The mother-infant interactions and the development of self-regulation : a comparative study of term and preterm infants

i The mother-infant interactions and the development of self-regulation: Ana Rita Ribeiro Pires A comparative study of term and preterm infants. Dissertation presented as partial requirement for obtaining the Master’s degree in Statistics and Information Management ii MEGI MGI i NOVA Information Management School Instituto Superior de Estatística e Gestão de Informação Universidade Nova de Lisboa THE MOTHER-INFANT INTERACTIONS AND THE DEVELOPMENT OF SELF-REGULATION: A COMPARATIVE STUDY OF TERM AND PRETERM INFANTS. by Ana Rita Ribeiro Pires Dissertation presented as partial requirement for obtaining the Master’s degree in Statistics and Information Management, with a specialization in Information Analysis and Management. Advisor: Ana Cristina Costa (NOVA IMS, Universidade Nova de Lisboa) Co Advisor: Marina Fuertes (School of Education, Lisbon Polytechnic Institute) November 2018 ii ACKNOWLEDGEMENTS For these last two years, there were several people who contributed for the accomplishment of this master degree. Each one of them influenced me in a direct or indirect way and for that I would like to express my deeply gratitude, namely: Professor Ana Cristina Costa, for agreed to be my advisor in this dissertation, for her availability, advices and suggestions and for being the main responsible for my accomplishment. My first thankful note is for you, for offering help whenever I needed, for all of the kind words and patience during this last year and half. FCT / FEDER, for the financial support to the projects where are inserted the data used in this dissertation. Professor Marina Fuertes, for providing the data and the first insights of the educational and psychological theories studied in this thesis. Without your availability to share the data and willingness to have further analysis on it would have been impossible to move forward with this work. Professor João Moreira, for his kindness and for putting me in contact with Professor Marina Fuertes. My parents, Madalena Pires e Manuel Pires, for allowing me to make my own mistakes and learn from them, for my childhood, for my education and for the constant support in my adult life. David Matias, who agreed to be my everyday support for the good and bad times. Tiago Ramalho, my best friend, for that special coffee, for being always present even when we are in different countries. Ana Paula Cerejo, a family friend, who was responsible for the choice of the area of study of this thesis by introducing me to Professor João Moreira. Margarida Tedim and Paula Torrão, who I’ve met in the beginning of my master, for all the home meals and late night conversations. Inês Nunes and Carlos Patarrana, for volunteering to be critical reviewers, for all the suggestions and for being the amazing persons that they are. And for all those who had been with me for all these past months. iii ABSTRACT This dissertation aims to study the similarities between term and preterm infants in the motherinfant interactions and the infant’s development of the self-regulation. There are three main experiences related to the infant’s self-regulation development and the mother-infant interactions: Still-Face Experiment, Free-Play Situation and the Strange Situation Experiment. The Apgar scores (measure of the general health of the newborn) and the Care-Index scores were found to be significantly different for preterm and term infants. The Apgar scores were also different across the majority of the infant’s weight groups and mother’s age groups. Also, some of the Care-Index maternal scales (maternal sensitivity and passivity) were found to be significantly different for the majority of education levels. The relation between the Still-Face Experiment and the Strange Situation Experiment was already discussed in several studies and is supported by this dissertation: Self-Comfort, Socially Negative and Socially Positive behaviors are related with the Insecure, Resistant and Secure attachments. Both preterm and term infants displayed a Socially Positive behavior and a Secure attachment in the majority of cases. However, tests shown that the results of these experiences could depend on the sample type (preterm or term infants). When analyzed separately, preterm and term infants seem to have similar results in each analysis. However, this was not found to be true: the results for the Still-Face Experiment, Free-Play Situation and the Strange Situation Experiment were significantly different for preterm and term infants. KEYWORDS Preterm infants; Term infants; Apgar Score; Still-Face Experiment; Care-Index; Strange Situation Experiment. iv INDEX 1. Introduction ............................................................................................................................. 1 2. Literature review ...................................................................................................................... 4 3. Methodology ............................................................................................................................ 9 3.1. Data Collection ................................................................................................................. 9 3.2. Data Preparation ............................................................................................................ 11 3.3. Descriptive Analysis ........................................................................................................ 13 3.4. Exploratory Analysis ....................................................................................................... 13 3.5. Nonparametric Tests ...................................................................................................... 14 3.5.1. Ordinal Variables..................................................................................................... 14 3.5.2. Categorical Variables .............................................................................................. 16 3.5.3. Preterm and Term Infants ...................................................................................... 16 4. Results and Discussion ........................................................................................................... 18 4.1. Descriptive Analysis ........................................................................................................ 18 4.2. Exploratory Analysis ....................................................................................................... 23 4.2.1. Average Scores ........................................................................................................ 23 4.2.2. Correlation Matrix .................................................................................................. 27 4.3. Nonparametric Tests ...................................................................................................... 29 4.3.1. Ordinal Variables..................................................................................................... 29 4.3.2. Categorical Variables .............................................................................................. 34 4.3.3. Preterm and Term Infants ...................................................................................... 34 5. Conclusions ............................................................................................................................ 38 6. References .............................................................................................................................. 41 7. Appendix ................................................................................................................................ 44 7.1. Strange Situation Experience Schematics ...................................................................... 44 7.2. Autoregulation Patterns and Behavioral Styles .............................................................. 44 7.3. The Apgar Score .............................................................................................................. 45 7.4. Descriptive Analysis ........................................................................................................ 46 7.4.1. Infants’ Demographics ............................................................................................ 46 7.4.2. Mothers’ Demographics ......................................................................................... 49 7.4.3. Fathers’ Demographics ........................................................................................... 50 7.5. Nonparametric Tests ...................................................................................................... 51 7.5.1. Ordinal Variables..................................................................................................... 51 v LIST OF TABLES Table 3.1 - Sample Demographics ........................................................................................................... 9 Table 3.2 - Description of Variables ...................................................................................................... 