- Email: [email protected]
Costs associated with hospital-acquired bacteraemia in an Indian hospital: a case–control study
Journal of Hospital Infection (2009) 71, 143e148
Available online at www.sciencedirect.com
www.elsevierhealth.com/journals/jhin
Costs associated with hospital-acquired bacteraemia in an Indian hospital: a caseecontrol study* A. Kothari a,*, V. Sagar b, V. Ahluwalia c, B.S. Pillai c, M. Madan d a
Dr BL Kapur Memorial Hospital, New Delhi, Delhi, India Regions Hospital, St Paul, Minnesota, USA c Max Heart & Vascular Institute, New Delhi, Delhi, India d Ranbaxy Laboratories Ltd, Gurgaon, Haryana, India b
Received 30 July 2008; accepted 22 October 2008 Available online 5 December 2008
KEYWORDS Bacteraemia; Costs; Hospitalacquired infection; Mortality; Developing countries
Summary Studies from around the world have shown that hospital-acquired infections increase the costs of medical care, morbidity and mortality. The aim of this study was to determine cost and attributable mortality associated with hospital-acquired bacteraemia in a tertiary care centre in India. This was a retrospective caseecontrol, cost utility analysis set in the cardiothoracic unit of a 200-bedded tertiary care cardiac hospital. Cases included adult patients who underwent coronary artery bypass graft and/or valve replacement surgery who developed bacteraemia (indicated by positive blood cultures) during postoperative stay (N ¼ 24). Controls were age- and sex-matched adult patients who underwent similar procedures but who did not develop bacteraemia (N ¼ 48). Data were collected from patient medical records and other administrative databases for cost analysis. Prolongation of hospital stay, attributable mortality and extra costs associated with hospital-acquired bacteraemia were analysed. Statistical analysis was done using Fisher’s exact test and unpaired t-test. Patients with hospital-acquired bacteraemia experienced a significantly longer total hospital stay [mean: 22.9 days; 95% confidence interval (CI): 17.2e28.6; P < 0.0001], significantly longer ICU stay (mean: 11.3 days; 95% CI: 9.0e13.6; P < 0.0001), a significantly higher mortality (mean: 54%; P < 0.0001) and cost significantly more (mean: US $14,818; 95%
* Part of this study has been previously presented in the 2007 Annual Meeting of the Infectious Diseases Society of America, Abstract No. 814. * Corresponding author. Address: A. Kothari, A-71, Kirti Nagar, New Delhi, 110015, India. Tel.: þ91 11 9818009668; fax: þ91 11 25447063. E-mail address: [email protected]
0195-6701/$ - see front matter ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2008.10.022
144
A. Kothari et al. CI: 10,663e18,974; P < 0.0001) than controls. We conclude that hospitalacquired bacteraemia significantly increases mortality and costs of hospitalisation in lower income developing countries. Our study demonstrates that costs associated with HAIs are similar between developing and developed countries. Better infection control planning and infrastructure may offset some of these costs. ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction Studies from around the world have shown that hospital-acquired infections (HAIs) increase the cost of medical care due to prolongation of hospital stay, and increased morbidity and mortality. Studies of economic cost of HAIs have been undertaken in the USA, England, Germany, Denmark, Spain, France, Belgium, and Taiwan, among other countries.1e13 Previous studies in the USA have estimated that a confirmed HAI resulted in an extra cost of US $15,275 per case.14 Another recent study from the UK stated that patients with nosocomial infections incurred hospital costs 2.9 times higher than uninfected patients, equivalent to an additional £3,154.6 Estimates of the cost of nosocomial infection incllude £930 million per annum in the UK and from US $296 million to 2.3 billion in the USA.6,15 HAIs are a serious problem in developing countries like India.16 However, there is no published literature on analysis of costs associated with HAI in India. The aim of this study was to determine the cost and the attributable mortality associated with hospital-acquired bacteraemia in a tertiary care centre in India.
Methods This was a retrospective caseecontrol study done in 2006. The setting was the cardiothoracic unit of a 200-bedded private, tertiary care cardiac hospital in New Delhi, India. Patients requiring emergency cardiac surgery were excluded from the study. This is because elective cardiothoracic surgical patients without complications have a stable duration of stay as well as total cost of hospitalisation. This allows for a comparison between case and control groups. This is in contrast to general medical and surgical patients, where both duration of stay as well as cost of hospitalisation is variable according to the diagnosis. The
unit did 842 adult cardiac surgical cases in 2006. Preoperative antibiotic prophylaxis used was cefazolin 1 g preoperatively 0.5e1 h before procedure and was continued postoperatively for 48 h.
Inclusion criteria Cases included all adult patients undergoing coronary artery bypass grafting (CABG) and/or valve replacement surgery (mitral and/or aortic valve) who developed clinically significant bacteraemia during postoperative stay (N ¼ 24). Age- and sex-matched adult patients undergoing similar procedures who did not develop detectable bacteraemia during stay were chosen as the control group (N ¼ 48) in a 2:1 ratio.
Preoperative risk evaluation All patients undergoing cardiac surgery were evaluated preoperatively using standard European System for Cardiac Operative Risk Evaluation (EuroSCORE).17e19 The EuroSCORE calculation was done using the EuroSCORE website (www.euroscore.org).
Definitions Hospital-acquired bacteraemia was defined as a bloodstream infection that developed 48 h after hospital admission. One set of positive blood cultures in a patient with clinical signs, who was prescribed specific antibiotics for treatment, was considered as hospital-acquired bacteraemia. If the bacteraemia was caused by a commensal micro-organism, at least two positive sets of blood cultures within a 72 h period with the same organism was required.
