Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Health-Related Disparities in the Metropolitan Region Ruhr: Large-Scale Spatial Model of Local Asthma Prevalence, Accessibility of Health Facilities, and Socioeconomic and Environmental Factors

Health-Related Disparities in the Metropolitan Region Ruhr: Large-Scale Spatial Model of Local... This paper investigates the area of the Metropole Ruhr in terms of spatial distributions of environmental factors that can prevent or cause a significantly lower or higher rate of respiratory diseases such as asthma. Environmental factors can have negative impact, like air pollution, and positive, like the access to urban green areas. In the second part of the analysis, the accessibility of pharmacies, hospitals, and medical facilities that oe ff r a special treatment for people with respiratory diseases will be spatially analysed and associated to those detected urban areas of higher and lower prevalence. The results of both approaches are spatially blended with socioeconomic and socio-demographic values of the respective residents. With this it is possible to point out whether accessibility of health facilities is a suitable and equitable for all people diagnosed with asthma regardless of their educational or migration background, their employment rate, salary or age. Consequently, all values will be disaggregated from large spatial units, such as city districts municipalities or neighbourhoods, to small city blocks, to assess large-scale spatial variability. This provides the opportunity of a point-by-point investigation and statistical analysis with a high level of detail that significantly exceeds previous study results. In the sociological context of environmental justice this highly interdisciplinary study contributes to the assessment of fair health conditions for people in densely populated conurbations. Keywords Environmental justice · Equal health · Disaggregation · Network analysis · Asthma · Prevalence Zusammenfassung Gesundheitsbezogene Disparitäten in der Metropolregion Ruhr: Großräumiges Modell der lokalen Asthma-Prävalenz, Erreichbarkeit von Gesundheitseinrichtungen sowie sozioökonomischen und Umweltfaktoren. Dieser Beitrag untersucht das Gebiet der Metropole Ruhr hinsichtlich der räumlichen Verteilung von Umweltfaktoren, die eine signifikant niedrigere oder höhere Rate von Lungenkrankheiten wie Asthma oder Bronchitis verhindern oder verursachen können. Dabei auftretende Umweltfaktoren können negativ konnotiert sein, wie z.B. Luftverschmutzung, aber auch positive Eigenschaften haben, wie der Zugang zu städtischen Grünflächen. Im zweiten Teil der Analyse wird die Erreichbarkeit von Apotheken, Krankenhäusern und Arztpraxen, die eine spezielle Behandlung für Menschen mit Lungenkrankheiten anbieten, räumlich analysiert und mit den ermittelten städtischen Gebieten mit höherer Prävalenz verknüpft. Die Ergebnisse beider Ansätze werden räumlich mit sozioökonomischen und soziodemographischen Charakteristiken der jeweiligen Bewohner zusammengeführt. Damit wird aufgezeigt, ob ein angemessener und gerechter Zugang zu lungenspezifischen Gesundheitseinrichtungen für alle Menschen besteht, unabhängig von ihrem Bildungs- oder Migrationshintergrund, ihrer Erwerbsquote, ihrem Einkommen oder ihrem Alter. Abschließend werden alle Werte von den zumeist großen räumlichen Einheiten, wie Stadtteilen oder Nachbarschaften, lokal gewichtet in kleine Stadtblöcke disaggregiert. Dies bietet die Möglichkeit einer punktuellen Untersuchung mit hohem * Annette Ortwein annette.ortwein@uni-marburg.de Philipps University Marburg, Institute for Health Services Research and Clinical Epidemiology, Marburg, Germany Ruhr University Bochum, Geomatics Research Group, Bochum, Germany Vol.:(0123456789) 1 3 474 PFG (2022) 90:473–490 Detailgrad, der deutlich über bisherige Studienergebnisse hinausgeht. Im soziologischen Kontext der Umweltgerechtigkeit leistet diese hochgradig interdisziplinäre Studie einen wichtigen Beitrag zur Beurteilung der gesundheitlichen Chancen- gleichheit von Menschen in dicht besiedelten Ballungsräumen. 1 Introduction Fachgesellschaften 2020). Gender disparities were found to vary among age groups. While boys are more likely to be With ongoing urbanization and the adoption of modern life- affected than girls, this relationship reverses into its opposite styles worldwide, the global burden of disease will most after puberty, leading to the assumption that sex hormones likely increase in the near future (Masoli et al. 2004), and play an important role in developing asthma (Carey et al. will not be distributed equally across regions. There is a 2007; Fuseini and Newcomb 2017). need for high-resolution comparisons to identify populations A higher body mass index (BMI) is associated with a particularly at risk and/or susceptible to adverse environ- higher risk of asthma in adults. As the BMI was found to mental and access-related factors within the cities to guide be an inadequate measure for children, fat mass measures policy makers by identifying local health disparities in exhibit a similar relationship in children (Guibas et al. 2013). densely populated areas. This paper will address identify- Higher asthma prevalence can be found in groups with ing areas of health disparities and inequity based on access low socioeconomic status, while allergies as a common to healthcare and environmental factors for asthma patients comorbidity potentially influencing asthma were associated in the Metropole Ruhr. with higher socioeconomic status. Laussmann et al. (2012: 310) state that children living in rural areas or smaller cities 1.1 Definition and Epidemiology of Asthma suffered less often from asthma, which is suspected to be caused, inter alia, by increased air pollution in cities. Fur- Asthma bronchiale is a noncommunicable chronic respira- thermore, exposition to traffic and congestion could influ- tory disease affecting approximately 262 million people ence triggering of respiratory diseases (Nowak and Mutius worldwide (WHO 2021). Asthma symptoms like wheezing, 2004: 511). cough, chest tightness and shortness of breath are caused by chronic inflammation and narrowing of the air passages 1.2 Environmental Risk Factors and Beneficiaries (NVL 2020; WHO 2021). Due to its high prevalence, asthma is one of the major noncommunicable chronic diseases While allergies like rhinitis and eczema, smoking, and obe- (WHO 2021). No single universal definition of asthma in sity are considered risk factors for both developing asthma epidemiological studies has been agreed upon (Pekkanen and exacerbations, other factors are unique to either condi- and Pearce 1999; Toelle et al. 1992). For the purpose of this tion. Thus, distinguishing between exposure to risk factors study and in accordance with the DEGS1, KiGGS Wave for acute exacerbation of asthma and developing asthma is 2, and WiDO studies, the definition of Asthma prevalence necessary for evaluating risk factors, beneficiaries, and dis- is limited to 12-month asthma prevalence with prescribed ease burden. For asthma patients, experiencing symptoms, medication (Robert Koch Institute 2015, 2019; Wissen- being exposed to triggers, and exacerbations can result in schaftliches Institut der AOK 2020). a lower overall quality of life as well as negative effects According to GEDA, the 12-month prevalence of the on social interactions, limitations of activities, and reduced adult population in Germany was 6.2% as captured via self- productivity (Stanescu et al. 2019). disclosure of the participants, while in DEGS1, 5% of the participants reported a diagnosis of asthma. Secondary data 1.2.1 Air Quality of the statutory health insurance show a 5.9% prevalence of diagnosed asthma in adults. KiGGS Wave 2 and statu- Asthma patients are particularly at risk regarding nega- tory health insurance data show similar prevalence of 4.0% tive impact of indoor and outdoor air pollutants (Bun- and 5.1%, respectively, in children and adolescents (German desärztekammer (BÄK), Kassenärztliche Bundesvereini- National Cohort (GNC) Consortium 2014; Hoffmann 2007; gung (KBV), Arbeitsgemeinschaft der Wissenschaftlichen Langer et al. 2020; Akmatov et al. 2018; RKI 2017; Wis- Medizinischen Fachgesellschaften 2020; Masoli et al. 2004, senschaftliches Institut der AOK). In the last 10 years, the 2004). Indoor environmental allergens with a link to asthma number of cases and the age-adjusted mortality for ICD-10 onset and/or exacerbation include moulds, house dust mites, codes J-46 (status asthmaticus) and J-45 (asthma bronchiale) and chemicals, while ozone, nitrogen dioxide and PM2.5/ decreased in all age groups in Germany (Bundesärztekam- PM10 are the most commonly studied outdoor pollutants mer (BÄK), Kassenärztliche Bundesvereinigung (KBV), (Guarnieri and Balmes 2014; WHO 2021). Guarnieri and Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Balmes (2014) state, that while direct inflammatory effects 1 3 PFG (2022) 90:473–490 475 of air pollutants on airway neuroreceptors occur at very 1.4 Access to Healthcare high concentrations not commonly experienced in Ger- many, ozone, nitrogen dioxide and PM2.5 can induce airway Timely access to relevant healthcare services influences responsiveness and (allergic) inflammation at lower con- treatment outcome, quality of care, and utilization, and centrations and are associated with oxidative stress; leading inadequate access to health care has been associated with to the well-founded assumption that exposition to pollut- increased morbidity, hospitalization rates, and avoidable ants are associated with exacerbation and onset of asthma deaths in asthma patients, especially when combined with through oxidative injury to the airways. lower socioeconomic status (Bryant-Stephens 2009; CDC The short-term exposure to air pollutants is associated 2018; Evans et  al. 1999; Haselkorn et  al. 2008; Jones with an increased number of (emergency) hospitalization. and Bentham 1997; Levy et al. 2006; Strunk et al. 2002). A 10 µg/m increase in PM2.5 was associated with an 1.5% According to German clinical practice guidelines, disease increase in risk of emergency admission (Bundesärztekam- management, therapy, symptoms, and adherence should mer (BÄK), Kassenärztliche Bundesvereinigung (KBV), be controlled regularly, to make adjustments as necessary Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen (Bundesärztekammer (BÄK), Kassenärztliche Bundesver- Fachgesellschaften 2020). Due to potential confounders, this einigung (KBV), Arbeitsgemeinschaft der Wissenschaftli- relationship might not be causal for lower concentrations chen Medizinischen Fachgesellschaften 2020). Asthma costs commonly found in Germany. and future risks of severe exacerbations are linked to the Therefore, the national guideline recommends to limit patient’s asthma control level (Luskin et al. 2014). Due to exposure to the aforementioned air pollutants in occupa- limited capacities of specialists, basic diagnosis and pul- tional, indoor, and outdoor settings (Bundesärztekammer monary function testing like spirometry is often performed (BÄK), Kassenärztliche Bundesvereinigung (KBV), Arbe- by general practitioners. Structured treatment programs are itsgemeinschaft der Wissenschaftlichen Medizinischen advised to lie within the treatment scope of the general prac- Fachgesellschaften 2020). titioner or paediatrist, unless the patient’s health status is highly instable, or asthma is classified as severe. In the case 1.3 Greenness of comorbidities, consulting another specialist may be neces- sary. Hospitalisation may be indicated based on the severity Studies suggest that greenness can influence asthma both of exacerbation(s) or severe infections affecting the respira- negatively and positively. As a direct effect, urban green tory system. Along with general practitioners, pharmacists means seasonal exposure to allergens such as weed and grass are encouraged to instruct the patients in inhalator and pollen, associated with increased asthmatic symptoms (Del- medication use, and, when possible monitor adherent and laValle et al. 2012; Lovasi et al. 2013), as well as reduced air correct use (Bundesärztekammer (BÄK), Kassenärztliche pollutant concentrations and urban heat islands (Nieuwen- Bundesvereinigung (KBV), Arbeitsgemeinschaft der Wis- huijsen et al. 2017; Shanahan et al. 2015). While exposure senschaftlichen Medizinischen Fachgesellschaften 2020). to allergens can cause allergic rhinoconjunctivitis and, there- Accessibility of these health facilities and medication such fore, increase the risk of asthmatic episodes, early-life expo- as inhalers is, therefore, crucial for adequate treatment of sure is also suspected to prevent allergies and strengthen the asthma. Thus, improving the distribution of health services immune system in accordance with the hygiene and environ- and medication is an important step towards equity of health mental hypothesis (Mutius 2016; Ruokolainen et al. 2015). care (Corburn and Cohen 2012; Masoli et al. 2004). An increased tree density was also found to be associ- ated with a lower prevalence of asthma and a lower risk of asthma-caused hospitalization, although the latter was found 2 Data Base and Study Area to be insignificant after controlling for confounders (Lovasi et al. 2008). Increasing forest and agriculture cover within a The Metropolitan Region Ruhr is a densely populated region 2–5 km range has been found to be associated with less risk in North Rhine-Westphalia, located in the west of Germany. of atopic sensitization (Ruokolainen et al. 2015). In an area of 4439 km , more than 5 million inhabitants live Indirect effects of greenness on wellbeing have also been in the four districts and eleven cities, making the Metropoli- taken into consideration by multiple studies. An increase in tan Region Ruhr the largest conurbation in Germany. Settle- urban green in the surrounding living area was found to be ments make up for 29.6% of the area, and 9.5% are dedicated associated with a lower prevalence of obesity and excessive to traffic (Regionalverband Ruhr 2021) (Fig. 1). screen time in children (Dadvand et al. 2014). A great number of different datasets is combined in this study (see Table 1), either directly associated with the final results and hence providing values and attributes that help 1 3 476 PFG (2022) 90:473–490 Fig. 1 Study area and population density analysing our declared research goal, or mandatory auxil- attributes, like e.g. accessibilities to urban green or medical iary data sets, that are crucial to process some of the values facilities, within each geometry as well as the comparison of from large spatial units (source zones) to smaller ones (tar- certain geometries among each other, as the city block rep- get zones), when conducting the disaggregation (see chap- resents a homogeneous structure type within a city (Lehner ter 3.1). Almost all data sets are open source to provide a and Blaschke 2019). Additionally, it provides the possibility maximum level of reproducibility as this study is not only to detect clusters of similar value ranges and deduce urban delivering results for a sophisticated assessment of environ- hot spots of inequalities. This can then be followed by an mental justice issues in the context of asthmatic prevalence, isolated focussed investigation of selective city blocks that but also provides a workflow that enables the implementa- includes secondary attributes to evaluate the quality of life tion to other areas of investigation. Since most official Ger - and health in the respective zone and analyse its structure for man socioeconomic data sets rely on the raster-based census a detection of zones of the socioeconomically and sanitary data from 2011 and hence are not suitable to be combined deprived (Theofilou 2013). with more recent data sets that are part of this study, the To achieve that objective, the whole workflow (see Fig.  2) commercial data set is a necessary exception from the open is subdivided into three major steps, of which the second data approach. Once new census data are available, it can be one is the disaggregation of the (1) socioeconomic variables replaced due to the universal disaggregation method adopted (absolute number of inhabitants, age, gender, heritage, edu- in this paper. The datasets’ specifications and processing cation, etc.) from PLZ8 units to city block level, and (2) the will be described in the respective subsections of the next prevalence for asthmatic illnesses depicted by the absolute chapter. number of affected people divided by the number of inhab- itants per spatial unit, being transferred from county areas to again the city block level. To do this, a lot of secondary 3 Spatial Distribution Modelling data sets are necessary (Langford 2006), see also Table 1) to create two ancillary data sets in the first step (one for each The primary goal is to assign all values that are necessary to disaggregation process) that hierarchizes all target zones and evaluate the environmental, socioeconomic and health con- results in a sophisticated distribution of values from the large ditions to one large-scale spatial unit, represented in this source into the smaller target zones. study by the residential city blocks, derived from the Urba- As the last step, a raster-based network analysis is con- nAtlas. This enables the comparison of various different ducted to quantify the accessibility of urban greens on the 1 3 PFG (2022) 90:473–490 477 1 3 Table 1 Data sets used in this study, non-commercial data unless indicated otherwise (*) Data set Short description Spatial resolution (number Coverage Year(s) of refer- Source References of sample points) ence Microm socioeconomic Absolute and relative values PLZ8 spatial units Germany 2017 Microm GmbH* Microm Gmbh (2020) variables depicting various attributes of the socioeconomic status of inhabitants per respective unit Noise map Corridors of noise assigned with Polygon features referring North Rhine-Westphalia 2019 LANUVhttps:// www. lanuv. nrw. de/ consistent dB values to networks and facilities Street network streets, paths, etc. with a certain line features of various Germany 2022 openstreetmap.orghttps:// www. opens treet map. hierarchy classes org/ Footprints houses areas of buildings of all kinds Polygon features North Rhine-Westphalia 2022 Geobasis.NRW Land NRW (2022) NO -concentration sample stationary measurements of NO2- Point features (50) North Rhine-Westphalia 2019 Geobasis.NRW Land NRW (2022) points concentrations (LUQS) PM -concentration sam- stationary measurements of Point features (12) North Rhine-Westphalia 2019 Geobasis.NRW Land NRW (2022) 2,5 ple points PM2,5-concentrations (LUQS) PM -concentration sam- stationary measurements of Point features (23) North Rhine-Westphalia 2019 Geobasis.NRW Land NRW (2022) ple points PM10-concentrations (LUQS) Digital landscape model topographic and thematic objects Polygon features of vari- Germany 2019 LANUV, BKG BKG (2020) (DLM) of landscape and elevation ous classes Health atlas asthma preva- asthma prevalence per county Counties (NUTS 3) Germany 2018 AOK WiDO Wissenschaftliches Institut lences based on medically treated der AOK (2020) patients POI hospitals Addresses of hospitals Point features Germany 2019 DESTATIShttps:// www. desta tis. de/ POI pharmacies Addresses of pharmacies Point features Germany 2019 Apotheken-Umschauhttps:// www. apoth eken- umsch au. de/ apoth ekenf inder/ POI pneumologists Addresses of pneumologists Point features Study Area 2022 Googlehttps:// www. maps. google. com POI pneumologists Addresses of pneumologists Point features Study Area 2022 Netzwerk schweres https:// www. asthma. de/ exper Asthmatensu che POI pneumologists Addresses of pneumologists Point features Study Area 2022 Federation of https:// lunge natlas. de/ Pneumologists in Germany POI general practitioners Addresses of general practicion- Point features Study Area 2022 Googlehttps:// www. maps. google. ers com Sentinel-2 imagery Satellite imagery from Coperni- raster image with 10 m Germany 2019 ESA European Commission, cus' Sentinel-2 mission resolution in respective Copernicus (2020) bands UrbanAtlas high-resolution land use maps for polygon features of various Study Area 2018 Copernicus European Commission, over 300 large urban zones and classes Copernicus (2020) their surroundings Standard land value ground values and further attrib- polygon features North Rhine-Westphalia 2018 Geobasis.NRW Land NRW (2022) utes for existing real estate 478 PFG (2022) 90:473–490 one hand, and specific medical facilities such as hospitals, pharmacies, pneumologists and general practitioners on the other hand. The resulting distances are assigned as averages to each city block geometry. In doing so, every city block geometry has a specific value for each attribute derived from the disaggregation and the network analysis. Few of these attributes are then used to conduct a statistical analysis to point out the whereabouts of city blocks with boldly correlating values and to identify places of significantly high or low levels of accumulated advantages or disadvantages. 3.1 Three‑Class Dasymetric Mapping As the majority of the data sets and values introduced in chapter two is assigned to spatial units that do not match the high-resolution approach of this study, they need to be disaggregated, before incorporating them into the following analysis. A disaggregation can be conducted in many ways, of which all depend on the availability of ancillary data sets, that improve the respective accuracy the more distinct and suitable they can describe and hierarchise the target zones (Eicher and Brewer 2001; Li et al. 2016; Li and Corcoran 2011; Moos et al. 2021). In this study two different data sets are disaggregated from the bigger source zones (PLZ8, microm GmbH 2020, and municipalities, BKG 2020) to the smaller target zones (city blocks, European Commission and Copernicus 2020, see Fig. 3), while using the method of three-class dasymetric mapping that has been evaluated in various studies (Lang- ford 2006; Mennis and Hultgren 2005, 2006; Moos 2020). Furthermore, this method also has been implemented and evaluated by Burian et al. (2021) into the disaggregator, a tool for the ArcGIS Pro environment that is used in this study. Additionally, it can be transferred to any other source zone property, like e.g. rasterized census data, which can help to update this study as soon as free and contemporary data sets are available. 3.1.1 Ancillary Data There are several ways to conduct the disaggregation of intensive or extensive values from larger to smaller units, and none of them claims to depict the real world with its results. In fact, a disaggregation is always just an approxima- tion to the real state and hence does only try to reconstruct a resolution that cannot be achieved by the given data (Ken- nedy and Kennedy 2004; Schulte 2008). But within these different approaches there are huge differences concern- ing the accuracy of the subsequent disaggregation results. The application of proper ancillary data sets determines the precision and usability of the values assigned to the target zones. Hence, it is crucial to incorporate ancillary data into 1 3 Table 1 (continued) Data set Short description Spatial resolution (number Coverage Year(s) of refer- Source References of sample points) ence DEGS 1 Cross-sectional health survey of individual population data Germany 2008–2011 RKI Robert Koch Institute, adults in Germany Department of Epidemiol- ogy and Health Monitoring (2015) KiGGs Wave 2 Cross-sectional health survey individual population data Germany 2014–2017 RKI Robert Koch Institute, of children and adolescents in Department of Epidemiol- Germany ogy and Health Monitoring (2019) BIK classes Regional classes based on polygon features per Germany 2021 BIKhttps:// www. bik- gmbh. de/ relations between city and sur- municipalitycms/ regio nalda ten/ bik- roundingsregio nen PFG (2022) 90:473–490 479 Fig. 2 Workflow Fig. 3 Source zones (black outlines) and target zones (red) the disaggregation process that describes and scales the tar- 3.1.1.1 Potential Living Area The first data set that is disag- get zones in terms of their potential allocation as accurate gregated is the microm data set which contains socioeco- as possible. This can be done in different classes, from a nomic values like age distribution or educational status and single class approach (areal interpolation, Goodchild and the absolute number of inhabitants. As relative attributes, Lam 1980) up to a three class dasymetric mapping approach like percentages of inhabitants with a certain characteris- that is applied in this study and that does not only include tic, highly depend on the absolute number of inhabitants per the size and distribution of the source an target zones, but spatial unit and, however. the absolute number of inhabitants also the respective usage type and its scalability respective highly depends on the amount of people that can live in each to all other zones within the area of investigation (Langford target zone, the ancillary data should quantify the potential 2006; Li et al. 2016; Moos 2020). living area in each target zone. With this, it is possible to As in this study two different data sets are disaggregated, distinguish all target zones (city blocks) from one another two different approaches of ancillary data sets are prepared and put them into a distinct hierarchy, starting with small before performing the disaggregation of the different values houses of mixed use, that can only contain a small number with the disaggregator (Burian et al. 2021). 1 3 480 PFG (2022) 90:473–490 of households, up to single dedicated housing blocks with a propensity score 1:1 matching to prepare the dataset for lot of floors that include dozens of them. logistic regression. The variables included in the propen- To create a matching hierarchy, the target zones are first sity score are overall health status, sex, west/east/Berlin, filtered with regard to their type of use, while every target and number of persons in Household. zone that does not include at least on house with the smallest For DEGS1, this resulted in 7856 cases that were included type of residential use, no matter if mixed with e.g. industry in propensity score matching, with 131 missing cases. The or not, is excluded from the final pool of city blocks via the sampling without replacement produced 17 exact matches implementation of the digital landscape model (DLM). The (325 match tries, 94.769% rejection rate), and 192 fuzzy remaining houses inside the city blocks are then attributed matches (match tolerance 0.4, 308 match tries, 37.662% with their respective type of use—also with regard to the rejection rate), while leaving 2 unmatched due to missing DLM—and the number of floors—coming from the stand- keys. ard land values—, developed and evaluated by Moos 2020. For KiGGS, 5,840 cases were included in propensity As the second last step, each house receives a certain fac- score matching, while 9,183 cases were missing. When sam- tor which is then used to calculate the absolute size of the pling without replacement, 24 exact matches (686 match potential living area which in turn is then aggregated to city tries, 96.501% rejection rate), and 228 fuzzy matches were block level. obtained (match tolerance 0.4, 662 match tries, 65.559% This results in a data set that contains the city blocks rejection rate), and 267 observations remained unmatched as target zones with the aforementioned distinct hierarchy due to missing keys. A graph of propensity scores across that refers back to footprint size, type of use and number of treatment and comparison groups was examined, and com- floors. mon support can be assumed for both study groups. For fur- ther analysis, matched DEGS1 and KiGGS Wave 2 were 3.1.1.2 Regression Analysis A logistic regression of the out- combined in one data set. come “diagnosis of asthma within the last 12 month includ- Eight hundred twenty-v fi e cases identie fi d during the pro - ing medical prescription” was carried out to be used as the pensity score matching were included in the analysis. The ancillary dataset for disaggregation of regional prevalences variables age group, BIK, binary CASMIN, unemployed, to building blocks. For DEGS1 (Robert Koch Institute, housing in sqm, noise pollution in the last 12 month (traf- Department of Epidemiology and Health Monitoring 2015) fic), noise pollution in the last 12 month (industry), building and KiGGS Wave 2 (Robert Koch Institute, Department of type, and partner in household were identified from litera- Epidemiology and Health Monitoring 2019) data provided ture research as being relevant to the distribution of asthma by the Robert Koch Institute, propensity score matching patients and available as aggregated measures on building is performed prior to inclusion in the logistic regression blocks level. The specified logistic regression including all model. The propensity score can be used to control for variables resulted in an overall accuracy of observed vs. pre- imbalances in the study groups, predicting the exposure of a dicted values of 65.3% correct (65.2% correctly specified as subject without including the outcome by means of a logis- absence, 65.5% as presence of asthma). tic regression, to then sample controls based on similarities It is important to note that the model does not reflect (Cepeda et al. 2003: 280; Rosenbaum and Rubin 1985). The causal relationship but is an aid to model probabilities of matched subjects are more closely related regarding their belonging to one class or the other. Stepwise inclusion or distribution of covariates than randomly selected subjects, removal according to Wald test did not result in improved therefore not being independent observations (Austin 2009; model accuracy. Rubin and Thomas 2000). To create an ancillary data set that can put all city blocks This study uses this property to construct matched sam- within an administrative district into an elaborate hierarchy ples that are similar in covariates that are not part of the that provides a proper disaggregation of the prevalence for disaggregation process but are thought to have an influ- asthmatic illnesses, it is necessary to collect several differ - ence on the individual outcome, while not influencing the ent values from variables and assign them to the city blocks. aggregated measure per building block. Thus, controlling After the assignment, all values can be factorised, summa- for differences that cannot be matched in the building rized and exponentially prorated, using values that are cal- block dataset, e.g. the participant’s sex, which is evenly culated via a comprehensive regression analysis. distributed on building block level, but is likely to be Variables that are included into the equation of the regres- correlated with age and asthma in an individual. There- sion analysis and their respective origin data set are shown fore, the effect sizes of the logistic regression cannot be in Table 2. assumed to be free of confounding effects, and are thus not All variables are spatially joined with the city block to be interpreted as such. As the use of a matched test can geometries which results in a final data set that contains an result in a lower type-I error rate (Austin 2009), we use averaged value for each variable in each city block geometry. 1 3 PFG (2022) 90:473–490 481 Table 2 Variables in the Variable KiGGS Wave 2 /DEGS1  S.E Sig e Matched building equation of the regression block data source analysis Age class − 0.166 0.060 0.006 0.847 Microm* BIK region − 0.171 0.032 < 0.001 1.187 BIK class CASMIN status (educational level) − 0.043 0.041 0.296 0.958 Microm* Unemployment rate − 0.018 0.290 0.952 0.983 Microm* Potential living area per inhabitant − 0.199 0.051 < 0.001 0.820 Microm* Noise pollution level (street) 0.034 0.084 0.685 1.035 Noise map Noise pollution (industry) 0.032 0.172 0.852 1.033 Noise map Type of use (building) − 0.113 0.210 0.589 0.893 DLM Partner living in household − 0.0.91 0.236 0.699 0.913 Microm* Constant 0.997 0.556 0.073 2.710 – For each of these geometries all values x are then multiplied depict the exact value from the source zone, as there may be with their respective regression coefficient β and summed some boundary values that come from a neighbouring zone. up via the formula 3.2 Network Analysis = c + ( ∗ x ) i i (1 + e ) As a further variable that can be queried for each city i=1 block geometry, the mean distance to several different resulting in a hierarchy of values per city block that depicts areas or points are calculated and added to the respective the combination of all factors and can be used as an ancil- city block. In this study they are divided into two differ - lary data set for the follow up disaggregation of the regional ent parts—the distance to urban greens (> 1 ha and > 10 ha) prevalence data set provided by WiDO (Wissenschaftliches and the distance to pharmacies, hospitals, pneumologists Institut der AOK 2020). and general practitioners. The definition of urban green is adopted by Grunewald et al. (2017), who followed numerous 3.1.2 Disaggregation approaches, supposing that urban green is an accessible and coherent area of at least 1 ha for recreational areas and larger After the ancillary data for both data sets are prepared, in than 10 ha for larger urban green spaces. As the Metropole the final step the disaggregation of all respective values is Ruhr covers a huge area and a classical network analysis that conducted in two steps—one for each data set. The basic requires a proper network data set of the whole region would underlying operation weighs all target zones within a source be both very time and hardware consuming, for this study zone according to their respective values—coming from the the network analysis is based on a raster-based approach. ancillary data—and distributes the value from the source Besides the fact that with this approach the processing time zone apportioned due to their place in the hierarchy to all is significantly enhanced, it can also easily include the acces- target zones. sibility or distance measurement of target areas, which is not There are many cases, where the boundaries of the given a designated task in the vector-based approach (Fuglsang geometries from both the source zones as well as the target et al. 2011; Mulrooney et al. 2017). zones are not the same but intersecting each other which As an overall preparation for all data sets, a street network leads to the circumstance that the ancillary value of each data set is filtered in respect of the operationality for pedes- target zone cannot be used at large. For this issue, the ancil- trians on the one hand and for driving vehicles on the other lary value is reduced according to the relative area of the hand. After this filtering the two respective street networks target zone that is still included by the respective source are rasterised to provide the fundament for the upcoming zone. Therefore, a target zone can be split up to two or more network analysis. parts during the disaggregation process (see Fig. 4). But each of the parts gets its fraction value from the 3.2.1 Urban Green respective superior source zone and after the disaggrega- tion the single parts are again merged while adding up all The accuracy of a raster-based network analysis is highly values from each part and assign the sum to the final target dependent on the resolution of the underlying network raster zone geometry. As a consequence, the sum of all intersecting data set, as the distances from and to the locations of facili- target zones within a given source zone does not necessarily ties are calculated using the length through all respective 1 3 482 PFG (2022) 90:473–490 Fig. 4 Schematic split up and reunited target zones Fig. 5 Schematic workflow of a network analysis for accessibili- ties to urban green areas < 1 ha raster cells that are crossed in the unit of the given coordi- to use it as a