Infants Born Prematurely and Learning and Behavioral Problems

Rationale

Studies from the literature review suggest that an increasing number of infants who were born prematurely would have learning and behavioral problems when they reach school age. This may be due to various reasons such as parental education, family income, and marital status.

Objective

To determine whether children who were prematurely born have a poorer motor and mental ability measured by motor and a mental score of Development Quotients (DQ) compared to those children who were born full-term.

Research Design Case-Control Study Design

On consent, mothers of both preterm and full-term infants completed a self-administered questionnaire to answer the questions in demographic information including education, family income, and marital status. Development Quotients of Infants were assessed based on: Motor Score and Mental Scores.

Sample

Eligible subjects were infants and their mothers. Infants who were prematurely born at Mater mothers hospital in Brisbane were recruited into the study (N =74). Inclusion required that preterm infants were aged 8 months; to be classified as having a birth weight less than 1200 grams and gestational age less than 32 weeks when they were born. Full-term born infants (N=74) who had a birth weight of 2500 -4000 grams, gestational age 36-42 weeks, and who were born in the same hospital were randomly selected. Preterm and full-term infants were matched in variables: age and gender.

Data Collected

Infants were preterm or not: (1 = preterm infant group; 2 = full-term infant group) DQ was measured by a motor and mental ability, with higher scores meaning higher DQ: Physical Score (70-130), Mental Score (70-130). Demographic variables were measured: Mothers education, Family income, Mothers marital status and Mothers depression level.

Research Questions:

Is the premature born infants group different from the full-term infants group in the mothers education, income, marital status and depression level?
Are there any differences between premature-born children and full-term infants in motor and mental scores?

Is premature born infants group different from the full-term infants group in mothers education, income, marital status and maternal depression level?

H₀: There is no difference between the premature born infants group and full-term infants group in terms of mothers education, income, marital status and depression level.
H₁: There is a difference between premature born infants group and full-term infants group in terms of mothers education, income, marital status and depression level.

A Chi-Square analysis was conducted to test the hypotheses since a Chi-Square test is used to test the independence of two variables. In this case, the two groups of infants (preterm and full-term infants) were assessed in terms of differences in mothers marital status, education, income and depression levels. Most important is that Pearson Chi-Square is very useful in determining the existence of relationships between two categorical variables. It is notable that since it is impossible to compare means for categorical variables (like marital status and depression level are categorical variables in our case); Chi-Square test is the best option since it compares frequencies in the various categories of data (Polit, 2010). The Chi-Square test also produces the Likelihood Ratio output which is also useful for testing relationship between categorical variables. In addition, Chi-Square requires that the expected number for each cell to be at least five and this was met by the data used in this case. Since Chi-Square provides cross-tabulations of each test variable against the infant groups, the test was the most appropriate for identifying any differences in the groups. The Chi-Square test was conducted at a 2-tailed significance level of p<.05.

After conducting a Chi-Square analysis, there were several findings regarding differences in the preterm infants and full-term infant groups in terms of mothers marital status, annual income, education level and depression level. The observed frequency for married or defacto mothers was 60 whereas the expected frequency was 58.0 in the preterm group. The observed frequency of single mothers in the preterm group was 14 whereas the expected frequency was 16.0. In the full-term group, the observed frequency of married/defacto mothers was 56 and the expected frequency was 58.0. On the other hand, the observed frequency in single mothers belonging to the full-term group was 18 and the expected frequency was 16.0. It was identified that married mothers in the preterm infant group constituted 51.7% compared to 48.3% in the full-term infant group. Single mothers in the preterm infants group made up 43.8% compared to 56.2% in the full-term infant group. The Chi-Square value was (.638, 1 df) and the 2-sided significance value was.424 (greater than.05). This implies that there was no significant difference in the marital status among mothers of preterm infants compared to mothers of full-term infants. The Likelihood ratio and the Fishers Exact Test value were also non-significant, p>.05.

