Postpartum Depression Screening at Well-Child Visits: Validity of a 2-Question Screen and the PHQ-9

Postpartum depression is a disease that family physicians are extremely well-suited to deal with. The condition is common and has a significant impact on the entire family. The postpartum woman suffers the debilitating effects of the depression, their infants are adversely impacted, and even spouses seem to be put at risk for their own emotional problems. Postpartum depression can be screened for, diagnosed, and treated by family physicians. For me, its diagnosis and treatment is extremely rewarding. Despite all of this, detection rates continue to be low, mainly due to low screening rates. The barriers to screening and diagnosis are well documented in the literature. Chief among them is the lack of time busy clinicians face on a daily basis.

This is why this study by Gjerdingen et al. is so important. The 2 question screen takes very little time to complete, does not require special training to administer or score, and is very sensitive. While there is legitimate reason to debate which screening method (EPDS or PHQ- 9) should follow, the real take home for me is that this two-step method might be more acceptable to providers because the first step is so easy. Breaking down the time barrier is crucial if screening rates are to rise.

Perhaps this study will open the door to more universal screening. It is important for such simple, efficient methods such as this to find their way into common practice. One way to insure this is for educators to instill their importance early. Students and residents need to get the message that screening for postpartum depression is quick, easy, and mandatory.

Competing interests:   None declared

Screening for postpartum depression (PPD) has been recommended and implemented, however, screening efforts are greatly affected by the availability of valid and reliable measures. Sensitivity and positive predictive value (PPV) are particularly important for screening. Gjerdingen and colleagues demonstrated that the modified PHQ-2 screener has excellent sensitivity and low PPV while the PHQ-9 simple scoring showed good sensitivity and low PPV using the gold standard, a diagnostic interview.

There have been two recently published articles examining the PHQ with postpartum samples. These articles were published after the Boyd and colleagues review(1) in which the PHQ-9 was not considered. The modified PHQ-2 screener has been shown to demonstrate good sensitivity and low PPV with the Edinburgh Postnatal Depression Scale (EPDS) as the criterion for postpartum women(2). However, Hanusa et al.(3) found that the PHQ-9 simple scoring demonstrated good PPV but correctly identified only 31% of the women with Major Depressive Disorder. Their main findings provided support for use of the EPDS in comparison to the PHQ-9. Taken together, these studies do not provide convincing psychometric evidence for the use of the PHQ in PPD screening, however more research is needed.

The Gjerdingen et al. article raises several issues concerning the implementation of PPD screening at well-child visits. First, the use of the complex scoring of the PHQ-9 resulted in reduced sensitivity of the measure. It is likely that clinical judgment is needed for the complex scoring and this may compromise the validity for detecting true depressive symptoms. Second, the PHQ-2 sceener showed a higher sensitivity when compared to the PHQ-2 providing support for the screener. Third, in busy practice settings, it is potentially burdensome for physicians or clinic staff to conduct two separate screenings even with brief measures. Screening procedures should be simple and practical for the staff implementing them. Moreover, there is no discussion of which staff at the well-child visit could be responsible for screening for PPD. Fourth, the women who dropped out of the study were those with the least resources (e.g., positive depression screen, lower income, less likely to be married) but are likely the most in need of postpartum support. Similarly, other studies have shown greater depressive symptoms in low- income women(4). Finally, it is critical to discuss mental health referral and treatment as a part of the screening process.

References
1. Boyd RC, Le HN, Somberg R. Review of screening instruments for postpartum depression. Arch Womens Ment Health 2005; 8(3):141–153.
2. Bennett IM, Coco A, Coyne JC. Efficiency of a two-item pre-screen to reduce the burden of depression screening in pregnancy and postpartum: an IMPLICIT network study. J Am Board Fam Med. 2008; 21(4):317-25.
3. Hanusa BH, Scholle SH, Haskett RF, Spadaro K, Wisner KL. Screening for depression in the postpartum period: a comparison of three instruments. J Womens Health (Larchmt). 2008;17(4):585-96.
4. Boyd RC, Worley H. Utility of the Postpartum Depression Screening Scale among low-income ethnic minority women. In: Rosenfield AI Ed., New research on postpartum depression. New York: Nova Science Publishers 2007;151-166.

Competing interests:   None declared

It is good to hear from Dr. Anne Buist and Dr. Donna Stewart, both of whom have made important contributions to women's mental health research.