recreational area (Jalkanen et al. 2020; Mark- nate system. The higher the resolution of the raster cells in evych et al. 2014; Neuvonen et al. 2007). the raster network data set, the more detailed the route is As for this approach the following calculations rely on calculated and hence the more precise the calculated route distances and not on travel times there is no need to put the distance. street network into a certain hierarchy which is why the cost For this analysis, after filtering the street network data set for crossing each raster cell is generalised and put to the is rasterised with a 10 m-resolution which implicates a dis- universal value ‘1’, which results in a cost path analysis that tance of ten meters per raster cell in each x- and y-direction focuses only on the shortest distance and not on the usage of and 14.1421 m in diagonal direction. To calculate the acces- potentially faster paths. The following path distance analy- sibilities to urban green, the DLM data set (see Table 1) is sis (or distance accumulation analysis) then calculates the filtered regarding the classes of green areas in urban space distances for each raster cell in the network to the nearest (grasslands, forests and other vegetation) and their respec- urban green and assigns the respective values to each raster tive size. After all urban green areas with common bounda- cell. All raster cells are then converted to point geometries ries have been merged, all areas smaller than one hectare are and with a mean value operation assigned to their respective dropped. The threshold of one hectare has been evaluated as overlaying city block geometry (see Fig. 5). the minimum size of an urban green space that is necessary 1 3 PFG (2022) 90:473–490 483 3.2.2 Medical Facilities 3.3 Spatial and Non‑spatial Statistics Access to emergency care is defined as driving or walk - Empirical Bayesian kriging was performed for ground- ing distance to the next hospital, as is accessibility to medi- measured NO2, PM10 and PM2.5 data from Geobasis.NRW cation by the distance to pharmacies, access to diagnostic (Land NRW 2022). The resulting continuous surface layers procedures monitoring and routine care by the distance to are regarded as a proxy for air pollution. general practitioners and specialists. Healthcare sites were Moran’s I was calculated to determine the level of spa- combined from different sources. Hospital addresses were tial autocorrelation in the modelled data set. Additionally, obtained from the DESTATIS German registry of hospi- Local bivariate relationships were calculated to determine tals and complemented by hospitals included in the open the pattern and nature of associations on a local level. To data POIs of Metropole Ruhr. Pharmacy locations were col- contrast local statistics with overall trends in ANOVA and lected through the online search tool of Apotheken Umschau MANOVA, environmental, access, and socioeconomic vari- (https://w ww.a pothe ken-u mscha u.d e/a pothe ken finder). Gen- ables were clustered based on data-inherent characteristics. eral Practitioners and pneumologists were identified through Raster-based formats i.e. air pollution layers, were clustered internet research based on google maps (maps.google.de, through an unsupervised ISO algorithm, with a minimum search strings “Allgemeinarzt”, “General Practitioner”, class size of 3, maximum groups of 5, 20 maximum itera- “Hausarzt” for GP, “Pneumologe”, “Lungenarzt”, “Lungen- tions, and a sampling interval of 10. Clusters of feature- facharzt” for specialist care), as well as specialist search on based access measures were determined by multivariate lung atlas (www.lung enatlas. de ) and network severe asthma unsupervised clustering, the optimal number of classes was (asthma.de/expertensuche), including all health sites that determined by comparing pseudo-F statistics. were located within the study area. The dataset was cleaned All reported p values are Bonferroni-corrected to account by excluding health sites with matching names (similar- for multiple testing. ity > = 90%) and addresses. Nevertheless, due to different naming conventions and multiple affiliations, not all health sites could be uniquely identified, thus duplicates cannot 4 Results be ruled out in the resulting data set. The consolidated list of health facility addresses was geocoded in R using Open The distribution of predicted asthma prevalence varies sig- Streetmap Data as street location information. nificantly within the study area (Global Moran’s I 0,055543, While the capacity of hospitals can be assessed through expected − 0,000025, z-score 82,211043, p < 0,0001). Clus- number of beds and staff, the number of practising physi- ters of significantly higher prevalence within a neighbour - cians and/or full-time equivalents cannot be determined hood of 100 building blocks are found in the metropolitan from the data sources at hand. To address these limitations, areas, forming a belt around the city centres of Duisburg, the following analyses are based on the occurrence of one Mülheim/Ruhr, Essen, Bochum, and Dortmund (west to or more physicians at a given location, and the distance to east). Asthma prevalence in these city centres is significantly the nearest health facility. lower than in the surroundings. Most clusters of low preva- The subsequent network analysis follows the same rules lence can be found towards the northwest of the study area as described in Chapter 3.2.1, except that the facilities are (see Fig. 6). Within the clusters, low outliers in high preva- points instead of polygons. This changes the one parameter lence clusters are more common than high-low-outliers. that to calculate the shortest distance it can be necessary to leave the given network raster data set, as some facilites 4.1 Accessibility of Health Facilities maybe inside of buildings and hence do not intersect with the network. In these cases, the orthogonal line from the Visualising the results of the network analysis regarding nearest street network segment to the facility point is taken the combined accessibility of hospitals and pneumologists into the calculation, adding the same raster values and reso- (see Fig. 7) reveals a distinct pattern. Before describing lution from the regular data set (10 × 10 m). this pattern, it must be stated that regions that are close to To determine classes of access to health facilities, the the border of the Metropolitan Region Ruhr (< 5 km) are unique distance bands per facility type were integrated in an excluded from the analysis (indicated by the white band unsupervised classification. It was assumed that the walking that fades out into the Ruhr area), since the network analy- distance should be the decisive factor for distances < 1000 m, sis did not incorporate facility locations outside the bound- while distances beyond were covered by vehicle distance. aries and hence could not consider that some inhabitants 3 distinct classes were determined for Metropole Ruhr, of inside the Metropolitan Region Ruhr might visit specialists high, low, and medium accessibility. beyond its borders. 1 3 484 PFG (2022) 90:473–490 Fig. 6 Asthma prevalence and air pollution clusters Fig. 7 Combined accessibilities of hospitals and pneumologists for the metropolitan region Ruhr The accessibility of both medical facilities is very good A similar picture is depicted in the following two Figs. 8 in the city centres, especially in the major cities, marked and 9. All respective city centres have an expected high rate by the biggest blue squares, which was rather expected. of accessibility to both pharmacies and general practitioners There are visible gaps, where the colours tend to be rather while the overall distribution of general practitioners is a lit- yellow and red, indicating longer distances to both facili- tle less dense which can be seen for example in the particular ties, especially in the outskirts in the densely populated dominant yellow and red colours around the centres and in centres. Certain small and medium-sized towns in the the northern part of the study area. Nevertheless, there are north-western, north-eastern and the southern parts are also certain small regions with a clear lack of pharmacies, lacking both facilities. Particularly the red band that is although these regions scale down to a few city blocks in also covered by the detail map in its southern part shows, the surrounding of the dedicated city centres. Taking a more that the distribution of specialists and hospitals for people extensive view away from the centres it becomes visible that with asthmatic illnesses is not equally distributed when in the rather rural areas of the study area like the northern, evaluating the whole area. north-western and southern parts access to basic medical 1 3 PFG (2022) 90:473–490 485 Fig. 8 Accessibility of pharmacies for the metropolitan region Ruhr Fig. 9 Accessibility of general practitioners for the metropolitan region Ruhr facilities is not as equally distributed as a densely populated differences (0.2572, 95% CI [0.1868, 0.3276], p < 0.05) conurbation like the Metropolitan Region Ruhr could indi- are smaller than between 1 and 3 (0.9592, 95% CI [0.8406, cate. Especially in the aforementioned red band shown in 1.0778], p < 0.05), but positive in both pairwise compari- the lower left corner of the detail map, evoked by missing sons. Both the mean prevalence and the mean absolute hospitals and pneumologists in the local area, this becomes number of patients with asthma is significantly higher clearly visible, as this region lacks all chosen facilities not closer to the medical facilities (Group 1–Group 2: mean only in few small parts but to a bigger extent. difference 2.430, 95% CI [1.963, 2.898], p < 0.05; Group For the combined measure of access to health facilities 1–Group 3: 5.498, 95% CI [4.710, 6.286], p < 0.05; par tial derived from data-driven multinomial clustering, ANOVA η 0.01 with adjustment for potential confounding effects shows significant differences between the 3 distinct classes of purchasing power). at the 0.05 level, with an effect size of 0.04 (partial η with adjustment for potential confounding effects of purchas - ing power, 95% CI). Between groups 1 and 2, the mean 1 3 486 PFG (2022) 90:473–490 Table 3 Air pollution cluster signatures (Means) more complex and varies from negative linear to concave to complex. Visually, the map in Fig. 10 shows a quite conclu- Class Class means sive spatial distribution of colours from the bivariate legend (n) PM 2.5 PM 10 NO2 that represents the two variables prevalence of asthmatic diseases and purchasing power. Especially the highly popu- Group 1 (1714) 11,43,713 17,90,335 25,93,830 lated parts of the Metropolitan Region Ruhr appear in yellow Group 2 (1801) 11,78,807 18,2082 29,52,017 colours, representing low values of purchasing power and Group 3 (959) 11,68,524 18,06,655 34,72,627 higher prevalence. In contrast to that, along the valley of the river Ruhr, blue colours indicate higher purchasing power 4.2 Air Pollution and mainly lower but at a few spots also high prevalence. The southern edge of the more yellow zone follows the The three groups of air quality determined by data-driven so called “socioeconomic equator” of the Ruhr area (Bogu- mil 2020; Kersting et al. 2009; Ziegler 2018) and therefore clustering feature increasing NO2, while PM 2.5 and PM 10 feature a higher mean in group 2 than group 3 (see Table 3). underlines its existence. The transition zone between the socioeconomic equator and the Ruhr valley is a narrow band For all three groups depicted in Fig.  6, the prevalence means differ significantly from each other at the 0.05 level, of colours for medium values. North-western and south-eastern rural areas are domi- but the measured effect size is very small (Tukey-HSD partial η 0.004, 95% CI [0.003, 0.005], p < 0.05 with adjustment for nated by medium values also. In the south-eastern part the city of Hagen and the valley of the river Ennepe westwards potential confounding effects of purchasing power, η 0.007, p < 0.05 without adjustment). Post-hoc-tests show that the from Hagen show lower purchasing power and higher preva- lence just as the central Metropolitan Region Ruhr. mean prevalence is lower in group 1 than in group 2 (− 0.255, 95% CI [− 0.3359, − 0.1746], p < 0.05) and 3 (− 0.450, 95% 4.4 Access to Urban Green CI –[0.5320, − 0.3681], p < 0.05). Thus, prevalence is associ- ated with the distinct air pollution patterns on a small scale. Among the 3 classes of access to urban green, group 1 has This relationship could not be confirmed on a larger scale, neither for absolute individual NO2, PM2.5 and PM10 values direct access to urban green and a high mean NDVI sur- rounding the building block, while group 2 lies in a neigh- nor for the identified clusters, as no significant associations can be reported for local analyses within 100 building blocks. bourhood with low NDVI and a comparatively high distance to green areas. Group 3 has medium access and features 4.3 Purchasing Power medium NDVI. Mean asthma prevalence is lower in group 1 compared to groups 2 (− 0.344, 95% CI [− 0.432, − 0.256], Overall, prevalence is negatively correlated with purchasing p < 0.05) and 3 (− 0.633, 95% CI [− 0.706, 0.559], p < 0.05). 2* With an effect size of 0.011 (partial η ), areas with simi- power (− 0.025, 95% CI [− 0.035; − 0.015[, p > 0.05). At the local level, the relationship between the two variables is lar access to urban green and medium NDVI are associated Fig. 10 Asthma prevalence and purchasing power bivariate 1 3 PFG (2022) 90:473–490 487 Fig. 11 Number of asthma patients and access to urban green bivariate with prevalence. When examining the spatial distribution of factor is not captured for the adult population, as movement the absolute number of patients with access to urban green patterns have not been included in the study. (Fig. 11), clusters of low prevalence and a greater distance to In the study area, air quality clusters are associated with urban green are found predominantly in city centres, clusters prevalence on a larger scale, linking higher 12 month preva- of high prevalence and a large distance can be found along lence to higher exposition to air pollutants. These findings major motorways (see Figs. 9 and 10). are in line with the literature, although it has to be noted that effect size is small and no causal relationship can be confirmed nor denied due to the scale and uncertainty of the 5 Discussion and Conclusion modelled air pollution values, as well as interdependencies with other variables and potential confounders like purchas- Expectedly, the spatial distribution of prevalence does nei- ing power. It can be stated that the air pollution measurement ther exhibit an overall pattern for the whole study area nor location grid is too coarse for establishing individual local uniform values. On local level, the spatial distribution var- links through kriging. All three pollutants are suspected to ies with purchasing power and accessibility of urban green, cause exacerbations and excess (emergency) hospitalization and the number of patients living close to a facility exceeds rates, so that exposition to air pollution should be monitored those in remote areas. The smaller the distance to the medi- closely. Thus knowing, how many inhabitants potentially cal facilities, the higher the predicted prevalence. As the suffer most from air pollution and are most likely to need definition of prevalence includes only asthma cases that are healthcare as a result, is crucial to resource planning and diagnosed and medically treated, the population at the out- should be investigated further. skirts could suffer from underdiagnosis and/or undertreat- Several studies discovered and verified a specific hidden ment due to lack of access to health facilities. In addition, role of the motorway A40 that is crossing the Metropolitan one has to bear in mind that diagnosis, coding schemes, and Region Ruhr from east to west, as it is often called the treatment varies among physicians. Therefore, the observed socioeconomic equator. This is not only due to income but prevalence based on ICD-10 codes and drug prescription also to language, land prices and other factors, and this does not necessarily cover the real regional 12-month study adds a few more to that list, while not rejecting the asthma prevalence (Masoli et al. 2004). At the same time, fact there is no true evidence that the motorway itself is the distance to urban green is lower on average in remote areas, cause (Jeworutzki et al. 2017; Kersting et al. 2009). In fact, where distance to health facilities is often greater than in the interplay of position and status is something that is a city centres. Green spaces can be indirectly beneficial for highly complex issue and there is a certain need of more not developing asthma in the first place, as allergies are studies that raise and try to answer the question whether less reported in rural areas. On the other hand, proximity to the A40 is a true cause for all those visible disparities or green spaces also leads to greater exposure to allergens for rather something that constitutes a fictional but also still patients with allergy-induced or severe asthma. It has to be physical border within a highly heterogeneous area like the noted, that early-life exposure to allergens as a beneficiary Metropolitan Region Ruhr. 1 3 488 PFG (2022) 90:473–490 Austin PC (2009) Type I error rates, coverage of confidence inter - A limitation of this study lies in the modelled nature vals, and variance estimation in propensity-score matched of the spatially disaggregated measures. Dispersity and analyses. Int J Biostat. https://doi. or g/10. 2202/ 1557- 4679. 