Methodology

Methods
Explanatory variable definitions
Outcome variable definitions
Data analysis

Methods

Participants

The children in this study participated in the National Longitudinal Survey of Children and Youth (NLSCY). The NLSCY is a long-term study of Canadian children that collects information on a wide range of topics including the social, emotional, behavioural, physical and cognitive development of children from birth to early adulthood.

At each collection cycle of the NLSCY, a new cohort of 0 and 1-year-olds is added and followed every two years until the ages of 4 and 5. For this study, children who were 0 and 1 years old in Cycle 3 (1998), Cycle 4 (2000), Cycle 5 (2002) and Cycle 6 (2004) were selected and pooled together. From this pooled sample, only those children who were first-born and whose biological mother was the person interviewed at each cycle were retained for this study. This resulted in a final sample of 3,382 children who represented about 556,000 children in the Canadian population. For more information about the NLSCY and the sample selection process for this study, see the section About the National Longitudinal Survey of Children and Youth.

Explanatory variables

In this study of children's developmental outcomes, it was important to take into account the influence of various socio-demographic, behavioural and developmentally-related characteristics of the children and their mothers. A full description of these explanatory variables is provided in the subsection Explanatory variable definitions, with a summary provided below. Note that all of these variables were based on data reported by the mother.

Maternal age group

For the present discussion, late childbearing is defined as giving birth to a first child at or after age 35. For all analyses, the reference group is women who first gave birth between the ages of 25 and 29 years.

However, to facilitate comparison with data from other studies, results are presented for other maternal age groups. The following terms are used in this report to define the groups of women and children:

• Teenaged mother = Woman who was under age 20 when she had her first child
• Young mother = Woman who was aged 20 to 24 when she had her first child
• Reference mother = Woman who was aged 25 to 29 when she had her first child
• Middle mother = Woman who was aged 30 to 34 when she had her first child
• Older mother = Woman who was aged 35 or older when she had her first child

Socio-demographic characteristics

Three socio-demographic characteristics were examined in this study: the mother's educational attainment (high school diploma or less versus more education), the level of household income (low-income versus not low-income), and the number of parents in the household (single-parent versus dual-parent household). All characteristics were assessed at the first interview when the child was aged 0 to 1.

Perinatal and postnatal measures

Variables related to the birth of the child included method of delivery, gestational age, and birth weight. These variables were assessed for all children at the first interview and were dichotomized to identify those who were born by caesarean, who were born preterm (258 days gestation or less), or who were born at low birth weight (2,499 grams or less).

Mothers were asked about duration of breastfeeding, and children were grouped into three categories: children not breastfed at all, children breastfed for 6 months or less, and children breastfed for more than 6 months.

Maternal health

The two indicators of maternal health included in this study were asthma and depressive symptoms. At first interview, women were asked whether they had been diagnosed with asthma by a health professional. Depressive symptoms were based on women's reports of the occurrence and severity of symptoms in the previous week, and were assessed at the same time as the selected outcome.

Parenting practices and family functioning

Mothers were asked about parenting practices and family functioning. This study examined positive parenting practices, ineffective parenting practices and a measure of overall family functioning. Each of these was measured at the same time as the selected outcome.

Daily reading to the child

Based on mother's reports, an indicator of whether the child was read to daily at ages 4 to 5 was included in certain analyses.

Outcome variables

A series of variables in the NLSCY were selected to characterize children's physical, behavioural, and cognitive development across early childhood. A short description of each is given below. For a full description of these outcome variables and their associated items, see the subsection Outcome variable definitions.

Children's physical health and development

Physical health and development included whether the child received specialized care following his/her birth, the child's general health status (excellent/very good or good/fair/poor) and whether the child had been diagnosed with asthma. Receipt of specialized care at birth was assessed at the first interview, while children's general health and asthma diagnosis were assessed at all ages.

The attainment of several developmental milestones was examined, along with the child's motor and social skill development at ages 0 to 1 and 2 to 3. All of these outcomes were based on maternal reports.

