Research Methods in Healthcare Epidemiology and Antimicrobial Stewardship: Use of Administrative and Surveillance Databases

Measuring the Direct Medical Costs of Hospital-Onset Infections Using an Analogy Costing Framework

BACKGROUND

The majority of recent estimates on the direct medical cost attributable to hospital-onset infections (HOIs) has focused on device- or procedure-associated HOIs. The attributable costs of HOIs that are not associated with device use or procedures have not been extensively studied.

OBJECTIVE

We developed simulation models of attributable cost for 16 HOIs and estimated the total direct medical cost, including nondevice-related HOIs in the USA for 2011 and 2015.

DATA AND METHODS

We used total discharge costs associated with HOI-related hospitalization from the National Inpatient Sample and applied an analogy costing methodology to develop simulation models of the costs attributable to HOIs. The mean attributable cost estimate from the simulation analysis was then multiplied by previously published estimates of the number of HOIs for 2011 and 2015 to generate national estimates of direct medical costs.

RESULTS

After adjusting all estimates to 2017 US dollars, attributable cost estimates for select nondevice-related infections attributable cost estimates ranged from $7661 for ear, eye, nose, throat, and mouth (EENTM) infections to $27,709 for cardiovascular system infections in 2011; and from $8394 for EENTM to $26,445 for central nervous system infections in 2016 (based on 2015 incidence data). The national direct medical costs for all HOIs were $14.6 billion in 2011 and $12.1 billion in 2016. Nondevice- and nonprocedure-associated HOIs comprise approximately 26-28% of total HOI costs.

CONCLUSION

Results suggest that nondevice- and nonprocedure-related HOIs result in considerable costs to the healthcare system.

[Data processing of the Brazilian National System of Controlled Product Management for drug utilization research with antimicrobials]

The Brazilian National System of Controlled Product Management (SNGPC) stores data on the dispensing of manufactured and compounded drugs and pharmaceutical inputs, whether controlled and antimicrobial, based on the records of private pharmacies and drugstores. This study assessed the quality of SNGPC data from the dispensing records of manufactured antibiotics, aiming to propose their use in drug utilization researchs (DURs), with a descriptive and retrospective design, analyzing the raw dataset of the SNGPC from January 2014 to December 2020. A total of 475,805,207 drug-dispensing records were collected. On average, antibiotics corresponded to 54.5% of the total records. The quality dimension "unreported" was systematically identified in the variables "active ingredient", "sex", "age" and "ICD-10". The amount of vials/bottles and packages ranged from one to 536 units and the amount of pharmaceutical inputs dispensed, from one to 7,500 units. Results show that 25% of the records exceed an individual therapy and the SNGPC has no critical mechanism to avoid dispensations outside the therapeutic standard for the class. Despite vulnerabilities due to data quality, which can be overcome, the SNGPC allows for the construction of different analytical plans, involving time and other aggregations, in the analysis of community use of antimicrobials and controlled drugs, which makes it a powerful source of data for DUR.

Using Clinical Vignettes and a Modified Expert Delphi Panel to Determine Parameters for Identifying Non-Traumatic Spinal Cord Injury in Health Administrative and Electronic Medical Record Databases

OBJECTIVE

To obtain expert consensus on the parameters and etiologic conditions required to retrospectively identify cases of non-traumatic spinal cord injury (NTSCI) in health administrative and electronic medical record (EMR) databases based on the rating of clinical vignettes.

DESIGN

A modified Delphi process included 2 survey rounds and 1 remote consensus panel. The surveys required the rating of clinical vignettes, developed after chart reviews and expert consultation. Experts who participated in survey rounds were invited to participate in the Delphi Consensus Panel.

SETTING

An international collaboration using an online meeting platform.

PARTICIPANTS

Thirty-one expert physicians and/or clinical researchers in the field of spinal cord injury (SCI).

MAIN OUTCOME MEASURE(S)

Agreement on clinical vignettes as NTSCI. Parameters to classify cases of NTSCI in health administrative and EMR databases.

RESULTS

In health administrative and EMR databases, cauda equina syndromes should be considered SCI and classified as a NTSCI or TSCI based on the mechanism of injury. A traumatic event needs to be listed for injury to be considered TSCI. To be classified as NTSCI, neurologic sufficient impairments (motor, sensory, bowel, and bladder) are required, in addition to an etiology. It is possible to have both a NTSCI and a TSCI, as well as a recovered NTSCI. If information is unavailable or missing in health administrative and EMR databases, the case may be listed as "unclassifiable" depending on the purpose of the research study.