1146 diversity will most likely to be underestimated due to BKG (2020) Verwaltungsgebiete 1:250 000 (VG250). https:// gdz. missing modelling parameters that are not available on bkg. bund. de/. Accessed 2 Mar 2022 Bogumil J (2020) Die Zukunft des Ruhrgebietes. Zukunft denken building block level. As the parameters used for model- und verantworten. Springer VS, Wiesbaden, pp 543–551 ling local asthma prevalence could not be integrated as Bryant-Stephens T (2009) Asthma disparities in urban environments. explanatory variables to further investigate the impact of J Allerg Clin Immunol 123:1199 socioeconomic disparities or traffic exposure on the out- Bundesärztekammer (BÄK), Kassenärztliche Bundesvereinigung (KBV), Arbeitsgemeinschaft der Wissenschaftlichen Medizinis- come, the analysis revealed intra-urban patterns that could chen Fachgesellschaften (2020) Nationale VersorgungsLeitlinie be directly attributed to these variables. Nevertheless, the Asthma – Langfassung, 4th edn. Bundesärztekammer (BÄK); results of the spatial disaggregation pattern analysis are in Kassenärztliche Bundesvereinigung (KBV); Arbeitsgemein- line with most associations from population-based cross- schaft der Wissenschaftlichen Medizinischen Fachgesellschaf- ten (AWMF) sectional as well as epidemiological studies, and reveal Burian J, Pászto V, Zapletal L (2021) Disaggregator—tool for aggre- potential areas of inequity among patients with asthma in gation and disaggregation of spatial data Earth Science Infor- a densely populated conurbation setting. matics. Springer, New York Carey MA, Card JW, Voltz JW, Arbes SJ, Germolec DR, Korach KS, Acknowledgements The authors would like to thank Prof. Dr. Carsten Zeldin DC (2007) It’s all about sex: gender, lung development Jürgens, Prof. Dr. med. Max Geraedts, and Prof. Dr. Jörg Bendix for and lung disease. Trends Endocrinol Metab 18:308–313. https:// their continued support. doi. org/ 10. 1016/j. tem. 2007. 08. 003 Cepeda MS, Boston R, Farrar JT, Strom BL (2003) Comparison of logistic regression versus propensity score when the number of Author Contributions All authors contributed to the study conception events is low and there are multiple confounders. Am J Epide- and design. Material preparation, data collection and analysis were miol 158:280–287. https:// doi. org/ 10. 1093/ aje/ kwg115 performed by Annette Ortwein and Nicolai Moos. First draft of the CDC (2018) Asthma as the Underlying Cause of Death. AsthmaStats. manuscript was written by Annette Ortwein and Nicolai Moos, and all https://www .cdc. go v/as thma/as thma_s tats/docum ents/ As thmS tat_ authors commented on previous versions of the manuscript. All authors Morta lity_ 2001- 2016-H. pdf. Accessed 2 Mar 2022 read and approved the final manuscript. European Commission, Copernicus (2020) Urban Atlas 2018. land. copernicus.eu/local/urban-atlas/urban-atlas-2012. Accessed 21 Funding Open Access funding enabled and organized by Projekt Feb 2022 DEAL. No additional funding was received for this study. Corburn J, Cohen AK (2012) Why we need urban health equity indi- cators: integrating science, policy, and community. PLoS Med Availability of Data and Material The datasets based on open data that 9:e1001285. https:// doi. org/ 10. 1371/ journ al. pmed. 10012 85 were generated and analysed during the current study are available Dadvand P, Villanueva CM, Font-Ribera L, Martinez D, Basagaña from the corresponding author on reasonable request. X, Belmonte J, Vrijheid M, Gražulevičienė R, Kogevinas M, Nieuwenhuijsen MJ (2014) Risks and benefits of green spaces for children: a cross-sectional study of associations with sed- Declarations entary behavior, obesity, asthma, and allergy. Environ Health Perspect 122:1329–1335. https://doi. or g/10. 1289/ ehp. 13080 38 Conflict of interest The authors declare no conflicts of interests. DellaValle CT, Triche EW, Leaderer BP, Bell ML (2012) Effects of ambient pollen concentrations on frequency and severity of Open Access This article is licensed under a Creative Commons Attri- asthma symptoms among asthmatic children. Epidemiology bution 4.0 International License, which permits use, sharing, adapta- 23:55–63. https:// doi. org/ 10. 1097/ EDE. 0b013 e3182 3b66b8 tion, distribution and reproduction in any medium or format, as long Eicher CL, Brewer CA (2001) Dasymetric mapping and areal inter- as you give appropriate credit to the original author(s) and the source, polation: implementation and evaluation. Cartogr Geogr Inf Sci provide a link to the Creative Commons licence, and indicate if changes 28:125–138. https:// doi. org/ 10. 1559/ 15230 40017 82173 727 were made. The images or other third party material in this article are Evans R, Gergen PJ, Mitchell H, Kattan M, Kercsmar C, Crain E, included in the article's Creative Commons licence, unless indicated Anderson J, Eggleston P, Malveaux FJ, Wedner H (1999) A otherwise in a credit line to the material. If material is not included in randomized clinical trial to reduce asthma morbidity among the article's Creative Commons licence and your intended use is not inner-city children: results of the National Cooperative Inner- permitted by statutory regulation or exceeds the permitted use, you will City Asthma Study. J Pediatr 135:332–338. https:// doi. org/ 10. need to obtain permission directly from the copyright holder. To view a 1016/ S0022- 3476(99) 70130-7 copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . Fuglsang M, Hansen HS, Münier B (2011) Accessibility analysis and modelling in public transport networks—a raster based approach. In: Murgante B, Gervasi O, Iglesias A, Taniar D, Apduhan BO (eds) Computational science and its applications - ICCSA 2011: International conference, Santander, Spain, References June 20–23, 2011; proceedings, vol 6782. Springer, Berlin, pp 207–224 Akmatov MK, Holstiege J, Steffen A, Bätzing J (2018) Diagnosepräva- Fuseini H, Newcomb DC (2017) Mechanisms driving gender differ - lenz und - inzidenz von Asthma bronchiale – Ergebnisse einer ences in asthma. Curr Allergy Asthma Rep 17:19. https://doi. or g/ Studie mit Versorgungsdaten aller gesetzlich Versicherten in 10. 1007/ s11882- 017- 0686-1 Deutschland (2009–2016). https:// doi. org/ 10. 20364/ VA- 18. 08 1 3 PFG (2022) 90:473–490 489 German National Cohort (GNC) Consortium (2014) The German the German Health and Examination Survey for Children and National Cohort: aims, study design and organization. Eur J Epi- Adolescents (KiGGS)). Bundesgesundheitsbl 55:308–317. https:// demiol 29:371–382. https:// doi. org/ 10. 1007/ s10654- 014- 9890-7doi. org/ 10. 1007/ S00103- 011- 1430-y Goodchild MF, Lam NS-N (1980) Areal Interpolation: a variant of the Lehner A, Blaschke T (2019) A generic classification scheme for urban traditional spatial problem. Geoprocessing 1(3):297–312 structure types. Remote Sens 11:173. https:// doi. org/ 10. 3390/ Grunewald K, Richter B, Meinel G, Herold H, Syrbe R-U (2017) Pro-rs110 20173 posal of indicators regarding the provision and accessibility of Levy M, Heffner B, Stewart T, Beeman G (2006) The efficacy of green spaces for assessing the ecosystem service “recreation in asthma case management in an urban school district in reducing the city” in Germany. Int J Biodivers Sci Ecosyst Serv Manag school absences and hospitalizations for asthma. J Sch Health 13:26–39. https:// doi. org/ 10. 1080/ 21513 732. 2017. 12833 61 76:320–324. https:// doi. org/ 10. 1111/j. 1746- 1561. 2006. 00120.x Guarnieri M, Balmes JR (2014) Outdoor air pollution and asthma. Li T, Corcoran J (2011) Testing dasymetric techniques to spatially Lancet 383:1581–1592. https:// doi. org/ 10. 1016/ S0140- 6736(14) disaggregate the regional population forecasts for South East 60617-6 Queensland. J Spat Sci 56:203–221. https:// doi.or g/10. 1080/ 14498 Guibas GV, Manios Y, Xepapadaki P, Moschonis G, Douladiris N, 596. 2011. 623343 Mavrogianni C, Papadopoulos NG (2013) The obesity-asthma Li T, Pullar D, Corcoran J, Stimson R (2016) A comparison of spatial link in different ages and the role of body mass index in its inves- disaggregation techniques as applied to population estimation for tigation: findings from the Genesis and Healthy Growth Studies. South East Queensland (SEQ). Australia. https://doi. or g/10. 4225/ Allergy 68:1298–1305. https:// doi. org/ 10. 1111/ all. 1224503/ 57E9A ECEBA 789 Haselkorn T, Lee JH, Mink DR, Weiss ST (2008) Racial disparities Lovasi GS, Quinn JW, Neckerman KM, Perzanowski MS, Rundle A in asthma-related health outcomes in severe or difficult-to-treat (2008) Children living in areas with more street trees have lower asthma. Ann Allergy Asthma Immunol 101:256–263. https:// prevalence of asthma. J Epidemiol Community Health 62:647– doi. org/ 10. 1016/ S1081- 1206(10) 60490-5 649. https:// doi. org/ 10. 1136/ jech. 2007. 071894 Hoffmann F (2007) Prevalence of asthma among German adults: Lovasi GS, O’Neil-Dunne JPM, Lu JWT, Sheehan D, Perzanowski MS, analysis of the German National Telephone Survey. J Asthma MacFaden SW, King KL, Matte T, Miller RL, Hoepner LA, Perera 44:433–436. https:// doi. org/ 10. 1080/ 02770 90070 14219 22 FP, Rundle A (2013) Urban tree canopy and asthma, wheeze, rhi- Jalkanen J, Fabritius H, Vierikko K, Moilanen A, Toivonen T (2020) nitis, and allergic sensitization to tree pollen in a New York City Analyzing fair access to urban green areas using multimodal birth cohort. Environ Health Perspect 121:494–500. https:// doi. accessibility measures and spatial prioritization. Appl Geogr org/ 10. 1289/ ehp. 12055 13 124:102320. https:// doi. org/ 10. 1016/j. apgeog. 2020. 102320 Luskin AT, Chipps BE, Rasouliyan L, Miller DP, Haselkorn T, Doren- Jeworutzki S, Knüttel K, Niemand C, Schmidt B-J, Schräpler J-P, baum A (2014) Impact of asthma exacerbations and asthma trig- Terpoorten T (2017) Wege zur Metropole Ruhr: Räumlich seg- gers on asthma-related quality of life in patients with severe or regierte Bildungsteilhabe in NRW und im Ruhrgebiet – zentrale difficult-to-treat asthma. J Allergy Clin Immunol Pract. https:// Ergebnisse, Bochumdoi. org/ 10. 1016/j. jaip. 2014. 02. 011 Jones AP, Bentham G (1997) Health service accessibility and deaths Markevych I, Tiesler CMT, Fuertes E, Romanos M, Dadvand P, Nieu- from asthma in 401 local authority districts in England and wenhuijsen MJ, Berdel D, Koletzko S, Heinrich J (2014) Access to Wales, 1988–92. Thorax 52:218–222 urban green spaces and behavioural problems in children: results Kennedy E, Kennedy LW (2004) Using dasymetric mapping for from the GINIplus and LISAplus studies. Environ Int 71:29–35. spatially aggregated crime data. J Quant Criminol 20:243–262. https:// doi. org/ 10. 1016/j. envint. 2014. 06. 002 https:// doi. org/ 10. 1023/B: JOQC. 00000 37733. 74321. 14 Masoli M, Fabian D, Holt S, Beasley R (2004) The global burden of Kersting V, Meyer C, Strohmeier P, Terpoorten T (2009) Die A 40 – asthma: executive summary of the GINA Dissemination Commit- der Sozialäquator des Ruhrgebiets. In: Prossek A, Schumacher tee report. Allergy 59:469–478. https:// doi. org/ 10. 1111/j. 1398- J (eds) Atlas der Metropole Ruhr: Vielfalt und Wandel des 9995. 2004. 00526.x Ruhrgebiets im Kartenbild. Emons, Köln Mennis J, Hultgren T (2005) Dasymetric mapping for disaggregating Land NRW (2022) Open Data - Digitale Geobasisdaten NRW. https:// coarse resolution population data. https:// cites eerx. ist. psu. edu/ www. openg eodata. nrw. de/. Accessed 21 Feb 2022viewd oc/ downl oad? doi= 10.1. 1. 496. 4768& rep= rep1& type= pdf. Langer S, Horn J, Kluttig A, Mikolajczyk R, Karrasch S, Schulz H, Accessed 2 Mar 2022 Wichmann H-E, Linseisen J, Jaeschke L, Pischon T, Fricke J, Mennis J, Hultgren T (2006) Intelligent dasymetric mapping and its Keil T, Ahrens W, Günther K, Kuß O, Schikowski T, Schmidt application to areal interpolation. Cartogr Geogr Inf Sci 33:179– B, Jöckel K-H, Michels KB, Franzke C-W, Becher H, Jagodz- 194. https:// doi. org/ 10. 1559/ 15230 40067 79077 309 inski A, Castell S, Kemmling Y, Lieb W, Waniek S, Wirkner K, microm GmbH (2020) PLZ8: NRW Region. Accessed 22 Nov 2020 Löffler M, Kaaks R, Greiser KH, Berger K, Legath N, Meinke- Moos N (2020) Soziogeomatik: Möglichkeiten und Grenzen der Ver- Franze C, Schipf S, Leitzmann M, Baurecht H, Weigl K, Ami- wendung von Erdbeobachtungsdaten und Geodaten zusammen tay E, Gottschick C (2020) Häufigkeit von Asthma bronchiale mit soziodemographischen und sozioökonomischen Daten. Dis- und Alter bei der Erstdiagnose – erste Ergebnisse der NAKO sertation, Ruhr-University Bochum Gesundheitsstudie (Occurrence of bronchial asthma and age at Moos N, Juergens C, Redecker AP (2021) Geo-spatial analysis of popu- initial asthma diagnosis-first results of the German National lation density and annual income to identify large-scale socio- Cohort). Bundesgesundheitsbl 63:397–403. https:// doi. org/ 10. demographic disparities. IJGI 10:432. https:// doi. org/ 10. 3390/ 1007/ s00103- 020- 03105-yijgi1 00704 32 Langford M (2006) Obtaining population estimates in non-census Mulrooney T, Beratan K, McGinn C, Branch B (2017) A comparison reporting zones: an evaluation of the 3-class dasymetric method. of raster-based travel time surfaces against vector-based network Comput Environ Urban Syst 30:161–180. https:// doi. or g/ 10. calculations as applied in the study of rural food deserts. Appl 1016/j. compe nvurb sys. 2004. 07. 001 Geogr 78:12–21. https:// doi. org/ 10. 1016/j. apgeog. 2016. 10. 006 Laussmann D, Haftenberger M, Langen U, Eis D (2012) Einfluss - Neuvonen M, Sievänen T, Tönnes S, Koskela T (2007) Access to green faktoren für Asthma bronchiale bei Kindern und Jugendlichen areas and the frequency of visits—a case study in Helsinki. Urban Forest Urban Green 6:235–247. https:// doi. org/ 10. 1016/j. ufug. in Deutschland. Ergebnisse der KiGGS-Studie (Determinants of 2007. 05. 003 asthma among children and adolescents in Germany. Results of 1 3 490 PFG (2022) 90:473–490 Nieuwenhuijsen MJ, Khreis H, Triguero-Mas M, Gascon M, Dadvand Schulte B (2008) Dasymetrische Modellierung der Verteilung von P (2017) Fifty shades of Green: pathway to healthy urban liv- Populationen. Semesterarbeit, Technische Fachhochschule Berlin ing. Epidemiology 28:63–71. https://doi. or g/10. 1097/ EDE. 00000 Shanahan DF, Fuller RA, Bush R, Lin BB, Gaston KJ (2015) The 00000 000549 health benefits of urban nature: how much do we need? Bioscience Nowak D, von Mutius E (2004) Asthma bronchiale im Kindes- und 65:476–485. https:// doi. org/ 10. 1093/ biosci/ biv032 Erwachsenenalter: Risikofaktoren, Diagnose, Standardtherapie Stanescu S, Kirby SE, Thomas M, Yardley L, Ainsworth B (2019) A (Bronchial asthma in children and adults: risk factors, diagnosis systematic review of psychological, physical health factors, and and standard treatment). Dtsch Med Wochenschr 129:509–516. quality of life in adult asthma. NPJ Prim Care Respir Med 29:37. https:// doi. org/ 10. 1055/s- 2004- 820540https:// doi. org/ 10. 1038/ S41533- 019- 0149-3 NVL (2020) Nationale Versorgungsleitline Asthma. https:// www. leitl Strunk RC, Ford JG, Taggart V (2002) Reducing disparities in asthma inien. de/ themen/ asthma. Accessed 17 Aug 2021 care: priorities for research–National Heart, Lung, and Blood Pekkanen J, Pearce N (1999) Defining asthma in epidemiological stud - Institute workshop report. J Allerg Clin Immunol 109:229–237. ies. Eur Respir J 14:951–957. https://doi. or g/10. 1034/j. 1399- 3003. https:// doi. org/ 10. 1067/ mai. 2002. 120950 1999. 14d37.x Theofilou P (2013) Quality of life: definition and measurement. EJOP RKI (2017) 12-month prevalence of asthma among adults in Germany. 9:150–162. https:// doi. org/ 10. 5964/ ejop. v9i1. 337 https:// doi. org/ 10. 17886/ RKI- GBE- 2017- 064 Toelle BG, Peat JK, Salome CM, Mellis CM, Woolcock AJ (1992) Robert Koch Institute, Department of Epidemiology and Health Moni- Toward a definition of asthma for epidemiology. Am Rev Respir toring (2015) German Health Interview and Examination Survey Dis 146:633–637. https:// doi. org/ 10. 1164/ ajrccm/ 146.3. 633 for Adults (DEGS1): Scientific Use File first Version von Mutius E (2016) The microbial environment and its influence on Robert Koch Institute, Department of Epidemiology and Health Moni- asthma prevention in early life. J Allergy Clin Immunol 137:680– toring (2019) The German Health Survey for Children and Ado- 689. https:// doi. org/ 10. 1016/j. jaci. 2015. 12. 1301 lescents (KiGGS Wave 2): Scientific Use File first version WHO (2021) Asthma. https:// www. who. int/ news- room/ fact- sheets/ Rosenbaum PR, Rubin DB (1985) The bias due to incomplete match-detail/ asthma. Accessed 2 Mar 2022 ing. Biometrics 41:103. https:// doi. org/ 10. 2307/ 25306 47 Wissenschaftliches Institut der AOK (WIdO) (2020) Gesundheits- Rubin DB, Thomas N (2000) Combining propensity score matching atlas Deutschland - Asthma: Verbreitung in der Bevölkerung with additional adjustments for prognostic covariates. J Am Stat Deutschlands und seinen Regionen.Ursachen, Folgen und Assoc 95:573–585. https://doi. or g/10. 1080 /01621 459. 2000 .10474 Präventionsmöglichkeiten. https:// www. wido. de/ filea dmin/ Datei 233en/ Dokum ente/ Publi katio nen_ Produ kte/ Buchr eihen/ Gesun dheit Regionalverband Ruhr (2021) Regionalstatistik. https://www .r vr.r uhr/ satlas/wido_ int_ g esundheit satlas_ as thma_1120. pdf . Accessed 02 daten- digit ales/ r egio nals t atis t ik/ news- de t ail- r egio nals t atis t ik. Mar 2022 Accessed 2 Mar 2022 Ziegler E (2018) 5. Visuelle Mehrsprachigkeit in Migrationsgesells- Ruokolainen L, von Hertzen L, Fyhrquist N, Laatikainen T, Lehtomäki chaften: monolinguale Norm vs. plurilinguale Norm. In: Lenz AN, J, Auvinen P, Karvonen AM, Hyvärinen A, Tillmann V, Niemelä Plewnia A (eds) Variation – Normen – Identitäten. De Gruyter, O, Knip M, Haahtela T, Pekkanen J, Hanski I (2015) Green areas Berlin, Germany, pp 127–156 around homes reduce atopic sensitization in children. Allergy 70:195–202. https:// doi. org/ 10. 1111/ all. 12545 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png "PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science" Springer Journals

Health-Related Disparities in the Metropolitan Region Ruhr: Large-Scale Spatial Model of Local Asthma Prevalence, Accessibility of Health Facilities, and Socioeconomic and Environmental Factors

Loading next page...
