Dietary data in the 1970 British Cohort Study

Learn about the 1970 British Cohort Study (BCS70) and its dietary measurements

Summary of cohort
Dietary data collection
Estimation of nutrient intake
Response
Key findings

Summary of cohort

The 1970 British Cohort Study (BCS70) is the third birth cohort study established in Britain. The BCS70 began as the British Births Survey, designed to examine the causes of neonatal morbidity and compare results with those of NSHD [62]. The target sample included just over 17,000 babies born in England, Scotland, Wales and Northern Ireland in one week in April 1970 [63]. Since the initial birth survey there have been nine further major data collections: 1975 (5 years), 1980 (10 years), 1986 (16 years), 1996 (26 years), 1999-2000 (29-30 years), and 2016 (46 years). Data collection on sub-samples has also been conducted at 22 and 42 months and 21 years [62-64]. Cohort members who were born in Northern Ireland were included in the birth survey but dropped from later sweeps. The sample was augmented to include immigrants who were born in the same week in April 1970 at ages 5 (n=68), 10 (n=270), 16 (n=57) and 26 years (n=8). It is worth noting that in 1986, there were postal strikes and industrial action by teachers, and therefore the overall response to this sweep was likely reduced [62, 64]. Information from this study has come from a number of sources (e.g. participants, parents, schools, medical records, doctors) and a variety of instruments (e.g. interview, self-completion, medical records, diaries). Although extensive information has been gathered, the ad hoc funding during the 1980s and 1990s made it difficult to develop consistent content and timing for each of the follow-ups [62].

Dietary data collection

Information about dietary intake was collected during the birth survey and at ages 10 (1980), 16 (1986), 30 (2000), 42 (2012) and 46 (2016) years using different methods. A summary of the different diet-related questions asked at each time point are outlined in the paragraphs and table below.

Infant diet

During the birth survey, the carer was asked questions relating to breast feeding and alternatives (questionnaire: https://cls.ucl.ac.uk/wp-content/uploads/2017/07/BCS70-Birth-Questionnaire.pdf).

Diet in childhood

Fifteen separate survey documents were used to capture information from participants when they were 10 years old. In the pupil questionnaire form, participants were asked about the frequency of consumption of certain foods and questions relating to purchasing of snacks, bringing snacks to school, and whether the participant received free school meals were also asked (see the table below).

The sweep at 16 years, known as ‘Youthscan’, consisted of 19 separate survey documents. Diet-related questions were asked during the Student Test Booklet/Student Score Form (Documents B & C, section titled ‘What I Eat’ and ‘Soft Drink Special’), Health-Related Behaviour Student Self-Completion (Document F), ‘Home and All That’ Student Self-Completion (Document G), Maternal Self-competition Form (Document P). In the Maternal Self-Completion Form, the participant’s carer was asked an extensive amount of diet related questions about the teenager as well as cooking and preparation habits. These questions have not been outlined in detail in the table below as there is overlap with the self-completion dietary questions, but can be found on the relevant questionnaire: https://cls.ucl.ac.uk/wp-content/uploads/2017/07/docp.pdf.

In addition to the self-completed dietary habits, the teenagers were asked to keep a 4 day estimated diet diary. The diaries were distributed through schools or sent by post to the participant’s home where appropriate. Participants were provided with instructions both within the diary and as an accompanying letter. Examples were provided throughout the diary. Participants were instructed to complete the diary on a Friday, Saturday, Sunday and Monday of one week. They were asked to include information about the type, brand, amount and time of all food eaten along with additional information about their eating habits. The diary had separate sections for before breakfast, breakfast, mid-morning between breakfast and midday meal, midday meal, mid-afternoon between midday meal and evening meal, evening meal and evening snacks. it was noted that many participants were revising for exams at home or were on holiday from school during the period of recording [65].

Diet in adulthood

At age 30, the frequency of consumption of a limited number of specified foods were recorded. These questions were identical to the questions asked in NCDS when participants were 42 years as a simultaneous survey of both cohorts was conducted in 1999-2000 using the same questionnaire. Participants were also asked if they were a vegetarian and/or on any special diet.

A small number of diet-related questions were asked during the self-completion questionnaire in 2012 when participants were 42 years.

