Using a geographic information system to enhance patient access to point-of-care diagnostics in a limited-resource setting
© Ferguson et al. 2016
Received: 16 June 2015
Accepted: 5 February 2016
Published: 1 March 2016
Rapid and accurate diagnosis drives evidence-based care in health. Point-of-care testing (POCT) aids diagnosis by bringing advanced technologies closer to patients. Health small-world networks are constrained by natural connectivity in the interactions between geography of resources and social forces. Using a geographic information system (GIS) we can understand how populations utilize their health networks, visualize their inefficiencies, and compare alternatives.
This project focuses on cardiac care resource in rural Isaan, Thailand. A health care access analysis was created using ArcGIS Network Analyst 10.1 from data representing aggregated population, roads, health resource facilities, and diagnostic technologies. The analysis quantified cardiac health care access and identified ways to improve it using both widespread and resource-limited strategies.
Results indicated that having diagnostic technologies closer to populations streamlines critical care paths. GIS allowed us to compare the effectiveness of the implementation strategies and put into perspective the benefits of adopting rapid POCT within health networks.
Geospatial analyses derive high impact by improving alternative diagnostic placement strategies in limited-resource settings and by revealing deficiencies in health care access pathways. Additionally, the GIS provides a platform for comparing relative costs, assessing benefits, and improving outcomes. This approach can be implemented effectively by health ministries seeking to enhance cardiac care despite limited resources.
KeywordsAcute myocardial infarction Spatial care path™ Health care access Geographic information system (GIS)
We developed a geographic information system (GIS) as a framework for (a) understanding how populations utilize a health small-world network in a limited-resource setting, (b) quantifying population access to health resources, (c) visualizing inefficiencies, and (d) evaluated alternative scenarios that build on a wide array of previous applications [1–7]. Prior studies are limited because they do not focus on social pressures of a health system, such as culturally related health decisions, or how diagnostic information is obtained.
The complexities of health networks are not explained by the sum of their parts and often are defined by the natural connectivity that arises from their element interactions . Health networks are thought of as small-world networks (SWN), which are not completely random or regularly connected systems , and may not be efficient in terms of the urgent care of its populations. The goal of this study, therefore, is to show that a GIS can help understand the inefficiencies that exist within health network SWNs, and then, used to enhance strategic organization and outcomes.
Point-of-care technologies (POCT) result from the miniaturization of conventional laboratory diagnostic tests into portable forms. Many POCT are characterized by factors such as, affordable, sensitive, specific, user-friendly, rapid/robust, equipment-free, and deliverable (ASSURED) criteria . ASSURED qualities allow POCT to be used in non-laboratory settings by non-technical staff without infrastructure, such as in-wall electricity, and in a wide range of environments. POCT has been implemented in national disaster caches for disaster preparedness , emergency settings for care optimization , and limited-resource settings [13, 14].
Kost et al.  attempted to quantify SWN relationships for cardiac support in health facilities in Northeastern Isaan, Thailand. Their study revealed isolated regions with inadequate support for their populations. They suggested several placement strategies for POCT that would improve patient outcomes. While the study was useful for understanding the SWN, it did not attempt to quantify population access to these diagnostic resources nor evaluate how the recommended placement strategies would affect this access.
GIS provides the ability to effectively implement POCT in health networks to streamline decision making at the point of care. This can (a) improve patient outcomes, (b) save resources including money and time, and (c) ensure that the health networks are sufficiently robust for a disaster or emergency event. This project demonstrates how a GIS can quantify health care access to help make decisions on how to integrate POCT.
To build and evaluate a spatial analysis that defines health care access.
Define the current health care access to cardiac care within an area used in the Kost et al.  study.
Evaluate implementation strategies that improve health care access to cardiac care.
Determine the outcomes of the implementation strategies against current access.