After conducting a 2*2 cross-tabulation of infants groups and mothers education level, observations were made. The observed count of preterm infant mothers who had a lower than 10^th grade level of education was 10 and an expected count of 8.0 compared to an observed count of 6 and an expected count of 8.0 in full-term infant mothers of the same education level. As such, preterm infant mothers with lower than 10^th grade education level constitute 62.5% compared to 37.5% in full-term infant mothers. The observed count of mothers with an education level of 10-12^th grade was 38 (60.3%) against 25 (39.7%) in the full-term infants group. For preterm infant mothers with an education level of Tafe and Diploma constituted an observed count of 14 (50%) against an observed count of 14 (50%) in the full-term infant mothers group. There was an observed frequency of 12 (29.3%) mothers in the pre-term infants group with a tertiary level of education against an observed frequency of 29 (70.7%) mothers in the full-term infants group. The Chi-Square value for this test was 10.731, three degrees of freedom and 2-sided p value of.013. The Likelihood ratio was 10.98, 3 df and 2-sided p value of.012 (less than.05). This indicates that there was a significant difference in education level among mothers of preterm infants and full-term infants. In essence, full-term infants group mothers had attained higher levels of education compared to preterm infants group mothers.

A cross tabulation of infants groups and mothers annual family income showed the following observations: An observed count of 35 (81.4%) of mothers in the pre-term infants group had an annual income less than 30,000 compared to an observed count of 8 (18.6%) in the full-term infants group. For pre-term infants group mothers, there was an observed frequency of 30 (55.6%) of mothers who had an annual income of 30,000-49,999 compared to an observed frequency of 24 (44.4%) of mothers in the full-term infants group. There was an observed frequency of 9 (17.6%) preterm infants group mothers who had an annual income of 50,000 to 79,999 compared to 42 (82.4%) of full-term infants group mothers in the same income level. Table 6 indicates the Chi-Square value for this analysis as 38.97, 2 degrees of freedom and a 2-sided significance value of.001. The Likelihood ratio was 42.13, 2 df and 2-sided significance value of.001. This indicates that there existed a significant difference in income levels among preterm infants group mothers and full-term infants group mothers. Precisely, most preterm infants mothers had an annual income of less than 49,000 whereas most full-term infants mothers had an annual income greater than 49,000.

After conducting a cross tabulation of infant groups and mothers depression levels, several observations were made. It was evident that 53 (86.9%) of preterm infant group mothers were depressed compared to 8 (13.1%) of full-term infant group mothers. In addition, there was an observed frequency of 21 (24.1%) of preterm infant group mothers who had no depression compared to 66 (75.9%) of full-term infant group mothers who were not depressed. From Table 7, it is evident that the Chi-Square value for this cross tabulation was 56.47, 1 df and a 2-sided significance value of.001. There was a significant difference in mothers depression level between preterm infant and full-term infant mothers (Fishers Exact Test, p<.05). This is also reflected by the significant Likelihood ratio value of 61.60, p<.05. To be specific, there were more depressed mothers in the preterm infants group compared to full-term infants group.

In summary, the differences in marital status of preterm infants group and full-term infants group mothers were not significant (there were almost equal married and single preterm infant mothers as there were in the full term infants group). There were more preterm infants mothers with an education level below 10^th grade compared to full-term mothers. Most of the preterm infant mothers had an education level of 10-12^th grade whereas there were equal numbers of preterm and full-term infant mothers who had a Tafe and Diploma. In addition, most full-term infant group mothers had a tertiary education with only a few preterm infant group mothers having a tertiary level education. In terms of annual income, most preterm infant group mothers had an annual income of not more than 49,000 whereas most full-term infant mothers had an annul income of at least 50,000. Finally, preterm infant group mothers were different from full-term infant group mothers in terms of depression levels with most preterm infant group mothers being depressed and a majority of full-term infant mothers having no depression. Conclusively, the alternate hypothesis, there is a difference between premature born infants group and full-term infants group in terms of mothers education, income, marital status and depression level, was accepted.

Differences between premature born children and full-term infants in motor and mental scores, if there are any

H₀: Premature born children and full-term infants do not differ significantly in their Motor and Mental Scores
H₁: Premature born children and full-term infants differ significantly in their Motor and Mental Scores.

An independent t-test was the most suitable analysis for testing the above hypothesis since the samples under test were from different populations. In addition, independent t-test was suitable for determining differences in means thus the differences in mean scores for motor and mental scores in premature and full-term infants would be easy to compute. The test was conducted at 95% confidence interval and a significance level of p<.05. Most important is that independent t-test is an indispensable parametric test used in testing null hypotheses by comparing means (Field, 2009) and it was assumed that there was normal distribution of data.

