OBJECTIVES. Parenting is recognized as a key mediator in both health and educational outcomes. Much is known on the value of support and group work in benefiting parenting, but little is known on the effect of written information. A randomized, controlled trial was conducted to evaluate the effect of a parenting newsletter, sent monthly to the parents' home from birth to 1 year, on maternal well-being and parenting style. We tested the hypothesis that mothers receiving the newsletter would show less stress and better parenting characteristics than controls.
METHODS. Parents of first infants born in a North East England District General Hospital between February and October 2003 who consented to take part in the study were randomly allocated to either the intervention or control arm. Those in the intervention arm were sent 12 monthly issues of an age-paced parenting newsletter containing information on emotional development, parent-child interaction, and play. Both the intervention and control group received normal parenting support. Mothers in both groups completed the Well-being Index, Parenting Daily Hassles Scale, and the Adult-Adolescent Parenting Inventory at birth and at 1 year.
RESULTS. One hundred eighty-five mothers were recruited, with 94 randomly assigned to the intervention group, and 91 controls. Allowing for differences at recruitment, there were significant differences between the groups at 1 year: the intervention mothers had lower frequency and intensity of perceived hassles and fewer inappropriate expectations of the infant on the Adult-Adolescent Parenting Inventory than the control mothers.
CONCLUSIONS. A monthly parenting newsletter sent directly to the home in the first year of life seems to help parents to understand their infant better and feel less hassled. This intervention is low cost and can be applied to all parents, so it is nonstigmatizing.
Parenting is widely recognized to be a key factor in early child development and influences physical health and growth, mental and social well-being, and educational achievement.1,2 Good parenting greatly benefits later child development, whereas inconsistent, critical, or neglectful parenting may lead to difficult behavior, emotional problems, conduct disorders, and school failure.1 Hence, there has been much recent emphasis on parenting education in government policy in the United Kingdom.3 There is now considerable evidence available on which methods work in parenting support.4 Some of the ingredients of successful programs, from a review of the international evidence, 4 include the following:
Interventions with a strong theory-base and clearly articulated model of the predicted mechanism of change
Interventions that have measurable, concrete objectives as well as overarching aims
Universal interventions for problems at the less severe end of the spectrum
Behavioral interventions that focus on specific parenting skills and practical take-home tips
Cognitive interventions for changing beliefs, attitudes and self-perceptions about parenting
Group work, where the issues involved are suitable to be addressed in a ‘public’ format
Individual work, where problems are severe or entrenched or parents are not ready/able to work in a group.
Since 1999, the government in the United Kingdom has invested major funding in an early intervention program for disadvantaged communities (Sure Start), which is based on interagency support for parents of children younger than 3 years old, using a community participation model. However, because each Sure Start program was developed with local autonomy and allocation of funding to communities was not randomized, it has been difficult to identify which aspects of support might have a positive impact on parenting skills.5
Little research has been conducted on educational measures to support parenting. The only purely educational intervention (ie, the provision of information to increase knowledge and understanding) covered in the United Kingdom Policy Research Bureau review4 looked at home visitation by health visitors implementing individualized parenting education, including an information package and educational leaflets, at 5 home visits between 1 and 18 months.6 A randomized, controlled trial found that the program did not significantly reduce potentially problematic behavior in children at either 9 months or 2 years.