Data collection Data were collected using patient medical records, hospital information system and other administrative databases for cost analysis. Measurable
Cost hospital-acquired bacteraemia India
145
Table I Causative organisms for episodes of bacteraemia in cases
Patient demographics
Micro-organisms
N
%
Candida spp. Acinetobacter baumannii Pseudomonas aeruginosa Klebsiella pneumoniae Enterococcus faecalis Staphylococcus aureus Stenotrophomonas maltophila Enterobacter spp.
10 5
35 17
There were 24 patients who fulfilled the inclusion criteria of cases during the time period of the study. The mean age of the cases was 64.8 years (SD: 9.9). The mean age of the 48 controls was 60.9 years (SD: 9).
4
14
EuroSCORE
3 2 2 2
10 7 7 7
1
3
The mean EuroSCORE of the cases, as calculated preoperatively, was 2.93 (SD: 2.08); and that of controls was 3.07 (SD: 1.61). There was no statistically significant difference between the EuroSCOREs of the two groups (P ¼ 0.7521; 95% CI: e1.03 to 0.74).
Causative organisms elements included prolongation of hospital stay, increase in mortality and extra costs associated with hospital-acquired bacteraemia. All costs were calculated in Indian rupees (Rs) and were then converted to US $ using the existing exchange rate in 2006 (Rs45.30 per US $) before doing statistical analysis.
The 24 cases had 29 episodes of hospital-acquired bacteraemia during their stay in the hospital (Table I). Five patients had two episodes of bacteraemia during hospitalisation. Sources of infection included central venous catheters (62%), VAPs (17%), and CAUTIs (7%) and were unknown in 14% of cases.
Mortality rate
Statistical analysis Statistical analysis was done using GraphPad Software. Continuous variables were defined as mean and SD. Comparative tests on variables were done using Fisher’s exact test and unpaired t-test.
Of the 24 cases with hospital-acquired bacteraemia, 13 died (54%). By comparison, none of the patients in the control group had any mortality (0%). This difference was highly significant (P < 0.0001).
Length of stay (LOS)
Results Nosocomial infection rate The overall central line-related bloodstream infection (CRBSI) rate in the cardiothoracic unit was 25 per 1000 central line days in 2006. Similarly, the rate of ventilator-associated pneumonia (VAP) was 0.7 per 1000 ventilator days and that of catheter-associated urinary tract infection (CAUTI) was 11.6 per 1000 catheter days. The surgical site infection rate in adult cardiac surgical patients was 0.6%. Table II
The mean LOS of the cases was 33.2 days (SD: 19.8) (Table II). For the control group, the mean LOS was 10.3 days (SD: 1.9). The difference in the mean LOS in the two groups was 22.9 days, which was highly significant (P < 0.0001; 95% CI: 17.2e28.6 days).
Cost analysis In the control group, i.e. adult patients who underwent CABG and/or valve replacement and did not develop bacteraemia, the total mean cost during hospital stay was US $8,055 (SD: 1,638) (Table III). Among cases, i.e. patients who
Comparison of length of stay (LOS) in cases and controls Mean (SD) LOS (days) Controls (N ¼ 48)
Intensive care unit High dependency unit Ward Total CI, confidence interval.
3.2 0.8 6.2 10.3
(1.7) (1.1) (1.6) (1.9)
P-value
95% CI
Cases (N ¼ 24) 14.5 12.9 5.7 33.2
(7.6) (13.9) (7.6) (19.8)
<0.0001 <0.0001 0.6093 <0.0001
9.0e3.6 8.0e16.1 2.8 to 1.6 17.2e28.6
146 Table III
A. Kothari et al. Cost analysis in controls and cases (US $) Controls Mean
Bed charges 855 Investigations 699 Pharmacy 1,821 Consults 18 OR charges 4,476 Blood bank 164 Bedside procedures 22 Dialysis e
Cases
%
Mean
%
10.6 8.7 22.6 0.2 55.6 2 0.3 e
3,274 3,160 9,875 407 4,473 940 410 334
14.3 13.8 43.2 1.8 19.5 4.1 1.8 1.5
OR, operating room.
underwent CABG and/or valve replacement and developed bacteraemia, the total mean cost during hospital stay was US $22,873 (SD: 14,350) (Table III). The difference in the mean total costs between the two groups was US $14,818, which was highly significant (P < 0.0001; 95% CI: US $10,663e18,974). A subset analysis was done of patients with hospital-acquired bacteraemia who survived (N ¼ 11) and age- and sex-matched controls without bacteraemia (N ¼ 22) undergoing similar elective procedures in a 2:1 ratio. The mean age of the survivors was 65.3 years (SD: 10.6) and that of matched controls was 62.3 years (SD: 6.9). The mean EuroSCORE of the survivors, as calculated preoperatively, was 3.12 (SD: 2.81); and that of controls was 2.75 (SD: 1.52). There was no statistically significant difference between the EuroSCOREs of the two groups (P ¼ 0.6219; 95% CI: e1.15 to 1.90). The results of this analysis, which are presented in Table IV, also showed that difference in total stay between the two groups was 21.9 days, ICU stay 10.7 days and HDU stay 8.2 days, all of which were highly significant. The difference in ward stay was 3 days, but this was not significant. The cost difference between these groups was US $9,946, which was highly significant.
