Spatial prevalence and associations among respiratory diseases in Maine

A Bayesian spatio-temporal analysis of neighborhood pediatric asthma emergency department visit disparities

Asthma disparities have complex, neighborhood-level drivers that are not well understood. Consequently, identifying particular contextual factors that contribute to disparities is a public health goal. We study pediatric asthma emergency department (ED) visit disparities and neighborhood factors associated with them in South Carolina (SC) census tracts from 1999 to 2015. Leveraging a Bayesian framework, we identify risk clusters, spatially-varying relationships, and risk percentile-specific associations. Clusters of high risk occur in both rural and urban census tracts with high probability, with neighborhood-specific associations suggesting unique risk factors for each locale. Bayesian methods can help clarify the neighborhood drivers of health disparities.

A geographic identifier assignment algorithm with Bayesian variable selection to identify neighborhood factors associated with emergency department visit disparities for asthma

BACKGROUND

Ecologic health studies often rely on outcomes from health service utilization data that are limited by relatively coarse spatial resolutions and missing geographic information, particularly neighborhood level identifiers. When fine-scale geographic data are missing, the ramifications and strategies for addressing them are not well researched or developed. This study illustrates a novel spatio-temporal framework that combines a geographic identifier assignment (i.e., geographic imputation) algorithm with predictive Bayesian variable selection to identify neighborhood factors associated with disparities in emergency department (ED) visits for asthma.

METHODS

ED visit records with missing fine-scale spatial identifiers (~ 20%) were geocoded using information from known, coarser, misaligned spatial units using an innovative geographic identifier assignment algorithm. We then employed systematic variable selection in a spatio-temporal Bayesian hierarchical model (BHM) predictive framework within the NIMBLE package in R. Our novel methodology is illustrated in an ecologic case study aimed at identifying neighborhood-level predictors of asthma ED visits in South Carolina, United States, from 1999 to 2015. The health outcome was annual ED visit counts in small areas (i.e., census tracts) with primary diagnoses of asthma (ICD9 codes 493.XX) among children ages 5 to 19 years.

RESULTS

We maintained 96% of ED visit records for this analysis. When the algorithm used areal proportions as probabilities for assignment, which addressed differential missingness of census tract identifiers in rural areas, variable selection consistently identified significant neighborhood-level predictors of asthma ED visit risk including pharmacy proximity, average household size, and carbon monoxide interactions. Contrasted with common solutions of removing geographically incomplete records or scaling up analyses, our methodology identified critical differences in parameters estimated, predictors selected, and inferences. We posit that the differences were attributable to improved data resolution, resulting in greater power and less bias. Importantly, without this methodology, we would have inaccurately identified predictors of risk for asthma ED visits, particularly in rural areas.

CONCLUSIONS

Our approach innovatively addressed several issues in ecologic health studies, including missing small-area geographic information, multiple correlated neighborhood covariates, and multiscale unmeasured confounding factors. Our methodology could be widely applied to other small-area studies, useful to a range of researchers throughout the world.

The Use of Geographic Data to Improve Asthma Care Delivery and Population Health

The authors examine uses of geographic data to improve asthma care delivery and population health and describe potential practice changes and areas for future research.

Practical utility of general practice data capture and spatial analysis for understanding COPD and asthma

BACKGROUND

General practice-based (GP) healthcare data have promise, when systematically collected, to support estimating local rates of chronic obstructive pulmonary disease (COPD) and asthma, variations in burden of disease, risk factors and comorbid conditions, and disease management and quality of care. The use of GP information systems for health improvement has been limited, however, in the scope and quality of data. This study assessed the practical utility of de-identified clinical databases for estimating local rates of COPD and asthma. We compared COPD and asthma rates to national benchmarks, examined health related risk factors and co-morbidities as correlates of COPD and asthma, and assessed spatial patterns in prevalence estimates at the small-area level.

METHODS

Data were extracted from five GP databases in western Adelaide, South Australia, for active patients residing in the region between 2012 and 2014. Prevalence estimates were computed at the statistical area 1 (SA1) spatial unit level using the empirical Bayes estimation approach. Descriptive analyses included summary statistics, spatial indices and mapping of geographic patterns. Bivariate associations were assessed, and disease profiles investigated to ascertain multi-morbidities. Multilevel logistic regression models were fitted, accounting for individual covariates including the number of comorbid conditions to assess the influence of area-level socio-economic status (SES).

