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

BACKGROUND: In clinical practice, challenges in identifying patients with uncomplicated urinary tract infections (uUTIs) at risk of antibiotic nonsusceptibility may lead to inappropriate prescribing and contribute to antibiotic resistance. We developed predictive models to quantify risk of nonsusceptibility to 4 commonly prescribed antibiotic classes for uUTI, identify predictors of nonsusceptibility to each class, and construct a corresponding risk categorization framework for nonsusceptibility. METHODS: Eligible females aged ≥12 years with Escherichia coli-caused uUTI were identified from Optum's de-identified Electronic Health Record dataset (1 October 2015-29 February 2020). Four predictive models were developed to predict nonsusceptibility to each antibiotic class and a risk categorization framework was developed to classify patients' isolates as low, moderate, and high risk of nonsusceptibility to each antibiotic class. RESULTS: Predictive models were developed among 87 487 patients. Key predictors of having a nonsusceptible isolate to ≥3 antibiotic classes included number of previous UTI episodes, prior ß-lactam nonsusceptibility, 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 nonsusceptibility to nitrofurantoin, trimethoprim-sulfamethoxazole, ß-lactams, and fluoroquinolones, respectively. Across classes, the proportion of patients categorized as having high-risk isolates was 3- to 12-fold higher among patients with nonsusceptible isolates versus susceptible isolates. CONCLUSIONS: Our predictive models highlight factors that increase risk of nonsusceptibility to antibiotics for uUTIs, while the risk categorization framework contextualizes risk of nonsusceptibility to these treatments. Our findings provide valuable insight to clinicians treating uUTIs and may help inform empiric prescribing in this population.

Unmet needs in uncomplicated urinary tract infection in the United States and Germany: a physician survey.

BACKGROUND: Uncomplicated urinary tract infections (uUTIs/acute cystitis) are among the most common infections in women worldwide. There are differences in uUTI treatment guidelines between countries and understanding the needs of physicians in diverse healthcare systems is important for developing new treatments. We performed a survey of physicians in the United States (US) and Germany to understand their perceptions of, and management approaches to uUTI. METHODS: This was a cross-sectional online survey of physicians in the US and Germany who were actively treating patients with uUTI (≥ 10 patients/month). Physicians were recruited via a specialist panel and the survey was piloted with 2 physicians (1 US, 1 Germany) prior to study commencement. Data were analyzed with descriptive statistics. RESULTS: A total of 300 physicians were surveyed (n = 200 US, n = 100 Germany). Across countries and specialties, physicians estimated 16-43% of patients did not receive complete relief from initial therapy and 33-37% had recurrent infections. Urine culture and susceptibility testing was more common in the US and among urologists. The most commonly selected first-line therapy was trimethoprim-sulfamethoxazole in the US (76%) and fosfomycin in Germany (61%). Ciprofloxacin was the most selected following multiple treatment failures (51% US, 45% Germany). Overall, 35% of US and 45% of German physicians agreed with the statement "I feel there is a good selection of treatment options" and ≥ 50% felt that current treatments provided good symptom relief. More than 90% of physicians included symptom relief amongst their top 3 treatment goals. The overall impact of symptoms on patients' lives was rated "a great deal" by 51% of US and 38% of German physicians, increasing with each treatment failure. Most physicians (> 80%) agreed that antimicrobial resistance (AMR) is serious, but fewer (56% US, 46% Germany) had a high level of confidence in their knowledge of AMR. CONCLUSIONS: Treatment goals for uUTI were similar in the US and Germany, although with nuances to disease management approaches. Physicians recognized that treatment failures have a significant impact on patients' lives and that AMR is a serious problem, though many did not have confidence in their own knowledge of AMR.

Antimicrobial Resistance Trends in Urine Escherichia coli Isolates From Adult and Adolescent Females in the United States From 2011 to 2019: Rising ESBL Strains and Impact on Patient Management.

