SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Objective To investigate the relationship between maternal physical and emotional health problems six to nine months after childbirth.

Design Statewide postal survey, incorporating the Edinburgh Postnatal Depression Scale, distributed to women six to seven months after childbirth, with telephone interview follow up of a sub-sample of participants at seven to nine months postpartum.

Participants The postal survey was distributed to all women who gave birth in a two-week period in Victoria, Australia in September 1993, except those who had a stillbirth or known neonatal death. Follow up interviews were conducted with respondents to the postal survey who provided contact details and expressed interest in participating in further research selected to recruit three groups of equivalent size according to scores on the Edinburgh Postnatal Depression Scale: a low scoring group (< 9); a borderline group (9–12); and a group with scores indicating probable depression (> 13).

Results The response rate to the postal survey was 62.5% (n= 1336). Respondents were representative of the total sample in terms of mode of delivery, parity and infant birthweight; young women, single women and women of non-English speaking background were under-represented. The participation rate in telephone follow up interviews was: 89.1% (n= 204), comprising 66 women with an Edinburgh Postnatal Depression Scale score of < 9; 72 women scoring 9–12; and 66 scoring ≥ 13. The point prevalence of depression at six to seven months postpartum was 16.9% (225/1331, 95% CI 14.9–18.9%). Physical and related health problems associated with significantly increased odds of (OR 3.42 [2.2–5.3]); urinary incontinence (OR 2.23 [1.5–3.4]); back pain (OR 2.20 [1.6–3.0]); sexual problems (OR 2.16 [1.6–3.0]); more coughs, colds and minor illnesses than usual (OR 2.78 [1.9–4.1]); bowel problems (OR 1.93 [1.3–2.9]) and relationship difficulties (OR 3.88 [2.8–5.4]). At follow up, three physical health factors were associated with statistically significant linear trends with poorer levels of emotional wellbeing. These were: tiredness (χ2 for linear trend = 12.38, P < 0.001); urinary incontinence (χ2 for linear trend = 5.63, P= 0.02); and more minor illnesses than usual (χ2 for linear trend = 7.94, P= 0.005).

Conclusion The study confirms a link between maternal emotional wellbeing and physical health and recovery in the postnatal period and has important implications for clinical practice. Strategies for encouraging greater disclosure of physical and emotional health problems, assessment of the physical health of recent mothers presenting with depression, and emotional health of recent mothers presenting with other morbidity should be high priorities for all health professionals in contact with mothers in the year following childbirth.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Links between depression and acute and chronic illness, especially when this involves major surgery, are well recognised in both the general medical and psychiatric literature1,2. It is, therefore, somewhat surprising that the role of physical health and recovery after childbirth as contributing factors to maternal depression has been so poorly researched. Prevalence studies have focused on a long list of potential contributing factors to maternal depression including obstetric events, family history of psychiatric illness, personality factors, sociodemographic characteristics, life events and social support3, overlooking women's postpartum physical health almost entirely.

Recently attention has been drawn to the extent of maternal physical morbidity after childbirth, and the under-recognition of many common problems in primary care4–10. This paper draws on a large population-based survey of recent mothers conducted in Victoria, Australia and a follow up study using telephone interviews at seven to nine months postpartum to explore the relationship between maternal physical health problems and depression at six to seven months postpartum.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Postal survey

All maternity hospitals and home birth practitioners in the state of Victoria, Australia were asked to facilitate the mailing of the questionnaires, six to seven months post-natally, to women who gave birth in a two-week period in September 1993, excluding those who had a stillbirth or whose baby was known to have died. All but three hospitals (of 133 maternity hospitals in Victoria) agreed to participate. Ethics approval for the postal survey and for follow up interviews at seven to nine months postpartum was obtained from 25 hospitals with individual ethics committees, and from the research and ethics committees of La Trobe and Monash Universities.

The postal survey was designed to obtain feedback from recent mothers about their experiences of maternity care, as well as information about a range of maternal and infant health outcomes. Data on sociodemographic characteristics, past reproductive history, events in the current pregnancy, physical health problems and emotional wellbeing after the birth, and on use of primary health care services were collected.

