Geographical access to health care facility among residents of Kalutara District in 2019- A cross sectional study

doi:10.21203/rs.3.rs-4827521/v1

Download PDF

Research Article

Geographical access to health care facility among residents of Kalutara District in 2019- A cross sectional study

https://doi.org/10.21203/rs.3.rs-4827521/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Achieving universal health equity remains a global challenge. The World Health Organization (WHO) advocates for universal health coverage (UHC) to ensure accessible, quality healthcare for all, regardless of financial status. Geographic Information System (GIS) technology has been instrumental in assessing healthcare accessibility, revealing gaps and informing policy decisions. This study investigates the geospatial distribution of Western healthcare facilities in Kalutara district, Sri Lanka, evaluating their accessibility.

Methods

A descriptive cross-sectional study was conducted from May 2019 to October 2019. The study geospatially mapped all healthcare institutions practicing Western medicine within Kalutara district and a 5km buffer zone beyond its boundaries. Data were collected using Epicollect5, a mobile application, ensuring accuracy within 10-15 meters. Variables included geospatial information and service availability. Data analysis was performed using ArcGIS 10.3, assessing coverage within buffer zones of 2km to 4km radii around each healthcare institution.

Results

General practitioners (GPs) covered 87.8% and 98.1% of the population within areas of 2km and 4km radii. All government institutions covered 21.8% and 67.6% of 2km and 4km radial buffer areas, while the coverage of private hospitals amounted to 3% and 8.8% respectively (i.e. within 2 km and 4km). It improved up to 64.1% and 90.0% by the inclusion of general practices. Majority of government hospitals and all private hospitals had inward care and 24x7 care. Blood banks were available only in 83.3% of specialized government hospitals. Of general practices, 10.6% had 24x7 care and 37.5% of private hospitals had ICU facilities. The study highlighted the significant role of general practitioners (GPs) in providing accessible healthcare, especially after routine working hours.

Conclusions:

GP settings contributed mainly in ensuring geographical access to healthcare. Hence, GP services need to be considered in all attempts of improving healthcare service delivery.

Universal health coverage

Healthcare Accessibility

General practitioners

Kalutara district

The goal of maintaining the health of every human being becomes a global challenge. The opportunities offered for the people by the system they live in, in achieving complete well-being, function as a critical factor in enabling them to face this challenge successfully. Given these facts, the World Health Organization (WHO) emphasizes that fair opportunities must be provided for everyone to achieve their maximum health potential without anyone being placed in a disadvantaged position (1). Concerns on equity in achieving health were in the global priority agenda a century ago. It was realized during the last few decades that in the recent context of development, unless anticipated carefully, marginalized populations may not fully receive the benefits of available health services due to lack of access, quality issues and financial hardships (2).

In order to have health equity the concept of Universal Health Coverage (UHC) has been developed by WHO in 2005. Main aim of the UHC to provide accessible, quality healthcare to all irrespective of their financial status. UHC is defined as “the ability of all people and communities to use the promotive, preventive, curative, rehabilitative and palliative health services they need, of sufficient quality to be effective, while also ensuring that the use of these services does not expose the user to financial hardship” (3). In providing the UHC the access to the health services, quality of health services and financial risk protection are considered.

Access to health services assures that all citizens of a country are treated with equity in physical accessibility, affordability and acceptability of the health services provided. Geospatial analysis is widely used to measure proximity, assess health service availability, identify resource gaps, and guide public health policies. By providing detailed spatial data, this approach enhances strategic planning and decision-making, ultimately improving the efficiency and effectiveness of health interventions and the allocation of resources. A recent study done in united States of America depicted around 5% of the eligible population does not have access to lung cancer screening facilities within 40 miles (4).

The application of GIS technology in the health industry has transformed the understanding of health facilities and geographic accessibility. GIS integrates and analyzes diverse data types such as health facility locations, population demographics, transportation networks, and geography, providing crucial insights into health service distribution and accessibility. GIS identifies gaps in coverage, optimizes health service routing, and determines efficient emergency transport routes (5). It helps map hospital locations and population distribution, guiding policymakers in allocating new facilities to ensure universal access. GIS also visualizes disease patterns and hotspots, aiding targeted interventions (6).

