Comparison of two procedures to estimate the hospital stay attributable to nosocomial infection: matched cohort study versus analysis of covariance of the total unmatched cohort

Estimating excess length of stay due to healthcare-associated infections: a systematic review and meta-analysis of statistical methodology

BACKGROUND

Healthcare-associated infection (HCAI) affects millions of patients worldwide. HCAI is associated with increased healthcare costs, owing primarily to increased hospital length of stay (LOS) but calculating these costs is complicated due to time-dependent bias. Accurate estimation of excess LOS due to HCAI is essential to ensure that we invest in cost-effective infection prevention and control (IPC) measures.

AIM

To identify and review the main statistical methods that have been employed to estimate differential LOS between patients with, and without, HCAI; to highlight and discuss potential biases of all statistical approaches.

METHODS

A systematic review from 1997 to April 2017 was conducted in PubMed, CINAHL, ProQuest and EconLit databases. Studies were quality-assessed using an adapted Newcastle-Ottawa Scale (NOS). Methods were categorized as time-fixed or time-varying, with the former exhibiting time-dependent bias. Two examples of meta-analysis were used to illustrate how estimates of excess LOS differ between different studies.

FINDINGS

Ninety-two studies with estimates on excess LOS were identified. The majority of articles employed time-fixed methods (75%). Studies using time-varying methods are of higher quality according to NOS. Studies using time-fixed methods overestimate additional LOS attributable to HCAI. Undertaking meta-analysis is challenging due to a variety of study designs and reporting styles. Study differences are further magnified by heterogeneous populations, case definitions, causative organisms, and susceptibilities.

CONCLUSION

Methodologies have evolved over the last 20 years but there is still a significant body of evidence reliant upon time-fixed methods. Robust estimates are required to inform investment in cost-effective IPC interventions.

The economic burden of nurse-sensitive adverse events in 22 medical-surgical units: retrospective and matching analysis

AIMS

The aim of this study was to assess the economic burden of nurse-sensitive adverse events in 22 acute-care units in Quebec by estimating excess hospital-related costs and calculating resulting additional hospital days.

BACKGROUND

Recent changes in the worldwide economic and financial contexts have made the cost of patient safety a topical issue. Yet, our knowledge about the economic burden of safety of nursing care is quite limited in Canada in general and Quebec in particular.

DESIGN

Retrospective analysis of charts of 2699 patients hospitalized between July 2008 - August 2009 for at least 2 days of 30-day periods in 22 medical-surgical units in 11 hospitals in Quebec.

METHODS

Data were collected from September 2009 to August 2010. Nurse-sensitive adverse events analysed were pressure ulcers, falls, medication administration errors, pneumonia and urinary tract infections. Descriptive statistics identified numbers of cases for each nurse-sensitive adverse event. A literature analysis was used to estimate excess median hospital-related costs of treatments with these nurse-sensitive adverse events. Costs were calculated in 2014 Canadian dollars. Additional hospital days were estimated by comparing lengths of stay of patients with nurse-sensitive adverse events with those of similar patients without nurse-sensitive adverse events.

RESULTS

This study found that five adverse events considered nurse-sensitive caused nearly 1300 additional hospital days for 166 patients and generated more than Canadian dollars 600,000 in excess treatment costs.

CONCLUSION

The results present the financial consequences of the nurse-sensitive adverse events. Government should invest in prevention and in improvements to care quality and patient safety. Managers need to strengthen safety processes in their facilities and nurses should take greater precautions.

Economic burden of surgical site infections at a European university hospital

OBJECTIVE

To quantify the economic burden of in-hospital surgical site infections (SSIs) at a European university hospital.

DESIGN

Matched case-control study nested in a prospective observational cohort study.

SETTING

Basel University Hospital in Switzerland, where an average of 28,000 surgical procedures are performed per year.

METHODS

All in-hospital occurrences of SSI associated with surgeries performed between January 1, 2000, and December 31, 2001, by the visceral, vascular, and traumatology divisions at Basel University Hospital were prospectively recorded. Each case patient was matched to a control patient by age, procedure code, and National Nosocomial Infection Surveillance System risk index. The case-control pairs were analyzed for differences in cost of hospital care and in provision of specialized care.