 
/lp/springer-journals/health-related-disparities-in-the-metropolitan-region-ruhr-large-scale-NWAqT9maxU
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
ISSN
2512-2789
eISSN
2512-2819
DOI
10.1007/s41064-022-00213-z
Publisher site
See Article on Publisher Site

Abstract

This paper investigates the area of the Metropole Ruhr in terms of spatial distributions of environmental factors that can prevent or cause a significantly lower or higher rate of respiratory diseases such as asthma. Environmental factors can have negative impact, like air pollution, and positive, like the access to urban green areas. In the second part of the analysis, the accessibility of pharmacies, hospitals, and medical facilities that oe ff r a special treatment for people with respiratory diseases will be spatially analysed and associated to those detected urban areas of higher and lower prevalence. The results of both approaches are spatially blended with socioeconomic and socio-demographic values of the respective residents. With this it is possible to point out whether accessibility of health facilities is a suitable and equitable for all people diagnosed with asthma regardless of their educational or migration background, their employment rate, salary or age. Consequently, all values will be disaggregated from large spatial units, such as city districts municipalities or neighbourhoods, to small city blocks, to assess large-scale spatial variability. This provides the opportunity of a point-by-point investigation and statistical analysis with a high level of detail that significantly exceeds previous study results. In the sociological context of environmental justice this highly interdisciplinary study contributes to the assessment of fair health conditions for people in densely populated conurbations. Keywords Environmental justice · Equal health · Disaggregation · Network analysis · Asthma · Prevalence Zusammenfassung Gesundheitsbezogene Disparitäten in der Metropolregion Ruhr: Großräumiges Modell der lokalen Asthma-Prävalenz, Erreichbarkeit von Gesundheitseinrichtungen sowie sozioökonomischen und Umweltfaktoren. Dieser Beitrag untersucht das Gebiet der Metropole Ruhr hinsichtlich der räumlichen Verteilung von Umweltfaktoren, die eine signifikant niedrigere oder höhere Rate von Lungenkrankheiten wie Asthma oder Bronchitis verhindern oder verursachen können. Dabei auftretende Umweltfaktoren können negativ konnotiert sein, wie z.B. Luftverschmutzung, aber auch positive Eigenschaften haben, wie der Zugang zu städtischen Grünflächen. Im zweiten Teil der Analyse wird die Erreichbarkeit von Apotheken, Krankenhäusern und Arztpraxen, die eine spezielle Behandlung für Menschen mit Lungenkrankheiten anbieten, räumlich analysiert und mit den ermittelten städtischen Gebieten mit höherer Prävalenz verknüpft. Die Ergebnisse beider Ansätze werden räumlich mit sozioökonomischen und soziodemographischen Charakteristiken der jeweiligen Bewohner zusammengeführt. Damit wird aufgezeigt, ob ein angemessener und gerechter Zugang zu lungenspezifischen Gesundheitseinrichtungen für alle Menschen besteht, unabhängig von ihrem Bildungs- oder Migrationshintergrund, ihrer Erwerbsquote, ihrem Einkommen oder ihrem Alter. Abschließend werden alle Werte von den zumeist großen räumlichen Einheiten, wie Stadtteilen oder Nachbarschaften, lokal gewichtet in kleine Stadtblöcke disaggregiert. Dies bietet die Möglichkeit einer punktuellen Untersuchung mit hohem * Annette Ortwein annette.ortwein@uni-marburg.de Philipps University Marburg, Institute for Health Services Research and Clinical Epidemiology, Marburg, Germany Ruhr University Bochum, Geomatics Research Group, Bochum, Germany Vol.:(0123456789) 1 3 474 PFG (2022) 90:473–490 Detailgrad, der deutlich über bisherige Studienergebnisse hinausgeht. Im soziologischen Kontext der Umweltgerechtigkeit leistet diese hochgradig interdisziplinäre Studie einen wichtigen Beitrag zur Beurteilung der gesundheitlichen Chancen- gleichheit von Menschen in dicht besiedelten Ballungsräumen. 1 Introduction Fachgesellschaften 2020). Gender disparities were found to vary among age groups. While boys are more likely to be With ongoing urbanization and the adoption of modern life- affected than girls, this relationship reverses into its opposite styles worldwide, the global burden of disease will most after puberty, leading to the assumption that sex hormones likely increase in the near future (Masoli et al. 2004), and play an important role in developing asthma (Carey et al. will not be distributed equally across regions. There is a 2007; Fuseini and Newcomb 2017). need for high-resolution comparisons to identify populations A higher body mass index (BMI) is associated with a particularly at risk and/or susceptible to adverse environ- higher risk of asthma in adults. As the BMI was found to mental and access-related factors within the cities to guide be an inadequate measure for children, fat mass measures policy makers by identifying local health disparities in exhibit a similar relationship in children (Guibas et al. 2013). densely populated areas. This paper will address identify- Higher asthma prevalence can be found in groups with ing areas of health disparities and inequity based on access low socioeconomic status, while allergies as a common to healthcare and environmental factors for asthma patients comorbidity potentially influencing asthma were associated in the Metropole Ruhr. with higher socioeconomic status. Laussmann et al. (2012: 310) state that children living in rural areas or smaller cities 1.1 Definition and Epidemiology of Asthma suffered less often from asthma, which is suspected to be caused, inter alia, by increased air pollution in cities. Fur- Asthma bronchiale is a noncommunicable chronic respira- thermore, exposition to traffic and congestion could influ- tory disease affecting approximately 262 million people ence triggering of respiratory diseases (Nowak and Mutius worldwide (WHO 2021). Asthma symptoms like wheezing, 2004: 511). cough, chest tightness and shortness of breath are caused by chronic inflammation and narrowing of the air passages 1.2 Environmental Risk Factors and Beneficiaries (NVL 2020; WHO 2021). Due to its high prevalence, asthma is one of the major noncommunicable chronic diseases While allergies like rhinitis and eczema, smoking, and obe- (WHO 2021). No single universal definition of asthma in sity are considered risk factors for both developing asthma epidemiological studies has been agreed upon (Pekkanen and exacerbations, other factors are unique to either condi- and Pearce 1999; Toelle et al. 1992). For the purpose of this tion. Thus, distinguishing between exposure to risk factors study and in accordance with the DEGS1, KiGGS Wave for acute exacerbation of asthma and developing asthma is 2, and WiDO studies, the definition of Asthma prevalence necessary for evaluating risk factors, beneficiaries, and dis- is limited to 12-month asthma prevalence with prescribed ease burden. For asthma patients, experiencing symptoms, medication (Robert Koch Institute 2015, 2019; Wissen- being exposed to triggers, and exacerbations can result in schaftliches Institut der AOK 2020). a lower overall quality of life as well as negative effects According to GEDA, the 12-month prevalence of the on social interactions, limitations of activities, and reduced adult population in Germany was 6.2% as captured via self- productivity (Stanescu et al. 2019). disclosure of the participants, while in DEGS1, 5% of the participants reported a diagnosis of asthma. Secondary data 1.2.1 Air Quality of the statutory health insurance show a 5.9% prevalence of diagnosed asthma in adults. KiGGS Wave 2 and statu- Asthma patients are particularly at risk regarding nega- tory health insurance data show similar prevalence of 4.0% tive impact of indoor and outdoor air pollutants (Bun- and 5.1%, respectively, in children and adolescents (German desärztekammer (BÄK), Kassenärztliche Bundesvereini- National Cohort (GNC) Consortium 2014; Hoffmann 2007; gung (KBV), Arbeitsgemeinschaft der Wissenschaftlichen Langer et al. 2020; Akmatov et al. 2018; RKI 2017; Wis- Medizinischen Fachgesellschaften 2020; Masoli et al. 2004, senschaftliches Institut der AOK). In the last 10 years, the 2004). Indoor environmental allergens with a link to asthma number of cases and the age-adjusted mortality for ICD-10 onset and/or exacerbation include moulds, house dust mites, codes J-46 (status asthmaticus) and J-45 (asthma bronchiale) and chemicals, while ozone, nitrogen dioxide and PM2.5/ decreased in all age groups in Germany (Bundesärztekam- PM10 are the most commonly studied outdoor pollutants mer (BÄK), Kassenärztliche Bundesvereinigung (KBV), (Guarnieri and Balmes 2014; WHO 2021). Guarnieri and Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Balmes (2014) state, that while direct inflammatory effects 1 3 PFG (2022) 90:473–490 475 of air pollutants on airway neuroreceptors occur at very 1.4 Access to Healthcare high concentrations not commonly experienced in Ger- many, ozone, nitrogen dioxide and PM2.5 can induce airway Timely access to relevant healthcare services influences responsiveness and (allergic) inflammation at lower con- treatment outcome, quality of care, and utilization, and centrations and are associated with oxidative stress; leading inadequate access to health care has been associated with to the well-founded assumption that exposition to pollut- increased morbidity, hospitalization rates, and avoidable ants are associated with exacerbation and onset of asthma deaths in asthma patients, especially when combined with through oxidative injury to the airways. lower socioeconomic status (Bryant-Stephens 2009; CDC The short-term exposure to air pollutants is associated 2018; Evans et  al. 1999; Haselkorn et  al. 2008; Jones with an increased number of (emergency) hospitalization. and Bentham 1997; Levy et al. 2006; Strunk et al. 2002). A 10 µg/m increase in PM2.5 was associated with an 1.5% According to German clinical practice guidelines, disease increase in risk of emergency admission (Bundesärztekam- management, therapy, symptoms, and adherence should mer (BÄK), Kassenärztliche Bundesvereinigung (KBV), be controlled regularly, to make adjustments as necessary Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen (Bundesärztekammer (BÄK), Kassenärztliche Bundesver- Fachgesellschaften 2020). Due to potential confounders, this einigung (KBV), Arbeitsgemeinschaft der Wissenschaftli- relationship might not be causal for lower concentrations chen Medizinischen Fachgesellschaften 2020). Asthma costs commonly found in Germany. and future risks of severe exacerbations are linked to the Therefore, the national guideline recommends to limit patient’s asthma control level (Luskin et al. 2014). Due to exposure to the aforementioned air pollutants in occupa- limited capacities of specialists, basic diagnosis and pul- tional, indoor, and outdoor settings (Bundesärztekammer monary function testing like spirometry is often performed (BÄK), Kassenärztliche Bundesvereinigung (KBV), Arbe- by general practitioners. Structured treatment programs are itsgemeinschaft der Wissenschaftlichen Medizinischen advised to lie within the treatment scope of the general prac- Fachgesellschaften 2020). titioner or paediatrist, unless the patient’s health status is highly instable, or asthma is classified as severe. In the case 1.3 Greenness of comorbidities, consulting another specialist may be neces- sary. Hospitalisation may be indicated based on the severity Studies suggest that greenness can influence asthma both of exacerbation(s) or severe infections affecting the respira- negatively and positively. As a direct effect, urban green tory system. Along with general practitioners, pharmacists means seasonal exposure to allergens such as weed and grass are encouraged to instruct the patients in inhalator and pollen, associated with increased asthmatic symptoms (Del- medication use, and, when possible monitor adherent and laValle et al. 2012; Lovasi et al. 2013), as well as reduced air correct use (Bundesärztekammer (BÄK), Kassenärztliche pollutant concentrations and urban heat islands (Nieuwen- Bundesvereinigung (KBV), Arbeitsgemeinschaft der Wis- huijsen et al. 2017; Shanahan et al. 2015). While exposure senschaftlichen Medizinischen Fachgesellschaften 2020). to allergens can cause allergic rhinoconjunctivitis and, there- Accessibility of these health facilities and medication such fore, increase the risk of asthmatic episodes, early-life expo- as inhalers is, therefore, crucial for adequate treatment of sure is also suspected to prevent allergies and strengthen the asthma. Thus, improving the distribution of health services immune system in accordance with the hygiene and environ- and medication is an important step towards equity of health mental hypothesis (Mutius 2016; Ruokolainen et al. 2015). care (Corburn and Cohen 2012; Masoli et al. 2004). An increased tree density was also found to be associ- ated with a lower prevalence of asthma and a lower risk of asthma-caused hospitalization, although the latter was found 2 Data Base and Study Area to be insignificant after controlling for confounders (Lovasi et al. 2008). Increasing forest and agriculture cover within a The Metropolitan Region Ruhr is a densely populated region 2–5 km range has been found to be associated with less risk in North Rhine-Westphalia, located in the west of Germany. of atopic sensitization (Ruokolainen et al. 2015). In an area of 4439 km , more than 5 million inhabitants live Indirect effects of greenness on wellbeing have also been in the four districts and eleven cities, making the Metropoli- taken into consideration by multiple studies. An increase in tan Region Ruhr the largest conurbation in Germany. Settle- urban green in the surrounding living area was found to be ments make up for 29.6% of the area, and 9.5% are dedicated associated with a lower prevalence of obesity and excessive to traffic (Regionalverband Ruhr 2021) (Fig. 1). screen time in children (Dadvand et al. 2014). A great number of different datasets is combined in this study (see Table 1), either directly associated with the final results and hence providing values and attributes that help 1 3 476 PFG (2022) 90:473–490 Fig. 1 Study area and population density analysing our declared research goal, or mandatory auxil- attributes, like e.g. accessibilities to urban green or medical iary data sets, that are crucial to process some of the values facilities, within each geometry as well as the comparison of from large spatial units (source zones) to smaller ones (tar- certain geometries among each other, as the city block rep- get zones), when conducting the disaggregation (see chap- resents a homogeneous structure type within a city (Lehner ter 3.1). Almost all data sets are open source to provide a and Blaschke 2019). Additionally, it provides the possibility maximum level of reproducibility as this study is not only to detect clusters of similar value ranges and deduce urban delivering results for a sophisticated assessment of environ- hot spots of inequalities. This can then be followed by an mental justice issues in the context of asthmatic prevalence, isolated focussed investigation of selective city blocks that but also provides a workflow that enables the implementa- includes secondary attributes to evaluate the quality of life tion to other areas of investigation. Since most official Ger - and health in the respective zone and analyse its structure for man socioeconomic data sets rely on the raster-based census a detection of zones of the socioeconomically and sanitary data from 2011 and hence are not suitable to be combined deprived (Theofilou 2013). with more recent data sets that are part of this study, the To achieve that objective, the whole workflow (see Fig.  2) commercial data set is a necessary exception from the open is subdivided into three major steps, of which the second data approach. Once new census data are available, it can be one is the disaggregation of the (1) socioeconomic variables replaced due to the universal disaggregation method adopted (absolute number of inhabitants, age, gender, heritage, edu- in this paper. The datasets’ specifications and processing cation, etc.) from PLZ8 units to city block level, and (2) the will be described in the respective subsections of the next prevalence for asthmatic illnesses depicted by the absolute chapter. number of affected people divided by the number of inhab- itants per spatial unit, being transferred from county areas to again the city block level. To do this, a lot of secondary 3 Spatial Distribution Modelling data sets are necessary (Langford 2006), see also Table 1) to create two ancillary data sets in the first step (one for each The primary goal is to assign all values that are necessary to disaggregation process) that hierarchizes all target zones and evaluate the environmental, socioeconomic and health con- results in a sophisticated distribution of values from the large ditions to one large-scale spatial unit, represented in this source into the smaller target zones. study by the residential city blocks, derived from the Urba- As the last step, a raster-based network analysis is con- nAtlas. This enables the comparison of various different ducted to quantify the accessibility of urban greens on the 1 3 PFG (2022) 90:473–490 477 1 3 Table 1 Data sets used in this study, non-commercial data unless indicated otherwise (*) Data set Short description Spatial resolution (number Coverage Year(s) of refer- Source References of sample points) ence Microm socioeconomic Absolute and relative values PLZ8 spatial units Germany 2017 Microm GmbH* Microm Gmbh (2020) variables depicting various attributes of the socioeconomic status of inhabitants per respective unit Noise map Corridors of noise assigned with Polygon features referring North Rhine-Westphalia 2019 LANUVhttps:// www. lanuv. nrw. de/ consistent dB values to networks and facilities Street network streets, paths, etc. with a certain line features of various Germany 2022 openstreetmap.orghttps:// www. opens treet map. hierarchy classes org/ Footprints houses areas of buildings of all kinds Polygon features North Rhine-Westphalia 2022 Geobasis.NRW Land NRW (2022) NO -concentration sample stationary measurements of NO2- Point features (50) North Rhine-Westphalia 2019 Geobasis.NRW Land NRW (2022) points concentrations (LUQS) PM -concentration sam- stationary measurements of Point features (12) North Rhine-Westphalia 2019 Geobasis.NRW Land NRW (2022) 2,5 ple points PM2,5-concentrations (LUQS) PM -concentration sam- stationary measurements of Point features (23) North Rhine-Westphalia 2019 Geobasis.NRW Land NRW (2022) ple points PM10-concentrations (LUQS) Digital landscape model topographic and thematic objects Polygon features of vari- Germany 2019 LANUV, BKG BKG (2020) (DLM) of landscape and elevation ous classes Health atlas asthma preva- asthma prevalence per county Counties (NUTS 3) Germany 2018 AOK WiDO Wissenschaftliches Institut lences based on medically treated der AOK (2020) patients POI hospitals Addresses of hospitals Point features Germany 2019 DESTATIShttps:// www. desta tis. de/ POI pharmacies Addresses of pharmacies Point features Germany 2019 Apotheken-Umschauhttps:// www. apoth eken- umsch au. de/ apoth ekenf inder/ POI pneumologists Addresses of pneumologists Point features Study Area 2022 Googlehttps:// www. maps. google. com POI pneumologists Addresses of pneumologists Point features Study Area 2022 Netzwerk schweres https:// www. asthma. de/ exper Asthmatensu che POI pneumologists Addresses of pneumologists Point features Study Area 2022 Federation of https:// lunge natlas. de/ Pneumologists in Germany POI general practitioners Addresses of general practicion- Point features Study Area 2022 Googlehttps:// www. maps. google. ers com Sentinel-2 imagery Satellite imagery from Coperni- raster image with 10 m Germany 2019 ESA European Commission, cus' Sentinel-2 mission resolution in respective Copernicus (2020) bands UrbanAtlas high-resolution land use maps for polygon features of various Study Area 2018 Copernicus European Commission, over 300 large urban zones and classes Copernicus (2020) their surroundings Standard land value ground values and further attrib- polygon features North Rhine-Westphalia 2018 Geobasis.NRW Land NRW (2022) utes for existing real estate 478 PFG (2022) 90:473–490 one hand, and specific medical facilities such as hospitals, pharmacies, pneumologists and general practitioners on the other hand. The resulting distances are assigned as averages to each city block geometry. In doing so, every city block geometry has a specific value for each attribute derived from the disaggregation and the network analysis. Few of these attributes are then used to conduct a statistical analysis to point out the whereabouts of city blocks with boldly correlating values and to identify places of significantly high or low levels of accumulated advantages or disadvantages. 3.1 Three‑Class Dasymetric Mapping As the majority of the data sets and values introduced in chapter two is assigned to spatial units that do not match the high-resolution approach of this study, they need to be disaggregated, before incorporating them into the following analysis. A disaggregation can be conducted in many ways, of which all depend on the availability of ancillary data sets, that improve the respective accuracy the more distinct and suitable they can describe and hierarchise the target zones (Eicher and Brewer 2001; Li et al. 2016; Li and Corcoran 2011; Moos et al. 2021). In this study two different data sets are disaggregated from the bigger source zones (PLZ8, microm GmbH 2020, and municipalities, BKG 2020) to the smaller target zones (city blocks, European Commission and Copernicus 2020, see Fig. 3), while using the method of three-class dasymetric mapping that has been evaluated in various studies (Lang- ford 2006; Mennis and Hultgren 2005, 2006; Moos 2020). Furthermore, this method also has been implemented and evaluated by Burian et al. (2021) into the disaggregator, a tool for the ArcGIS Pro environment that is used in this study. Additionally, it can be transferred to any other source zone property, like e.g. rasterized census data, which can help to update this study as soon as free and contemporary data sets are available. 3.1.1 Ancillary Data There are several ways to conduct the disaggregation of intensive or extensive values from larger to smaller units, and none of them claims to depict the real world with its results. In fact, a disaggregation is always just an approxima- tion to the real state and hence does only try to reconstruct a resolution that cannot be achieved by the given data (Ken- nedy and Kennedy 2004; Schulte 2008). But within these different approaches there are huge differences concern- ing the accuracy of the subsequent disaggregation results. The application of proper ancillary data sets determines the precision and usability of the values assigned to the target zones. Hence, it is crucial to incorporate ancillary data into 1 3 Table 1 (continued) Data set Short description Spatial resolution (number Coverage Year(s) of refer- Source References of sample points) ence DEGS 1 Cross-sectional health survey of individual population data Germany 2008–2011 RKI Robert Koch Institute, adults in Germany Department of Epidemiol- ogy and Health Monitoring (2015) KiGGs Wave 2 Cross-sectional health survey individual population data Germany 2014–2017 RKI Robert Koch Institute, of children and adolescents in Department of Epidemiol- Germany ogy and Health Monitoring (2019) BIK classes Regional classes based on polygon features per Germany 2021 BIKhttps:// www. bik- gmbh. de/ relations between city and sur- municipalitycms/ regio nalda ten/ bik- roundingsregio nen PFG (2022) 90:473–490 479 Fig. 2 Workflow Fig. 3 Source zones (black outlines) and target zones (red) the disaggregation process that describes and scales the tar- 3.1.1.1 Potential Living Area The first data set that is disag- get zones in terms of their potential allocation as accurate gregated is the microm data set which contains socioeco- as possible. This can be done in different classes, from a nomic values like age distribution or educational status and single class approach (areal interpolation, Goodchild and the absolute number of inhabitants. As relative attributes, Lam 1980) up to a three class dasymetric mapping approach like percentages of inhabitants with a certain characteris- that is applied in this study and that does not only include tic, highly depend on the absolute number of inhabitants per the size and distribution of the source an target zones, but spatial unit and, however. the absolute number of inhabitants also the respective usage type and its scalability respective highly depends on the amount of people that can live in each to all other zones within the area of investigation (Langford target zone, the ancillary data should quantify the potential 2006; Li et al. 2016; Moos 2020). living area in each target zone. With this, it is possible to As in this study two different data sets are disaggregated, distinguish all target zones (city blocks) from one another two different approaches of ancillary data sets are prepared and put them into a distinct hierarchy, starting with small before performing the disaggregation of the different values houses of mixed use, that can only contain a small number with the disaggregator (Burian et al. 2021). 1 3 480 PFG (2022) 90:473–490 of households, up to single dedicated housing blocks with a propensity score 1:1 matching to prepare the dataset for lot of floors that include dozens of them. logistic regression. The variables included in the propen- To create a matching hierarchy, the target zones are first sity score are overall health status, sex, west/east/Berlin, filtered with regard to their type of use, while every target and number of persons in Household. zone that does not include at least on house with the smallest For DEGS1, this resulted in 7856 cases that were included type of residential use, no matter if mixed with e.g. industry in propensity score matching, with 131 missing cases. The or not, is excluded from the final pool of city blocks via the sampling without replacement produced 17 exact matches implementation of the digital landscape model (DLM). The (325 match tries, 94.769% rejection rate), and 192 fuzzy remaining houses inside the city blocks are then attributed matches (match tolerance 0.4, 308 match tries, 37.662% with their respective type of use—also with regard to the rejection rate), while leaving 2 unmatched due to missing DLM—and the number of floors—coming from the stand- keys. ard land values—, developed and evaluated by Moos 2020. For KiGGS, 5,840 cases were included in propensity As the second last step, each house receives a certain fac- score matching, while 9,183 cases were missing. When sam- tor which is then used to calculate the absolute size of the pling without replacement, 24 exact matches (686 match potential living area which in turn is then aggregated to city tries, 96.501% rejection rate), and 228 fuzzy matches were block level. obtained (match tolerance 0.4, 662 match tries, 65.559% This results in a data set that contains the city blocks rejection rate), and 267 observations remained unmatched as target zones with the aforementioned distinct hierarchy due to missing keys. A graph of propensity scores across that refers back to footprint size, type of use and number of treatment and comparison groups was examined, and com- floors. mon support can be assumed for both study groups. For fur- ther analysis, matched DEGS1 and KiGGS Wave 2 were 3.1.1.2 Regression Analysis A logistic regression of the out- combined in one data set. come “diagnosis of asthma within the last 12 month includ- Eight hundred twenty-v fi e cases identie fi d during the pro - ing medical prescription” was carried out to be used as the pensity score matching were included in the analysis. The ancillary dataset for disaggregation of regional prevalences variables age group, BIK, binary CASMIN, unemployed, to building blocks. For DEGS1 (Robert Koch Institute, housing in sqm, noise pollution in the last 12 month (traf- Department of Epidemiology and Health Monitoring 2015) fic), noise pollution in the last 12 month (industry), building and KiGGS Wave 2 (Robert Koch Institute, Department of type, and partner in household were identified from litera- Epidemiology and Health Monitoring 2019) data provided ture research as being relevant to the distribution of asthma by the Robert Koch Institute, propensity score matching patients and available as aggregated measures on building is performed prior to inclusion in the logistic regression blocks level. The specified logistic regression including all model. The propensity score can be used to control for variables resulted in an overall accuracy of observed vs. pre- imbalances in the study groups, predicting the exposure of a dicted values of 65.3% correct (65.2% correctly specified as subject without including the outcome by means of a logis- absence, 65.5% as presence of asthma). tic regression, to then sample controls based on similarities It is important to note that the model does not reflect (Cepeda et al. 2003: 280; Rosenbaum and Rubin 1985). The causal relationship but is an aid to model probabilities of matched subjects are more closely related regarding their belonging to one class or the other. Stepwise inclusion or distribution of covariates than randomly selected subjects, removal according to Wald test did not result in improved therefore not being independent observations (Austin 2009; model accuracy. Rubin and Thomas 2000). To create an ancillary data set that can put all city blocks This study uses this property to construct matched sam- within an administrative district into an elaborate hierarchy ples that are similar in covariates that are not part of the that provides a proper disaggregation of the prevalence for disaggregation process but are thought to have an influ- asthmatic illnesses, it is necessary to collect several differ - ence on the individual outcome, while not influencing the ent values from variables and assign them to the city blocks. aggregated measure per building block. Thus, controlling After the assignment, all values can be factorised, summa- for differences that cannot be matched in the building rized and exponentially prorated, using values that are cal- block dataset, e.g. the participant’s sex, which is evenly culated via a comprehensive regression analysis. distributed on building block level, but is likely to be Variables that are included into the equation of the regres- correlated with age and asthma in an individual. There- sion analysis and their respective origin data set are shown fore, the effect sizes of the logistic regression cannot be in Table 2. assumed to be free of confounding effects, and are thus not All variables are spatially joined with the city block to be interpreted as such. As the use of a matched test can geometries which results in a final data set that contains an result in a lower type-I error rate (Austin 2009), we use averaged value for each variable in each city block geometry. 1 3 PFG (2022) 90:473–490 481 Table 2 Variables in the Variable KiGGS Wave 2 /DEGS1  S.E Sig e Matched building equation of the regression block data source analysis Age class − 0.166 0.060 0.006 0.847 Microm* BIK region − 0.171 0.032 < 0.001 1.187 BIK class CASMIN status (educational level) − 0.043 0.041 0.296 0.958 Microm* Unemployment rate − 0.018 0.290 0.952 0.983 Microm* Potential living area per inhabitant − 0.199 0.051 < 0.001 0.820 Microm* Noise pollution level (street) 0.034 0.084 0.685 1.035 Noise map Noise pollution (industry) 0.032 0.172 0.852 1.033 Noise map Type of use (building) − 0.113 0.210 0.589 0.893 DLM Partner living in household − 0.0.91 0.236 0.699 0.913 Microm* Constant 0.997 0.556 0.073 2.710 – For each of these geometries all values x are then multiplied depict the exact value from the source zone, as there may be with their respective regression coefficient β and summed some boundary values that come from a neighbouring zone. up via the formula 3.2 Network Analysis = c + ( ∗ x ) i i (1 + e ) As a further variable that can be queried for each city i=1 block geometry, the mean distance to several different resulting in a hierarchy of values per city block that depicts areas or points are calculated and added to the respective the combination of all factors and can be used as an ancil- city block. In this study they are divided into two differ - lary data set for the follow up disaggregation of the regional ent parts—the distance to urban greens (> 1 ha and > 10 ha) prevalence data set provided by WiDO (Wissenschaftliches and the distance to pharmacies, hospitals, pneumologists Institut der AOK 2020). and general practitioners. The definition of urban green is adopted by Grunewald et al. (2017), who followed numerous 3.1.2 Disaggregation approaches, supposing that urban green is an accessible and coherent area of at least 1 ha for recreational areas and larger After the ancillary data for both data sets are prepared, in than 10 ha for larger urban green spaces. As the Metropole the final step the disaggregation of all respective values is Ruhr covers a huge area and a classical network analysis that conducted in two steps—one for each data set. The basic requires a proper network data set of the whole region would underlying operation weighs all target zones within a source be both very time and hardware consuming, for this study zone according to their respective values—coming from the the network analysis is based on a raster-based approach. ancillary data—and distributes the value from the source Besides the fact that with this approach the processing time zone apportioned due to their place in the hierarchy to all is significantly enhanced, it can also easily include the acces- target zones. sibility or distance measurement of target areas, which is not There are many cases, where the boundaries of the given a designated task in the vector-based approach (Fuglsang geometries from both the source zones as well as the target et al. 2011; Mulrooney et al. 2017). zones are not the same but intersecting each other which As an overall preparation for all data sets, a street network leads to the circumstance that the ancillary value of each data set is filtered in respect of the operationality for pedes- target zone cannot be used at large. For this issue, the ancil- trians on the one hand and for driving vehicles on the other lary value is reduced according to the relative area of the hand. After this filtering the two respective street networks target zone that is still included by the respective source are rasterised to provide the fundament for the upcoming zone. Therefore, a target zone can be split up to two or more network analysis. parts during the disaggregation process (see Fig. 4). But each of the parts gets its fraction value from the 3.2.1 Urban Green respective superior source zone and after the disaggrega- tion the single parts are again merged while adding up all The accuracy of a raster-based network analysis is highly values from each part and assign the sum to the final target dependent on the resolution of the underlying network raster zone geometry. As a consequence, the sum of all intersecting data set, as the distances from and to the locations of facili- target zones within a given source zone does not necessarily ties are calculated using the length through all respective 1 3 482 PFG (2022) 90:473–490 Fig. 4 Schematic split up and reunited target zones Fig. 5 Schematic workflow of a network analysis for accessibili- ties to urban green areas < 1 ha raster cells that are crossed in the unit of the