Evaluation of a healthy lifestyle intervention among disadvantaged preschoolers

30 Jul 2012

Evaluation is defined as "the process by which we judge the worth or value of something" (Suchman, 1967). Improving nutrition, reducing obesity and increasing the level of exercise were three priority population health objectives identified by the Ministry of Health in the early 2000's. The Healthy Eating – Healthy Action: Oranga Kai - Oranga Pumau Strategy (the HEHA Strategy) (Ministry of Health, 2003b)) and accompanying Implementation Plan (Ministry of Health, 2004b) was the New Zealand Governments response to these issues. In 2005, all District Health Boards (DHBs) were invited to submit proposals to access the HEHA Innovation Fund. The second phase of the successful Southland DHBs proposal was the health promotion programme Healthy Me and You. A quasi-experimental study measuring the effectiveness of Healthy Me and You was conducted between 2008 and 2009 with the recipients of the intervention recruited from within Southland DHB boundary and a matched comparison group obtained from the adjoining Otago DHB. The programme was specifically designed for socially disadvantaged caregivers and their preschool age children in relation to increasing physical activity levels and fruit and vegetable intake. Healthy Me and You was delivered once a week over six consecutive weeks for two hours in a small group community setting. Twelve programmes were run in six locations during 2008 across Southland. Participants were referred to the study by health and social service agencies in both regions. The content and design of the intervention was informed by conducting five focus groups with parents and health professionals; using the social cognitive theory as a framework to design the content; and finally piloting the programme and amending where necessary. At baseline, data were collected on 135 caregivers and their child aged three-to-four years (n=67 intervention group; n=68 comparison group). The majority of caregivers were female; a third were from Maori or Pacific origin; with a median age 33 of years. Just under half were in some paid employment; with a third having a tertiary qualification while a quarter reported having none. Most children lived with at least two adults and another child. Data were collected a second time after six months, with 113 participants (n=54, intervention group: n=59, comparison group) retained. Information (caregiver and child characteristics and behaviours) were collected using three instruments: NL-1000 accelerometer; fruit and vegetable tick-list; and two interview schedules (core plus supplementary interviews) conducted eight days apart. These included household characteristics; child family food environment; caregiver and child sedentary behaviour, dietary habits and physical activity(PA). Outcome measures investigated were child PA (in minutes), number of fruit offerings to child by caregiver, and number of vegetable offerings to child by caregiver.In order to gain a better understanding of the participants, predictors of the outcome variables were first explored before examining the effectiveness of the intervention. Regression models (tobit and poisson) were used to model the outcome variables. As these were all obtained from repeated measures, a random participant effect was included. Univariate screening was carried out using p<0.20 for inclusion in subsequent models. Results showed children's PA levels reduced as they grew older; were higher for non-Maori/non-Pacific; and increased as the number of caregiver steps increased or if their caregiver was in paid employment. All effect sizes were however small. Likewise several behaviours were shown to increase fruit offerings or vegetable offerings but they were all of limited practical significance due to the large number of additional offerings required to produce an increase of either one fruit or vegetable offering. No statistically significant effect of the intervention was found associated with any of the three key outcome variables while controlling for their respective baseline values. Each outcome was then explored further by considering other characteristic-type covariates (caregiver and child) with adjusted models producing the same result. Changes at an individual child level were identified, however these were small and were not of any practical importance. Boys did more PA, as did children with older caregivers or children who had a male caregiver. It was also found that as caregiver education levels increased, fruit offerings also increased compared to those with no or minimal education. Offerings of vegetables were found to decrease if the child was male. An investigation into whether there were any subgroups of participants, who benefited more from the intervention, was also carried out. Results showed that there was no statistically significant subgroup effect on child PA or child vegetable offering by the caregiver, while controlling for corresponding baseline values. A very small increase in fruit offerings was however found (p=0.046) with respect to increasing attendance rates by caregivers but this was considered to be of marginal practical importance (RR=1.1).These findings show that although health promotion programmes are regularly used to promote healthy lifestyles for children, for socially disadvantaged caregivers with preschoolers in New Zealand, programmes such as Healthy Me and You were not found to be as effective as initially thought. As there is currently little evidence available in this area of research further research is recommended to enable researchers and policy makers to be more informed so that our limited resources available can be used effectively to tackle the major public health issue of obesity.

Publication Details
Published year only: 
Subject Areas
Geographic Coverage