One innovative form of informational support is an age-paced newsletter, focusing on parent-child relationships and infants' cognitive development. Such a newsletter, posted to the family home, has been in use in Wisconsin for over 20 years.7 Each issue covers topics that are important in child development at a specific stage, written in popular newspaper style. Newsletters are distributed monthly in the first year and then every second month to parents of a first infant who are thought to have the greatest need for information. Evaluation showed that the newsletter was more useful to parents than any other source of information,7 and that parents receiving it showed more positive attitudes against corporal punishment than those not receiving it (D. Riley, PhD, unpublished data, 1997). However this was not a randomized study. Our pilot-controlled evaluation showed that the newsletter is extremely popular with both mothers and fathers and is perceived as more useful a source of information than relatives or friends, general physicians or health visitors, group support, or books and magazines.8,9
We conducted a randomized trial on the effects of receiving Baby Express in a cohort of first time mothers in a socioeconomically disadvantaged area (based on United Kingdom indicators including relatively low income, high unemployment, and low housing quality). The hypothesis tested was that receiving a parenting newsletter over the first year of the infant's life would significantly reduce parent stress, improve well-being, and positively alter self-reported style of parenting in comparison with controls who did not receive the newsletter.
Recruitment of first-time mothers was conducted by midwives antenatally and by the project researchers (Ms Welsh and Ms Keane) on the postnatal ward at a North East England District Hospital from February to October 2003. Parents who agreed to participate were randomly assigned to be in the intervention group (receiving copies of Baby Express newsletter) or to be a control (normal services only, no newsletter). Participants were offered a shopping voucher as a sign of appreciation.
The randomization was conducted by a separate researcher by using a computerized random numbers table. The allocation to intervention or control was placed in a sealed envelope handed to the parent by the researcher after she had signed consent to take part. When the interviews were conducted, the researcher was not blind to group. However, a standard proforma was used for the interviews, with rating scales for the questionnaires, to avoid introducing bias. The researcher usually did not interview parents at both time points and was not aware of baseline scores.
The research was given a favorable opinion by the Northern and Yorkshire Multi-center Research Ethics committee.
Baby Express (Fig 1) is an 8-page glossy newspaper written by a local health journalist using the model of “Parenting the first year” developed by Riley7 in Madison, Wisconsin. The original content was adapted and updated for an audience in the United Kingdom, and illustrations added together with a guide to local services. The reading age level equates to a United Kingdom tabloid newspaper, around 10 years. Baby Express was sent monthly to the home in the first year and 2 monthly for the following 4 years. The researcher gave out the first edition of the newsletter at the first visit to the home.
Information was obtained by questionnaires administered in the home by the researchers shortly after the child's birth and at 1 year.
Demographic data included infant age and gender, maternal age, previous employment, and ethnic group. Level of deprivation is measured by the Townsend Index.10 Individual household postcodes were matched to standardized norms for the North of England using the 2001 census (indices based on unemployment, overcrowding, lack of ownership of a car, not owning their house). The range of scores in the sample was −4.69 (affluent) to 7.89 (very deprived). Level of education was coded as basic school leaving qualifications (up to O level, General Certificate of Secondary Education [GCSE]), advanced school leaving qualifications (A level, Higher National Diploma, etc) or a university degree. The US equivalents of these levels would be a high school diploma, some college attendance, and a college degree after 4 years.
The Well-being Index (General Health Questionnaire [GHQ12]11) has been extensively used in research and has high sensitivity for detecting postpartum depression.12 The 12 items ask about feelings over the past few weeks (eg, “been able to concentrate on whatever you are doing?”), on a 4-point scale presented as “better than usual” to “much less than usual.” The total score range is 0-36 and high scores indicate a lower sense of well-being.
Parenting Daily Hassles Scale13 (Infant Version)
The 20 items describe events that routinely occur and can be a “hassle” to parents (eg, “continually cleaning messes of food,” “my child is fussy and/or cries all the time”). Each item is scored in 2 ways: frequency on a 4-point scale from rarely to constantly, total score range 20 to 80; and intensity of hassle on a 5-point scale from “no hassle” to “big hassle”; total score range: 20–100. The scales have good internal consistency (Crnic and Greenberg report α = .81 and .90, respectively). There are no clinical cutoff points.14 A high score on either frequency or intensity indicates a high level of “hassle” that the mother experiences with the infant. We hypothesized that intensity was the more likely to change with intervention.
