Recommendation on instrument-based screening for depression during pregnancy and the postpartum period

this show that the of to for the following very

Unlike usual communication and discussion during a patient visit, screening for depression involves formally asking all patients a defined set of questions or administering a questionnaire with a pre-defined cut-off score to identify patients who may have depression. Patients with positive screening scores must be followed up by a health care provider to conduct a more detailed mental health assessment or make an appropriate referral. Treatment for depression during pregnancy and the postpartum period can include supportive interventions, psychotherapy, lifestyle changes, and antidepressant medications.

INTERVENTION:
Interventions that use a single question, small sets of questions, or a screening questionnaire (validated or nonvalidated) with a pre-defined cut-off score to identify patients who may have depression, but who have not reported their symptoms to healthcare providers or who have otherwise not been identified as possibly depressed by healthcare providers. Exceeding a cut-off score must be followed by a specified action such as more thorough assessment or referral.

COMPARISON:
No depression screening. Usual care during pregnancy and the postpartum period typically includes informally asking pregnant and postpartum individuals about their personal and family history of mental illness (e.g., anxiety, depression) and current mood-related symptoms.

MAIN OUTCOMES:
Critical: • Symptoms of depression (using continuous or dichotomous measures) or a diagnosis of major depressive disorder (MDD) using a validated diagnostic interview • Health-related quality of life (using validated tools) • Reported or observed capacity to parent (e.g., attachment, responsiveness to infant, positive regard of infant/fetus) Important: • Suicidality (suicidal ideation, attempt, or completion) • False positive screens (positive screens in absence of depressive disorder), overdiagnosis, or overtreatment The evidence review did not identify any studies of screening for depression during pregnancy.
The evidence review found one study of screening for depression among postpartum individuals (described as "mothers" or "women" in the study). Evidence from 1 RCT (n=462) indicated that the effects of screening on the number of patients identified as depressed, depression scores, and minor psychiatric conditions at 6-months follow-up (4-months post screening) were very uncertain (5).
The low-certainty evidence for mean infant body weight (kg) at six months suggests that screening for depression may lead to little to no difference compared to no screening (mean weight difference 0.05 kg higher, 95% CI 0.1 kg lower to 0.2 kg higher; SMD 0.06 higher, 95% CI 0.12 lower to 0.24 higher) (5).
Results from this study show that the effects of screening compared to usual care for the following outcomes was very lowcertainty: -relationship with partner and other supports (e.g., marital satisfaction scores); -reported/observed capacity to parent (e.g., parenting stress scores, parental distress scores); -childbearing parent-child interactions (e.g., parent-child dysfunctional interaction scores); -infant outcomes (e.g., number of child hospitalizations).The evidence review did not identify any studies that reported on suicidality.

Explanations (Tables 1 and 2)
a. Questionnaire was completed by participants.
b. Selective reporting identified. One of the primary outcomes listed in clinicaltrials.gov was listed as a secondary outcome in the publication. The primary outcome reported as such was statistically significant, and the other was not statistically significant.
c. Due to uncertainty and lack of empirical evidence to support, we have elected not to calculate an optimal information size Further, GRADE suggests a rule of thumb of a minimum of 400 events for dichotomous outcomes and 400 participants for continuous outcomes. As there were 462 participants in this study and without a known threshold for continuous outcomes, a judgement of "serious concern" was applied.
d. No information on how this information was collected, so many domains were unclear.
e. Reported by participants/caregivers.

UNDESIRABLE EFFECTS
How substantial are the undesirable anticipated effects?
No trials that met inclusion criteria assessed the harms of screening for depression during pregnancy or the postpartum period (e.g., false positives, overdiagnosis, overtreatment, or labeling/stigma). In one RCT, no adverse events were reported in either group of screened and not screened participants (5). However, it was unclear how adverse events were defined, collected, and reported in the study so the effect could not be estimated and evidence was considered very low-certainty.
The evidence review did not assess harms for infants related to screening their childbearing parent for depression during pregnancy or the postpartum period.

Explanations (Tables 1 and 2)
a. Questionnaire was completed by participants. b. Selective reporting identified. One of the primary outcomes listed in clinicaltrials.gov was listed as a secondary outcome in the publication. The primary outcome reported as such was statistically significant, and the other was not statistically significant. c. Due to uncertainty and lack of empirical evidence to support, we have elected not to calculate an OIS. Further, GRADE suggests a rule of thumb of a minimum of 400 events for dichotomous outcomes and 400 participants for continuous outcomes. As there were 462 participants in this study and without a known threshold, a judgement of serious concern was applied. d. No information on how this information was collected, so many domains were unclear. e. Reported by participants/caregivers.

