Effect of prenatal mental health problems on early speech sound acquisition. Title. Wong, Kei-yan, Gillian; 黃己恩

Transcription

1 Title Effect of prenatal mental health problems on early speech sound acquisition Author(s) Wong, Kei-yan, Gillian; 黃己恩 Citation Wong, K. G. [ 黃己恩 ]. (2012). Effect of prenatal mental health problems on early speech sound acquisition. (Thesis). University of Hong Kong, Pokfulam, Hong Kong SAR. Issued Date 2012 URL Rights This work is licensed under a Creative Commons Attribution- NonCommercial-NoDerivatives 4.0 International License.; The author retains all proprietary rights, (such as patent rights) and the right to use in future works.

2 Effect of prenatal mental health 1 Effect of prenatal mental health problems on early speech sound acquisition Wong Kei Yan Gillian A dissertation submitted in partial fulfillment of the requirements for the Bachelor of Science (Speech and Hearing Sciences), The University of Hong Kong, June 30, 2012.

3 Effect of prenatal mental health 2 Abstract The study investigated prospectively the effect of prenatal mental health problems in terms of anxiety and depression on speech sound acquisition in 26 mother-child dyads. Prenatal anxiety and depression were assessed using validated screening tools during late pregnancy. Speech sound production ability was assessed in terms of the number of atypical phonological patterns produced in a standardized speech assessment when the children reached the age of 2. Multiple regression analyses illustrated that prenatal depression uniquely explained 30.8% of the variance in speech sound acquisition after controlling for the child s sex and postnatal maternal depression level. However, the regression model for the anxiety measures was not significant. In conclusion, clinical level of prenatal depression was significantly associated with more atypical phonological patterns. The current findings contribute to the understanding of the etiology of functional speech sound disorders (SSD). At the clinical level, prenatal depression may be taken as a risk factor for SSD and the importance of enhancing mental health in pregnant women with depression is empathized. Early phonological assessment and intervention of children exposed to depression during late pregnancy are also recommended.

4 Effect of prenatal mental health 3 Speech sound disorders (SSD) are characterized by having difficulties producing speech sounds of a language beyond the expected developmental age (Bankson, Bernthal, & Flipsen, 2009a). SSD are often diagnosed in early childhood from age 3 to 6 and can be broadly categorized into organically-based and functional SSD. While organically-based SSD can be readily related to obvious clinical features such as cleft palate and Down syndrome, functional SSD cannot be attributed to any identifiable etiology and therefore has been described as having an unknown origin (Flipsen, Bankson, & Bernthal, 2009). Given the unexplained deficits of functional SSD, this group of clients remains to be a challenge to researchers regarding its contributing factors and maintaining causes (Shriberg, Tomblin, & McSweeny, 1999), as well as to clinicians regarding its early identification and differentiation from typical development. Prevalence studies of functional SSD in young children have been burgeoning in recent years. In an epidemiological study in the US, the prevalence of functional SSD based on 1,328 monolingual English-speaking children at 6 years of age was 3.8% (Shriberg et al., 1999). Using the same diagnostic criteria, the prevalence for the younger counterparts at age 3 was 15.6% (Campbell et al., 2003). These prevalence estimates suggest SSD are so highly prevalent in childhood that they occupy a significant proportion of the caseload among speech therapists (American Speech-Language-Hearing Association, 2010). Risk factors for functional SSD have been extensively studied using group comparison and correlational studies. Odds ratios (ORs) were employed to identify the risk factors for SSD. An OR is a relative estimate of risk which provides information on the extent to which an individual will develop a condition upon exposure to a condition under investigation when compared with the unexposed one. To represent a clinically significant increase in risk, an OR of 2.0 or above was used as the general cut-off (Streiner, 1998). Campbell et al. (2003)

5 Effect of prenatal mental health 4 investigated the risk factors for functional SSD in 3-year-old children and reported that children born to mothers with lower level of education had a higher risk of having SSD (OR = 2.58). Fox, Dodd, and Howard (2002) found in a retrospective cross-sectional study that environmental factors such as birth complications (OR= 9.35) and sucking habits (OR= 2.49) were significant risk factors for functional SSD. To, Cheung & McLeod (2010) investigated demographic risk factors such as parental education, household income, number of siblings and primary caretakers for inferior speech development in Hong Kong preschoolers and found that these extrinsic factors explained only less than 6% of the variance in speech sound ability, leaving a large percentage of unexplained variance. It may suggest that SSD might be associated more with intrinsic factors rather than merely induced by the environment. Strong aggregation of SSD within families suggests that SSD may be heritable. Fox et al. (2002) identified significant odds ratios (OR= 4.38) for positive family history of developmental communication disorders. Consistent findings were found in Campbell et al. (2003) and Lewis et al. (2007) who reported elevation of odds by almost two times given a positive history of developmental SSD and/or language disorders (LD) of first-degree relatives. In a systematic review, Lewis et al. (2006) concluded that concordance rates of SSD were significantly higher for monozygotic (MZ) than dizygotic (DZ) twins. Given that MZ twins are genetically the same and DZ twins share merely around 50% of the segregating genes, a higher concordance rate of MZ over DZ implies a genetic link to the disorder. In addition, male sex is associated with a greater risk for SSD. For instance, in a demographically representative population of 6-year-old children, Shriberg et al. (1999) reported that functional SSD was around 1.5 times more prevalent among boys than girls. Campbell et al. (2003) also noted significant OR (OR= 2.19) for male sex, suggesting that boys are nearly twice as likely as their female peers to have SSD.

