Anxiety measures validated in perinatal populations: A systematic review

“…1978); RDC = Research Diagnostic Criteria; CIS = Clinical Interview Schedule (Goldberg et al., 1970); HRSD = Hamilton Rating Scale for Depression (Hamilton, 1960); PSE = Present State Examination (Wing et al., 1974); PAS = Psychiatric Assessment Schedule; SPI = Standardised Psychiatric Interview.”

“Sample:PC = primary care e.g. women at antenatal clinic; R = subsample of a research study; C = community Self-report instruments: CC-D&A = Trait anxiety & depression scales (Costello & Comrey, 1967); BAI = Beck Anxiety Inventory (Beck & Steer, 1991); BDI = Beck Depression Inventory (Beck, 1961); ENES = Eysenck Neurot…”

“…has shown robust psychometric properties in non-perinatal samples and had substantially better validity coefficients than the GHQ-12 in discriminating cases of anxiety and mood disorders in the Australian National Survey of Mental Health and Well-being (Kessler et al., 2002; Furukawa et al., 2003).”

“Only the GHQ-28 has a specific anxiety subscale.”

“…that DASS and DASS-21 scores discriminated between diagnoses of major depressive disorder, panic disorder, social anxiety disorder , obsessive compulsive disorder and specific phobia; and there is a clear three-factor structure with items loading onto the expected factors (Antony et al., 1998).”

“Self-report measures of a construct should correlate more highly with another measure of the same construct, than with a purported measure of a different construct (Watson et al., 2008, Clark and Watson, 1991).”

“%) in Goldberg et al. (1997).”

“Six and four item versions of the state scale of the STAI have been constructed and validated for use in pregnant populations (Marteau and Bekker, 1992).”

“The scale was designed to include minimal reference to somatic symptoms (Snaith, 2003).”

“…has shown robust psychometric properties in non-perinatal samples and had substantially better validity coefficients than the GHQ-12 in discriminating cases of anxiety and mood disorders in the Australian National Survey of Mental Health and Well-being (Kessler et al., 2002; Furukawa et al., 2003).”

“In a UK study of 8,323 women, 13% of women reported anxiety symptoms at either eight weeks or eight months postpartum and of these two-thirds had experienced anxiety in pregnancy (Heron et al., 2004).”

“Validity of any self-report measure depends on recalibration for the population under study (Geisinger, 1994).”

“However, the utility of the EPDS appears to vary in different samples (Gibson et al., 2009) and there is evidence that the EPDS has three items that also measure anxiety.”

“A large prospective study (n = 12,361) confirms that antenatal anxiety is one of the strongest predictors of postnatal depression indicating that antenatal screening for anxiety may be helpful in reaching women at risk of postnatal depression (Milgrom et al., 2008).”

“The Hopkins-Symptom Checklist 25 (Derogatis et al., 1974) is a self-report inventory of 10 anxiety and 15 depression items derived from the SCL-90.”

“54 5th month postnatal (1994a) 50.0 84.3 - - - (Nott & Cutts, 1982) Unmodified 6/7 80 89 50 97 19 Author modified 6/7 83 87 53 96 16 (Sharp, 1988)a 5/6 77 78 52 96 22 Note: aClinical Interview Schedule cut-off point of 17/18; bSampling strategy prohibited calculation of positive predictive value, therefore the authors…”

“Firstly, the Inventory of Depression and Anxiety Symptoms (IDAS; Watson et al., 2007) is a new multi-dimensional measure of symptoms of anxiety and depression, containing 64 items which yielded 10 factors in multiple samples relating to symptoms of: suicidality, lassitude (fatigue, lack of energy,…”

“Although prevalence data of anxiety disorders is limited, a review tentatively suggests that rates of OCD and GAD are higher in perinatal populations than in the general population, whilst rates of panic disorder and posttraumatic stress disorder (PTSD) are comparable (Ross and McLean, 2006).”

“Pregnancy specific stress has predicted poor birth outcomes better than state anxiety and perceived stress (Lobel et al., 2008).”

“…linked with intrauterine artery resistance leading to low birthweight (Teixeira et al., 2009; Field et al., 2010), low apgar scores (Berle et al., 2005), suboptimal changes in fetal heart rate and motor activity (DiPietro, 2010) and complications of labour and delivery (Johnson and Slade, 2003).”