Current health care access for cardiac events
Mean (SD) travel time for current health access
Travel time (min)
Populated places to diagnosis
Diagnosis to care
Populated places to care
Figure 1 (middle frame) shows the total travel time from populated places to diagnosis then to care. In the western areas of Bueng Kan there are groupings of populated places that have under 2 h to reach care. Immediately surrounding these locations, travel time of these populated places increases to between 5 and 6 h. Individuals who travel east must then backtrack to reach care. A longer time to diagnosis in order to decrease overall time to care may lead to a more appropriate evaluation.
The longest travel times to care mostly fall in Bueng Kan. A small area of higher travel times can also be found in northern Sakhon Nakhon which may have resulted from individuals from this area travelling 1.5–2 h to reach a diagnosis and then backtrack to reach care. The locations further away from Srinagarind Hospital, where you would expect longer travel times, particularly eastern Nakhon Phanom, have relatively lower travel times.
Figure 1 (bottom frame) shows the histogram for the travel times from populated places to care and provides a means to understand the distribution of calculated routes.
Widespread implementation strategy
Mean (SD) travel time comparison between different policy implementation strategies
All health resource facilities
Average travel time from pop to diagnosis (min)
Percent decrease over current access
Average travel time from diagnosis to care (min)
Percent decrease over current access
Total travel time (min)
Mean travel time percent decrease over current access
The quickest routes are generated when POCT is implemented in every health resource facility; however it offers little advantage over integrating POCT in every hospital. The difference in average travel times from populated places to diagnosis between the two strategies was 9.3 min which is negligible when compared to the travel time from populated places to care. The difference between these two strategies would only result in a 2.6 % decrease, potentially a reason to only implement POCT within hospitals instead of every health resource facility.
Histograms in Fig. 2 show the travel time to care distributions for the two widespread implementation strategies (white bars) compared to the current health care access (black bars). In both cases we see the improvements in overall travel times.
Limited-resource implementation strategy
Travel time comparison between different low-resource implementation strategies
Current position with 5 more
Current position with 10 more
Rearrange with 5 more
Rearrange with 10 more
Mean (SD) travel from population to diagnosis
Decrease compared to current
Mean (SD) travel from diagnosis to care
Decrease compared to current
Mean (SD) travel time from population to care
Decrease compared to current
The first analysis rearranges existing POCT resources to better support the population in the study area, and resulted in a decrease of travel time of 16.7 %. The second and third analyses involve keeping the current POCT where they are and adding an additional five and ten POCT, optimally located, to the health network. These analyses decreased travel time from populated places to care by 18.4 and 20.5 % respectively. It is worth noting that comparing rearranging existing resources to adding 5 additional POCT only results in a decrease of 1.7 %, suggesting little benefit.
In the last two analyses, rearranging existing resources while adding five and ten POCT, the overall effect on health care access is the same. The differences being a decrease to diagnosis by 3.7 with an increase to care by while the addition of the second additional five POCT decreases the time to diagnosis by 3.7 min negated by and increase to care of the same. Thus, it could be concluded that adding the five extra tests does not warrant the costs of adopting them, since overall it does not add any benefit.
This study successfully demonstrated that a GIS helps understand the role of diagnostic technologies within a SWN in rural Thailand. Although preliminary in nature, this analysis could help make informed decisions to improve health networks worldwide. It is important, however, to recognize some important limitations of the spatial analysis.
Complete, accurate, and appropriate data is one of the most important and biggest challenges for this analysis. OSM data was used for both populated places and roads because no other data was available. While OSM data is often criticized for its incompletes and inaccuracy in developing countries like Thailand, it may still be the most appropriate dataset for this type of analysis due to its availability.
The purpose of the road data is to help understand the generalized travel patterns for the population. This can be done by using a wide range of data sets from different sources. Government data sets like the US topologicaly intergrated geographic encoding and referencing (TIGER) provide a standardized data source collected by the government, however lack the integration of travel times or may not be available in some countries. Additionally, commercial data sets like Tom Tom®, Google Maps® and HERE® can provide the necessary travel time data, but are costly. When available government or commercial data sets may present a viable option to replace OSM used in this analysis.