6 Effect of prenatal mental health 5 Group comparison, familial aggregation and twins studies are now advanced to molecular genetic studies which attempt to locate the exact chromosomal genes that contribute to the development of SSD. Genome-wide linkage analysis and fine mapping of chromosome loci identified disruptions of the FOXP2 gene as the causation of SSD in some affected individuals (Newbury & Monaco, 2010). However, such monogenic mutations could not be the sole attribution to SSD in the general population due to their scarcity (Lewis et al., 2006). In a review examining the comorbidity rates among SSD, LD and dyslexia, Pennington and Bishop (2009) pointed out that children with SSD were 2.3 to 6.1 and 2.6 times more likely to have a comorbid diagnosis of LD and dyslexia respectively than the non-ssd controls. Such high rates of comorbidity may reflect overlapping genetic bases for these disorders. It is generally accepted that SSD are attributed to complex interactions of polygenic and environmental risk and protective factors (Lewis et al., 2006; Shriberg, 2010). However, the contributing factors that trigger the phenotype of SSD are still unknown (Shriberg, 2010). In a recent review of risk factors for SSD in preschool children, Harrison and McLeod (2010) located only one study (Fox et al., 2002) that explored prenatal factors as a potential predictor of SSD. Fox et al. (2002) studied prenatal problems retrospectively as a risk factor for SSD in 65 preschoolers with functional SSD and 48 normally speaking peers. Their findings showed that prenatal difficulties reported by mothers in terms of extreme prenatal stress, maternal infections and/or fetal-damaging drugs during pregnancy were a significant risk factor for functional SSD. With preliminary significant findings by Fox et al. (2002), it is reasonable to speculate that prenatal problems such as prenatal mental health problems may contribute to SSD in children, thus meriting a more precise and well-controlled investigation of the unique role of prenatal problems in speech sound development. Also, with scarcity of research in the

7 Effect of prenatal mental health 6 area, a pioneer in studying prenatal contributions could provide new insights into the variables that link to SSD. Prenatal mental health problems and postnatal outcomes Pregnancy specifically places a woman at a higher risk for mental health problems (Blegen, Hummelvoll, & Severinsson, 2010). Much recent evidence has attested that prenatal mental health problems such as severe anxiety and stressful life events are linked to negative outcomes of neonates, including premature birth and low birth weight, to name but a few (Brown, Yelland, Sutherland, Baghurst, & Robinson, 2011; Khashan et al., 2008; Zhu, Tao, Hao, Sun, & Jiang, 2010). These negative consequences could be long lasting, with a growing body of research suggests that prenatal mental health problems are significant risk factors for various well-defined developmental disorders such as autistic spectrum disorder (Beversdorf et al., 2005; Kinney, Miller, Crowley, Huang, & Gerber, 2008; Ward, 1990) and attention deficit hyperactivity disorder (Motlagh et al., 2010; O Connor, Heron, Golding, Beveridge, & Glover, 2002; Rodriguez & Bohlin, 2005; Ronald, Pennell, & Whitehouse, 2011; Van Den Bergh & Marcoen, 2004). Preliminary evidence that delineates the negative effects of prenatal mental health problems on offspring s language development is also emerging. In a longitudinal study, Laplante et al. (2004) investigated the effect of prenatal mental health problems as induced by a severe ice storm on 58 2-year-old children s language development using parent-reported questionnaires. Their study revealed that prenatal mental health problems uniquely accounted for 12.1% and 17.3% of the variance in the children s expressive and receptive language abilities respectively after controlling for confounding factors such as maternal postpartum depression. In a subsequent follow-up study, Laplante, Brunet, Schmitz, Ciampi, and King (2008) demonstrated persistent negative effects of prenatal mental health problems on the 5-year-old children s language functioning using

8 Effect of prenatal mental health 7 validated receptive language measures. Aim of the Present Study Taken together preliminary evidence and previous research findings, the objective of the study was to examine in a prospective longitudinal study the unique contribution of prenatal mental health problems to offspring s speech sound ability. The present study focused on children at 2 years of age when phoneme acquisition is very rapid (McLeod, 2009). In a recent literature review of the association between prenatal mental health problems and outcomes of the offspring, Huizink, Mulder, and Buitelaar (2004) hypothesized that different sensitive periods of prenatal contributions may exist for different postnatal outcomes. With considerable evidence suggesting that perisylvian region of the dominant hemisphere, the cortical structures that are actively involved in speech and language development develops and matures in late stages of gestation (Cohen-Sacher, Lerman-Sagie, Lev, & Malinger, 2006), the present study focused on exposure to mental health problems during late pregnancy, and more specially, the third trimester of pregnancy. Methods Participants This study was a prospective longitudinal design of 2-year-old children born to women recruited during their early pregnancy in a project investigating prenatal mental health problems and their impacts on the mothers (Lee et al., 2007). The women were then screened for anxiety and depression symptoms at 3 time points, namely the first trimester (around 12 weeks), the second trimester (around 20 weeks) and the third trimester (around 36 weeks) of pregnancy. Prenatal inclusion criteria included pregnant Chinese ethnicity women aged over 18 years and prenatal exclusion criteria included women who had significant medical illnesses, were conceived through in-vitro fertilization, and were considering termination of

