Recommendation 1. Nurses provide individualized, flexible postpartum care based on the identification of depressive symptoms and maternal preference (Level of Evidence = Ia).
Systematic reviews of randomized control trials examining the effects of preventive psychosocial or psychological interventions for reducing the risk of developing postnatal depression suggest that intensive, professionally based support may prevent women from developing postnatal depression. In this body of research, the reviewers indicate that the preventive approaches were diverse: antenatal and postnatal classes, lay home visits, or early postpartum follow-up (Boath, Bradley, & Henshaw, 2005; Chabrol & Callahan, 2007; Dennis, 2005). Supportive home visits made postnatally were identified as being particularly effective when the population included vulnerable mothers and when the visits were individualized.
MacArthur et al. (2002), in a randomized controlled trial of 2,064 women, assessed community-based postpartum care in the U.K. The intervention was designed to be flexible and meet the individual needs of the new mother. Individual maternal needs were determined by midwives using a symptom checklist and assessment. Women in the control group received standard care of midwifery home visits with follow-up from health visitors and general practitioners. Significant group differences were found in favor of the intervention group with 14.4% of mothers scoring above 12 on the Edinburgh Postnatal Depression Scale (EPDS) in comparison to 21.3% of mothers in the control group suggesting that the delivery of flexible and mother-centered care may improve women’s mental health outcomes.
Armstrong, Fraser, Dadds, and Morris (1999) conducted a trial of “at-risk” families. The subjects included mothers living in circumstances such as violence, financial stress, unstable housing, ambivalent about pregnancy, and having had limited antenatal care. Mothers in the intervention group (n = 90) received nursing home visits weekly to 6 weeks postpartum, then every 2 weeks to 12 weeks postpartum, and then monthly to 24 weeks postpartum. Mothers in the control group (n = 91) were given instructions on how to access existing community child health services. At 6 weeks postpartum, intensive nursing home visits resulted in significantly lower mothers’ EPDS scores compared to those in the control group. These differences, however, were not maintained at the 16-week postpartum follow-up. Armstrong et al. hypothesize that continuation of biweekly visits might enable the initial preventive effect to be maintained to 24 weeks postpartum.
Recommendation 2. Nurses initiate preventive strategies in the early postpartum period (Level of Evidence = Ia).
Researchers evaluating the effect of intervention onset report that interventions initiated in the antenatal period fail to reduce the risk of women developing postpartum depression in comparison to interventions delivered only in the postnatal period (Dennis, 2005; Shaw, Levitt, Wong, Kaczorowski, & The McMaster University Postpartum Research Group, 2006). For example, Matthey, Kavanagh, Howie, Barnett, and Charles (2004), in a study of 268 prenatal primiparous mothers, found no differences in depressive symptoms at 6 months postpartum between mothers who received additional psychosocial intervention and those who received routine care or baby play information sessions. By way of caution, much of the antenatal prevention studies have significant methodological limitations such as heterogeneous samples, diverse screening measures for depression, and high attrition rates.
Maternal mood in the immediate postpartum period (or up to 2 weeks postpartum) is a significant predictor of postpartum depression (Beck, 2002; Dennis, 2005). Mothers who scored high on the EPDS at 7 days postpartum were more likely to have depressive symptoms at 1 and 2 months postpartum (Dennis, 2005). Similar findings have been found in other studies of mothers with diverse ethnic orientations (Yamashita, Yoshida, Nakano, & Tashiro, 2000). Beck’s metasynthesis of qualitative research suggests that mothers who talk about postpartum blues as overwhelming also describe experiences of postpartum depression. Such findings support the importance of early interventions such as referral or support in order to minimize the impact of symptoms for the mother and her family.
Confirming depressive symptom among postpartum women.
Recommendation 3. The EPDS is the recommended self-report tool to confirm depressive symptoms in postpartum mothers (Level of Evidence = III).
Various tools such as the Beck Depression Inventory, the Postpartum Depression Screening Scale (PDSS), the Hospital Depression and Anxiety Scale, and the Hamilton Rating Scale for Depression have been clinically used to screen for depressive symptoms in postpartum mothers. The panel recommends the EPDS in clinical practice (RNAO, 2005). It has demonstrated reliability, sensitivity (or proportion of mothers correctly identified as having depressive symptoms), specificity (proportion of mothers correctly identified as not having depressive symptoms), and positive predictive value (proportion of mothers who were diagnosed with postpartum depression) (Boyce, Stubbs, & Todd, 1993; Cox, Holden, & Sagovsky, 1987; Harris, Huckle, Thomas, Johns, & Fung, 1989; Thompson, Harris, Lazarus, & Richards, 1998). Ease of administration, acceptability by different cultures, international recognition, and availability at no cost were identified as supportive for the clinical applicability of the EPDS (Dennis, 2003a).