While road data may be obtained from government or commercial sources, information about where people may originate is a source of considerable potential error. In this study populated places from OSM were used as origin points for travel. However, it is acknowledged that this data source may not be the most appropriate because there is no means to judge the accuracy or completeness of this of dataset and the spatial scale of aggregation may not be appropriate for this analysis.
To overcome incompleteness and inaccuracy of OSM data, census data can be used if available. For this study, no census data of appropriate spatial scale was available. When available, census aggregations may provide a better and more standardized way of obtaining origins of travel. However, even when official census data is available, the collection timeframe and the scale of spatial aggregation may suffer to the same problems OSM data has.
If population data cannot be obtained through outside sources, then it must be collected. While this can be a time and resource intensive process whose cost may not justify the benefits, a recent study showed that low cost collections of village locations can be done easily and rapidly .
No matter what dataset is used in this analysis to represent populations or road networks, consideration for how this data effects the analysis must be considered. Having inappropriate data may cause the analysis to miss subsets populations who could benefit from increased access to health resources.
The analysis assumes that an individual chooses to take the quickest path to their. There are several different ways in which an individual may realistically divert from this assumption. First, a patient who self-identifies as critically ill may decide to travel in a more direct path knowing their ending destination will be Srinagarind Hospital. Second, a patient may not have the knowledge of or does not trust certain roads or health resource facilities and thus alters their path, from a more optimal one.
These diversions from the assumed paths represent the concept of SWN, where individual paths through the network directly reflect the cultural or social pressures that exist. A way to accurately include this into the analysis is to survey the population or health professionals to better understand their attitudes towards their health network and understanding the critical paths.
Another assumption is that the quickest path to diagnosis will lead to the quickest path to care. The analysis only optimizes the time to diagnosis. Traveling to the closest facility with appropriate diagnosis may not be the most effective overall care path. Thus, another way to pre-position supplies is to optimize travel from populated places such that travel to care is optimized.
While the analysis takes relatively little processing power or time, the software cost may be prohibitive for use by the rural health networks that would benefit the most. The open source community has software packages that can easily replicated the workflows used in this analysis. While the cost for the software is free the technical requirements are considerably higher to implement something similar. Future efforts will be done to implement an open source application that can be easily applied.
Recommendations for the future
Health costs have risen consistently over the last few years. This increase is partly associated with technological developments. At the same time, integration of new technology is often slow, difficult, and expensive. Unfortunately, health networks may not have the means to understand the benefits of adopting technologies or to compare those benefits to the costs of adoption . The research described here is the basis of what has become the spatial care path™ (SCP). The SCP is defined as the most efficient route available to individual patients within health networks, and evaluates health networks in a spatial context to improve decision-making and reduce costs .
In this analysis we applied SWN analysis in a simplified fashion. More advanced methods incorporating SWN dynamics and detailed location-allocation models could greatly improve this analysis. A recent study published uses a modified P-median model to understand where best to place new facilities within health networks . The limitations of these models are that they focus on generalized health access instead of focusing on the SCP for individual diseases. The SCP analysis could be incorporated into these advanced methodologies to provide a unique perspective on the spatial benefits.
The SCP concept has been expanded to include other disease paths including diabetes management and the Ebola outbreak [22, 23]. Research is currently underway to understand how variations in data completeness, health network infrastructure, and topological variations impact the results of the analysis described here. This project will explore if a compelling argument can be made using this type of analysis to motivate the integration of POCT into health networks. One of the additional outcomes of this study is a discussion on the data sources used and their limitations, ideally motivating the creation of more appropriate datasets in the future. The hope for the SCP is that it will fit into future decision making for all health networks by supporting the evaluation of care paths in a spatial framework.