Problem and positive behaviours

Indicators of physical aggression, emotional disorder and anxiety, hyperactivity and inattention, and positive behaviour were examined in this study. All were measured at ages 2 to 3 and 4 to 5, with the exception of positive behaviour which was measured at ages 4 to 5 only. All behavioural outcomes were based on maternal reports.

Cognitive development

Three direct assessments of cognitive ability were administered to children ages 4 to 5. The Peabody Picture Vocabulary Test - Revised (PPVT-R) was used to assess receptive vocabulary skills (Dunn & Dunn, 1981). Children's number knowledge was evaluated using the Number Knowledge Assessment instrument. The 'Who Am I?' instrument was used to assess copying and symbol use (DeLemos, 2002).

Explanatory variable definitions

i) Maternal age group

Although discussions of late childbearing trends are becoming more common in the literature, its definition varies greatly. The age range used to differentiate "late" childbearing from "typical" may depend upon historical patterns of childbearing in a particular country or region, or may be an arbitrary decision based on availability of information. Three general definitions of late childbearing can be found in the literature: first birth at or after (i) age 30 (Heck et al., 1997), (ii) age 35 (Astolfi & Zonta, 2002; Garrison, Blalock, Zarski, & Merritt, 1997; Prysak et al., 1995), or (iii) age 40 (Ziadeh & Yahaya, 2001). Because the average age of women at childbirth in Canada is near 30, defining late childbearing as first birth at or after age 30 was deemed too early for the Canadian context. For the purposes of the present study, "late" or older maternal age was defined as a first birth occurring at or after age 35.

Women interviewed by the NLSCY were asked how old they were when they gave birth to their first child. This variable was used to classify women into five maternal age group categories: (i) women who were under age 20 when they had their first child (teenaged mothers); (ii) women who were ages 20 through 24 when they had their first child (young mothers); (iii) women who were ages 25 through 29 when they had their first child (reference mothers); (iv) women who were ages 30 through 34 when they had their first child (middle mothers); and (v) women who were age 35 and up when they had their first child (older mothers).

ii) Sociodemographic characteristics

Low household income (ages 0 to 1)

The household income variable used in this report is based on the ratio of self-reported pre-tax household income to the low-income cut-off (LICO) level as reported by Statistics Canada for the size and location of the child's household. LICOs are derived based on expenditure-to-income patterns, and are updated annually. LICO thresholds differ depending on family size and the region of residence (size of urban or rural centre) in which the family is located.

Using information available in the NLSCY, household income was dichotomized in analyses:

• (0) not low-income (ratio of income to LICO was equal to or greater than 1).
• (1) low-income (ratio of income to LICO was less than 1)

Low maternal educational attainment (ages 0 to 1)

Women reported the highest level of education that they had achieved. Responses were categorized in the following manner: less than high school education; high school degree or diploma; some post-secondary education; and college or university graduate (incl. trade schools).

Mothers' highest level of education was dichotomized in analyses:

• (0) some post-secondary, college or university degree (including trade)
• (1) less than secondary, secondary school graduation

Single-parent household (ages 0 to 1)

Single-parent households were those headed by the biological mother. Dual-parent households included the biological mother and the biological, step, foster or adoptive father. This variable was dichotomized in analyses:

• (0) dual-parent household
• (1) single-parent household

iii) Perinatal and postnatal measures

Each of these questions were asked of the mother during her first interview with the NLSCY, when the child was aged 0 to 1.

Method of delivery

Based on responses that women provided at their first interview, the method of birth delivery was dichotomized as:

• (0) vaginal
• (1) caesarean section

Gestational age

A series of questions were used to determine the child's gestational age in days.

1. "Was this child born before, after or on the due date?"
2. If the child was not born on the due date, the mother was asked, "How many days before or after the due date was this child born?"