9 Effect of prenatal mental health 8 pregnancy. The current study recruited 31 women and their offspring. Of the 31 children, 5 children were not able to complete 70% of the assessment in the study. This resulted in a total of 26 mother-child dyads. These 26 children included 7 boys and 19 girls aged from 24 to 28 months (M = 25.2 months, SD = 1.0). All the children were born full term as reported by the mothers. Measures Prenatal anxiety. The validated Chinese adaptation of the anxiety subscale of the Hospital Anxiety and Depression Scale (HADS-A; Leung, Ho, Kan, Hung, & Chen, 1993; Zigmond & Snaith, 1983) was used to identify possible and probable cases of anxiety disorder. The HADS-A is a self-administered questionnaire composed of seven 4-point Likert scales. With increasing severity from 0 to 3 scores of each item, a maximum score of 21 could be obtained (Couto et al., 2009). The recommended optimal cut-off score is 8 or higher (Bjelland, Dahl, Haug, & Neckelmann, 2002). Based on this cut-off point, 15 women were at or above the clinical cut-off of prenatal anxiety. Prenatal depression. The Edinburgh Depression Scale (EDS; Murray & Cox, 1990) was used for screening clinically significant depression. The EDS is a self-administered screening instrument by rating one s feeling over the past week. It is composed of ten 4-point Likert scales and it yields an overall score ranging from 0 to 30 with increasing severity (Murray & Cox, 1990). The Chinese version of the scale has already been validated among Chinese women in the territory with satisfactory psychometric properties (Lee et al., 1998). A cut-off score of 11 or higher was chosen for evaluation of prenatal depression. The cut-off score was selected for its high sensitivity and specificity (88% and 92% respectively) in detecting major depression disorder in pregnant women (Bunevicius et al., 2009). Based on this cut-off point, six women were at or above the clinical cut-off of prenatal depression.

10 Effect of prenatal mental health 9 Child s speech sound ability. The Hong Kong Cantonese Articulation Test (HKCAT; Cheung, Ng, & To, 2006), a standardized assessment, was employed to assess the children s speech sound ability. The children s speech sound productions were compared with reference to adult s productions using relational analysis (Bankson, Bernthal, & Flipsen, 2009b). The number of atypical phonological patterns used by the children, which are referred to as patterns realized by less than 5% of children in the population (To, Cheung, & McLeod, in press), was computed. Atypical phonological patterns are considered as qualitative measures of phonology (McIntosh & Dodd, 2008). In a recent longitudinal study of phonological acquisition, McIntosh and Dodd (2008) pointed out that qualitative but not quantitative measures of phonology (e.g. percentages of consonants correct) of 2-year-old children were predictive of a clinical diagnosis of SSD when the children reached the age of 3. Potential covariates. The mental health profiles when the children were at 2 years of age and the sex of the child were chosen as potential covariates. The child s sex was included because a significant relationship between male sex and an elevated risk for SSD has been well-documented in a plethora of studies (e.g., Campbell et al., 2003; Harrison & McLeod, 2010; Lewis et al., 2006). The effect of postnatal maternal mental health on children s speech sound acquisition is less explored and is inconsistently found to be associated with speech sound development in children. In a systematic literature review of risk factors and protective factors for SSD and LD, Harrison and McLeod (2010) identified 5 out of 22 studies that explored the association between postnatal maternal mental health and the two communication disorders with mixed findings. Taken into considerations the potential contribution of postnatal maternal mental health to speech sound acquisition, the maternal mental health at the time of testing (i.e., around two years postpartum) was included as the second predictor variable. Postnatal anxiety was screened by HADS-A adopting the

11 Effect of prenatal mental health 10 recommended cut-off score of 8 or higher as discussed earlier on for prenatal anxiety. Postnatal depression was screened by EDS adopting a cut-off score of 13 or higher since sensitivity and specificity were reported to be higher using the cut-off score (Cox, Chapman, Murray, & Jones, 1996). Procedures Women in the database of the previous project were contacted by telephone to invite to take part in the present study when their children reached around 2 years of age. For those women who agreed to participate in the study, they were asked to have the child assessed either at their home or in a quiet room in the Division of Speech and Hearing Sciences at The University of Hong Kong. The assessors who administered the test were blinded to the mental health profiles of the mothers in order to control for potential bias that might affect the validity of assessment results. Prior to the child assessment, parental consent was obtained and the mother filled in the HADS-A and EDS questionnaires. Upon establishing rapport, each child completed the HKCAT (Cheung et al., 2006). If the child could not name the item spontaneously, a model was given to elicit an imitated response. Each child took approximately minutes to complete the test depending on his/her responsiveness and cooperativeness. Speech samples obtained were recorded on a Sony digital voice recorder ICD-UX81F for later transcriptions and subsequent reliability checking. The transcribers who transcribed the samples and did the reliability check were also blinded to the mental statuses of the mothers. Reliability To determine inter-rater transcription agreement, assessments of three randomly selected children (12% of the samples) were transcribed independently by one trained transcriber based on the recordings. For intra-rater transcription agreement, the investigator

12 Effect of prenatal mental health 11 re-transcribed another three children s assessments about one month after the initial transcriptions. The reliability was calculated by dividing the number of agreements of phonemes by the total number of phonemes produced, followed by multiplication by 100%. The inter-rater and intra-rater point-to-point agreement across transcriptions were 89% and 92% respectively, which were considered to be satisfactory. The original transcriptions of the investigator were used for analysis. Results Descriptive Analyses The mean HADS-A scores of the anxious and non-anxious groups were 14.7 and 3.6 respectively while the mean EDS scores of the depressive and non-depressive groups were 11.2 and 6.9 respectively. The means of the number of atypical phonological patterns produced by children of the four groups (i.e. with and without exposure to prenatal anxiety and depression) in both sexes are summarized in Table 1. There were more children born to mothers who scored at or above the cutoff score for prenatal anxiety than below the cut-off. On the contrary, there were more children born to mothers who scored below the cut-off score for prenatal depression than at or above the cut-off. There were more girls than boys in the sample irrespective of prenatal mental health profiles. On average, boys produced more atypical phonological patterns than girls when they were grouped according to the cutoff scores of prenatal measures. There were 15 and 6 children whose mothers scored at clinical level of prenatal anxiety and depression respectively.

