Home Articles patial Surveillance of Epidemiological Disease: A Case Study in Ayutthaya Province, Thailand

patial Surveillance of Epidemiological Disease: A Case Study in Ayutthaya Province, Thailand

Spatial Surveillance of Epidemiological Disease: A Case Study in Ayutthaya Province, Thailand

Southanome Keola
[email protected] 

Mitstuharu Tokunaga
[email protected] 

Nitin Kumar Tripathi
[email protected]
Space Technology Applications 
and Research (STAR) program
School of Advance Technology (SAT)
Asian Institute of Technology (AIT), Thailand

Wisesjindawat Wisa
[email protected]
High Performance Computing Research 
and Development Division (RD-C)
National Electronics and Computer Technology
Center (NECTEC), Thailand

Ayutthaya province has been included in the UNESCO’s list of world heritage on 13 December 1991 (TAT, 1999). Due to rise in tourism the urbanisation increased rapidly and resulted in high population rise. This has further caused increase in some epidemiological diseases such as pneumonia, acute diarrhoea, food poisoning, conjunctivitis. Several researchers have studied different disease pattern in Thailand caused by changes in social and environmental conditions. The pattern of AIDS diffusion was found to be a hierarchical one in Chiang Mai, which is the most urbanised provinces, as its origin followed by a spread to lesser-urbanised areas, and it diffused along the main highways (Wisesjindawat, 2000). The Japanese Encephalitis increased due to the environment changes, particular climate in Chiangrai province of Thailand (Adsavakulchai, S. et al., 2001). The diarrhoea increased in children aged between 0-1 years due to less-breast feeding in Bang Pa In district, Ayutthaya province. This problem surfaced due to the collapse of economy. For survival many women became factory workers, therefore they had less time to breast feed their children, and that lead to diminished children immunities. Moreover they leave their children staying with elder people who lack appropriate nutrition knowledge. (Vilay, 1999).

In recent past several countries have implemented GIS in public health applications. This was instrumental in bringing forth the utility to this exciting technology. GIS can improve public health capabilities, and can assist in earlier detection of public health issues when compared to more traditional means. As issues can be visualised through maps, GIS can provide a mechanism to bring changes to communities and provide ability to do precise spatial analysis. GIS helps to ensure the accuracy of the spatial information and that public health information maps are clear and concise (Gould, 1993).

GIS has rich repertoire of powerful tools for decision making for public health issues. It has been applied in this work for spatial analysis of epidemiological surveillance disease. To meet the purpose of this study which is to find out the highest-ranking disease across the districts of Ayutthaya province together with the use of statistical and spatial analytical methods.

Study Area
Ayutthaya is the 11th largest province, out of 24 Central region provinces with an area of 2,547.62 sq. km. It is located in the latitude 14° 6′ 33″ N and longitude 100° 14′ 53″ E at the sea elevation of 3.50 meters near the Gulf of Thailand. Listed clockwise, bordering on Ayutthaya are Ang Thong, Lop Buri and Saraburi on the north, Saraburi on the east, Nonthaburi and Pathumthani on the south and Suphanburi on the west (Ayutthaya provincial Annual Report, 1999). The topographic features are mostly plain regions with many rivers and canals; no forest and mountainous land exist in this province (Fig. 1).

Table 1:The epidemiological diseases in Ayutthaya Province for the last decade

Local Administration consists 182 units which are 1 Changwat Administrative Organisation, 2 Urban (Phra Nakhon and Sena districts), 25 District Municipalities and 129 Sub-district Municipality Organisations. At present, the total population of Phra Nakhon Si Ayutthaya province is 736,517.