Based on these questions, gestational age was dichotomized as:

• (0) term birth (259 days or more)
• (1) preterm birth (258 days or less)

Birth weight

Based on maternal report of the child's birth weight in kilograms and grams, birth weight was dichotomized as:

• (0) not low birth weight (2,500 grams at birth or more)
• (1) low birth weight (2,499 grams or less at birth)

Duration of breastfeeding

At first interview, women were asked whether they had ever breastfed their child. For those who answered affirmatively, they were asked for how long the child was breastfed. Based on responses to these two questions, children were categorized into three groups: (1) those who were never breastfed, (2) those who were breastfed for 6 months or less, (3) those who were breastfed for more than 6 months.

Duration of breastfeeding was not assessed for children who were currently breastfed at the time of interview. In such cases, if the child was 7 months or older, their duration of breastfeeding was categorized as "more than 6 months". For children 0 to 6 months of age who were currently being breastfed at their last interview, duration of breastfeeding was categorized as unknown. To maximize sample size in regression models, children whose duration of breastfeeding was unknown were included in the models, but the value of the regression estimate was not presented in tables.

iv) Parenting and family characteristics

Positive parenting practices scale (ages 2 to 3 and 4 to 5)

This scale is based on maternal responses to a set of five individual items. The mother was asked: "The following questions have to do with things that <your child> does and ways that you, as a parent, react to him/her." The questions were:

1. How often do you praise <your child> by saying something like "Good for you!" or "What a nice thing you did!" or "That's good going!"?
2. How often do you and <your child> talk or play with each other, focusing attention on each other for five minutes or more, just for fun?
3. How often do you and <your child> laugh together?
4. How often do you do something special with <your child> that he/she enjoys?
5. How often do you play sports, hobbies or games with <your child>?

Response options were: (0) never, (1) about once a week or less, (2) a few times a week, (3) one or two times a day, and (4) many times each day. Scores were calculated for children ages 2 and over. Possible scores ranged from 0 to 20.

For children aged 2 to 3, the standardized, weighted Cronbach's alpha measure of internal reliability for the 5-item score was 0.65. For children aged 4 to 5, the measure of internal reliability was 0.68.

Ineffective parenting practices scale (ages 2 to 3 and 4 to 5)

This scale is based on maternal responses to a set of seven individual items. The mother was asked: "For the following questions, I would like you, as <your child>'s parent, to tell me how things go when you spend time with him/her." The questions were:

1. How often do you get annoyed with <your child> for saying or doing something he/she is not supposed to?
2. Of all the times that you talk to <your child> about his/her behaviour, what proportion is praise?
3. Of all the times that you talk to <your child> about his/her behaviour, what proportion is disapproval?
4. How often do you get angry when you punish <your child>?
5. How often do you think that the kind of punishment you give <your child> depends on your mood?
6. How often do you feel you are having problems managing <your child> in general?
7. How often do you have to discipline <your child> repeatedly for the same thing?

Response options for the first item where: (0) never, (1) about once a week or less, (2) a few times a week, (3) one or two times a day, and (4) many times each day. Response options for items ii through vii were: (0) never, (1) less than half the time, (2) about half the time, (3) more than half the time, and (4) all the time.¹ Possible scores ranged from 0 to 28. A high score indicates more hostile or ineffective interactions.

For children aged 2 to 3, the standardized, weighted Cronbach's alpha measure of internal reliability for the 7-item score was 0.67. For children aged 4 to 5, the measure of internal reliability was 0.69.

Family functioning scale (ages 2 to 3 and 4 to 5)

This scale is comprised of the 12-item General Functioning subscale of the McMaster Family Assessment Device (Epstein, Baldwin, & Bishop, 1983) and was used to provide a global assessment of family functioning and an indication of the quality of the relationships between parents or partners. The mother was asked: "The following statements are about families and family relationships. For each one, please indicate which response best describes your family." The statements were:

1. Planning family activities is difficult because we misunderstand each other.
2. In times of crisis we can turn to each other for support.
3. We cannot talk to each other about sadness we feel.
4. Individuals, in the family, are accepted for what they are.
5. We avoid discussing our fears or concerns.
6. We express feelings to each other.
7. There are lots of bad feelings in our family.
8. We feel accepted for what we are.
9. Making decisions is a problem for our family.
10. We are able to make decisions about how to solve problems.
11. We don't get along well together.
12. We confide in each other.

