Abstract
Fruit and vegetables play an important role in disease prevention. In New Zealand, 58% of adults do not consume at least five servings of fruits and vegetables a day, and 58% of low fruit and vegetable consumers are unmotivated to meet recommended intakes.
To increase a population’s fruit and vegetable consumption, cost-effective public health interventions are needed to address individual, social, and environmental factors influencing food choice decisions. Across health behaviours, the most effective communication interventions extensively use psychology theory and behavioural change techniques.
The Transtheoretical Model has been widely used in clinical and community settings, but the design and evaluation of these interventions to increase adult fruit and vegetable consumption requires further investigation using valid and reliable measures of multiple Transtheoretical Model constructs. The aim of this research is to identify and overcome challenges associated with developing and validating dual-purpose instruments to measure key Transtheoretical Model constructs.
Individual interviews were conducted with New Zealand adults to design self-administered questionnaires assessing fruit/vegetable intake, stage of change, decisional balance, and self-efficacy. Apart from the intake instruments, the validation studies were conducted amongst a reasonably nationally representative sample of New Zealand adults ages 25 to 60 years (response rates: 72% [n = 231]; 67.4% [n = 2132]). A purposeful sample of 100 adults ages 25 to 52 years was used for the intake validation study.
Four brief food frequency questionnaires were validated against a dietary history method to measure an individual’s usual daily intake of fruit or vegetable servings over the past month. The best performing fruit/vegetable instrument had adequate validity to assess an individual’s goal attainment and group intake, necessary for creating and evaluating individually tailored communications.
The staging assessment method categorised individuals into qualitatively distinct stages of change using a validated instrument to assess individual goal attainment, a pre-action stage measure of goal behaviour intention, and a post-action stage measure of behaviour duration. Validity was tested against an independent dietary assessment method and previously validated scales. The construct validity of these staging algorithms was established based on behavioural, decisional balance, and self-efficacy criteria.
In the scale development research, extensive qualitative research was conducted with the target audience to generate culturally appropriate items and to assess translation validity. Principal components analysis and structural equation modelling were used to develop and confirm the scales. The self-efficacy to eat fruit/vegetables scale had very good internal reliability and construct validity based on dietary intake. The decisional balance to eat more fruit and vegetables scale had a hierarchical structure of pros (health incentives and non-health incentives) and cons (disincentives). This reliable scale had sufficient construct validity for predicted patterns across the stages of change. The test-retest reliability of these instruments needs to be established.
Used appropriately, these instruments can be used in public health and clinical settings to screen, design communications, and evaluate interventions for New Zealand adults ages 25 to 60 years. Further research is needed to test the validity and reliability of these instruments in at-risk subgroups of this study population.