A GIS Open Source Application to Perform the Spatial Distribution

of Prevention Quality Indicators (PQIs)

L. Duarte

1,2

, M. Lobo

3,4

, J. Viana

3,4

, A. Freitas

3,4

and A. C. Teodoro

1,2

Institute of Earth Sciences, FCUP pole, Rua do Campo Alegre, Porto, Portugal

Department of Geosciences, Environment and Spatial Planning, FCUP, Porto, Portugal

Department of Community Medicine, Information and Health Decision Sciences (MEDCIDS), Faculty of Medicine,

University of Porto, Porto, Portugal

Center for Health Technology and Services Research (CINTESIS), Faculty of Medicine, University of Porto, Porto,

Portugal

Keywords: GIS, Health Care Indicators, PQI, Spatial Database.

Abstract: Geographical variations carry important information for improving and planning more equitable and

sustainable health care services. Geographic Information Systems (GIS) are crucial tools that provide intuitive

visual help which contributes to a better understanding of the spatial distribution of health risk factors,

resources, care and outcomes. The interest in GISs have stimulated the development of several applications

worldwide to publicly inform the geographical patterns of health. However, in Portugal, this type of tools

remains underdeveloped for public reporting of health information. The aim of this study was to develop a

GIS open source application for spatial analysis of healthcare indicators in Portugal, using hospital data

obtained from the Administração Central do Sistema de Saúde, I.P. Specifically, given their importance to

monitor the quality of primary health care, data regarding Prevention Quality Indicators (PQIs) will be used

to establish a proof of concept of this tool. The tool was connected to a spatial database in order to filter the

parameters. Several maps based on PQI information were created in order to test the application. It was

concluded that the spatial combination of all the data provided in a GIS software and through an intuitive

application can contribute to the analysis of quality of primary health care.

1 INTRODUCTION

Geographical variations carry important information

for improving and planning more equitable and

sustainable health care services. However,

understanding geographic patterns can be

cumbersome and error-prone if performed through

frequency tables and traditional plots, especially if the

information covers several data sources and variables

such as health care, environmental, social, and

economical variables.

Geographic Information Systems (GIS) are

important tools that contribute to a better

understanding of the spatial distribution of health risk

factors, resources, care and outcomes through

https://orcid.org/0000-0002-7537-6606

https://orcid.org/0000-0003-3890-7735

https://orcid.org/0000-0003-4696-1002

https://orcid.org/0000-0003-2113-9653

https://orcid.org/0000-0002-8043-6431

intuitive geospatial visual interfaces. Therefore,

carrying valuable insights for policymakers, health

providers and populations. The interest in GIS has

stimulated the development of several applications

worldwide to publicly inform on geographical

patterns of health (DAP, 2020; Atlas, 2020; SAHSU,

2020; HWM, 2020). The use of GIS in public health

is increasing as a response to the requirements of

queries and analysis of health indicators

(Maheswaran and Craglia, 2004).

However, in Portugal, this type of tools remains

underdeveloped for public reporting of health

information. GEOSAUDE (http://www.geosaude.

dgs.pt/) a web GIS, and PORDATA (https://www.

pordata.pt/), a statistical database, offer regional

Duarte, L., Lobo, M., Viana, J., Freitas, A. and Teodoro, A.

A GIS Open Source Application to Perform the Spatial Distribution of Prevention Quality Indicators (PQIs).

DOI: 10.5220/0009805801290134

In Proceedings of the 6th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2020), pages 129-134

ISBN: 978-989-758-425-1

129

variations health-related data interactive maps, within

Portugal.

The PORDATA only presents statistical data in the

form of numerical statistics, graphs and indicators.

This information is not spatially represented. The

GEOSAUDE is a powerful web GIS composed by

several indicators and options to filter the

information. However, it just allows the visualization

of the data. It does not allow to manipulate the

information.