BACKGROUND: Uncomplicated urinary tract infection (uUTI) is predominantly caused by Escherichia coli, which has increasing antimicrobial resistance (AMR) at the United States (US)-community level. As uUTI is often treated empirically, assessing AMR is challenging, and there are limited contemporary data characterizing period prevalence in the US. METHODS: This was a retrospective study of AMR using Becton, Dickinson and Company Insights Research Database (Franklin Lakes, New Jersey, US) data collected 2011-2019. Thirty-day, nonduplicate Escherichia coli urine isolates from US female outpatients (aged ≥12 years) were included. Isolates were evaluated for nonsusceptibility (intermediate/resistant) to trimethoprim-sulfamethoxazole, fluoroquinolones, or nitrofurantoin, and assessed for extended-spectrum ß-lactamase production (ESBL+) and for ≥2 or ≥3 drug-resistance phenotypes. Generalized estimating equations were used to model AMR trends over time and by US census region. RESULTS: Among 1 513 882 E. coli isolates, the overall prevalence of isolates nonsusceptible to trimethoprim-sulfamethoxazole, fluoroquinolones, and nitrofurantoin was 25.4%, 21.1%, and 3.8%, respectively. Among the isolates, 6.4% were ESBL+, 14.4% had ≥2 drug-resistance phenotypes, and 3.8% had ≥3. Modeling demonstrated a relative average yearly increase of 7.7% (95% confidence interval [CI], 7.2-8.2%) for ESBL+ isolates and 2.7% (95% CI, 2.2-3.2%) for ≥3 drug-phenotypes (both P < .0001). Modeling also demonstrated significant variation in AMR prevalence between US census regions (P < .001). CONCLUSIONS: Period prevalence of AMR among US outpatient urine-isolated E. coli was high, and for multidrug-resistance phenotypes increased during the study period with significant variation between census regions. Knowledge of regional AMR rates helps inform empiric treatment of community-onset uUTI and highlights the AMR burden to physicians.

Prescription opioid abuse and tampering in the United States: results of a self-report survey.

OBJECTIVE: The objective of this study is to estimate the prevalence and impact of prescription opioid abuse and tampering among US adults. METHODS: Participants from the US National Health and Wellness Survey were invited to complete an online survey assessing use, misuse, and abuse of prescription opioid medications in the preceding 3 months. A total of 25,864 adults were screened for self-reported opioid abuse. Prevalence was calculated using weights based on age, gender, race, and education. Respondents reporting abuse or medical use of prescription opioid medication in the prior 3 months (N = 1,242) completed a questionnaire assessing health care resource use and the Work Productivity and Activity Impairment questionnaire. RESULTS: The prevalence of prescription opioid abuse in the 3 months prior to the survey was estimated at 1.31% of US adults, with approximately half (0.67%) tampering during that time. Opioid abuse increased with younger age, male sex, minority race, psychiatric illness, alcoholism, cigarette smoking, being employed, and higher household income. Respondents abusing opioid medications reported greater impairment in work and nonwork activities and more health care use than nonusers. Tampering with opioid medication was associated with greater productivity loss and increased use of health care (all P < 0.05). CONCLUSIONS: Tampering with opioid medications to get high is associated with substantial loss of productivity and health care use. Technologies that reduce users' ability to tamper may reduce the burden of opioid abuse on the health care system.

Systematic literature review and meta-analysis of the efficacy and safety of prescription opioids, including abuse-deterrent formulations, in non-cancer pain management.