Table IV
Discussion In our study, which is the first attempt to determine costs associated with nosocomial infections from the Indian subcontinent, extra cost of each case of hospital-acquired bacteraemia was US $14,818. Even when a comparison was done between survivor cases and controls, the excess cost incurred by each case of BSI was US $9,946. Other studies done in North America and Europe have also estimated an extra cost per BSI to range from US $11,971 to 56,167 (Table V). Although the increase in costs shown in our study was due to several factors, pharmacy contributed the highest share (Table VI). This included not only antimicrobials and other drugs, but also medical consumables such as intravenous lines, catheters, etc. Other major factors for the increased costs were investigation charges and bed charges (due to increased LOS). In our study, patients with infections had an increased total LOS of 22.9 days, and an extra ICU stay of 11.3 days. This is in line with previous studies that have shown an extra ICU stay of 2e20 days and increased total LOS of 7.5e24 days (Table V). Attributable mortality due to hospital-acquired bacteraemia in our study was 54%, in contrast to Western literature with values between 4.4% to 41% (Table V). Only two studies have shown no significant differences in mortality between patients with nosocomial BSIs and controls.1,20 This could be due to the high preponderance of coagulasenegative staphylococci (CoNS) (63% and 46%) isolated from cases in these studies. CoNS are lower virulence pathogens compared with Gram-negative bacteria and candida, which were predominant in our study. The absence of mortality in our control group could be due to the fact that we chose elective cardiac surgical patients as our study population. This group has a lower mortality rate compared with other medical and surgical ICU patients.
Results of subset analysis of survivors and controls Subgroup of survivors with BSI (N ¼ 11) Controls (N ¼ 22) P-value Mean (SD) Mean (SD)
Length of stay (days) ICU HDU Ward Total Cost of hospitalisation (US $)
13.9 (7.2) 9 (9.4) 9.2 (9.9) 32.1 (20.3) 18,013 (8696)
3.2 (1.8) <0.0001 0.8 (1.2) 0.0003 6.2 (1.7) 0.1786 10.2 (1.8) <0.0001 8,067 (1790) <0.0001
BSI, bloodstream infection; CI, confidence interval; ICU, intensive care unit; HDU, high dependency unit.
95% CI
7.4e13.9 4.1e12.3 1.4 to 7.3 13.1e30.6 6,063e13,827
Cost hospital-acquired bacteraemia India
147
Table V Cost, mortality and length of stay associated with nosocomial bloodstream infections (BSIs): review of literature Author
Year Country
Setting
Cases
Rose et al.2 Spengler and Greenough4 Pittet et al.3
1977 USA 1978 USA
ICU and wards ICU and wards
40 81
1994 USA
ICU
86
Digiovine et al.1
1999 USA
ICU
68
Rello et al.20 Dimick et al.21
2000 Spain 2001 USA
ICU ICU
49 86
Orsi et al.22
2002 Italy
ICU and wards
105
Pirson et al.12
2005 Belgium ICU and wards
46
32.2%
Warren et al.23
2006 USA
ICU
41
23%
ICU
144
16%
Laupland et al.24 2006 Canada
Attributable mortality 28% 14 times greater 35% 4.4% (not significant) Not significant 35% (total) 33% (ICU) 35.2e40.9%
Extra length of stay
Extra cost per BSI
19 days total 14 days total
US $4,370 US $3,600
24 days total 8 days ICU 10 days ICU
US $40,000
19.6 days total 22 days total 20 days ICU 19.1e19.9 days total 21.1 days total 7.2 days ICU 7.5 days total 2.4 days ICU 2 days ICU
V3,124 US $56,167 Total US $71,443 ICU V16,356
US $34,508
V12,853 US $11,971 US $12,321
ICU, intensive care unit.
Gram-negative organisms contributed to 51% of isolates in our study, with candida contributing 35%. While there has been an increasing trend in Gram-negative septicaemia, most Western literature shows a predominance of Gram-positive organisms as causative agents of bacteraemia.25,26 Several different studies from India in various teaching hospitals have shown Gram-negative organisms as being the predominant cause of bacteraemia, with rates ranging between 66 and 81%.27e29 There are, however, certain drawbacks in our study. We had a small sample size, and findings may be different with larger studies. Also, we did not calculate the opportunity cost to the hospital, as well as cost of loss of productivity. Depending on the setting and the socio-economic situation, the opportunity cost (to the health service) and the loss of productivity may far exceed the direct cost of hospitalisation. This would make a stronger
Table VI Comparison of pharmacy costs between cases and controls Mean cost (US $) % increase Controls Drugs (excluding antimicrobials) Antimicrobials Medical consumables Total cost
Cases
180
1,687
937
59 1,582 1,821
2,680 5,508 9,875
4,542 348 542
case for more investment in infection control as a means of decreasing healthcare costs. We estimate that hospital-acquired bacteraemia leads to an excess burden of US $980,000 in our cardiothoracic unit. This study illustrates that even though the cost of healthcare is much lower in India than the Western countries, the costs of HAI are similar in both regions.30 Considering the socio-economic situation in India, where procurement of health insurance (both government and private) is limited and per capita income is very low, nosocomial infections place an even greater burden on resources than in the West. There is a need for larger studies in the developing world on costs associated with HAI. Infrastructure in hospitals and knowledge about optimal infection control practices are deficient in the healthcare sector in India. The results of this study emphasise the need for greater resource allocation for infrastructure and training in infection control and development of local guidelines for prevention of nosocomial infections.
Acknowledgements We thank Dr T.D. Chugh, Dr D. Shuckla and A. Pruthi for their contributions during the preparation of the manuscript. Conflict of interest statement None declared.