“In two samples of women interviewed at 6 weeks postpartum Matthey et al. (2003) found 16.2% and 10.4% of women had anxiety disorders only, whereas 5.6% and 2.6% had major or minor depression only, and 4.2% and 2.1% had comorbid depression and anxiety highlighting the importance of screening for…”

“Quality of studies was assessed using a combined checklist based on the quality assessment of diagnostic accuracy studies (QUADAS) (Whiting et al., 2003) and a checklist developed by Mirza and Jenkins (2004).”

“In general, internal consistency is good, ranging from 0.91 – 0.95 for the state subscale of the full version (Marteau and Bekker, 1992; Grant et al., 2008) and 0.82 for the 6 item version (Marteau and Bekker, 1992).”

“Test-retest reliability (the ability of a test to yield consistent scores over time) should be greater than .80 over one to two weeks or .70 or greater over one month (Boyd et al., 2005).”

“A study using diagnostic interviews with women eight weeks postpartum found 8.2% of women had generalized anxiety disorder (GAD), 2.7% had obsessive compulsive disorder (OCD), 1.4% had panic disorder and 4.1% social anxiety disorder (Wenzel et al., 2005).”

“…review indicates that postnatal maternal anxiety is associated with deleterious outcomes for children such as alterations in emotional and conduct problems, and adverse effects on cognitive and social developments, although effects on infant temperament are inconclusive (Glasheen et al., 2010).”

“…women at antenatal clinic; R = subsample of a research study; C = community Self-report instruments: CC-D&A = Trait anxiety & depression scales (Costello & Comrey, 1967); BAI = Beck Anxiety Inventory (Beck & Steer, 1991); BDI = Beck Depression Inventory (Beck, 1961); ENES = Eysenck Neuroticism…”

“Prenatal anxiety has been linked with intrauterine artery resistance leading to low birthweight (Teixeira et al., 2009; Field et al., 2010), low apgar scores (Berle et al., 2005), suboptimal changes in fetal heart rate and motor activity (DiPietro, 2010) and complications of labour and delivery…”

“One study shows that women with anxiety disorders in pregnancy are nearly three times more likely to present with postnatal depression (Sutter-Dallay et al., 2004).”

“34 weeks gestation HADS - Factorial Test-retest Internal consistency 8 Kitamura et al., 1989d 13 Japan 108 PC Antenatal 12 weeks gestation GHQ-30 SADS (RDC) Criterion: Sens, Spec, MR, PPV, NPV Discriminant Function Analysis - 8 Kitamura et al., 1993d 14 Japan 108 PC Antenatal 12 weeks gestation GHQ-30 - Compared 4 dif…”

“Cut-offs used in perinatal studies ranged from 5 in a Nigerian sample (Abiodun, 1993) to 7/8 in a Japanese sample (Kitamura et al., 1989, 1994a, 1994b).”

“Longer term effects of high antenatal anxiety can include externalizing problems and self-reported anxiety at ages 8-9 (Van den Bergh and Marcoen, 2004) and high impulsivity and lower scores on cognitive tests at ages 14-15 (Van den Bergh et al., 2005).”

“A number of measures have been developed that are specific to pregnancy and consider aspects related to anxiety such as worry or daily hassles but do not measure anxiety itself (Cambridge Worry Scale: Green et al., 2003; Pregnancy Experience Scale: DiPietro et al., 2004).”

“There is also evidence that the depression subscale of the EPDS correlates more highly with other measures of anxiety than the anxiety subscale does, indicating poor convergent validity (Brouwers et al., 2001).”

“Self-report measures of a construct should correlate more highly with another measure of the same construct, than with a purported measure of a different construct (Watson et al., 2008, Clark and Watson, 1991).”

“A number of measures have been developed that are specific to pregnancy and consider aspects related to anxiety such as worry or daily hassles but do not measure anxiety itself (Cambridge Worry Scale: Green et al., 2003; Pregnancy Experience Scale: DiPietro et al., 2004).”

“…has been linked with intrauterine artery resistance leading to low birthweight (Teixeira et al., 2009; Field et al., 2010), low apgar scores (Berle et al., 2005), suboptimal changes in fetal heart rate and motor activity (DiPietro, 2010) and complications of labour and delivery (Johnson and…”

“Reliability is a measure of the properties of a test within a particular sample rather than a property of a test per se and therefore should be computed in each sample (Vacha-Haase et al., 2000).”

