Economic burden of inpatients infected with Klebsiella pneumoniae carbapenemase

ABSTRACT Objective: To estimate the direct medical costs of drug therapy of Klebsiella pneumoniae carbapenemase (KPC) infection patients in hospital-based context. Methods: A cost-of-illness study conducted with a prospective cohort design with hospitalized adults infected by KPC. Data collection was performed using an instrument composed of sociodemographic data, clinical and prescription medication. Estimates of the direct costs associated to each treatment were derived from the payer's perspective, in the case of federal public hospitals from Brazil, and included only drug costs. These costs were based on the average price available at the Brazilian Price Database Health. No discount rate was used for the cost of drugs. The costs are calculate in American Dollar (US$). Results: A total of 120 inpatients participated of this study. The total drug cost of these inpatients was US$ 367,680.85. The systemic antimicrobial group was responsible for 59.5% of total costs. The direct drug cost per patients infected by KPC was conservatively estimated at nearly US$ 4,100.00, and about of 60% of costs occurred during the period of infection. Conclusion: The findings of our study indicate a thoughtful economic hazard posed by KPC that all healthcare sectors have to face. The increasing worldwide incidence of these bacteria represents a growing burden that most health systems are unable to deal with. There is an imperative need to develop protocols and new antimicrobials to treatment of KPC, aiming to rearrange resources to increase the effectiveness of healthcare services.


