Falko Tesch

Burden and risk factors of ambulatory or hospitalized CAP: A population based cohort study

BACKGROUND Data on incidence, risk factors and outcome of community-acquired pneumonia (CAP) including outpatients is sparse. METHODS We conducted a cohort study on 1.837.080 adults insured by a German statutory health insurance in 2010-2011. CAP was identified via ICD-10-GM codes, ambulatory cases were validated by antibiotic prescription within 7 days. Primary outcomes were incidence, hospitalisation and 30-day all-cause mortality. Evaluated risk factors included age, sex and comorbidities. Evaluation was done by multivariate regression analysis adjusting for these factors and health care utilization. RESULTS CAP incidence was 9.7 per 1000 person years, hospitalisation rate 46.5%, and 30-day mortality 12.9%. 30-day mortality of ambulatory cases was 5% (with 27% subsequently hospitalized for another diagnosis before death). 30-day mortality of hospitalized patients was 21.9%, but in-hospital mortality 17.2%. Risk factors for CAP included age, male sex and all evaluated comorbidities with highest risk for neurologic (OR 2.4), lung (OR 2.3) or immunosuppressive (OR 2.1) disease. Mortality risk was highest for neurologic (OR 2.3) and malignant (OR 2.0) disease. CONCLUSIONS CAP constitutes a major burden in terms of incidence, morbidity and all-cause mortality in hospitalized and ambulatory patients. Interventions to raise awareness for disease impact also in ambulatory patients with risk factors are warranted.

Increased attention‐deficit/hyperactivity symptoms in atopic dermatitis are associated with history of antihistamine use

Epidemiologic evidence indicates a relevant association between atopic dermatitis (AD) and attention‐deficit/hyperactivity disorder (ADHD). Underlying mechanisms and ways to best identify subgroups of AD patients at risk for ADHD are poorly understood.

Outdoor air pollution, greenspace, and incidence of ADHD: A semi-individual study

BACKGROUND Attention deficit hyperactivity disorder (ADHD) is a frequently occurring neurodevelopmental disorder, symptoms of which first appear in early childhood. Etiology of ADHD is not well understood. We investigated whether outdoor air pollution and greenspace affect ADHD incidence in children residing in Saxony. METHODS 66,823 children, all beneficiaries of the statutory health insurance company AOK PLUS and born between 2000 and 2004, were followed until 2014. We considered any child with at least one ICD-10-GM F90 diagnosis by a child/adolescent psychiatrist, neuropaediatrician, or psychotherapist an ADHD case. Childrens home addresses were known up to their four-digit postal code area. Population-weighted mean values of particulate matter with diameter of < 10 μm (PM10), nitrogen dioxide (NO2), and MODIS Normalized Difference Vegetation Index (NDVI) were calculated for 186 postal code areas. Associations with each exposure were assessed by two-level adjusted Poisson regression models. RESULTS 2044 children (3.06%) were diagnosed with ADHD within the observation period. An increase of PM10 and NO2 by 10 μg/m3 raised the relative risk of ADHD by a factor of 1.97 [95% CI: 1.35-2.86] and 1.32 [1.10-1.58], respectively. A 0.1-unit increase in NDVI decreased the relative risk of ADHD by a factor of 0.82 [0.68-0.98]. Better access to child/adolescent psychiatrists was the most important confounder that increased ADHD risk across all models. CONCLUSION Our results provide some evidence that air pollution might affect ADHD. Future studies with more detailed address information and better control for confounders, in particular socioeconomic status and parental psychopathology, should replicate the observed associations.