The Portuguese administrative hospital database

from Administração Central do Sistema de Saúde,

I.P. (Central Administration of the Health System,

ACSS, 2020), is an essential tool to support hospital

funding based on diagnosis-related groups (DRG). It

also represents a fundamental source of data of the

official health statistics (Ferreira et al., 2017; OECD,

2019). This database is easily accessible and well

documented, with a high population and temporal

coverage, containing information regarding all

inpatient and outpatient hospitalizations of

Portuguese public hospitals in the mainland territory,

since 2000. As it contains data regarding the area of

residence of each patient (geographic location), it

may be easily usable in a GIS environment. This

possibility enhances the interoperability between the

data source and environmental data and even other

geographical variables. Therefore, potential value

may be added to this database through its integration

in GIS. The objective of this work was to develop a

GIS open source application which allows to easily

connect to a database of health quality indicators and

spatially represent them. The spatial representation of

the data will allow to analyse the indicators in a

national level. This analysis allows to relate other

factors with these health indicators.

1.1 Ambulatory Care Sensitive

Conditions (ACSCs)

The Ambulatory Care Sensitive Conditions (ACSCs)

are conditions for which good outpatient care can

potentially prevent or reduce the need for inpatient or

emergency care due to complications or more severe

diseases associated with these conditions (ACSC,

2016). For instance, diabetic complications may arise

if diabetes is not adequately monitored or if education

regarding patient self-management is not provided

(PQIO, 2020; AHRQ Quality Indicators, 2020).

The Agency for Healthcare Research and Quality

(AHRQ) developed a set of indicators based on

hospital administrative data – the Prevention Quality

Indicators (PQIs) – to measure quality of care for

several common ACSCs and compare local health

care systems across communities (PQIO, 2020).

In Portugal, 12,3% (n=1003602) of inpatient

hospitalizations were attributable to PQI-related

ACSCs, exhibiting several regional variation patterns

depending on the condition. The heart failure

hospitalizations were more common in the most

northern and interior regions of Portugal and in the

central Portugal (WHO, 2016; Sarmento et al., 2015;

Rocha et al., 2019). The low hospitalization rates

were reported to cluster closer to the coastal zones

and around bigger cities; higher hypertensive heart

disease hospitalization rates were reported in the

interior regions of the country (WHO, 2016;

Sarmento et al., 2015; Rocha et al., 2019). Therefore,

in Portugal, important insights may be gained

regarding the quality of health care in ACSCs from

monitoring regional variations of PQIs. This could be

used to screen potential problems in primary health

care system, direct further investigations to assess

causes of problems, and to compare performance of

regional community health care, which may assist in

the definition of public health recommendations and

ultimately improve health care in ACSCs.

To our knowledge, at the moment, there is no GIS

covering ACSCs care-related metrics such as the

PQIs. In this context, the development of a GIS

application under GIS software environment can help

to analyse and generate more information.

1.2 Objective

The aim of this study was the development of a GIS

open source application for spatial analysis of

healthcare indicators in Portugal, using hospital data

obtained from the ACSS. Specifically, given their

importance to monitor the quality of primary health

care data regarding PQIs will be used to establish a

proof of concept of this tool. The tool was connected

to a spatial database in order to filter the parameters.

2 MATERIAL AND METHODS

The GIS open source application was developed

under the open source software QGIS version 3.10,

(QGIS, 2019). Several Python libraries and

Application Programming Interfaces (APIs) were

used to develop the application, such as QGIS API

and Qt API (QGIS, 2020; Qt API, 2020). QGIS

supports spatial databases such as PostgreSQL and

PostGIS (PostGIS, 2020). The most recent versions

of QGIS provides a Qt Tool named Qt Designer for

designing and building graphical user interfaces

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

130

(GUIs) with Qt Widgets (Qt Designer, 2020). The Qt

Designer was used to create the application GUI. The

main objective of this GIS application was to improve

the connection to a spatial database, implemented in

PostGIS and spatially represent the PQIs. This

database is composed by multiple data. The focus is

to represent spatially the information and provide

some functionalities that can be useful to analyse the

database provided.