A recent study in India revealed that 11% of villages lacked ambulatory and immunization care within a 2 km and for a 5 km radius. Travel time accessibility analysis showed that walking provided the lowest level of accessibility. Enabling motorized travel improved accessibility, with the highest accessibility for privately owned vehicles. Even in the best-case scenario, around two-thirds of the population remained without access to all service packages (7).

Geographic access to government health services in Sri Lanka is considered satisfactory compared to the regional peers. Households are on average 2.5 kilometers from a maternity clinic, four kilometers from a government dispensary, and 6.5 kilometers from a hospital. About 93% of the population is within 15 kilometers of a hospital. Outpatient services are open until four pm, but emergency care is available twenty-four hours seven days per week (24×7) (8). A study conducted in the rural sector of Sri Lanka revealed that the physical proximity of health care services does not guarantee the utilization of that particular service due to bypassing of institutions (9). Since the Sri Lankan health system has no gatekeeping mechanism, the accessibility to health care services cannot be entirely evaluated by the geographical accessibility to the nearest hospital.

This study was conducted in Kalutara district of the Western province. Kalutara, the fifth most populated district in the country, has a diverse ethnic and religious population distributed across urban (8.9%), rural (88%), and estate (3.1%) sectors, providing a sampling frame that approximates the national population structure (10). Furthermore, Kalutara district has a District General Hospital (DGH), a network of secondary and primary care hospitals and PMCUs. It also consists of a network of private hospitals and also a wider network of GPs and laboratory services catering the health care needs of the population. Thus, as a health system study, this study has fulfilled both requirements in relation to the patients as well as the health system holistically. Hence this study was done to analyze geospatial distribution of Western healthcare facilities to residents of Kalutara district.

This was conducted as a descriptive cross-sectional enumerative study to geo-spatially map all categories of healthcare institutions practicing Western medicine in Kalutara district.

The geo-spatial mapping was done from May 2019 to October 2019. All healthcare institutions practicing OPD services related to Western medicine (Government hospitals, private hospitals, general practitioners) in and around a 5km buffer zone from district boundary of Kalutara district. This buffer zone was included with expert inputs as facilities in this area might be accessed by the residents who reside within Kalutara closer to its border.

Curative healthcare institutions practicing western medicine (i.e. District General Hospitals, Base Hospitals, Divisional Hospitals and Primary Medical Care Units), healthcare institutions registered under the medical ordinance (i.e. Private hospitals), healthcare providers registered under medical ordinance (i.e. General practitioners: full time and part time) and nursing homes were included. Institutions registered but not in operational state (closed down or yet to be opened) channeled consultation centers, mobile medical services, medical centers in occupational settings, dental clinics, MOH clinics were excluded from the study. A census of all allopathic healthcare institutions which meet the eligibility criteria was included in the mapping. No sampling method was applicable as all eligible healthcare institutions were included.

A mobile application (i.e. “Epicollect5”) with offline data collection capability was used as the study tool. Epicollect5 has been developed by the Imperial College London and is financially supported by the Welcome Trust Fund. It includes a standard procedure on collecting geospatial data points of locations as to be unique and identical for each location. Both Android and iPhone devices are supported by it. This system has a documented accuracy of within 10 to 15 meters if standard procedures are adhered to when the data collectors physically visited each healthcare institution for the acquisition of relevant data. The variables included geospatial information on the healthcare institutions as well as information on their service availability (facility profile and 24×7 availability to the public). The list of facilities was cross-checked with the Ministry of Health, RDHS, GMOA Branch Unions, GP Association and the list of facilities available at Private Medical Institution Regulatory Authority (PMIRA).

Ten pre-intern medical graduates were trained by the principal investigator and the supervisor who were thorough in using the data collecting software. Each data collector used a free mobile application (Epicollect5) which works in smart phones with offline data collection capability (11).

The data was collected in real time to the prepared dataset. Geospatial analysis was performed with ArcGIS 10.3 application. Spatial maps were developed for each category. The institution layer files were overlaid with those available for GN divisions’ layer files with population information for assessing coverage at selected distances. The institutions were stratified as: specialized government hospitals (secondary and tertiary government hospitals), other primary government institutions, all government institutions, general practices, all private institutions and all healthcare institutions.