The independent t-test descriptive statistics indicated that the mean Motor Score for Development Quotient in preterm infants was 93.80 with a standard deviation of 10.86, whereas the mean Motor Score for full-term infants was 1.00E2 with a standard deviation of 10.02. The mean Mental IQ Score of Development Quotient for preterm infants was 93.47 with a standard deviation of 9.39 and a mean of 1.12E2 with a standard deviation of 6.49 for full-term infants.

Taking the equal variances assumed values for Motor Score; the Levenes test for equality of variances was not significant. As such, the variances can be assumed to be equal (F=.002, p>.05). This implies that variances for Motor Scores in preterm and full-term infants were equal (significance =.96). From the equal variances assumed row for Motor Score, it is evident that the t value for Motor Scores was -3.62 with 146 degrees of freedom and a 2-tailed significance level of.001. The independent t-test conducted to test the hypothesis that premature born children and full-term infants differed significantly in their Motor and Mental Scores show that the mean Motor Score for preterm infants (93.80, SD = 10.86) is statistically significantly different (t = -3.62, df = 146, 2-tailed p =.001) from that of full-term infants (M = 1.00E2, SD = 10.02) (Table 8).

The Levenes test for equality of variances for Mental IQ score was significant (F =9.05, p<.05), indicating that the variances for Mental Scores for preterm and full-term infants were unequal. Due to the unequal variances between the two groups, a test for unequal variances was used. The equal variances not assumed indicated the t value for Mental Scores as -13.66 with 129.75 degrees of freedom and a 2-tailed significance value of.001. The independent t-test conducted to test the hypothesis that premature born children and full-term infants differ significantly in their Motor and Mental Scores showed that the mean Mental Score for preterm infants (93.47, SD = 9.39) was statistically significantly different (t = -13.66, df = 129.75, 2-tailed p =.001) from that of full-term infants (M = 1.12E2, SD = 6.49) (Table 8).

In summary, the alternate hypothesis- premature born children and full-term infants differ significantly in their Motor and Mental Scores- was accepted. Since it was also identified that there were differences in mothers education level, income and depression levels in preterm and full-term infant groups, it is possible to attribute differences in Mental and Motor Scores to these differences. This would be in line with studies from literature review which suggest that learning and behavioural problems in premature born children may be due to various reasons such as parental education, family income, and depression level.

References

Field, A. (2009). Discovering statistics using SPSS (3rd Ed.). Thousand Oaks, CA: Sage Publications.

Polit, D. F. (2010). Statistics and data analysis for nursing research (Second Edition). Upper Saddle River, NJ: Pearson Education.

Appendix

Table 6: Chi-Square Test for Infant Groups and Annual Family Income

Chi-Square Tests
	Value	df	Asymp. Sig. (2-sided)
Pearson Chi-Square	38.973^a	2	.000
Likelihood Ratio	42.130	2	.000
Linear-by-Linear Association	38.215	1	.000
N of Valid Cases	148
a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 21.50.

Table 7: Chi-Square Tests for Infant Groups and Mothers Depression Level

Chi-Square Tests
	Value	df	Asymp. Sig. (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Pearson Chi-Square	56.473^a	1	.000
Continuity Correction^b	53.991	1	.000
Likelihood Ratio	61.603	1	.000
Fishers Exact Test				.000	.000
Linear-by-Linear Association	56.091	1	.000
N of Valid Cases^b	148
a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 30.50.
b. Computed only for a 2×2 table

Table 8: Independent Samples Test for Infant Groups and Motor and Mental Scores

Independent Samples Test
		Levenes Test for Equality of Variances		t-test for Equality of Means
		F	Sig.	t	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	95% Confidence Interval of the Difference
		F	Sig.	t	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	Lower	Upper
Motor score of Development Quotient	Equal variances assumed	.002	.962	-3.619	146	.000	-6.21622	1.71761	-9.61081	-2.82162
Motor score of Development Quotient	Equal variances not assumed			-3.619	145.077	.000	-6.21622	1.71761	-9.61099	-2.82144
Mental IQ score of Development Quotient	Equal variances assumed	9.053	.003	-13.655	146	.000	-18.12162	1.32712	-20.74447	-15.49877
Mental IQ score of Development Quotient	Equal variances not assumed			-13.655	129.754	.000	-18.12162	1.32712	-20.74722	-15.49602