Perinatal depression has been defined as encompassing ‘major and minor depressive episodes that occur either during pregnancy or within the first 12 months after delivery’ (Gavin et al, 2005: 1071), and includes new episodes of depression and pre-existing depression (Matthey, 2004). It is important to identify perinatal depression as there is evidence to suggest it is associated with self-reported poor maternal health in pregnancy (Orr et al, 2007) poor obstetric and neonatal outcomes (Marcus, 2009) and has a substantial impact on the mother and her partner (Boath et al, 1998), mother–baby interaction (Murray et al, 1996) and the longer-term emotional and cognitive development of the infant (Mensah and Kiernan, 2011). However, it is estimated that less than 50% of cases of maternal depression are identified in routine clinical practice (Oates et al, 2011).
Screening strategies to identify perinatal depression have been advocated (Leverton, 2005), although in the UK, no single identification strategy meets the National Screening Committee (NSC) criteria for use as a screening test (Hill, 2010). While diagnostic utility represents an important criterion to appraise the accuracy of a test, an appropriate test must demonstrate it is acceptable to the population in whom the test is used (NSC, 2014), and as such is directly influenced by the views and experiences of this population.
Several self-report, perinatal-specific questionnaires designed to identify perinatal depression exist, of which the 10-item Edinburgh Postnatal Depression Scale (EPDS) (Cox et al, 1987) is the most extensively validated and widely used (Hewitt et al, 2010). Although women's acceptability of the EPDS has been demonstrated (Matthey et al, 2005; Buist et al, 2006; Gemmill et al, 2006; Leigh and Milgrom, 2007; Slade et al, 2010), other qualitative evidence has identified concerns with questionnaire administration, feelings of personal intrusion and manipulation of responses to mask low mood and stigmatisation (Shakespeare et al, 2003; Cubison and Munro, 2005; Poole et al, 2006; Chew-Graham et al, 2009). In addition, the EPDS is perceived as both a barrier and facilitator to support and referrals (Armstrong and Small, 2010).
In recent years, perinatal mental health policy has advocated a shift away from use of multi-item measures to the use of brief case-finding questions (National Institute for Health and Care Excellence (NICE), 2007). These two brief questions, often referred to as the ‘Whooley’ questions, enquire about depressed mood and loss of interest in activities (Whooley et al, 1997) (Box 1), plus an additional question about the need for help asked of women who answer ‘yes’ to either of the initial questions (Arroll et al, 2005). The guidance has recently been updated (NICE, 2014) and recommends that health professionals use the case-finding questions to identify perinatal depression; the use of the ‘help’ question for positive responses to either question is not recommended in the updated guidance. The new guidance also asks health professionals to consider use of two additional questions to identify perinatal anxiety.
In the UK, most antenatal and postnatal care is provided by community midwives until 10 days after delivery, when care is provided by community health visitors. In addition, GPs offer a 6-week postnatal health check. Therefore, substantial opportunity exists for these health professionals to discuss maternal mood at routine contacts at health clinics and at child health and immunisation clinics.
A systematic review of qualitative studies of women's views of perinatal depression strategies highlighted that setting, timing, confidentiality and prior notification of questionnaire administration were important to women's acceptability of an identification method. However, it was unable to identify evidence about the acceptability of the Whooley questions (Brealey et al, 2010).
A recent study has provided some limited evidence of the case-finding questions’ acceptability in a perinatal setting to identify depression (Darwin et al, 2015). The study validated the use of case-finding questions in 191 pregnant women during a routine antenatal appointment at 13 weeks' gestation and also examined women's views of the case-finding questions using semi-structured interviews with 22 women at 16 and 33 weeks' gestation and repeated around 10 postnatal weeks.
Emergent themes focused on disclosure of symptoms. The case-finding questions in this study were not administered in the postnatal period, so views about the case-finding questions elicited at the postnatal interview relate to women's recall regarding the experience of being asked the questions during pregnancy.