RESULTS

A total of 6,283 procedures were performed: 187 SSIs were detected in inpatients, 168 of whom were successfully matched with a control patient. For case patients, the mean additional hospital cost was SwF-19,638 (95% confidence interval [CI], SwF-8,492-SwF-30,784); the mean additional postoperative length of hospital stay was 16.8 days (95% CI, 13-20.6 days); and the mean additional in-hospital duration of antibiotic therapy was 7.4 days (95% CI, 5.1-9.6 days). Differences were primarily attributable to organ space SSIs (n = 76).

CONCLUSIONS

In a European university hospital setting, SSIs are costly and constitute a heavy and potentially preventable burden on both patients and healthcare providers.

Are we really seeing the total costs of surgical site infections? A Spanish study

To identify overall costs generated by surgical site infections (SSI) patients, including indirect costs. A prospective study of case series of patients who have undergone major surgical treatment was undertaken. Patients who suffered SSI were compared with controls (nested case-control design). Centers for Disease Control and Prevention definitions were followed and SSI established. Overall costs and indirect related morbidity/mortality costs were estimated. The study was performed in a general, tertiary hospital (Valencia, Spain) for 4.5 years. Surgical site infections patients were 9.02% of the total people who underwent surgery. Their stays were prolonging by 14 days, and resources were used more intensely and for longer periods than in controls. Excess hospital costs were $10,232 per patient of which 37% corresponded to prolonged stays. Health costs only accounted for 10% of overall costs; $97,433 per patient including indirect social costs. Studies merely assessing excess costs due to prolonged stays of SSI patients do not reflect the entire scenario as they simply represent 35% of real hospital costs. A comprehensive appraisal shows that total healthcare expenditures represent a tenth of overall costs, which strengthens the claims that investment in preventing SSI would be highly cost-effective.

Postoperative Wundinfektionen

Surgical site infections can be traced to discrepancies in one specific hospital department: the operating suite. Therefore, prevention is often viewed as resting completely on the surgeon. However, the source of micro-organisms responsible for surgical site infections can be endogenous or exogenous. Most infections are believed to be the former, i.e. caused by micro-organisms already resident in the patient's body. Therefore the surgeon can be regarded as suspect only in exceptional cases and usually himself a victim. Prevention is possible not only for exogenous surgical site infection but also many endogenous infections. A multicenter surveillance of infection rates at 130 operative departments participating for at least 4 years in the German National Nosocomial Infection Surveillance System was conducted. A significant 25% reduction in the 3rd year was observed compared with patients who underwent surgery within the 1st year of participation. However, surgeons alone cannot achieve such a decrease, and a team approach is required under most circumstances.

Behavior of cross-sectional surveys in the hospital setting: a simulation model

OBJECTIVE

To date, it has not been adequately proven whether the published formulas used to obtain incidence from the prevalence of nosocomial infections provide a good estimate of real incidence. With the hypothesis that within the hospital setting prevalence may be lower than incidence, the aim of this study was to analyze the behavior of point prevalence as it relates to cumulative incidence and duration of infection.

DESIGN

Hospital simulation study.

METHODS

By randomly selecting a sample of infected patients within a specific range of cumulative incidences and infection durations, we constructed a simulated hospital population, allowing us to estimate daily point prevalences and their maximum and minimum values. The association between the different components of stay and cumulative incidence was evaluated to obtain a more accurate estimate of incidence.

RESULTS

Prevalence can be lower than, equal to, or higher than the corresponding incidence. For all incidence levels, prevalence was increasing with duration. Between 14 and 20 days of infection duration, prevalence was consistently lower than incidence. Prevalence duration of infection was approximately half the time of the total duration.

CONCLUSIONS

The existing formulas relating incidence and prevalence can frequently be inadequate. Until a validated system for converting prevalence into incidence is available, we do not believe their use is appropriate.