Data on self-reported morbidity were obtained from responses to a question which asked “Thinking about your own health and how you have been feeling since the birth, has any of the following been a problem for you?” A precoded list of health issues included both physical health problems, and items for ‘family planning/contraception’, ‘relationship with your husband/partner’ and ‘sexual problems’. Maternal emotional wellbeing was assessed using the Edinburgh Postnatal Depression Scale (EPDS), a 10-item scale which has been found to have a good sensitivity and specificity assessed against psychiatric diagnosis in three settings in the United Kingdom11–13 and in an Australian validation14,15. A score of ≥ 13 on this scale was regarded as indicative of probable depression.

Three mailings took place at two week intervals. Each one included a covering letter inviting women to participate in the study, a copy of the questionnaire, a brief explanation of the study in six community languages and a reply paid return envelope. The survey form was not made available in translation, as this strategy was found to be unsuccessful in a postal survey conducted in 198916.

Telephone interviews

A sub-sample of women completing the main postal survey was invited to take part in a follow up telephone interview at seven to nine months postpartum. The follow up study was known as the Life as a Mother Project. The purpose of these interviews conducted in this project was to investigate the early postnatal experiences, physical health, emotional wellbeing and use of health services of mothers in the first nine months postpartum. The method of sampling and content of the interview schedule were specifically designed to explore how physical health and recovery after birth, and access to social and community support impacts on emotional wellbeing.

Women invited to participate in follow up interviews were drawn from respondents to the postal survey who had indicated an interest in taking part in further research. A research officer not directly involved in the study (F.B.) coded the EPDS as soon as the questionnaires were received. The method of selection involved taking every alternate questionnaire received from women happy to take part in further research (i.e. those providing their name and contact phone number on the questionnaire), and entering individual identification numbers and contact details into a separate register. Another separate listing was used to record EPDS scores alongside each woman's identification number.

The aim was to recruit three groups of equivalent size, categorising women according to their scores on the EPDS:

  • 1
    A low scoring group (EPDS < 9).
  • 2
    A group with scores borderline for depression (EPDS 9–12).
  • 3
    A group with scores indicating probable depression (EPDS ≥ 13).

A record was kept of all women approached to take part, and of completed interviews. This record was checked weekly by F.B. in order to maintain an up-to-date listing of the numbers of completed interviews within each EPDS sub-group. Interviewing was halted when a minimum of 66 interviews had been completed in each category.

The interviews were conducted by the authors and another experienced researcher (M.M.) all of whom were blinded to women's scores on the EPDS. Women were contacted by phone and invited to take part in a telephone interview about their experiences since the birth. Those interested in taking part were mailed an information sheet outlining the purpose and nature of the study, and a consent form. An appointment for the interview was made one to two weeks in advance to allow time for women to receive and mail back the consent forms.

Analysis

Data from the postal survey and the telephone interviews were analysed using SPSS for Windows17 and the Statcalc component of Epi Info Version 618. Statistical analysis of the postal survey involved χ2 comparisons and calculation of odds ratios. The telephone interview data were analysed using χ2 for linear trends as a test for trend across the three categories of EPDS scores (< 9/9−12/≥ 13). Power calculations based on the final sample size of 204 (with group sizes of 66, 72 and 66, respectively) indicated that for comparison of any two of the groups with respect to dichotomous outcomes, the sample size was adequate to detect an odds ratio of three for common exposures (alpha = 0.05, beta = 0.20). Where there are trends across the three groups (nondepressed/ borderline/ depressed), the power of the study is increased and smaller odds ratios of two may be detected. For ordered categorical outcomes the power is similarly increased and the sample size is adequate for the identification of smaller associations19.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Postal survey

A total of 1366 completed questionnaires were returned. Twenty-four of these were excluded because the baby's birth date fell outside the study period. Five duplicate questionnaires, and a questionnaire returned by a woman who had had a stillbirth were also excluded. Another 86 questionnaires were returned unable to be delivered at the mailing address. Excluding these questionnaires from the denominator, the response rate to the survey was 62.5% (1336/2138).

