We demonstrated a substantial burden of RSV-associated pneumonia among children in two rural Provinces in Thailand during the study period; 1 in 187 children aged 1–11 months and 1 in 450 children aged 1–4 years were hospitalized annually for radiographically-confirmed RSV pneumonia. RSV accounted for 25% of hospitalized radiographically-confirmed pneumonia among children <1 year of age and 21% among children 1–4 years of age. In addition, RSV caused an almost equal number of additional hospitalizations among children for lower respiratory tract illnesses that were not radiographically-proven pneumonia. Age <1 year and 1–4 years were strong predictors of RSV-associated illness. In contrast, <2% of enrolled adults ≥65 years of age had RSV infection.
There are few population-based studies that report rates of RSV-associated radiographically-confirmed pneumonia. Our adjusted rate of RSV radiographically-confirmed pneumonia for all adults (3.6/100,000) was higher than that reported in the United States in the 1990s (1.6/100,000); however, we used a more sensitive assay, RT-PCR
. Rates of RSV-associated radiographically-confirmed pneumonia among children have not been reported from the U.S. or other industrial countries, or from developing countries. However, our rates of hospitalized lower respiratory illness (enrolled patients) were similar to rates of hospitalizations for the clinical syndrome of pneumonia and severe pneumonia in Kenyan children (children <1 year: 1107/100,00 and children <5 year: 293/100,000)
. Additional studies to document rates of RSV disease, including radiographically-confirmed pneumonia, in developing and industrial countries would put RSV respiratory disease and pneumonia into perspective with other commonly recognized causes of pneumonia. Given the recognized global burden of pneumonia, especially among children, the information from this study should inform planning for pneumonia intervention strategies.
To facilitate comparison of our rates with other reports that capture all hospitalized RSV infections (pneumonia, bronchiolitis, etc), we estimated the number of RSV acute lower respiratory tract infections that we missed by limiting our enrollment to patients who had a chest radiograph within 48 hours of admission and added these patients to our incidence estimates. The proportion of RSV infections among hospitalized children aged <5 years who met the case definition for acute lower respiratory tract illness but did not have a chest radiograph was 11% (HC Baggett, personal communication). Combining patients with and without a chest radiograph our estimated rate of hospitalized RSV acute lower respiratory tract infection among children <5 years would be 883/100,000. This crude estimation is consistent with rates for RSV-associated hospitalizations and severe disease among children aged <5 years reported in a recently published system review 
Prevention and treatment of childhood pneumonia are important components of the vision of the Global Action Plan for the Prevention and Control of Pneumonia (GAPP) 
. However, at this time there is no effective and safe vaccine that prevents RSV infection in children and there are no proven treatment modalities. Available prophylactic measures against RSV, including palivizumab, are very expensive and only recommended for children at the highest risk for severe RSV disease
. New interventions to prevent and treat RSV illness could reduce hospitalizations for radiographically-confirmed pneumonia by up to 22% among Thai children <5 years of age and reduce at least a similar proportion of hospitalizations for other RSV lower respiratory tract illness.
We demonstrated annual peaks in RSV circulation occurring between July and October in Thailand. Patients admitted to a hospital with lower respiratory tract illness during these months were more likely to have RSV infection compared to patients with a negative RSV test. These peak months are similar to reports from other countries with tropical and subtropical climates
and other reports from Thailand
and correspond to the rainy season.
Unlike previous reports
, we found no differences in severity of illness among patients with RSV group A and B infections. However, our analysis was limited by small numbers. We were unable to evaluate correlations between RSV load in nasopharyngeal specimens and severity, as reported by others 
. We found more RSV group B infections among elderly adults than group A. Clinical studies from one candidate vaccine among the elderly, PFP-2 (purified fusion protein), found lower neutralizing titers to group B viruses compared to group A 
Our results are limited by some potential biases. We likely underestimated the incidence of hospitalized RSV infections. We did not enroll hospitalized patients who met the clinical case definition for acute lower respiratory illness but did not have a chest radiograph ordered within 48 hours of admission; we may have missed cases of bronchiolitis. Also, serology was performed for a limited time in one province but added a substantial number of cases during that time period. The lack of serology testing for the remainder of the study period likely reduced the number of RSV infections that we detected. This is especially true for older children and adults where approximately 50% of infections were only detected by serology when both RT-PCR and serology were performed. We did not adjust for the lack of serology in later years due to the small number of older children and adults who had both serology and RT-PCR performed. Very young children and very ill patients were less likely to enroll in the etiology study (H. Baggett, personal communication). Thus, we may have underestimated infections among children <6 months of age, and we likely underestimated the number of deaths, especially pediatric deaths.
The burden of hospitalized RSV-associated pneumonia and other RSV-associated lower respiratory tract illness among children <5 years was substantial in rural Thailand. Efforts to inexpensively and effectively prevent and treat RSV infection in children <5 years of age, including development of effective vaccines and treatment modalities could substantially reduce the global number of pediatric pneumonias and respiratory hospitalizations