10 Table 3.3 - Baby's Gestational Age in Groups ....................................................................................... 11 Table 3.4 - Baby's Weight in Groups ..................................................................................................... 12 Table 3.5 - Parents' Age in Groups ........................................................................................................ 12 Table 3.6 - Education Levels in Groups ................................................................................................. 12 Table 4.1 - Care-Index Scores by Intervals and Sample Type ................................................................ 20 Table 4.2 - Preterm: Still-Face vs. Strange Situation ............................................................................. 21 Table 4.3 - Term: Still-Face vs. Strange Situation .................................................................................. 21 Table 4.4 - Average Apgar Scores by Sample Type and Baby's Weight Group ..................................... 23 Table 4.5 - Average Apgar Scores by Sample Type and Mother's Age Group ....................................... 24 Table 4.6 - Mother’s Care-Index Scores by Sample Type and Mother's Age Group ............................. 25 Table 4.7 - Mother’s Care-Index Scores by Still-Face and Strange Situation Scores ............................. 26 Table 4.8 - Child's Care-Index Scores by Still-Face and Strange Situation Scores ................................. 26 Table 4.9 - Spearman's Correlation: Preterm Care-Index Scores .......................................................... 27 Table 4.10 - Spearman's Correlation: Term Care-Index Scores ............................................................. 28 Table 4.11 - Mann-Whitney Test Scores by Sample Type ..................................................................... 29 Table 4.12 - Kruskal-Wallis Test Scores by Baby Weight ....................................................................... 29 Table 4.13 - Nemenyi Test Scores: Conclusion for Apgar-1 by Baby’s Weight Group .......................... 30 Table 4.14 - Nemenyi Test Scores: Conclusion for Apgar-5 by Baby’s Weight Group .......................... 31 Table 4.15 - Kruskal-Wallis Test Scores by Mothers’ Age Group .......................................................... 31 Table 4.16 - Nemenyi Test Scores: Conclusion for Apgar-1 by Mother’s Age Group ............................ 32 Table 4.17 - Nemenyi Test Scores: Conclusion for Apgar-5 by Mother’s Age Group ............................ 32 Table 4.18 - Kruskal-Wallis Test Scores by Mothers’ Education Level .................................................. 33 Table 4.19 - Nemenyi Test Scores: Conclusion for Maternal Sensitivity by Mother’ Education Level . 33 Table 4.20 - Nemenyi Test Scores: Conclusion for Maternal Passivity by Mother’ Education Level .... 33 Table 4.21 - Pearson's Chi-Square Test Scores by Sample Type ........................................................... 34 Table 4.22 - Preterm: Mann-Whitney Test Scores by Baby Gender ..................................................... 34 Table 4.23 - Preterm: Pearson's Chi-Square Test Scores by Baby Gender ............................................ 35 Table 4.24 - Term: Mann-Whitney Test Scores by Baby Gender .......................................................... 35 Table 4.25 - Term: Pearson's Chi-Square Test Scores by Baby Gender ................................................. 35 Table 4.26 - Frequencies of the Still-Face Experiment by Sample Type................................................ 36 Table 4.27 - Frequencies of the Strange Situation Experiment by Sample Type .................................. 36 Table 4.28 - McNemar-Bowker Test results for Preterm vs. Term Infants ........................................... 36 Table 7.1 - Strange Situation Description .............................................................................................. 44 Table 7.2 - Match between Autoregulation Patterns and Behavioral Styles ........................................ 44 Table 7.3 - Apgar Score Form ................................................................................................................ 45 vi Table 7.4 - Infants' Demographics: Categorical Variables ..................................................................... 46 Table 7.5 - Infants' Demographics: Quantitative Variables ................................................................... 46 Table 7.6 - Distribution: Apgar Scores by Sample Type ........................................................................ 47 Table 7.7 - Frequency Distribution: Still-Face Patterns of Child Behavior Scores by Sample Type ....... 47 Table 7.8 - Frequency Distribution: Still-Face Behavioral Styles 1-7 Scores by Sample Type ............... 47 Table 7.9 - Distribution: Care-Index by Sample Type (Maternal Scores) .............................................. 48 Table 7.10 - Distribution: Care-Index by Sample Type (Child Scores) ................................................... 48 Table 7.11 - Frequency Distribution: Strange Situation Scores by Sample Type .................................. 48 Table 7.12 - Mothers' Demographics .................................................................................................... 49 Table 7.13 - Fathers' Demographics ...................................................................................................... 50 Table 7.14 - Nemenyi Test for Baby Weight Groups and Apgar-1 (Critical Value = 495,0) ................... 51 Table 7.15 - Nemenyi Test for Baby Weight Groups and Apgar-5 (Critical Value = 286,3) ................... 52 Table 7.16 - Nemenyi Test for Mother’s Age Groups and Apgar-1 (Critical Value = 503,0) ................. 53 Table 7.17 - Nemenyi Test for Mother’s Age Groups and Apgar-5 (Critical Value = 305,4) ................. 53 vii LIST OF FIGURES Exhibit 4.1 - Average Apgar Score by Sample Type ............................................................................... 18 Exhibit 4.2 - Still-Face Patterns by Sample Type ................................................................................... 18 Exhibit 4.3 - Still-Face Behavioral Styles by Sample Type ..................................................................... 19 Exhibit 4.4 - Average Care-Index Scores by Sample Type ..................................................................... 19 Exhibit 4.5 - Strange Situation Scores by Sample Type ......................................................................... 21 Exhibit 4.6 - Parents' Overall Demographics ......................................................................................... 22 Exhibit 4.7 - Agreement of the Still-Face Experiment ........................................................................... 37 Exhibit 4.8 - Agreement of the Strange Situation Experiment .............................................................. 37 LIST OF EQUATIONS Equation 3.1 - Sturge’s Rule .................................................................................................................. 11 Equation 3.2 - Class Width .................................................................................................................... 11 5 Still-Face Experiment Edward Tronick along with his colleagues introduced the Still-Face Experiment in a paper entitled “Infant Emotions in Normal and Perturbated Interactions” (Tronick, Adamson, Als & Brazelton, 1975). In their experiment both mother and infant are videotaped while interacting with each other. At some point, the mother stays “completely unresponsive, with a flat expressionless face for 3 minutes”. The infant’s first response was to catch his mother’s attention and, when unsuccessful, starts to ignore her. The Still-Face Experiment, formally known as the Still-Face Paradigm, comprises three episodes: a face-to-face play episode, a still-face episode and a reunion episode. The aim of this experiment is to assess the infants’ response to stressful events. Since Tronick’s experiment, there were several studies conducted on the Still-Face basis and on the two birth types (preterm and term birth). One of those studies, analysed the response to maternal mirroring, i.e. coping behavior, in the still face and replay tasks (Bigelow & Walden, 2009). The experimenters concluded that maternal mirroring had a positive reaction in the still face and replay tasks, supporting their prediction that maternal mirroring would enhance visual attention and affection. This reaction would support the type B attachment – secure. A more recent study, conducted by Fuertes, Beeghly, Santos & Tronick (2012), was looking for “Predictors of infant positive, negative and self-direct coping during face to face still-face in a Portuguese preterm sample”. This study was based on the premise that there were three different categories of coping behavior during Tronick’s Face to Face Still-Face Paradigm (Tronick et. al, 1978): a Positive Other-Directed Coping, a Negative Other-Directed Coping and a Self-Directed Coping. These coping behaviors could be related to three types of