❚ INTRODUCTION
Klebsiella pneumoniae carbepemase (KPC) is a multidrug resistant bacteria, with a costly therapy and high mortality rate. (1,2) The World Health Organization (WHO) published that carbapenem-resistant Enterobacteriaceae has the highest level of priority that new antibiotics are urgently needed. (3) The incidence of KPC increased quickly on the last years, from 1% (2001) to 30% (2008) of all hospital infections. Cases have been reported in other regions of the world, including Europe, (4,5) Asia, (6,7) Australia, (8) and South America. (9)(10)(11) Patients with long periods of hospitalization, mechanical ventilation, undergoing organ or stem cell transplantation, and treatment with antimicrobial agents are more likely to develop KPC infection. (12) This bacterium is involved in extra-intestinal infections, urinary tract infection, pneumoniae, bloodstream infections, surgical wound infections, endocarditis and sepsis, and the mortality can be higher than 40% in 30 days. (4,6) The annual economic burden of multidrug resistant bacteria's amounted to more than US$ 45 billion, only considering direct and indirect costs. (13) Annually, just the pharmaceutical purchases can represent 70% of out-of-pocket health costs in India, 43% in Pakistan, and 20% in Brazil. (13) The economic cost of therapy and the high mortality rates of KPC make this infection a relevant health problem. (14)(15)(16) This  (17) The study participants were adult inpatients with KPC infection during hospitalization, confirmed by laboratory testing. Data collection was performed using an instrument including sociodemographic data (sex, age, colour or race, marital status, and educational level), clinical data (site of infection, length of hospital stay, periods of infection, reinfection and reasons for hospital discharge) and prescription medication (drug per day, dose and route of administration). These data were obtained by accessing the participant's records after antibiogram confirming KPC. The time line was considered until patient's discharge.
The drugs were initially classified by the World Health Organization Anatomical Therapeutic Chemical (ATC) classification system, which divides substances into different groups according to the organ or system they act on and in their chemical, pharmacological and therapeutic properties. We use the fifth level of this system to identify the drugs (subgroup for chemical substance). For other analyses they were grouped according to the first level (main anatomical group), second level (main therapeutic group), third level (therapeutic/pharmacological subgroup) or fourth level (chemical/therapeutic/pharmacological subgroup). The infection period was verified through the antibiogram to classify the treatment period in prior to infection, during infection, and after infection.
A standardized approach by the SUS was used to address treatment costs. Estimates of the associated direct costs of each treatment were derived from the payer's perspective, in the case of Public Federal Hospitals, and included only drug costs.
For the costs of drugs, all medications were included, either for treatment of infection or of comorbidities. Drug costs were based on the average price available in at Departamento de Informática of the SUS (DATASUS; http://bps.saude.gov.br/login.jsf), and for the study, we used the purchase prices of federal public agencies during the study period. No discount rate was used for the cost of drugs. The costs were calculated in United States Dollars (US$), using a reference of US$ 1=R$ 3,20 (Brazilian reals).
The daily cost of each drug was achieved by dividing the prescribed daily dose by the drug dosage, and then multiplied by the cost of the drug. Moreover, for the cost of the total treatment, the total number of days at hospital was added, as the formula: therapy daily cost = × cost of drug y daily dose of drug y presentantion of drug y The descriptive analysis was performed to present the prescribed drugs according to ATC. To verify if there is difference in the total cost according to the period of treatment and site of infection, the Kruskal-Wallis test was used. A significance level of 5% was adopted, and the analyses was made using the Statistical Package for the Social Sciences (SPSS) 21.0 software.
The mean time of initial prescription of antimicrobial was between 6 to 9 days, with a range between 2 to 25 days. Pneumonia presented a mean of 9.4 days (min 2 days; max 24 days) of initial prescription of antimicrobials, followed by intra-abdominal infection ( Three groups of drugs (systemic antimicrobials, blood and hematopoietic organs, and digestive tract and metabolism) accounted for 85% of total cost of patients, but just accounted for 21.4% of total sum of items (Table 2).
Four antimicrobials (amikacin, meropenem, polymyxin B and vancomycin) accounted for 48.6% of prescriptions of this group, whereas other antimicrobials were prescribed for less than 5 days (erythromycin, benzylpenicillin).   As estimated, costs improved at least 72% during the infection, as compared to other periods. These results are more significant than the impact of the prescription. Furthermore, incorrect doses or duration of treatment with antimicrobials can increase costs up to 36%. (14,15) A significant percentage of total costs (59.5%) were due to systemic antimicrobials during hospitalization, but this figure accounts for only 7.1% of all drugs administrated. The cost of systemic antimicrobials during the infection represents 41.2% of all costs. This aspect shows the importance of antimicrobial stewardship programs considering cost-effectiveness analysis. Such programs are capable to reduce doses by 26% and expenses by 81%, representing a huge economic impact in health sector. (18) In addition, this intervention in prescription of antimicrobial therapies can reduce the spectrum of multidrug resistant bacteria.
When comparing the sites of infection, we observed significant difference only during the period of infection. The bloodstream infection presented the highest costs, followed by osteoarticular, pneumonia, intra-abdominal, skin and soft tissues, and urinary tract infection. Patients with bloodstream infection received highly doses of costly antimicrobials, such as meropenem, tigecycline and ciprofloxacin, when compared to the other sites of infection, although they did not present the highest mean treatment time. The literature shows that the treatment of bloodstream infection is more expensive, and to the use of antimicrobials to treatment of multidrug resistant bacteria can increase costs by up to 1.6 fold. (19)(20)(21) Additionally, they use more complementary drugs to fight the clinical effects of KPC infection. (22,23) The mean time of prescription of initial antimicrobial therapy varied between 2 and 14 days. This difference in prescription days can be related to the use of second or third antimicrobial options for treating KPC infections. In most clinical cases, prolonged therapy can be beneficial; however prolonged duration of antibiotic einstein (São Paulo). 2019;17(4):1-8 therapy is associated with increased resistance, drugrelated effects, high costs and adverse drug reactions. (24) The use of 7 or 8 days of antibiotic therapy did not increase the risk of adverse clinical outcomes, and may reduce the emergence of resistant organisms, as compared to a prolonged course of more than 10 days. (25,26) To our knowledge, this is the first study conducted in Latin America to measure the direct costs of KPC treatment, and it is also the first study worldwide to compare the cost of different periods of infection in this population. Most previous studies included only the cost of antimicrobials during treatment (27)(28)(29) or the general cost of hospitalization plus drug therapy. (30) Finally, our micro cost analysis allowed us to control possibly confounding influences on outcomes of cost studies, and better understand the real cost of each drug class.
While the current study enhances our understanding of the economic burden of KPC, some limitations should be considered. First, this study was conducted only in one hospital and may not be representative of all Brazil patients with KPC infection. Second, the DATASUS values may not be 100% equivalent to costs of drugs at other hospitals. Third, other cost categories, such as serum monitoring of patients, cost of materials, inputs required for preparation and administration of antibiotics, treatment of adverse reactions, and additional hospitalization time due to the presence of infectious diseases were not included in the analysis. However, there are no available cost studies with primary data collected from the Brazilian Unified Health System that could be used for comparison. Since our study did not collect data from patients without KPC, we cannot ensure that the costs were attributed exclusively to this infection.
The findings of our study indicate a thoughtful economic hazard posed by KPC that all healthcare sectors have to face. There are indirect and intangible costs that were not shown. The increasing incidence of KPC worldwide represents a growing burden that most health systems are unable to deal with. There is an imperative need to develop protocols and new antimicrobials to treat KPC, aiming to rearrange resources to increase the effectiveness of healthcare. Otherwise, the cost of treating multidrug resistant bacteria will not be feasible in a near future, with severe consequences to the population.
This study brought advances in knowledge of costs of treatment of patients with KPC infection. We verified increased costs during the period of infection in more than one drug class, and the difference in treatment cost according to the site of infection. Moreover, the direct costs were checked according to ATC classification.

❚ CONCLUSION
The direct drug cost per patient infected by Klebsiella pneumoniae carbepemase is conservatively estimated at nearly US$ 4,100.00, and about 60% of costs are during the period of infection. The results of this study have implications for the public health system, especially because the payment are made by the Brazilian Unified Health System. This system makes the payment according to reasons for hospitalization, however the amount reimbursed does not cover the actual value of the treatment. In addition, the treatment of some infections is not covered by this method when acquired at hospital. These data could also be used to develop appropriate cost-effectiveness models, which are needed to improve quality of treatment with reduced costs.