Adult-Adolescent Parenting Inventory
The Adult-Adolescent Parenting Inventory (AAPI) is a measure of parenting attitudes that was developed in the United States as a predictor of child abuse.15 The parent indicates their agreement with 40 statements, on a 5-point scale. Parenting style is summarized as 5 constructs: (1) inappropriate expectations of children (score range: 7–35), (2) parental lack of empathy toward children's needs (score range: 10–50), (3) strong belief in the use of corporal punishment as a means of discipline (score range: 11–55), (4) reversing parent-child role responsibilities (score range: 7–35), and (5) oppressing children's power and independence (score range: 5–25). The scales have good internal consistency (α >.80).15 Low scores on any construct would be concerning and indicative of potential abuse. The scale has been used in measurement of outcome of parenting interventions, for example, for adolescent mothers.16
Sample size17 was calculated from pilot data for the Parenting Daily Hassles Scale.8 Comparison between intervention and control groups at 6 months using the t test gave an effect size of 0.50 in terms of lower intensity of daily hassles for the former. For 80% power at P < .05, 64 subjects were required in each arm of the study.
Differences on independent and dependent measures between those who remained in the study at year 1 and those who were lost to follow-up were calculated by using χ2 or t tests. Similarly, whether randomization had created equivalent groups was calculated by using χ2 or t tests.
The change between baseline and year 1 scores was calculated for each individual, for each questionnaire summary score, and the data checked for normal distribution. Linear regression was used to compare these change scores between the intervention group and control group, with adjustment for highest qualification achieved and employment as recorded at baseline (described later). The goodness of fit of the regression models was assessed by using residual analysis.
Mean scores (SD) are presented for intervention and control groups separately, at baseline and year 1. The mean difference (95% confidence interval) in change scores between the intervention and control groups is also presented.
The analysis was conducted by using SPSS and Stata. Because of the number of separate calculations, the significance level was set at P < .01.
After randomization, 94 mothers were included in the intervention group, with 91 controls. At 1 year, there were 81 (13 lost) in the intervention group and 71 controls (20 lost) (see Fig 2). The reasons for loss to follow-up were predominantly difficulty in tracing mothers who did not respond to repeated telephone calls or letters. Two children were withdrawn from the control group (1 child diagnosed with cerebral palsy, 1 taken into care). Only 4 mothers were from an ethnic minority, as is typical of the local resident population.
There were some significant differences between those who remained in the study and the 33 who were lost to follow-up on independent measures of marital status, age, and level of education (Table 1). Over half of the very young mothers (aged 15–19 years), and 29% of mothers with basic school-leaving qualifications (or no qualifications) had dropped out at 1 year. However, the difference in deprivation levels (Townsend Index score) between those lost or remaining in the study was not significant.
There were some significant differences between the research sample and those lost to follow-up also on dependent measures. Those remaining in the study had significantly higher scores for frequency and intensity of Daily Hassles. There was, however, no difference in the baseline scores on the other 2 questionnaires.
Randomization to Groups
The randomization produced mostly equivalent groups. There were no differences between the intervention (n = 94) and control (n = 91) parents at baseline with respect to maternal age, marital status, number employed, and community level of deprivation. There was however, a significant difference in level of education, where 36% of intervention mothers held a degree or postgraduate degree compared with 16% of control mothers (χ2 = 12.23; P = .002). Whether a mother had been employed did not differ significantly between groups (χ2 = 0.33; P = .412), but there was an almost significant trend toward intervention mothers having been more likely to work full-time (81%) as opposed to control mothers (61%) (χ2 = 7.30; P = .026). The pattern of equivalence between intervention and control groups was the same for those successfully followed up (level of education, χ2 = 12.95, P = .002; employment pattern, χ2 = 8.81, P = .012; no other differences). Therefore, subsequent analyses of dependent measures were adjusted for highest qualification achieved and employment pattern.