In the latest data collection in 2016-2017 (46 years) dietary intake was collected using an online diet diary. The Oxford WebQ, developed by the Cancer and Epidemiology Unit at the University of Oxford was used [66]. Briefly, participants are asked to report all food and drink consumed prior to the day they completed it along with portion sizes. Participants completed two 24-hour recalls. They were randomly allocated to complete the survey on one weekday and one weekend.

Diet-related questions in BCS70

	1980		1986			1999-2000	2012	2016
	Age 10y		Age 16y			Age 29-30y	Age 42y	Age 46y
Questionnaire source*	Pupil form	B & C	F	Questionnaire	Diet-Diary	Questionnaire	Questionnaire	Questionnaire
4-day diet-diary					✔
Online diet-diary								✔
24-hour recall**			✔
Frequency of fresh fruit						✔
Frequency of salad/raw vegetables						✔
Frequency of cooked vegetables						✔
Frequency of white bread	✔	✔
Frequency of brown bread	✔
Frequency of wholemeal/granary bread		✔
Frequency of wholemeal bread or rolls						✔
Frequency of other bread or rolls						✔
Frequency of breakfast cereal		✔
Frequency of butter	✔	✔
Frequency of margarine	✔	✔
Frequency of cheese	✔	✔
Frequency of eggs	✔	✔				✔
Frequency of pulses						✔
Frequency of meat	✔	✔
Frequency of red meat						✔
Frequency of chicken or turkey		✔
Frequency of poultry						✔
Frequency of fish	✔	✔				✔
Frequency of chips		✔				✔
Frequency of food fried in vegetable oil						✔
Frequency of food fried in hard fat						✔
Frequency of chocolate/sweets	✔					✔
Frequency of sweets		✔
Frequency of chocolate		✔
Frequency of puddings		✔
Frequency of biscuits and cakes of all kinds						✔
Frequency of cakes or buns		✔
Frequency of sweet biscuits		✔
Frequency of crisps		✔
Cups/glasses tea	✔
Frequency of tea		✔
Cups/glasses cocoa	✔
Frequency of cocoa		✔
Cups/glasses coffee	✔
Frequency of coffee
Cups/glasses milk	✔
Frequency of milk		✔
Cups/glasses cola	✔
Frequency of cola		✔
Frequency of low calorie or sugar free drinks		✔
Frequency of fresh fruit juice		✔
Frequency of fruit squash		✔
Cups/glasses water	✔
Frequency of water		✔
Type of lunch yesterday†			✔
Type of breakfast yesterday‡			✔
Number of days consuming breakfast							✔
How much milk sugar and sweetener you add to tea or coffee				✔
Are you a vegetarian/special diet				✔		✔
Number of times per week you get food from takeaway				✔
Frequency of takeaways							✔
Frequency of home-cooked meal							✔
Frequency of ready meals							✔
Frequency of other convenience foods							✔

Response options.
In 1980 responses for frequency questions were: nearly every day, quite often, sometimes, and hardly ever.
In 1986 responses for frequency questions ranged from zero to seven times per week. A separate ‘soft drinks’ section assessed the consumption of drinks with a variety of responses in 1986.
In 2000 responses for frequency questions were: more than one per day, one per day, 3-6 days per week, 1-2 days per week, less than one day a week, occasionally, never.
In 2003, responses were: one per day, one per day, several times a week, once or twice a week, at least once a month, less often, never.

Notes.
* Link provided to CLOSER Discovery (interactive version) or original study PDF as applicable.
** Recall asked about specific items including amounts of: meat, fish, eggs/cheese, milk, tea/coffee, cereal, bread, soup, potatoes, baked beans, rice/spaghetti, packets of crisps, ice lollies, ice cream, mousse, sweets, chocolate biscuits of bars, biscuits, cakes or tarts, fruit pies, puddings, squash or cordial, fizzy drinks, fruit juice, fresh fruit, raw vegetables or salads, cooked vegetables (not potatoes), alcoholic drinks, other items.
† Over counter in school/packed lunch in school/outside school takeaway/outside school packed lunch/at home/did not have lunch.
‡ Nothing, just something to drink, cereal or bread and drink, cereal and bread and drink, just an egg and drink, cooked breakfast with bread and drink, cooked breakfast with cereal and bread and drink.