Point of care technologies enables diagnostic decisions in new locations streamlining care paths, improving outcomes, and reducing inefficiencies. Spatial analysis estimated current mean travel time to cardiac care to be over 3.5 h. This could be reduced by as much as 25 % with integration of POCT, demonstrating the utility of a GIS to help make decisions. Although not without its limitations, this spatial analysis can help health networks (a) understand the benefit of integrating POCT within a rural health system, and (b) provide evidence of the potential benefits which can be used to justify the costs.
This analysis uses a two-step method that defines health care access by quantifying how an individual would travel from (a) a place of origin to a location of diagnosis and then (b) from the diagnosis to a location of care. For this analysis travel is considered on road only as other travel medians (air or railway) are cost prohibitive or not available for the majority of the population. This process involves four data types: (a) populated places, (b) roads, (c) health resource facilities, and (d) cardiac diagnostic resources. Two implementation strategies were used to improve this health access. The first evaluated a widespread POCT integrations strategy; the second analyzed a limited resource strategy.
A portion of the region of Isaan, Thailand, was used as the study area because it was previously investigated using paper-based surveys by the researchers in the Kost et al.  study. The Isaan region, often referred to simply as the Northeast region of the country, had 21,305,000 people in 2010 and is considered to contain the country’s most rural and poorest areas.
Roads and populated places
User submitted open data that emphasizes local knowledge. Contains both line and point data with attributes
Downloaded May 29th, 2014
MapMagic 2013: Thailand
Health resource facilities
Proprietary point data collected and sold by a Thai based company
Assumed to represent the facilities that exist as of 2013
Global Administrative Areas
Lines representing political boundaries. Note: Bueng Kan was established in 2011 and the new boundary was added manually
Downloaded May 29th, 2014
Kost et al. 
Diagnostic technology locations
Resources for available cardiac diagnostic resource were surveyed in Isaan Region. Contains attributed point data
Collected throughout 2009 and 2010.
OpenStreetMap populated place descriptions (descriptions modified from OpenStreetMap places metadata)
Largest urban settlement in province which usually has more than 100,000 people
A distinct section of an urban settlement with unknown population
A second tier urban settlement of local importance with more than 10,000 people
A smaller distinct settlement, smaller than a town with <10,000 people
A smaller rural community with 100–200 people
OpenStreetMap road descriptions
Estimated speed (MPH)
The most important roads in a country’s system that aren’t motorways
The next most important roads in a country’s system. (Often link larger towns.)
The next most important roads in a country’s system. (Often link smaller towns and villages.)
The next most important roads in a country’s system
The least most important through roads in a country’s system—i.e. minor roads of a lower classification than tertiary, but which serve a purpose other than access to properties
Roads which are primarily lined with and serve as an access to housing
For access roads to, or within an industrial estate, camp site, business park, car park etc
Residential streets where pedestrians have legal priority over cars, speeds are kept very low and children are allowed to play on the street
A road where the mapper is unable to ascertain the classification from the information available
Roads added to the network to ensure connectivity between health facilities and population aggregation locations
Health resource facilities
Health resource facility by province and type
Health promoting hospitals
Diagnostic technology locations
Data on the location of point-of-care technology within the Isaan region was collected by Kost et al. . These locations may not represent current POCT status within the provinces; however the locations can be used to demonstrate the effects for the implementation strategies.
Implementation of the spatial analysis required several stages: (a) creating a road network, (b) quantifying health care access under current conditions, (c) quantifying health care access under widespread integration strategies, and (d) using a location-allocation algorithm to select best implementation sites for new POCT locations under limited resource strategies.
It was necessary to ensure that the point locations for health facilities and population centers fell on the network, otherwise travel between points could not be determined. An additional 2331 road segments were created as straight lines from the populated place and health resource facilities points to the nearest OSM road segment. These connector roads accounted for 738.3 miles of new roads, with an average length of 0.32 miles and longest length of 6.3 miles.