Since the development of this recommendation, two systematic reviews have been conducted to evaluate screening tools for postpartum depression in the postnatal period (Boyd, Le, & Somberg, 2005; Gaynes et al., 2005). Boyd et al. reviewed 36 studies evaluating eight self-report measures used to assess for depressive symptoms in the postpartum period. Their findings concurred with the RNAO’s recommendation that the EPDS is the most extensively studied postpartum assessment tool with moderate psychometric soundness. Although the PDSS also demonstrates promise as a valid tool, further independent psychometric testing is required before any recommendations can be made regarding its use (Boyd et al.; RNAO, 2005). Comparatively, Gaynes et al.’s (2005) systematic review of 10 studies assessed the accuracy of different screening tools for detecting depression and did not identify any one particular tool as superior. They found that the estimates of sensitivity and specificity appeared equivalent and they were unable to combine the results of different studies for meta-analysis due to the different tools and cutoffs used. They concluded that there are various screening instruments (including the EPDS) that can be used to identify perinatal depression, most accurately major depression.
It is noteworthy that both systematic reviews identified that many included studies suffered from small sample size. Overall, identifying that the evidence base is quite limited, more research is required including larger samples that are representative of the population, diverse racial and ethnic mix, and other demographic variables. See Table 3 for future potential research studies. Additionally, the systematic reviews had different inclusion/exclusion criteria, thus the difference in number of studies included (Gaynes, 10; Boyd, 36). The main difference was Gaynes et al.’s (2005) inclusion criteria that specified a clinical assessment or structured clinical interview to confirm depressive symptoms. Boyd et al. (2005) and RNAO (2005) did not require a clinical interview and therefore included studies with self-reported depressive symptoms.
Table 3. Potential Research Questions
| 1. Can health providers reliably identify women at risk for postpartum depression?|
| 2. Are supportive preventive interventions more effective when provided by peers or health professionals?|
| 1. Is any one tool superior in confirming depressive symptoms in postpartum women?|
| 2. What are the effects of false-negative and false-positive postpartum depression scores?|
| 3. Does the repeated administration of tools to assess depressive symptoms alter the validity, sensitivity, or specificity of the tool?|
| 1. What interventions do mothers perceive as beneficial in the treatment of postpartum depression?|
| 2. Are any treatment interventions more effective than others?|
| 3. What are the long-term treatment effects (medication and nonmedication) for postpartum depression on mothers and infants?|
| 4. Are multifaceted interventions more effective than single interventions?|
Recommendation 4. The EPDS can be administered anytime throughout the postpartum period (birth to 12 months) to confirm depressive symptoms (Level of Evidence = III).
For the majority of women, the onset of depression usually occurs within the first few weeks or months after delivery (Cooper, Campbell, Day, Kennerley, & Bond, 1988; Goodman, 2004; McIntosh, 1993; Small, Brown, Lumley, & Astbury, 1994). However, for some mothers, the inception of depressive symptoms occurred after 12 weeks (Small et al., 1994) and others much later between 6 and 12 months (Cooper et al., 1988). A systematic review of 30 studies evaluating the prevalence and incidence of postpartum depression found that the prevalence of major and minor depressions began to rise following delivery with the highest increase of 12.9% at 3 months, declining slightly in the 4th through 7th month (9.9%-10.6%), and declining even further (approximately 6.6%) from the 8th to 12th month (Gaynes et al., 2005). As such, the presence of depressive symptoms remains fairly high for the first 6 months postpartum before it starts to decrease. It is noteworthy that research has not specified an “optimal” clinical time to employ the EPDS or the implications of repeatedly applying the EPDS to assist with assessment. The validity, sensitivity, and specificity of repeated use with a postpartum mother are unknown.
Recommendation 5. Nurses encourage postpartum mothers to complete the EPDS tool by themselves in privacy (Level of Evidence = III).
Current research suggests that EPDS scores are more reliable when mothers complete the EPDS alone, without others present (Cox & Holden, 2003). Researchers report that mothers are often reluctant to disclose emotional problems (Beck, 2002; Brown & Lumley, 2000) or to obtain professional assistance (Small, Johnston, & Orr, 1997). Furthermore, the EPDS can be effectively administered during a phone conversation with a mother (Dennis, 2003b; Zelkowitz & Milet, 1995). The RNAO Best Practice Guideline (2005) provides sample lead-in questions and administration protocol.
