Abstract
Thailand is facing air pollution problems from particulate matter of diameter less than 2.5 micron or
PM2.5 exceeding standard levels every year. Particularly Nakhon Phanom province of Health Region 8 has
been significantly affected. This study aimed to evaluate the PM2.5 surveillance system in Nakhon Phanom
both qualitatively and quantitatively to develop policies and operational guidelines and to serve as a case
study for Health Region 8. This research employed a mixed-method approach by conducting in-depth
interviews with 11 public health officials and stakeholders and reviewing 508 medical records of patients
suspected of PM2.5-related diseases at Nakhon Phanom Hospital between October 1, 2020 and March 31,
2023. The qualitative analysis revealed a high level of acceptance of the surveillance system, primarily due
to growing public concern regarding the health impacts of air pollution. The system was found to be
relatively simple and flexible in implementation. The quantitative results showed that out of 485 medical
records, 442 matched the reporting definitions according to the ICD-10 (International Classification of Disease
10th revision) criteria, and 355 matched the disease definitions. The system demonstrated high coverage
(90.7%), indicating that a substantial proportion of potential cases were included in the surveillance
data. However, the positive predictive value was moderate (72.9%), suggesting that while the system is
generally accurate, there is room for improvement in differentiating between diseases directly attributable
to PM2.5 exposure and other factors. Additionally, no clear correlation was found between PM2.5 levels and
the number of reported patients. The absence of clear diagnostic criteria for diseases related to PM2.5 exposure adversely impacts data accuracy and consistency. In conclusion, the PM2.5 surveillance system in
Nakhon Phanom province was functioning satisfactorily, with high coverage and reasonable data
accuracy. However, the study highlights the need to improve the diagnostic criteria for PM2.5-related
diseases and the development of learning machine artificial intelligence for coding Z58.1. Furthermore,
refinements to the surveillance system would better support the development of evidence-based policies
aimed at mitigating the health impacts of air pollution.