given coordi- to use it as a recreational area (Jalkanen et al. 2020; Mark- nate system. The higher the resolution of the raster cells in evych et al. 2014; Neuvonen et al. 2007). the raster network data set, the more detailed the route is As for this approach the following calculations rely on calculated and hence the more precise the calculated route distances and not on travel times there is no need to put the distance. street network into a certain hierarchy which is why the cost For this analysis, after filtering the street network data set for crossing each raster cell is generalised and put to the is rasterised with a 10 m-resolution which implicates a dis- universal value ‘1’, which results in a cost path analysis that tance of ten meters per raster cell in each x- and y-direction focuses only on the shortest distance and not on the usage of and 14.1421 m in diagonal direction. To calculate the acces- potentially faster paths. The following path distance analy- sibilities to urban green, the DLM data set (see Table 1) is sis (or distance accumulation analysis) then calculates the filtered regarding the classes of green areas in urban space distances for each raster cell in the network to the nearest (grasslands, forests and other vegetation) and their respec- urban green and assigns the respective values to each raster tive size. After all urban green areas with common bounda- cell. All raster cells are then converted to point geometries ries have been merged, all areas smaller than one hectare are and with a mean value operation assigned to their respective dropped. The threshold of one hectare has been evaluated as overlaying city block geometry (see Fig. 5). the minimum size of an urban green space that is necessary 1 3 PFG (2022) 90:473–490 483 3.2.2 Medical Facilities 3.3 Spatial and Non‑spatial Statistics Access to emergency care is defined as driving or walk - Empirical Bayesian kriging was performed for ground- ing distance to the next hospital, as is accessibility to medi- measured NO2, PM10 and PM2.5 data from Geobasis.NRW cation by the distance to pharmacies, access to diagnostic (Land NRW 2022). The resulting continuous surface layers procedures monitoring and routine care by the distance to are regarded as a proxy for air pollution. general practitioners and specialists. Healthcare sites were Moran’s I was calculated to determine the level of spa- combined from different sources. Hospital addresses were tial autocorrelation in the modelled data set. Additionally, obtained from the DESTATIS German registry of hospi- Local bivariate relationships were calculated to determine tals and complemented by hospitals included in the open the pattern and nature of associations on a local level. To data POIs of Metropole Ruhr. Pharmacy locations were col- contrast local statistics with overall trends in ANOVA and lected through the online search tool of Apotheken Umschau MANOVA, environmental, access, and socioeconomic vari- (https://w ww.a pothe ken-u mscha u.d e/a pothe ken finder). Gen- ables were clustered based on data-inherent characteristics. eral Practitioners and pneumologists were identified through Raster-based formats i.e. air pollution layers, were clustered internet research based on google maps (maps.google.de, through an unsupervised ISO algorithm, with a minimum search strings “Allgemeinarzt”, “General Practitioner”, class size of 3, maximum groups of 5, 20 maximum itera- “Hausarzt” for GP, “Pneumologe”, “Lungenarzt”, “Lungen- tions, and a sampling interval of 10. Clusters of feature- facharzt” for specialist care), as well as specialist search on based access measures were determined by multivariate lung atlas (www.lung enatlas. de ) and network severe asthma unsupervised clustering, the optimal number of classes was (asthma.de/expertensuche), including all health sites that determined by comparing pseudo-F statistics. were located within the study area. The dataset was cleaned All reported p values are Bonferroni-corrected to account by excluding health sites with matching names (similar- for multiple testing. ity > = 90%) and addresses. Nevertheless, due to different naming conventions and multiple affiliations, not all health sites could be uniquely identified, thus duplicates cannot 4 Results be ruled out in the resulting data set. The consolidated list of health facility addresses was geocoded in R using Open The distribution of predicted asthma prevalence varies sig- Streetmap Data as street location information. nificantly within the study area (Global Moran’s I 0,055543, While the capacity of hospitals can be assessed through expected − 0,000025, z-score 82,211043, p < 0,0001). Clus- number of beds and staff, the number of practising physi- ters of significantly higher prevalence within a neighbour - cians and/or full-time equivalents cannot be determined hood of 100 building blocks are found in the metropolitan from the data sources at hand. To address these limitations, areas, forming a belt around the city centres of Duisburg, the following analyses are based on the occurrence of one Mülheim/Ruhr, Essen, Bochum, and Dortmund (west to or more physicians at a given location, and the distance to east). Asthma prevalence in these city centres is significantly the nearest health facility. lower than in the surroundings. Most clusters of low preva- The subsequent network analysis follows the same rules lence can be found towards the northwest of the study area as described in Chapter 3.2.1, except that the facilities are (see Fig. 6). Within the clusters, low outliers in high preva- points instead of polygons. This changes the one parameter lence clusters are more common than high-low-outliers. that to calculate the shortest distance it can be necessary to leave the given network raster data set, as some facilites 4.1 Accessibility of Health Facilities maybe inside of buildings and hence do not intersect with the network. In these cases, the orthogonal line from the Visualising the results of the network analysis regarding nearest street network segment to the facility point is taken the combined accessibility of hospitals and pneumologists into the calculation, adding the same raster values and reso- (see Fig. 7) reveals a distinct pattern. Before describing lution from the regular data set (10 × 10 m). this pattern, it must be stated that regions that are close to To determine classes of access to health facilities, the the border of the Metropolitan Region Ruhr (< 5 km) are unique distance bands per facility type were integrated in an excluded from the analysis (indicated by the white band unsupervised classification. It was assumed that the walking that fades out into the Ruhr area), since the network analy- distance should be the decisive factor for distances < 1000 m, sis did not incorporate facility locations outside the bound- while distances beyond were covered by vehicle distance. aries and hence could not consider that some inhabitants 3 distinct classes were determined for Metropole Ruhr, of inside the Metropolitan Region Ruhr might visit specialists high, low, and medium accessibility. beyond its borders. 1 3 484 PFG (2022) 90:473–490 Fig. 6 Asthma prevalence and air pollution clusters Fig. 7 Combined accessibilities of hospitals and pneumologists for the metropolitan region Ruhr The accessibility of both medical facilities is very good A similar picture is depicted in the following two Figs. 8 in the city centres, especially in the major cities, marked and 9. All respective city centres have an expected high rate by the biggest blue squares, which was rather expected. of accessibility to both pharmacies and general practitioners There are visible gaps, where the colours tend to be rather while the overall distribution of general practitioners is a lit- yellow and red, indicating longer distances to both facili- tle less dense which can be seen for example in the particular ties, especially in the outskirts in the densely populated dominant yellow and red colours around the centres and in centres. Certain small and medium-sized towns in the the northern part of the study area. Nevertheless, there are north-western, north-eastern and the southern parts are also certain small regions with a clear lack of pharmacies, lacking both facilities. Particularly the red band that is although these regions scale down to a few city blocks in also covered by the detail map in its southern part shows, the surrounding of the dedicated city centres. Taking a more that the distribution of specialists and hospitals for people extensive view away from the centres it becomes visible that with asthmatic illnesses is not equally distributed when in the rather rural areas of the study area like the northern, evaluating the whole area. north-western and southern parts access to basic medical 1 3 PFG (2022) 90:473–490 485 Fig. 8 Accessibility of pharmacies for the metropolitan region Ruhr Fig. 9 Accessibility of general practitioners for the metropolitan region Ruhr facilities is not as equally distributed as a densely populated differences (0.2572, 95% CI [0.1868, 0.3276], p < 0.05) conurbation like the Metropolitan Region Ruhr could indi- are smaller than between 1 and 3 (0.9592, 95% CI [0.8406, cate. Especially in the aforementioned red band shown in 1.0778], p < 0.05), but positive in both pairwise compari- the lower left corner of the detail map, evoked by missing sons. Both the mean prevalence and the mean absolute hospitals and pneumologists in the local area, this becomes number of patients with asthma is significantly higher clearly visible, as this region lacks all chosen facilities not closer to the medical facilities (Group 1–Group 2: mean only in few small parts but to a bigger extent. difference 2.430, 95% CI [1.963, 2.898], p < 0.05; Group For the combined measure of access to health facilities 1–Group 3: 5.498, 95% CI [4.710, 6.286], p < 0.05; par tial derived from data-driven multinomial clustering, ANOVA η 0.01 with adjustment for potential confounding effects shows significant differences between the 3 distinct classes of purchasing power). at the 0.05 level, with an effect size of 0.04 (partial η with adjustment for potential confounding effects of purchas - ing power, 95% CI). Between groups 1 and 2, the mean 1 3 486 PFG (2022) 90:473–490 Table 3 Air pollution cluster signatures (Means) more complex and varies from negative linear to concave to complex. Visually, the map in Fig. 10 shows a quite conclu- Class Class means sive spatial distribution of colours from the bivariate legend (n) PM 2.5 PM 10 NO2 that represents the two variables prevalence of asthmatic diseases and purchasing power. Especially the highly popu- Group 1 (1714) 11,43,713 17,90,335 25,93,830 lated parts of the Metropolitan Region Ruhr appear in yellow Group 2 (1801) 11,78,807 18,2082 29,52,017 colours, representing low values of purchasing power and Group 3 (959) 11,68,524 18,06,655 34,72,627 higher prevalence. In contrast to that, along the valley of the river Ruhr, blue colours indicate higher purchasing power 4.2 Air Pollution and mainly lower but at a few spots also high prevalence. The southern edge of the more yellow zone follows the The three groups of air quality determined by data-driven so called “socioeconomic equator” of the Ruhr area (Bogu- mil 2020; Kersting et al. 2009; Ziegler 2018) and therefore clustering feature increasing NO2, while PM 2.5 and PM 10 feature a higher mean in group 2 than group 3 (see Table 3). underlines its existence. The transition zone between the socioeconomic equator and the Ruhr valley is a narrow band For all three groups depicted in Fig.  6, the prevalence means differ significantly from each other at the 0.05 level, of colours for medium values. North-western and south-eastern rural areas are domi- but the measured effect size is very small (Tukey-HSD partial η 0.004, 95% CI [0.003, 0.005], p < 0.05 with adjustment for nated by medium values also. In the south-eastern part the city of Hagen and the valley of the river Ennepe westwards potential confounding effects of purchasing power, η 0.007, p < 0.05 without adjustment). Post-hoc-tests show that the from Hagen show lower purchasing power and higher preva- lence just as the central Metropolitan Region Ruhr. mean prevalence is lower in group 1 than in group 2 (− 0.255, 95% CI [− 0.3359, − 0.1746], p < 0.05) and 3 (− 0.450, 95% 4.4 Access to Urban Green CI –[0.5320, − 0.3681], p < 0.05). Thus, prevalence is associ- ated with the distinct air pollution patterns on a small scale. Among the 3 classes of access to urban green, group 1 has This relationship could not be confirmed on a larger scale, neither for absolute individual NO2, PM2.5 and PM10 values direct access to urban green and a high mean NDVI sur- rounding the building block, while group 2 lies in a neigh- nor for the identified clusters, as no significant associations can be reported for local analyses within 100 building blocks. bourhood with low NDVI and a comparatively high distance to green areas. Group 3 has medium access and features 4.3 Purchasing Power medium NDVI. Mean asthma prevalence is lower in group 1 compared to groups 2 (− 0.344, 95% CI [− 0.432, − 0.256], Overall, prevalence is negatively correlated with purchasing p < 0.05) and 3 (− 0.633, 95% CI [− 0.706, 0.559], p < 0.05). 2* With an effect size of 0.011 (partial η ), areas with simi- power (− 0.025, 95% CI [− 0.035; − 0.015[, p > 0.05). At the local level, the relationship between the two variables is lar access to urban green and medium NDVI are associated Fig. 10 Asthma prevalence and purchasing power bivariate 1 3 PFG (2022) 90:473–490 487 Fig. 11 Number of asthma patients and access to urban green bivariate with prevalence. When examining the spatial distribution of factor is not captured for the adult population, as movement the absolute number of patients with access to urban green patterns have not been included in the study. (Fig. 11), clusters of low prevalence and a greater distance to In the study area, air quality clusters are associated with urban green are found predominantly in city centres, clusters prevalence on a larger scale, linking higher 12 month preva- of high prevalence and a large distance can be found along lence to higher exposition to air pollutants. These findings major motorways (see Figs. 9 and 10). are in line with the literature, although it has to be noted that effect size is small and no causal relationship can be confirmed nor denied due to the scale and uncertainty of the 5 Discussion and Conclusion modelled air pollution values, as well as interdependencies with other variables and potential confounders like purchas- Expectedly, the spatial distribution of prevalence does nei- ing power. It can be stated that the air pollution measurement ther exhibit an overall pattern for the whole study area nor location grid is too coarse for establishing individual local uniform values. On local level, the spatial distribution var- links through kriging. All three pollutants are suspected to ies with purchasing power and accessibility of urban green, cause exacerbations and excess (emergency) hospitalization and the number of patients living close to a facility exceeds rates, so that exposition to air pollution should be monitored those in remote areas. The smaller the distance to the medi- closely. Thus knowing, how many inhabitants potentially cal facilities, the higher the predicted prevalence. As the suffer most from air pollution and are most likely to need definition of prevalence includes only asthma cases that are healthcare as a result, is crucial to resource planning and diagnosed and medically treated, the population at the out- should be investigated further. skirts could suffer from underdiagnosis and/or undertreat- Several studies discovered and verified a specific hidden ment due to lack of access to health facilities. In addition, role of the motorway A40 that is crossing the Metropolitan one has to bear in mind that diagnosis, coding schemes, and Region Ruhr from east to west, as it is often called the treatment varies among physicians. Therefore, the observed socioeconomic equator. This is not only due to income but prevalence based on ICD-10 codes and drug prescription also to language, land prices and other factors, and this does not necessarily cover the real regional 12-month study adds a few more to that list, while not rejecting the asthma prevalence (Masoli et al. 2004). At the same time, fact there is no true evidence that the motorway itself is the distance to urban green is lower on average in remote areas, cause (Jeworutzki et al. 2017; Kersting et al. 2009). In fact, where distance to health facilities is often greater than in the interplay of position and status is something that is a city centres. Green spaces can be indirectly beneficial for highly complex issue and there is a certain need of more not developing asthma in the first place, as allergies are studies that raise and try to answer the question whether less reported in rural areas. On the other hand, proximity to the A40 is a true cause for all those visible disparities or green spaces also leads to greater exposure to allergens for rather something that constitutes a fictional but also still patients with allergy-induced or severe asthma. It has to be physical border within a highly heterogeneous area like the noted, that early-life exposure to allergens as a beneficiary Metropolitan Region Ruhr. 1 3 488 PFG (2022) 90:473–490 Austin PC (2009) Type I error rates, coverage of confidence inter - A limitation of this study lies in the modelled nature vals, and variance estimation in propensity-score matched of the spatially disaggregated measures. Dispersity and analyses. Int J Biostat. https://doi. or g/10. 2202/ 1557- 4679. 