Interview participants were restricted to only those women identified as ‘possible’ or ‘probable’ cases of depression identified by the self-reported EPDS measure; no gold-standard diagnostic criteria were used to confirm a depressive episode.
The acceptability of a case-finding approach for identifying perinatal depression such as the two brief case-finding questions, in line with NSC criteria for other screening strategies, is that case-finding questions should be acceptable to all members of the population in which they will be used (Mant and Fowler, 1990). This paper presents the results of a qualitative content analysis of the views of pregnant women and new mothers who were asked the NICE-recommended case-finding questions to identify perinatal depression, and were administered diagnostic criteria to confirm presence or absence of depression during both their antenatal care and during the early postnatal period in order to examine the acceptability of the questions.
Methods
This study used intra-method mixing (Johnson and Turner, 2003). This variation of mixed methods technique employs a survey to elicit both qualitative data in the form of free-text comments and quantitative data in the form of three closed questions, to examine pregnant women's and new mothers' views of the recommended case-finding questions to identify perinatal depression.
The case-finding questions were administered to pregnant women and new mothers in the context of a diagnostic test accuracy (DTA) study where the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) standardised diagnostic criteria were used to confirm presence or absence of depression (American Psychiatric Association, 1994); the DTA methods have been described elsewhere (Mann et al, 2012).
Pregnant women (approximately 26–28 weeks' gestation) were recruited over 7 weeks between September and November 2010 at the maternity unit of a general hospital where more than 90% of local women receive their antenatal care. Written informed consent was obtained from all participants.
In the antenatal study phase, the case-finding questions and acceptability survey were administered concurrently as a self-report questionnaire in the antenatal clinic. In the postnatal study phase, study participants were re-contacted after birth of their baby at 5–6 weeks postnatally and invited to complete a self-report postal questionnaire, which consisted of the same case-finding questions and acceptability survey used in the antenatal phase of the study.
The acceptability questionnaire was adapted from a survey developed to examine the acceptability of the EPDS in newly-delivered Australian women (Gemmill et al, 2006) and was concurrently administered with the case-finding questions to capture women's views after the case-finding questions were self-completed.
A simple one-page form asked women to respond to three closed questions and one directive free-text question (Box 2).
Participants were invited to respond to a directive free-text question to provide women the opportunity to express their opinions regarding the case-finding questions. Directive free-text questions have been defined as an appropriate method to capture the views and opinions of contextual issues that are considered relevant to survey respondents (Mackichan et al, 2010; Ong et al, 2006). They are useful for allowing respondents the opportunity to voice their opinions and corroborate answers to closed questions (O'Cathain and Thomas, 2004), and can be considered useful for eliciting contextual information at the time a questionnaire is administered.
Recommendations for reporting free-text comments were adopted, which consist of reporting the minimum and maximum length of sentences written by participants, the number of participants who made comments within each theme and the number of participants who failed to make any free-text responses (O'Cathain and Thomas, 2004; Garcia et al, 2004).
Free-text comments were analysed by qualitative content analysis (Elo and Kyngäs, 2008) using the conventional content analysis method (Hsieh and Shannon, 2005). The comments were read multiple times in order to gain immersion in the data. Content-characteristic words in free-text comments were identified and free-text comments sorted into categories, from which emergent themes were identified. Elo and Kyngäs (2008) recommend use of exemplar comments in content analysis, and these are provided with the descriptive text accompanying each theme.
In order to contexualise the exemplar comment used for each theme, respondents were identified according to four possible outcomes related to the identification of cases of depression.
Exemplar comments are presented according to the agreement between the case-finding questions and the DSM-IV standardised diagnostic criteria, which were the gold-standard criteria used to determine whether depression was present or absent in respondents. There are four possible outcomes when comparing agreement between the questions and the gold-standard criteria:
Exemplar comments are reported exactly as they were written by participants. The three closed questions asked about the desirability and comfort of asking pregnant women and new mothers the case-finding questions, were analysed in PASW statistics using descriptive statistics.
