Fig 1.
Flow diagram of patient inclusion and exposure category assignment for PEPV noncompletion analysis, 2019-2022.
(A) Stepwise inclusion of patients from the initial cohort of 242,648 individuals who attended the three IPC Rabies Prevention Centers (RPCs) between 2019 and 2022. Exclusions were applied based on species not known to transmit rabies (humans, rodents, lagomorphs, etc.), missing district data, confirmed negative virology results in the exposing animal, Category I or undefined exposures, late PEP initiation with the animal still alive at day 10 post-exposure, and missing data for key variables, resulting in a final analytic cohort of 239,874 patients. (B) Classification of patients (n = 241,639) based on WHO exposure categories using information on the animal species, mode of exposure (bite, scratch, lick), skin surface (intact or broken/mucosal), wound depth (superficial or deep), and presence of bleeding. Exposure categories were assigned as Category II, III, or mixed (II or III); a small number of cases were classified as “incoherent” due to inconsistent data or as “category cannot be determined” to indicate cases where exposure type was missing and could not be classified into a WHO exposure categories. Color-coded boxes represent categorical decisions, while icon labels indicate the exposure classification pathway based on animal species, exposure tyspe, and clinical characteristics..
Table 1.
Annual number of patients seeking post-exposure prophylaxis vaccination (PEPV) and number of patients with incomplete PEPV, by year and by Rabies Prevention Center (RPC).
Table 2.
Socio-demographic characteristics of PEPV patients and noncompletion rates.
Fig 2.
Temporal trends in patient attendance and PEPV noncompletion across the three IPC Rabies Prevention Centers (RPCs), 2019-2022.
Monthly trends show the total number of patients with WHO-defined Category II or III exposures requiring PEPV (black line, left y-axis) and the corresponding percentage of patients who did not complete the full PEPV regimen (pink line, right y-axis) at each RPC: Battambang, Kampong Cham, and Phnom Penh. Key contextual events are annotated, including the opening of the Kampong Cham RPC, a high-profile social media event (Feb–Apr 2019), Covid-19 detection and containment phases, nationwide lockdowns, and the full reopening of the country. Data include all patients eligible for PEPV from 2019 to 2022 (n = 239,874).
Fig 3.
Travel time from districts to RPCs.
(A) Map of travel time from districts to the RPC that is temporally the closest. (B) Number of districts by travel time. (C) Travel time to the closest RPC by RPC catchment area. (D) Number of districts temporally closest to each RPC. Travel time was computed using the terra and gdistance packages in R software, based on the Malaria Atlas Project, Global Motorized Friction Surface 2019. The Malaria Atlas Project maps are under the “Creative Commons Attribution 3.0 Unported License” (https://malariaatlas.org/open-access-policy/). The source of the basemap shapefile is https://gadm.org/download_country.html. The data are freely available for academic use such as publishing of academic research articles https://gadm.org/license.html.
Fig 4.
Patient travel time to the three RPCs from 2019 to 2022.
The figure compares the variability of travel time between patients living within the same province hosting the RPC (intra-provincial, in brown) and those living outside (inter-provincial, in pink). For each RPC, the violin shows the density of patient travel time to the visited RPC, and the boxplot shows median and interquartile range (IQR) of patient travel time.
Fig 5.
Spatial distribution of PEPV attendance, incidence, and noncompletion rates in Cambodia, 2019 to 2022.
(A) Total number of patients seeking PEPV per province. (B) Mean annual incidence of PEPV patients per 100,000 inhabitants. (C) PEPV noncompletion rates by province, limited to districts with >20 patients. Panels A and B were based on the total number of patients included in the descriptive, spatial, and incidence analyses (n = 241,845), while Panel C includes only patients with WHO-defined category II or III exposures necessitating PEPV (n = 239,874)). The maps were created using R software. The source of the basemap shapefile is https://gadm.org/download_country.html. The data are freely available for academic use such as publishing of academic research articles https://gadm.org/license.html.
Fig 6.
Spatial distribution of the median age of patients seeking PEPV, 2019-2022.
The map shows the median age of patients by district across Cambodia, based on all included visits (n = 241,845). Shading indicates age categories, and districts with fewer than 20 patients are shown with a dotted pattern. Median ages ranged from 13 to 24 years, with younger median ages observed in districts near newly established RPCs. The map was created using R software. The source of the basemap shapefile is https://gadm.org/download_country.html. The data are freely available for academic use such as publishing of academic research articles https://gadm.org/license.html.
Table 3.
Characteristics of the animals involved in exposure events and their association with PEPV noncompletion.
Table 4.
Exposure characteristics, use of RIG, and timing of PEPV initiation.
Fig 7.
Forest plot of risk factors associated with PEPV noncompletion.
Multivariate logistic regression results showing adjusted odds ratios (OR) and 95% confidence intervals (CI) for factors associated with noncompletion of PEPV. The reference category (OR=1.0) is indicated for each variable. Filled circles indicate statistically significant associations (p ≤ 0.05), while open circles represent non-significant associations (p > 0.05). Variables retained in the final model include patient demographics, exposure context, animal characteristics, health system access, and time-based contextual factors. The model was selected based on stepwise selection and lowest AIC and includes 239,874 patients with Category II and III exposures.