Differences Between Intervention and Control Groups on Dependent Measures
The mean values at baseline and year 1 for the GHQ12, Daily Hassles, and APPI questionnaires are presented in Table 2. The groups did not differ significantly at baseline except for frequency of Daily Hassles (t = 2.65, P = .009), with an almost significant difference for intensity (t = 2.48, P = .014). In both cases, intervention group mothers reported higher levels of hassle at baseline. The evaluation of the baseline to year 1 change for each measure is presented in Table 3.
From baseline to year 1, the mean total GHQ12 score in each group fell, indicating an improvement in the sense of well-being in both groups of participants. At year 1, the intervention group's mean total score was lower than that of the controls; however, the difference in the amount of change between the intervention and control groups was not significant after adjustment for highest qualification achieved and previous employment pattern.
By the end of year 1, both groups reported a reduction in the intensity of daily hassles, but more so in the intervention group than in the control group. There was a significantly larger reduction in the perceived intensity of daily hassles in the intervention group, with a medium effect size.17 Similar results were found for the frequency of daily hassles, where change reported by the intervention group mothers was significantly more than in the control group, with a small-to-medium effect size.
There was a significant difference in the change in scores over the year of the study between intervention and control groups for construct A (inappropriate expectations of children), where the change score in the intervention group was significantly larger than in the control group. Both groups' scores increased, ie, improved, from baseline to year 1, but more so for the intervention group. However, the effect size was small, and the mean year 1 value for the intervention group remained lower than the mean baseline value for the control group. No difference in change between intervention and control groups was found for the other constructs.
Little work has been conducted previously on the value to parents of educational materials, including the publication given out to all new parents in the United Kingdom (“Birth to Five”18). Despite apparent access to a wide range of information on parenting issues, parents say there is little information on the emotional needs of very young children.19
Yet educational materials are known to be of value in promoting health, particularly if combined with other methods,20,21 and the age-paced principle of providing small amounts of information at a time when parents need it most would seem more likely to be successful than offering it all at once in a book.
Our evaluation suggests significantly greater improvements in several outcomes for the mothers who received Baby Express during the first year of their new infant's life. Given randomized allocation to study groups, these improvements may be taken to result from the intervention. Interpretation of results is complicated by the fact that intervention group mothers reported higher levels of perceived daily hassles at the start of the study, and both groups showed an improvement over the first year. However, improvements were more marked in the intervention group.
The reduction in how stressful daily hassles seemed was a considerable effect. It would translate on average as 3 types of hassle (eg, feeding difficulties, demanding constant attention, or being hard to manage in public) changing from “big” to “little.” The reason that intervention mothers felt less stressed may relate to greater understanding of the meanings of infant behavior from reading Baby Express. The effect size suggests that the influence of this intervention over a large population would be important. The similar change in reported (in)appropriate expectations of children may add extra weight to these findings. Change in the other constructs of the AAPI, such as belief in the use and value of corporal punishment, may well become more relevant during subsequent years of the child's life.
In many families, information alone is unlikely to bring about significant changes, but if used as part of a package of parenting support (eg, in a Children's Centre), it could add value particularly in relation to emotional development and behavior. There is substantial research evidence that degree of maternal acceptance is related to more competent child functioning.22
LIMITATIONS OF THE STUDY
The attrition of 33 parents from the total cohort may have influenced the results. That younger, less qualified, less settled participants were harder to trace is not surprising; however, the remaining sample did not differ from the local community level of deprivation. The cohort had higher levels of perceived daily hassles at baseline than those who dropped out, which may have increased the size of the reduction over time; nevertheless, it would be difficult to argue that this had biased the finding of greater reduction in the intervention group as fewer dropped out from this group.
Outcomes as reported in this article all rely on report by mothers and do not include direct observations of parenting style.22 A subsample has been studied playing with their children (data to be reported separately).