attachment discovered by Ainsworth: (B) Secure, (C) Resistant and (A) Insecure, respectively. Among others, they reached to the conclusion that both internal and external factors contribute to the infant’s self-regulation, i.e., gestational birth weight, heart rate, gestational age and maternal Care-Index Scores. According to their study, a Positive Other-Directed coping was positively correlated with gestational age (ρ = 0,304) and birth weight (ρ = 0,385) and with the Maternal Sensitivity Behavior - Care-Index Score - (ρ = 0,321); SelfDirected Coping was positively correlated with Maternal Controlling Behavior - Care-Index Score - (ρ = 0,308). The three categories of coping behavior are evaluated according to micro-expressions and behaviors (Fuertes, Beeghly, Santos & Tronick, 2012): - Positive Other-Directed Coping: looking at their mother’s face and/or body, reaching towards the mother, making positive vocalizations, smiling; - Negative Other-Directed Coping: arching and/or pushing away from the mother, making vocal protests, crying; - Self-Directed Coping: mouthing his/her hands and/or objects, closing his/her eyes, looking away from the mother, self-touching and/or squeezing one hand against the other. The mother-infant interactions are evaluated at each second accordingly to a point system: one point for an isolate behavior, three points for two behaviors happening simultaneously (depending on the intensity), and five points for three or more behaviors happening simultaneously. 6 The three autoregulation behaviors presented by Tronick et. al (1978) could also be divided into seven behavioral styles (Seixas, Barbosa & Fuertes, 2017) – see Appendix 7.2 Autoregulation Patterns and Behavioral Styles: - Positive Other-Directed Coping: (7) Positive, (6) Less Positive, (5) Positive with more difficulty in calm down; - Negative Other-Directed Coping: (4) Moderately Positive but does not calm down (3) Does not calm down/Very Negative/Interrupted Situation; - Self-Directed Coping: Self-Comfort/Look Away (2) and (1) No differentiation between episodes. Conradt & Ablow (2010) highlighted the importance of studying the mothers’ response in this context. In their discoveries, they have found out that “maternal sensitivity during and following distress uniquely predicted infant physiological and behavioral reactivity and regulation”. That is, maternal sensitivity is related with infant’s secure attachment, which translates into greater social competences and less problematic behaviors. According to Fuertes et al. (2009) about 70% of the infants with secure behaviors (Strange Situation Experiment) are related with Positive Other-Directed Coping behaviors (Still-Face Experiment) and mothers’ sensitivity (Care-Index). The mothers’ response to the Still-Face Experiment was already studied by Hsu & Jeng (2013) but only related with the infants’ birth status. They took into account that the type of sample (i.e. term or preterm) could negatively influence the mother’s response due to stereotyping. Their results were supported by the hypothesis that the Still-Face Experiment would affect in a different way the mothers of term and preterm and thus revealing that the experience would affect not only the infants’ response, but also the mothers’ response. There are also several studies that relate the infant’s demographics with the Still-Face results: eye contact is kept for longer periods of time in infants with six month compared with infants with three months (Toda & Fogel, 1993); Positive Other-Directed Coping behaviors for preterm infants with higher gestational age (Fuertes, Beeghly, Lopes dos Santos & Tronick, 2011); girls have a bigger capability of distinguish emotional expressions (McClure, 2000) and to respond to social stimuli (Alexander & Wilcox, 2012). These relations between the infants’ demographics and the Still-Face results are not always obtained, which could indicated a cultural difference. 7 Care-Index applied to the Free-Play Situation Crittenden & Ainsworth (1983) developed the thesis that “anxious (or insecure) attachment is a critical concept in regard to both the origin of family maltreatment and the rehabilitation of families (…) because attachment theory is a developmental theory, it is responsive to differences in the nature and effects of anxious attachment at different points in the lifespan”. This thesis was based on Bowlby’s Attachment Theory regarding the child’s behavior through the years. Later on, Patricia Crittenden developed the Care-Index to predict patterns of attachment. The CareIndex could be applied in a Free-Play Situation, from birth to 15 months, where the adult (parent) is expected to play with the child as he usual does, during three to five minutes, with no restrictions (Crittenden, 2005). This allows the observer to better evaluate the real behavior between the parent and the child. The child is evaluated in terms of Cooperation, Compulsion, Difficulty and Passivity; the parent is evaluated in terms of Sensitivity, Control and Passivity. With the scores ranging from 0 to 14, the observer assesses facial and vocal expressions, position of the body, among others aspects. These two evaluation types could be related, in the sense that both mother and infants reactions could influence one another: Infants who are more communicative give their mothers the opportunity to be more reactive, that is, the opportunity to display more sensitive reactions (Bigelow & Power, 2016). In her doctoral dissertation (Rotas de Vinculação: O desenvolvimento do comportamento interactivo e a organização da vinculação no primeiro ano de vida do bebé5, 2008), Fuertes found strong positive correlation between the mothers’ sensitivity and child’s cooperation (ρ = 0,861); between maternal control and child’s compulsion (ρ = 0,733) and passivity (ρ = -0,729); between mother’s passivity and child’s compulsion (ρ = -0,517) and passivity (ρ = 0,584). Fuertes also suggested an association between Maternal Care-Index Scores and the mothers’ level of education: higher education levels translate into higher sensitivity scores and smaller passivity scores. 5 Routes of Binding: The development of interactive behavior and the organization of binding in the first year of the infant's life 8 Medical Concepts and Definitions According to the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NIH), the expected gestational age of an infant is 40 weeks (about nine months) however that is not always the case. The World Health Organization (WHO) estimates that 15 million infants (from 5% up to 18% of infants born) are born before 37 weeks of gestation every year. Infants born with and after 37 weeks are considered to be “term” infants; between 32 and 37 weeks are considered “late preterm” infants; between 28 and 32 weeks are considered “very preterm”; and infants with gestational age inferior to 28 weeks are considered to be “extremely preterm”. The Center for Disease Control and Prevention (CDC) states that an infant that does not complete the 37 weeks of gestational age could develop breathing problems, along with development delay, vision problems and hearing impairment. Although this is considered a global problem, the majority of cases (60%) happen in Africa and South Asia, and within poorer families. There are several factors that could contribute to earlier birth, such as the mother’s previous pregnancies, multiple gestations (e.g. twins) and several medical conditions; and although 40% of the cases are of unknown nature, Tommy’s Organization states that “25% of preterm births are planned because the mother and/or baby are suffering life-threatening complications”. The newborns could be classified according to the Apgar Score which is used to assess the general health of the newborn and his response to resuscitation if needed. Introduced by Dr. Virginia Apgar in 1952, the Apgar Score is applied in the first and fifth minute after birth – Apgar-Index 1 and 5, respectively. This method takes into account five components: appearance (skin color), pulse (heart rate), grimace response (reflexes), activity (muscle tone) and respiration (breathing rate and effort). For each component is given a score of 0, 1 and 2 resulting in a possible score of 0 up to 10 – example of the Apgar Table in Appendix 7.3. According to NIH, only 1% of newborns born in developed countries have an Apgar Score inferior to 7 at five minutes. This conclusion could be related with the social and economic development of these countries (e.g. more knowledge, awareness and access to better medical care). A study in Germany (Straube et al., 2010) suggested that some biological maternal factors, such as overweight mothers, existence of previous children and the increasing of mother’s age (+35), could have influence on the Apgar-Index 5. The same was suggested for the maternal education level in Brazil (Almeida, Pedreira & Almeida, 2016). 