A comparison of respondents with all women who gave birth in the study period using the state's perinatal data collection (excluding those who gave birth at non-participating hospitals and women whose infant was stillborn or known to have died in the neonatal period, estimated to be < 9%) showed that the respondents were largely representative in terms of important obstetric characteristics: the number of previous live births, birth-weight of infants in the current pregnancy and method of birth. The following groups of women were under-represented: women born overseas of non-English speaking background, single women and women aged < 25 years. A detailed description of the study sample and methods is given elsewhere16.

Point prevalence of depression

Two hundred and twenty-five women responding to the postal survey scored ≥ 13 on the EPDS at six to seven months postpartum, a point prevalence of 16.9% (225/1331, 95% CI 14.9–18.9%). Five women did not complete all EPDS items, and had total scores in the range where a missing item may have resulted in a score above or below the cut off score for probable depression of 12.5. These women were excluded from subsequent analysis. The mean score on the EPDS was 7.2 (SD 5–21), with a median of 6 and range of 0 to 30.

Association between depression and health

Table 1 shows self-reported maternal health problems in the first six to seven months postpartum for women scoring above and below the cut off score of 12.5 on the EPDS. A number of health problems were associated with greater odds of scores indicating probable depression (EPDS score ≥ 13). Tiredness and relationship problems were associated with greater than threefold increases in the likelihood of women scoring ≥ 13. Urinary incontinence, back pain, experiencing more colds and minor illnesses than usual, and sexual problems were each associated with greater than twofold increases in the likelihood of women scoring above the cut off point for probable depression. Smaller significant increases in the likelihood of women being depressed were associated with bowel problems (constipation or faecal incontinence), perineal pain and problems with contraception. Physical health problems not associated with depression were: mastitis, haemorrhoids and pain associated with a caesarean section. None of the 75 women who reported having no physical health problems had a score of ≥ 13 on the EPDS.

Table 1.  Physical health problems during the first six postnatal months and emotional wellbeing at six to seven months postpartum (n= 1331). Values are given as n (%) or OR [95% CI]. EPDS = Edinburgh Postnatal Depression Scale
 EPDS Score 
 <13≥ 13Odds ratio
  1. *Denominator excludes women who had an elective or emergency caesarean section

  2. Denominator excludes women who had a spontaneous or operative vaginal birth

  3. Denominator excludes women who did not begin breastfeeding

  4. §χ2= 16.23; P≤ 0.0001

Painful perineum*   
  No687 (85.0)121 (15.0)1.00
  Yes220 (80.0)55 (20.0)1.42 [1.0–2.1]
Pain from caesarean section   
  No81 (85.3)14 (14.7)1.00
  Yes112 (76.2)35 (23.8)1.81 [0.9–3.8]
Urinary incontinence   
  No1001 (84.5)184 (15.5)1.00
  Yes100 (70.9)41 (29.1)2.23 [1.5–3.4]
Bowel problems   
  No978 (84.4)181 (15.6)1.00
  Yes123 (73.7)44 (26.3)1.93 [1.3–2.9]
Haemorrhoids   
  No831 (83.0)170 (17.0)1.00
  Yes270 (83.1)1.00 [0.7–1.4]Tiredness/ exhaustion
  No377 (93.3)27 (6.7)1.00
  Yes723 (78.5)198 (21.5)3.42 [2.2–5.3]
More colds and minor illnesses than usual   
  No996 (85.1)174 (14.9)1.00
  Yes105 (67.3)51 (32.7)2.78 [1.9–4.1]
Back pain   
  No659 (87.9)91 (12.1)1.00
  Yes442 (76.7)134 (23.3)2.20 [1.6–3.0]
  Mastitis*   
  No827 (83.0)169 (17.0)1.00
  Yes177 (86.8)27 (13.2)1.75 [0.5–1.2]
Problems with contraception   
  No1017 (83.8)196 (16.2)1.00
  Yes84 (74.3)29 (25.7)1.79 [1.1–2.9]
Sexual problems   
  No841 (86.2)135 (13.8)1.00
  Yes260 (74.3)90 (25.7)2.16 [1.6–3.0]
Relationship problems with partner   
  No947 (87.3)138 (12.7)1.00
  Yes154 (63.9)87 (36.1)3.88 [2.8–5.4]
One or more health problems   
  No75 (100)0 (0.0)undefined§
  Yes1026 (82.0)225 (18.0) 