OF EVIDENCE
What is the overall certainty of the evidence of effects? X Very low ○ Low ○ Moderate ○ High

○ No included studies
There is very low certainty that the evidence of effects represents the true effects of screening people for depression during pregnancy or up to one year postpartum. We are very uncertain about the absolute effects of screening on the critical outcomes of interest. This means that the true effects of screening are likely substantially different from the study data.
The very serious concerns with Risk of Bias with the Leung (2010) study include: Blinding of outcome assessors (self-report questionnaires) and selective outcome reporting (one of the primary outcomes listed in the protocol was reported as a secondary outcome in the study publication) (5).
There were also serious concerns with imprecision due to small sample size (n=462) (5).

VALUES
Is there important uncertainty about or variability in how much people value the main outcomes? ○ Important uncertainty or variability ○ Possibly important uncertainty or variability X Probably no A systematic review on patients' values and preferences was not conducted.
Two phases of knowledge translation activities were conducted among pregnant and postpartum individuals.
KT Phase 1 results (6): • A total of 15 participants (6 currently pregnant, 9 postpartum, all identifying as female gender) were recruited to participate in the online surveys and focus group discussions via teleconference to rate the importance of key outcomes. Most participants were from Ontario (n=9), with others from British Columbia (n=5) and Alberta (n=1). A total of 3 participants reported previous or current diagnoses of or treatment for depression. • Suicidality and 3 infant-related outcomes (infant health and development; mother-child interactions; capacity to parent) were rated the most critical to consider when deciding whether or not to get screened for depression. • Harms of screening (false positives, overdiagnosis, overtreatment, and labelling) were rated the lowest (but still • There was a very strong preference to be screened for depression among participants (median score to the question: "…how much would you want to be screened during pregnancy or the postpartum period" was '9 = Very Much').
• Many believed that screening at several times during pregnancy/postpartum period, and by several health care providers (e.g., midwives, doulas, nurse practitioners, pediatricians), would be beneficial.
KT Phase 2 results (7): • A total of 14 participants (4 currently pregnant, 10 postpartum, all identifying as female gender) were recruited to participate in the online surveys and focus group discussions via teleconference to rate the importance of key outcomes. Participants were from Ontario (n=7), British Columbia (n=4), Quebec (n=1), Prince Edward Island (n=1), and New Brunswick (n=1). A total of 5 participants reported a previous diagnosis or treatment of depression by a health professional and 2 participants reported currently receiving treatment for depression.
• In the final survey, participants rated almost all screening benefits as 'critical' to consider during decision making, whereas they rated all screening harms slightly lower, but still 'important'.
• Participants reported a preference for screening. When asked "Considering the risk of many of the harms and benefits of screening for depression during the pregnancy and postpartum period are not well known, how much would you want to be screened", participants median score was 7.5, where 5='Neutral' and 9='Very Much'.
• "[W]hile participants rated their preference to be screened fairly highly in the survey, focus group discussions indicated that participants felt most strongly about having a discussion with a healthcare provider about their mental health and wellbeing, rather than a formal screening process. They felt a discussion about depression with a primary health care provider during the pregnancy and [postpartum] period is critical[.]" • "[Participants] identified types of information that would be helpful as part of a screening discussion, including: differentiating between depression and the "baby blues", providing information on the prevalence of depression in the pregnancy and [postpartum] period, describing symptoms to be aware of, and describing potential next steps following a diagnosis (e.g., duration and availability of treatment)." • "Participants expressed concerns that without screening, they may not be capable of identifying symptoms of depression, or may not take initiative to seek input from a primary care provider, especially considering all of the stresses and changes one goes through as a new parent." The evidence was very uncertain from this single RCT regarding whether screening for depression may be beneficial for reducing the number of postpartum individuals classified as having depression 4 months after screening or for reducing their EPDS score (5).