CONCLUSION

The Delphi panel provided guidelines to appropriately classify cases of NTSCI in health administrative and EMR databases.

Divergences between healthcare-associated infection administrative data and active surveillance data in Canada

BACKGROUND

Although Canada has both a national active surveillance system and administrative data for the passive surveillance of healthcare-associated infections (HAI), both have identified strengths and weaknesses in their data collection and reporting. Active and passive surveillance work independently, resulting in results that diverge at times. To understand the divergences between administrative health data and active surveillance data, a scoping review was performed.

METHOD

Medline, Embase and Cumulative Index to Nursing and Allied Health Literature along with grey literature were searched for studies in English and French that evaluated the use of administrative data, alone or in comparison with traditional surveillance, in Canada between 1995 and November 2, 2020. After extracting relevant information from selected articles, a descriptive summary of findings was provided with suggestions for the improvement of surveillance systems to optimize the overall data quality.

RESULTS

Sixteen articles met the inclusion criteria, including twelve observational studies and four systematic reviews. Studies showed that using a single source of administrative data was not accurate for HAI surveillance when compared with traditional active surveillance; however, combining different sources of data or combining administrative with active surveillance data improved accuracy. Electronic surveillance systems can also enhance surveillance by improving the ability to detect potential HAIs.

CONCLUSION

Although active surveillance of HAIs produced the most accurate results and remains the gold-standard, the integration between active and passive surveillance data can be optimized. Administrative data can be used to enhance traditional active surveillance. Future studies are needed to evaluate the feasibility and benefits of potential solutions presented for the use of administrative data for HAI surveillance and reporting in Canada.

A computerized indicator for surgical site infection (SSI) assessment after total hip or total knee replacement: The French ISO-ORTHO indicator

OBJECTIVE

The French National Authority for Health (HAS), with a multidisciplinary working group, developed an indicator 'ISO-ORTHO' to assess surgical site infections (SSIs) after total hip arthroplasty or total knee arthroplasty (THA/TKA) based on the hospital discharge database. We present the ISO-ORTHO indicator designed for SSI automated detection and its relevance for quality improvement and hospital benchmarks.

METHODS

The algorithm is based on a combination of International Statistical Classification of Diseases, Tenth Revision (ICD-10) and procedure codes of the hospital stay. The target population was selected among adult patients who had a THA or TKA between January 1, 2017, and September 30, 2017. Patients at very high risk of SSI and/or with SSI not related to hospital care were excluded. We searched databases for SSIs up to 3 months after THA/TKA. The standardized infection ratio (SIR) of observed versus expected SSIs was calculated (logistic regression) and displayed as funnel plot with 2 and 3 standard deviations (SD) after adjustment for 13 factors known to increase SSI risk.

RESULTS

In total, 790 hospitals and 139,926 THA/TKA stays were assessed; 1,253 SSI were detected in the 473 included hospitals (incidence, 0.9%: 1.0% for THA, 0.80% for TKA). The SSI rate was significantly higher in males (1.2%), in patients with previous osteo-articular infection (4.4%), and those with cancer (2.3%), obesity, or diabetes. Most hospitals (89.9%) were within 2 SD; however, 12 hospitals were classified as outliers at more than +3 SD (1.6% of facilities), and 59 hospitals (7.9%) were outliers between +2 SD and +3 SD.

CONCLUSION

ISO-ORTHO is a relevant indicator for automated surveillance; it can provide hospitals a metric for SSI assessment that may contribute to improving patient outcomes.

Gradual decline in outpatient antibiotic prescriptions in paediatrics: A data warehouse-based 11-year cohort study

AIM

To describe trends in antibiotic (AB) prescriptions in children in primary care over 11 years, using a large data warehouse.

METHODS

A retrospective cohort study assessed outpatient AB prescriptions 2007-2017, using the Massachusetts Health Disparities Repository. The evolution of paediatric outpatient AB prescriptions was assessed using time-series analyses through annual per cent change (APC) for the population and for children with or without comorbid condition.

RESULTS

About 25 000 children were followed in primary care with 31 248 AB prescriptions reported in the data warehouse. The youngest children had more AB prescriptions. Penicillins were prescribed most frequently (46%), then macrolides (28%). One third of children had comorbid conditions, receiving significantly more antibiotics (30.3 vs 21.0 AB/100 child-years, relative risk: 1.43, 95% CI: 1.40, 1.46). Overall AB prescription decreased over the period (APC = -5.34%, 95% CI: -7.10, -3.54), with similar trends for penicillins (APC = -5.49; 95% CI: -8.27, -2.62) and macrolides (APC = -6.46; 95% CI: -8.37, -4.58); antibiotic prescribing declined more in children with comorbid conditions.