Telephone follow up

Two hundred and four women participated in follow up telephone interviews conducted between seven and nine months postpartum. The final sample (n= 204) comprised 66 women with a score on the EPDS of < 9, 72 women with a score of 9–12, and 66 with a score of ≥ 13. Attempts were made to contact a total of 229 women. Thus, the final participation rate for the study was 89.1% (204/229). Table 2 outlines reasons for non-inclusion according to women's scores on the EPDS. Reasons for declining to participate included: not having access to a phone, being too busy (e.g. looking after the baby, working full-time, sick children, or an elderly parent with a terminal illness), insufficient fluency in English, not wanting to talk on the phone, and one woman was recovering from being in hospital. Women who made appointments, but subsequently cancelled them, gave similar reasons. Two changed their minds after making the appointment, one because her husband did not wish her to take part. The 16 women who either declined to take part or cancelled the interview included many who were relatively socially disadvantaged. Nine had a family income of less than A$ 30,000 in the year before the baby's birth, eight had not completed the final year of secondary education, three were of a non-English speaking background and two were without a partner. Nine were primiparae. The group of nine women who had either moved and/or had a phone number which had been disconnected included seven with a family income of less than A$ 30,000 per annum, six who had not completed the final year of secondary education, two women without a partner, one woman of non-English speaking background and three women with only one child.

Table 2.  Participation in the Life as a Mother Project (LAMP) telephone interviews
 EPDS score
 ≤99–12>13
Agreed to take part667266
Moved/unable to conatct045
cancelled appontment132
Declined109
TOTAL approached687982

Women who provided their name and telephone number on the survey forms (61% of respondents), in order for us to contact them for the purposes of follow up research, did not differ in terms of their EPDS scores from women who did not give their name and telephone number (Mann-Whitney P= 0.483). Similarly, there was no significant difference in the depression scores for the 66 women in the telephone interview sample who scored above the cut off score for depression and all other women who scored as depressed (EPDS ≥ 13) in the postal survey (Mann-Whitney P= 0.631).

Comparisons were also made between the women in the follow up sample who scored above the cut off score for depression and other women in the postal survey sample with scores of ≥ 13 on the EPDS, in terms of a number of sociodemographic characteristics. There were no differences between the groups in terms of age (χ2= 1.25, 2df, P= 0.5), marital status (χ2= 0.61, 2df, P= 0.7), education (completed secondary school: χ2= 1.26, 1df, P= 0.2), tertiary level qualifications (χ2= 3.08,2df, P= 0.2), income (χ2= 2.54,3df, P= 0.5), country of birth (Australia/overseas-English speaking background/overseas-non-English speaking background: χ2= 3.9, 2df, P= 0.2), place of residence (metropolitan/rural: χ2= 0.07, 1df, P= 0.8), health insurance status (χ2= 0.0, 1df, P= 0.9), or parity (χ2= 2.72, 1df, P= 0.1).

Table 3 shows the number and proportions of women reporting specific health problems in the six weeks before the interview according to their score on the EPDS (scores of > 9/scores of 9–12/scores ≥ 13). Four variables show statistically significant linear trends with poorer levels of emotional wellbeing: urinary incontinence, having more colds and minor illnesses than usual, extreme tiredness/exhaustion, and relationship problems.