RESULTS

For 33,725 active patients, prevalence estimates were 3.4% for COPD and 10.3% for asthma, 0.8% higher and 0.5% lower for COPD and asthma, respectively, against 2014-15 National Health Survey (NHS) benchmarks. Age-specific comparisons showed discrepancies for COPD in the '64 years or less' and 'age 65 and up' age groups, and for asthma in the '15-25 years' and '75 years and up' age groups. Analyses confirmed associations with individual-level factors, co-morbid conditions, and area-level SES. Geographic aggregation was seen for COPD and asthma, with clustering around GP clinics and health care centres. Spatial patterns were inversely related to area-level SES.

CONCLUSION

GP-based data capture and analysis has a clear potential to support research for improved patient outcomes for COPD and asthma via knowledge of geographic variability and its correlates, and how local prevalence estimates differ from NHS benchmarks for vulnerable age-groups.

Brief adult respiratory system health status scale-community version (BARSHSS-CV): developing and evaluating the reliability and validity

Background

The evaluation of respiratory system health status in hospitalized patients is usually based on many laboratory examinations and imaging examinations. Medical examinations require a lot of manpower, material resources, financial resources, and may cause a certain degree of mechanical damage and radiation damage. It is not easily used widely and economically to assess the respiratory health status of community adults. Therefore, researchers developed a brief adult respiratory system health status scale-community version (BARSHSS-CV) and tested its reliability and validity.

Methods

Using clinical characteristics and pathogenic factors of respiratory system diseases as a theoretical basis and through reference to relevant literature, researchers developed an initial scale. A randomized cluster sampling strategy was used to recruit adults in the communities of Baoding City, Shijiazhuang City, Cangzhou city and Chifeng City in China. Researchers randomly selected 1 district from each city. Subsequently, 4 communities were respectively randomly selected from 4 districts. Then, researchers conducted the questionnaire survey in 4 communities. Finally, researchers investigated 615 community adults. 584 valid questionnaires were recovered. By applying exploratory factor analysis, confirmatory factor analysis, content validity index, Cronbach’s α coefficient, mean inter-item correlation coefficient and test-retest reliability, researchers tested the reliability and validity of scale and created the final BARSHSS-CV.

Results

BARSHSS-CV Cronbach’s α=0.951, content validity = 0.933, test-retest reliability = 0.963 and factor cumulative contribution rate = 67.168% by exploratory factor analysis. By confirmatory factor analysis, Chi square value (χ2) was 442.117, degrees of freedom (df) was 161, Chi square value/degrees of freedom (χ2 /df) was 2.746, root-mean-square error of approximation (RMSEA) was 0.065, goodness of fit index (GFI) was 0.902, incremental fit index (IFI) was 0.955, comparative fit index (CFI) was 0.955, normed fit index (NFI) was 0.931, Tueker-Lewis index (TLI) was 0.947. BARSHSS-CV consisted of 20 items and 3 dimensions.

Conclusions

BARSHSS-CV with good test-retest reliability and content/construct validity is a brief and economical tool for assessing the state of respiratory system amongst adult communities. BARSHSS-CV may help medical staff in community primary medical institutions quickly, conveniently and economically assess the status of respiratory system and the main problems of respiratory system in community adults.

Electronic supplementary material

The online version of this article (10.1186/s12913-018-3505-z) contains supplementary material, which is available to authorized users.

Trends and area variations in Potentially Preventable Admissions for COPD in Spain (2002-2013): a significant decline and convergence between areas

BACKGROUND

Potentially Preventable Hospitalizations (PPH) are hospital admissions for conditions which are preventable with timely and appropriate outpatient care being Chronic Obstructive Pulmonary Disease (COPD) admissions one of the most relevant PPH. We estimate the population age-sex standardized relative risk of admission for COPD-PPH by year and area of residence in the Spanish National Health System (sNHS) during the period 2002-2013.

METHODS

The study was conducted in the 203 Hospital Service Areas of the sNHS, using the 2002 to 2013 hospital admissions for a COPD-PPH condition of patients aged 20 and over. We use conventional small area variation statistics and a Bayesian hierarchical approach to model the different risk structures of dependence in both space and time.

RESULTS

COPD-PPH admissions declined from 24.5 to 15.5 per 10,000 persons-year (Men: from 40.6 to 25.1; Women: from 9.1 to 6.4). The relative risk declined from 1.19 (19 % above 2002-2013 average) in 2002 to 0.77 (30 % below average) in 2013. Both the starting point and the slope were different for the different regions. Variation among admission rates between extreme areas dropped from 6.7 times higher in 2002 to 4.6 times higher in 2013.

CONCLUSIONS

COPD-PPH conditions in Spain have undergone a strong decline and a reduction in geographical variation in the last 12 years, suggesting a general improvement in health policies and health care over time. Variability among areas still remains, with a substantial room for improvement.