Study design
We prospectively collected cases of pneumonia due to A. baumannii and P. aeruginosa, and matched them with two control patients, one with pneumonia due to non-multidrug resistant A. baumannii or P. aeruginosa, and one with no infection. We used the multistate model utilised by Stewardson et al19. Enrolled patients were categorized into three states: MDRO infection (R-group), susceptible organism infection (S-group), and no-infection (N-group). We compared the clinical and economic aspects of the R-group and S-group, and the R-group and N-group, respectively, and estimated the additional burden of pneumonia due to MDRO, compared to that of susceptible organisms, and no infection. Our study conforms to the Consolidated Health Economic Evaluation Reporting Standards (CHEERS), and all methods were carried out in accordance with relevant guidelines and regulations.
Setting
We collected data from ten secondary and tertiary hospitals in South Korea, selected by considering the regional distribution of hospitals. The study was performed from September 2017 to February 2018. We used a currency exchange rate of 1110 Korean won /1 US dollar for the calculations.
Participants
We prospectively identified and collected the data of all patients with pneumonia due to A. baumannii, and P. aeruginosa, regardless of antibiotic susceptibilities. After collection, the R-group and S-group were selected by pre-defined criteria. In brief, MDR was defined as Acinetobacter or Pseudomonas isolates resistant to at least three classes of antimicrobial agents, including carbapenem, aminoglycosides, and fluoroquinolones. Pneumonia was defined as a case in which bacteria grew in respiratory specimens such as sputum, transtracheal aspiration fluid, and bronchoscopy washings, and susceptible antibiotics were administered against the organisms. The criteria for selected matched susceptible or no-infection control cases were: same principal diagnosis at the time of admission, same major surgery or intervention during the hospitalization, age (years ± 10), sex, and admission date within 60 days.
If the control cases of susceptible infection or no infection experienced invasive bacterial infection during the same hospital admission, then the control case was excluded, and another was selected. Cases with a total length of stay (LOS) of 180 days or longer were excluded from matching.
Variables
Data collection variables were baseline characteristics, route of admission, LOS before and after infection, and underlying disease. We also collected data on severity of infection through the Sequential Organ Failure Assessment (SOFA) score, and 90-day mortality. Hospital costs of patients in each group were collected.
Statistical analysis
Estimation of additional hospital and caregiver costs of MDRO pneumonia
We estimated the total additional direct medical costs of MDRO pneumonia (R-group) by subtracting the mean hospital cost of the S-group or N-group from that of the corresponding R-group.
The cost of caregiving was calculated by multiplying the daily cost of the hired caregiver and the excess length of stay (LOS) (caregiver fee was $59.1 per day [65,000 Korean won] as taken from the caregiver association).
Estimation of numbers and mortality of MDRO pneumonia cases nationwide
The estimation methods are described in the online only supplementary methods. In brief, we calculated the ratio of cases of MDRO bacteraemia between the 10 study hospitals and the national survey which we undertook previously. We then assumed that the ratio of pneumonia in the 10 study hospitals to the nationwide results was the same as that of MDRO bacteraemia (unpublished data). The number of occurrences of pneumonia and the number of deaths nationwide were estimated, based on this ratio.
We estimated the mortality due to MDRO pneumonia according to age distribution. We estimated the mortality rate for each age group in the R-group patients, and then calculated the ratio of patients by age group among the total deaths. The estimated number of deaths of each age cohort of MDRO pneumonia on a nationwide scale was calculated by multiplying the estimated number of MDRO pneumonia cases and the 90-day mortality rate from our data.
Estimation of productivity loss due to death
The productivity loss due to unexpected death was calculated from the number of deaths associated with MDRO pneumonia and the annual mean wages reported by the Ministry of Labor in Korea (Labor Statistics of Korea, Ministry of Employment and Labor 2017; available from http://wage.go.kr/index.jsp). The productivity loss due to the unexpected death of a given patient was the sum of the annual wages up to the time that patient would have reached 65 years of age if he or she had not died. The annual discount rate was 5%.
Estimation of nationwide socioeconomic burden of MDRO pneumonia
The socioeconomic burden of MDRO pneumonia was estimated by the sum of the additional hospital cost, caregiver cost, and productivity loss due to unexpected death. The additional hospital and caregiver costs were calculated by multiplying the additional cost due to MDRO pneumonia and the estimated annual number of MDRO pneumonia patients.
Ethical review
This study was approved by the Institutional Review Board (IRB) of Seoul National University Bundang Hospital (IRB no. B-1804-463-105) and the IRBs of each participating hospital. Informed consent was waived by the following IRBs; Seoul National University Bundang Hospital, Ewha Womans University Mokdong Hospital, Hallym University Sacred Heart Hospital, Seoul Metropolitan Government-Seoul National University Boramae Medical Center, Inje University Ilsan Paik Hospital, Chungnam National University Hospital, Chonnam National University Hospital, Kangwon National University Hospital, Seoul National University Hospital, and Pusan National University Hospital.