OBJECTIVE: This study was conducted to compare safety and efficacy outcomes between opioids formulated with technologies designed to deter or resist tampering (i.e., abuse-deterrent formulations [ADFs]) and non-ADFs for commonly prescribed opioids for treatment of non-cancer pain in adults. METHODS: PubMed and Cochrane Library databases were searched for opioid publications between September 1, 2001 and August 31, 2011, and pivotal clinical trials from all years; abstracts from key pain conferences (2010-2011) were also reviewed. One hundred and ninety-one publications were initially identified, 68 of which met eligibility criteria and were systematically reviewed; a subset of 16 involved a placebo group (13 non-ADFs vs placebo, 3 ADFs vs placebo) and reported both efficacy and safety outcomes, and were included for a meta-analysis. Summary estimates of standardized difference in mean change of pain intensity (DMCPI), standardized difference in sum of pain intensity difference (DSPID), and odds ratios (ORs) of each adverse event (AE) were computed through random-effects estimates for ADFs (and non-ADFs) vs placebo. Indirect treatment comparisons were conducted to compare ADFs and non-ADFs. RESULTS: Summary estimates for standardized DMCPI and for standardized DSPID indicated that ADFs and non-ADFs showed significantly greater efficacy than placebo in reducing pain intensity. Indirect analyses assessing the efficacy outcomes between ADFs and non-ADFs indicated that they were not significantly different (standardized DMCPI [0.39 {95% confidence interval (CI) 0.00-0.76}]; standardized DSPID [-0.22 {95% CI -0.74 to 0.30}]). ADFs and non-ADFs both were associated with higher odds of AEs than placebo. Odds ratios from indirect analyses comparing AEs for ADFs vs non-ADFs were not significant (nausea, 0.87 [0.24-3.12]; vomiting, 1.54 [0.40-5.97]; dizziness/vertigo, 0.61 [0.21-1.76]; headache, 1.42 [0.57-3.53]; somnolence/drowsiness, 0.47 [0.09-2.58]; constipation, 0.64 [0.28-1.49]; pruritus 0.41 [0.05-3.51]). CONCLUSION: ADFs and non-ADFs had comparable efficacy and safety profiles, while both were more efficacious than placebo in reducing pain intensity.

Impact of a pregabalin step therapy policy among medicare advantage beneficiaries.

BACKGROUND: Managed healthcare organizations often utilize formulary management strategies such as prior authorization and step therapy to guide appropriate medication use and to control medication expenditures. The objective of this study was to examine clinical and economic outcomes associated with implementation of a pregabalin step therapy (ST) policy among Medicare Advantage Prescription Drug (MAPD) members. METHODS: Pharmacy and medical claims data from Humana (restricted cohort; ST policy implemented 01/01/2009) and Thomson Reuters MarketScan(®) (unrestricted cohort) were analyzed for MAPD members aged 65 to 89 years receiving treatment for painful diabetic peripheral neuropathy (pDPN), postherpetic neuralgia (PHN) or fibromyalgia (FM). Difference-in-differences (DID) was used to examine year-over-year changes in disease-related and all-cause utilization and costs. Regression analyses examined medication utilization and healthcare expenditures after controlling for between-group compositional differences. RESULTS: We identified 13,911 members in the restricted cohort and matched to members from unrestricted health plans. FM (51.0%) and pDPN (41.8%) were the most common diagnoses. Members in the unrestricted cohort were older and had a greater level of comorbidity than members in the restricted cohort. The restricted cohort demonstrated greater year-over-year decrease in pregabalin utilization and increase in year-over-year gabapentin utilization compared with the unrestricted cohort. ST restriction was associated with an increase in disease-related pharmacy costs and a decrease in total medical costs for the restricted cohort compared with the unrestricted cohort. There was no difference between cohorts in total healthcare cost. CONCLUSION: After controlling for differences in age and comorbidity burden between the groups, implementation of a pregabalin ST restriction was associated with increased disease-related pharmacy costs and decreased total medical costs; however, there was no net difference in total healthcare cost or total pharmacy cost.

Prevalence of diagnosed opioid abuse and its economic burden in the veterans health administration.