2.1 Data Sources/Database

In the database, the PQIs were estimated based on the

AHRQ definitions, using the hospital administrative

data and population estimates between 2014 and 2017

(INE, 2020). The hospital data cover all

hospitalizations of mainland Portuguese public

hospitals, containing demographic and clinical

information of the patients, such as age, sex,

residence, diagnoses, procedures, and the patient

disposition after discharge. Until 2016 the diagnoses

and procedures were coded according to the

International Classification of Diseases, 9

Revision,

Clinical Modification (ICD-9-CM). In 2016, a few

pilot hospitals initiated the transition to the ICD-10-

CM. During this transitional year the two clinical

classification systems coexist. This is an existent

attribute in the database of this study. As of the 1

January 2017, all public hospitals were instructed to

code in ICD-10-CM.

The PQIs encompass sixteen ambulatory care

sensitive conditions (Table 1). These are then

aggregated into four additional composite indicators

(Table 2).

Table 1: PQIs description.

PQI Code Description

PQI01 Diabetes Short-Term Complications

PQI02 Perforated Appendix

PQI03 Diabetes Lon

-Term Com

lications

PQI05 COPD or Asthma in Older Adults

PQI07 H

ertension

PQI08 Heart Failure

PQI09 Low Birth Weight

PQI10 Dehydration

PQI11 Bacterial Pneumonia

PQI12 Urinar

Tract Infection

PQI14 Uncontrolled Diabetes

PQI15 Asthma in Younger Adults

PQI16

Lower-Extremity Amputation Among

Patients with Diabetes

Table 2: PQIs composites description.

PQI Code Descri

tion

PQI90 Prevention Qualit

Overall Com

osite

PQI91

Prevention Quality Acute Composite

(PQI11, PQI12)

PQI92

Prevention Quality Chronic Composite

(PQI1, PQI3, PQI5, PQI7, PQI8,

PQI14, PQI15, PQI16

)

PQI93

Prevention Quality Diabetes Composite

(PQI1, PQI3, PQI14, PQI16)

The PQIs were computed by sex, year, month and

the residence of the patient. The unit of analysis

considered was the district and Nomenclature of

Territorial Units for Statistics (NUTSIII).

2.2 PostGIS Database

The PostGIS database is a spatial database extender

for PostgreSQL object-relational database (PostGIS,

2019). PortGIS allows to query in Structured Query

Language (SQL) language and supports geographic

objects. It is released under the GNU General Public

License (GPL). There are several plugins developed

in QGIS which uses PostGIS such as DB Style

Manager, Fast SQL layer, PostGIS manager, among

others (QGIS plugins, 2020).

2.3 Application Development

As referred before, the application interface was

created through Qt Designer. Also, a PostGIS

database was created, and connected to the

application.

The application was developed using Python

programming language (Python, 2020). Several

libraries were also used such as PyQt5 and QGIS API.

The SQL language was implemented in order to

perform a selection based on the query applied by the

user. The SQL conditions run through pgsql2shp.exe

from PostgreSQL. In order to define the SQL queries,

a batch file is automatically created to run the exe

with the parameters defined by the user in the GUI.

The batch file is automatically saved in the plugin

folder and it is built in the moment that the user chose

the variables. Through the os.system function the

batch file runs and the shapefile is created and

automatically added to the canvas.

The application is composed by a button on QGIS

environment which opens a dialog composed by two

tabs: Symbology and Mapping. The first one allows to

represent spatially the PQIs data by a certain level

(district or NUTS) and using two types of symbology

(Figure 1), Categorized or Graduated; the Mapping

tab (Figure 2) allows to: i) incorporate a base map in

A GIS Open Source Application to Perform the Spatial Distribution of Prevention Quality Indicators (PQIs)

131

the QGIS canvas in order to overlap with other

information; ii) add some additional information such

as the connection to Bing Aerial Map, the location of

hospital facilities and/or primary health care facilities

(as point shapefile); iii) create a layout in order to

print the map; iv) convert the shapefile to KML

format and; v) to export a bar plot with the variation

of PQIs along the years (existent years in the

database).

Figure 1: Symbology tab of the HCQI application.

Figure 2: Mapping tab of the HCQI application.