148 Funding sources None.
References 1. Digiovine B, Chenoweth C, Watts C, Higgins M. The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit Care Med 1999;160:976e981. 2. Rose R, Hunting KJ, Townsend TR, Wenzel RP. Morbidity/mortality and economics of hospital acquired blood-stream infections: a controlled study. South Med J 1977;70:1267e1269. 3. Pittet D, Tarara D, Wenzel RP. Nosocomial bloodstream infections in critically ill patients: excess length of stay, extra costs and attributable mortality. J Am Med Assoc 1994;271: 1598e1601. 4. Spengler RF, Greenough WB. Hospital costs and mortality attributed to nosocomial bacteremias. J Am Med Assoc 1978;240:2455e2458. 5. Coello R, Glenister H, Fereres J. The cost of infection in surgical patients; a caseecontrol study. J Hosp Infect 1993;25: 239e250. 6. Plowman RP, Graves N, Griffin MA. The rate and cost of hospital-acquired infections occurring in patients admitted to selected specialities of a district general hospital in England and the national burden imposed. J Hosp Infect 2001;47: 198e209. 7. Davies T, Cottingham J. The cost of hospital infection in orthopedic patients. J Infect 1979;1:330e338. 8. Kappstein I, Schulgen G, Fraedrich G, Schlosser V, Schumacher M, Daschner FD. Added hospital stay due to wound infections following cardiac surgery. Thorac Cardiovasc Surg 1992;40:148e151. 9. Poulsen KB, Bremmelgaard A, Sorensen AI, Raahave D, Petersen JV. Estimated costs of postoperative wound infections: a caseecontrol study of marginal hospital and social security costs. Epidemiol Infect 1994;113:282e295. 10. Medina M, Martinez-Gallego G, Sillero-Arenas M, DelgadoRodriguez M. Risk factors and length of stay attributable to hospital infections of the urinary tract in general surgery patients. Enferm Infecc Microbiol Clin 1997;15:310e314. 11. Girard R, Fabry J, Meynet R, Lambert DC, Sepetjan M. Costs of nosocomial infection in a neonatal unit. J Hosp Infect 1983;4:361e366. 12. Pirson M, Dramaix M, Struelens M, Riley TV, Leclercq P. Costs associated with hospital-acquired bacteraemia in a Belgian hospital. J Hosp Infect 2005;59:33e40. 13. Sheng WH, Wang JT, Lu DCT, Chie WC, Chen YC, Chang SC. Comparative impact of hospital-acquired infections on medical costs, length of hospital stay and outcome between community hospitals and medical centres. J Hosp Infect 2005;59:205e214. 14. Roberts RR, Scott RD, Cordell R, et al. The use of economic modeling to determine the hospital costs associated with nosocomial infections. Clin Infect Dis 2003;36:1424e1432.
A. Kothari et al. 15. Mermel LA. Correction: catheter related bloodstreaminfections. Ann Intern Med 2000;133:395. 16. Mehta A, Rosenthal VD, Mehta Y, et al. Device-associated nosocomial infection rates in intensive care units of seven Indian cities. Findings of the International Nosocomial Infection Control Consortium (INICC). J Hosp Infect 2007; 67:168e174. 17. Nashef SAM, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16:9e13. 18. Geissler HJ, Holzl P, Marohl S, et al. Risk stratification in heart surgery: comparison of six score systems. Eur J Cardiothorac Surg 2000;17:400e406. 19. Toumpoulis IK, Anagnostopoulos CE, Swistel DG, DeRose Jr JJ. Does EuroSCORE predict length of stay and specific postoperative complications after cardiac surgery? Eur J Cardiothorac Surg 2005;27:128e133. 20. Rello J, Ochagavia A, Sabanes E, et al. Evaluation of outcome of intravenous catheter-related infections in critically ill patients. Am J Respir Crit Care Med 2000;162: 1027e1030. 21. Dimick JB, Pelz RK, Consunji R, Swoboda SM, Hendrix CW, Lipsett PA. Increased resource use associated with catheter-related bloodstream infection in the surgical intensive care unit. Arch Surg 2001;136:229e234. 22. Orsi GB, Di Stefano L, Noah N. Hospital-acquired, laboratory-confirmed bloodstream infection: increased hospital stay and direct costs. Infect Control Hosp Epidemiol 2002; 23:190e197. 23. Warren DK, Quadir WW, Hollenbeak CS, Elward AM, Cox MJ, Fraser VJ. Attributable cost of catheter-associated bloodstream infections among intensive care patients in a nonteaching hospital. Crit Care Med 2006;34:2084e2089. 24. Laupland KB, Lee H, Gregson DB, Manns BJ. Cost of intensive care unit-acquired bloodstream infections. J Hosp Infect 2006;63:124e132. 25. Albrecht SJ, Fishman NO, Kitchen J, et al. Reemergence of gram-negative health care-associated bloodstream infections. Arch Intern Med 2006;166:1289e1294. 26. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 2004;39:309e317. 27. Mathur P, Kapil A, Das B. Nosocomial bacteraemia in intensive care unit patients of a tertiary care centre. Indian J Med Res 2005;122:305e308. 28. Kumar S, Rizwi M, Vidhani S, Sharma VK. Changing face of septicaemia and increasing drug resistance in blood isolates. Indian J Pathol Microbiol 2004;47:441e446. 29. Mehta M, Dutta P, Gupta V. Antimicrobial susceptibility pattern of blood isolates from a teaching hospital in north India. Jpn J Infect Dis 2005;58:174e176. 30. Puskas JD, Williams WH, Mahoney EM, et al. Off-pump vs conventional coronary artery bypass grafting: early and 1-year graft patency, cost, and quality-of-life outcomes: a randomized trial. J Am Med Assoc 2004;291:1841e1849.