CONCLUSION

Outpatient AB prescribing decline was gradual and consistent in paediatrics over the period. Prescription differences persisted between age groups, conditions and indication. The availability of routine care data through data warehouse fosters the surveillance automation, providing inexpensive fast tools to design appropriate antimicrobial stewardship.

Utility of the Current Procedural Terminology Codes for Prophylactic Stabilization for Defining Metastatic Femur Disease

INTRODUCTION

Cohorts from the electronic health record are often defined by the Current Procedural Terminology (CPT) codes. The error prevalence of CPT codes for patients receiving surgical treatment of metastatic disease of the femur has not been investigated, and the predictive value of coding ontologies to identify patients with metastatic disease of the femur has not been adequately discussed.

METHODS

All surgical cases at a single academic tertiary institution from 2010 through 2015 involving prophylactic stabilization of the femur or fixation of a pathologic fracture of the femur were identified using the CPT and International Classification of Disease (ICD) codes. A detailed chart review was conducted to determine the procedure performed as documented in the surgical note and the patient diagnosis as documented in the pathology report, surgical note, and/or office visit notes.

RESULTS

We identified 7 CPT code errors of 171 prophylactic operations (4.1%) and one error of 71 pathologic fracture fixation s(1.4%). Of the 164 prophylactic operations that were coded correctly, 87 (53.0%) had metastatic disease. Of the 70 pathologic operations that were coded correctly, 41 (58%) had metastatic disease.

DISCUSSION

The error prevalence was low in both prophylactic stabilization and pathologic fixation groups (4.1% and 1%, respectively). The structured data (CPT and ICD-9 codes) had a positive predictive value for patients having metastatic disease of 53% for patients in the prophylactic stabilization group and 58% for patients in the pathologic fixation group. The CPT codes and ICD codes assessed in this analysis do provide a useful tool for defining a population in which a moderate proportion of individuals have metastatic disease in the femur at an academic medical center. However, verification is necessary.

Measuring Surgical Site Infection From Linked Administrative Data Following Hip and Knee Replacement

BACKGROUND

Surgical site infections (SSIs) in hip and knee arthroplasty are increasing internationally. Current trends in SSI monitoring use single source administrative databases with data collection points commonly at 30 or 90 days. We hypothesize that SSI rates are being under-reported due to methodological biases.

METHODS

Data from multiple administrative data sets were contrasted and compared to look at the 90-day SSI rates for hip and knee arthroplasty in a single province from 2001 to 2015. SSI rates were calculated over time by year, and the differences in infection rates between single and multiple administrative data sets were calculated as an estimate of under reporting rates of SSIs. Days until diagnosis was measured for those diagnosed with an infection within 1 year.

RESULTS

Combining administrative data sets indicates that hospital-based data underestimate SSI rates by 0.44 (P < .0001) of a percentage point over all years, a clinically significant result given the overall infection rate of 2.2% over the period. Less than 50% of hip and knee arthroplasty was recorded as infected by 30 days and approximately 75% of cases were recorded as infected by 90 days.

CONCLUSION

Single source administrative data sets and short follow-up periods underestimate SSI rates. Administrative data sets should be combined and a minimum follow-up period of 90 days should be used to more accurately track SSI rates in hip and knee arthroplasty.

Using data linkage methodologies to augment healthcare-associated infection surveillance data

Background and objectives:

The landscape of antimicrobial resistance (AMR) surveillance is changing rapidly. The primary objective of this study was to assess the benefit of linking population-based infection prevention and control surveillance data on methicillin-resistant Staphylococcus aureus (MRSA) to hospital discharge abstract data (DAD). We assessed the value of this novel data linkage for the characterization of hospital-acquired (HA) and community-acquired MRSA (CA-MRSA) cases.

Methods:

Incident inpatient MRSA surveillance data for all adults (≥18 years) from 4 acute-care facilities in Calgary, Alberta, between April 1, 2011, and March 31, 2017, were linked to DAD. Personal health number (PHN) and gender were used to identify specific individuals, and specimen collection time-points were used to identify specific hospitalization records. A third common variable on admission date between these databases was used to validate the linkage process. Descriptive statistics were used to characterize HA-MRSA and CA-MRSA cases identified through the linkage process.