13 Effect of prenatal mental health 12 Table 1 Atypical Phonological Patterns (Mean and (SD)) Produced by Children With and Without Prenatal Exposure to Mental Health problems Without Anxiety With Anxiety Boys Girls Total Boys Girls Total (n = 3) (n = 8) (n = 11) (n = 4) (n = 11) (n = 15) Atypical patterns 11.3 (9.5) 7.5 (6.7) 8.5 (7.3) 20.5 (12.1) 5.5 (5.6) 9.5 (10.1) Without Depression With Depression Boys Girls Total Boys Girls Total (n = 5) (n = 15) (n = 20) (n = 2) (n = 4) (n = 6) Atypical patterns 11.6 (8.3) 4.7 (5.6) 6.4 (6.8) 29.0 (7.1) 12.5 (3.0) 18.0 (9.4) Hierarchal Multiple Regression Analyses The data could fit in a regression equation as the predictor variables (i.e., mental health statuses and sex) were dichotomous and the dependent variable (i.e., the number of atypical phonological patterns) was continuous. Hierarchical regressions were used so as to control for potential covariates that might have associations with the dependent variable but were not the focus of the study. The variable of sex was coded as 1 for female and 0 for male and the variables of maternal mental health statuses were coded as 1 for at clinical level and 0 for below clinical level of mental health problems. Stepwise analyses were conducted. In the first step, sex was entered into the equation due to ample studies that consistently reported male sex as a significant risk factor for SSD (Harrison & McLeod, 2010). Postnatal maternal mental health status measured at the time of testing was entered next and followed by prenatal mental health status in the last step.

14 Effect of prenatal mental health 13 The results of the multiple regression analysis for anxiety are shown in Table 2. Tests for multicollinearity indicated a low level of multicollinearity (tolerance =.864,.864 and for sex, postnatal anxiety and prenatal anxiety respectively). In the first step, sex uniquely accounted for 27.6% of the variance in atypical phonological patterns and with a negative beta value of That means, male sex (with the coding of 0 ) was associated with more atypical phonological patterns. In the second step, postnatal anxiety did not significantly increase the proportion of variance explained. Adding the variable of prenatal anxiety in the last step also did not significantly increase the proportion of variance accounted for. Table 2 Hierarchical Multiple Regression Model Investigating Whether Prenatal Anxiety Explained Variance in Speech Sound Acquisition Above and Beyond Potential Covariates Predictor variables ß R 2 ΔR 2 F ΔF Step1# ** Sex -.525** Step *.288 Sex Postnatal anxiety -.563**.102 Step Sex Postnatal anxiety Prenatal anxiety -.563** Note. # df(1, 24) df(2, 23) df(3, 22); **p <.01 *p <.05

15 Effect of prenatal mental health 14 The results of the multiple regression analysis for depression are shown in Table 3. Tests for multicollinearity also indicated a low level of multicollinearity (tolerance =.893,.864 and.941 for sex, postnatal depression and prenatal depression respectively). Again, sex accounted for 27.6% of the variance in atypical phonological patterns in the first step. In the second step, postnatal depression was added and did not significantly increase the proportion of variance explained. When prenatal depression was added in the last step, the amount of variance accounted for elevated to 58.8%. Controlling for sex and postnatal depression, prenatal depression uniquely accounted for 30.8% of the variance: clinical level of prenatal depression was significantly associated with more atypical phonological patterns. Table 3 Hierarchical Multiple Regression Model Investigating Whether Prenatal Depression Explained Variance in Speech Sound Acquisition Above and Beyond Potential Covariates Predictor variables ß R 2 ΔR 2 F ΔF Step1# ** Sex -.525** Step *.138 Sex Postnatal depression -.505* Step **** *** Sex Postnatal depression Prenatal depression -.419** *** Note. # df(1, 24) df(2, 23) df(3, 22); ****p <.001 ***p <.005 **p <.01 *p <.05

16 Effect of prenatal mental health 15 Discussion The study was the first attempt to explore systematically the unique influence of prenatal mental health problems during late pregnancy on offspring s speech sound development after controlling for the child s sex and postnatal maternal mental health problems. As clearly demonstrated in the results, sex explained 27.6% of the variance in speech sound acquisition of the children. This parallels with a considerable amount of literature that suggests a negative association between male sex and speech sound development in young children. More important in the study, prenatal depression during the third trimester of gestation uniquely accounted for 30.8% of the variance in the children s speech sound acquisition in terms of the number of atypical phonological patterns produced, after controlling for the child s sex and postnatal depression. That means when compared with children unexposed to prenatal depression, children who were born to mothers with clinical level of prenatal depression during late pregnancy produced significantly more atypical phonological patterns. The number of atypical phonological patterns is a better predictor that differentiates between children with SSD and those who can resolve the speech errors by themselves than other measures such as the percentage of correct phonemes produced (McIntosh & Dodd, 2008). The findings reported in the present study supported the observation in Fox et al. (2002) who reported that prenatal mental health problems were associated with children s SSD. They evaluated prenatal mental health problems using a yes/no question in a retrospective manner. Conversely, the prospective longitudinal design of the study made use of timely assessments of well-defined mental health problems of woman participants during pregnancy. Moreover, the outcome measures of the study were assessed objectively by the investigator rather than through parental report measures which are often subject to informational bias due to overreporting by the mothers (Charman, Drew, Baird, & Baird, 2003). In particular, mothers

