Cardiovascular diseases are the leading cause of death in most industrialised countries and in developing countries the trend in cardiovascular-related deaths is increasing. World-wide, type 2 diabetes mellitus (T2DM) is an emerging cause of disability and premature death. Both these conditions are closely associated with the consumption of energy-dense foods and food products that are poor in nutrients, as well as with a sedentary lifestyle. Pharmacological and surgical interventions can improve the outcome and delay the progression of the disease, but in terms of population-level prevention there is no substitute for the adoption of a healthy lifestyle. SETTING The underlying studies were conducted in Västerbotten (the VIP study), and in Norrbotten and Västerbotten combined (the MONICA Project). Norrbotten andVästerbotten are the two northernmost counties in Sweden. Since the mid-1980sthe prevalence of cardiovascular disease has decreased and diabetes rates haveremained stable in this region, despite of an unbroken trend of increasing body weight. OBJECTIVE The aim of this thesis is to describe changes in reported dietary habits, estimatetheir relative importance as risk factors for diabetes and cardiovascular disease, and finally to identify lifestyle components as potential targets for intervention. RESULTS The first paper describes changes in self-reported food consumption between 1986 and 1999. During this period, the population in question switched from products with high saturated fatty acid content (e.g. milk containing 3% fat, butter) to foods containing less saturated fat (e.g. milk containing 1.5% fat, vegetable oil, low-fat margarine); pasta and rice were consumed more often, and potatoes were consumed less. Convenience foods (e.g. hamburgers, snacks, sweets) became more popular, whilst traditional dishes (e.g. potato dumplings, black pudding, blöta) decreased in popularity. Fruit and vegetable intake remained low. In paper two we study the effects of these changes in food intake on the risk of developing T2DM using body fat distribution as an early indicator. Increased consumption of convenience foods was associated with unfavourable changes (smaller hip circumference and larger waist circumference), whereas the increased consumption of vegetable oil and pasta was associated with low-risk fat distribution. In the third paper we report studies on the association between fat consumption and T2DM. We used the pattern of fatty acids in the membranes of red blood cells as a marker of fat intake. In addition to confirming earlier findings (markers of the intake of saturated fat are associated with increased risk of T2DM and markers of unsaturated fat are associated with reduced T2DM risk), we also identified associations between two markers of milk-derived saturated fat intake and enterolactone, a biomarker of dietary fibre intake, and the risk of developing myocardial infarction. Our results indicate that moderately high levels of enterolactone intake in men are associated with lower risk of experiencing myocardial infarction. Manuscript 5 ranks education level, physical activity, smoking status, and self-reported intake of dietary fibre and fatty acids according to their effects on body fat distribution. Increased levels of physical activity, a higher education level and a reduced intake of saturated fat from meat were ranked as the most strongly associated factors in both men and women. Increased intake of dietary fibre from grains in women, and increased intake of dietary fibre from fruits and vegetables in men, was also inversely associated with average waist circumference. CONCLUSION Both questionnaire-based and biological markers of the risk of developing diabetes or cardiovascular disease have been identified. Based on available population level measurements, reduced consumption of convenience foods, increased consumption of whole-grain products, fruits and vegetables, vegetable oil and pasta as well as increased physical activity are potential goals for interventions in northern Sweden.

Background: Neonates born with low birth weight or preterm are at an increased risk of long-term adverse health outcomes. Research studies on the association of fish consumption during pregnancy and birth outcomes, have led to inconsistent conclusions. Maternal dietary intakes during pregnancy have a significant impact on foetal development and growth. The aim of this project is to determine levels of maternal fish intake at <18 weeks during pregnancy and to determine the association with birth outcomes in pregnant women from Johannesburg, South Africa. Methods: This Master‘s study is nested in a larger study with a longitudinal observational research design was conducted on 250 pregnant women in Johannesburg, South Africa. For this Master‘s study, data from the first 102 participants were used. Data for this study were collected early in pregnancy (<18 week‘s gestation) and at birth. The birth data were collected by the study mid-wife. Maternal fish consumption during early pregnancy was measured using a Quantitative Food Frequency Questionnaire (QFFQ). Correlation analysis was used to examine the association between maternal fish consumption during early pregnancy and neonatal anthropometry (birth weight, crown heel length and head circumference) and gestational age at birth. Results: Majority (88.1%) of the mothers were black South Africans between the ages of 18 and 39 with a mean age of 28 ± 5 years. At enrolment, the mean BMI of the women was 27.8±5.8kg/m2 having a mean height of 158.8±6.7cm and a mean weight of 70.4±15.2kg. Most of them were unmarried (45.4%), living in households of 2 – 5 members (86.3%), wage-earning (44.6%) and had Grade 11 or 12 schooling (58.4%). Most (76.5%) of the pregnant women consumed fish rarely (once a month) and the overall median fish intake was 4.8g/day (0; 25). In the study sample 12.5% of new-borns had a low birth weight (<2500g), the percentages of preterm births were 1.0% - extremely preterm (<28 weeks), 2.0% - very preterm (28 – <32 weeks) and 10.0% - moderate to late preterm (32 – 37 weeks). The mean birth weight was 2999.2±624.4g with boys having a mean birth weight of 3157.3±571g and girls at 2819±671g. The new-borns‘ mean gestational age at birth was 38.8±2.4weeks (271.6days). The percentage of new-born head circumference ≤ 31.49cm was 9.2% and the mean head circumference was 34.3±3.6cm with the boys having a mean head circumference of 34.5±2.4cm and the girls 34.1±4.3cm. In this study sample, 3.7% of new-borns were born with crown heel length of 31 – 40cm and the mean crown heel length mean was 49.5±4.6cm with the boys having a mean crown heel length of 49.8±4.9cm and the girls having mean crown heel length of 49.3±4.3cm. In this study, there were no statistically significant associations between fish consumption at early pregnancy and birth outcomes such as gestational age at birth (r=0.051; p=0.625), birth weight (r=-0.043; p=0.695) and crown heel length (r=0.008; p=0.943). There was a positive association between maternal fish consumption in early pregnancy and head circumference of the new-born which tended towards statistical significance (r=0.193; p=0.079). Conclusions: In this study of pregnant women living in Johannesburg, a few women consumed fish at early pregnancy compared with women who did not consume fish during pregnancy. We found no statistically significant association in this study between fish consumption at early pregnancy and birth outcomes. / Life and Consumer Sciences / MCS (Consumer Science)