BACKGROUND: Air pollution causes many diseases and deaths. It is important to see how air pollution affects obesity, which is common worldwide. Therefore, we analyzed data from a smartphone application for intentional weight loss, and then we validated them. METHODS: Our analysis was structured in two parts. We analyzed data from a cohort registered to a smartphone application in 10 large cities of the world and matched it with the annual pollution values. We validated these results using daily pollution data in United States and matching them with user information. Body mass index (BMI) variation between final and initial login time was considered as outcome in the first part, and daily BMI in the validation. We analyzed: daily calories intake, daily weight, daily physical activity, geographical coordinates, seasons, age, gender. Weather Underground application programming interface provided daily climatic values. Annual and daily values of particulate matter PM10 and PM2.5 were extracted. In the first part of the analysis, we used 2,608 users and then 995 users located in United States. RESULTS: Air pollution was highest in Seoul and lowest in Detroit. Users decreased BMI by 2.14 kg/m2 in average (95% confidence interval, −2.26 to −2.04). From a multilevel model, PM10 (β=0.04, P=0.002) and PM2.5 (β=0.08, P < 0.001) had a significant negative effect on weight loss when collected per year. The results were confirmed with the validation (βAQI*time=1.5×10–5; P < 0.001) by mixed effects model. CONCLUSION: This is the first study that shows how air pollution affects intentional weight loss applied on wider area of the world.
Air Pollution* ; Body Mass Index ; Cohort Studies ; Mobile Applications ; Motor Activity ; Obesity ; Particulate Matter ; Seasons ; Seoul ; Smartphone ; United States ; Weather ; Weight Loss

BACKGROUND: There have been equivocal results in studies of the effects of dipeptidyl peptidase-4 inhibitors (DPP-4i) on fractures. In this study, we analyzed the effect of DPP-4i on bone fracture risk in a Korean population. METHODS: We extracted subjects (n = 11,164) aged 50 years or older from the National Health Insurance Service–National Sample Cohort 2.0 from 2009 to 2014. Our control group included subjects without diabetes (n = 5,582), and our treatment groups with diabetes included DPP-4i users (n = 1,410) and DPP-4i non-users (n = 4,172). The primary endpoint was the incidence of a composite outcome consisting of osteoporosis diagnosis, osteoporotic fractures, vertebral fractures, non-vertebral fractures, and femoral fractures. The secondary endpoint was the incidence of each individual component of the composite outcome. Survival analysis was performed with adjustment for age, gender, diabetes complications severity index, Charlson comorbidity index, hypertension medication, and dyslipidemia treatment. RESULTS: The incidence of the composite outcome per 1,000 person-years was 0.089 in DPP-4i users, 0.099 in DPP-4i non-users, and 0.095 in controls. There was no significant difference in fracture risk between DPP-4i users and DPP-4i non-users or controls after the adjustments (P > 0.05). The incidences of osteoporosis diagnosis, osteoporotic fractures, vertebral fractures, non-vertebral fractures, and femoral fractures were not significantly different between DPP-4i users and non-users. The results of subgroup analyses by gender and age were consistent. CONCLUSION: DPP-4i had no significant effect on the risk of fractures in a Korean population.
Cohort Studies ; Comorbidity ; Diabetes Complications ; Diabetes Mellitus, Type 2 ; Diagnosis ; Dipeptidyl-Peptidase IV Inhibitors ; Dyslipidemias ; Femoral Fractures ; Fractures, Bone ; Hypertension ; Incidence ; National Health Programs ; Osteoporosis ; Osteoporotic Fractures