17 Effect of prenatal mental health 16 who are depressed might be more judgmental or less sensitive to their child s development (Keim et al., 2011). Given the aim of this study was to compare children s ability in depressive and non-depressive mothers, objective assessments of the children s speech sound productions using direct observations could better ascertain the validity of the findings. In addition, the conclusion of the study was drawn independent of postnatal maternal depression. As maternal depression are reported to be negatively associated with the quality of mother-child interactions (Korja et al., 2008), which may in turn pose a pervasive adverse influence on a child s speech and language development (Westerlund & Lagerberg, 2008), controlling for postnatal mental health problems in the study could strengthen the prenatal contributions to offspring s phonological acquisition. All in all, the current study added a stronger piece of evidence to the claim about the detrimental effect of prenatal mental health problems on children s speech development. The current study also contributes to our understanding of the etiology of functional SSD which is regarded as having an unknown origin. Functional SSD is an etiologically complex disorder and is contributed by multiple genes that interact with each other and with the environment (Lewis, 2006; Lewis, 2010). Despite certain degree of genetic influences, no single gene has been identified as the casual factor of functional SSD in the general population (Smith, 2011). With this unexplained causation yet potential area for research, this study sheds light on the possibility that epigenetic modifications might contribute to the development of functional SSD. According to Lewis (2012), epigenetics refer to non-genetic factors such as maternal environmental alternations that mediate the expression of genes. Through this epigenetic reprogramming during sensitive period of development, the developmental trajectory of an offspring might be altered (Bloomfield, 2011), thereby affecting the postnatal outcomes. It is hence hypothesized in the study that epigenetic

18 Effect of prenatal mental health 17 modifications may play a role in controlling the association between prenatal exposure to depression during late pregnancy and speech sound production in the children. Despite undetermined physiologic processes of prenatal mental health problems in humans (Talge et al., 2007), the potential role of increased activity of maternal hypothalamic-pituitary-adrenal (HPA) axis in mediating the epigenetic effects has been widely discussed and reviewed (Huizink et al., 2004). The adrenal cortex of the HPA axis produces and releases cortisol into the bloodstream as the final product and passes to the fetus through the placenta (Talge et al., 2007). Although normal range of cortisol levels are essential to neuronal maturations, chronically high concentration of maternal cortisol induced by mental health problems may pose detrimental influences to the offspring s brain development (Huizink et al., 2004). Under normal circumstances, the fetal brain is protected from maternal cortisol by 11-β-hydroxysteroid dehydrogenase (11-β-HSD) which oxidizes cortisol into its biologically non-toxic form (Field, 2011). However, 11-β-HSD is substantially reduced towards the end of pregnancy, thus allowing greater transduction of maternal cortisol from the pregnant mother to the fetus (Huizink, Robles de Medina, Mulder, Visser, & Buitelaar, 2003). This, together with the rapid development of the perisylvian region of the brain which is responsible for speech and language development and becomes detectable by sonographic images during late pregnancy (Cohen-Sacher et al., 2006), might account for the deleterious effects of prenatal depression during the third trimester of pregnancy in the study. The effect of prenatal depression was consistent with a considerable amount of research which outlined the adverse effects of depression during pregnancy on various aspects of offspring s outcomes (Field et al., 2004; Hernandez-Reif, Field, Diego, & Ruddock, 2006; Tikotzky, Chambers, Gaylor, & Manber, 2010). As suggested by Talge et al. (2007), although evidence showing the association between prenatal mental health problems and offspring s

19 Effect of prenatal mental health 18 outcomes is promising, the timing of pregnancy that are the most vulnerable to prenatal contributions varies depending on the outcome measures. With relatively little studies identifying the timing of prenatal contributions (O Donnell, O Connor, & Glover, 2009), the findings of the study shed light on the phonological aspect of postnatal development as sensitive to prenatal mental health problems during the third trimester of pregnancy, which is comparable to some of the studies reported for symptoms of ASD (Kinney et al., 2008), ADHD (de Bruijn, van Bakel, & van Baar, 2009; O Connor et al., 2002) and mixed handedness, an indicator of disturbed brain development (Obel, Hedegaard, Henriksen, Secher, & Olsen, 2003). On the other hand, contributions of prenatal anxiety were not evident in the current study despite expectations of similar effects on the offspring (Field, 2011). The separate effects of prenatal depression and anxiety are difficult to interpret as currently no single factor of prenatal mental health problems has been identified conclusively in literature for its contribution to offspring s development (Field et al., 2010), not to mention the differential effects of selected prenatal mental health problems. Nevertheless, it is hypothesized that different operating mechanisms of prenatal anxiety and depression might account for the discrepancies in results (Field et al., 2003). Clinical implications The study demonstrated that prenatal mental health problems in terms of depression in the third trimester of gestation uniquely accounted for 30.8% of the variance in atypical speech sound productions of 2-year-old children, above and beyond the child s sex and postnatal maternal depression. These findings highlight the significance of enhancing psychological well-being of women with clinical levels of depression especially in late pregnancy. In their randomized-control studies, Field, Diego, Hernandez-Reif, Deeds, and Figueiredo (2009) showed that non-pharmacological interventions such as massage therapy

20 Effect of prenatal mental health 19 by the pregnant women s trained significant others could reduce their depression symptoms. Not only could interventions as such benefit the women s mental health, it could also minimize long term consequences of prenatal mental health problem on their children s phonological development in a cost-efficient manner. In addition, prenatal mental health problems can be included in protocols of developmental surveillance programs for early assessment of children at risk of speech and language problems. This is to initiate intervention early when improvement is more likely to occur with a developing phonological system (Broomfield & Dodd, 2005), therefore reducing the persistence and severity of speech production difficulties in young children (Harrison & McLeod, 2010). This would in turn lead to fewer frustrations upon entering school due to unintelligible speech (McIntosh & Dodd, 2008). Limitations of the study Certain limitations exist in the present study. First, the conclusion was drawn solely based on a limited number of mother-child dyads. Therefore, the small number of dyads involved may not be representative of the target population. This small sample size also limited the number of covariates that can be included in the study. This is attributed to the notion that the more predictors included into the model, the larger the sample size is required to provide sufficient statistical power. Future investigations are hence recommended to include a larger group of cohorts for inclusion of more potential confounding variables to enhance the generalization of findings. Moreover, both the HADS (Zigmond & Snaith, 1983) and EDS (Murray & Cox, 1990) could also be applied to non-childbearing population (Bjelland et al., 2002). As beginning evidence suggested that measures of mental health problems specific to pregnancy could predict their contributions to offspring s development better (Buss, Davis, Hobel, & Sandman, 2011), future studies may employ mental health