BALANCE
The same study reported very low-certainty evidence on differences between those screened and not screened for the outcomes: mental health symptoms (as measured by the General Health Questionnaire), their relationship with partner and other supports, reported/observed capacity to parent, childbearing parent-child interactions, or infant outcomes (number of hospitalizations) (5). Low-certainty evidence suggested little to no effect of screening on infant body weight (5).
No evidence was found on the undesirable effects of screening such as suicidality, false positives, overdiagnosis, overtreatment, or labeling/stigma. No adverse events in either the screened and not screened group were recorded in the one RCT identified in the systematic review (5). There would be costs associated with a recommendation for routine screening of patients for depression during pregnancy and the postpartum period. These costs could include items such as: clinician and patient time, printing and technological costs of administering the screening instrument, and further costs related to increased demand for psychiatry/psychology resources, and treatment (e.g., counselling and/or medications). However, the total costs of a depression screening program in Canada are not known.

REQUIRED
If we choose a Recommendation AGAINST Screening: The resource requirements for following a recommendation against screening are unknown. It is possible that resource savings (e.g., primary care provider time, unnecessary follow-up and treatment costs) could occur in primary care settings that had previously administered formal screening instruments to all patients during pregnancy and the postpartum period. However, implications may vary depending on current practice and are not known with certainty.

X No included studies
A cost-effectiveness systematic review was not completed. However, given that universal screening is not current practice in all jurisdictions, implementation of screening will certainly include some costs (the exact costs are unknown). their healthcare provider. In the judgement of the task force, discussions regarding mental health and wellbeing can occur within the context of usual care and are consistent with a recommendation against screening.

COST EFFECTIVENESS
The task force believes this recommendation would be acceptable to some stakeholders, such as primary care providers and policymakers, as it highlights the lack of evidence to support screening but affirms the clinical practice of inquiring into depression, anxiety, and mood. However, a recommendation against screening may contradict current practice or policy in some jurisdictions. As such, some providers may feel discomfort about de-implementing screening due to concerns about 'missing' cases of this important health issue. However, based on the evidence examined, carrying out formal screening is not an effective solution to this concern.
Note that a recommendation against screening is not a recommendation against the role of clinicians in alerting patients about mood changes that can occur during pregnancy and the postpartum period.

FEASIBILITY
Is the intervention feasible to implement?
Recommendation FOR or AGAINST screening: A recommendation against screening is considered feasible by the task force. We do not know the degree to which this would represent a change in practice for clinicians. While most provinces/territories have guidance that recommend using a screening instrument, we do not know how many clinicians are following this. Nevertheless, even with a recommendation against routine screening, clinicians should still remain alert to signs or symptoms of depression in their patients as part of their usual vigilance and should alert all patients that mood can change during pregnancy and the postpartum period even though these changes don't necessarily mean the patient has a depressive disorder on for the intervention

RECOMMENDATION
The task force recommends against instrument-based depression screening using a questionnaire with cut-off score to distinguish "screen positive" and "screen negative" administered to all individuals during pregnancy and the postpartum period (up to 1 year after childbirth) (conditional recommendation, very low-certainty evidence).
This recommendation assumes that, as part of usual care during pregnancy and the postpartum period, care providers will inquire about and be attentive to mental health and wellbeing.

JUSTIFICATION
The supporting systematic review suggests that the additional benefit of screening all patients with a questionnaire with a cut-off score compared to usual care (which should include inquiry into mood and mental health) during primary care visits is very uncertain. Although no evidence was found on the harms of screening in our systematic review, evidence from other sources described below suggest the time and focus on screening could reduce opportunities to discuss other aspects of health during a perinatal primary care encounter as providers would be evaluating and potentially referring all patients who screen positive, in many cases unnecessarily. Screening could lead to an increase in false positives, false negatives, unnecessary referrals and diagnostic evaluation, and overdiagnosis for some patients.
A false positive can occur when the patient meets a screening cut-off score and is sent for additional psychiatric evaluation, which finds they do not actually meet the diagnostic criteria for depression. A recent individual patient data meta-analysis provides accuracy information for the EPDS, the tool used in the one trial we identified (8). Based on a prevalence of 8%, screening 100 patients with the EPDS using the common cut-off score of 13 would result in 5 true positives, 3 false negatives, 5 false positives, and 87 true negatives (8,9). This means that some patients who are screened will be sent for an unnecessary additional assessment.
Overdiagnosis could occur in patients with mild temporary symptoms, who might meet a screening cut-off score, leading to further evaluation and possible referral to specialty mental health services, but who would not benefit as the symptoms would subside on their own. Given the significant challenges to accessing mental health services in Canada, the unnecessary redirection of resources from the treatment of patients with mental health disorders could be an unintended harm of screening. Spending even one to two1-2 minutes per clinical encounter reviewing the results of a formal screening instrument with no proven value could consume a significant amount of time during a 15-minute encounter. In the task force's Force's view, this could detract from the ability of the clinician to have a meaningful and empathetic discussion about the health of the patient. As noted above, about 10% of all patients screened using a questionnaire and cut-off score would have to receive additional assessment or referrals, and thus the resource implications also extend beyond the initial clinical encounter. The task force is mindful of the resource constraints faced by our primary health care system and as such makes recommendations against interventions when the resource implication of a particular health intervention are certain to be important and benefits have not been demonstrated (10).