Table 3.  Physical health problems in the six weeks prior to the Life as a Mother Project (LAMP) telephone interview at seven to nine months postpartum and associations with emotional wellbeing at six to seven months postpartum (n= 204).. Values are given as n (%), unless otherwise indicated
 EPDS score  
 ≤ 99–12> 13χ2 for linear trendp
  1. *Denominator excludes women who had an elective or emergency caesarean section

  2. Reference category is women reporting no symptoms

  3. The LAMP interview schedule did not include questions about pain from a caesarean wound, mastitis or problems with contraception

Painful perineum*     
  No51 (92.7)51 (85.0)42 (89.4)  
  Yes4 (7.3)9 (15.0)5 (10.6)NS 
Urinary incontinence     
  No56 (84.8)56 (77.8)44 (66.7)5.630.02
  Yes10 (15.2)16 (22.2)22 (33.3)  
Bowel problems*     
  No58 (87.9)65 (90.3)57 (86.4)  
  Constipation7 (10.6)5 (6.9)6 (9.1)NS 
  Incontinence1 (1.5)2 (2.8)3 (4.5)  
Haemorrhoids     
  No53 (80.3)63 (87.5)51 (77.3)NS 
  Yes13 (19.7)9 (12.5)15 (22.7)  
Tiredness/ exhaustion     
  No24 (36.4)19 (26.4)6 (9.1)12.38≤0.001
  Yes42 (63.6)53 (73.6)60 (90.9)  
More colds and minor illnesses than usual     
  No58 (87.9)59 (81.9)44 (66.7)7.940.005
  Yes8 (12.1)13 (18.0)22 (33.0)  
Back pain     
  No34 (51.5)36 (50.0)25 (37.9)NS 
  Yes32 (48.5)36 (50.0)41 (62.1)  
Sexual problems     
  No46 (69.7)50 (69.4)38 (57.6)NS 
  Yes20 (30.3)22 (30.6)28 (42.4)  
Problems in relationship with partner     
  No54 (81.8)51 (70.8)32 (48.5)15.01≤0.001
  Yes12 (18.2)34 (51.5)   

Back pain, which was common in all three groups (around 50% or greater), was slightly more common among women scoring ≥ 13 on the EPDS, but this difference was not statistically significant. Similarly, sexual problems affected a third or more women in all categories, with slightly higher prevalence in the group with scores above the cut off point for depression, again not a statistically significant difference.

Table 4 examines the associations between maternal physical health problems at seven to nine months postpartum, and relationship problems based on data from the Life as a Mother Project interviews. Four variables were associated with a greater likelihood of having relationship problems. These were: persisting perineal pain; urinary incontinence; more coughs, colds and other minor illnesses than usual; and sexual problems. Two variables—exhaustion and back pain—had results bordering on statistical significance.

Table 4.  Relationship between maternal physical health at seven to nine months postpartum and relationship problems (n= 204)*. Values are given as n (%) or OR [95% CI].
 Relationship problems 
 NoYesOdds ratio
  1. *Denominators vary because of missing values.

Painful perineum   
 No128 (93.4)9 (66)1.00
 Yes55 (83.3)11 (167)284 [1.0–8.2]
Urinary incontinence   
 No112 (71.8)44 (28.2)1.00
 Yes25 (52.1)23 (47.9)2.34 [1.1–4.8]
Haemorrhoids   
 No113 (67.7)54 (37.3)1.00
 Yes24 (66.7)13 (36.1)1.13 [0.5–2.5]
Tiredness/exhaustion   
 No37 (77.1)11 (22.9)1.00
 Yes99 (63.4)56 (36.1)1.90 [0.9–4.3]
More colds and minor illnesses than usual   
 No117 (72.7)44 (27.3)1.00
 Yes20 (46.5)23 (53.5)3.06 [1.5–6.5]
Back pain   
 No69 (72.6)26 (27.4)1.00
 Yes68 (62.4)41 (37.6)1.60 [0.9–3.0]
Sexual problems   
 No106 (79.1)28 (20.9)1.00
Yes31 (44.3)39 (55.7)4.76 [2.4–9.4]

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

A Medline search uncovered few other studies investigating the relationship between physical and emotional health after childbirth. A study conducted by Jacobson et al.20 in 1965 reported associations between postpartum mental disorders and three physical health factors: ‘genital pains’, puerperal fever lasting longer than two days, and ‘feelings of abdominal laxity’. Playfair and Gowers21 found a strong correlation between depression and ‘external stress after the birth’, including ‘ill health of the patient’ in a study based on a sample of 618 women drawn from general practice clinics in the UK. Kumar et al.22 investigated the impact of chronic illness using a life events scale and found no association with depression in a group of first time mothers. In a more recent study, Lanzi et al.23 report a strong association between maternal chronic health problems and depression in mothers of kindergarten-aged children attending a Head Start program in the United States.

