Indirect covariate balance and residual confounding: An applied comparison of propensity score matching and cardinality matching.

PURPOSE: Propensity score matching (PSM) is subject to limitations associated with limited degrees of freedom and covariate overlap. Cardinality matching (CM), an optimization algorithm, overcomes these limitations by matching directly on the marginal distribution of covariates. This study compared the performance of PSM and CM. METHODS: Comparative cohort study of new users of angiotensin-converting enzyme inhibitor (ACEI) and ß-blocker monotherapy identified from a large U.S. administrative claims database. One-to-one matching was conducted through PSM using nearest-neighbor matching (caliper = 0.15) and CM permitting a maximum standardized mean difference (SMD) of 0, 0.01, 0.05, and 0.10 between comparison groups. Matching covariates included 37 patient demographic and clinical characteristics. Observed covariates included patient demographics, and all observed prior conditions, drug exposures, and procedures. Residual confounding was assessed based on the expected absolute systematic error of negative control outcome experiments. PSM and CM were compared in terms of post-match patient retention, matching and observed covariate balance, and residual confounding within a 10%, 1%, 0.25% and 0.125% sample group. RESULTS: The eligible study population included 182 235 (ACEI: 129363; ß-blocker: 56872) patients. CM achieved superior patient retention and matching covariate balance in all analyses. After PSM, 1.6% and 28.2% of matching covariates were imbalanced in the 10% and 0.125% sample groups, respectively. No significant difference in observed covariate balance was observed between matching techniques. CM permitting a maximum SMD <0.05 was associated with improved residual bias as compared to PSM. CONCLUSION: We recommend CM with more stringent balance criteria as an alternative to PSM when matching on a set of clinically relevant covariates.

Performance characteristics of code-based algorithms to identify urinary tract infections in large United States administrative claims databases.

BACKGROUND: In real-world evidence research, reliability of coding in healthcare databases dictates the accuracy of code-based algorithms in identifying conditions such as urinary tract infection (UTI). This study evaluates the performance characteristics of code-based algorithms to identify UTI. METHODS: Retrospective observational study of adults contained within three large U.S. administrative claims databases on or after January 1, 2010. A targeted literature review was performed to inform the development of 10 code-based algorithms to identify UTIs consisting of combinations of diagnosis codes, antibiotic exposure for the treatment of UTIs, and/or ordering of a urinalysis or urine culture. For each database, a probabilistic gold standard was developed using PheValuator. The performance characteristics of each code-based algorithm were assessed compared with the probabilistic gold standard. RESULTS: A total of 2 950 641, 1 831 405, and 2 294 929 patients meeting study criteria were identified in each database. Overall, the code-based algorithm requiring a primary UTI diagnosis code achieved the highest positive predictive values (PPV; >93.8%) but the lowest sensitivities (<12.9%). Algorithms requiring three UTI diagnosis codes achieved similar PPV (>0.899%) and improved sensitivity (<41.6%). Algorithms requiring a single UTI diagnosis code in any position achieved the highest sensitivities (>72.1%) alongside a slight reduction in PPVs (<78.3%). All-time prevalence estimates of UTI ranged from 21.6% to 48.6%. CONCLUSIONS: Based on these findings, we recommend use of algorithms requiring a single UTI diagnosis code, which achieved high sensitivity and PPV. In studies where PPV is critical, we recommend code-based algorithms requiring three UTI diagnosis codes rather than a single primary UTI diagnosis code.

Development and performance characteristics of novel code-based algorithms to identify invasive Escherichia coli disease.