After adjusting for the overlaps, the percentage geographical area and the number of people around each stratum, within buffer zones with a radius of 2km, 2.5km, 3km, 3.5km and 4km were calculated. The updated GN level population information obtained from the District Secretariat office was used to calculate the population under each of these buffer zones. An assumption of equal distribution within the GN area was made for the calculation.

Ethical approval

was obtained by the Ethics Review Committee of National Institute of Health Sciences (NIHS/ERC/19/82R)

The distribution of healthcare institutions within the Kalutara district and an area including a five-kilometer radial buffer zone around it is shown in table 1and Fig. 1. General practices were the most commonly found healthcare institution. The ratio between full time to part time GPs was approximately 1: 9 and there were 72 (10.3%) full time GPs.

Table 1

Distribution of healthcare institutions providing western medical services in Kalutara district and within a 5km radial buffer zone beyond it
Type of Institution	Frequency	Percentage
Specialized Government Hospitals	6	0.8
Primary Government Institutions
-Divisional Hospitals	22	2.9
-Estate Hospitals	4	0.5
-Primary Medical Care Units	18	2.4
Private Hospitals	8	1.1
General Practices	697	92.3
Total	755	100.0

*An “Estate hospital” is a hospital that treats patients from a particular estate it serves. Treatment is done by an estate medical officer.

The percentage of the population covered by each level of healthcare institutions, when a buffer zone of 2km was considered around each health institution within that particular level, is shown in Table 2. It also shows the percentage of the geographical area covered. In preparing this, it had been assumed that the population density in a given GN division was constant throughout it.

Within a buffer zone of 2km radius, specialized government hospitals covered 9.2% of the population in Kalutara district. Even with the addition of other government institutions such as PMCU and District Hospitals, it covered only 34.4% of the population. When the general practices were added, it covered 89.0% of the population which was 54.6% more than the population covered by government institutions alone. The increment of the geographical area coverage in a 2km radial buffer zone by adding general practices to government institutions is shown in Fig. 2.

Table 2

Population coverage and geographical area coverage considering a 2km radial buffer area around each type of healthcare institutions concerned
	Population	Percentage (%)	Area(km²)	Percentage (%)
Specialized Government Hospitals	112628	9.2	47.2	2.9
Primary Government Institutions	313976	25.7	302.1	18.9
All Government Institutions (Excluding Overlap)	419838	34.4	348.3	21.8
Private Hospitals	149178	12.2	47.6	3.0
General Practices	1073047	87.8	979.7	61.3
All Private Institutions (Excluding Overlap)	1073047	87.8	979.7	61.3
All Healthcare Institutions (Excluding Overlap)	1087456	89.0	1025.4	64.1

*calculated population is rounded to the nearest integer.

The distribution of population and geographical areas covered similarly within 2.5km, 3km, 3.5km and 4km buffer zones are shown in Table 3. These show that, when a buffer zone of 2.5 km is considered, all government institutions covered less than 48% of the population. The increment of geographical area coverage by adding general practices to government institutions in 2.5km radial buffer is shown in Fig. 3

Table 3

Population coverage and geographical area coverage considering a 2.5km radial buffer area around each type of healthcare institutions concerned
Radial buffering area	Type of healthcare institution	Population	Percentage (%)	Area (km2)	Percentage (%)
2.5km	Specialized government hospitals	173395	14.2	72.2	4.5
	Primary Government Institutions	428346	35.0	459.1	28.7
	All government institutions (Excluding overlap)	577823	47.3	525.8	32.9
	Private hospitals	200001	16.4	66.8	4.2
	General Practices	1132683	92.7	1135.3	71.0
	All private institutions (Excluding overlap)	1132683	92.7	1135.3	71.0
	All healthcare institutions (Excluding overlap)	1143794	93.6	1177.6	73.7
3km	Specialized government hospitals	229272	18.8	101.7	6.4
	Primary Government Institutions	547657	44.8	633.4	39.6
	All government institutions (Excluding overlap)	728922	59.6	719.4	45.0
	Private hospitals	249836	20.4	126.9	7.9
	General Practices	1166913	95.5	1255.5	78.6
	All private institutions (Excluding overlap)	1166913	95.5	1255.5	78.6
	All healthcare institutions (Excluding overlap)	1174873	96.1	1294.4	81.0
3.5km	Specialized government hospitals	277982	22.7	135.1	8.4
	Primary Government Institutions	665904	54.5	811.5	50.8
	All government institutions (Excluding overlap)	858616	70.3	909.1	56.9
	Private hospitals	295864	24.2	113.8	7.1
	General Practices	1187158	97.1	1347.5	84.3
	All private institutions (Excluding overlap)	1187158	97.1	1347.5	84.3
	All healthcare institutions (Excluding overlap)	1192818	97.6	1382.6	86.5
4km	Specialized government hospitals	323907	26.5	172.2	10.8
	Primary Government Institutions	782193	64.0	980.9	61.4
	All government institutions (Excluding overlap)	972714	79.6	1080.7	67.6
	Private hospitals	340541	27.9	140.8	8.8
	General Practices	1198665	98.1	1407.5	88.1
	All private institutions (Excluding overlap)	1198665	98.1	1407.5	88.1
	All healthcare institutions (Excluding overlap)	1202094	98.4	1438.5	90.0
*calculated population is rounded to the nearest integer.