Estimation of nutrient intake

Nutrient intakes from the 16 year dietary data were estimated as outlined by Crawley et al. [65]. It is notable that the procedures used to code this data were not supported by specially developed software as discussed in the NSHD section of this guide. Briefly, coding was completed using standard food portion sizes [67] from a list of 1250 foods. McCance and Widdowson’s The composition of Foods 4^th Edition was used as the nutrient database [38] along with additional information on the nutrient contents of other products as outlined by Crawley et al. [65]. There was no evidence of differences between weekday and weekend nutrient intakes and it was noted that many participants were revising for exams at home or were on holiday from school during the period of recording [65].

Nutrients from the 46 year online diet diary were estimated automatically within the Oxford WebQ, [66] from McCance and Widdowson’s The Composition of Foods and its supplement reports [25, 39-41, 66, 68-72].

Response

Response to dietary measures in BCS70

Original cohort consisted of 18,640 participants (including those not born in Great Britain, added during school years).

Year	Age (y)	N interviewed	Response to at least one diet question (n(%))
1970	Birth	17,196	17,175 (99.9%)
1980	10	14,869	12,695 (85%)
1986	16	11,615	6,651*
2000	30	11,261	11,205 (99.5%)
2012	42	9,841	8,721 (88.6%)
2016	46	8,581	5,950 (69.3%)

Notes.
* Participant response, carer questionnaire n=8,993. n=4760 competed the 4 day unweighted diet diary.

Key findings

Diets of teenagers

The majority of diet-related analyses in BCS70 were cross-sectional and conducted in the 1990s using the 16 year diet diaries [65, 73-79]. Teenager’s intakes of fats and extrinsic sugars exceed the 1991 Department of Health recommendations and intakes of milk sugars, starch and non-starch polysaccharides were lower than recommended [65]. Regular breakfast consumption at 16 years was associated with lower intakes of fat while non-consumers had lower micronutrient intakes [79]. Higher fibre breakfast cereals were more likely to be consumed in London and the Southeast than in Scotland and the North and less likely to be consumed as the socioeconomic position declined [79]. When looking at regional differences, dietary intake of Scottish teenagers was different compared to the rest of Great Britain even after accounting for smoking, alcohol, family size and family tenure [73]. These teenagers had lower intakes of fibre, some micronutrients, non-processed vegetables, and polyunsaturated fat spread and higher intakes of soft drinks, chips and white bread [73]. Two studies examining the dietary intakes of dieting teenagers found that their total energy and micronutrient intake were lower than non-dieters but their protein intake as a percentage of energy was higher [74, 75]. Teenagers whose parents smoked had diets that were lower in fibre, vitamin C, folate and magnesium with lower intakes of fruit juices, wholemeal bread and vegetables compared to those with non-smoking parents [77]. In a similar study among teenage smokers, the authors found they consumed less fibre and vitamin C, as well as fewer puddings, biscuits, wholemeal bread, fruit juices [78]. Eating takeaway meals twice or more per week, consuming two or more soft drinks per day and a history of dieting to lose weight at 16 years was associated with an increase in BMI z-scores between 16 and 30 years [80].

Comparison of diet between NCDS and 1970 British birth cohort

In a study that compared dietary intake between NCDS and the 1970 British birth cohort, the authors identified and compared clusters of health behaviours, including diet, smoking, alcohol and physical activity among participants in their thirties. The authors used principal component analyses to summarise dietary intake into three variables; fruit and vegetable, chips and fried food, sweets, chocolate and biscuits. They identified three clusters that were similar among men and women: a risky group, a moderate smokers group and a mainstream group. The mainstream group included not smoking, frequent fruit and vegetable consumption, less frequent consumption of chips and fried food and being more physically active [81]. Consumption of sweet foods, however, was also common in the mainstream cluster. Cluster patterns were similar between men and women and across cohorts. More people in the BCS70 fell into the mainstream group and tended to have healthier behaviours (except alcohol) than NCDS.

Learn about the other studies covered by this guide and their dietary measurements:

Get background detail on this guide:

Learn about harmonisation in the context of dietary data:

Discussion & illustration of dietary data harmonisation

Further information:

This page is part of the CLOSER resource: ‘A guide to the dietary data in eight CLOSER studies’.