Gram-Negative Bacterial Infection, Infection Resistant to Drugs
Conditions
Keywords
Enterobacteriaceae, Pseudomonas, Acinetobacter, Stenotrophomonas, Mortality, Morbidity, Cost of Illness, Causality, Multicenter Study, Cohort Studies, Case-Control Studies, Prospective Studies, Follow-up Studies
Brief summary
This study aims to assess how large an additional disease burden and what extra costs are generated by antibiotic resistance in patients suffering from infections caused by gram-negative bacteria, such as Escherichia coli and Pseudomonas aeruginosa, in hospitals in the Netherlands.
Detailed description
This study addresses the following three aims: 1. To provide a more accurate estimate than currently available of the incremental disease burden and attributable costs of antibiotic-resistant as compared to antibiotic-sensitive gram-negative bacteria (i.e. Enterobacteriaceae and non-fermenters). This analysis is focused on gram-negative infections for which patients are hospitalized. In a less detailed manner, the same analysis of disease burden and costs can be performed for acquiring a gram-negative infection during hospitalization. 2. To identify determinants associated with resistance in gram-negative infections, to the extent that they are confounders of the relation between resistance and outcome. 3. To adapt and optimize existing methodology to measure the burden of resistance, among others by calculating disability-adjusted life years (DALYs) which incorporate not merely mortality, but also morbidity. GRAND-ABC is designed as a prospective parallel matched cohort, which will run for a year in each of the eight participating hospitals. The primary cohort is a random sample of all Gram-negative infections occurring in a participating hospital during the study period. This cohort can be divided on the basis of the primary determinant status (whether the Gram-negative pathogen is resistant or not based on Dutch guideline for multi-drug resistant organisms; Werkgroep Infectiepreventie (WIP). Bijzonder resistente micro-organismen (BRMO). December 2012. http://www.wip.nl/free\_content/Richtlijnen/130424\_BRMO.pdf) into two parallel subcohorts. Each patient in each of the subcohorts will be matched to one patient without a gram-negative infection. Together these will form the secondary cohort of non-infected patients: patients admitted to the hospital during the study period who are within the same risk set as the infected patients. For all patients data collection will be performed by review of medical files, which will cover the entire admission during which they were included in the study, and all cause 30 day mortality. Data collection for the hospital stay covers confounders and effect modifiers of the associations studied, and feeds into the outcomes costs, DALYs and length of stay. For the cohort with gram-negative infections, data on infection parameters and antibiotic treatment parameters are also collected. In addition, the subcohort with infections by multi-drug resistant organisms and a random 20% of the subcohort with infections by sensitive organisms will be selected for follow-up, consisting of sending questionnaires and renewed medical file review 30 days after the index culture date. In the case of ongoing sequelae of the gram-negative infection, this procedure is repeated 90 days after the index culture date. These questionnaires will feed into the outcomes costs, DALYs and quality-adjusted life years (QALYs).
Interventions
None listed
Sponsors
ZonMw: The Netherlands Organisation for Health Research and Development
UMC Utrecht
Study design
Observational model
OTHER
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Please refer to descriptions of two cohorts
Exclusion criteria
* Patients on children's wards * Patients admitted to wards for long-term care or psychiatric wards, that were not subsequently admitted to acute care wards as a consequence of the infection * Patients admitted to wards that are excluded for logistic reasons such as the non-availability of electronic patient files, that were not subsequently admitted to included acute care wards as a consequence of the infection * Patients that have been included in the cohort of infected patients during the same hospitalization or within the past 30 (if not eligible for follow-up) or 90 (if eligible for follow-up) days
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| All cause mortality | Up to 30 days | Death (whether in-hospital or after discharge) from any cause, as apparent from medical record or municipal registry. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Costs | Hospital stay (all patients; expected average 1 week) and up to 90 days (follow-up patients) | Costs generated from societal perspective, including: * Direct costs within healthcare sector, based on relating standard Dutch reference prices to (1) restricted chart review for all patients, (2) extensive collection of cost data in two of eight participating hospitals, (3) health care use after discharge, as recorded from medical files and reported by patients in questionnaires * Other direct costs (own out-of-pocket expenses and time invested by caregivers), as reported by patients in questionnaires * Indirect productivity losses, as reported by patients in questionnaires * (possibly) Decision-analytic modelling of costs generated by sequelae not within 90 days of index culture date Questionnaires are not available for the non-infected cohort, and therefore the costs for acquiring gram-negative infections can only be calculated from the hospital perspective. |
| DALYs | Up to 90 days | Years of Life Lost (YLL) and Years Lived with Disability (YLD) attributable to infection as apparent from an outcome tree of health outcomes related to gram-negative infections. This tree is preconceived, but modifiable according to observed sequelae in the study. A mathematical model for this outcome will be constructed that incorporates, apart from the observed sequelae in the study, factors such as (1) preexisting decreased life expectancy and quality of life due to comorbidity, (2) known transition parameters between health outcomes from literature, (3) confounding effects of comorbidity on transition parameters, (4) sequelae not observed within the time frame of data collection for the study, (5) 'baseline' change in health outcomes during hospitalization as apparent from the non-infected cohort, and (6) occupancy of several health outcome simultaneously. This outcome will not be calculated for the non-infected cohort. |
| Length of stay | Hospital stay (expected average 1 week) | Number of days until hospital discharge. |
Other
| Measure | Time frame | Description |
|---|---|---|
| QALYs | Up to 90 days | Measured by EuroQol 5 Dimensions 5 Levels (EQ-5D-5L) in questionnaires, as a confirmation of the DALY model results. This outcome will not be calculated for the non-infected cohort. |
Countries
Netherlands
Outcome results
None listed