Table 3. Population coverage and geographical area coverage considering a 2.5km radial buffer area around each type of healthcare institutions concerned

As shown in Table 3 even within a buffer zone of a 3km radius, the percentage of the population covered was less than 60%. Private hospitals alone only covered around 20% of the population. However, the general practices alone could cover more than 95% of the population. The increment of the geographical area coverage by adding general practices to government institutions in a 3km radial buffer area is shown in Fig. 4.

Similarly, when the radius of the buffer zone was considered to be 3.5km, government institutions covered around 70% of the population and private hospitals covered about one fourth as shown in Table 3. The coverage by general practices exceeded 97%. The increment of the geographical area coverage by adding general practices to government institutions in a 3.5km radial buffer area is shown in Fig. 5.

Within a buffer zone with a radius of 4km, all government institutions covered nearly four fifth of the population and private hospitals alone covered less than 30%. The increment of the geographical area coverage by adding general practices to government institutions with a 4km radial buffer zone is shown in Table 3 and Fig. 6.

Table 4

Comparison of population coverage and geographical area coverage considering buffer areas of different radii around each healthcare institution concerned (all healthcare institutions combined)
	Population	Percentage (%)	Area(km²)	Percentage (%)
2 km radial buffer	466798	38.2	363.1	22.7
2.5 km radial buffer	618953	50.6	541.8	33.9
3 km radial buffer	762277	62.4	733.8	45.9
3.5 km radial buffer	887066	72.6	922.3	57.7
4 km radial buffer	993191	81.3	1092.7	68.3

*calculated population is rounded to the nearest integer.

Even within a buffer zone with a radius of 2.5km, GPs ensured that the population coverage was more than 93%, which would have otherwise been around 50%. Other primary care government institutions had contributed in raising the population percentage covered by the specialized intuitions alone, even though the contribution was not that high as that of GPs. All government institutions provided services to more than 400,000 people at a 2km radial buffer zone and when the buffer radius was increased to 4km the covered population increased to nearly one million. In order to increase the population provided the services at a 2km radial buffer zone to six hundred thousand, GPs’ contribution needed to be taken into consideration.

This is the first documented study on geo-spacial distribution of Western healthcare facilities to residents of Kalutara district. Geospatial mapping of the healthcare institutions within component 1 marks a milestone in research related to UHC in Sri Lanka as it marks a collaboration of new technology and healthcare research. Similar work has been quoted as milestones in research related to UHC. (11, 12, 13, 14).

Obtained locations from this mapping enabled valid estimations of the percentages of the population covered which would not have been possible if a secondary data source was used. In addition, coupling of GIS technology, accurate geo locations and usage of lowest administrative level population densities made it possible to get reliable estimates for the population under care. Furthermore, rather than being a paper-based method which would need entering of data as an additional step, usage of compatible mobile devices enabled an automated data entering followed by subsequent quick, complete and reliable analyses.

In the mapping of the curative healthcare institutions within the Kalutara district and a five-kilometer radial buffer zone around geospatially, as expected it revealed that the clinics of the general practitioners were the most common in numbers being nearly 10 times higher in number than all other government and private institutions. This highlights the importance of the GPs in the provision of primary healthcare and in ensuring the access to health services. This was further proven by the estimation of the percentage of population covered with geospatial mapping. When the GP stations were added, the percentage population covered by the health institutions increased significantly. There was no improvement of UHC accessibility in terms of geography by the private hospitals, which were located near the main hospitals; rather it was vastly improved by the GP settings.