“Clinician judgement and use of the gold standard will also have an effect on the utility of validity coefficients (Joiner et al., 2005).”

“5% of women reported intense anxiety or panic attacks occasionally or often between 6 and 9 months postpartum in a computer assisted telephone interview (Woolhouse et al., 2009).”

“Therefore an evaluation must be made as to whether the sample with whom the test was designed is sufficiently similar to the test sample to ensure little variability in functioning of the self-report measure (Myers and Winters, 2002).”

“A Dutch study improved the ability of the K10 to identify anxiety disorders by adding five further items pertaining to GAD, social anxiety disorder, panic and agoraphobia (Donker et al., 2010).”

“…linked with intrauterine artery resistance leading to low birthweight (Teixeira et al., 2009; Field et al., 2010), low apgar scores (Berle et al., 2005), suboptimal changes in fetal heart rate and motor activity (DiPietro, 2010) and complications of labour and delivery (Johnson and Slade, 2003).”

“For example, the EPDS is used as a measure of depression but evidence suggests a few items on the EPDS measure anxiety (Phillips et al., 2009; Jomeen and Martin, 2005).”

“The EPDS and GHQ-12 showed good concurrent validity (r = 0.80) (Navarro et al., 2007).”

“In a sample of women in early pregnancy the K-10 identified 69 - 76% of diagnosed cases of anxiety disorders (Spies et al., 2009).”

“…linked with intrauterine artery resistance leading to low birthweight (Teixeira et al., 2009; Field et al., 2010), low apgar scores (Berle et al., 2005), suboptimal changes in fetal heart rate and motor activity (DiPietro, 2010) and complications of labour and delivery (Johnson and Slade, 2003).”

“Sample:PC = primary care e.g. women at antenatal clinic; R = subsample of a research study; C = community Self-report instruments: CC-D&A = Trait anxiety & depression scales (Costello & Comrey, 1967); BAI = Beck Anxiety Inventory (Beck & Steer, 1991); BDI = Beck Depression Inventory (Beck, 1961); ENES = Eysenck Neurot…”

“Barnett and Parker (1986) reported further discriminant validity through inspection of hospital files where nursing staff considered a significantly higher percentage of women in the high anxiety group to have ‘mood problems’ than in the moderate or low anxiety groups.”

“There is also preliminary evidence that a further proportion of women experience clinically significant anxiety symptoms that do not fit into DSM-IV anxiety diagnoses (Phillips et al., 2009).”

“1 month postnatal Kitamura et al., 1994bd 16 Japan 108 PC Antenatal i.1st trimester ii.3rd trimester GHQ-30 SADS (RDC) Criterion: Sens, Spec, MR, PPV, NPV - 9 Levin, 1991 17 USA 266 PC Postnatal measurement of antenatal anxiety i.After birth whilst mothers still in hospital Pregnancy Anxiety Scale (PAS) - Factorial -…”

“Research into perinatal anxiety uses a variety of self-report measures including the State Trait Anxiety Inventory (STAI; Spielberger et al., 1970), Crown-Crisp Experiential Index (Crisp et al., 1978) and Beck Anxiety Inventory (Beck et al., 1988).”

“Self-report measures created for use in populations in western countries are often validated for use in different languages and cultures (Prince, 2008).”

“In a small sample of 3 or 6 month postpartum women the SCL-90-R anxiety subscale scores distinguished women with anxiety disorders from women with no psychiatric diagnosis but did not distinguish anxiety cases from depression cases in post-hoc tests (Muzik et al., 2000).”

“Jomeen and Martin (2004) carried out principal components analysis (PCA) and confirmatory factor analysis (CFA).”

“There is also preliminary evidence that a further proportion of women experience clinically significant anxiety symptoms that do not fit into DSM-IV anxiety diagnoses (Phillips et al., 2009).”

“However a Japanese study in pregnancy and at different time points postpartum found no cut-off yielded satisfactory sensitivity (Kitamura et al., 1994a).”

“…2nd HADS - Factorial Concurrent Internal consistency 8 trimester EPDS Kaaya et al., 2002 11 Tanzania 903 R Antenatal 8-26 weeks gestation HSCL-25 SF-36 SCID (DSM-IV) Factorial Concurrent Internal consistency 10 Karimova & Martin, 2003 12 UK, Uzbekista n 100 PC Antenatal i.12 weeks gestation ii.”