PURPOSE: Evaluation of novel code-based algorithms to identify invasive Escherichia coli disease (IED) among patients in healthcare databases. METHODS: Inpatient visits with microbiological evidence of invasive bacterial disease were extracted from the Optum© electronic health record database between January 1, 2016 and June 30, 2020. Six algorithms, derived from diagnosis and drug exposure codes associated to infectious diseases and Escherichia coli, were developed to identify IED. The performance characteristics of algorithms were assessed using a reference standard derived from microbiology data. RESULTS: Among 97 194 eligible records, 25 310 (26.0%) were classified as IED. Algorithm 1 (diagnosis code for infectious invasive disease due to E. coli) had the highest positive predictive value (PPV; 96.0%) and lowest sensitivity (60.4%). Algorithm 2, which additionally included patients with diagnosis codes for infectious invasive disease due to an unspecified organism, had the highest sensitivity (95.5%) and lowest PPV (27.8%). Algorithm 4, which required patients with a diagnosis code for infectious invasive disease due to unspecified organism to have no diagnosis code for non-E. coli infections, achieved the most balanced performance characteristics (PPV, 93.6%; sensitivity, 78.1%; F1 score, 85.1%). Finally, adding exposure to antibiotics in the treatment of E. coli had limited impact on performance algorithms 5 and 6. CONCLUSION: Algorithm 4, which achieved the most balanced performance characteristics, offers a useful tool to identify patients with IED and assess the burden of IED in healthcare databases.

Adaptation and validation of a coding algorithm for the Charlson Comorbidity Index in administrative claims data using the SNOMED CT standardized vocabulary.

OBJECTIVES: The Charlson comorbidity index (CCI), the most ubiquitous comorbid risk score, predicts one-year mortality among hospitalized patients and provides a single aggregate measure of patient comorbidity. The Quan adaptation of the CCI revised the CCI coding algorithm for applications to administrative claims data using the International Classification of Diseases (ICD). The purpose of the current study is to adapt and validate a coding algorithm for the CCI using the SNOMED CT standardized vocabulary, one of the most commonly used vocabularies for data collection in healthcare databases in the U.S. METHODS: The SNOMED CT coding algorithm for the CCI was adapted through the direct translation of the Quan coding algorithms followed by manual curation by clinical experts. The performance of the SNOMED CT and Quan coding algorithms were compared in the context of a retrospective cohort study of inpatient visits occurring during the calendar years of 2013 and 2018 contained in two U.S. administrative claims databases. Differences in the CCI or frequency of individual comorbid conditions were assessed using standardized mean differences (SMD). Performance in predicting one-year mortality among hospitalized patients was measured based on the c-statistic of logistic regression models. RESULTS: For each database and calendar year combination, no significant differences in the CCI or frequency of individual comorbid conditions were observed between vocabularies (SMD ≤ 0.10). Specifically, the difference in CCI measured using the SNOMED CT vs. Quan coding algorithms was highest in MDCD in 2013 (3.75 vs. 3.6; SMD = 0.03) and lowest in DOD in 2018 (3.93 vs. 3.86; SMD = 0.02). Similarly, as indicated by the c-statistic, there was no evidence of a difference in the performance between coding algorithms in predicting one-year mortality (SNOMED CT vs. Quan coding algorithms, range: 0.725-0.789 vs. 0.723-0.787, respectively). A total of 700 of 5,348 (13.1%) ICD code mappings were inconsistent between coding algorithms. The most common cause of discrepant codes was multiple ICD codes mapping to a SNOMED CT code (n = 560) of which 213 were deemed clinically relevant thereby leading to information gain. CONCLUSION: The current study repurposed an important tool for conducting observational research to use the SNOMED CT standardized vocabulary.

Applied comparison of large-scale propensity score matching and cardinality matching for causal inference in observational research.