Participants' sociodemographic data were obtained from background information collected at recruitment. Potential differences in age and ethnicity were examined between women who provided free-text comments and those who did not using independent samples t-test and chi-squared test respectively. Level of statistical significance was set at P<0.05.
Ethical approval
Ethical approval for the study was granted by Yorkshire and the Humber–Leeds Central Research Ethics Committee (Ref: 10/H1313/17).
Results
The participants' sociodemographic characteristics are presented in Table 1.
| Characteristic | n (%) |
|---|---|
| Ethnicity | |
| White British | 81 (53.3) |
| White other | 5 (3.3) |
| Mixed (White and Black) | 4 (2.6) |
| Mixed (White and South Asian) | 3 (2.0) |
| Black | 6 (3.9) |
| Indian | 5 (3.3) |
| Pakistani | 38 (25.0) |
| Bangladeshi | 5 (3.3) |
| Other | 5 (3.3) |
| Parity | |
| Primiparous | 73 (48) |
| Marital status | |
| Married (first marriage) | 79 (52.0) |
| Remarried | 10 (6.6) |
| Single (never married) | 60 (39.9) |
| Divorced | 3 (2.0) |
| Cohabitation | |
| Living with baby's father | 122 (80.3) |
| Living with another partner | 1 (0.7) |
| Not living with partner (in a relationship) | 20 (13.2) |
| Not living with a partner (not in a relationship) | 9 (5.9) |
| Housing | |
| Buying house (mortgage/loan) | 68 (44.7) |
| Own outright | 13 (8.6) |
| Rents | 54 (35.5) |
| Live there rent-free | 17 (11.2) |
| Mother's highest educational qualification | |
| None | 22 (14.5) |
| School | 35 (23.0) |
| Further | 31 (20.4) |
| Higher | 49 (32.2) |
| Other | 12 (7.9) |
| Don't know | 3 (2.0) |
| Mother's employment status | |
| Currently employed | 94 (61.8) |
| Previously employed | 34 (22.4) |
| Never employed | 24 (15.8) |
| Mother's smoking status | |
| Currently a smoker | 22 (14.5) |
| Not currently smoker | 42 (27.6) |
| Never a smoker | 88 (57.9) |
| Self-reported history of depression diagnosis | |
| Never been diagnosed by GP with depression | 128 (84.2) |
| 1 or ³2 episodes of depression | 24 (15.8) |
| Mother's age in years, mean (standard deviation) | 27.4 (5.8) |
In the antenatal study phase, 155 self-report surveys were returned, of which 152 were available for analysis (three women withdrew consent within 2 days of completing the survey). Twenty (13.2%) pregnant participants returned their acceptability questionnaire without making a free-text comment about the acceptability of the case-finding questions. There was no difference between participants who provided a comment to the antenatal free-text question and those who returned their questionnaire without a response in terms of age (P=0.856) and ethnicity (P=0.878).
Of the 152 women eligible for postnatal study follow-up, four were excluded (three births were >6 weeks premature and the infants were neonatal inpatients; and one infant was in the child safeguarding service) and two withdrew consent. Of the postnatal respondents, 97 (66.4%) completed the self-report survey. Women who did not complete the postnatal study phase were younger (P=0.002), less educated (P=0.02) and less likely to be employed (P=0.01) than those women who participated postnatally.
Five postnatal participants (5.2%) returned their acceptability questionnaire without making a free-text comment with regard to answering the two case-finding questions. Of the antenatal and postnatal participants, 94.7% (n=144) and 92.8% (n=90) answered ‘yes’ to the question about the desirability of asking women about their mental wellbeing, respectively; eight (5.3%) antenatal and seven (7.2%) postnatal participants answered ‘not sure’. None of the participants gave a negative response to this question. Of the antenatal and postnatal participants, 99.3% (n=151) and 96.9% (n=94) respectively indicated that they felt fairly to very comfortable answering the case-finding questions.
The analysis of the free-text comments is presented according to four themes.