21 Effect of prenatal mental health 20 measures that capture concerns specific to being pregnant (e.g. I am worried about the health of my baby ) for optimal investigation of their effects on children s outcomes. Thirdly, the study could not directly identify the underlying biomechanism by which prenatal mental health problems contribute to offspring s development. Future studies could try to document women s biochemistry profiles such as urinary cortisol or dopamine level across stages of pregnancy for their contributions. Finally, the study only focused on 2-year-old children exposed to different degrees of prenatal mental health problems. Follow-up phonological assessments of the cohorts of children are therefore necessary to determine whether the influences of prenatal mental health problems are intensified, maintained or lessen as the children develop.

22 Effect of prenatal mental health 21 Acknowledgments I would like to express my gratitude to my supervisor, Dr. Carol To, for her inspirations throughout the study. I am also grateful to the colleagues from Department of Psychiatry of The University of Hong Kong for their assistance with participant recruitment. More important, I would like to show my appreciation to the mothers and children who joined the study. My sincere thanks should also be devoted to my family members and my classmates, notably Ms. Cherry Lee, Ms. Eunice Siu and Ms. Mona Tong for their support during completion of the study. References American Speech-Language-Hearing Association. (2010). Schools survey report: SLP caseload characteristics trends Rockville, MD: Author. Bankson, N. W., Bernthal, J. E., & Flipsen P. (2009a). Introduction to the study of speech sound disorders. In J. E. Bernthal, N. W. Bankson, & P. Flipsen (Eds.), Articulation and phonological disorders: Speech sound disorders in children (6 th ed.). (pp. 1-4). Boston, MA: Allyn & Bacon. Bankson, N. W., Bernthal, J. E., & Flipsen P. (2009b). Phonological assessment procedures. In J. E. Bernthal, N. W. Bankson, & P. Flipsen (Eds.), Articulation and phonological disorders: Speech sound disorders in children (6 th ed.). (pp ). Boston, MA: Allyn & Bacon. Beversdorf, D. Q., Manning, S. E., Hillier, A., Anderson, S. L., Nordgren, R. E., Walters, S. E.,... Bauman, M. L. (2005). Timing of prenatal stressors and autism. Journal of Autism and Developmental Disorders, 35(4), Bjelland, I., Dahl, A. A., Haug, T. T., & Neckelmann, D. (2002). The validity of the Hospital Anxiety and Depression Scale. An updated literature review. Journal of Psychosomatic

23 Effect of prenatal mental health 22 Research, 52(2), Blegen, N. E., Hummelvoll, J. K., & Severinsson, E. (2010). Mothers with mental health problems: A review. Nursing and Health Sciences, 12(4), Bloomfield, F. H., (2011). Epigenetic modifications may play a role in the developmental consequences of early life events. Journal of Neurodevelopmental Disorders, 3(4), Broomfield, J., & Dodd, B. (2004). Children with speech and language disability: caseload characteristics. International Journal of Language & Communication Disorders, 39(3), Brown, S. J., Yelland, J. S., Sutherland, G. A., Baghurst, P. A., Robinson, J. S. (2011). Stressful life events, social health issues and low birthweight in an Australian population-based birth cohort: Challenges and opportunities in antenatal care. BMC Public Health, 11, 196. Bunevicius, A., Kusminskas, L., Pop, V. J., Pedersen, C. A., Buevicius, R. (2009). Screening for antenatal depression with the Edinburg Depression Scale. Journal of Psychosomatic Obstetrics and Gynaecology, 30(4), Buss, C., Davis, E. P., Hobel, C. J., & Sandman, C. A. (2011). Maternal pregnancy-specific anxiety is associated with child executive function at 6-9 years age. Stress, 14(6), Campbell, T. F., Dollaghan, C. A., Rockette, H. W., Paradise, J. L., Feldman, H. M., Shriberg, L. D.,... Kurs-Lasky, M. (2003). Risk factors for speech delay of unknown origin in 3-year-old children. Child Development, 74(2), Charman, T., Drew, A., Baird, C., & Baird, G. (2003). Measuring early language development in preschool children with autism spectrum disorder using the MacArthur

24 Effect of prenatal mental health 23 Communicative Development Inventory (Infant Form). Journal of Child Language, 30(1), Cheung, P., Ng, A., & To, C. (2006). Hong Kong Cantonese Articulation Test. Hong Kong: City University of Hong Kong. Cohen-Sacher, B., Lerman-Sagie, T., Lev, D., & Malinger, G. (2006). Sonographic developmental milestones of the fetal cerebral cortex: A longitudinal study. Ultrasound Obstetrics & Gynecology, 27(5), Couto, E. R., Couto, E., Vian, B., Gregório Z, Nomura, M. L., Zaccaria, R., & Passini, R. (2009). Quality of life, depression and anxiety among pregnant women with previous adverse pregnancy outcomes. São Paulo Medical Journal, 127(4), Cox, J. L., Chapman, G., Murray, D., & Jones, P. (1996). Validation of the Edinburg Postnatal Depression Scale (EPDS) in non-postnatal women. Journal of Affection Disorder, 39(3), de Bruijn, A. T., van Bakel J. H., & van Baar, A. L. (2009). Sex differences in the relation between prenatal maternal emotional complaints and child outcome. Early Human Development, 85(5), Field, T. (2011). Prenatal depression effects on early development: A review. Infant Behavior & Development, 34(1), Field, T., Diego, M., Dieter, J., Hernandez-Reif, M., Schanberg, S., Kuhn, C.,... Bendell, D. (2004). Prenatal depression effects on the fetus and the newborn. Infant Behavior & Development, 27(2), Field, T., Diego, M., Hernandez-Rief, M., Deeds, O., & Figueiredo, B. (2009). Pregnancy massage reduce prematurity, low birthweight and postpartum depression. Infant Behavior & Development, 32(4),