BACKGROUND: Cardinality matching (CM), a novel matching technique, finds the largest matched sample meeting prespecified balance criteria thereby overcoming limitations of propensity score matching (PSM) associated with limited covariate overlap, which are especially pronounced in studies with small sample sizes. The current study proposes a framework for large-scale CM (LS-CM); and compares large-scale PSM (LS-PSM) and LS-CM in terms of post-match sample size, covariate balance and residual confounding at progressively smaller sample sizes. METHODS: Evaluation of LS-PSM and LS-CM within a comparative cohort study of new users of angiotensin-converting enzyme inhibitor (ACEI) and thiazide or thiazide-like diuretic monotherapy identified from a U.S. insurance claims database. Candidate covariates included patient demographics, and all observed prior conditions, drug exposures and procedures. Propensity scores were calculated using LASSO regression, and candidate covariates with non-zero beta coefficients in the propensity model were defined as matching covariates for use in LS-CM. One-to-one matching was performed using progressively tighter parameter settings. Covariate balance was assessed using standardized mean differences. Hazard ratios for negative control outcomes perceived as unassociated with treatment (i.e., true hazard ratio of 1) were estimated using unconditional Cox models. Residual confounding was assessed using the expected systematic error of the empirical null distribution of negative control effect estimates compared to the ground truth. To simulate diverse research conditions, analyses were repeated within 10 %, 1 and 0.5 % subsample groups with increasingly limited covariate overlap. RESULTS: A total of 172,117 patients (ACEI: 129,078; thiazide: 43,039) met the study criteria. As compared to LS-PSM, LS-CM was associated with increased sample retention. Although LS-PSM achieved balance across all matching covariates within the full study population, substantial matching covariate imbalance was observed within the 1 and 0.5 % subsample groups. Meanwhile, LS-CM achieved matching covariate balance across all analyses. LS-PSM was associated with better candidate covariate balance within the full study population. Otherwise, both matching techniques achieved comparable candidate covariate balance and expected systematic error. CONCLUSIONS: LS-CM found the largest matched sample meeting prespecified balance criteria while achieving comparable candidate covariate balance and residual confounding. We recommend LS-CM as an alternative to LS-PSM in studies with small sample sizes or limited covariate overlap.

Enhancement and external validation of algorithms using diagnosis codes to identify invasive Escherichia coli disease.

OBJECTIVE: To assess the predictive accuracy of code-based algorithms for identifying invasive Escherichia coli (E. coli) disease (IED) among inpatient encounters in US hospitals. METHODS: The PINC AI Healthcare Database (10/01/2015-03/31/2020) was used to assess the performance of six published code-based algorithms to identify IED cases among inpatient encounters. Case-confirmed IEDs were identified based on microbiological confirmation of E. coli in a normally sterile body site (Group 1) or in urine with signs of sepsis (Group 2). Code-based algorithm performance was assessed overall, and separately for Group 1 and Group 2 based on sensitivity, specificity, positive and negative predictive value (PPV and NPV) and F1 score. The improvement in performance of refinements to the best-performing algorithm was also assessed. RESULTS: Among 2,595,983 encounters, 97,453 (3.8%) were case-confirmed IED (Group 1: 60.9%; Group 2: 39.1%). Across algorithms, specificity and NPV were excellent (>97%) for all but one algorithm, but there was a trade-off between sensitivity and PPV. The algorithm with the most balanced performance characteristics included diagnosis codes for: (1) infectious disease due to E. coli OR (2) sepsis/bacteremia/organ dysfunction combined with unspecified E. coli infection and no other concomitant non-E. coli invasive disease (sensitivity: 56.9%; PPV: 56.4%). Across subgroups, the algorithms achieved lower algorithm performance for Group 2 (sensitivity: 9.9%-61.1%; PPV: 3.8%-16.0%). CONCLUSIONS: This study assessed code-based algorithms to identify IED during inpatient encounters in a large US hospital database. Such algorithms could be useful to identify IED in healthcare databases that lack information on microbiology data.

Comparison of Clinical Outcomes of Gripping Surface Technology Staple Reloads versus Standard Staple Reloads Used with Manual Linear Surgical Staplers.