1146 diversity will most likely to be underestimated due to BKG (2020) Verwaltungsgebiete 1:250 000 (VG250). https:// gdz. missing modelling parameters that are not available on bkg. bund. de/. Accessed 2 Mar 2022 Bogumil J (2020) Die Zukunft des Ruhrgebietes. Zukunft denken building block level. As the parameters used for model- und verantworten. Springer VS, Wiesbaden, pp 543–551 ling local asthma prevalence could not be integrated as Bryant-Stephens T (2009) Asthma disparities in urban environments. explanatory variables to further investigate the impact of J Allerg Clin Immunol 123:1199 socioeconomic disparities or traffic exposure on the out- Bundesärztekammer (BÄK), Kassenärztliche Bundesvereinigung (KBV), Arbeitsgemeinschaft der Wissenschaftlichen Medizinis- come, the analysis revealed intra-urban patterns that could chen Fachgesellschaften (2020) Nationale VersorgungsLeitlinie be directly attributed to these variables. Nevertheless, the Asthma – Langfassung, 4th edn. Bundesärztekammer (BÄK); results of the spatial disaggregation pattern analysis are in Kassenärztliche Bundesvereinigung (KBV); Arbeitsgemein- line with most associations from population-based cross- schaft der Wissenschaftlichen Medizinischen Fachgesellschaf- ten (AWMF) sectional as well as epidemiological studies, and reveal Burian J, Pászto V, Zapletal L (2021) Disaggregator—tool for aggre- potential areas of inequity among patients with asthma in gation and disaggregation of spatial data Earth Science Infor- a densely populated conurbation setting. matics. Springer, New York Carey MA, Card JW, Voltz JW, Arbes SJ, Germolec DR, Korach KS, Acknowledgements The authors would like to thank Prof. Dr. Carsten Zeldin DC (2007) It’s all about sex: gender, lung development Jürgens, Prof. Dr. med. Max Geraedts, and Prof. Dr. Jörg Bendix for and lung disease. Trends Endocrinol Metab 18:308–313. https:// their continued support. doi. org/ 10. 1016/j. tem. 2007. 08. 003 Cepeda MS, Boston R, Farrar JT, Strom BL (2003) Comparison of logistic regression versus propensity score when the number of Author Contributions All authors contributed to the study conception events is low and there are multiple confounders. Am J Epide- and design. Material preparation, data collection and analysis were miol 158:280–287. https:// doi. org/ 10. 1093/ aje/ kwg115 performed by Annette Ortwein and Nicolai Moos. First draft of the CDC (2018) Asthma as the Underlying Cause of Death. AsthmaStats. manuscript was written by Annette Ortwein and Nicolai Moos, and all https://www .cdc. go v/as thma/as thma_s tats/docum ents/ As thmS tat_ authors commented on previous versions of the manuscript. All authors Morta lity_ 2001- 2016-H. pdf. Accessed 2 Mar 2022 read and approved the final manuscript. European Commission, Copernicus (2020) Urban Atlas 2018. land. copernicus.eu/local/urban-atlas/urban-atlas-2012. Accessed 21 Funding Open Access funding enabled and organized by Projekt Feb 2022 DEAL. No additional funding was received for this study. Corburn J, Cohen AK (2012) Why we need urban health equity indi- cators: integrating science, policy, and community. PLoS Med Availability of Data and Material The datasets based on open data that 9:e1001285. https:// doi. org/ 10. 1371/ journ al. pmed. 10012 85 were generated and analysed during the current study are available Dadvand P, Villanueva CM, Font-Ribera L, Martinez D, Basagaña from the corresponding author on reasonable request. X, Belmonte J, Vrijheid M, Gražulevičienė R, Kogevinas M, Nieuwenhuijsen MJ (2014) Risks and benefits of green spaces for children: a cross-sectional study of associations with sed- Declarations entary behavior, obesity, asthma, and allergy. Environ Health Perspect 122:1329–1335. https://doi. or g/10. 1289/ ehp. 13080 38 Conflict of interest The authors declare no conflicts of interests. DellaValle CT, Triche EW, Leaderer BP, Bell ML (2012) Effects of ambient pollen concentrations on frequency and severity of Open Access This article is licensed under a Creative Commons Attri- asthma symptoms among asthmatic children. Epidemiology bution 4.0 International License, which permits use, sharing, adapta- 23:55–63. https:// doi. org/ 10. 1097/ EDE. 0b013 e3182 3b66b8 tion, distribution and reproduction in any medium or format, as long Eicher CL, Brewer CA (2001) Dasymetric mapping and areal inter- as you give appropriate credit to the original author(s) and the source, polation: implementation and evaluation. Cartogr Geogr Inf Sci provide a link to the Creative Commons licence, and indicate if changes 28:125–138. https:// doi. org/ 10. 1559/ 15230 40017 82173 727 were made. The images or other third party material in this article are Evans R, Gergen PJ, Mitchell H, Kattan M, Kercsmar C, Crain E, included in the article's Creative Commons licence, unless indicated Anderson J, Eggleston P, Malveaux FJ, Wedner H (1999) A otherwise in a credit line to the material. If material is not included in randomized clinical trial to reduce asthma morbidity among the article's Creative Commons licence and your intended use is not inner-city children: results of the National Cooperative Inner- permitted by statutory regulation or exceeds the permitted use, you will City Asthma Study. J Pediatr 135:332–338. https:// doi. org/ 10. need to obtain permission directly from the copyright holder. To view a 1016/ S0022- 3476(99) 70130-7 copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . Fuglsang M, Hansen HS, Münier B (2011) Accessibility analysis and modelling in public transport networks—a raster based approach. In: Murgante B, Gervasi O, Iglesias A, Taniar D, Apduhan BO (eds) Computational science and its applications - ICCSA 2011: International conference, Santander, Spain, References June 20–23, 2011; proceedings, vol 6782. Springer, Berlin, pp 207–224 Akmatov MK, Holstiege J, Steffen A, Bätzing J (2018) Diagnosepräva- Fuseini H, Newcomb DC (2017) Mechanisms driving gender differ - lenz und - inzidenz von Asthma bronchiale – Ergebnisse einer ences in asthma. Curr Allergy Asthma Rep 17:19. https://doi. or g/ Studie mit Versorgungsdaten aller gesetzlich Versicherten in 10. 1007/ s11882- 017- 0686-1 Deutschland (2009–2016). https:// doi. org/ 10. 20364/ VA- 18. 08 1 3 PFG (2022) 90:473–490 489 German National Cohort (GNC) Consortium (2014) The German the German Health and Examination Survey for Children and National Cohort: aims, study design and organization. Eur J Epi- Adolescents (KiGGS)). Bundesgesundheitsbl 55:308–317. https:// demiol 29:371–382. https:// doi. org/ 10. 1007/ s10654- 014- 9890-7doi. org/ 10. 1007/ S00103- 011- 1430-y Goodchild MF, Lam NS-N (1980) Areal Interpolation: a variant of the Lehner A, Blaschke T (2019) A generic classification scheme for urban traditional spatial problem. Geoprocessing 1(3):297–312 structure types. Remote Sens 11:173. https:// doi. org/ 10. 3390/ Grunewald K, Richter B, Meinel G, Herold H, Syrbe R-U (2017) Pro-rs110 20173 posal of indicators regarding the provision and accessibility of Levy M, Heffner B, Stewart T, Beeman G (2006) The efficacy of green spaces for assessing the ecosystem service “recreation in asthma case management in an urban school district in reducing the city” in Germany. Int J Biodivers Sci Ecosyst Serv Manag school absences and hospitalizations for asthma. J Sch Health 13:26–39. https:// doi. org/ 10. 1080/ 21513 732. 2017. 12833 61 76:320–324. https:// doi. org/ 10. 1111/j. 1746- 1561. 2006. 00120.x Guarnieri M, Balmes JR (2014) Outdoor air pollution and asthma. Li T, Corcoran J (2011) Testing dasymetric techniques to spatially Lancet 383:1581–1592. https:// doi. org/ 10. 1016/ S0140- 6736(14) disaggregate the regional population forecasts for South East 60617-6 Queensland. J Spat Sci 56:203–221. https:// doi.or g/10. 1080/ 14498 Guibas GV, Manios Y, Xepapadaki P, Moschonis G, Douladiris N, 596. 2011. 623343 Mavrogianni C, Papadopoulos NG (2013) The obesity-asthma Li T, Pullar D, Corcoran J, Stimson R (2016) A comparison of spatial link in different ages and the role of body mass index in its inves- disaggregation techniques as applied to population estimation for tigation: findings from the Genesis and Healthy Growth Studies. South East Queensland (SEQ). Australia. https://doi. or g/10. 4225/ Allergy 68:1298–1305. https:// doi. org/ 10. 1111/ all. 1224503/ 57E9A ECEBA 789 Haselkorn T, Lee JH, Mink DR, Weiss ST (2008) Racial disparities Lovasi GS, Quinn JW, Neckerman KM, Perzanowski MS, Rundle A in asthma-related health outcomes in severe or difficult-to-treat (2008) Children living in areas with more street trees have lower asthma. Ann Allergy Asthma Immunol 101:256–263. https:// prevalence of asthma. J Epidemiol Community Health 62:647– doi. org/ 10. 1016/ S1081- 1206(10) 60490-5 649. https:// doi. org/ 10. 1136/ jech. 2007. 071894 Hoffmann F (2007) Prevalence of asthma among German adults: Lovasi GS, O’Neil-Dunne JPM, Lu JWT, Sheehan D, Perzanowski MS, analysis of the German National Telephone Survey. J Asthma MacFaden SW, King KL, Matte T, Miller RL, Hoepner LA, Perera 44:433–436. https:// doi. org/ 10. 1080/ 02770 90070 14219 22 FP, Rundle A (2013) Urban tree canopy and asthma, wheeze, rhi- Jalkanen J, Fabritius H, Vierikko K, Moilanen A, Toivonen T (2020) nitis, and allergic sensitization to tree pollen in a New York City Analyzing fair access to urban green areas using multimodal birth cohort. Environ Health Perspect 121:494–500. https:// doi. accessibility measures and spatial prioritization. Appl Geogr org/ 10. 1289/ ehp. 12055 13 124:102320. https:// doi. org/ 10. 1016/j. apgeog. 2020. 102320 Luskin AT, Chipps BE, Rasouliyan L, Miller DP, Haselkorn T, Doren- Jeworutzki S, Knüttel K, Niemand C, Schmidt B-J, Schräpler J-P, baum A (2014) Impact of asthma exacerbations and asthma trig- Terpoorten T (2017) Wege zur Metropole Ruhr: Räumlich seg- gers on asthma-related quality of life in patients with severe or regierte Bildungsteilhabe in NRW und im Ruhrgebiet – zentrale difficult-to-treat asthma. J Allergy Clin Immunol Pract. https:// Ergebnisse, Bochumdoi. org/ 10. 1016/j. jaip. 2014. 02. 011 Jones AP, Bentham G (1997) Health service accessibility and deaths Markevych I, Tiesler CMT, Fuertes E, Romanos M, Dadvand P, Nieu- from asthma in 401 local authority districts in England and wenhuijsen MJ, Berdel D, Koletzko S, Heinrich J (2014) Access to Wales, 1988–92. Thorax 52:218–222 urban green spaces and behavioural problems in children: results Kennedy E, Kennedy LW (2004) Using dasymetric mapping for from the GINIplus and LISAplus studies. Environ Int 71:29–35. spatially aggregated crime data. J Quant Criminol 20:243–262. https:// doi. org/ 10. 1016/j. envint. 2014. 06. 002 https:// doi. org/ 10. 1023/B: JOQC. 00000 37733. 74321. 14 Masoli M, Fabian D, Holt S, Beasley R (2004) The global burden of Kersting V, Meyer C, Strohmeier P, Terpoorten T (2009) Die A 40 – asthma: executive summary of the GINA Dissemination Commit- der Sozialäquator des Ruhrgebiets. In: Prossek A, Schumacher tee report. Allergy 59:469–478. https:// doi. org/ 10. 1111/j. 1398- J (eds) Atlas der Metropole Ruhr: Vielfalt und Wandel des 9995. 2004. 00526.x Ruhrgebiets im Kartenbild. Emons, Köln Mennis J, Hultgren T (2005) Dasymetric mapping for disaggregating Land NRW (2022) Open Data - Digitale Geobasisdaten NRW. https:// coarse resolution population data. https:// cites eerx. ist. psu. edu/ www. openg eodata. nrw. de/. Accessed 21 Feb 2022viewd oc/ downl oad? doi= 10.1. 1. 496. 4768& rep= rep1& type= pdf. Langer S, Horn J, Kluttig A, Mikolajczyk R, Karrasch S, Schulz H, Accessed 2 Mar 2022 Wichmann H-E, Linseisen J, Jaeschke L, Pischon T, Fricke J, Mennis J, Hultgren T (2006) Intelligent dasymetric mapping and its Keil T, Ahrens W, Günther K, Kuß O, Schikowski T, Schmidt application to areal interpolation. Cartogr Geogr Inf Sci 33:179– B, Jöckel K-H, Michels KB, Franzke C-W, Becher H, Jagodz- 194. https:// doi. org/ 10. 1559/ 15230 40067 79077 309 inski A, Castell S, Kemmling Y, Lieb W, Waniek S, Wirkner K, microm GmbH (2020) PLZ8: NRW Region. Accessed 22 Nov 2020 Löffler M, Kaaks R, Greiser KH, Berger K, Legath N, Meinke- Moos N (2020) Soziogeomatik: Möglichkeiten und Grenzen der Ver- Franze C, Schipf S, Leitzmann M, Baurecht H, Weigl K, Ami- wendung von Erdbeobachtungsdaten und Geodaten zusammen tay E, Gottschick C (2020) Häufigkeit von Asthma bronchiale mit soziodemographischen und sozioökonomischen Daten. Dis- und Alter bei der Erstdiagnose – erste Ergebnisse der NAKO sertation, Ruhr-University Bochum Gesundheitsstudie (Occurrence of bronchial asthma and age at Moos N, Juergens C, Redecker AP (2021) Geo-spatial analysis of popu- initial asthma diagnosis-first results of the German National lation density and annual income to identify large-scale socio- Cohort). Bundesgesundheitsbl 63:397–403. https:// doi. org/ 10. demographic disparities. IJGI 10:432. https:// doi. org/ 10. 3390/ 1007/ s00103- 020- 03105-yijgi1 00704 32 Langford M (2006) Obtaining population estimates in non-census Mulrooney T, Beratan K, McGinn C, Branch B (2017) A comparison reporting zones: an evaluation of the 3-class dasymetric method. of raster-based travel time surfaces against vector-based network Comput Environ Urban Syst 30:161–180. https:// doi. or g/ 10. calculations as applied in the study of rural food deserts. Appl 1016/j. compe nvurb sys. 2004. 07. 001 Geogr 78:12–21. https:// doi. org/ 10. 1016/j. apgeog. 2016. 10. 006 Laussmann D, Haftenberger M, Langen U, Eis D (2012) Einfluss - Neuvonen M, Sievänen T, Tönnes S, Koskela T (2007) Access to green faktoren für Asthma bronchiale bei Kindern und Jugendlichen areas and the frequency of visits—a case study in Helsinki. Urban Forest Urban Green 6:235–247. https:// doi. org/ 10. 1016/j. ufug. in Deutschland. Ergebnisse der KiGGS-Studie (Determinants of 2007. 05. 003 asthma among children and adolescents in Germany. Results of 1 3 490 PFG (2022) 90:473–490 Nieuwenhuijsen MJ, Khreis H, Triguero-Mas M, Gascon M, Dadvand Schulte B (2008) Dasymetrische Modellierung der Verteilung von P (2017) Fifty shades of Green: pathway to healthy urban liv- Populationen. Semesterarbeit, Technische Fachhochschule Berlin ing. Epidemiology 28:63–71. https://doi. or g/10. 1097/ EDE. 00000 Shanahan DF, Fuller RA, Bush R, Lin BB, Gaston KJ (2015) The 00000 000549 health benefits of urban nature: how much do we need? Bioscience Nowak D, von Mutius E (2004) Asthma bronchiale im Kindes- und 65:476–485. https:// doi. org/ 10. 1093/ biosci/ biv032 Erwachsenenalter: Risikofaktoren, Diagnose, Standardtherapie Stanescu S, Kirby SE, Thomas M, Yardley L, Ainsworth B (2019) A (Bronchial asthma in children and adults: risk factors, diagnosis systematic review of psychological, physical health factors, and and standard treatment). Dtsch Med Wochenschr 129:509–516. quality of life in adult asthma. NPJ Prim Care Respir Med 29:37. https:// doi. org/ 10. 1055/s- 2004- 820540https:// doi. org/ 10. 1038/ S41533- 019- 0149-3 NVL (2020) Nationale Versorgungsleitline Asthma. https:// www. leitl Strunk RC, Ford JG, Taggart V (2002) Reducing disparities in asthma inien. de/ themen/ asthma. Accessed 17 Aug 2021 care: priorities for research–National Heart, Lung, and Blood Pekkanen J, Pearce N (1999) Defining asthma in epidemiological stud - Institute workshop report. J Allerg Clin Immunol 109:229–237. ies. Eur Respir J 14:951–957. https://doi. or g/10. 1034/j. 1399- 3003. https:// doi. org/ 10. 1067/ mai. 2002. 120950 1999. 14d37.x Theofilou P (2013) Quality of life: definition and measurement. EJOP RKI (2017) 12-month prevalence of asthma among adults in Germany. 9:150–162. https:// doi. org/ 10. 5964/ ejop. v9i1. 337 https:// doi. org/ 10. 17886/ RKI- GBE- 2017- 064 Toelle BG, Peat JK, Salome CM, Mellis CM, Woolcock AJ (1992) Robert Koch Institute, Department of Epidemiology and Health Moni- Toward a definition of asthma for epidemiology. Am Rev Respir toring (2015) German Health Interview and Examination Survey Dis 146:633–637. https:// doi. org/ 10. 1164/ ajrccm/ 146.3. 633 for Adults (DEGS1): Scientific Use File first Version von Mutius E (2016) The microbial environment and its influence on Robert Koch Institute, Department of Epidemiology and Health Moni- asthma prevention in early life. J Allergy Clin Immunol 137:680– toring (2019) The German Health Survey for Children and Ado- 689. https:// doi. org/ 10. 1016/j. jaci. 2015. 12. 1301 lescents (KiGGS Wave 2): Scientific Use File first version WHO (2021) Asthma. https:// www. who. int/ news- room/ fact- sheets/ Rosenbaum PR, Rubin DB (1985) The bias due to incomplete match-detail/ asthma. Accessed 2 Mar 2022 ing. Biometrics 41:103. https:// doi. org/ 10. 2307/ 25306 47 Wissenschaftliches Institut der AOK (WIdO) (2020) Gesundheits- Rubin DB, Thomas N (2000) Combining propensity score matching atlas Deutschland - Asthma: Verbreitung in der Bevölkerung with additional adjustments for prognostic covariates. J Am Stat Deutschlands und seinen Regionen.Ursachen, Folgen und Assoc 95:573–585. https://doi. or g/10. 1080 /01621 459. 2000 .10474 Präventionsmöglichkeiten. https:// www. wido. de/ filea dmin/ Datei 233en/ Dokum ente/ Publi katio nen_ Produ kte/ Buchr eihen/ Gesun dheit Regionalverband Ruhr (2021) Regionalstatistik. https://www .r vr.r uhr/ satlas/wido_ int_ g esundheit satlas_ as thma_1120. pdf . Accessed 02 daten- digit ales/ r egio nals t atis t ik/ news- de t ail- r egio nals t atis t ik. Mar 2022 Accessed 2 Mar 2022 Ziegler E (2018) 5. Visuelle Mehrsprachigkeit in Migrationsgesells- Ruokolainen L, von Hertzen L, Fyhrquist N, Laatikainen T, Lehtomäki chaften: monolinguale Norm vs. plurilinguale Norm. In: Lenz AN, J, Auvinen P, Karvonen AM, Hyvärinen A, Tillmann V, Niemelä Plewnia A (eds) Variation – Normen – Identitäten. De Gruyter, O, Knip M, Haahtela T, Pekkanen J, Hanski I (2015) Green areas Berlin, Germany, pp 127–156 around homes reduce atopic sensitization in children. Allergy 70:195–202. https:// doi. org/ 10. 1111/ all. 12545 1 3

Journal

"PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science"Springer Journals

Published: Oct 1, 2022

Keywords: Environmental justice; Equal health; Disaggregation; Network analysis; Asthma; Prevalence

References