OBJECTIVE: Evaluate prevalence and risk-adjusted healthcare costs of diagnosed opioid abuse in the national Veterans Health Administration (VHA). Costs were compared between patients with and without diagnosed opioid abuse. DESIGN: Medical and pharmacy claims analysis of VHA data (10/01/2006 to 09/30/2010) were retrospectively analyzed. Prevalence was calculated as the percent of patients with diagnosed opioid abuse for the entire VHA membership and those with noncancer pain diagnoses, compared between patients prescribed opioids prior to abuse diagnosis and those not prescribed opioids through the VHA system. Healthcare utilization and costs were estimated using matching techniques and generalized linear models to control for clinical and demographic differences between patients with and without diagnosed opioid abuse. Separate comparisons were made (with diagnosed abuse vs. without) for each cohort: patients with/without opioid prescriptions. RESULTS: Five-year diagnosed opioid abuse was 1.11%. Among patients prescribed opioids, 5-year abuse prevalence was 3.04%. Pain patients prescribed opioids had the highest abuse rate at 3.26%. Adjusted annual healthcare costs for diagnosed opioid abuse patients were higher than for those without diagnosed abuse, (prescribed opioids overall healthcare costs: $28,882, with diagnosed abuse vs. $13,605 for those without; not prescribed opioids: $25,197 vs. $6350, P-value< 0.0001; opioid-specific healthcare costs for patients prescribed opioids: $8956 vs. $218; patients not prescribed opioids: $8733 vs. $20). CONCLUSIONS: Diagnosed opioid abuse prevalence is almost 7-fold higher in the veteran's administration population than in commercial health plans and translates to a significant economic burden. Appropriate interventions should be considered to prevent and reduce opioid abuse.

Differences in healthcare utilization and associated costs between patients prescribed vs. nonprescribed opioids during an inpatient or emergency department visit.

OBJECTIVES: Compare healthcare resource utilization (HCRU) and costs between patients prescribed opioids (RxOP) and those who were not (NoRxOP) during an emergency department (ED) or inpatient visit. METHODS: Retrospective cohort analysis was performed (January 2006 to September 2010). Continuously eligible RxOP patients in ED/inpatient settings (January 2007 to September 2009) were included if age was ≥ 12 years by initial prescription date (or random date between first ED/inpatient admission and September 30, 2009 [NoRxOP patients]). Healthcare resource utilization and costs for 12 months after initial prescription were compared. Univariate descriptive analyses were performed for baseline and outcome variables and compared using appropriate tests. Risk adjustment compared HCRU between RxOP and NoRxOP cohorts for the postindex period. RESULTS: Of 27,599 eligible patients, RxOP patients (n = 18,819) were younger, less likely to be male, more likely to reside in southern United States and to have Preferred Provider Organization health plans, and had lower comorbidity index scores, compared with NoRxOP patients (n = 8,780). RxOP patients were less likely to have nonpain-related comorbidities and more frequently diagnosed with pain-related comorbidities. Unmatched and propensity-matched RxOP patients experienced higher HCRU and costs in all subcategories (total, inpatient, outpatient ED, physician, pharmacy, other outpatient settings). Opioid abuse frequency was low in patients with common diagnoses/procedures within 3 months before initial prescription (0.48%). Average time to abuse was < 1 year (201 days). CONCLUSION: Most patients were prescribed opioids initially during ED/inpatient visits and incurred higher HCRU than those not prescribed opioids. Among those with diagnosed opioid abuse after initiating opioids, time to diagnosis was rapid (range: 14 to 260 days) for patients with common diseases and procedures.

The prevalence of diagnosed opioid abuse in commercial and Medicare managed care populations.