9 3. METHODOLOGY 3.1. DATA COLLECTION The data presented are inserted in two investigation projects: “Vinculação em bebés prematuríssimos: o impacto dos fatores perinatais, da auto-regulação infantil, da sensibilidade materna e do desenvolvimento infantil” (PTDC/MHC-PED/1424/2014) and “Touchpoints: Estudo sobre os contributos da autorregulação infantil, do comportamento materno e da intervenção com touchpoints na qualidade da vinculação no 1º ano de vida” (PTDC/PSI-EDD/110682/2009), financed by the Fundação da Ciência e Tecnologia (FCT) and coordinated by Professor Marina Fuertes. The data collection was authorized by the Pedro Hispano Hospital Ethics Board and gathered by Marina Fuertes and Joana Lopes in the scope of their doctoral dissertations in Psychology by the Faculty of Psychology and Education Science of the University of Porto. The transcription and quotation of the data was performed by Marina Fuertes, Joana Lopes, Pedro Lopes dos Santos and Anabela Faria. The data used in this dissertation are based on interviews and observations; were inserted in the IBM SPSS Statistics software (.sav files) and later on converted to .xls files to facilitate their preparation. The sample is composed by 54 preterm and 84 full term infants and their parents (138 dyads). For each dyad (infant-mother) was collected the demographic variables listed in Table 3.1. Infant Mother Father Code Age Age Sample Type Civil Status Level of Education Gender Level of Education Occupation Gestational Age Occupation Birth Weight Number of children Apgar-Index 1 Apgar-Index 5 Type of Birth Number of Siblings Table 3.1 - Sample Demographics The Apgar-Index was applied at the first and fifth minute of the infant’s life (Apgar Index 1 and 5, respectively), as it is considered to be a measure of the general health after birth. The infants were evaluated in the Still-Face Experiment (applied at 3 months), Free-Play Situation evaluated with the Care-Index (applied at 9 months) and in the Strange Situation Experiment (applied at 12 months). The first and third experiments were classified as qualitative variables since they could take three possible scores each. In the Still-Face Experiment, the infant’s behavior is categorized as Socially Positive (i.e. Positive Other-Directed Coping behavior), Socially Negative (i.e. Negative Other-Directed Coping behavior) or Self-Comfort (i.e. Self-Directed Coping behavior); in the Strange Situation Experiment the infant could be classified as Secure, Insecure or Resistant. 10 The Free-Play Situation is evaluated according to the Care-Index. The Care-Index includes three maternal (M) scales (Sensitivity, Control and Passivity) and four child (C) scales (Cooperation, Compulsion, Difficulty and Passivity), each of them with a score of 0 to 14. All the variables listed above were summarized and described below according to the Table 3.2. Variable Description Units / Categories DB_Code Sample's Code Identification N.D. Baby_Type Sample Type Preterm vs. Term Gender_B Baby's Gender Male vs. Female Age_Group_B Baby's Gestational Age in Groups Weeks Weight_Group_B Birth Weight in Groups Grams Apgar_1 Apgar Score at the 1st minute Ordinal: 0 – 10 Apgar_5 Apgar Score at the 5th minute Ordinal: 0 – 10 Birth_Type Birth Type Induced vs. Spontaneous Siblings_B Number of baby' siblings Numeric Age_M Mother's Age Years Age_Group_M Mother's Age in Groups Years Status_M Mother's Civil Status Single, Married, Widow, Divorced Education_Group_M Mother's Academic Qualifications in Groups Categorical Occupation_M Mother's Occupation Categorical Children_M Number of Mother's Children Numeric Age_F Father's Age Years Age_Group_F Father's Age in Intervals Years Education_Group_F Father's Academic Qualifications in Groups Categorical Occupation_F Father's Occupation Categorical Still_Face_3m Still-Face Patterns of Child Behavior (3 months) Categorical Still_Face_3m_1_7 Still-Face Behavioral Styles 1-7 (3 months) Categorical M_Sensitivity Care-Index: Maternal Sensitivity Scale (9 months) Ordinal: 0 – 14 M_Control Care-Index: Maternal Control Scale (9 months) Ordinal: 0 – 14 M_Passivity Care-Index: Maternal Passivity Scale (9 months) Ordinal: 0 – 14 C_Cooperation Care-Index: Scale of Child Cooperation (9 months) Ordinal: 0 – 14 C_Compulsion Care-Index: Child Compulsion Scale (9 months) Ordinal: 0 – 14 C_Difficulty Care-Index: Children's Difficulty Temperament Scale (9 months) Ordinal: 0 – 14 C_Passivity Care-Index: Child Passivity Scale (9 months) Ordinal: 0 – 14 Strange_Group Strange Situation Classification (12 months) Categorical Table 3.2 - Description of Variables 11 3.2. DATA PREPARATION The analysis of the data was conducted using the SAS Enterprise Guide software (version 7.1), Microsoft Excel (version 2007), JMP software (Student Edition 12.1.0) and the IBM SPSS Statistics software (version 22.0). After converting the .sav files to .xls files, using the Import SPSS file data task in SAS, was necessary to normalize both samples regarding language and variables’ labels to allow a meaningful comparison between the two samples. The WHO presented three categories for preterm infants: extremely preterm (less than 28 weeks); very preterm (between 28 and 32 weeks); moderate to late preterm (32 to 37 weeks) and the Baby Corner Community acknowledge the existence of five terms to describe the infant’s gestational age: Extremely Preterm for infants born before 25 weeks; Preterm for infants born before 34 weeks; Late Preterm for when the infant is born between 34 and 37 weeks; Full-term for when the infant is born between 38 and 40 weeks; Post-Term for infants born after 40 weeks of gestation. For study purposes, the infants’ gestational age was divided into four different class intervals (see Table 3.3), associated with two birth groups: Preterm and Term. Baby's Gestational Age (in weeks) Gestational Age Definition* Birth Groups [ 31 – 34 [ Preterm Preterm [ 34 – 37 [ Late Preterm Preterm [ 37 – 40 [ Full Term Term [40 – 43 ] Post-Term Term Table 3.3 - Baby's Gestational Age in Groups * Baby Corner Community Due to the extensive possibilities, infants’ weight was also grouped according to the Sturge’s Rule (see Table 3.4). With a total of 138 observations , the Sturge’s Rule (Equation 3.1) gives us eight classes which, computed with the range of the data, gives of a width of 407 (Equation 3.2). Equation 3.1 - Sturge’s Rule Equation 3.2 - Class Width 12 In the present study, the mothers’ age range from 18 up to 42 years old and the fathers’ age range from 21 up to 51. Given this, and using the Sturge’s Rule (1), the parents’ age was grouped into eight intervals with a width (2) of four years (see Table 3.5): Birth Weight (in grams) [ 1050 - 1457 [ [ 1457 - 1864 [ [ 1864 - 2271 [ [ 2271 - 2678 [ [ 2678 - 3085 [ [ 3085 - 3492 [ [ 3492 - 3899 [ [ 3899 - 4350 ] Table 3.4 - Baby's Weight in Groups Parents’ Age (in years) [ 18 – 22 [ [ 22 – 26 [ [26 – 30 [ [ 30 – 34 [ [ 34 – 38 [ [ 38 – 42 [ [ 42 – 46 [ [ 46 – 51 ] Table 3.5 - Parents' Age in Groups Our samples were classified in a different way regarding the parents’ education level. The Preterm sample was classified in levels and the Term sample was classified in number of school years. In order to standardize the data, the parents’ education was classified according to the Levels of Education indicated below in Table 3.6. Level of Education Number of School Years Basic (1st Cycle) 1 to 4 years Basic (2nd Cycle) 5 to 6 years Basic (3rd Cycle) 7 to 9 years High School 10 to 12 years University Above 12 years Table 3.6 - Education Levels in Groups 13 3.3. DESCRIPTIVE ANALYSIS An exploratory analysis provides an overall view of the infants and their parents’ characteristics. The goal was to answer the three main questions about each sample type: - Which is the frequency distribution for each sample? - How are the samples distributed according to gender, gestational age and weight and Apgar indexes? - How are the samples distributed according to their Still-Face, Care-Index and Strange Situation scores? - How are the parents distributed according to their demographics? The Descriptive Analysis of the JMP software describes the data used in this study regarding all variables involved. This analysis is mainly focused on the two sample - Preterm and Term infants - and their variables (Apgar Scores, Still-Face Experiment, Care-Index applied in the Free-Play Situation, Strange Situation Experiment), as well as their parents overall demographics. 3.4. EXPLORATORY ANALYSIS The Exploratory Analysis goes beyond the simple description of the data as it could raise some suspicions regarding a possible relationship between a set of variables used in this research. The first section resulted from the Microsoft Excel’s Pivot Tables analysis: Average Scores for ApgarIndexes vs. mothers’ age and well as the infants’ weight; Average Maternal Care-Index Scores vs. mothers’ group age; Still-Face and Strange Situation Experiment and the Care-Index Scores. For the second section was used the Multivariate Analysis of the JMP software, where is presented the correlation scores for the Care-Index Scores. Taking into account that the Care-index is measured with ordinal values, the Spearman’s Coefficient was used to determine the correlation between variables. 14 3.5. NONPARAMETRIC TESTS The main variables of interest are the Apgar-Index 1 and 5 (ordinal variables), Still-Face Scores (nominal variable), Care-Index Scores (ordinal variable) and the Strange Situation Scores (nominal variable). Taking these types of variables into account, nonparametric approaches for ordinal data and specific techniques for nominal data must be applied – JMP Software. 