Available online at www.sciencedirect.com
www.elsevierhealth.com/journals/jhin
Costs associated with hospital-acquired bacteraemia in an Indian hospital: a caseecontrol study* A. Kothari a,*, V. Sagar b, V. Ahluwalia c, B.S. Pillai c, M. Madan d a
Dr BL Kapur Memorial Hospital, New Delhi, Delhi, India Regions Hospital, St Paul, Minnesota, USA c Max Heart & Vascular Institute, New Delhi, Delhi, India d Ranbaxy Laboratories Ltd, Gurgaon, Haryana, India b
Received 30 July 2008; accepted 22 October 2008 Available online 5 December 2008
KEYWORDS Bacteraemia; Costs; Hospitalacquired infection; Mortality; Developing countries
Summary Studies from around the world have shown that hospital-acquired infections increase the costs of medical care, morbidity and mortality. The aim of this study was to determine cost and attributable mortality associated with hospital-acquired bacteraemia in a tertiary care centre in India. This was a retrospective caseecontrol, cost utility analysis set in the cardiothoracic unit of a 200-bedded tertiary care cardiac hospital. Cases included adult patients who underwent coronary artery bypass graft and/or valve replacement surgery who developed bacteraemia (indicated by positive blood cultures) during postoperative stay (N ¼ 24). Controls were age- and sex-matched adult patients who underwent similar procedures but who did not develop bacteraemia (N ¼ 48). Data were collected from patient medical records and other administrative databases for cost analysis. Prolongation of hospital stay, attributable mortality and extra costs associated with hospital-acquired bacteraemia were analysed. Statistical analysis was done using Fisher’s exact test and unpaired t-test. Patients with hospital-acquired bacteraemia experienced a significantly longer total hospital stay [mean: 22.9 days; 95% confidence interval (CI): 17.2e28.6; P < 0.0001], significantly longer ICU stay (mean: 11.3 days; 95% CI: 9.0e13.6; P < 0.0001), a significantly higher mortality (mean: 54%; P < 0.0001) and cost significantly more (mean: US $14,818; 95%
* Part of this study has been previously presented in the 2007 Annual Meeting of the Infectious Diseases Society of America, Abstract No. 814. * Corresponding author. Address: A. Kothari, A-71, Kirti Nagar, New Delhi, 110015, India. Tel.: þ91 11 9818009668; fax: þ91 11 25447063. E-mail address: [email protected]
0195-6701/$ - see front matter ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2008.10.022
144
A. Kothari et al. CI: 10,663e18,974; P < 0.0001) than controls. We conclude that hospitalacquired bacteraemia significantly increases mortality and costs of hospitalisation in lower income developing countries. Our study demonstrates that costs associated with HAIs are similar between developing and developed countries. Better infection control planning and infrastructure may offset some of these costs. ª 2008 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction Studies from around the world have shown that hospital-acquired infections (HAIs) increase the cost of medical care due to prolongation of hospital stay, and increased morbidity and mortality. Studies of economic cost of HAIs have been undertaken in the USA, England, Germany, Denmark, Spain, France, Belgium, and Taiwan, among other countries.1e13 Previous studies in the USA have estimated that a confirmed HAI resulted in an extra cost of US $15,275 per case.14 Another recent study from the UK stated that patients with nosocomial infections incurred hospital costs 2.9 times higher than uninfected patients, equivalent to an additional £3,154.6 Estimates of the cost of nosocomial infection incllude £930 million per annum in the UK and from US $296 million to 2.3 billion in the USA.6,15 HAIs are a serious problem in developing countries like India.16 However, there is no published literature on analysis of costs associated with HAI in India. The aim of this study was to determine the cost and the attributable mortality associated with hospital-acquired bacteraemia in a tertiary care centre in India.
Methods This was a retrospective caseecontrol study done in 2006. The setting was the cardiothoracic unit of a 200-bedded private, tertiary care cardiac hospital in New Delhi, India. Patients requiring emergency cardiac surgery were excluded from the study. This is because elective cardiothoracic surgical patients without complications have a stable duration of stay as well as total cost of hospitalisation. This allows for a comparison between case and control groups. This is in contrast to general medical and surgical patients, where both duration of stay as well as cost of hospitalisation is variable according to the diagnosis. The
unit did 842 adult cardiac surgical cases in 2006. Preoperative antibiotic prophylaxis used was cefazolin 1 g preoperatively 0.5e1 h before procedure and was continued postoperatively for 48 h.
Inclusion criteria Cases included all adult patients undergoing coronary artery bypass grafting (CABG) and/or valve replacement surgery (mitral and/or aortic valve) who developed clinically significant bacteraemia during postoperative stay (N ¼ 24). Age- and sex-matched adult patients undergoing similar procedures who did not develop detectable bacteraemia during stay were chosen as the control group (N ¼ 48) in a 2:1 ratio.
Preoperative risk evaluation All patients undergoing cardiac surgery were evaluated preoperatively using standard European System for Cardiac Operative Risk Evaluation (EuroSCORE).17e19 The EuroSCORE calculation was done using the EuroSCORE website (www.euroscore.org).
Definitions Hospital-acquired bacteraemia was defined as a bloodstream infection that developed 48 h after hospital admission. One set of positive blood cultures in a patient with clinical signs, who was prescribed specific antibiotics for treatment, was considered as hospital-acquired bacteraemia. If the bacteraemia was caused by a commensal micro-organism, at least two positive sets of blood cultures within a 72 h period with the same organism was required.
