Abstract
BACKGROUND: An obesity paradox has been described in relation to adverse clinical outcomes (e.g., mortality) with lower body mass index (BMI). AIMS: We sought to evaluate the association between BMI and weight loss with long-term all-cause mortality in adult populations under the care of family physicians. METHODS: LIPIDOGRAM studies were conducted in primary care in Poland in 2004, 2006, and 2015 and enrolled a total of 45,615 patients. The LIPIDOGRAM Plus study included 1627 patients recruited in the LIPIDOGRAM 2004 and repeated measurements in 2006 edition. Patients were classified by BMI categories as underweight, normal weight, overweight and class I, II, or III (obesity). Follow-up data up to December 2021 were obtained from the Central Statistical Office. Differences in all-cause mortality were analyzed using Kaplan‒Meier and Cox regression analyses. RESULTS: Of 45,615 patients, 10,987 (24.1%) were normal weight, 320 (0.7%) were underweight, 19,134 (41.9%) were overweight, and 15,174 (33.2%) lived with obesity. Follow-up was available for 44,620 patients (97.8%, median duration 15.3 years, 61.7% females). In the crude analysis, long-term all-cause mortality was lowest for the normal-weight group (14%) compared with other categories. After adjusting for comorbidities, the highest risk of death was observed for the class III obesity and underweight categories (hazard ratio, HR 1.79, 95% CI [1.55-2.05] and HR 1.57, 95% CI [1.22-2.04]), respectively. The LIPIDOGRAM Plus analysis revealed that a decrease in body weight (by 5 and 10%) over 2 years was associated with a significantly increased risk of death during long-term follow-up-HR 1.45 (95% CI 1.05-2.02, p = 0.03) and HR 1.67 (95% CI 1.02-2.74, p < 0.001). Patients who experienced weight loss were older and more burdened with comorbidities. CONCLUSIONS: Being underweight, overweight or obese is associated with a higher mortality risk in a population of patients in primary care. Patients who lost weight were older and more burdened with cardiometabolic diseases, which may suggest unintentional weight loss, and were at higher risk of death in the long-term follow-up. In nonsmoking patients without comorbidities, the lowest mortality was observed in those with a BMI < 25 kg/m(2), and no U-curve relationship was observed.