OSM data is not provided in a fully developed network format. To ensure connectivity between all elements, the Network Analyst Integrate Tool was used to snap together all roads segments within 10 m. No turns were included in the network, meaning no cost was added to the routes generated from moving to one element to another and that it is possible to turn in any direction at any junction. The small-scale alterations to the road network dataset were designed to create a functional network and are expected to have little effect on the results.
Quantifying health care access
Once a network has been created it is possible to use the closest facility tools of the ArcGIS Network Analyst extension to determine the fastest route between pairs of origins and destinations. Total health care access can then be determined as the summation of travel times, weighted by population, from (a) each populated place to diagnosis, and (b) each point of diagnosis to care.
Current health care access was determined by calculating the total shortest distances between populated places, health resource facilities indicated to have POCT from the Kost et al. study , and Srinagarind hospital in Khon Kaen.
Exploring health care access under widespread integration strategies involved two analyses to evaluate how implementing POCT at different administrative levels will affect health care access. These simulate situations in which the health network adopted policies that dictated certain facilities had to implement POCT. Two scales of integration were used. The first quantified health care access if POCT existed in every hospital. The second analysis quantified health care access if POCT existed in every health resource facility (hospitals and health promoting hospitals).
Rearranging existing POCT to optimal locations.
Optimizing placement of 5 additional POCT while keeping the existing resources where they are.
Optimizing placement of 10 additional POCT while keeping the existing resources where they are.
Optimizing placement of 5 additional POCT while allowing the existing resources to be rearranged.
Optimizing placement of 10 additional POCT while allowing the existing resources to be rearranged.
geographic information systems
affordable, sensitive, specific, user-friendly, rapid/robust, equipment-free, and deliverable
spatial care path™
WJF carried out the method creation, data analysis, and writing. KK helped conceive the study and draft the manuscript. GK provided expert review and guidance for the manuscript. All authors read and approved the final manuscript.
Spatial care path™ is a trademark by William Ferguson and Gerald Kost, Knowledge Optimization®, Davis, CA. Figures and tables were provided courtesy and permission of Knowledge Optimization®, Davis, California, and Visual Logistics, a division of Knowledge Optimization®. We would like to thank the UC Davis Library System for providing funding for this article to be open access. Additionally, we would like to thank Dr. Darren Ruddell and Dr. Yao-Yi for their contributions.
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Clark RA, Coffee N, Turner D, Eckert KA, van Gaans D, Wilkinson D, et al. Application of geographic modeling techniques to quantify spatial access to health services before and after an acute cardiac event. Health Serv Outcomes Res. 2012;125:2006–14.Google Scholar
- Coffee N, Turner D, Clark RA, Eckert K, Coombe D, Hugo G, et al. Measuring accessibility to cardiac services using geographic information systems. Appl Geogr. 2012;34:445–55.View ArticleGoogle Scholar
- Ranisinghe I, Turbull F, Tonkin A, Clark RA, Coffee N, Brieger D. Comparative effectiveness of population interventions to improve access to reperfusion for ST-segment-elevation myocardial infarction in Australia. Circ Cardiovasc Qual Outcomes. 2012;5:429–36.View ArticleGoogle Scholar