Response options were: (0) strongly agree, (1) agree, (2) disagree or (3) strongly disagree. Responses to odd-numbered items were reverse-coded. Possible scores ranged from 0 to 36. A high score indicates greater family dysfunction.

For children aged 2 to 3, the standardized, weighted Cronbach's alpha measure of internal reliability for the 12-item score was 0.91. For children aged 4 to 5, the internal reliability measure was 0.92.

Mother's depressive symptoms scale (ages 2 to 3 and 4 to 5)

The NLSCY contains an adapted 12-item version of the Centre for Epidemiology Studies - Depression scale (CES-D). This scale is used to measure the occurrence and severity of symptoms associated with depression during the previous week. It is based on responses to a set of 12 individual items. The mother was asked: "The next set of statements describes feelings or behaviours. For each one, please tell me how often you felt or behaved this way during the past week." The statements were:

1. I did not feel like eating; my appetite was poor.
2. I felt that I could not shake off the blues even with help from my family or friends.
3. I had trouble keeping my mind on what I was doing.
4. I felt depressed.
5. I felt that everything I did was an effort.
6. I felt hopeful about the future.
7. My sleep was restless.
8. I was happy.
9. I felt lonely.
10. I enjoyed life.
11. I had crying spells.
12. I felt that people disliked me.

Response options were: (0) rarely or none of the time (less than 1 day), (1) some or a little of the time (1 to 2 days), (2) occasionally or a moderate amount of time (3 to 4 days), and (3) most or all of the time (5 to 7 days). Responses to items vi, viii, and x were reverse-coded. Possible scores ranged from 0 to 36. A high score indicates the presence of depressive symptoms.

For children aged 2 to 3, the standardized, weighted Cronbach's alpha measure of internal reliability for the 12-item score was 0.78. For children aged 4 to 5, the internal reliability measure was 0.77.

Child was read to daily (ages 4 to 5)

Child was read to daily was based on the mother's response to the following question:

"How often do you (or your spouse): Read aloud to him/her or listen to him/her read or try to read?" The response options were: (i) rarely or never; (ii) a few times a month; (iii) once a week, (iv) a few times a week; or (v) daily.

In analyses, daily reading was dichotomized as:

• (0) child was read to less often than daily (including rarely or never, a few times a month, once a week, and a few times a week)
• (1) child was read to daily

Outcome variable definitions

i) Physical health and development

All physical health and development outcomes were based on mothers' reports.

Received special care at birth

For children aged 0 to 23 months, or if it was the first interview for the child, mothers were asked, "Did this child receive special medical care following his birth?".

• (0) No
• (1) Yes

For children who were reported as receiving special care at birth, mothers were asked what type of special care was received. Options were (i) intensive care, (ii) ventilation or oxygen, (iii) transfer to a specialized hospital, or (iv) other type of care. Mothers could report that their child received more than one type of specialized care. Mothers also indicated the number of days that their child was in this type of care following birth.

General health (ages 0 to 1, 2 to 3 and 4 to 5)

"In general, would you say this child's health is: (1) Excellent; (2) Very good; (3) Good; (4) Fair; (5) Poor"?

Responses to this question were dichotomized as:

• (0) Good, Fair or Poor
• (1) Excellent or Very good

Asthma (ages 0 to 1, 2 to 3 and 4 to 5)

"Has this child ever had asthma that was diagnosed by a health professional?"

• (0) No
• (1) Yes

Developmental milestones

Mothers were asked to report the age at which their child achieved each of five developmental milestones.

