Analysing hospital travel times in central Athens using Google Maps services

Abstract Introduction The importance of timely care is well documented for numerous emergency conditions, including STEMI and ischemic stroke, where low symptom-to-balloon/symptom-to-needle times are crucial for mortality and disability. The study of all potential delays helps us understand the constraints we have to work under. Here, we use Google Maps services to map the travel times from central Athens areas to on-duty hospitals Methods We built our code in the Python programming language, using the Google Maps Distance Matrix API to perform real-time trip duration calculations based on real-life data. As reference points, we used a set of Athens neighbourhoods provided by the Municipality as open data; we considered only public secondary and tertiary health facilities as valid destinations, and based our calculations based on the available daily duty schedules. Results Our algorithm collected 43,200 data points in total over two weeks, using 144 starting points. The average trip durations to reach an on-duty department formed a right-skewed distribution (-0.424), with a mean of 19.55 minutes and a median of 19.95 minutes. The maximum average time was 26.78 minutes, and the overall maximum was 44 minutes. Average travel times to cardiology, general surgery, neurology and internal medicine ERs, which experience a heavy patient load, were higher than the total mean (20.60/22.06/21.31/20.51 mins respectively). We found no correlation between the average travel time and average distance from a hospital or the geographical location, but we were able to create a map with hotspots of high/low travel times. Conclusions Our approach to collecting accurate trip data has shown the impact of time-of-day and location on trips to hospitals, even for patients within the same larger area. As acute care can be time-sensitive, similar wide-scale modelling could be used to create systemic solutions, e.g. data-guided spatial distribution of facilities or transportation. Key messages Public APIs can be used to gather useful data about the context around our health systems. In Athens, a difference in position can mean up to 100% longer travel times to a hospital.


Background:
In national health systems based on primary care, cross-level clinical coordination (CC) is a priority, as it may improve quality of care.Evidence on the impact of information and communication technology (ICT)-based coordination mechanisms on CC is inconclusive.The implementation of those mechanisms increased during the pandemic.The aim is to adapt the validated COORDENA-CAT questionnaire, for measuring CC, to analyse the implementation of ICT-based coordination mechanisms and its impact on CC in three regions of Spain.

mechanisms i
creased during the pandemic.The aim is to adapt the validated COORDENA-CAT questionnaire, for measuring CC, to analyse the implementation of ICT-based coordination mechanisms and its impact on CC in three regions of Spain.


Methods:

The COORDENA-CAT questionnaire underwent a two-stage adaptation process:1) contents revision based on literature review, expert discussions, and pretest to adapt the contents and language and produce a version for each region; and 2) piloting the adapted version by self-administration of the qu

Methods:
The COORDENA-CAT questionnaire underwent a two-stage adaptation process:1) contents revision based on literature review, expert discussions, and pretest to adapt the contents and language and produce a version for each region; and 2) piloting the adapted version by self-administration of the questionnaire to primary and secondary care doctors in the health systems of two of the participating regions.

tionnaire
to primary and secondary care doctors in the health systems of two of the participating regions.


Results:

The adapted version (COORDENA-TICs) kept most of the original contents.Main changes were on coordination mechanisms section.Pretest showed an adequate level of comprehensiveness, comprehension, sequence of themes and questions, and length.A low non-response rate was observed, with little variab

Results:
The adapted version (COORDENA-TICs) kept most of the original contents.Main changes were on coordination mechanisms section.Pretest showed an adequate level of comprehensiveness, comprehension, sequence of themes and questions, and length.A low non-response rate was observed, with little variability or unexpected responses.The question on any difficulties encountered in answering the questionnaire revealed no relevant difficulties.The Shared Electronic Medical Record of each region was the most frequently used ICT-based coordination mechanism.Limited access to information and technical issues related to its use were the most common difficulties encountered.Suggestions for improving its use were receiving specific training on its use.

ity or un
xpected responses.The question on any difficulties encountered in answering the questionnaire revealed no relevant difficulties.The Shared Electronic Medical Record of each region was the most frequently used ICT-based coordination mechanism.Limited access to information and technical issues related to its use were the most common difficulties encountered.Suggestions for improving its use were receiving specific training on its use.