- Blandford JI, Kumar S, Luo W, MacEachren AM. It’s a long, long walk: accessibility to hospitals, maternity and integrated health centers in Niger. Int J Health Geogr. 2012;11:24.View ArticleGoogle Scholar
- Munoz UH, Kalestal C. Geographical accessibility and spatial coverage modeling of the primary health care network in the Western Province of Rwanda. Int J Health Geogr. 2012;11:40.View ArticleGoogle Scholar
- Nesbitt RC, Barysch S, Laub A, Soremekun S, Manu A, Bret Kirkwood, et al. Methods to measure potential spatial access to delivery care in low- and middle-income countries: a case study in rural Ghana. Int J Health Geogr. 2014;11:25.View ArticleGoogle Scholar
- Vanderschuren M, Mckune D. Emergency care facility access in rural areas within the golden hour?: Western Cape case study. Int J Health Geogr. 2015;14:5.View ArticlePubMedPubMed CentralGoogle Scholar
- Luke DA, Stamatakis KS. Systems science methods in public health: dynamics, networks, and agents. Annu Rev Public Health. 2012;33:357–76.View ArticlePubMedPubMed CentralGoogle Scholar
- Watts DJ, Strogatz SH. Collective dynamics of ‘small-world’ networks. Nature. 1998;393:440–3.View ArticlePubMedGoogle Scholar
- Peeling RW, Mabey D. Point-of-care tests for diagnosing infections in the developing world. Clin Microbiol Infect. 2010;16(8):1062–9.View ArticlePubMedGoogle Scholar
- Curtis CM, Louie RF, Vy JH, Ferguson WJ, Lam M, Truong AT, et al. Innovations in point-of-care testing for enhanced United States disaster caches. Am J Disaster Med. 2013;8(3):181–204.View ArticlePubMedGoogle Scholar
- Lewandrowski EL, Lewandrowski K. Implementing point-of-care testing to improve outcomes. J Hosp Adm. 2013;2(2):125–32.Google Scholar
- Garcia PJ, Carcamo CP, Chiappe M, Valderrama M, La Rosa S, Holmes KK, Mabey DCW, et al. Rapid syphilis tests as catalysts for health systems strengthening: a case study from Peru. Public Libr Sci. 2013;8(6):1–7.
- Mabey DC, Sollis KA, Kelly HA, Benzaken AS, Bitarakwate E, Changalucha J, et al. Point-of-care test to strengthen health systems and save newborn lives: the case of syphilis. Public Libr Science. 2012;9(6):1–6.
- Kost GJ, Yu JN, Tran NK. Theory, principles, and practice of optimizing point-of-care small-world networks for emergency and disaster care. Trans Biomed Eng. 2010;11(2):96–101.Google Scholar
- Law G. 2014. Provinces of Thailand. http://www.statoids.com/uth.html. Accessed 2 April 2015.
- Ferguson WJ. Modeling patient access to point-of-care diagnostic resources in a healthcare small-world network in rural Isaan, Thailand. Master Thesis, Spatial Sciences Institute, University of Southern California, Lost Angeles, CA. 2014.
- Munaneza F, Hirschhorn LR, Amoroso CL, Nyiranzinyoye L, Birru E, Mugunga JC, et al. Leveraging community health worker system to map a mountainous rural district in low resource setting: a low-cost approach to expand use of geographic information systems for public health. Int J Health Geogr. 2014;13:49.View ArticleGoogle Scholar
- Price PP, John AS. Point-of-care testing: making innovation work for patient-centered care. Washington DC: AACC Press; 2012.View ArticleGoogle Scholar
- Kost GJ, Ferguson WJ. Principles of point of care culture, the spatial care path,™ and enabling community and global resilience. Elect J Intl Fed Clin Chem. 2014;25:4–23. http://www.ifcc.org/media/260912/eJIFCC%20August%202014.pdf. Accessed 2 April 2015.
- Jia T, Tao H, Qin K, Wang Y, Liu C, Gao Q. Selecting the optimal healthcare centers with a modified P-median model: a visual analytic perspective. Int J Health Geogr. 2014;13:42.View ArticlePubMedPubMed CentralGoogle Scholar
- Kost GJ, Ferguson WJ, Hoe J, Truong AT, Banpavichit A, Kongpila S. The Ebola spatial care path:™ accelerating point-of-care diagnosis, decision making, and community resilience in outbreaks. Am J Disaster Med. 2015;10(2):121–43.
- Kost GJ, Ferguson WJ. Spatial care path™ strengthen links in the chain of global resilience—disaster caches, prediabetes, and Ebola virus disease. Point Care. 2016 (in press).