Results:

A total of 2,430 surveillance records (94.6%) were successfully linked to the correct hospitalization period. By linking surveillance and administrative data, we were able to identify key differences between patients with HA- and CA-MRSA. These differences are consistent with previously reported findings in the literature. Data linkage to DAD may be a novel tool to enhance and augment the details of base surveillance data.

Conclusion and recommendations:

This is the first Canadian study linking a frontline healthcare-associated infection AMR surveillance database to an administrative population database. This work represents an important methodological step toward complementing traditional AMR surveillance data practices. Data linkage to other data types, such as primary care, emergency, social, and biological data, may be the basis of achieving more precise data focused around AMR.

Impact of clinical decision support on empirical antibiotic prescribing for children with community-acquired pneumonia

AIM

To assess the impact of a computerised clinical decision support system (CDSS) on antibiotic use in hospitalised children with a presumptive diagnosis of uncomplicated community-acquired pneumonia (CAP).

METHODS

Codes associated with lower respiratory tract infection were used to identify cases of presumed uncomplicated CAP requiring admission to a tertiary paediatric hospital. Random sampling of the periods between 1 October 2010 and 30 September 2012 (pre-CDSS) and 1 October 2012 and 30 September 2014 (post-CDSS) determined the sequence of case assessment by two independent investigators. Initial antibiotic therapy, associated CDSS approvals and documented signs of clinical deterioration prior to antibiotic decision-making were recorded.

RESULTS

Statistically significant differences between cases pre- and post-CDSS implementation were minimal. High fever was observed in 57.5% (77/134) cases pre-CDSS and 45.8% (49/107) cases post-CDSS (P = 0.07). Supplemental oxygen was used in 30.6% pre-CDSS and 54.2% post-CDSS cases (P < 0.001). Narrow-spectrum penicillins were prescribed most often, with no statistically significant change post-CDSS implementation (81.3% pre-CDSS, 77.6% post-CDSS, P = 0.47). Macrolides were used consistently throughout the study period (53.7% pre-CDSS, 61.7% post-CDSS; P = 0.21).

CONCLUSION

CDSS implementation did not reduce already low rates of broad-spectrum antibiotic use for uncomplicated CAP.

A Discrete Event Simulation Model of Patient Flow in a General Hospital Incorporating Infection Control Policy for Methicillin-Resistant Staphylococcus Aureus (MRSA) and Vancomycin-Resistant Enterococcus (VRE)

BACKGROUND

Hospitalized patients are assigned to available staffed beds based on patient acuity and services required. In hospitals with double-occupancy rooms, patients must be additionally matched by gender. Patients with methicillin-resistant Staphylococcus aureus (MRSA) or vancomycin-resistant Enterococcus (VRE) must be bedded in single-occupancy rooms or cohorted with other patients with similar MRSA/VRE flags.

METHODS

We developed a discrete event simulation (DES) model of patient flow through an acute care hospital. Patients are matched to beds based on acuity, service, gender, and known MRSA/VRE colonization. Outcomes included time to bed arrival, length of stay, patient-bed acuity mismatches, occupancy, idle beds, acuity-related transfers, rooms with discordant MRSA/VRE colonization, and transmission due to discordant colonization.

RESULTS

Observed outcomes were well-approximated by model-generated outcomes for time-to-bed arrival (6.7 v. 6.2 to 6.5 h) and length of stay (3.3 v. 2.9 to 3.0 days), with overlapping 90% coverage intervals. Patient-bed acuity mismatches, where patient acuity exceeded bed acuity and where patient acuity was lower than bed acuity, ranged from 0.6 to 0.9 and 8.6 to 11.1 mismatches per h, respectively. Values for observed occupancy, total idle beds, and acuity-related transfers compared favorably to model-predicted values (91% v. 86% to 87% occupancy, 15.1 v. 14.3 to 15.7 total idle beds, and 27.2 v. 22.6 to 23.7 transfers). Rooms with discordant colonization status and transmission due to discordance were modeled without an observed value for comparison. One-way and multi-way sensitivity analyses were performed for idle beds and rooms with discordant colonization.

CONCLUSIONS

We developed and validated a DES model of patient flow incorporating MRSA/VRE flags. The model allowed for quantification of the substantial impact of MRSA/VRE flags on hospital efficiency and potentially avoidable nosocomial transmission.