Data collection Data were collected using patient medical records, hospital information system and other administrative databases for cost analysis. Measurable
Cost hospital-acquired bacteraemia India
145
Table I Causative organisms for episodes of bacteraemia in cases
Patient demographics
Micro-organisms
N
%
Candida spp. Acinetobacter baumannii Pseudomonas aeruginosa Klebsiella pneumoniae Enterococcus faecalis Staphylococcus aureus Stenotrophomonas maltophila Enterobacter spp.
10 5
35 17
There were 24 patients who fulfilled the inclusion criteria of cases during the time period of the study. The mean age of the cases was 64.8 years (SD: 9.9). The mean age of the 48 controls was 60.9 years (SD: 9).
4
14
EuroSCORE
3 2 2 2
10 7 7 7
1
3
The mean EuroSCORE of the cases, as calculated preoperatively, was 2.93 (SD: 2.08); and that of controls was 3.07 (SD: 1.61). There was no statistically significant difference between the EuroSCOREs of the two groups (P ¼ 0.7521; 95% CI: e1.03 to 0.74).
Causative organisms elements included prolongation of hospital stay, increase in mortality and extra costs associated with hospital-acquired bacteraemia. All costs were calculated in Indian rupees (Rs) and were then converted to US $ using the existing exchange rate in 2006 (Rs45.30 per US $) before doing statistical analysis.
The 24 cases had 29 episodes of hospital-acquired bacteraemia during their stay in the hospital (Table I). Five patients had two episodes of bacteraemia during hospitalisation. Sources of infection included central venous catheters (62%), VAPs (17%), and CAUTIs (7%) and were unknown in 14% of cases.
Mortality rate
Statistical analysis Statistical analysis was done using GraphPad Software. Continuous variables were defined as mean and SD. Comparative tests on variables were done using Fisher’s exact test and unpaired t-test.
Of the 24 cases with hospital-acquired bacteraemia, 13 died (54%). By comparison, none of the patients in the control group had any mortality (0%). This difference was highly significant (P < 0.0001).
Length of stay (LOS)
Results Nosocomial infection rate The overall central line-related bloodstream infection (CRBSI) rate in the cardiothoracic unit was 25 per 1000 central line days in 2006. Similarly, the rate of ventilator-associated pneumonia (VAP) was 0.7 per 1000 ventilator days and that of catheter-associated urinary tract infection (CAUTI) was 11.6 per 1000 catheter days. The surgical site infection rate in adult cardiac surgical patients was 0.6%. Table II
The mean LOS of the cases was 33.2 days (SD: 19.8) (Table II). For the control group, the mean LOS was 10.3 days (SD: 1.9). The difference in the mean LOS in the two groups was 22.9 days, which was highly significant (P < 0.0001; 95% CI: 17.2e28.6 days).
Cost analysis In the control group, i.e. adult patients who underwent CABG and/or valve replacement and did not develop bacteraemia, the total mean cost during hospital stay was US $8,055 (SD: 1,638) (Table III). Among cases, i.e. patients who
Comparison of length of stay (LOS) in cases and controls Mean (SD) LOS (days) Controls (N ¼ 48)
Intensive care unit High dependency unit Ward Total CI, confidence interval.
3.2 0.8 6.2 10.3
(1.7) (1.1) (1.6) (1.9)
P-value
95% CI
Cases (N ¼ 24) 14.5 12.9 5.7 33.2
(7.6) (13.9) (7.6) (19.8)
<0.0001 <0.0001 0.6093 <0.0001
9.0e3.6 8.0e16.1 2.8 to 1.6 17.2e28.6
146 Table III
A. Kothari et al. Cost analysis in controls and cases (US $) Controls Mean
Bed charges 855 Investigations 699 Pharmacy 1,821 Consults 18 OR charges 4,476 Blood bank 164 Bedside procedures 22 Dialysis e
Cases
%
Mean
%
10.6 8.7 22.6 0.2 55.6 2 0.3 e
3,274 3,160 9,875 407 4,473 940 410 334
14.3 13.8 43.2 1.8 19.5 4.1 1.8 1.5
OR, operating room.
underwent CABG and/or valve replacement and developed bacteraemia, the total mean cost during hospital stay was US $22,873 (SD: 14,350) (Table III). The difference in the mean total costs between the two groups was US $14,818, which was highly significant (P < 0.0001; 95% CI: US $10,663e18,974). A subset analysis was done of patients with hospital-acquired bacteraemia who survived (N ¼ 11) and age- and sex-matched controls without bacteraemia (N ¼ 22) undergoing similar elective procedures in a 2:1 ratio. The mean age of the survivors was 65.3 years (SD: 10.6) and that of matched controls was 62.3 years (SD: 6.9). The mean EuroSCORE of the survivors, as calculated preoperatively, was 3.12 (SD: 2.81); and that of controls was 2.75 (SD: 1.52). There was no statistically significant difference between the EuroSCOREs of the two groups (P ¼ 0.6219; 95% CI: e1.15 to 1.90). The results of this analysis, which are presented in Table IV, also showed that difference in total stay between the two groups was 21.9 days, ICU stay 10.7 days and HDU stay 8.2 days, all of which were highly significant. The difference in ward stay was 3 days, but this was not significant. The cost difference between these groups was US $9,946, which was highly significant.