Recommendation 6. An EPDS cutoff score greater than 12 may be used to determine depressive symptoms among English-speaking women in the postpartum period. This cutoff criterion should be interpreted cautiously with mothers who (a) are non-English speaking, (b) use English as a second language, and/or (c) are from diverse cultures (Level of Evidence = III).
Research has demonstrated that the EPDS is most effective in the confirmation of depressive symptoms (sensitivity and specificity) when the recommended cutoff score of greater than 12 is used in the postpartum period among English-speaking mothers in the U.K. (Cox et al., 1987), Canada (Zelkowitz & Milet, 1995), and Australia (Boyce et al., 1993). An EPDS score greater than 12 does not mean a mother has postpartum depression, nor does a score translate into the severity of depressive symptoms (Cox & Holden, 2003). Conversely, a score of 11 does not indicate that the mother is not experiencing emotional distress particularly when the health professional has identified concerns. Rather, an EPDS score indicates the mother’s perception of her mood.
In addition, different cutoff scores may be required for non-English versions of the EPDS (Dennis, 2003b). For example, with Japanese subjects, using a 12/13 cutoff score, no mothers obtained an EPDS score of 13 or higher (Okano et al., 1998; Yoshida, Yamashita, Ueda, & Tashiro, 2001). Researchers suggest that, due to cultural expectations, Japanese mothers may be reluctant to disclose depressive symptoms and, therefore, a lower cutoff of 8/9 may be more suitable for this population (Okano et al., 1998). When used with Arabic women, a threshold score of 11/12 was adopted to confirm depressive symptoms at 7 days postpartum (Ghubash, Abou-Saleh, & Daradkeh, 1997). A cutoff score of 9/10 was considered most appropriate at 6 weeks postpartum for mothers of Chinese descent (Lee et al., 1998). These research findings highlight the social and cultural expectations and context of motherhood and reinforce the importance of clinical judgment when interpreting the EPDS score for all mothers regardless of language.
While the systematic review by Gaynes et al., (2005) did not recommend any specific cutoff value, they emphasized that the relative cost, or value, of errors in screening tests is of utmost importance in setting cutoff scores. For example, a false-negative result may lead to a prolonged and untreated depression, whereas a false-positive result may lead to unnecessary treatment or stress associated with an incorrect screen. Thus, the screening test should try to maximize the most effective cutoff.
The EPDS may be used as part of a comprehensive assessment to identify depressive symptoms early so that referral and treatment may be initiated.
Recommendation 7. The EPDS must be interpreted in combination with clinical judgment to confirm postpartum mothers with depressive symptoms (Level of Evidence = III).
The clinical use of the EPDS is not intended to replace health professionals’ comprehensive assessment of a mother/family during the postpartum period. Instead, the guideline recommends that the EPDS be used as an adjunct to clinical evaluation. Elliott and Leverton (2000) emphasize that when used alone, “the EPDS is just a piece of paper, a checklist” (p. 305). When combined with knowledge in prevention, identification, and treatment of postpartum depression, however, the EPDS can facilitate discussion about needs, family and role transitions, and supports (Cox & Holden, 2003; Lundh & Gyllang, 1993). Additionally, the EPDS may be used as a means of identifying improving or worsening symptoms and to assist decision making regarding engaging support systems, providing postpartum depression information, and referring and securing treatment.
Recommendation 8. Nurses should provide immediate assessment for self-harm ideation/behavior when a mother scores positive (e.g., from 1 to 3) on the EPDS self-harm item number 10 (Level of Evidence = IV).
Question number 10 on the EPDS assesses any thoughts specific to self-harm ideation. In a Canadian population of 594 mothers, 4.5% (n = 27) expressed suicidal thoughts at 1 week, with a similar amount at 4.3% (n = 23) at 4 weeks, which increased slightly to 6.3% (n = 32) at 8 weeks (Dennis, 2004a). Additional researchers have identified similar percentages of suicidal ideation (Morris-Rush, Freda, & Bernstein, 2003). Depression is a major risk factor for suicide, and any positive score (1-3) related to suicidal ideation or self-harm must be further assessed in a timely manner. Although various suicide assessment scales are available for use in practice, the developers of the EPDS recommend additional questions for assessment (see Table 4). Agency policies and procedures must be in place to assist health providers with decision making in the plan of care for mothers who score positive on self-harm (Cox & Holden, 2003).