3.5.1. Ordinal Variables To verify if the two independent samples (Preterm and Term infants) come from the same (or identical) population regarding the Apgar-Indexes and the Care-Index scores (applied in the Free-Play Situation), Wilcoxon-Mann-Whitney tests (also known as Mann-Whitney U test) were applied. Hypotheses: H0: Preterm and Term infants have identical scores of Apgar-Index 1 H1: Preterm and Term infants have distinct scores of Apgar-Index 1 H0: Preterm and Term infants have identical scores of Apgar-Index 5 H1: Preterm and Term infants have distinct scores of Apgar-Index 5 H0: Preterm and Term infants have identical scores of Care-Index H1: Preterm and Term infants have distinct scores of Care-Index To verify the existence of the relationship between the infants’ weight and their Apgar Scores and the relationship between the mothers’ age and the infants’ Apgar Scores the Kruskal-Wallis Test was applied: Hypothesis: H0: All Baby Weight Groups have identical scores of Apgar-Index 1 H1: There’s at least one Baby Weight Group with a distinct score of Apgar-Index 1 H0: All Baby Weight Groups have identical scores of Apgar-Index 5 H1: There’s at least one Baby Weight Group with a distinct score of Apgar-Index 5 H0: All Mothers’ Age Groups have identical scores of Apgar-Index 1 H1: There’s at least one of the Mothers’ Age Group with a distinct score of Apgar-Index 1 H0: All Mothers’ Age Groups have identical scores of Apgar-Index 5 H1: There’s at least one of the Mothers’ Age Group with a distinct score of Apgar-Index 5 21 The last evaluation was the Strange Situation experiment at the twelfth month of the infants’ life. According to Exhibit 4.5, preterm infants demonstrate more insecurity than term infants (38,89% vs. 11,90%), which are more Secure and Resistant (61,90% vs. 44,44% and 26,19% vs. 16,67%, respectively). Comparing the first and last experiments – Still-Face and Strange Situation Experiment – and using the cross-tabulation – see Table 4.2 –, we verify that the majority of preterm displays a socially positive behavior and has a secure attachment (40,74%) or a self-comfort behavior and an insecure attachment (25,93%). Strange Situation Insecure Resistant Secure Total Still-Face Self-Comfort 14 0 1 15 Socially Negative 0 9 1 10 Socially Positive 7 0 22 29 Total 21 9 24 54 Table 4.2 - Preterm: Still-Face vs. Strange Situation Taking into account Table 4.3 below we see that, like preterm infants, the majority of term infants displays a socially positive behavior and has a secure attachment (40,48%). However, there is a significant part that displays a socially negative behavior but has a secure attachment (17,86%). Strange Situation Insecure Resistant Secure Total Still-Face Self-Comfort 5 3 3 11 Socially Negative 1 10 15 26 Socially Positive 4 9 34 47 Total 10 22 52 84 Table 4.3 - Term: Still-Face vs. Strange Situation 0,00% 10,00% 20,00% 30,00% 40,00% 50,00% 60,00% 70,00% Secure Insecure Resistant 44,44% 38,89% 16,67% 61,90% 11,90% 26,19% Frequency (%) Type of Attachments Preterm Term Exhibit 4.5 - Strange Situation Scores by Sample Type 22 For better understanding of the experiments results, mothers were interviewed regarding their demographics and their infant’s father demographics. This information is summarized in Exhibit 4.6. Most mothers have ages between 28 and 33 years old (38,41%), a superior education, i.e. University degree (49,28%), and have an intellectual and/or scientific profession (29,71%). These statements are also true for each sample type except for the occupation: for preterm’s mothers the biggest percentage is for the Operators of assembly plants and machines and assembly workers – see Appendix 7.4.2 Mothers’ Demographics. A typical father is considered to also have between 28 and 33 years (35,51%), a superior education (36,96%), and to be a Specialist in Intellectual and/or Scientific Profession (21,74%). However, preterm’s fathers are best characterized for having the 2nd or 3rd cycle and for being unskilled workers – see Appendix 7.4.3 Fathers’ Demographics. Mean Age: 30,16 Mean No. Children: 1,52 Education Group Mode: University Occupation Mode: Specialist in Intellectual and/or Scientific Profession Mean Age: 32,90 Education Group Mode: University Occupation Mode: Specialist in Intellectual and/or Scientific Profession Exhibit 4.6 - Parents' Overall Demographics 23 4.2. EXPLORATORY ANALYSIS 4.2.1. Average Scores Since their development is not completely done, preterm infants could have low weight, which could be related to their low Apgar Scores. Looking at Table 4.4, the average Apgar Score does not seem to increase along with the several weight groups – statistical tests will be made to assess the relationship between these variables. Sample Type Baby's Weight Group Apgar-1 Apgar-5 Preterm [ 1050 - 1457 [ 7,57 9,00 [ 1457 - 1864 [ 7,64 8,40 [ 1864 - 2271 [ 8,07 8,75 [ 2271 - 2678 [ 7,60 8,25 [ 2678 - 3085 [ 8,00 9,00 Term [ 2271 - 2678 [ 9,00 10,00 [ 2678 - 3085 [ 9,19 10,00 [ 3085 - 3492 [ 9,06 9,97 [ 3492 - 3899 [ 9,10 10,00 [ 3899 - 4350 ] 9,22 9,89 Table 4.4 - Average Apgar Scores by Sample Type and Baby's Weight Group 24 According to NIH, a woman’s age is one of the risk factors in pregnancy that could lead to a premature birth, which could lead to smaller Apgar Scores (Exhibit 4.1). The organization puts women older than 35 years as risk groups, “because they are more likely to have other conditions (such as high blood pressure and diabetes)”. Looking at Table 4.5, is not possible to verify if the mother’s age could influence the infant’s Apgar Scores. For assessing the relationship between these two variables, statistical tests will be made. Sample Type Mother's Age Group Apgar-1 Apgar-5 Preterm [ 18 - 22 [ 7,44 8,83 [ 22 - 26 [ 7,71 8,50 [ 26 - 30 [ 7,44 8,38 [ 30 - 34 [ 8,42 8,83 [ 34 - 38 [ 8,40 9,00 [ 38 - 42 ] 7,25 8,33 Term [ 18 - 22 [ 9,00 10,00 [ 22 - 26 [ 9,00 10,00 [ 26 - 30 [ 8,89 9,94 [ 30 - 34 [ 9,18 9,97 [ 34 - 38 [ 9,25 10,00 [ 38 - 42 ] 9,20 10,00 Table 4.5 - Average Apgar Scores by Sample Type and Mother's Age Group 25 The general impression is that preterm infants require extra care due to the consequences of their inferior gestational age. This could lead to the hypothesis that mother’s behavior differs for each sample type. However, in Exhibit 4.4 - Average Care-Index Scores by Sample Type was shown that mothers of both preterm and term infants are more sensitive and less passive. In the preterm sample, the sensitivity score is more evident in mothers between 34 and 38 and between 38 and 42 years old for the passive score; in the term sample the sensitivity score is predominant in mothers between 18 and 22 years old and 22 and 26 years old, respectively – see Table 4.6. Sample Type Mother's Age Group M_Sensitivity M_Control M_Passivity Preterm [ 18 - 22 [ 5,89 5,78 2,33 [ 22 - 26 [ 6,14 3,43 4,43 [ 26 - 30 [ 6,88 3,76 3,35 [ 30 - 34 [ 7,17 4,50 2,33 [ 34 - 38 [ 7,60 6,00 0,40 [ 38 - 42 ] 5,50 5,00 3,50 Term [ 18 - 22 [ 11,00 2,00 1,00 [ 22 - 26 [ 8,33 2,50 3,00 [ 26 - 30 [ 9,06 4,11 0,83 [ 30 - 34 [ 9,18 3,21 1,61 [ 34 - 38 [ 9,19 2,19 2,69 [ 38 - 42 ] 9,60 3,80 0,60 Table 4.6 - Mother’s Care-Index Scores by Sample Type and Mother's Age Group For further conclusions, statistical tests should be made. 26 When analyzing the Still-Face and Strange Situation experiments, we see that sensitive mothers could have higher change of having infants with socially positive behavior and secure infants; controlling mothers could have infants who exhibit self-comfort behavior and insecure attachment; and higher passive scores could translate into infants with socially negative behavior and resistant attachment – see Table 4.7. Experiment Scores M_Sensitivity M_Control M_Passivity Still-Face Self-Comfort 6,04 5,46 2,50 Socially Negative 7,75 3,33 2,89 Socially Positive 9,12 3,32 1,58 Strange Situation Insecure 5,87 5,61 2,52 Resistant 7,32 3,65 3,03 Secure 9,47 2,99 1,54 Table 4.7 - Mother’s Care-Index Scores by Still-Face and Strange Situation Scores The same pairs could be achieved by analyzing the infants’ Care-Index Score – see Table 4.8. Cooperative infants could have more socially positive behavior and a secure attachment; compulsive infants have higher chance of having self-comfort behaviors and an insecure attachment; difficult infants could display self-comfort behavior and a resistant attachment; passive infants have socially negative behavior and resistant attachment. Experiment Scores C_Cooperation C_Compulsion C_Difficulty C_Passivity Still-Face Self-Comfort 5,81 3,62 2,58 2,08 Socially Negative 7,83 1,06 1,86 3,22 Socially Positive 9,37 1,37 1,53 1,96 Strange Situation Insecure 5,81 3,90 2,19 2,10 Resistant 7,39 0,94 2,35 3,32 Secure 9,68 1,13 1,43 1,99 Table 4.8 - Child's Care-Index Scores by Still-Face and Strange Situation Scores 27 4.2.2. Correlation Matrix Looking at Table 4.9, it is possible to see strong positive correlations in the preterm sample between the Child’s Cooperation and the Mother’s Sensitivity (0,84), between the Child’s