Our reason for exploring the relationship between physical health problems and depression in the current study was the finding of a strong association between maternal physical health and emotional wellbeing in an earlier Victoria study following up women who took part in the 1989 Survey of Recent Mothers24,25. Analyses of the 1993 Survey and of the Life as a Mother Project telephone interview data reinforce the findings of this earlier study. A similar pattern of association between the presence of physical health problems and depression at six months postpartum has been identified by Small et al.26 based on home interviews with over 300 Vietnamese, Turkish and Filipino women giving birth in Victoria.

It is difficult to establish the full extent of maternal morbidity in the postnatal period. While severe physical morbidity such as postpartum haemorrhage and eclampsia generally result in hospital admission, more common morbidities such as urinary incontinence, perrneal pain and chronic exhaustion usually do not. Although women living in countries like the United Kingdom and Australia have frequent contact with general practitioners and community health nurses in the first year after childbirth10,27, several recent studies have documented women's reluctance to discuss their own health and recovery at these visits8,9. Poor disclosure of physical and emotional health problems after childbirth means that studies based on medical or hospital records are likely to under-represent the extent of maternal morbidity. Studies using self-reporting measures potentially provide better estimates of maternal health problems.

A major strength of the two studies reported here is that they draw on a large and representative population-based sample. Although the response rate for the postal survey was lower than anticipated based on an earlier survey, respondents were largely representative in terms of important obstetric characteristics16. Low response rates were received from younger women, women of non-English speaking backgrounds and women without a partner limiting the generalisation of the findings for these groups16. An analysis of the relationship between obstetric events in the current pregnancy and maternal physical and emotional health problems in the first six postnatal months based on the survey data is available9.

It is not possible to assess directly the extent to which responders to the postal survey may differ from nonre-spondents in terms of the prevalence of depression. In this study and a previous survey of recent mothers in Victoria we have compared early, intermediate and late responders to see whether there were any trends in relation to depression. In the 1989 Survey of Recent Mothers there were no differences in the proportion of depressed women (according to their scores on the Edinburgh Postnatal Depression Scale) related to the timing of response28. In contrast, women who responded later to the 1993 Survey of Recent Mothers were more likely to be depressed (χ2 for linear trend = 7.14, P= 0.008). Thus, we cannot rule out the possibility that nonrespondents may be more likely to be depressed, but equally we cannot necessarily extrapolate from late responders to nonrespondents.

The final sample size of the Life as a Mother Project study was adequate for identifying major differences between the groups. The participation rate of 89%, the absence of any statistically significant differences between the women included in the Life as a Mother Project sample and the main 1993 Survey of Recent Mothers sample strengthens the reliability of the study findings. Nonetheless, it is clear from the raw numbers that women whose scores on the EPDS were borderline (9–12), or above the cut off score for depression (≥ 13) more frequently declined to participate, or could not be contacted.

Reported prevalence of maternal morbidity in other studies varies widely depending on the method of ascertainment. Studies using interview methods or very detailed questioning usually report higher prevalence of physical problems, than studies using written questionnaires, especially where very general questions are used. It is likely that the postal survey results under-estimate the full extent of morbidity in the first six months post-partum. The higher prevalence reported in the Life as a Mother Project interviews is at least partially explained by the use of interviews, rather than a written questionnaire. The anonymity of the telephone interview—as opposed to a face to face interview—may also be an advantage, where the issues being discussed are of a very personal or sensitive nature.