Table IV
Discussion In our study, which is the first attempt to determine costs associated with nosocomial infections from the Indian subcontinent, extra cost of each case of hospital-acquired bacteraemia was US $14,818. Even when a comparison was done between survivor cases and controls, the excess cost incurred by each case of BSI was US $9,946. Other studies done in North America and Europe have also estimated an extra cost per BSI to range from US $11,971 to 56,167 (Table V). Although the increase in costs shown in our study was due to several factors, pharmacy contributed the highest share (Table VI). This included not only antimicrobials and other drugs, but also medical consumables such as intravenous lines, catheters, etc. Other major factors for the increased costs were investigation charges and bed charges (due to increased LOS). In our study, patients with infections had an increased total LOS of 22.9 days, and an extra ICU stay of 11.3 days. This is in line with previous studies that have shown an extra ICU stay of 2e20 days and increased total LOS of 7.5e24 days (Table V). Attributable mortality due to hospital-acquired bacteraemia in our study was 54%, in contrast to Western literature with values between 4.4% to 41% (Table V). Only two studies have shown no significant differences in mortality between patients with nosocomial BSIs and controls.1,20 This could be due to the high preponderance of coagulasenegative staphylococci (CoNS) (63% and 46%) isolated from cases in these studies. CoNS are lower virulence pathogens compared with Gram-negative bacteria and candida, which were predominant in our study. The absence of mortality in our control group could be due to the fact that we chose elective cardiac surgical patients as our study population. This group has a lower mortality rate compared with other medical and surgical ICU patients.
Results of subset analysis of survivors and controls Subgroup of survivors with BSI (N ¼ 11) Controls (N ¼ 22) P-value Mean (SD) Mean (SD)
Length of stay (days) ICU HDU Ward Total Cost of hospitalisation (US $)
13.9 (7.2) 9 (9.4) 9.2 (9.9) 32.1 (20.3) 18,013 (8696)
3.2 (1.8) <0.0001 0.8 (1.2) 0.0003 6.2 (1.7) 0.1786 10.2 (1.8) <0.0001 8,067 (1790) <0.0001
BSI, bloodstream infection; CI, confidence interval; ICU, intensive care unit; HDU, high dependency unit.
95% CI
7.4e13.9 4.1e12.3 1.4 to 7.3 13.1e30.6 6,063e13,827
Cost hospital-acquired bacteraemia India
147
Table V Cost, mortality and length of stay associated with nosocomial bloodstream infections (BSIs): review of literature Author
Year Country
Setting
Cases
Rose et al.2 Spengler and Greenough4 Pittet et al.3
1977 USA 1978 USA
ICU and wards ICU and wards
40 81
1994 USA
ICU
86
Digiovine et al.1
1999 USA
ICU
68
Rello et al.20 Dimick et al.21
2000 Spain 2001 USA
ICU ICU
49 86
Orsi et al.22
2002 Italy
ICU and wards
105
Pirson et al.12
2005 Belgium ICU and wards
46
32.2%
Warren et al.23
2006 USA
ICU
41
23%
ICU
144
16%
Laupland et al.24 2006 Canada
Attributable mortality 28% 14 times greater 35% 4.4% (not significant) Not significant 35% (total) 33% (ICU) 35.2e40.9%
Extra length of stay
Extra cost per BSI
19 days total 14 days total
US $4,370 US $3,600
24 days total 8 days ICU 10 days ICU
US $40,000
19.6 days total 22 days total 20 days ICU 19.1e19.9 days total 21.1 days total 7.2 days ICU 7.5 days total 2.4 days ICU 2 days ICU
V3,124 US $56,167 Total US $71,443 ICU V16,356
US $34,508
V12,853 US $11,971 US $12,321
ICU, intensive care unit.
Gram-negative organisms contributed to 51% of isolates in our study, with candida contributing 35%. While there has been an increasing trend in Gram-negative septicaemia, most Western literature shows a predominance of Gram-positive organisms as causative agents of bacteraemia.25,26 Several different studies from India in various teaching hospitals have shown Gram-negative organisms as being the predominant cause of bacteraemia, with rates ranging between 66 and 81%.27e29 There are, however, certain drawbacks in our study. We had a small sample size, and findings may be different with larger studies. Also, we did not calculate the opportunity cost to the hospital, as well as cost of loss of productivity. Depending on the setting and the socio-economic situation, the opportunity cost (to the health service) and the loss of productivity may far exceed the direct cost of hospitalisation. This would make a stronger
Table VI Comparison of pharmacy costs between cases and controls Mean cost (US $) % increase Controls Drugs (excluding antimicrobials) Antimicrobials Medical consumables Total cost
Cases
180
1,687
937
59 1,582 1,821
2,680 5,508 9,875
4,542 348 542
case for more investment in infection control as a means of decreasing healthcare costs. We estimate that hospital-acquired bacteraemia leads to an excess burden of US $980,000 in our cardiothoracic unit. This study illustrates that even though the cost of healthcare is much lower in India than the Western countries, the costs of HAI are similar in both regions.30 Considering the socio-economic situation in India, where procurement of health insurance (both government and private) is limited and per capita income is very low, nosocomial infections place an even greater burden on resources than in the West. There is a need for larger studies in the developing world on costs associated with HAI. Infrastructure in hospitals and knowledge about optimal infection control practices are deficient in the healthcare sector in India. The results of this study emphasise the need for greater resource allocation for infrastructure and training in infection control and development of local guidelines for prevention of nosocomial infections.
Acknowledgements We thank Dr T.D. Chugh, Dr D. Shuckla and A. Pruthi for their contributions during the preparation of the manuscript. Conflict of interest statement None declared.