Using Cost of Infection as a Tool to Demonstrate a Difference in Prophylactic Antibiotic Efficacy: A Prospective Randomized Comparison of the Pharmacoecomonic Effectiveness of Ceftriaxone and Cefotaxime Prophylaxis in Abdominal Surgery

The purpose of this study was to test the hypothesis that cost, as well as frequency of infection, could be used to demonstrate a difference in the performance of prophylactic antibiotics. In a prospective, randomized, double-blind study, 1013 patients undergoing abdominal surgery were given 1 g of intravenous ceftriaxone (R) or cefotaxime (C) at induction of anesthesia, and an additional 500 mg of metronidazole for colorectal surgery. Infection was checked for during the hospital stay and at 30 days postoperatively. The inpatient, outpatient, and community costs of infection were prospectively collected. The frequency of wound infection for appendectomies when additional metronidazole was not administered was greater with cefotaxime (R 6%, C 18%, p < 0.05), but the cost of infection was the same (average cost R $994 +/- SD $1101, C $878 +/- $1318). For all other procedures, the frequency of wound infection was similar (R 8%, C 10%), but the cost was less with ceftriaxone (R $887 +/- $1743, C $2995 +/- $6592, p < 0.05). Ceftriaxone decreased the frequency but not the cost of chest and urinary infection (frequency R 6%, C 11%, p < 0.02, cost R $1273 +/- 2338, C $1615 +/- 4083). Differences in both the frequency and cost of all infection are also presented. Ceftriaxone decreased either the frequency or the cost of different postoperative infections. The cost of infection can increase the discriminatory power of trials comparing antibiotic effectiveness.

Factors related to hospital stay among patients with nosocomial infection acquired in the intensive care unit

OBJECTIVES

To assess the influence of nosocomial infection on length of stay in the intensive care unit (ICU) and to determine the relative effect of other factors on extra length of hospitalization associated with nosocomial infection.

DESIGN

Prospective cohort multicenter study in the context of the ENVIN-UCI project.

SETTING

Medical or surgical ICUs of 49 different hospitals in Spain.

METHODS

All consecutive patients (N = 6,593) admitted to ICUs of the participating hospitals who stayed for more than 24 hours during a 3-month period (from January 15 to April 15, 1996) were included. Length of ICU stay was compared between patients with and without nosocomial infections. RESULTS Uninfected patients (N = 5,868) had a median stay in the ICU of 3 days, whereas the median for infected patients (N = 725) was 17 days (P < .001). The median for infected patients with one episode of nosocomial infection was 13 days. The greatest length of stay (40 days) was among patients admitted to the ICU because of medical diseases, with an infection acquired before admission to the ICU, and with the largest number of nosocomial infection episodes. In extended stays, nosocomial infection was significantly associated with length of hospitalization (day 21; odds ratio, 22.38; 95% confidence interval 16.6 to 30.4), whereas an effect of variables related to severity of illness on admission (Acute Physiology and Chronic Health Evaluation II score, urgent surgery, and infection prior to ICU admission) was not found.

CONCLUSIONS

The presence of nosocomial infection and the number of infection episodes were the variables with the strongest association with prolonged hospital stay among ICU patients.

Relationship between hospital infection and long-term mortality in general surgery: a prospective follow-up study

A prospective study of 1431 patients admitted to a general surgery department were followed up for a median of 6.2 years after discharge (7679 person-years of follow-up). We collected information on underlying conditions, including severity of illness, and healthcare-related variables. Relative rates of death and their 95% confidence interval (CI) were estimated using person-years as the denominator. Multiple-risk factors adjusted for relative rates (RR) were obtained using Poisson regression analysis. There were 172 deaths during the follow-up period after hospital discharge (2/100 person-years). Follow-up was complete in 91% of the cohort. There were no important differences in demographic characteristics or risk factors between patients followed up and those lost to follow-up. The death rate in patients with any hospital-acquired infection was 5.3/100 person-years, and the relative rate was 3.07 (95% CI: 2.20-4.24). After adjusting for the main predictors of mortality, we found an effect modification by the presence of chronic disease (P = 0.01 for the product-term between hospital infection and the diagnosis of chronic diseases). Among patients without any underlying chronic disease, hospital-acquired infection was related to a significantly higher long-term mortality (RR = 2.47, 95% CI: 1.24-4.91). In these patients, surgical wound infection yielded a RR of mortality of 3.44 (95% CI: 1.63-7.27). Among patients with underlying chronic disease no association between hospital infection and long-term mortality was found. No evidence of an important modification of the relative rate along the follow-up period was observed. In conclusion surgical patients without chronic disease developing hospital-acquired infection have an increased risk of long-term mortality.