An age-paced newsletter, delivered monthly to first-time parents of new infants, seems to be an effective public health measure to improve early parenting quality. The study has shown, however, that those most at risk are hardest to recruit and keep track of. The newsletter is aimed at first-time mothers, which is thought to be one of the reasons for the success of the home visitation program in the United States.23 It is a low-cost universal intervention that carries no stigma and has been shown to be very popular with parents.7,8Baby Express holds considerable potential as a means of improving parent-child interaction and thereby relationships in later childhood.
This work was supported by the Sure Start Unit of the United Kingdom Government Department of Children, Schools and Families.
We are very grateful to the producers of Baby Express (Heather Neil and Kate Welford) for their excellent work, to Peter James for additional statistics support, to the Children's Foundation for their backing throughout, and to the parents who have given us generously of their time over the period of the study. Sincere thanks also to Dave Riley in Madison, Wisconsin, for his assistance and ongoing support.
- Accepted April 22, 2008.
- Address correspondence to Tony Waterston, MD, FRCPCH, Newcastle University, School of Clinical Medical Sciences, Sir James Spence Institute, Royal Victoria Infirmary, Newcastle NE1 4LP, United Kingdom. E-mail:
The authors have indicated they have no financial interests relevant to this article to disclose.
What's Known on This Subject
The provision of parental stimulation in infancy is correlated with later outcomes in emotional well-being and child development. Support to parents is known to benefit this provision. The value of written information to parents has not been evaluated.
What This Study Adds
An age-paced parenting newsletter sent to parents monthly in the first year of life was read and enjoyed. Parents receiving the newsletter were less likely to be “hassled” by the new infant and showed more positive parenting characteristics than controls.
- ↵Royal College of Paediatrics and Child Health. Helpful Parenting. London, United Kingdom: Royal College of Paediatrics and Child Health; 2002
- ↵Department for Education and Skills. Every Parent Matters. London, United Kingdom: Department for Education and Skills; 2007
- ↵Moran P, Ghate D, van der Merwe A. What Works in Parenting Support? A Review of the International Evidence. London, United Kingdom: Policy Research Bureau; 2004
- ↵Hewitt K, Mason L, Snelson W, Crawford, W. Parent education in preventing behaviour problems. Health Vis.1991;64 (12):415– 417
- ↵Townsend P, Phillimore P, Beattie A. Health and Deprivation. London, United Kingdom: Croom Helm, Beckenham; 1988
- ↵Goldberg D, Williams PA. User's Guide to the General Health Questionnaire. London United Kingdom: NFER-Nelson; 1988
- ↵Department of Health. Framework for the Assessment of Children in Need and their Families. London, United Kingdom: The Stationery Office; 2000
- ↵Bavolek SJ, Keene RG. Adult-Adolescent Parenting Inventory. Assessing High Risk Parenting Attitudes and Behaviours. Park City, UT: Family Development Resources Inc; 2001
- ↵Department of Health. Birth to Five Book. London, United Kingdom: Department of Health; 2006
- ↵Nicholas D, Marden M. Parents and their information needs: A case study: parents of children under the age of five. J Library Info Sci.1998;30 (1):35– 47
- ↵Serra C, Cabezas C, Bonfill X, Pladevall-Vila M. Interventions for preventing tobacco smoking in public places. Cochrane Database Syst Rev.2008;(3):CD001294
- ↵Brunner EJ, Thorogood M, Rees K, Hewitt G. Dietary advice for reducing cardiovascular risk. Cochrane Database Syst Rev.2007;(4):CD002128
- ↵Melhuish E, Belsky J, Anning A, Ball M, Barnes J, Romaniuk H, et al., and the NESS Research Team. Variation in community intervention programmes and consequences for children and families: the example of Sure Start Local Programmes. J Child Psychol Psychiatry.2007;48 (6):543– 551
- ↵Olds DL, Robinson J, Pettitt L, et al. Effects of home visits by paraprofessionals and by nurses: age 4 follow up of a randomized trial. Pediatrics.2004;114 (6):1560– 1568
- Copyright © 2009 by the American Academy of Pediatrics