SUBGROUP CONSIDERATIONS
This recommendation applies to all patients, including patients who have characteristics that may suggest elevated risk of depression (e.g., trauma in early life, family history of depression) (11). This recommendation does not apply to pregnant and postpartum individuals with a personal history or current diagnosis of depression or another mental health disorder, those currently receiving assessment or treatment for mental disorders, those receiving care in psychiatric or other mental health settings, or those who are seeking services due to symptoms of depression.

IMPLEMENTATION CONSIDERATIONS
The term 'screening' in this recommendation refers to a routine process in which primary care providers administer an instrument such as a questionnaire to every pregnant or postpartum individual not already reporting symptoms of depression and then use a cut-off score to determine a follow-up action for those at or above the cut-off score. The task force recommends against the addition of such a screening process because of the absence of evidence that it adds value beyond discussions about overall wellbeing, depression, anxiety, and mood that are currently a part of established perinatal clinical care.
Ten provinces and territories in Canada provide guidance documents (e.g., best practice recommendations, care pathways, perinatal records) that suggest asking patients about current depression, anxiety, or mood during pregnancy or the postpartum period as part of usual clinical care (Appendix 3,4). Nine provinces and territories provide guidance documents that suggest primary care providers (e.g., public health nurses, family physicians, midwives) screen patients with instruments such as the Edinburgh Postnatal Depression Scale (EPDS) during pregnancy or the postpartum period (Appendix 3,5). This guidance includes recommended cut-off scores and follow-up actions as part of screening. Among these nine provinces and territories, seven also provide a place to enter scores for depression screening instruments in medical record forms used during pregnancy or the postpartum period (Appendix 3,5). Guidance on which questionnaires to use, when to administer them, and pre-defined cut-off scores varies across provinces and territories.
Some jurisdictions included screening as part of their standard perinatal care without fully defining it as such, whereas we use the term 'screening' to indicate a comprehensive, systematic process applied in a standard way and with defined follow up diagnostic processes for all individuals in a defined group (e.g., all patients in a clinical setting with specific characteristics). As screening practices vary across Canada (Appendices 3,4,5), jurisdictions may reconsider the use of such screening in settings where it is currently implemented.
As noted above, the task force definition of screening in this context means that the recommendation against screening emphasizes the importance of good clinical practice where clinicians should inquire and be alert to changes in physical and mental health symptoms of their patients. Given the health implications of depression during pregnancy and the post-partum period, it is essential that providers inquire about and be attentive to mental health and wellbeing. If providers are uncertain about how to engage in these discussions with patients, they may consider referring to questionnaires for discussion prompts (without engaging in formal screening by using the questionnaire score for determining subsequent actions).

MONITORING AND EVALUATION
Clinician awareness of this recommendation against screening is a performance measure for this guideline. The task force will monitor evidence related to this guideline and will update the recommendation if new evidence becomes available that could influence its direction or strength.

RESEARCH PRIORITIES
There is only 1 randomized controlled trial assessing the benefits and harms of screening for depression versus no screening during the postpartum period. There were none during pregnancy. Trials that compare screening to usual clinical care, where those identified as depressed in either arm receive the same level of care, are needed in order to isolate the effectiveness of screening as an intervention. Outcomes should include both maternal and infant-related benefits and harms. As experiences of pregnancy and the postpartum period can vary based on factors such as culture, ethnicity, socioeconomic status, geographic region, and other social determinants of health (12), studies that reflect and provide evidence for the diversity of the Canadian population would also be helpful.