The healthcare costs of antimicrobial resistance in Lebanon: a multi-centre prospective cohort study from the payer perspective

BMC Infectious Diseases

Christine Roques¹  Yasmina Yared²  Natalia Dirani³  Katia Iskandar⁴  Laurent Molinier⁵  Rola Husni-Samaha⁶  Matta Matta⁷  Roula Matta⁸  Inas Mostafa⁹  Rana Rizk^1,10  Pascale Salameh^1,11  Souheil Hallit^1,12  Rachel Abdo^1,13 

[1] Department of Bioprocédés et Systèmes Microbiens, Laboratoire de Génie Chimique, Université Paul Sabatier Toulouse III, UMR 5503, Toulouse, France;Department of Bactériologie-Hygiène, Centre Hospitalier Universitaire, Toulouse, Hôpital Purpan, Toulouse, France;Department of Clinical Pharmacy, Geitaoui Hospital, Beirut, Lebanon;Department of Infectious Diseases, Dar El Amal University Hospital, Baalbeck, Lebanon;Department of Mathématiques Informatique et Télécommunications, Université Toulouse III, Paul Sabatier, INSERM, UMR 1295, F-31000, Toulouse, France;INSPECT-LB: Institut National de Santé Publique, d’Épidémiologie Clinique et de Toxicologie-Liban, Beirut, Lebanon;Department of Pharmacy, Lebanese University, Mount Lebanon, Beirut, Lebanon;Department of Medical Information, Centre Hospitalier Universitaire, INSERM, UMR 1027, Université Paul Sabatier Toulouse III, F-31000, Toulouse, France;Department of Medicine, Lebanese American University, Byblos, Lebanon;Department of Infection Control, Lebanese American University Medical Center, Beirut, Lebanon;Department of Medicine, St Joseph University, Beirut, Lebanon;Department of Pharmacy, Lebanese University, Mount Lebanon, Beirut, Lebanon;Department of Quality and Safety, Nabatieh Governmental Hospital, Nabatieh, Lebanon;INSPECT-LB: Institut National de Santé Publique, d’Épidémiologie Clinique et de Toxicologie-Liban, Beirut, Lebanon;Department of Health Services Research, School CAPHRI, Care and Public Health Research Institute, Maastricht University, 6200, Maastricht, MD, The Netherlands;INSPECT-LB: Institut National de Santé Publique, d’Épidémiologie Clinique et de Toxicologie-Liban, Beirut, Lebanon;Department of Pharmacy, Lebanese University, Mount Lebanon, Beirut, Lebanon;Medical School, University of Nicosia, Nicosia, Cyprus;INSPECT-LB: Institut National de Santé Publique, d’Épidémiologie Clinique et de Toxicologie-Liban, Beirut, Lebanon;Faculty of Medicine and Medical Sciences, Holy Spirit University of Kaslik (USEK), Jounieh, Lebanon;INSPECT-LB: Institut National de Santé Publique, d’Épidémiologie Clinique et de Toxicologie-Liban, Beirut, Lebanon;Medical School, University of Nicosia, Nicosia, Cyprus;

关键词: Antimicrobial resistance;    Healthcare cost;    Length of stay;    In-hospital mortality;   

DOI  :  10.1186/s12879-021-06084-w

来源: Springer

【 摘 要 】

BackgroundOur aim was to examine whether the length of stay, hospital charges and in-hospital mortality attributable to healthcare- and community-associated infections due to antimicrobial-resistant bacteria were higher compared with those due to susceptible bacteria in the Lebanese healthcare settings using different methodology of analysis from the payer perspective .MethodsWe performed a multi-centre prospective cohort study in ten hospitals across Lebanon. The sample size consisted of 1289 patients with documented healthcare-associated infection (HAI) or community-associated infection (CAI). We conducted three separate analysis to adjust for confounders and time-dependent bias: (1) Post-HAIs in which we included the excess LOS and hospital charges incurred after infection and (2) Matched cohort, in which we matched the patients based on propensity score estimates (3) The conventional method, in which we considered the entire hospital stay and allocated charges attributable to CAI. The linear regression models accounted for multiple confounders.ResultsHAIs and CAIs with resistant versus susceptible bacteria were associated with a significant excess length of hospital stay (2.69 days [95% CI,1.5–3.9]; p < 0.001) and (2.2 days [95% CI,1.2–3.3]; p < 0.001) and resulted in additional hospital charges ($1807 [95% CI, 1046–2569]; p < 0.001) and ($889 [95% CI, 378–1400]; p = 0.001) respectively. Compared with the post-HAIs analysis, the matched cohort method showed a reduction by 26 and 13% in hospital charges and LOS estimates respectively. Infections with resistant bacteria did not decrease the time to in-hospital mortality, for both healthcare- or community-associated infections. Resistant cases in the post-HAIs analysis showed a significantly higher risk of in-hospital mortality (odds ratio, 0.517 [95% CI, 0.327–0.820]; p = 0.05).ConclusionThis is the first nationwide study that quantifies the healthcare costs of antimicrobial resistance in Lebanon. For cases with HAIs, matched cohort analysis showed more conservative estimates compared with post-HAIs method. The differences in estimates highlight the need for a unified methodology to estimate the burden of antimicrobial resistance in order to accurately advise health policy makers and prioritize resources expenditure.

【 授权许可】

CC BY