Development of Predictive Models to Inform a Novel Risk Categorization Framework for Antibiotic Resistance in E. coli-Causing Uncomplicated Urinary Tract Infection.

BACKGROUND: In clinical practice, challenges in identifying patients with uncomplicated urinary tract infections (uUTIs) at risk of antibiotic non-susceptibility may lead to inappropriate prescribing and contribute to antibiotic resistance. We developed predictive models to quantify risk of non-susceptibility to four commonly prescribed antibiotic classes for uUTI, identify predictors of non-susceptibility to each class, and construct a corresponding risk categorization framework for non-susceptibility. METHODS: Eligible females aged ≥12 years with E. coli-caused uUTI were identified from Optum's de-identified Electronic Health Record dataset (10/1/2015‒2/29/2020). Four predictive models were developed to predict non-susceptibility to each antibiotic class and a risk categorization framework was developed to classify patients' isolates as low, moderate, and high risk of non-susceptibility to each antibiotic class. RESULTS: Predictive models were developed among 87487 patients. Key predictors of having a non-susceptible isolate to ≥3 antibiotic classes included number of previous UTI episodes, prior ß-lactam non-susceptibility, prior fluoroquinolone treatment, census bureau region, and race. The risk categorization framework classified 8.1%, 14.4%, 17.4%, and 6.3% of patients as having isolates at high risk of non-susceptibility to nitrofurantoin, trimethoprim-sulfamethoxazole, ß-lactams, and fluoroquinolones, respectively. Across classes, the proportion of patients categorized as having high-risk isolates was 3-12 folds higher among patients with non-susceptible isolates versus susceptible isolates. CONCLUSIONS: Our predictive models highlight factors that increase risk of non-susceptibility to antibiotics for uUTIs, while the risk categorization framework contextualizes risk of non-susceptibility to these treatments. Our findings provide valuable insight to clinicians treating uUTIs and may help inform empiric prescribing in this population.

Combining the Power of Artificial Intelligence with the Richness of Healthcare Claims Data: Opportunities and Challenges.

Combinations of healthcare claims data with additional datasets provide large and rich sources of information. The dimensionality and complexity of these combined datasets can be challenging to handle with standard statistical analyses. However, recent developments in artificial intelligence (AI) have led to algorithms and systems that are able to learn and extract complex patterns from such data. AI has already been applied successfully to such combined datasets, with applications such as improving the insurance claim processing pipeline and reducing estimation biases in retrospective studies. Nevertheless, there is still the potential to do much more. The identification of complex patterns within high dimensional datasets may find new predictors for early onset of diseases or lead to a more proactive offering of personalized preventive services. While there are potential risks and challenges associated with the use of AI, these are not insurmountable. As with the introduction of any innovation, it will be necessary to be thoughtful and responsible as we increasingly apply AI methods in healthcare.

The impact of psoriasis on health care costs and patient work loss.

BACKGROUND: There are few comprehensive estimates of the cost of psoriasis in the United States. OBJECTIVE: We sought to quantify the incremental direct medical and indirect work loss costs associated with psoriasis. METHODS: A de-identified claims database from 31 self-insured employers during the period 1998 to 2005 was used. Patients with at least two psoriasis diagnosis claims (N = 12,280) were compared with 3 control subjects (matched on year of birth and sex) without psoriasis. Multivariate two-part regression analysis was used to isolate the incremental cost of psoriasis by controlling for comorbidities and other confounding factors. RESULTS: After multivariate adjustment, the incremental direct and indirect costs of psoriasis were approximately $900 and $600 (P < .001) per patient per year, respectively. LIMITATIONS: The database used in this study does not contain information on patient out-of-pocket costs or loss of productivity costs at work. CONCLUSION: The incremental cost of psoriasis is approximately $1500 per patient per year, with work loss costs accounting for 40% of the cost burden.

The direct and indirect cost burden of atopic dermatitis: an employer-payer perspective.

The goal of this study was to quantify the incremental direct medical and indirect work-loss costs associated with patients diagnosed with atopic dermatitis (AD). A de-identified administrative claims database was used comprising 5.1 million covered beneficiaries from 31 Fortune 500 self-insured employers between 1998 and 2005. Patients with at least two AD diagnosis claims (N = 13,749) were compared with three matched controls (based on yr of birth and gender) with no AD diagnosis (N = 41,247). In addition, a multivariate two-part regression analysis was used to isolate the cost increase attributable to AD by controlling for confounding factors such as age, gender, health plan type, comorbidities, organ transplantation, industry of employer, region, and year. Direct medical and indirect work-loss costs for the AD group were higher on average by $88 and $64 per patient per month, respectively (both P< .001). After multivariate adjustment, the total incremental cost per patient per month for the AD group was $83 (direct: $52, P< .001; indirect: $31, P< .001). Employer-payers experience a significant annual cost burden of $991 per patient attributable to AD. Employee disability and increased sick days account for 38% of the cost burden.

Does use of a PACS increase the number of images per study? A case study in ultrasound.

The purpose of this study was to determine if the use of a picture archiving and communications system (PACS) in ultrasonography increased the number of images acquired per examination. The hypothesis that such an increase does occur was based on anecdotal information; this study sought to test the hypothesis. A random sample of all ultrasound examination types was drawn from the period 1998 through 1999. The ultrasound PACS in use (ACCESS; Kodak Health Information Systems, Dallas, TX) records the number of grayscale and color images saved as part of each study. Each examination in the sample was checked in the ultrasound PACS database,.and the number of grayscale and color images was recorded. The comparison film-based sample was drawn from the period 1994 through 1995. The number of examinations of each type selected was based on the overall statistics of the section; that is, the sample was designed to represent the approximate frequency with which the various examination types are done. For film-based image counts, the jackets were retrieved, and the number of grayscale and color images were counted. The number of images obtained per examination (for most examinations) in ultrasound increased with PACS use. This was more evident with some examination types (eg, pelvis). This result, however, has to be examined for possible systematic biases because ultrasound practice has changed over the time since the authors stopped using film routinely. The use of PACS in ultrasonography was not associated with an increase in the number of images per examination based solely on the use of PACS, with the exception of neonatal head studies. Increases in the number of images per study was otherwise associated with examinations for which changes in protocols resulted in the increased image counts.