Theme 1: General acceptability comment
Fifty antenatal and 37 postnatal free-text responses related to a general acceptability comment. The comments were one-word answers, such as ‘ok’, ‘no problem’, ‘fine’, or one sentence where no other opinion was offered.
‘I felt fine answering the questions. These did not bother me at all.’ (TN)
Theme 2: Completing and answering the questions
There were 42 antenatal and 18 postnatal free-text responses that referred to completing and answering the case-finding questions.
Comments related to general completion of the questions and referred to the level of difficulty, relevance and non-intrusiveness completing the questions. The majority of women reported that they found the questions easy, simple and straightforward to answer.
‘The questions are simple and basic and therefore require little thought … you don't really feel under any pressure to answer them.’ (TN)
Some women commented about the non-intrusiveness of the questions and referred to the normality of completing these types of questions.
‘Normal questions … didn't feel any different to answering any questions in general.’ (FP)
Several pregnant women referred to feeling more comfortable answering the questions in written format within the antenatal clinic and referred to the privacy this afforded them to be able to admit how they were feeling or answer the questions comfortably.
‘It felt better knowing that, on paper, I wasn't been judged and could get help if I needed.’ (TP)
Seven women (11.7%) thought that the questions were too brief and questioned if the generic format would elicit accurate information about depression.
‘Easy to answer but a little bit too general. Didn't like the “yes” or “no” answers as believe you can feel things to a certain extent.’ (TN)
Only one comment was overtly negative with specific criticisms related to the 1 month recall period and the unclear and ambiguous format of the questions.
Theme 3: The value of asking the questions
Thirty-nine antenatal and 17 postnatal participants referred to the value of asking the questions.
Comments related to the helpfulness to identify potential problems with mental health in pregnancy and postnatally. The questions were seen as a way for women to admit to problems, confide in someone, a potential means of help-seeking and a way to alert the health professional to difficulties.
‘I felt fairly comfortable, I feel these questions need to be asked as not all women have partners/someone to turn to, to talk, or they have these but get no help at all. A questionnaire might be easier to ask for help than 1:1 or on the phone.’ (FP)
Theme 4: Expressing own personal feelings
Eighteen antenatal respondents and 39 postnatal respondents referred to their own personal feelings when answering the questions.
Women felt positive about having the opportunity to share, express and reflect on their personal feelings and experiences; comments were all positive in affective tone. Women indicated that answering the questions had made them more aware of their feelings, which helped them understand or admit their feelings.
‘I felt fine answering them because they are not questions your nearest relatives etc ask you. It feels like you're getting a little bit off your chest by admitting you've felt a bit down.’ (TP)
Some antenatal respondents used the questions to reflect on their perception of their current pregnancy or past experience of pregnancy. Postnatal respondents justified their responses to the questions in terms of emotional and physical feelings and experiences as a consequence of perceived factors associated with motherhood, such as hormonal changes after delivery, physical and emotional stress, and fatigue associated with caring for a newborn baby.
‘I find it hard to know how to answer these questions as a lot of the feelings could be put down to hormonal changes following the birth or extreme tiredness with having sleepless nights or just finding the transition from one to two children a difficult one.’ (FP)
Comments of women who met the criteria for depression
Of the 17 antenatal participants who were depressed according to the DSM-IV criteria, 16 provided comments that were positive in affective tone, and one respondent did not comment. All 18 postnatal participants who were depressed, according to the DSM-IV criteria, provided comments. Fourteen of the postnatal comments were positive and indicated the questions were acceptable. They felt that the questions were easy to complete, helpful and important to ask. The questions provided participants with the opportunity to reflect or become aware of their own feelings.
‘It's a good idea to ask those questions because sometimes you don't have anyone to talk to or understand why you feel like you do.’ (TP)
Comments indicating discomfort with questions
Of the 132 antenatal participants who provided free-text comments, there were two respondents whose free-text comments indicated that they felt some discomfort answering the questions. However, in contrast, both of these respondents reported feeling ‘comfortable’ and ‘fairly comfortable’ when asking question 2 of the acceptability questionnaire (see Box 2).
