Associations among 25-year trends in diet, cholesterol and BMI from 140,000 observations in men and women in Northern Sweden

Abstract

Background: In the 1970s, men in northern Sweden had among the highest prevalences of cardiovascular diseases (CVD) worldwide. An intervention program combining population- and individual-oriented activities was initiated in 1985. Concurrently, collection of information on medical risk factors, lifestyle and anthropometry started. Today, these data make up one of the largest databases in the world on diet intake in a population-based sample, both in terms of sample size and follow-up period. The study examines trends in food and nutrient intake, serum cholesterol and body mass index (BMI) from 1986 to 2010 in northern Sweden.

Methods: Cross-sectional information on self-reported food and nutrient intake and measured body weight, height, and serum cholesterol were compiled for over 140,000 observations. Trends and trend breaks over the 25-year period were evaluated for energy-providing nutrients, foods contributing to fat intake, serum cholesterol and BMI.

Results: Reported intake of fat exhibited two significant trend breaks in both sexes: a decrease between 1986 and 1992 and an increase from 2002 (women) or 2004 (men). A reverse trend was noted for carbohydrates, whereas protein intake remained unchanged during the 25-year period. Significant trend breaks in intake of foods contributing to total fat intake were seen. Reported intake of wine increased sharply for both sexes (more so for women) and export beer increased for men. BMI increased continuously for both sexes, whereas serum cholesterol levels decreased during 1986 - 2004, remained unchanged until 2007 and then began to rise. The increase in serum cholesterol coincided with the increase in fat intake, especially with intake of saturated fat and fats for spreading on bread and cooking.

Conclusions: Men and women in northern Sweden decreased their reported fat intake in the first 7 years (1986-1992) of an intervention program. After 2004 fat intake increased sharply for both genders, which coincided with introduction of a positive media support for low carbohydrate-high-fat (LCHF) diet. The decrease and following increase in cholesterol levels occurred simultaneously with the time trends in food selection, whereas a constant increase in BMI remained unaltered. These changes in risk factors may have important effects on primary and secondary prevention of cardiovascular disease (CVD).

Figure 1

Number of observations in the study cohort. Number of observations excluded by various criteria and observations remaining for analyses are given.

Figure 2

Mean intake of fat, protein and carbohydrate expressed as energy from each nutrient in per cent of total energy intake. Mean intakes, adjusted for age and BMI for the period 1986 to 2010 for men and women. Information was not available for 1987 to 1990 since the VIP FFQ was not fully harmonized until 1991, and the reduced FFQ version used in MONICA year 1990 was not acceptable for nutrient estimation. Black dots (●) show years with a trend shift as indicated by the Joint Point statistical programme.

Figure 3

Reported mean intake of various types of fats used for spreading on bread or cooking. Reported weekly mean intake frequencies, adjusted for age and BMI for the period 1986 to 2010 for men and women. Information was not available for 1987 and 1988 since the VIP FFQ was not fully harmonized until 1991. Black dots (●) show years with a trend shift as indicated by the Joint Point statistical programme.

Figure 4

Reported mean intake of various types of non-fermented milk. Reported weekly mean intake frequencies, adjusted for age and BMI for the period 1986 to 2010 for men and women. Information was not available for 1987 and 1988 since the VIP FFQ was not fully harmonized until 1991. Black dots (●) show years with a trend shift as indicated by the Joint Point software.

Figure 5

Reported mean intake of various types of oils used for salad dressing or cooking, cream and cream products, and non-sweet snacks. Reported weekly mean intake frequencies, adjusted for age and BMI for the period 1986 to 2010 for men and women. Information was not available for 1987 and 1988 since the VIP FFQ was not fully harmonized until 1991. Black dots (●) show years with a trend shift as indicated by the Joint Point software.

Figure 6

Reported mean intake of various types of alcohol containing products. Reported weekly mean intake frequencies, adjusted for age and BMI for the period 1986 to 2010 for men and women. Information was not available for 1987 and 1988 since the VIP FFQ was not fully harmonized until 1991. Black dots (●) show years with a trend shift as indicated by the Joint Point software.

Figure 7

(A) Mean body mass index (BMI) by study year and proportions with BMI > 27; (B) mean total cholesterol in serum by study year and proportions reporting intake of cholesterol lowering medication by study year. Means, adjusted for age for the period 1986 to 2010 for men and women. Black dots (●) show years with a trend shift as indicated by the Joint Point software.