Conclusions:

COORDENA-TICs questionnaire is a useful tool to assess utilization of ICT-based coordination mechanisms and its impact on CC from the perspective of primary and secondary care doctors.It will allow comparisons across areas, regions


Introduction:

The importance of timely care is well documented for numerous emergency conditions, including STEMI and ischemic stroke, where low symptom-to-balloon/symptomto-needle times are crucial for mortality and disability.The study of all potential delays helps us understand the constraints we have to work under.Here, we use Google Map

Conclusions:
COORDENA-TICs questionnaire is a useful tool to assess utilization of ICT-based coordination mechanisms and its impact on CC from the perspective of primary and secondary care doctors.It will allow comparisons across areas, regions

Introduction:
The importance of timely care is well documented for numerous emergency conditions, including STEMI and ischemic stroke, where low symptom-to-balloon/symptomto-needle times are crucial for mortality and disability.The study of all potential delays helps us understand the constraints we have to work under.Here, we use Google Maps services to map the travel times from central Athens areas to on-duty hospitals Methods:

services to ma
the travel times from central Athens areas to on-duty hospitals Methods:

We built our code in the Python programming language, using the Google Maps Distance Matrix API to perform real-time trip duration calculations based on real-life data.As reference points, we used a set of Athens neighbourhoods provided by the Municipality as open data; we considered only public secondary and tertiary health facilities as valid destinations, and based our calculations based on the available daily duty schedules.


Results:

Our algorithm collected 43,200 data points in total over two weeks, using 144 starting points.The average trip durations to reach an on-duty department formed a right-skewed distribution (-0.424), with a mean of 19.55 minutes and a median of 19.95 minutes.The maximum average time was 26.78 minutes, and the overall maximum was 44 minut We built our code in the Python programming language, using the Google Maps Distance Matrix API to perform real-time trip duration calculations based on real-life data.As reference points, we used a set of Athens neighbourhoods provided by the Municipality as open data; we considered only public secondary and tertiary health facilities as valid destinations, and based our calculations based on the available daily duty schedules.

Results:
Our algorithm collected 43,200 data points in total over two weeks, using 144 starting points.The average trip durations to reach an on-duty department formed a right-skewed distribution (-0.424), with a mean of 19.55 minutes and a median of 19.95 minutes.The maximum average time was 26.78 minutes, and the overall maximum was 44 minutes.Average travel times to cardiology, general surgery, neurology and internal medicine ERs, which experience a heavy patient load, were higher than the total mean (20.60/22.06/21.31/20.51mins respectively).We found no correlation between the average travel time and average distance from a hospital or the geographical location, but we were able to create a map with hotspots of high/low travel times.

.Average
ravel times to cardiology, general surgery, neurology and internal medicine ERs, which experience a heavy patient load, were higher than the total mean (20.60/22.06/21.31/20.51mins respectively).We found no correlation between the average travel time and average distance from a hospital or the geographical location, but we were able to create a map with hotspots of high/low travel times.


Conclusions:

Our approach to collecting accurate trip data has shown the impact of time-of-day and location on trips to hospitals, even for patients within the same larger area.As acute care can be time-sensitive, similar wide-scale modelling could be used to create systemic solutions, e.g.data-guided spatial distribution of facilities or transport

Conclusions:
Our approach to collecting accurate trip data has shown the impact of time-of-day and location on trips to hospitals, even for patients within the same larger area.As acute care can be time-sensitive, similar wide-scale modelling could be used to create systemic solutions, e.g.data-guided spatial distribution of facilities or transportation.

Key messages:
sages:

Public APIs can be used to gather useful data about the context around our health systems.

In Athens, a difference in position can mean up to 100% longer travel times to a hospital.

iii360

European Journal of Public Health, Volume 32 Supplement 3, 2022 Public APIs can be used to gather useful data about the context around our health systems.
In Athens, a difference in position can mean up to 100% longer travel times to a hospital. iii360 European Journal of Public Health, Volume 32 Supplement 3, 2022 