‘I felt anxious because I do not usually like to express my feelings because I do not want people to think differently about me.’ (FP)
Four of the 97 postnatal participants indicated some discomfort completing the questions.
‘I felt slightly uncomfortable but I suppose it's a good way of detecting signs of PND or anxiety in new mothers.’ (TN)
Three participants who met DSM-IV depression criteria indicated some discomfort completing the questions. None of these comments were overtly negative; comments were qualified by reasons why the participant felt discomfort, for example, feeling it was difficult to admit feelings or admit they might be struggling, worry regarding being judged, and the realisation at actually ‘seeing’ their own unhappiness on paper.
‘… bit uncomfortable, haven't had time to think about myself, since the baby, but I should feel a lot happier than I do I have realised.’ (TP]
Discussion
This study examined the views of both pregnant women and new mothers who were administered brief case-finding questions recommended by NICE (2007; 2014) to identify depression in the antenatal and postnatal period. The overall findings indicate that the perinatal women in this study found the questions acceptable; they were perceived as easy and straightforward to answer. The women in this sample felt it was desirable to be asked questions about mental wellbeing and very few comments indicated discomfort or negativity. Women felt the questions could facilitate help-seeking and could alert health professionals to potential difficulties. Women also believed that the questions contributed to self-awareness and enabled them to reflect on and express their current feelings. Postnatal women tended to reflect on their circumstances regarding the transition to motherhood, or justified their responses in terms of challenging circumstances associated with a new baby, i.e. lack of sleep, physical complications during delivery or in the immediate postnatal period.
One study that has examined pregnant women's views of the case-finding questions administered in an antenatal setting (Darwin et al, 2015) restricted interviews to women with possible depression identified by the EPDS, a measure to identify self-reported symptoms, which is not considered to be the best available strategy for diagnosis of a depressive episode. The current study used standardised diagnostic criteria considered to be a gold-standard method to confirm presence or absence of depression. This identifies four possible outcomes associated with using the case-finding questions in all women who were asked these questions, and contextualises the findings of the themes and exemplar comments according to ‘true’ depression status of respondents. A key finding in this study is that the majority of perinatal women who were identified as meeting the diagnostic criteria for depression were positive about the use of the case-finding questions and found them acceptable. Only comments from one antenatal and four postnatal women indicated some discomfort and these were endorsed with justification of their responses. In the current study, none of the women who were identified with depression mentioned stigmatisation or inability to disclose fears associated with the brief case-finding questions. A small number of women in the current study who were non-depressed speculated that their feelings about the questions may be different dependent on their mental health and questioned the personal relevance, meaning of the questions and the benefit of asking generic-format questions. Only three non-depressed pregnant women who made comments regarding the format and meaning of the case-finding questions speculated that the questions might be difficult to answer if they had not been feeling emotionally well. Four non-depressed postnatal women did comment that their answers may be dependent on how they were feeling at the time, which suggests these women may have intentionally hidden their true feelings under different emotional circumstances.
Darwin et al (2015) highlighted women's concerns about the relevance of asking the questions in the context of maternity appointments that focus on pregnant women's physical health, responding to perinatal mental health questions truthfully, concerns about disclosure of symptoms of depression, and the meaning and consequences of assessment. This is in contrast to the current study, which found the majority of women found the questions acceptable when administered in the antenatal and postnatal period.
