics and Gynecology, 25 (3/4), 189-201.Lobel, M., Cannella, D. L., Graham, J. E., Devincent, C., Schneider, J., and Meyer, B. A., 2008. Pregnancy-specific stress, prenatal health behaviors, and birth outcomes. Health Psychology, 27 (5), 604-615.Lothian, S. 2002. Emotional processing deficits in colorectal cancer : A theoretical overview and empirical investigation. Thesis (PhD). Southampton: University of Southampton.


References and additional reading

National Institute for Health and Clinical Excellence. 2007. Antenatal and postnatal mental health: Clinical management and service guidelines. NICE clinical guideline 45. London: National Institute for Health and Clinical Excellence. O’Hara, M. W., and Swain, A. M., 1996. Rates and risk of postpartum depression: A meta-analysis. International Review of Psychiatry, 8 (1), 37.Rachman, S., 2001. Emotional processing, with special reference to post-traumatic stress disorder. International Review of Psychiatry, 13 (3), 164-171.Raleigh, J., 2004. A preliminary comparative study of emotional processing in women with fybromyalgia syndrome, rheumatoid arthritis and healthy subjects. Thesis (MSc). University of Southampton . Robertson, E., Grace, S., Wallington, T., and Stewart, D. E., 2004. Antenatal risk factors for postpartum depression: A synthesis of recent literature. General Hospital Psychiatry, 26 (4), 289-295.Rosenberg, M., 1989. Society and the adolescent self-image. Revised Edition ed. Middeltown, CT.: Wesleyan University Press.Ware, J., E, and Sherbourne, C., D. 1992. The MOS 36-item Short-Form Health Survey (SF-36). 1. Conceptual framework and item selection. Medical Care, 30 (6), 473-483.Wilkins, C. 2012. Emotional Processing in Childbirth. A longitudinal study of women’s management of emotions during pregnancy and the association with postnatal depression. Thesis (PhD) Bournemouth University

Www.bournemouth.ac.uk 1 DCP Annual Conference, 3 rd December 2015. London. Dr. Carol Wilkins, Lead Midwife for Education. Faculty of Health and Social.

Documents