Compulsion and the Mother’s Control (0,79) and between the Child and Mother’s Passivity (0,70). The Mother’s Control is negatively correlated with the Mother’s Passivity Scale (-0,63) and with the Child’s Passivity (-0,62) and Child’s Cooperation (-0,55); also the Child’s Compulsion is negatively correlated with the Mother’s Passivity (-0,55), Child’s Cooperation (-0,54), Difficulty (-0,54) and Passivity (-0,52). Variable by Variable Spearman ρ Prob>|ρ| M_Control M_Sensivity -0,4397 0,0009 M_Passivity M_Sensivity -0,3447 0,0107 M_Passivity M_Control -0,6258 <,0001 C_Cooperation M_Sensivity 0,8354 <,0001 C_Cooperation M_Control -0,5512 <,0001 C_Cooperation M_Passivity -0,1215 0,3814 C_Compulsion M_Sensivity -0,3538 0,0087 C_Compulsion M_Control 0,7934 <,0001 C_Compulsion M_Passivity -0,5460 <,0001 C_Compulsion C_Cooperation -0,5397 <,0001 C_Difficulty M_Sensivity -0,0607 0,6628 C_Difficulty M_Control -0,1113 0,4231 C_Difficulty M_Passivity 0,2986 0,0283 C_Difficulty C_Cooperation 0,0343 0,8054 C_Difficulty C_Compulsion -0,5380 <,0001 C_Passivity M_Sensivity -0,0580 0,6769 C_Passivity M_Control -0,6231 <,0001 C_Passivity M_Passivity 0,6958 <,0001 C_Passivity C_Cooperation 0,0142 0,9187 C_Passivity C_Compulsion -0,5219 <,0001 C_Passivity C_Difficulty -0,1107 0,4255 Table 4.9 - Spearman's Correlation: Preterm Care-Index Scores 28 In the term sample (Table 4.10) we have less obvious correlations. Like the preterm sample, the Child’s Cooperation and Mother’s Sensitivity have a strong positive correlation (0,90). However, when we look at the reimaging variables, none of them stands out. Nevertheless, the Mother’s Control was found to be negatively correlated with the Mother’s Sensitivity (-0,62) and with the Child’s Cooperation (-0,53). Variable by Variable Spearman ρ Prob>|ρ| M_Control M_Sensivity -0,6233 <,0001 M_Passivity M_Sensivity -0,3564 0,0009 M_Passivity M_Control -0,4445 <,0001 C_Cooperation M_Sensivity 0,8955 <,0001 C_Cooperation M_Control -0,5289 <,0001 C_Cooperation M_Passivity -0,3399 0,0016 C_Compulsion M_Sensivity -0,2424 0,0263 C_Compulsion M_Control 0,3457 0,0013 C_Compulsion M_Passivity -0,0884 0,4241 C_Compulsion C_Cooperation -0,1774 0,1064 C_Difficulty M_Sensivity -0,3138 0,0036 C_Difficulty M_Control 0,4154 <,0001 C_Difficulty M_Passivity -0,0934 0,3982 C_Difficulty C_Cooperation -0,3618 0,0007 C_Difficulty C_Compulsion 0,0017 0,9880 C_Passivity M_Sensivity -0,3094 0,0042 C_Passivity M_Control -0,0629 0,5700 C_Passivity M_Passivity 0,3944 0,0002 C_Passivity C_Cooperation -0,4357 <,0001 C_Passivity C_Compulsion -0,1865 0,0894 C_Passivity C_Difficulty -0,3745 0,0004 Table 4.10 - Spearman's Correlation: Term Care-Index Scores Given this, we can state that preterm’s’ mothers show a more obvious correlation between several variables in study and on the contrary, term infants’ mothers do not display a strong correlation between any variables, with the exception of the Maternal Sensitivity Scale and the Child’s Cooperation Scale, the Maternal Control and Sensitivity Scale and Child’s Cooperation Scale. 29 4.3. NONPARAMETRIC TESTS 4.3.1. Ordinal Variables For the ordinal variables under study (Apgar-Index 1, Apgar-Index 5, Care-Index Scores) and to check if the samples of Preterm and Term infants come from identical populations, the Mann-Whitney Test was applied – see Table 4.11. Variables Chi-Square Prob > ChiSq Apgar-1 51,8957 <,0001 Apgar-5 100,5765 <,0001 M_Sensitivity 23,9614 <,0001 M_Control 4,0391 0,0445 M_Passivity 2,4240 0,1195 C_Cooperation 24,1027 <,0001 C_Compulsion 20,7584 <,0001 C_Difficulty 0,0784 0,7794 C_Passivity 8,4712 0,0036 Table 4.11 - Mann-Whitney Test Scores by Sample Type As our threshold is a level of significance of 5%, we reject the null hypothesis for all variables except for M_Passivity and C_Difficulty. Since there is no statistical evidence to state that all variables groups come from identical populations, we can state that the Apgar Scores and Care-Index Scores are significantly different in the Preterm and Term infants. Section 4.2.1 Average Scores suggested the existence of a relation between the infants’ weight and their Apgar Scores (Table 4.4). For better assessment, the Kruskal-Wallis test was conducted – see Table 4.12. Variables Chi-Square Prob>ChiSq Apgar-1 50,0622 <,0001 Apgar-5 78,0718 <,0001 Table 4.12 - Kruskal-Wallis Test Scores by Baby Weight Given our results we can conclude that we reject the hypothesis of the Apgar-Indexes being identical in all infants’ weight groups, at the 5% significance level. 30 Taking into account the previous results, further tests were made in order to determine the differences between groups. For this, the nonparametric multiple comparisons test (post-hoc test) chosen was the Nemenyi test. According to Table 4.13 and for a level of significance of 5% and a we reject the null hypothesis of having identical scores for Apgar-1 for 17 out of 28 pairs (60,71% of the total pairs). Level + Level - Difference Between Rank Sums Absolute Difference Conclusion [ 1050 - 1457 [ [ 2678 - 3085 [ -1842,00 1842,00 Reject H0 [ 1050 - 1457 [ [ 3085 - 3492 [ -2421,00 2421,00 Reject H0 [ 1050 - 1457 [ [ 3492 - 3899 [ -1431,00 1431,00 Reject H0 [ 1457 - 1864 [ [ 2678 - 3085 [ -1646,50 1646,50 Reject H0 [ 1457 - 1864 [ [ 3085 - 3492 [ -2225,50 2225,50 Reject H0 [ 1457 - 1864 [ [ 3492 - 3899 [ -1235,50 1235,50 Reject H0 [ 1864 - 2271 [ [ 2678 - 3085 [ -1373,50 1373,50 Reject H0 [ 1864 - 2271 [ [ 3085 - 3492 [ -1952,50 1952,50 Reject H0 [ 1864 - 2271 [ [ 3492 - 3899 [ -962,50 962,50 Reject H0 [ 2271 - 2678 [ [ 2678 - 3085 [ -1176,00 1176,00 Reject H0 [ 2271 - 2678 [ [ 3085 - 3492 [ -1755,00 1755,00 Reject H0 [ 2271 - 2678 [ [ 3492 - 3899 [ -765,00 765,00 Reject H0 [ 2678 - 3085 [ [ 3085 - 3492 [ -579,00 579,00 Reject H0 [ 2678 - 3085 [ [ 3899 - 4350 ] 1269,00 1269,00 Reject H0 [ 3085 - 3492 [ [ 3492 - 3899 [ 990,00 990,00 Reject H0 [ 3085 - 3492 [ [ 3899 - 4350 ] 1848,00 1848,00 Reject H0 [ 3492 - 3899 [ [ 3899 - 4350 ] 858,00 858,00 Reject H0 Table 4.13 - Nemenyi Test Scores: Conclusion for Apgar-1 by Baby’s Weight Group 37 The SAS Enterprise Guide software gives us the possibility to verify the level of agreement between the two sample types. Accordingly to the Exhibit 4.7 below, we see two Still-Face behaviors with an exact agreement: Self-Comfort and Socially Positive behavior. However, looking at the Socially Negative behavior, we could only see a partial agreement, which could tell us that it might differ between the preterm and term infants. Exhibit 4.7 - Agreement of the Still-Face Experiment For the Strange Situation experiment (see Exhibit 4.8), we also have two type of attachment with an exact agreement and one without agreement – Resistant attachment. This result is aligned with the McNemar-Bowker Test results, since the hypothesis of the results of the Strange Situation being equal regardless the sample type is rejected. Exhibit 4.8 - Agreement of the Strange Situation Experiment Self-Comfort Socially Negative Socially Positive Preterm Infants Self-Comfort Socially Negative Socially Positive Term Infants 0 10 20 30 40 50 Cumulative Frequency Agreement of StillFace_Preterm and StillFace_Term Self-Comfort Socially Negative Socially Positive Preterm Infants Self-Comfort Socially Negative Socially Positive Term Infants 0 10 20 30 40 50 Cumulative Frequency Partial Agreement Exact Agreement Agreement of StillFace_Preterm and StillFace_Term Insecure Resistant Secure Preterm Infants Insecure Resistant Secure Term Infants 0 10 20 30 40 50 Cumulative Frequency Agreement of Strange_Preterm and Strange_Term Insecure Resistant Secure Preterm Infants Insecure Resistant Secure Term Infants 0 10 20 30 40 50 Cumulative Frequency Partial Agreement Exact Agreement Agreement of Strange_Preterm and Strange_Term 38 5. CONCLUSIONS This study comprises detailed analyses of a sample comprised by two smaller independent samples: preterm infants (39,13%) and term infants (60,87%), from which 56,52% are male and 43,48% are female. Independently of the sample type, according to our results, the Apgar score is smaller at the first minute when compared with the fifth minute, which is aligned with the theoretical review since it measures the general well-being of the infant at birth. The first experience was conducted at the third month of the infants’ life in which the following were observed: (1) the Still-Face patterns of child behavior, and (2) the Still-Face styles of child behavior. The majority of infants exhibits socially positive behaviors (53,70% of preterm infants and 55,95% of term infants). However, preterm infants displayed more self-comfort behaviors (27,78%) and term infants a more socially negative behavior (30,95%). Regarding the Still-Face styles, the majority of term infants exhibits a positive behavior, but with difficulty in calming down after the mother’s return (35,71%), followed by a moderate positive behavior (17,86%); however, most of preterm infants exhibits a self-comfort behavior (27,78%), followed by a positive with difficulty of calming down after the mother’s return (22,22%). In this study no infant displayed no differentiation between intervals. At their ninth month, the infants and their mothers were observed