In the second tab, Mapping, three check boxes

were added, and the user checks the box that pretends

to overlap and clicks on Add button to add the

information to the canvas. The Layout button allows

to create a layout with the information selected by the

user and composed with the main elements: north

arrow, scale, map and legend. The button Export to

KML connects to a new GUI. In this GUI, the user

selects the layer that pretends to convert in KML

format and save it. The Export Bar Plot connects to

other GUI (Figure 3).

Figure 3: Bar Plot functionality.

In this GUI, the user must choose a district, the

attribute to insert in the x axis and the attribute to

insert in the y axis. This functionality allows to create

a bar plot to a specific district and for the assigned

attribute. This functionality was implemented based

on Bar Plot algorithm from QGIS Processing

Toolbox (QGIS, 2020).

3 RESULTS

In order to test the application, a shapefile with the

district level was incorporated in the PostGIS

database. This shapefile allowed to select between

two levels: district or NUTSIII. Some maps based on

PQIs information were created in order to test the

application. Figure 4 presents the result of the

selection of PQI value for PQI90 in 2016 year with a

graduated symbology.

Other options were also tested, such as the

possibility of overlapping a web map to the data

already opened (Figure 4).

The proximity with primary care facilities and

hospitals are, among other factors, crucial in the

utilization of hospital care to treat ACSC (Carneiro,

2018). Therefore, the application was also

complemented with the possibility to add the national

hospital facilities and the primary health care

facilities. Figure 5 presents the overlapping of these

information with the aerial map (Bing Aerial Maps).

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

132

Figure 4: PQI90 in 2016 overlapped to Bing Aerial Map.

Figure 5: Overlapping of Bing Aerial Map with hospital and

primary care centres.

The implementation of these functionalities (the

addition of the hospital facilities and the health care

facilities) combined with the PQIs information,

provides the possibility to analyse the quality of

access to health care. Longer distances to primary

health care may represent a barrier to seek primary

health care, leading patients to use the hospital care

when conditions have worsened. Figure 6 presents the

overlapping between the facilities and the PQIs

information for PQI90 for 2016 year. In addition,

since the database is composed by data from 2014 to

2017, it would be very useful to analyse the PQIs

variation along the 4 years, so the possibility to create

bar plots with that variations was also tested. Figure

6 also presents a bar plot with the variation of PQIs

for Porto district.

Figure 6: Overlapping between hospital and primary care

centres and the information of PQIs.

From Figure 6 we can conclude that the number

of hospitalizations with ACSCs in Porto showed a

positive trend. Besides this possibility, the application

also converts a shapefile to KML format. This can be

very useful to open and overlap the data in Google

Maps® or Google Earth®.

4 CONCLUSIONS

The GIS application will be very useful to help the

health experts to understand the geographical

A GIS Open Source Application to Perform the Spatial Distribution of Prevention Quality Indicators (PQIs)

133

distribution of PQI values (by district or NUTS), even

apply filters in terms of PQI values, years and the

clinical codification system. The spatial combination

of all the data provided in a GIS software and through

an intuitive application can contribute to the analysis

of quality of primary health care. The implemented

filters may already provide insight to questions

regarding where (e.g. which districts?), when (e.g.

which year? Quality of care is getting better?) and

why (e.g. proximity to hospital? Distance to primary

health care facilities?) PQIs indicate better quality of

care. These are relevant questions to policymakers,

health care providers and the general population.

Even though among the list of indicators in

GEOSAUDE there is “the number of inpatient

hospitalizations with an ACSC”, this is still not

operational. Thus, to our knowledge there is no

application representing PQIs for Portugal.

In the future we intend to improve the GIS

application with the ability to infer and point out any

problems or areas needing further analysis in the data.

ACKNOWLEDGEMENTS

This work was funded through the Foundation for

Science and Technology, through the COMPETE

2020 programme and framed within the activities of

the UIDB/04683/2020. ICT financed through the

European Regional Development Fund (COMPETE

2020), with ref. POCI-01-0145-ERDF-007690.