25 Effect of prenatal mental health 24 Field, T., Diego, M., Hernandez-Reif, M., Figueiredo, B., Deeds, O., Ascencio, A.,... Kuhn, C. (2010). Comorbid depression and anxiety effects on pregnancy and neonatal outcome. Infant Behavior & Development, 33(1), Field, T., Diego, M., Hernandez-Reif, M., Schanberg, S., Kuhn, C., Yando, R., & Bendell, D. (2003). Pregnancy anxiety and comorbid depression and anger: Effects on the fetus and neonate. Depression and Anxiety, 17(3), Flipsen, P., Bankson, N. W., & Bernthal, J. E. (2009). Classifications and factors related to speech sound disorders. In J. E. Bernthal, N. W. Bankson, & P. Flipsen (Eds.), Articulation and phonological disorders: Speech sound disorders in children (6 th ed.). (pp ). Boston, MA: Allyn & Bacon.s Fox, A.V., Dodd, B., & Howard, D. (2002). Risk factors for speech disorders in children. International Journal of Language and Communication Disorder, 37(2), Harrison, L. J., & McLeod, S. (2010). Risk and protective factors associated with speech and language impairment in a nationally representative sample of 4- to 5- year-old children. Journal of Speech, Language, and Hearing Research, 53(2), Hernandez-Reif, M., Field, T., Diego, M., & Ruddock, M. (2006). Greater arousal and less attentiveness to face/ voice stimuli by neonates of depressed mothers on the Brazelton Neonatal Behavioral Assessment Scale. Infant Behavior & Development, 29(4), Huizink, A. C., Mulder, E. J. H., & Buitelaar, J. K. (2004). Prenatal stress and risk for psychopathology: Specific effects of induction of general susceptibility? Psychological Bulletin, 130(1), Huizink, A. C., Robles de Medina, P. G., Mulder, E. J., Visser, G. H., & Buitelaar, J. K. (2003). Stress during pregnancy is associated with developmental outcome in infancy.

26 Effect of prenatal mental health 25 Journal of Child Psychology and Psychiatry, and Allied Disciplines, 44(6), Keim, S. A., Daniels, J. L., Dole, N., Herring, A. H., Siega-Riz, A. M., & Scheidt, P. C. (2011). A prospective study of maternal anxiety, perceived stress, and depressive symptoms in relation to infant cognitive development, Early Human Development, 87(5), Khashan, A. S., McNamee, R., Abel, K. M., Pedersen, M. G., Webb, R. T., Kenny, L. C.,... Baker, P. N. (2008). Reduced infant birthweight consequent upon maternal exposure to severe life events. Psychosomatic Medicine, 70(6), Kinney, D., Miller, A. M., Crowley, D. J., Huang, E., & Gerber, E. (2008). Autism prevalence following prenatal exposure to hurricanes and tropical storms in Louisiana. Journal of Autism and Developmental Disorders, 38(3), Korja, R., Savonlahti, E, Ahlqvist-Björkroth, S., Stolt, S., Haataja, L., Lapinleimu, H.,... the PIPARI study group (2008). Maternal depression is associated with mother-infant interaction in preterm infants. Acta Paediatrica, 97(6), Laplante, D. P., Barr, R. G., Brunet, A., Galbaud du Fort, G., Meaney, M. L., Saucier, J.,... King, S. (2004). Stress during pregnancy affects general intellectual and language functioning in human toddlers. Pediatric Research, 56(3), Laplante, D. P., Brunet, A., Schmitz, N., Ciampi, & King, S. (2008). Project ice storm: Prenatal maternal stress affects cognitive and linguistic functioning in 5 1/2-year-old children. American Academy of Child and Adolescent Psychiatry, 47(9), Lee, A. M., Lam, S. K., Lau, S. M. S. M., Chong, C. S. Y., Chui, H. W., Fong. D. Y. T. (2007). Prevalence, course and risk factors antenatal anxiety and depression. Obstetrics & Gynecology, 110(5),

27 Effect of prenatal mental health 26 Lee, D. T., Yip, S. K., Chiu, H. F., Leung, T. Y., Chan, K. P., Chau, I. O,... Chung, T. K. (1998). Detecting postnatal depression in Chinese women. Validation of the Chinese version of the Edinburg Postnatal Depression Scale. British Journal of Psychiatry, 172, Leung, C. M., Ho, S., Kan, C. S., Hung, C. H., & Chen, C. N. (1993). Evaluation of the Chinese version of the Hospital Anxiety and Depression Scale. A cross-cultural perspective, Internal Journal of Psychosomatics, 40(1-4), Lewis, B. A. (2010). Genetic influences on speech sound disorders. In R. Paul & P. Flipsen (Eds.), Speech sound disorders in children: In honor of Lawrence D. Shriberg. (pp ). San Diego, CA: Plural Publishing. Lewis, B. A. (2012, March). The role of genetics in speech, language, and reading disorders [PDF document]. Paper presented at 2012 Ohio Speech-Language-Hearing Association Convention, Columbus, OH. Retrieved from Lewis, B. A., Freebairn, L. A., Hansen, A. J., Miscimara, L., Iyengar, S. K., & Taylor, H. G. (2007). Speech and language skills of parents of children with speech sound disorders. American Journal of Speech-Language Pathology, 16(2), Lewis, B. A., Shriberg, L. D., Freebairn, L.A., Hansen, A. J., Stein, C. M., Taylor, H. G., & Iyengar, S. K. (2006). The genetic bases of speech sound disorders: Evidence from spoken and written language. Journal of Speech, Language, and Hearing Research, 49(6), McIntosh, B., & Dodd, B. J. (2008). Two-year-olds phonological acquisition: Normative data. International Journal of Speech-Language Pathology, 10(6), McLeod, S. (2009). Speech sound acquisition. In J. E. Bernthal, N. W. Bankson, & P. Flipsen