Purpose: Linear surgical staplers reduce rates of surgical adverse events (bleeding, leaks, infections) compared to manual sutures thereby reducing patient risks, surgeon workflow disruption, and healthcare costs. However, further improvements are needed. Ethicon Gripping Surface Technology (GST) reloads, tested and approved by regulatory authorities in combination with powered staplers, may reduce surgical risks through improved tissue grip. While manual staplers are used in some regions due to affordability, clinical data on GST reloads used with manual staplers are unavailable. This study compared surgical adverse event rates of manual staplers with GST vs standard reloads. These data may be used for label changes in China and Latin America. Patients and Methods: Patients undergoing general or thoracic surgery between October 1, 2015 and August 31, 2021 using ECHELON FLEX™ manual staplers with GST or standard reloads were identified from the Premier Healthcare Database. GST reloads were compared to standard reloads for non-inferiority in bleeding and anastomotic leak for general surgery. Secondary outcomes included sepsis for general surgery, and bleeding and prolonged air leak for thoracic surgery. Covariate balancing was performed using stable balancing weights. Results: The general and thoracic surgery cohorts contained 4571 (GST: 2780; standard: 1791) and 814 (GST: 514; standard: 300) patients, respectively. GST reloads were non-inferior to standard reloads for bleeding and anastomotic leak (adjusted cumulative incidence ratio: 1.02 [90% CI: 0.71, 1.45] and 1.03 [90% CI: 0.72, 1.46], respectively) for general surgery. Compared with standard reloads, GST reloads had a similar incidence of sepsis (2.2% vs 2.1%) for general surgery and lower incidences of bleeding (9.5% vs 16.0%) and prolonged air leak (12.6% vs 14.0%,) for thoracic surgery. Conclusion: GST reloads, compared to standard reloads, used with ECHELON FLEX™ manual staplers had comparable perioperative bleeding and anastomotic leak for general surgery, and lower incidences of safety events for thoracic surgery.

Incidence, predictors, and economic burden of circular anastomotic complications in left-sided colorectal reconstructions involving manual circular staplers.

STUDY AIM: Manual circular staplers are widely used in colorectal surgery; however, limited literature exists examining complications related to circular anastomoses when such devices are used. The present study evaluated the incidence, predictors, and economic burden of circular anastomotic complications in left-sided colorectal reconstructions involving manual circular staplers. MATERIALS AND METHODS: Patients aged ≥18 years who underwent hemicolectomy, low anterior resection, or sigmoidectomy between 1 October 2016 and 31 December 2018 were identified from the Premier Healthcare Database. Manual circular stapler use was identified from hospital administrative billing records. Circular anastomotic complications were defined as a composite endpoint of multiple circular stapler use (proxy for stapler failure) or other circular anastomotic complications (anastomotic leak, bleeding, device/surgical complications, infection, and transfusion). Multivariable analyses were used to model the associations between circular anastomotic complications and total hospital costs, length of stay, operating room time, and 30-, 60-, and 90-day readmission rates. RESULTS: A total of 13,167 patients met the study criteria, of whom 2,984 (22.7%) had circular anastomotic complications. Predictors of circular anastomotic complications included age, procedure type, provider region, and select patient comorbidities. As compared with those who did not, patients who suffered circular anastomotic complications had significantly higher adjusted total hospital costs ($26,924 vs. $18,748; p < .0001), length of stay (7.79 vs. 4.99 days; p < .0001), operating room time (280 vs. 239 min; p < .0001), non-home discharge status (9.63% vs. 4.61%; p < .0001), and all-cause readmission at 30 days (12.2% vs. 8.7%; p < .0001), 60 days (16.0% vs. 11.6%; p < .0001), and 90 days (18.5% vs. 13.4%; p < .0001). LIMITATIONS: The present study is limited by the observational nature and potential for measurement error that is inherent to administrative healthcare databases. CONCLUSIONS: In this analysis of patients undergoing left-sided colorectal reconstructions involving a manual circular stapler, circular anastomotic complications were associated with adverse economic consequences.

Economic and clinical outcomes of spinal fusion surgeries with skin closure through skin staples plus waterproof wound dressings versus 2-octyl cyanoacrylate plus polymer mesh tape.