It was also supported by FEDER - Fundo Europeu de

Desenvolvimento Regional funds through the

COMPETE 2020 - Operacional Programme for

Competitiveness and Internationalisation (POCI),

and by Portuguese funds through FCT - Fundac¸ão

para a Ciência e a Tecnologia in the framework of the

project POCI-01-0145-FEDER-030766 (“1st.

IndiQare - Quality indicators in primary health care:

validation and implementation of quality indicators

as an assessment and comparison tool”).

REFERENCES

ACSC. Ambulatory care sensitive conditions in Portugal.

World Health Organization. WHO Reg Off Eur. 2016.

ACSS. Administração Central do Sistema de Saúde. 2020.

http://www.acss.min-saude.pt/. Accessed January 2020.

AHRQ Quality Indicators. Guide to Prevention Quality

Indicators: Hospital Admission for Ambulatory Care

Sensitive Conditions. 2020. https://www.ahrq.gov/

downloads/pub/ahrqqi/pqiguide.pdf. Accessed January

2020.

Atlas. Atlas de hospitalizaciones por enfermedad isquémica

coronaria y procedimientos cardiacos. 2020.

https://www.atlasvpm.org/atlas/. Accessed January

2020.

Carneiro, C.S. Hospitalisation of ambulatory care sensitive

conditions and access to primary care in Portugal.

Public health. 2018. Dec 1;165:117-24.

DAP. The Dartmouth Atlas of Health Care. 2020.

https://www.dartmouthatlas.org/. Accessed January

2020.

Ferreira, R., Macedo, M., Pinto, F., Neves, R., Andrade, C.,

Santos, G. Programa Nacional para as Doenças

Cérebro-Cardiovasculares. 2017.

HWM. Health Workforce Mapper. 2020. https://www.ama-

assn.org/about/research/health-workforce-mapper.

Accessed January 2020.

INE. Instituto Nacional de estatística. 2020.

https://www.ine.pt/xportal/xmain?xpgid=ine_main&x

pid=INE&xlang=pt. Accessed January 2020.

Maheswaran, R. and Craglia, M. (Eds), 2004. GIS in Public

Health Practice. CRC Press.

Moore, R., Lopes, J., 1999. Paper templates. In

TEMPLATE’06, 1st International Conference on

Template Production. SCITEPRESS.

OECD. Health at a Glance 2019: OECD Indicators,

OECD Publishing, Paris. 2019. https://doi.org/10.1787/

4dd50c09-en.

PostGIS. Spatial and Geographic objects for PostgreSQL.

https://postgis.net/. Accessed January 2020.

PQIO. Prevention Quality Indicators Overview. 2020.

https://www.qualityindicators.ahrq.gov/modules/pqi_o

verview.aspx. Accessed January 2020.

Python. Python Programming Language.

http://python.org/. Accessed January 2020.

QGIS. QGIS Project. 2020. http://www.qgis.org/. Accessed

January 2020.

QGIS plugins. Plugins tagged with: postgis. 2020.

https://plugins.qgis.org/plugins/tags/postgis/. Accessed

January 2020.

Qt API. Qt API. 2020. https://doc.qt.io/qt-5/reference-

overview.html. Accessed January 2020.

Qt Designer. Qt Documentation. 2020. https://doc.qt.io/qt-

5/qtdesigner-manual.html. Accessed January 2020.

Rocha JV, Nunes C, Santana R. Avoidable hospitalizations

in Brazil and Portugal: Identifying and comparing

critical areas through spatial analysis. PloS one.

2019;14(7).

Sarmento J, Alves C, Oliveira P, Sebastiao R, Santana R.

Caracterização e evolução dos internamentos evitáveis

em Portugal. Acta Médica Portuguesa. 2015;28(5):590-

600.

SAHSU. UK Small Area Health Statistics Unit. The

environment and health atlas for England and Wales.

2020.

http://www.envhealthatlas.co.uk/homepage/index.html

. Accessed January 2020.

World Health Organization (WHO). Ambulatory care

sensitive conditions in Portugal. WHO Reg Off Eur.

2016.

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

134