‘When mental health questions are integrated into routine care, women should be informed of their use in advance, along with the relevance and meaning of the assessment, possible intervention pathways or consequences of follow-up assessment or referral’
The majority of antenatal care in the UK context is provided by community midwives, with postnatal care provided by midwives, health visitors and GPs who offer a 6-week postnatal health check. Postnatal checks by GPs and health visitors at child health clinics and immunisation clinics present as an ideal opportunity where case-finding for perinatal depression and discussion of maternal mood could occur. However, a caveat to such ‘screening’ or case-finding activities in routine maternity care is that appropriate protocols and services should be available where women respond positively to the two case-finding questions. Lack of appropriate interventions or follow-up services is a cause for concern and, if not in place, may deter health professionals' willingness to undertake case-finding, and deter women from answering honestly. Women in this study reported that asking the questions could facilitate help-seeking, indicating that they were open to being asked these questions; however, Darwin et al (2015) found that the rushed nature of tasks relating to antenatal appointments created a barrier to potential disclosure of feelings; thus indicating that the barrier relates to the context and process of verbally asking the questions in a time-limited, face-to-face consultation, rather than the questions themselves.
In this study, a small number of women who were not depressed were critical of the questions and appeared to query the personal relevance and meaning of the questions. Prior knowledge of the use and benefits of using the case-finding questions may facilitate women feeling that they can answer the questions with honesty. The method of questioning may also be of relevance; the current study asked women to complete the questions on paper in privacy, which may have facilitated acceptability, rather than face-to-face as in the Darwin et al (2015) study.
The implication for clinical practice in maternity care settings would suggest that when mental health questions are integrated into routine care, women should be informed of their use in advance, along with the relevance and meaning of the assessment, possible intervention pathways or consequences of follow-up assessment or referral. Use of written response method to the questions, e.g. ‘paper and pencil’, response could be considered.
Limitations
The use of free-text question to elicit women's views at the time the case-finding questions were administered could be seen as a limitation of the study. Free-text questions have been criticised as data can range from one word to several sentences or paragraphs, data may be unwieldy, disparate in nature, difficult to abstract and therefore not fully satisfy the standards for qualitative methodology. Women were also limited to the amount of detail in comments by the amount of space to respond on the questionnaire. Use of face-to-face interviews would be useful to explore women's in-depth views of the case-finding questions in both antenatal and postnatal care. However, although free-text data ranged from one word to a paragraph, this study was transparent reporting this and comments consistently demonstrated women's acceptability of the questionnaires, therefore in line with other authors who have used this approach it does not negate their worth (Garcia et al, 2004; O'Cathain and Thomas, 2006).
In this sample, the non-responders in the postnatal phase were younger, less educated and less likely to be employed. This may reflect difficulties surrounding the competing demands associated with the transition to motherhood for younger mothers and a younger population more likely to be transient in nature (renting or living with friends). In addition, some women may not have responded due to postnatal depressive symptoms; conversely, some younger women (18–19 years) indicated that they were already being monitored for mood/psychological health in a pilot project in the local area for young, teenage mothers and therefore may have felt further mental health enquiry after the birth of the baby was inappropriate or unnecessary. Characteristics of non-responders observed in the current study were the same as reported in another DTA validation study of the case-finding questions in an American postnatal sample (Gjerdingen et al, 2009), suggesting younger women may be under-represented in these types of perinatal study.
A qualitative study using in-depth interviews in pregnant women and new mothers who have been asked the case-finding questions to identify perinatal depression, alongside standardised diagnostic criteria to confirm the presence or absence of depression, would be useful to contextualise this study's findings. It would also examine the acceptability of the two case-finding questions across the antenatal and postnatal care pathway, including the acceptability of the additional two mental health questions that have been recommended in the updated clinical guidance to identify perinatal anxiety (NICE, 2014).
Conclusion
The case-finding questions recommended by NICE (2007; 2014) to identify perinatal depression were acceptable to the majority of perinatal women in this study and the case-finding questions were perceived as simple and straightforward to answer. An important issue identified in the current study was that women who met the diagnostic criteria that confirmed presence of depression felt comfortable completing the questions. This is important because the focus of using such questions in routine practice is to identify and manage those women experiencing perinatal depression. However, information-sharing by health professionals such as midwives and health visitors regarding the use and benefits of a method such as the case-finding questions, the relevance of the questions, the purpose and consequence of mental health assessment, and purpose of potential follow-up of ‘possible’ cases of depression should be discussed with women and may facilitate the acceptability of this approach in routine practice.