Table 4. Assessment of Self-Harm
|1. How often do you have thoughts of harming yourself?|
|2. How severe are these feelings? How much have they been bothering you?|
|3. Have you had these kinds of feelings before? If so, what happened? How did you cope with them?|
|4. Have you made any previous suicide/self-harm attempts?|
|5. Have you thought about how you would harm yourself?|
|6. What support do you have at home?|
|7. (If she has a partner) Have you talked about your feelings with him/her?|
|8. Are you close to your parents/other family members? Do they know how you have been feeling?|
|9. Can you count on your partner and/or family members to give you emotional support?|
|10. Is there anyone else in your life whose support you can count on?|
|11. Have you told this person or anyone else about your feelings?|
|12. Could you phone this person and would he/she come if you felt the needed support?|
Recommendation 9. Nurses provide supportive weekly interactions and ongoing assessment focusing on maternal mental health needs of postpartum mothers experiencing depressive symptoms (Level of Evidence = Ib).
Diverse psychological/psychosocial interventions delivered on a weekly basis have been identified as effective in the treatment of postpartum depression. These include interpersonal psychotherapy (Klier, Muzik, Rosenblum, & Lenz, 2001), cognitive behavioral therapy (Honey, Bennett, & Morgan, 2002), and nondirective counseling (Cooper, Murray, Wilson, & Romaniuk, 2003; Holden, Sagovsky, & Cox, 1989; Wickberg & Hwang, 1996). While all the interventions evaluated were effective, nondirective counseling may be of particular interest as nurses can implement it into their practice. Nondirective counseling involves the art of presencing (i.e., being there), displaying nonjudgment as well as purposeful listening in order to promote a safe space for the exploration of the woman’s reality (RNAO, 2005).
Additional support for the effectiveness of nondirective counseling comes from qualitative literature of mothers’ expressed needs to disclose their emotional concerns with a compassionate listener (Ritter, Hobfoll, Lavin, Cameron, & Hulsizer, 2000; Small et al., 1994). With training, this interactive approach can be used by a variety of health care providers with women with mild-to-moderate depression (Holden et al., 1989; Wickberg & Hwang, 1996). For women experiencing severe depression, other treatments are often required. The degree of depression alone is not the sole indicator of a specific intervention (such as medication, cognitive behavioral therapy, or interpersonal psychotherapy); rather, treatment based on maternal needs and preferences is the guiding principle (Holden, 1996). Health providers (nurses and home visitors) working with mothers with postpartum depression require access to support from physicians, psychiatrists, psychologists, or community psychiatric nurses so that referrals can be made in a timely fashion (Holden, 1996).
More recent findings also provide additional support for the effectiveness of diverse treatments for postpartum depression. In particular, British Medical Journal ’s Clinical Evidence search and appraisal identified cognitive behavioral therapy, interpersonal therapy, and nondirective counseling as likely to be effective nondrug treatments for postpartum depression (Howard, 2006). Similarly, Bledsoe and Grote’s (2006) preliminary meta-analysis found diverse interventions to be effective for treating nonpsychotic major depression during pregnancy and postpartum. Their research involved 16 treatment trials with 922 participants, which assessed treatment effects during pregnancy (3 trials) and the postpartum period (13 trials). However, their findings combined both the antenatal and the postnatal studies and therefore cannot be generalizable to this postpartum treatment recommendation. Finally, a literature review by Chabrol and Callahan (2007) suggests equivalent effectiveness among the psychological interventions (nondirective counseling, cognitive behavioral therapy, and interpersonal therapy) with no one intervention identified as being optimal.
There are also some preliminary findings that suggest that additional therapies such as group interpersonal therapy (Reay et al., 2006), telecare support (Ugarriza & Schmidt, 2006), and exercise (Daley, Macarthur, & Winter, 2007) may potentially have therapeutic effects on depressive symptoms. However, these findings are preliminary and based on pilot work, and more research is required before any recommendations can be made regarding their use as an adjunct or primary treatment for postpartum depression.
Recommendation 10. Nurses facilitate opportunities for the provision of peer support for postpartum mothers with depressive symptoms (Level of Evidence = IIb).
Many studies identified a lack of support as a strong risk factor in the development of postpartum depression (Dennis, 2003b, 2004b). Presence of a trustworthy, available, and understanding listener and/or belonging to a group of similar others may be effective in reducing the potential severity of depressive symptoms particularly when women describe postpartum depression in terms of isolation, as being misunderstood and as being alone (Brugha et al., 1998; Eberhard-Gran, Eskild, Tambs, Schei, & Opjordsmoen, 2001). Limited research, however, has been conducted specifically evaluating the efficacy of social support and postpartum depression, and the research that has been conducted has been inconclusive (Chen, Tseng, Chou, & Wang, 2000; Dennis, 2003b, 2004a; Fleming, Klein, & Corter, 1992; Morgan, Matthey, Barnett, & Richardson, 1997).
The degree of depression alone should not dictate the type of intervention; rather, treatment should be based on maternal needs, preferences, and severity of symptoms.