Ayutthaya is the first place in Thailand to establish the Community Health Service Center (CHSC) in 1990. It is one kind of the integrated health care system which aims to reduce patient density in the central hospital. The main task is to provide primary health care services. With a growing population, one CHSC responds for 200,000 people. There are three CHSC in Pra Na Khorn Si Ayuthaya, and it will establish more in the future (The Report of Seeking Behavior to Community Health Care Service Center, 1996). Others health facilities are: 16 hospitals, 205 local clinics, private clinics and some traditional clinics. (Ayutthaya Provincial Annual Report, 1999). Ayutthaya was selected to be the study area based on its interesting nature of topography and health care network.

Fig 2A: The distribution of eight high ranking diseases within the past 10 years (1991 – 2000)

Fig 2B: The distribution of four high ranking diseases within the past 10 years (1991 – 2000)

The surveillance disease distribution and its tendency
Secondary data of epidemiological surveillance cases from 16 districts in Ayutthaya province was used for selection of diseases with highest incident ranking cases. The data for the epidemiological disease was collected from the “Patient Record Form 506”. There are 72 epidemiological surveillance diseases in “Patient Record Form 506”, and out of those 31 surveillance diseases occurred in Ayutthaya province. (Annual Provincial Report 1999).

The disease selection category has confirmed that disease used to occur and widely spread in the past and it continues to spread unabated.

The observation tendency from chart
The epidemiological surveillance diseases cases per 100,000 persons in ten years were processed, and classified in two groups:

  • The epidemiological surveillance disease cases per 100,000 persons – more than 100/100,000
  • The epidemiological surveillance disease cases per 100,000 persons less than 100/100,000 and have trend to increase

As a result, two groups classified as high-ranking diseases in which first group had 6 diseases and second group 2 diseases (Table 1). The disease distribution in last ten years shows the rising tendency. Pneumonia shows linear up trend and gives an indication for further rise in the future (Figure 2A and 2B).

The tendency of disease from the correlation analysis
Correlation analysis is the statistical tool which can be used to describe the degree to which one variable is related to another. Frequently, correlation analysis is used in conjunction with regression analysis to measure how well the least squares line fits the data. Correlation analysis can also be used by itself, however, to measure the degree of association between two variables (G. William and Patton, Carl V., 1991). Two measures are presented here for describing the correlation between two variables time and case incidence. And, the results of correlation coefficient are listed in Table 2. According to Table 2, R-value of pneumonia is very high and the tendency graph becomes nearly linear. It brings out important information that the highest-ranking disease in Ayutthaya province is pneumonia.

Fig. 3: The location quotient distribution of five diseases from 1996-2000.

Disease tendency found by spatial analysis
The comparison of disease distribution through district was found by applying the Location Quotient (Eq 1) model. The quotient is a computed ratio between the local observed data (district) and the observed data of some reference unit (province). This ratio is calculated for all districts to determine whether or not the local observed data has a greater share of that reference unit. The location quotient (LQ) is an index for comparing an area’s share of a particular activity with the area’s share of some basic or aggregate phenomenon. (G. William and Patton, Carl V., 1991).

Where:

  • NDistrict and NProvince represent number of incident cases in district and province respectively,
  • PDistrict and PProvince represent number of population of the district and province respectively. The result of location quotient can be considered in three cases, such as:
  • LQ < 1 indicates a relatively less concentration of the activity in the area X, compared to the region as a whole
  • LQ = 1 means the area has a share of the activity in accordance with its share of the base
  • LQ > 1 means the area has more share of the activity than found at regional level

    The result of applying LQ is shown in the Figure 3, and its legend were considered such as:

  • LQ from 0-0.05 is the low disease distribution
  • LQ from 0.05-1 is the middle disease distribution
  • LQ from 1-1.55 is the high disease distribution

According to the tendency of high disease distribution observed by above method is the ratio of disease distribution in district (Eq. 1), and the difference of highest density area (the area has LQ> 1) in 1996 to 2000 found pneumonia is the high disease distribution (Table 3).