[Nosocomial bacteremia in the adult patient. Study of associated costs]

OBJECTIVE

Nosocomial infection causes a prolonged hospital stay and an increase in care costs. The objective of this study was to determine the length of stay excess and costs attributable to nosocomial bacteremia.

PATIENTS AND METHODS

Retrospective study of clinical records of 148 patients with nosocomial bacteremia during 1996. A matched case-control study was performed. For matching, the following parameters were used: RDG, year of admission, age 10 years, main diagnosis and number of secondary diagnoses. Costs were determined by excess length of hospital stay and calculating alternative costs.

RESULTS

Matching was obtained for 100 cases (67.5%) and cost estimation was performed. Compared with cases, non-matched cases showed differences regarding significant issues for cost, such as hospital stay ( p = 0.01), number of empirical (p = 0.001) or definitive antibiotics (p = 0.03). The median hospital stay for cases was longer than for controls (35 vs 15.5 days, respectively; p = 0.000). When only survivor case-control pairs were considered (n = 75), cases remained in hospital for a median of 36 vs 15 days for controls (p = 0.000). Hospital stay days attributable to nosocomial bacteremia were 19.5 for all matched and 21 for matched survivor cases. Only 76% of cases had stay days attributable to bacteremia. Significant differences between cases and controls included: the mean total costs of admission (p = 0.000), cost of stay (p = 0.001), pharmaceutical expenses (p = 0.000), and cost of microbiological studies (p = 0.000), laboratory work-up (p = 0.001) and radiological studies (p = 0.000). Hospital stay represented more than 60% of costs, followed by pharmaceutical expenses. Cost differences between bacteremic patients and controls, calculated in function of stay median, was 4.424 euros (p = 0.000) and 4.744 euros (p = 0.000) for alternative costs. Ten cases showed a difference that represented more than half of the total difference.

CONCLUSIONS

Nosocomial bacteremia represent a stay prolongation and a significant economical burden. Hospital stay and pharmaceutical expenses accounted for the most part of the associated costs. The differences in costs obtained with both methods were small. Since not all selected cases were matched, there may be an error in the appreciation of the difference between cases and controls.

Stress-induced susceptibility to bacterial infection during cutaneous wound healing

Psychological stress delays wound healing and decreases immune/inflammatory responses required for bacterial clearance. To determine if stress increases the susceptibility to wound infection, female SKH-1 mice were subjected to restraint stress (RST) beginning 3 days prior to the placement of cutaneous wounds. Viable bacteria were quantified from harvested wounds. RST delayed healing by 30% and caused a 2- to 5-log increase in opportunistic bacteria (e.g., Staphylococcus aureus) when compared to wounds from control animals (p <.05). By day 7, 85.4% of the wounds from RST mice had bacterial counts predictive of infection compared to 27.4% from control mice (p <.001). To assess the role of RST-induced glucocorticoids in bacterial clearance, mice were treated with the glucocorticoid receptor antagonist RU486. RU486 reduced opportunistic bacteria by nearly 1 log in wounds from RST mice (p <.05). Thus, stress impairs bacterial clearance during wound healing, resulting in a significant increase in the incidence of opportunistic infection.

Estimation of extra hospital stay attributable to nosocomial infections: heterogeneity and timing of events

Infections acquired in hospital are likely to affect the duration of hospitalization. Suitable statistical methods for estimating the extra days spent in hospital due to nosocomial infections should allow modeling of the heterogeneity of the patient population and the timing of events, as failure to account for important covariates and failure to model adequately the timing of events may lead to biased results. Three approaches have been used in the past to estimate the extra stay: a comparison of duration of stay of infected and uninfected patients, matching of infected and uninfected patients with respect to potentially important determinants of the length of hospital stay, and matching for time-to-infection in addition to the other factors. While these approaches can allow for the heterogeneity of the patient population, none takes sufficient account of the real timing of events and may overestimate the effect of nosocomial infections. We explored the statistical methods available for analyzing time-to-event data and derived alternative methods to estimate the extra stay that appropriately account for heterogeneity and timing. Data from two prospective cohort studies on postoperative wound infection and on nosocomial pneumonia showed that the two-group comparison yields the highest estimates of extra stay (21 and 14 extra days), while matching for confounders and time reduced the estimates to 11 and 8 extra days; our methods yield even lower results (10-12 and 3-4 extra days).