OP I – 6 Outdoor air pollution, greenspace and incidence of adhd in saxony: a semi-individual cohort study

Background/aim Attention deficit hyperactivity disorder (ADHD) is a frequent mental disorder with onset in childhood and largely unknown aetiology. It has been speculated that apart of genetic and lifestyle factors, the environment could also contribute to the ADHD onset but evidence is very scarce. We investigated whether outdoor air pollution and greenspace affect ADHD incidence in children residing in Saxony. Methods 66 823 children from the population-based statutory health insurance company AOK PLUS born between 2000 and 2004 were followed until 2014. We considered any child with at least one ICD-10-GM F90 diagnosis ever by a child/adolescent psychiatrist, neuropaediatrician or psychotherapist an ADHD case. Children’s home addresses were known up to their four-digit postal code area. Population-weighted mean values of particulate matter with diameter of less than 10 µm (PM10), nitrogen dioxide (NO2) and MODIS Normalised Difference Vegetation Index (NDVI) were calculated for 186 postal code areas. Associations with each exposure were assessed by two-level adjusted Poisson regression models. Results 2044 children (3.06%) were diagnosed with ADHD within the observation period. Pearson correlation coefficients between PM10, NO2 and NDVI were ≥|0.80|. An increase of PM10 and NO2 by 10 µg/m³ raised the relative risk of ADHD by a factor of 1.97 [95% CI: 1.35–2.86] and 1.32 [1.10–1.58], respectively. 0.1-unit increase in NDVI decreased the relative risk of ADHD by a factor of 0.82 [0.68–0.98]. Better access to child psychiatrists was the most important confounder that increased ADHD risk across all models. Conclusion Our results provide some evidence that environmental factors might affect ADHD. Future studies with more detailed address information and better control for confounding, in particular for socioeconomic status and parental psychopathology, should clarify whether the observed associations are true.

Associations of myeloid hematological diseases of the elderly with Osteoporosis: A longitudinal analysis of routine health care data

BACKGROUND Myelodysplastic Syndromes (MDS) and Acute Myeloid Leukemia (AML) are hematological stem cell diseases mainly of the elderly. Studies indicate a close relationship between bone metabolism and hematopoietic stem cells within the osteo-hematopoietic niche. However, it remains unclear how the disturbed interaction within the osteo-hematopoietic niche affects bone homeostasis in MDS and AML patients. METHODS We utilized data of a large German statutory health insurance of approximately 2 million persons living in the German federal state of Saxony. Applying case definitions based on diagnosis, procedures and prescriptions we identified prevalent and incident cases with MDS, AML and osteoporosis (OSP) in persons aged ≥60 years. We applied time-to-event analyses to determine the relationship of MDS and AML with OSP with a specific focus on temporality. RESULTS Among all individuals aged ≥60 years (n = 891,095), 2.62% (n = 23,326), 0.14% (n = 1219) and 0.10% (n = 893) were identified with incident OSP, MDS and AML, respectively. The risk of incident OSP was significantly increased in patients with prevalent MDS (sex and age-adjusted model: HR = 1.87, 95%CI: 1.51-2.23). Conversely, patients with prevalent OSP had an increased risk to be diagnosed with incident MDS in the adjusted model (HR = 1.42, 1.19-1.65). For AML no significant associations were observed (adjusted models: inc. OSP with pre. AML; HR = 1.06, 0.65-1.47; inc. AML with pre. OSP; HR = 0.82, 0.41-1.23). DISCUSSION Our results could indicate a clinically relevant relationship between MDS and OSP in elderly patients, most likely resulting from a disturbed microenvironment within the osteo-hematopoietic niche. An alternative, non-causal explanation that MDS is caused by the medication prescribed for OSP can be partially ruled out, as the association between the two diseases remains if incident OSP cases are considered in patients with pre-existing MDS. These results need to be confirmed within other prospective studies and may allow then for comprehensive strategies for the prevention, early detection and clinical care of patients with MDS and OSP.

Do Changes in Bar Opening Hours Influence Violence in the Night? Evidence from 13 Bavarian Towns:

Changes in opening hours for on-premise drinking places may influence the level of alcohol-related violence in two ways. The increased availability of alcohol increases the risks associated with it, while restrictive opening hours may produce more occasions for crime due to overcrowding at closing time. We use a quasi-experimental design with data from 13 Bavarian towns with and without restrictive opening hours and a negative binomial panel model. The outcome measure is violent incidents reported by the police between 2:00 a.m. and 6:00 a.m. over the period of 12 years. Incidents at night disproportionally increase over the study period in the sample. After controlling for daily violence as well as for an interaction between policy regime and violence level, we find that restricted opening hours are only beneficial within settings of a low daily violence level.