How much is '5-a-day'? A qualitative investigation into consumer understanding of fruit and vegetable intake guidelines

BACKGROUND: Despite the known health benefits of fruit and vegetables (FV), population intakes remain low. One potential contributing factor may be a lack of understanding surrounding recommended intakes. The present study aimed to explore the understanding of FV intake guidelines among a sample of low FV consumers.

METHODS: Six semi-structured focus groups were held with low FV consumers (n = 28, age range 19-55 years). Focus groups were recorded digitally, transcribed verbatim and analysed thematically using nvivo (QSR International, Melbourne, Australia) to manage the coded data. Participants also completed a short questionnaire assessing knowledge on FV intake guidelines. Descriptive statistics were used to analyse responses.

RESULTS: The discussions highlighted that, although participants were aware of FV intake guidelines, they lacked clarity with regard to the meaning of the '5-a-day' message, including what foods are included in the guideline, as well as what constitutes a portion of FV. There was also a sense of confusion surrounding the concept of achieving variety with regard to FV intake. The sample highlighted a lack of previous education on FV portion sizes and put forward suggestions for improving knowledge, including increased information on food packaging and through health campaigns. Questionnaire findings were generally congruent with the qualitative findings, showing high awareness of the '5-a-day' message but a lack of knowledge surrounding FV portion sizes.

CONCLUSIONS: Future public health campaigns should consider how best to address the gaps in knowledge identified in the present study, and incorporate evaluations that will allow the impact of future initiatives on knowledge, and ultimately behaviour, to be investigated.

Machine learning classification of surgical pathology reports and chunk recognition for information extraction noise reduction

Background and aims: Machine learning techniques for the text mining of cancer-related clinical documents have not been sufficiently explored. Here some techniques are presented for the pre-processing of free-text breast cancer pathology reports, with the aim of facilitating the extraction of information relevant to cancer staging.

Materials and methods: The first technique was implemented using the freely available software RapidMiner to classify the reports according to their general layout: ‘semi-structured’ and ‘unstructured’. The second technique was developed using the open source language engineering framework GATE and aimed at the prediction of chunks of the report text containing information pertaining to the cancer morphology, the tumour size, its hormone receptor status and the number of positive nodes. The classifiers were trained and tested respectively on sets of 635 and 163 manually classified or annotated reports, from the Northern Ireland Cancer Registry.

Results: The best result of 99.4% accuracy – which included only one semi-structured report predicted as unstructured – was produced by the layout classifier with the k nearest algorithm, using the binary term occurrence word vector type with stopword filter and pruning. For chunk recognition, the best results were found using the PAUM algorithm with the same parameters for all cases, except for the prediction of chunks containing cancer morphology. For semi-structured reports the performance ranged from 0.97 to 0.94 and from 0.92 to 0.83 in precision and recall, while for unstructured reports performance ranged from 0.91 to 0.64 and from 0.68 to 0.41 in precision and recall. Poor results were found when the classifier was trained on semi-structured reports but tested on unstructured.

Conclusions: These results show that it is possible and beneficial to predict the layout of reports and that the accuracy of prediction of which segments of a report may contain certain information is sensitive to the report layout and the type of information sought.