“3% of whom had substantially lower trait scores at 3 months postpartum than at 2-3 days postpartum (Giakoumaki et al., 2009).”

“34 weeks gestation HADS - Factorial Test-retest Internal consistency 8 Kitamura et al., 1989d 13 Japan 108 PC Antenatal 12 weeks gestation GHQ-30 SADS (RDC) Criterion: Sens, Spec, MR, PPV, NPV Discriminant Function Analysis - 8 Kitamura et al., 1993d 14 Japan 108 PC Antenatal 12 weeks…”

“Aderibigbe & Gureje, 1992b 4 Nigeria 277 PC Antenatal Late 2nd –early 3rd trimester GHQ-28 PAS (DSM-III-R) Criterion: Sens, Spec, MR, PPV, NPV Concurrent Compared 2 different scoring methods Discriminant Function Analysis - 8 Barnett & Parker, 1985c 5 Australia 147 PC Postnatal i. 3-4 days ii.”

“One study found that the C-GHQ method of scoring resulted in higher sensitivity and specificity than the standard GHQ scoring in Nigerian women in the late second or early third trimester (Aderibigbe and Gureje, 1992).”

“For example the STAI has been used in perinatal samples with cut-offs of >40 (Grant et al., 2008) to >45 (Moss et al., 2009) or >48 (Field et al., 2010) to indicate disorder, without prior validation in this population.”

“Sample:PC = primary care e.g. women at antenatal clinic; R = subsample of a research study; C = community Self-report instruments: CC-D&A = Trait anxiety & depression scales (Costello & Comrey, 1967); BAI = Beck Anxiety Inventory (Beck & Steer, 1991); BDI = Beck Depression Inventory (Beck, 1961); ENES = Eysenck Neurot…”

“Anxiety disorders and subthreshold anxiety symptoms can be detrimental to the mother-infant relationship (Zelkowitz and Papageorgiou, 2005).”

“Hundley et al (1998) found test-retest correlations of between 0.37 - 0.85 on the trait scale.”

“The shorter versions of the GHQ do not result in greatly reduced sensitivity and specificity in non-perinatal populations (Ayers, 2001).”

“The K10 may be the best short screening measure due to its ability, over the GHQ-12, to differentiate anxiety disorders.”

“…9 Sagrestano et al., 2001 25 USA 166 PC Antenatal >20 weeks gestation PSAI (single items) STAI - Concurrent Discriminant Internal consistency 9 Sharp, 1988 26 UK 179 PC Antenatal 20 weeks gestation GHQ-30 CIS Criterion: Sens, spec, MR, PPV - 9 Spies et al., 2001 27 South Africa 129 PC…”

“One study found that the C-GHQ method of scoring resulted in higher sensitivity and specificity than the standard GHQ scoring in Nigerian women in the late second or early third trimester (Aderibigbe and Gureje, 1992).”

“…was constructed through an internet survey of 142 questions derived from case descriptions of postnatal psychiatric disturbance and DSM-IV-R, with participants asked to retrospectively rate the frequency with which they experienced symptoms in the first 30 days after birth (Marrs et al., 2009).”

“Factor analysis in a non-perinatal population suggests the scale may be measuring one latent factor rather than two distinct anxiety and depression subscales (Shevlin et al., 1998).”

“However a Japanese study in pregnancy and at different time points postpartum found no cut-off yielded satisfactory sensitivity (Kitamura et al., 1994a).”

“34 weeks gestation HADS - Factorial Test-retest Internal consistency 8 Kitamura et al., 1989d 13 Japan 108 PC Antenatal 12 weeks gestation GHQ-30 SADS (RDC) Criterion: Sens, Spec, MR, PPV, NPV Discriminant Function Analysis - 8 Kitamura et al., 1993d 14 Japan 108 PC Antenatal 12 weeks gestation GHQ-30 - Compared 4 dif…”

“66 Self-report measures do have limitations, such as potentially inflated prevalence estimates, but they also have high clinical utility in obstetric/midwifery, public health and primary care practices, where the majority of perinatal mental health problems are managed.”

“Although the prevalence of maternal anxiety during pregnancy is heterogeneous, anxiety symptoms and disorders are common in the perinatal period (Meades and Ayers, 2011; Ross and McLean, 2006; Giardinelli et al., 2012).”

“The Spielberger STAI has been validated in for use in perinatal populations.(22) Population normative data shows that the median score for both state and trait anxiety among women ages 19–39 is 34.”