in a Free-Play Situation. For this experience, it was applied the Care-Index for each dyad. It was verified that both preterm infants (6,80 on average) and term infants (9,26 on average) were more cooperative, and their mothers were more sensitive (6,65 on average for the preterm infants and 9,17 on average for term infants). On the other side of the children scale, preterm infants are less passive (1,76 on average) and term infants are less compulsive (0,60 on average). The Maternal Care-Index and the mothers’ age for the preterm sample was also analyzed: mothers with ages between 34 and 38 years old are more sensitive (7,60 on average) but also more controlling (6,00 on average) and less passive (0,40 on average). For the term infants, mothers between 18 and 22 years old were more sensitive (11,00 on average) and also the less controlling (2,00 on average). The third experience was the Strange Situation, conducted at their twelfth month, which showed that the majority of infants has a secure attachment (44,44% of preterm infants and 61,90% of term infants). However, preterm infants demonstrate to be more insecure (38,89%) than term infants who were more resistant (26,19%). When comparing the Still-Face and Strange Situation scores, the results show that the majority of preterm and term infants displayed socially positive behavior as well as a secure attachment (40,74% of preterm infants and 40,48% of term infants). These two experiences were also compared with the Care-Index Scores: sensitive mothers had infants which displayed socially positive behaviors (9,12 on average) and secure attachment (9,47 on average); controlling mothers had infants which displayed self-comfort behaviors (5,46 on average) and insecure attachment (5,61 on average); passive mothers had infants with socially negative behaviors (2,89 on average) and resistant attachment (3,03 on average). On the children scale, cooperative infants had also more socially positive behaviors (9,37 on average) and secure attachment (9,68 on average); compulsion infants displayed self-comfort behaviors (3,62 on average) and insecure attachment (3,90 on average); difficult infants displayed also self-comfort behaviors (2,58 on average) and a resistant attachment (2,35 on 39 average); passive infants had more socially negative behaviors (3,22 on average) and a resistant attachment (3,32 on average). These results are aligned with previous studies, since Fuertes, Beeghly, Santos & Tronick (2012) found positive correlation between positive behavior (Still-Face Experiment) and Maternal Sensitivity (Care-Index) and between self-comfort behavior (Still-Face Experiment) and Maternal Control (Care-Index). Conradt & Ablow (2010) also found a positive correlation between secure attachment (Strange Situation) and Maternal Sensitivity (Care-Index). In this study several significant correlations were found for both preterm infants and term infants. For the preterm sample: mother’s control is correlated with mother’s passivity (negative), child’s cooperation (negative), child’s compulsion (positive) and child’s passivity (negative); mother’s sensitivity is correlated with child’s cooperation (positive); mother’s passivity is correlated with child’s compulsion (negative) and child’s passivity (positive); child’s compulsion is correlated with child’s cooperation (negative), child’s difficulty (negative) and child’s passivity (negative). For the term sample: mother’s sensitivity is correlated with mother’s control (negative) and child’s cooperation (positive); mother’s control is correlated with child’s cooperation (negative). Some of these results are in line with the research done by Fuertes (2008), who found a correlation between the mothers’ sensitivity and child’s cooperation (positive), between maternal control and child’s compulsion (positive) and passivity (negative), and between mother’s passivity and child’s compulsion (negative) and passivity (positive). Using nonparametric methods, we verified that the Apgar scores and Care-Index scores were significantly different in the preterm and term samples. When considering both samples, the Apgar scores are significantly different for certain groups of infants’ weight and mothers’ age groups. However, the same could not be concluded for the Maternal Care-Index scores that did not display significant differences across all the mothers’ age groups. The same statistical tests were applied for the Maternal Care-Index scores and the mothers’ education level: maternal sensitivity and maternal passivity demonstrate to be significantly different for the majority of educational levels. With the Pearson’s Chi-Square test it was also demonstrated that preterm infants’ scores for the Still-Face Experiment are not independent from the term infants’ scores (at a 6% level of significance). The same can be concluded for the Strange Situation Experiment at a 5% level of significance). The same nonparametric tests were applied for the preterm and term samples, separately, using baby’s gender as an independent variable. The Apgar scores, Care-Index scores, Still-Face scores and Strange Situation scores didn’t shown significant differences between boys and girls for both the preterm and term samples, with the exception of the Child’s Passivity in the term sample. Using the cross-tabulation, it was verified that the majority of infants displayed a socially positive behavior (37,04%) and had a secure attachment (27,78%). When checking for the importance of the sample type in the Still-Face and Strange Situation experiences, we found that the results we found that the results differ for each experience, at a 10% significance level. The Social Negative behavior (Still-Face Experiment) and the Resistance attachment (Strange Situation Experiment) did not display an exact agreement. This conclusion make sense because, according to the literature review, Social Positive behavior is related with a Secure attachment, Social Negative behavior is related with Resistance attachment and Self-Comfort is related with an Insecure attachment. 40 The purpose of this Dissertation was to answer the following question: Are the Still-face Experiment, Free-Play Situation and the Strange Situation Experimente similar in preterm and term Infants? For a 10% significance level, we can conclude that all three experiences are different for the preterm and term infants. The Care-Index scores (Free-Play Situation) were found to be significantly different for preterm and term infants (at a 5% significance level). The results obtained in the Still-Face and Strange Situation experience were also found to be different for preterm and term infants, at a 10% and 5% significance level, respectively. These results indicate a strong possibility of significant differences in the mother-infant interactions and the development of self-regulation taking into account their gestational age. Some caution is warranted since the sample size was relatively small. Whilst these results indicate some potentially interesting avenues for research, more statistical work with larger samples would provide for more solid results. In psychology, the sample size (54 preterm infants and 84 term infants) is not considered small. That is, for an experimental and longitudinal study, the sample size is acceptable. However, from a statistical point of view, some variables could be benefit from a larger sample size. For solid results, further statistical tests should be made using larger and more heterogeneous samples. 41 6. REFERENCES Adamson, L. B., & Frick, J. E. (2003). The Still Face: A History of a Shared Experimental Paradigm. Infancy, 4(4), 451–473. https://doi.org/10.1207/S15327078IN0404_01 Almeida, N. K. O., Pedreira, C. E., & Almeida, R. M. V. R. (2016). Impact of maternal education level on risk of low Apgar score. Public Health, 140, 244–249. https://doi.org/10.1016/j.puhe.2016.04.009 Bigelow, A. E., & Power, M. (2016). Effect of Maternal Responsiveness on Young Infants’ Social Bidding-Like Behavior during the Still Face Task. Infant and Child Development, 25(3), 256–276. https://doi.org/10.1002/icd.1974 Bigelow, A., & Walden, L. (2009). Infants’ Response to Maternal Mirroring in the Still Face and Replay Tasks. Infancy, 14(5), 526–549. https://doi.org/10.1080/15250000903144181 Bowker, A.H. (1948). A test for symmetry in contingency tables. Journal of American Statistical Association, 43, 572-574 Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28(5), 759–775. https://doi.org/10.1037/0012-1649.28.5.759 Conradt, E., & Ablow, J. (2010). Infant physiological response to the still-face paradigm: Contributions of maternal sensitivity and infants’ early regulatory behavior. Infant Behavior and Development, 33(3), 251–265. https://doi.org/10.1016/j.infbeh.2010.01.001 Crittenden, P. M. (2005). Using the CARE-Index for Screening, Intervention, and Research, (XXIV), 99– 106. Crittenden, P. M., & Ainsworth, M. D. S. (1983). Child maltreatment and attachment theory. In D. Cicchetti & V. Carlson (Eds.), Child maltreatment (pp. 432–463). Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511665707.015 Fuertes, M. (2004). Rotas da vinculação: O desenvolvimento do comportamento interactivo e a organização da vinculação no primeiro ano de vida do bebé prematuro. Faculdade de Psicologia e de Ciências da Educação da Universidade do Porto. Fuertes, M. (2010). Exploratory study about atypical attachment: disorganization or adaptation? Psychologica, 1(52), 349–370. https://doi.org/10.14195/1647-8606_52-1_17 Fuertes, M., Beeghly, M., Santos, P. L. dos, & Tronick, E. (2012). Predictors of infant positive, negative and self-direct coping during face to face still-face in a Portuguese preterm sample. Análise Psicológica, 29(4), 553–565. https://doi.org/10.14417/ap.103 Fuertes, M., Faria, A., Soares, H., & Crittenden, P. (2009). Developmental and evolutionary assumptions in a study about the impact of premature birth and low income on mother–infant interaction. Acta Ethologica, 12(1), 1–11. https://doi.org/10.1007/s10211-008-0051-4 Fuertes, M., Santos, P. L. D., Beeghly, M., & Tronick, E. (2006). More Than Maternal Sensitivity Shapes Attachment: Infant Coping and Temperament. Annals of the New York Academy of Sciences, 1094(1), 292–296. https://doi.org/10.1196/annals.1376.037 42 Garcia, J. L. (1995). Freud’s Psychosexual Stage Conception: A Developmental Metaphor for Counselors. Journal of Counseling & Development, 73(5), 498–502. https://doi.org/10.1002/j.1556-6676.1995.tb01785.x Hsu, H.-C., & Jeng, S.-F. (2013). Differential Effects of Still-Face Interaction on Mothers of Term and Preterm Infants. Infant Mental Health Journal, 34(4), 267–279. https://doi.org/10.1002/imhj.21386 Ingleby, D. (1983). Freud and Piaget: The phoney war. New Ideas in Psychology, 1(2), 123–144. https://doi.org/10.1016/0732-118X(83)90005-3 Seixas, Í., Barbosa, M., & Fuertes, M. (2017). Contributos para a auto-regulação do bebé no Paradigma Face-to-Face Still-Face. Análise Psicológica, 35(4), 469–485. https://doi.org/10.14417/ap.1280 Simon, K., & Bowlby, J. (1951). Maternal Care and Mental Health. The Milbank Memorial Fund Quarterly, 29(4), 496. https://doi.org/10.2307/3348276 Statement, P. (2015). The Apgar Score. Pediatrics, 136(4), 819–822. https://doi.org/10.1542/peds.2015-2651 Straube, S., Voigt, M., Jorch, G., Hallier, E., Briese, V., & Borchardt, U. (2010). Investigation of the association of Apgar score with maternal socio-economic and biological factors: an analysis of German perinatal statistics. Archives of Gynecology and Obstetrics, 282(2), 135–141. https://doi.org/10.1007/s00404-009-1217-7 Svenvik, M., Brudin, L., & Blomberg, M. (2015). Preterm Birth: A Prominent Risk Factor for Low Apgar Scores. BioMed Research International, 2015, 1–8. https://doi.org/10.1155/2015/978079 Valera, A. P. S., Nunes, C., Oliveira, D. L. de, Cardoso, E. da R., Silva, H. C., Kuball, M. A. de S., … Santos, V. C. dos. (2012). John Bowlby: As sete características do apego, as quatro classificações dos padrões de apego e o cuidador. 43 Websites: Center for Disease Control and Prevention. Preterm Birth (URL: https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pretermbirth.htm, June 2017) Eunice Kennedy Shriver national Institute of Child Health and Human Development. Preterm Labor and Birth: Condition Information (URL: https://www.nichd.nih.gov/health/topics/preterm/conditioninfo/Pages/default.aspx, June 2017) Jean Piaget Society. A Brief Biography of Jean Piaget (URL: http://www.piaget.org/aboutPiaget.html, June 2017) Medline Plus. Apgar score (URL: https://medlineplus.gov/ency/article/003402.htm, October 2017) Medline Plus. Birth Weight (URL: https://medlineplus.gov/birthweight.html, October 2017) Real Statistics Using Excel. Follow-up Tests to Kruskal-Wallis (URL: http://www.realstatistics.com/one-way-analysis-of-variance-anova/kruskal-wallis-test/follow-up-tests-kruskalwallis/, April 2018) The American College of Obstetricians and Gynecologists. Definition of Term Pregnancy (URL: https://www.acog.org/Clinical-Guidance-and-Publications/Committee-Opinions/Committee-onObstetric-Practice/Definition-of-Term-Pregnancy, March 2018) The American Congress of Obstetricians and Gynecologists. The Apgar Score (URL: https://www.acog.org/Resources-And-Publications/Committee-Opinions/Committee-onObstetric-Practice/The-Apgar-Score, October 2017) The Baby Corner. Preterm, Full Term & Post-term Babies: What Do They Mean? (URL: http://www.thebabycorner.com/page/2952/, March 2018) Tommy’s Organization. Why we exist (URL: https://www.tommys.org/our-organisation/why-weexist, June 2017) World Health Organization. Preterm Birth (URL: http://www.who.int/mediacentre/factsheets/fs363/en/, June 2017) 44 7. APPENDIX 7.1. STRANGE SITUATION EXPERIENCE SCHEMATICS Episode Participants Duration Description 1 Mother, Infant, Stranger 30 seconds A third person introduces the playroom to the mother and her infant. 2 Mother, Infant 3 minutes The mother and the infants are left in the playroom. 3 Mother, Infant, Stranger 3 minutes The third person enters into the room and (1) seats, (2) talks with the mother, (3) approach the infant and (4) mother leaves the room. 4 Infant, Stranger 3 minutes maximum The third person interacts with the infant and comforts him if necessary. 5 Mother, Infant 3 minutes The mother enters into the room and the third person leaves. The mother interacts with the infant. 6 Infant 3 minutes maximum The mother leaves the room. 7 Infant, Stranger 3 minutes maximum The third person enters into the room and interacts with the infant. 8 Mother, Infant 3 minutes The mother enters into the room, pick-up the infant and the third person leaves the room. Table 7.1 - Strange Situation Description 7.2. AUTOREGULATION PATTERNS AND BEHAVIORAL STYLES Behavioral Styles Description of Behavioral Styles Autoregulation Patterns VII Prolonged and intense interaction in an emotionally positive mood, with distinct change during the Still-Face episode and subsequent recovery Positive OtherDirected Coping VI Good level of interactive participation in the first episode, with distinct disturbance during the Still-Face episode, without great difficulties in recovering subsequently V Good to moderate level of interactive participation in the first episode with disturbance during the Still-Face episode, with recovery in the 3rd episode IV Moderate level of interactive participation in the first episode with disturbance during the Still-Face episode and absence of subsequent recovery Negative OtherDirected Coping III Low quality interactive participation and progressive increase of discomfort (or negative manifestations) over the 3 Still-Face episodes II Low quality interactive participation or obvious signs of high disturbance in the first and last episode with distinct decrease in tension level during the StillFace episode Self-Directed Coping I Low interactive participation Table 7.2 - Match between Autoregulation Patterns and Behavioral Styles 45 7.3. THE APGAR SCORE Sign 0 1 2 1st minute 5th minute Color Blue or Pale Acrocyanotic Completely Pink Heart Rate Absent < 100 per minute > 100 per minute Reflex Irritability No Response Grimace Cry or Active Withdrawal Muscle Tone Limp Some Flexion Active Motion Respiration Absent Weak Cry; Hypoventilation Good, Crying Source: The American College of Obstetricians and Gynecologists Table 7.3 - Apgar Score Form 46 7.4. DESCRIPTIVE ANALYSIS 7.4.1. Infants’ Demographics Variable Preterm Term Total % Total Gender Female 22 38 60 43,48% Male 32 46 78 56,52% Age Group [ 31 -34 [ 20 0 20 14,49% [ 34 - 37 [ 34 0 34 24,64% [ 37 - 40 [ 0 35 35 25,36% [ 40 - 43 ] 0 49 49 35,51% Weight Group [ 1050 - 1457 [ 7 0 7 5,07% [ 1457 - 1864 [ 11 0 11 7,97% [ 1864 - 2271 [ 15 0 15 10,87% [ 2271 - 2678 [ 15 3 18 13,04% [ 2678 - 3085 [ 6 21 27 19,57% [ 3085 - 3492 [ 0 31 31 22,46% [ 3492 - 3899 [ 0 20 20 14,49% [ 3899 - 4350 ] 0 9 9 6,52% Birth Type Induced 37 29 66 47,83% Spontaneous 17 55 72 52,17% Table 7.4 - Infants' Demographics: Categorical Variables No. Siblings Preterm Term Mean 0,54 0,76 Std Err Mean 0,12 0,13 Std Dev 0,91 1,16 (Min. - Max.) ( 0 - 5) ( 0 - 6) Mode 0 0 N 54 84 Table 7.5 - Infants' Demographics: Quantitative Variables 53 Level + Level - Difference Between Rank Sums [ 18 - 22 [ [ 22 - 26 [ -330,5 [ 18 - 22 [ [ 26 - 30 [ -1650,5 [ 18 - 22 [ [ 30 - 34 [ -3074,5 [ 18 - 22 [ [ 34 - 38 [ -1311,5 [ 18 - 22 [ [ 38 - 42 ] -602,0 [ 22 - 26 [ [ 26 - 30 [ -1320,0 [ 22 - 26 [ [ 30 - 34 [ -2744,0 [ 22 - 26 [ [ 34 - 38 [ -981,0 [ 22 - 26 [ [ 38 - 42 ] -271,5 [ 26 - 30 [ [ 30 - 34 [ -1424,0 [ 26 - 30 [ [ 34 - 38 [ 339,0 [ 26 - 30 [ [ 38 - 42 ] 1048,5 [ 30 - 34 [ [ 34 - 38 [ 1763,0 [ 30 - 34 [ [ 38 - 42 ] 2472,5 [ 34 - 38 [ [ 38 - 42 ] 709,5 Table 7.16 - Nemenyi Test for Mother’s Age Groups and Apgar-1 (Critical Value = 503,0) Level + Level - Difference Between Rank Sums [ 18 - 22 [ [ 22 - 26 [ -305 [ 18 - 22 [ [ 26 - 30 [ -1167,5 [ 18 - 22 [ [ 30 - 34 [ -2236 [ 18 - 22 [ [ 34 - 38 [ -988,5 [ 18 - 22 [ [ 38 - 42 ] -581 [ 22 - 26 [ [ 26 - 30 [ -862,5 [ 22 - 26 [ [ 30 - 34 [ -1931 [ 22 - 26 [ [ 34 - 38 [ -683,5 [ 22 - 26 [ [ 38 - 42 ] -276 [ 26 - 30 [ [ 30 - 34 [ -1068,5 [ 26 - 30 [ [ 34 - 38 [ 179 [ 26 - 30 [ [ 38 - 42 ] 586,5 [ 30 - 34 [ [ 34 - 38 [ 1247,5 [ 30 - 34 [ [ 38 - 42 ] 1655 [ 34 - 38 [ [ 38 - 42 ] 407,5 Table 7.17 - Nemenyi Test for Mother’s Age Groups and Apgar-5 (Critical Value = 305,4)