"The next questions ask when your child started to do certain things. If you do not know the exact age, your best estimate is fine." The questions were:

1. At what age (in months) did this child first sit themselves up?
2. At what age (in months) did this child start eating solid food?
3. At what age (in months) did this child start feeding themselves by picking up small pieces of food with their fingers and putting them in their mouth? This includes feeding him/herself a cracker or a cookie.
4. At what age (in months) did this child take their first steps?
5. At what age (in months) did this child start saying their first words? By word I mean a sound or sounds a baby says consistently to mean someone or something such as 'baba' for 'bottle'.

All developmental milestone questions were asked for children aged 9 to 47 months at the time of interview, except for the question regarding first words which was limited to children aged 12 to 47 months at the time of interview.

Children above the 90^th percentile were said to be late in achieving their developmental milestones. Due to the fact that PMKs reported their child's age at achieving milestones in whole months, the cut-points above were rounded to the next highest whole number. The rounded cut-points for each milestone are as follows:

• Late to sit up by self: age ≥ 8 months
• Late to eat solid food: age ≥ 10 months
• Late to feed self: age ≥ 13 months
• Late to say first word: age ≥ 13 months
• Late to take first step: age ≥ 14 months

Motor and Social Development scale

For children aged 0 to 3 years, mothers responded to a series of age-specific questions that were meant to measure children's gross and fine motor skills, perception and cognition, communication and language, and social development. The Motor Social Development (MSD) scale has been used in other national surveys such as the National Longitudinal Survey of Youth in the United States, and the National Child Development Survey in England.

Based on the child's age in months, mothers were asked whether their child had ever accomplished various tasks or behaviours. Fifteen items are used for each age range, as indicated in Table C.1.

The raw MSD score was standardized across age groups to have a mean of 100 with a standard deviation of 15. To facilitate comparison across cycles, the NLSCY produces a standardized score that uses the norms developed in the first cycle (1994/95) of the NLSCY. The standardized score was used in the present study.

ii) Problem and positive behaviours

All behavioural outcomes were based on mothers' reports.

Physical aggression scale

This scale is one of the components of the NLSCY's behaviour checklist. It is based on responses to a set of eight individual items for 2- to 3-year-olds and six items for 4- to 5-year-olds. Asked to indicate whether a specific behaviour was (0) never or not true, (1) sometimes or somewhat true, or (2) often or very true for their child, the child's mother was presented with the following example behaviours:

For 2- to 3-year-olds, the scale ranges from 0 to 16, with higher scores indicating the presence of greater physical aggression and opposition. The standardized, weighted Cronbach's alpha measure of internal reliability for the 8-item score for 2- to 3-year-olds was 0.70.

For 4- to 5-year-olds, the scale ranges from 0 to 12, with higher scores indicating the presence of greater physical aggression and conduct disorder. The standardized, weighted Cronbach's alpha measure of internal reliability for the 6-item score for 4- to 5-year-olds was 0.76.

Hyperactivity and inattention scale

This scale is one of the components of the NLSCY's behaviour checklist. It is based on responses to a set of six individual items for 2- to 3-year-olds, and seven items for 4- to 5-year-olds. Asked to indicate whether a specific behaviour was (0) never or not true, (1) sometimes or somewhat true, or (2) often or very true for their child, the child's mother was presented with the following example behaviours:

For 2- to 3-year-olds, possible scores ranged from 0 to 12, with a higher score indicating the presence of greater hyperactivity and inattention. The standardized, weighted Cronbach's alpha measure of internal reliability for the 6-item score for 2- to 3-year-olds was 0.69.

For 4- to 5-year-olds, possible scores ranged from 0 to 14, with a higher score indicating the presence of greater hyperactivity and inattention. The standardized, weighted Cronbach's alpha measure of internal reliability for the 7-item score for 4- to 5-year-olds was 0.78.