The above finding is complemented by the arrangement of the healthcare service delivery in Sri Lanka. Government doctors are allowed to engage in private practice after their working hours (15). Given the relatively lower salary offered in the government sector, the GP practice has hence become a complementary income source for medical officers employed in the government sector. This has resulted in medical officers establishing general practices close and accessible to people at the same time increasing the retention of medical professionals within rural areas. In addition to this, some GPs may be involved in the practice due to other reasons like sharpening of clinical skills and as a social service.

Overall, around 90% of the GPs engaged in part time GP-practice after routine working hours in the government sector. This implies that the availability of them as GPs in the daytime was limited in contrast to the 10% minority who engaged in full-time GP practice. Even though this would not affect the working population who usually seek care after working hours, the dependent categories like elders, children would gain more benefits if the availability of the GPs is increased.

On the other hand, with the increase of the availability of GPs while access is improved, it raises a question as to how the health-related financial risk protection would be affected. The GP services, even though may not be as costly as that of private hospitals, would still incur out-of-pocket expenditure. Hence the overall influence of improving the GP coverage on the Universal Health Coverage is to be evaluated with further research. Anyhow the attempts of improving the quality of care as stated above will facilitate raising the UHC parameters.

Both primary level healthcare institutions as well as GPs contribute significantly in ensuring healthcare access. The quality of healthcare in these institutions can be ensured; the UHC would be easily ensured owing to the high level of accessibility as shown with geospatial mapping. Hence, the present study sheds light on the importance of auditing and ensuring quality service delivery by GPs. A good method of conduct of these audits can be “internal implementation” by the GP-setting itself. Firstly, policy planners may focus on preparing tools in the assessment of the quality and disseminate to the GPs to apply within the settings. This must be promoted as a quality improvement process rather than a fault-finding exercise. Secondly, measures must be taken in enabling the GPs in raising their quality-of-service delivery. Several measures which could be used include dissemination of latest evidence and preparation of GP-related clinical guidelines. Dissemination of evidence could be coupled with new technological advances like email. Other customized possibilities like offline-web-based workshops must be explored. Given the fact that most GPs use updated communication strategies like smart-phones, a relatively large proportion could be covered by these mechanisms.

However, it is well known that overcrowding is commonly found in secondary and tertiary care institutions. Clients are bypassing the other government institutions and reaching these. The above facts give rise to several implications in relation to UHCs when access as well as quality of care are concerned. Firstly, even though accessibility is there, Sri Lankan health clients seem to be underutilizing the lower level of government institutions.

The reason for this must be sought. It may be related to a perceived deficiency of the quality of care offered by these institutions. However due to the overcrowding at the specialized level hospitals, the quality is potentially compromised in return. If reasons are explored and the utilization of the lower level of healthcare institutions can be increased, it would help in improving healthcare access as well as help in raising the quality-of-service delivery in the specialized institutions.

The present study had several limitations. In measuring the access to care, only the geographical access was assessed. However, within the scope of this study, measuring the geographical access was sufficient. Furthermore, in describing the access by type of the healthcare institution, all private hospitals were categorized into a single group. This was due to the fact that a reliable “level of care” classification was not available for private institutions. Getting these into a single group would not have affected the internal validity but has some implications on the generalizability. Even though the primary level institutions in workplace settings were within the exclusion criteria, if those were included, additional information would have been acquired.

The highest proportion of population coverage for access of the healthcare facilities (i.e. more than 50%) was given by the general practitioners within and in the immediate surroundings of the Kalutara district. All government institutions covered 21.8% of the geographical area. It improved up to 64.1% by inclusion of general practices. (General practice coverage alone was 61.3%).

Currently underutilized government primary healthcare units need to be strengthened to improve the accessibility to the underprivileged population. Other primary healthcare services (e.g. preventive, promotive etc.) should be promoted.

Supplementary file

WHO

World Health Organization

UHC

Universal health coverage

GIS

Geographic Information System technology

GPs

General practitioners

Ethics approval and consent to participate- Obtained from Ethics review Committee of National Institute of Health Sciences (NIHS/ETC/19/82R)

Consent for publication-Not applicable

Availability of data and materials- Not applicable.