Some of the results from the telephone interviews, following up women between seven and nine months after birth, are particularly concerning. Among the 63 women whose scores on the EPDS were above the cut off score for depression, two-thirds had experienced back pain in the preceding six weeks, half reported problems in relating to their partner, greater than two-fifths reported sexual problems and a third reported current urinary incontinence. Although, there is a longitudinal component to this study, we lack sufficient information to determine whether relationships are causal, and more specifically, to determine the direction of effects. It is likely that the relationships between depression and the occurrence of individual health problems, and between depression and women's general health and recovery after childbirth, are complex. Depression may contribute to slower recovery and to problems such as tiredness, relationship problems and sexual difficulties, but it is also likely that physical health problems contribute to depression. It is noteworthy that perineal pain during the first six months post-partum is associated with higher odds of depression in the postal survey results, but persisting perineal pain reported at the seven to nine month follow up is not associated with depression. This finding should allay concerns that our results are potentially biased by women who are depressed being more likely to report physical problems in an effort to explain their poorer emotional wellbeing.

One of the most striking results of the LAMP study is the high prevalence of physical health problems, such as exhaustion, back pain, sexual problems, urinary incontinence and persisting perineal pain among women who were not depressed at the time of our six to seven month assessment. Two-thirds of women who had scores below the cut off point for depression reported being extremely tired or exhausted, almost half had back pain in the previous six week period, one in five had current urinary incontinence and one in ten reported current perineal pain. These figures reinforce the view that maternal morbidity following childbirth is extremely common, with many problems persisting into the second half year after the birth.

Thirty-one percent of women who were not depressed reported problems in their relationship with their partner, indicating that this also is a common health issue for women not concurrently affected by depression. Our results suggest that physical health problems, such as urinary incontinence and increased vulnerability to minor illnesses, as well as persisting perineal pain and sexual problems, may contribute to relationship difficulties in the first year after childbirth.

The fact that physical health has been largely overlooked as a potential contributing factor to maternal depression is probably a direct reflection of the extent to which common physical health problems of recent mothers have been under-reported and under-recognised by health professionals. Our findings, confirming a small number of other studies identifying a link between maternal emotional wellbeing, physical health and recovery in the postnatal period, have important implications for clinical practice. Strategies for encouraging greater disclosure of physical and emotional health problems, assessment of the physical health of recent mothers presenting with depression, and emotional health of recent mothers presenting with other morbidity, should be high priorities for all health professionals in contact with mothers in the year following childbirth.