BACKGROUND CONTEXT: Spinal fusion surgeries are one of the most common types of operations performed during inpatient stays in the United States. Successful wound closure, including watertight closure at the skin layer, plays in important role in patient outcomes. PURPOSE: To compare the economic and clinical outcomes of spinal fusion surgeries using one of two sutureless skin closure techniques: skin staples plus waterproof wound dressings (SSWWD) or 2-octyl cyanoacrylate plus polymer mesh tape (2OPMT). STUDY DESIGN/SETTING: Retrospective study using a multi-hospital database. PATIENT SAMPLE: Patients undergoing inpatient spinal fusion surgery for a spine disorder between October 1, 2015 and March 31, 2019. OUTCOME MEASURES: Total costs from the hospital perspective, operating room time (ORT), hospital length of stay (LOS), non-home discharge, infection/wound complications during the 90-day global period (index surgery through 90 days post-discharge), and 30/60/90-day all-cause readmissions. METHODS: Outcomes were compared between study groups using nearest neighbor propensity score matching with exact matching on 45 primary procedure/diagnosis code groupings and generalized estimating equations to account for hospital-level clustering. This study was sponsored by Ethicon, Inc., a Johnson & Johnson company; the authors are employees or consultants of Johnson & Johnson. RESULTS: A total of 11,991 patients met the study criteria (2OPMT=5,961; SSWWD=6,030), of which 3,602 were included in each post-match study comparison group (total=7,204). As compared with the SSWWD group, the 2OPMT group had statistically significant lower median ORT (240 vs. 270 minutes; p=0.002), mean LOS (3.35 [SD=2.6] vs. 3.86 [SD=2.8] days, p=0.031), risks of non-home discharge status (17.63% vs. 23.10%, p=0.035), overall infections/wound complications (1.37% vs. 2.48%, p=0.015), and surgical site infection (1.11% vs. 2.07%, p=0.023). Differences between the study groups in total hospital costs, all-cause readmissions, and other sub-components of the infection/wound complication composite outcome were statistically insignificant (p>0.05). CONCLUSIONS: In this retrospective observational study of patients undergoing elective inpatient spinal fusion surgery, the use of 2OPMT for skin closure was associated with significantly lower ORT, LOS, non-home discharge, and 90-day rates of infections/wound complications as compared with SSWWD.

Comparison of safety and utilization outcomes in inpatient versus outpatient laparoscopic sleeve gastrectomy: a retrospective, cohort study.

BACKGROUND: Laparoscopic sleeve gastrectomy (LSG) is the most common type of bariatric surgery performed in the United States and may be performed on an outpatient basis. Limited literature exists comparing outcomes of outpatient and inpatient LSG, and study results are conflicting. OBJECTIVES: To compare safety and utilization outcomes of outpatient versus inpatient LSG. SETTINGS: Retrospective, multihospital database study (Optum Pan-Therapeutics Database). METHODS: Patients 18 years of age and older who underwent LSG between October 1, 2015, and December 31, 2018, were identified from the Optum Pan-Therapeutics Database and classified as having undergone outpatient or inpatient surgery. Nearest neighbor propensity score matching and generalized estimating equations accounting for procedural physician-level clustering were used to compare the following outcomes between outpatient and inpatient LSG: all-cause 30-day patient morbidity, hospital readmission, readmission length of stay, bariatric reoperation. and mortality. RESULTS: We identified 22,945 patients (outpatient: 1542; inpatient: 21,403) meeting the study inclusion criteria. After propensity score matching, the inpatient and outpatient groups contained 1542 and 13,903 patients, respectively. Bariatric reoperation (n = 13) and mortality (n = 5) were rare events occurring in <.1% of all cases. Compared with the inpatient group, the outpatient group had a statistically significant lower readmission length of stay (4.63 versus 3.23 days; P = .0057). Otherwise, there was no significant association between procedure setting and 30-day overall morbidity (4.8% versus 5.3%; P = .5775) or hospital readmission (2.6% versus 2.1%; P = .1841). CONCLUSIONS: Safety and utilization outcomes were similar between outpatient and inpatient LSG, and outpatient LSG was associated with shorter hospital readmission length of stay.