148 Funding sources None.
References 1. Digiovine B, Chenoweth C, Watts C, Higgins M. The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit Care Med 1999;160:976e981. 2. Rose R, Hunting KJ, Townsend TR, Wenzel RP. Morbidity/mortality and economics of hospital acquired blood-stream infections: a controlled study. South Med J 1977;70:1267e1269. 3. Pittet D, Tarara D, Wenzel RP. Nosocomial bloodstream infections in critically ill patients: excess length of stay, extra costs and attributable mortality. J Am Med Assoc 1994;271: 1598e1601. 4. Spengler RF, Greenough WB. Hospital costs and mortality attributed to nosocomial bacteremias. J Am Med Assoc 1978;240:2455e2458. 5. Coello R, Glenister H, Fereres J. The cost of infection in surgical patients; a caseecontrol study. J Hosp Infect 1993;25: 239e250. 6. Plowman RP, Graves N, Griffin MA. The rate and cost of hospital-acquired infections occurring in patients admitted to selected specialities of a district general hospital in England and the national burden imposed. J Hosp Infect 2001;47: 198e209. 7. Davies T, Cottingham J. The cost of hospital infection in orthopedic patients. J Infect 1979;1:330e338. 8. Kappstein I, Schulgen G, Fraedrich G, Schlosser V, Schumacher M, Daschner FD. Added hospital stay due to wound infections following cardiac surgery. Thorac Cardiovasc Surg 1992;40:148e151. 9. Poulsen KB, Bremmelgaard A, Sorensen AI, Raahave D, Petersen JV. Estimated costs of postoperative wound infections: a caseecontrol study of marginal hospital and social security costs. Epidemiol Infect 1994;113:282e295. 10. Medina M, Martinez-Gallego G, Sillero-Arenas M, DelgadoRodriguez M. Risk factors and length of stay attributable to hospital infections of the urinary tract in general surgery patients. Enferm Infecc Microbiol Clin 1997;15:310e314. 11. Girard R, Fabry J, Meynet R, Lambert DC, Sepetjan M. Costs of nosocomial infection in a neonatal unit. J Hosp Infect 1983;4:361e366. 12. Pirson M, Dramaix M, Struelens M, Riley TV, Leclercq P. Costs associated with hospital-acquired bacteraemia in a Belgian hospital. J Hosp Infect 2005;59:33e40. 13. Sheng WH, Wang JT, Lu DCT, Chie WC, Chen YC, Chang SC. Comparative impact of hospital-acquired infections on medical costs, length of hospital stay and outcome between community hospitals and medical centres. J Hosp Infect 2005;59:205e214. 14. Roberts RR, Scott RD, Cordell R, et al. The use of economic modeling to determine the hospital costs associated with nosocomial infections. Clin Infect Dis 2003;36:1424e1432.
A. Kothari et al. 15. Mermel LA. Correction: catheter related bloodstreaminfections. Ann Intern Med 2000;133:395. 16. Mehta A, Rosenthal VD, Mehta Y, et al. Device-associated nosocomial infection rates in intensive care units of seven Indian cities. Findings of the International Nosocomial Infection Control Consortium (INICC). J Hosp Infect 2007; 67:168e174. 17. Nashef SAM, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg 1999;16:9e13. 18. Geissler HJ, Holzl P, Marohl S, et al. Risk stratification in heart surgery: comparison of six score systems. Eur J Cardiothorac Surg 2000;17:400e406. 19. Toumpoulis IK, Anagnostopoulos CE, Swistel DG, DeRose Jr JJ. Does EuroSCORE predict length of stay and specific postoperative complications after cardiac surgery? Eur J Cardiothorac Surg 2005;27:128e133. 20. Rello J, Ochagavia A, Sabanes E, et al. Evaluation of outcome of intravenous catheter-related infections in critically ill patients. Am J Respir Crit Care Med 2000;162: 1027e1030. 21. Dimick JB, Pelz RK, Consunji R, Swoboda SM, Hendrix CW, Lipsett PA. Increased resource use associated with catheter-related bloodstream infection in the surgical intensive care unit. Arch Surg 2001;136:229e234. 22. Orsi GB, Di Stefano L, Noah N. Hospital-acquired, laboratory-confirmed bloodstream infection: increased hospital stay and direct costs. Infect Control Hosp Epidemiol 2002; 23:190e197. 23. Warren DK, Quadir WW, Hollenbeak CS, Elward AM, Cox MJ, Fraser VJ. Attributable cost of catheter-associated bloodstream infections among intensive care patients in a nonteaching hospital. Crit Care Med 2006;34:2084e2089. 24. Laupland KB, Lee H, Gregson DB, Manns BJ. Cost of intensive care unit-acquired bloodstream infections. J Hosp Infect 2006;63:124e132. 25. Albrecht SJ, Fishman NO, Kitchen J, et al. Reemergence of gram-negative health care-associated bloodstream infections. Arch Intern Med 2006;166:1289e1294. 26. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 2004;39:309e317. 27. Mathur P, Kapil A, Das B. Nosocomial bacteraemia in intensive care unit patients of a tertiary care centre. Indian J Med Res 2005;122:305e308. 28. Kumar S, Rizwi M, Vidhani S, Sharma VK. Changing face of septicaemia and increasing drug resistance in blood isolates. Indian J Pathol Microbiol 2004;47:441e446. 29. Mehta M, Dutta P, Gupta V. Antimicrobial susceptibility pattern of blood isolates from a teaching hospital in north India. Jpn J Infect Dis 2005;58:174e176. 30. Puskas JD, Williams WH, Mahoney EM, et al. Off-pump vs conventional coronary artery bypass grafting: early and 1-year graft patency, cost, and quality-of-life outcomes: a randomized trial. J Am Med Assoc 2004;291:1841e1849.