Acknowledgements

We are grateful to the women of Victoria who gave generously of their time to complete the survey and to participate in telephone interviews; to the Victorian Health Promotion Foundation which funded the study; the hospitals and home birth practitioners who distributed the questionnaires; Ms F. Bruinsma for assistance with maintaining the register of women invited to participate in the Life as a Mother Project interviews; Dr M. Montague for assistance with conducting the interviews; the Victorian Perinatal Data Collection Unit for providing data on obstetric and social characteristics of women who gave birth in the study period; and members of the project reference group for their helpful advice throughout and for assistance with piloting. While we greatly appreciate the contributions others have made, the authors accept full responsibility for the design and conduct of the study, and for the analysis of the results.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References
  • 1
    Leung SS, Steinbeck KS, Morris SL, Kohn MR, Towns SJ, Bennett DL. Chronic illness perception in adolescence: implications for the doctor-patient relationship. J Paediatr Child Health 1997; 3: 107112.
  • 2
    Clarke DM, Russell PA, Polglase AL, McKenzie DP. Psychiatric disturbance and acute stress responses in surgical patients. Aust N Z J Surg 1997; 67: 115118.
  • 3
    Romito P. Unhappiness after childbirth. In: ChalmersI, EnkinM, KeirseMJNC, editors. Effective Care in Pregnancy and Childbirth. Oxford : Oxford University Press, 1989: 14331446.
  • 4
    MacArthur C. What does postnatal care do for women's health Lancet 1999; 353 (9150): 343344.
  • 5
    Glazener C, Abdalla M, Stroud P, Naji S, Templeton A, Russell IT. Postnatal maternal morbidity: extent, causes, prevention and treatment. Br J Obstet Gynaecol 1995; 102: 282287.
  • 6
    MacArthur C, Lewis M, Knox EG. Health after childbirth. Br J Obstet Gynaecol 1991; 8: 11931204.
  • 7
    MacArthur C, Lewis M, Knox EG. Health After Childbirth. London : HMSO, 1991.
  • 8
    Gjerdingen DK, Froberg DG, Chaloner KM, McGovern P. Changes in women's physical health during the first postpartum year. Arch Fam Med 1993; 2: 277283.
  • 9
    Brown S, Lumley J. Maternal health after childbirth: results of an Australian population based survey. Br J Obstet Gynaecol 1998; 105: 156161.
  • 10
    Glazener C, MacArthur C, Garcia J. Postnatal care: time for a change. Contemp Rev Obstet Gynaecol 1993; 5: 130136.
  • 11
    Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression: development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psych 1987; 150: 782786.
  • 12
    Harris B, Huckle P, Thomas R, John S, Fung H. The use of rating scales to identify postnatal depression. Br J Psychiatry 1989; 154: 813817.
  • 13
    Murray L, Carothers AD. The validation of the Edinburgh Postnatal Depression Scale on a community sample. Br J Psychiatry 1990; 157: 288290.
  • 14
    Boyce P, Stubbe R, Todd A. The Edinburgh Postnatal Depression Scale: validation in an Australian sample. Aust N Z J Psychiatry 1993; 27: 472476.
  • 15
    Hickey A, Boyce P, Ellwood D, Morris-Yates A. Early discharge and risk for postnatal depression. Med J Aust 1997; 167: 244247.
  • 16
    Brown S, Lumley J. The 1993 Survey of Recent Mothers: issues in design, analysis and influencing policy. Int J Qual Health Care 1997; 9: 265277.
  • 17
    SPSS for Windows, Version 6.1. Chicago , USA . SPSS Inc: June 1994.
  • 18
    Dean AG, Dean JA, Coulumbier D et al. Epi Info Version 6 (A word processing, database and statistics program for public health on IBM compatible microcomputers). Atlanta , Georgia : Centers for Disease Control and Prevention: 1995.
  • 19
    Campbell MJ, Julious SA, Altman DG. Estimating sample sizes for binary, ordered categorical and continuous outcomes in two group comparisons. BMJ 1965; 1: 16401643.
  • 20
    Jacobson L, Kaij L, Nilsson A. Postpartum mental disorders in an unselected sample: frequency of symptoms and predisposing factors. BMJ 1965; 1: 16401643.
  • 21
    Playfair HR, Gowers JI. Depression following childbirth — a search for predictive signs. J R Coll Gen Pract 1981; 31: 201208.
  • 22
    Kumar R, Robson KM. A prospective study of emotional disorders in childbearing women. Br J Psychiatry 1984; 144: 3547.
  • 23
    Lanzi RG, Pascoe JM, Keltner B, Ramey SL. Correlates of maternal depressive symptoms in a national Head Start program sample. Arch Pedatr Adolesc Med 1999; 153: 801807.
  • 24
    Small R, Brown S, Lumley J. Missing voices: what women say and do about depression after childbirth. J Reprod Infant Psychol 1994; 12: 89103.
  • 25
    Small R, Astbury J, Brown S, Lumley J. Depression after childbirth. Does social context matter Med J Aust 1994; 161: 473477.
  • 26
    Small R, Lumley J, Yelland J, Liamputtong-Rice P. Mothers in a New Country. Summary Report. Melbourne : Centre for the Study of Mothers' and Children's Health, 1997; and personal communication R. Small, 1999.
  • 27
    Gunn J, Lumley J, Young D. Visits to medical practitioners in the first six months of life. J Paediatr Child Health 1996; 32: 162166.
  • 28
    Brown S, Lumley J, Small R, Astbury J. Missing voices. The experience of motherhood. Melbourne : Oxford University Press, 1994.