Fig. 4: The prediction of spatial pneumonia incidence in the next five years

Pneumonia Forecasting for Next Five Years

Pneumonia and population forecasting by projection
Projection techniques are quantitative methods for estimating for the future. Techniques to project the future are essential for making plans and for most policy decisions. In this process will predict the LQ in each district, where linear projection is suitable to predict as the R is 0.94. The prediction of pneumonia and population in the next five years were calculated using the Gibbs Method as shown in Eq. 2.1 and 2.2. This method is described in population projection, written in Annual Rural-Regional Planning Workshop (Rural-regional development planning students, 2000).

Population projection
Dp = D1 + D1 * (rp /100*t) ……………………. (Eq. 2.1)
Where :

  • D = population projection
  • D1 = number of population in the current year (2000)
  • rp = growth rate (%)
  • t = time interval (eq in 2005, t=2005-2000=5)

Pneumonia projection
Dd = D1 + D1 * (rd /100*t) ……………………. (Eq. 2.2)
Where :

Dd = pneumonia projection
D1 = number of pneumonia in the current year ( year 2000)
rd = growth rate (%)
t = time interval (for example in year 2005, t=2005-2000=5)
With the result of pneumonia incidence and population trend in next five years 2001 – 2005, the case per 100,000 rate was calculated and is shown in Table 4. From this Table, it is observed that this rate will gradually increase.

Pneumonia prediction by spatial distribution
Gibbs Method was applied for the prediction of pneumonia cases and population in each district. Location Quotient was used to find the density of spatial distribution in the next five years (2001-2005). Pneumonia is predominant within the districts that have high population density such as Bang Pa In, Uthai, Wang Noi, Na Kho Luang, Sena, etc. (Figure 4). In the period of 2001 – 2005, the prediction shows that the diffusion of pneumonia spreads from the western districts to the eastern districts. Most of western districts have industrial areas with high density trend of population growth, while agricultural areas are located in eastern districts.

Conclusion
It is difficult to understand the issues related to epidemic diffusion simply by statistical analysis as it lacks spatial information. Therefore, combination of both statistics and GIS methods is very useful to researchers to model the health related issues as GIS provides efficient capability to visualise the spatial data.

In this research, statistical analysis and GIS methods for identifying the highest ranking epidemic disease in Ayutthaya province were employed. The main results show that the correlation coefficient of pneumonia was found as the highest, which is 0.94 and it is considered as a widely spread disease in term of spatial diffusion, especially in urban areas with high trend of population growth.

Acknowledgement
Authors would like to express their sincere gratitude to the ASEAN Foundation for financially supporting this research. We are extremely grateful to Dr Prakit Phothisan, Research Coordinator, Ayutthaya Provincial Health Care Department and Dr Kitiya Pasanvong for providing the data.

Reference

  • Ayutthaya Provincial Annual Report 1999. Provincial Health Care Department, p. 122.
  • Adsavakulchai S, Honda K, Nualchawee K,Murai S, Noomhorm A and Lertlum S 2001; A Study on Climate Impact Assessment on Human Health using Remotely Sensed Data, Asian Journal of Geoinformatics (a Quarterly Publication of ARSRIN); Volume 1, No. 3 March 2001. pp. 75-90
  • Gould Peter, 1993, The Slow Plague, A geography of the AIDS Pandemic, Why a geographer writes about AIDS. pp 2-10
  • Page, G. William and Patton, Carl V., 1991. Quick Answers to Quantitative Problems. Chapter 14 LQ, pp. 173-177.
  • Rural-regional Development Planning Students, 2000; Annual Rural-Regional Planning Workshop Report (September-December 2000). SERD, AIT, Thailand . pp. 168-174.
  • Vilay Prathumvan, 1999. Master thesis, Diarrhoea Pattern in Bang Pa-In district, Mahidol University, p. 150
  • Wisesjindawat Wisa, 2000; Master thesis, Diffusion of AIDS In Zone 10 of The Ministry of Public Health, Chulalonglorn University, ISBN 974-346-385-2, p 200.