Dr. Buist is correct - the EPDS is a very widely used and acceptable screen for postpartum depression, and in fact in the most commonly used screen for PPD worldwide. We wanted to validate the PHQ-9 in a postpartum sample, as it is becoming an increasingly popular tool in general depression research and practice, and is considered a diagnostic instrument, in that it contains all of the diagnostic criteria for depression. I hope that at some time the PHQ-9 and the EPDS will be compared head-to-head in a large postpartum depression study. Importantly though, either tool would be useful in identifying depressed mothers. Regarding the participant numbers for references 12 and 13, I agree that the "860,479" figure was misleading, and it would have been clearer if the text had read "860 and 479." I apologize for the confusion. The 860 figure came from the abstract of reference #12, which states that 860 postnatal women were surveyed.

I agree wholeheartedly with Dr. Stewart's comment, that positive screens need to be followed by clinical assessment and treatment, and that to screen without ensuring this type of follow-up is unethical and futile!

Competing interests:   None declared

Screening for postnatal depression has been the source of ongoing debate, first in the UK with Judy Shakespeare's review highlighting the high level of data and research required in the ideal world before a routine screening should be introduced, and over the last few years taken up by our group when beyondblue (a not for profit organisation) funded a large project in part research and in part a public health initiative (references 12 and 13 in this article refer to the beyondblue project). As the Director of this earlier project, I agree with the authors of this article that screening is important; we have shown it to be acceptable and feasable and the Australian government had supported our bid to have it introduced as "routine". Each state government is having to look at how to implement this however, and there remains controversy. A initial easier step (in this case the two questions) would be attractive if introducing screening on such a large scale. I think in our case however, the world wide use, acceptability and simplicity of the EPDS (EDINBURGH POSTNATAL DEPRESSION SCALE)would make this the obvious second step. Also, just to clarify; I am uncertain where the authors got 860,479 women from references 12 and 13. In my report on acceptability 860 women were approached and 501 responded; overall the project screened just over 40,000 women (though information was given to over 100,000.

Competing interests:   None declared

I appreciate the comments of Susan McRoberts and Pam Stenhjem. Pam was wise and perhaps a little unusual in being able to diagnose her own postpartum depression, because of her friend's previous depression and her own self-education. Most women with postpartum depression are not so fortunate and rely on health care providers to recognize the problem, which may not work if the provider doesn't screen, or if screening doesn't coincide with depressive symptoms. This article describes an efficient, validated method for making the diagnosis, but we also need to remember to give effective, consistent care and follow up to women who are diagnosed.

Competing interests:   None declared

Perinatal depression is a serious public health problem.Depression during pregnancy worsens maternal and fetal outcomes and is a strong predictor for postpartum depression, which has deleterious effects for the mother, her infant and family.

Gjerdingen et al in this issue of Annals of Family Medicine have demonstrated the feasibility and validity of administering a 2-question screen, followed by the PHQ-9 when indicated, to identify postpartum depression during well-child visits in family medicine and pediatric practices. Although the 2-question screen had previously been validated in other populations, the current study is the first to do in a postpartum sample.

Despite the limited response rate (33%), the study makes an important contribution to clinical practice. The 2-question screen had 100% sensitivity and 44% specificity and the PHQ-9 (complex scoring) led to sensitivity/specificity of 67%/92% respectively, when compared to the Structural Clinical Interview for DSM IV (SCID). Importantly, the lack of specificity on the 2-question screen could be remedied by administering the PHQ-9 to women screening positive on either of the 2 screening questions.

These results suggest that a time efficient 2 stage screening can readily be conducted in both family medicine and pediatric practices. What should not be forgotten, however, is that women with positive results should be clinically assessed and appropriate treatment plans implemented. To screen without appropriate follow-up is unethical and futile!

Competing interests:   Scientific Advisory Board, Eli Lilly,Wyeth, Boelinger I.

I think that the 2 question screen is a fantastic way to look for postpartum issues in women. The PHQ-9 is an invaluable tool in digging deeper into what is going on. Speaking from experience, it is excruciating to share the details of depression and anxiety with a doctor, even one you fully trust. These tools give women the words they need when they can't find them. It's easier to answer these questions than it is to come to the doctor with a list of symptoms. It's less intimidating and takes the pressure off the mother. Because someone already has the symptoms listed, it validates the woman's issues, causes her to feel like she isn't making it all up and like she isn't the first person to experience these scary things. These are tools that doctors should be using. These are not time consuming tests and can alter the course of a woman's postpartum health. I think that is time well spent.

Competing interests:   None declared