Competing interests - The authors declare that they have no competing interests

Funding - Financial assistance from Primary health care system strengthening project (PSSP),

Authors' contributions- HND- participated in the designing of the study, data collection, and performed the statistical analysis, drafting the manuscript, Padmal- Supervisor, MDUB- data collection and data entry DSKD- data collection and data entry

All authors read and approved the final manuscript.

Acknowledgements- Dr Palitha Abeykon, Dr. Buddhika Mahesh and Dr Sameera Senanayaka who always guided me in the correct path through all the stages of this project.

World Health Organization. (2020). Health Equity. Health topics. Retrieved from https://www.who.int/topics/health_equity/en/
Alshamsan R, Leslie H, Majeed A, Kruk M. Financial hardship on the path to Universal Health Coverage in the Gulf States. Health Policy. 2017;121(3):315–20. 10.1016/j.healthpol.2016.12.012.
World Health Organization. Universal health coverage and universal access. Bull World Health Organisation. 2013;91:546–A546.
Sahar L, Douangchai Wills VL, Liu KK, Fedewa SA, Rosenthal L, Kazerooni EA, et al. Geographic access to lung cancer screening among eligible adults living in rural and urban environments in the United States. Cancer. 2022;128(8):1584–94.
Kapwata T, Manda S. Geographic assessment of access to health care in patients with cardiovascular disease in South Africa. BMC Health Serv Res. 2018;18(1):1–10.
Dos Anjos Luis A, Cabral P. Geographic accessibility to primary healthcare centers in Mozambique. Int J Equity Health [Internet]. 2016;15(1):1–13. http://dx.doi.org/10.1186/s12939-016-0455-0
Verma VR, Dash U. Geographical accessibility and spatial coverage modelling of public health care network in rural and remote India. PLoS One [Internet]. 2020;15(10 October):1–23. http://dx.doi.org/10.1371/journal.pone.0239326
World Bank. (2018). Sri Lanka: Achieving Pro-Poor UHC Without Financial Reforms. Retrieved from.
Weerasinghe MC, Fernando DN. ACCESS TO CARE IN A PLURAL HEALTH SYSTEM: CONCERNS FOR POLICY REFORMS. J Coll Community Physicians Sri Lanka. 2009;14(1):39–45.
Department of Census & Statistics. (2015). Census of Population and Housing 2012.
National Cancer Control Programme, Ministry of Health (Sri Lanka). Cancer Incidence Data: Sri Lanka 2007. Colombo, Sri LankaCentre for Genomic Pathogen Surveillance (2020), What is Epicollect5. Retrieved from https://docs.epicollect.net/
Environmental Systems Research Institute. (2020). Create and Share Maps, Analytics, and Data. Retrieved from https://desktop.arcgis.com/en/
Lo J, Amour C, Buchner D, Cross D, Brenner JL, Matevelo D. Mapping health facility characteristics and location in rural Tanzania. Paediatr Child Health. 2017;22(Suppl 1):e18–9. 10.1093/pch/pxx086.047.
Fisher R, Myers P, B. A. Free and simple GIS as appropriate for health mapping in a low resource setting: a case study in eastern Indonesia. Int J Health Geogr. 2011;10:15. 10.1186/1476-072X-10-15.
de Silva WI. The age structure transition and the demographic dividend: an opportunity for rapid economic take-off in Sri Lanka. Sri Lanka J Adv Social Stud. 2013;2(1):3–46. 10.4038/sljass.v2i1.5117.

No competing interests reported.

Download PDF

Reviews received at journal
07 Sep, 2024
Reviewers agreed at journal
06 Sep, 2024
Reviewers agreed at journal
06 Sep, 2024
Reviewers invited by journal
23 Aug, 2024
Editor invited by journal
02 Aug, 2024
Editor assigned by journal
01 Aug, 2024
Submission checks completed at journal
01 Aug, 2024
First submitted to journal
30 Jul, 2024

You are reading this latest preprint version

Geographical access to health care facility among residents of Kalutara District in 2019- A cross sectional study

Status:

Version 1

Abstract

Figures

Background

Methodology

Results

Discussion

Conclusions & recommendations

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1