28 Effect of prenatal mental health 27 (Eds.), Articulation and phonological disorders: Speech sound disorders in children (6 th ed.). (pp ). Boston, MA: Allyn & Bacon. Motlagh, M. G., Katsovich, L., Thompson, N., Lin, H., Kim, Y. S., Scahill, L.,... Leckman, L. F. (2010). Severe psychosocial stress and heavy cigarette smoking during pregnancy: An examination of the pre- and perinatal risk factors associated with ADHD and Tourette syndrome. European Child and Adolescent Psychiatry, 19(10), Murray, D., & Cox, J. L. (1990). Screening for depression during pregnancy with the Edinburgh Depression Scale (EPDS). Journal of Reproductive and Infant Psychology, 8(2), Newbury, D. F., & Monaco, A. P. (2010). Genetic advances in the study of speech and language disorders. Neuron, 68(2), Obel, C., Hedgaard, M., Hernriksen, T. B., Secher, N. J., Olsen, J. (2003). Psychological factors in pregnancy and mixed-handedness in the offspring. Developmental Medicine and Child Neurology. 45(3), O Connor, T. G., Heron, J., Golding, J., Beveridge, M., & Glover, V. (2002). Maternal antenatal anxiety and children s behavioural/ emotional problems at 4 years. Report from the Avon Longitudinal Study of Parents and Children. The British Journal of Psychiatry, 180, O Donnell, K., O Connor, T. G., & Glover, V. (2009). Prenatal stress and neurodevelopment of the child: Focus on the HPA axis and role of the placenta. Developmental Neuroscience, 31(4), Pennington, B. F., & Bishop, D. V. (2009). Relations among speech, language, and reading disorders. Annual Review of Psychology, 60, Rodriguez, A., & Bohlin, G. (2005). Are maternal smoking and stress during pregnancy

29 Effect of prenatal mental health 28 related to ADHD symptoms in children? Journal of Child Psychology and Psychiatry, 46(3), Ronald, A., Pennell, C. E., & Whitehouse, A. J. O. (2011). Prenatal maternal stress associated with ADHD and autistic traits in early childhood. Frontiers in Psychology, 223(1), 1-8. Shriberg, L. D. (2010). Childhood speech sound disorders: From postbehaviorism to the postgenomic era. In R. Paul & P. Flipsen (Eds.), Speech sound disorders in children: In honor of Lawrence D. Shriberg. (pp. 1-33). San Diego, CA: Plural Publishing. Shriberg, L. D., Tomblin, J. B., & McSweeny, J. L. (1999). Prevalence of speech delay in 6-year-old children and comorbidity with language impairment. Journal of Speech, Language, and Hearing Research, 42(6), Smith, S. D. (2011). Approach to epigenetic analysis in language disorders. Journal of Neurodevelopmental Disorders, 3(4), Streiner, D. L. (1998). Risky business: Making sense of estimates of risk. Canadian Journal of Psychiatry, 43(4), Talge, N. M., Neal, C., Glover, V., & the Early Stress, Translational Research and Prevention Science Network: Fetal and Neonatal Experience on Child and Adolescent Mental Health. (2007). Journal of Child Psychology and Psychiatry, and Allied Disciplines, 48(3-4): Tikotzky, L., Chambers, A. S., Gaylor, E., & Manber, R. (2010). Maternal sleep and depressive symptoms: Links with infant Negative Affectivity. Infant Behavior & Development, 33(4), To, C. K. S., Cheung, P. S. P., & McLeod, S. (2010). Demographic factors in relation to speech sound development in Cantonese-speaking children. Paper presented at the Hong Kong Speech & Hearing Symposium, Hong Kong.

30 Effect of prenatal mental health 29 To, C. K. S., Cheung, P.S.P., & McLeod, S. (in press). A population study of children s acquisition of Hong Kong Cantonese consonants, vowels, and tones. Journal of Speech, Language and Hearing Research. Van Den Bergh, B. R., & Marcoen, A. (2004). High antenatal maternal anxiety is related to ADHD symptoms, externalizing problems, and anxiety in 8-and 9-year-olds. Child Development, 75(4), Ward, A. J. (1990). A comparison and analysis of the presence of family problems during pregnancy of mothers of autistic children and mothers of normal children. Child Psychiatry and Human Development, 20(4), Westerlund, M., & Lagerberg, D. (2008). Expressive vocabulary in 18-month-old children in relation to demographic factors, mother and child characteristics. Child: Care, Health and Development, 34(2), Zhu, P., Tao, F., Hao, J., Sun, Y., & Jiang, X. (2010). Prenatal life events stress: Implications for preterm birth and infant birthweight. American Journal of Obstetrics and Gynecology, 203(1), 34.e1-34.e8. Zigmond, A. S., & Snaith, R. P. (1983). The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica, 67(6),