Emotional disorder and anxiety scale

This scale is one of the components of the NLSCY's behaviour checklist. It is based on responses to a set of six individual items for 2- to 3-year-olds and seven items for 4- to 5-year-olds. Asked to indicate whether a specific behaviour was (0) never or not true, (1) sometimes or somewhat true, or (2) often or very true for their child, the child's mother was presented with the following example behaviours:

For 2- to 3-year-olds, possible scores ranged from 0 to 12, with a higher score indicating the presence of greater emotional disorder and anxiety. The standardized, weighted Cronbach's alpha measure of internal reliability for the 6-item score for 2- to 3-year-olds was 0.64.

For 4- to 5-year-olds, possible scores ranged from 0 to 14, with a higher score indicating the presence of greater emotional disorder and anxiety. The standardized, weighted Cronbach's alpha measure of internal reliability for the 7-item score for 4- to 5-year-olds was 0.67.

Positive behaviour scale (ages 4 to 5)

This scale is made up of six items from the Positive Behaviour checklist to assess positive behaviour of children aged 3 to 5, including perseverance and independence. The mother was asked: "For the next set of questions, please think about <your child>'s behaviour over the past month or two. How often does <your child>:

1. Keep his/her temper?
2. Listen well and pay attention?
3. Show self-control?
4. Finish things he/she starts?
5. Persist with solving a problem, even when things go wrong for a while?
6. Make an effort to do something, even if he/she doesn't feel confident about it?"

Response options were: (0) never, (1) sometimes or (2) often. Possible scores ranged from 0 to 12. A high score indicates greater perseverance and independence.

The standardized, weighted Cronbach's alpha measure of internal reliability for the 6-item score was 0.69.

iii) Cognitive development

All measures of cognitive development are direct assessments of the child.

Receptive vocabulary score

The measure of receptive vocabulary was the standard score on the Peabody Picture Vocabulary Test - Revised (PPVT-R). Scores had a mean value of 100 and a standard deviation of 15. More information about the PPVT-R in the NLSCY may be found in the Microdata User Guide for the NLSCY for cycle 6, pages 141 to 143 (Statistics Canada).

Number knowledge score

The measure of number knowledge was the age-standardized score on the 22-question (30-item) Number Knowledge Assessment instrument. Scores had a mean value of 100 and a standard deviation of 15. More information about the Number Knowledge Assessment instrument in the NLSCY may be found in the Microdata User Guide for the NLSCY for cycle 6, pages 143 to 145 (Statistics Canada).

Copying and symbol use score

The measure of copying and symbol use was the standard score on Who Am I?, an instrument that evaluates the developmental level of young children (DeLemos, 2002). Scores had a mean value of 100 and a standard deviation of 15. More information about the copying and symbol use instrument in the NLSCY may be found in the Microdata User Guide for the NLSCY for cycle 6, pages 145 to 147 (Statistics Canada).

Data analysis

For each outcome measure, means or proportions were generated overall and by maternal age group. In all analyses, children whose mothers were aged 25 to 29 at the child's birth formed the reference category.

Multivariate regression models were generated for each outcome. Included in each model was the mother's age at the child's birth, a group of socio-demographic characteristics, and selected maternal characteristics or behaviours thought to be associated with the outcome of interest. Because respondents were pooled across several cycles of the NLSCY, a control variable indicating cycle of entry to the NLSCY was also included in each model (although values for this estimate are not given herein).

Logistic models were fit to dichotomous outcomes, while linear models were fit to continuous outcomes. All analyses were conducted using SAS-callable SUDAAN. To account for the complex survey design of the NLSCY, an appropriate survey weight was used for all estimates, and bootstrap weights were used to produce variance estimates.

In the NLSCY, the child is the unit of analysis. Therefore, throughout this paper, any information relating to the mother or the household should be interpreted as a characteristic of the child. For example, when referring to mothers who suffered from gestational diabetes, the estimate reflects the percentage of children whose mothers suffered from the condition, not the percentage of mothers with the condition. While this distinction may not always be made explicitly in the text, the reader should be aware that it is always implied.

---

Note

1. Response categories were reverse coded for item 2.