PURPOSE: To measure the prevalence of diagnosed opioid abuse and prescription opioid use in a multistate managed care organization. METHODS: This retrospective claims data analysis reviewed the prevalence of diagnosed opioid abuse and the parallel prevalence of prescription opioid use in half-year intervals for commercial and Medicare members enrolled with Humana Inc., from January 1, 2008 to June 30, 2010. Diagnosis of opioid abuse was defined by ≥ 1 medical claim with any of the following ICD-9-CM codes: 304.0 ×, 304.7 ×, 305.5 ×, 965.0 ×, excluding 965.01, and opioid use was defined by ≥ 1 filled prescription for an opioid. The prevalence of opioid abuse was defined by the number of members with an opioid abuse diagnosis, divided by the number of members enrolled in each 6-month interval. RESULTS: The 6-month prevalence of diagnosed opioid abuse increased from 0.84 to 1.15 among commercial and from 3.17 to 6.35 among Medicare members, per 1,000. In contrast, there was no marked increase in prescription opioid use during the same time period (118.0 to 114.8 for commercial members, 240.6 to 256.9 for Medicare members, per 1,000). The prevalence of diagnosed opioid abuse was highest among members younger than 65 years for both genders in commercial (18- to 34-year-olds) and Medicare (35- to 54-year-olds) populations. CONCLUSIONS: Despite a stable rate of prescription opioid use among the observed population, the prevalence of diagnosed opioid abuse is increasing, particularly in the Medicare population.

Cost drivers of prescription opioid abuse in commercial and Medicare populations.

OBJECTIVE: Growth in the number of patients with pain conditions, and the subsequent rise in prescription opioid use for treatment, has been accompanied by an increase in diagnosed opioid abuse. Understanding what drives the incremental healthcare costs of members diagnosed with prescription opioid abuse may assist in developing better screening techniques for abuse. DESIGN: This retrospective analysis examined costs, resource use, and comorbidities 365 days pre- and postdiagnosis in prescription opioid users diagnosed with abuse (cases) vs. their matched nondiagnosed controls. Inclusion criteria for cases were diagnosis of opioid abuse (ICD-9-CM: 304.0x, 304.7x, 305.5x, 965.0x). Multivariate analysis used generalized linear modeling with log-transformed cost as dependent variable, controlling for comorbidities. RESULTS: Final sample sizes were 8,390 cases and 16,780 matched controls. Postindex abuse-related costs were $2,099 for commercial members, $539 for Medicare members aged < 65, and $170 for Medicare members aged ≥ 65. A higher percentage of cases had pain conditions (82.0% vs. 57.4% commercial, 95.9% vs. 87.5% Medicare members aged < 65, 92.9% vs. 82.4% Medicare members aged ≥ 65, P < 0.0001), and a higher numbers of cases had multiple opioid prescribers (3.7 vs. 1.4 commercial, 3.3 vs. 2.2 Medicare < 65, 2.2 vs. 1.6 Medicare ≥ 65, P < 0.0001) than controls preindex. Cases had higher rates of substance abuse and psychiatric diagnoses pre- and postindex (P < 0.0001, all comparisons). Adjusted costs were 28% higher for cases than for controls (P < 0.0001). CONCLUSION: Costs of members diagnosed with prescription opioid abuse are driven by higher pain and psychiatric comorbidities relative to nonabuse controls.

Pain conditions ranked by healthcare costs for members of a national health plan.

Healthcare resource utilization (HCRU) and associated costs specific to pain are a growing concern, as increasing dollar amounts are spent on pain-related conditions. Understanding which pain conditions drive the highest utilization and cost burden to the healthcare system would enable providers and payers to better target conditions to manage pain adequately and efficiently. The current study focused on 36 noncancer chronic and 14 noncancer acute pain conditions and measured the HCRU and costs per member over 365 days. These conditions were ranked by per-member costs and total adjusted healthcare costs to determine the most expensive conditions to a national health plan. The top 5 conditions for the commercial line of business were back pain, osteoarthritis (OA), childbirth, injuries, and non-hip, non-spine fractures (adjusted annual total costs for the commercial members were $119 million, $98 million, $69 million, $61 million, and $48 million, respectively). The top 5 conditions for Medicare members were OA, back pain, hip fractures, injuries, and non-hip, non-spine fractures (adjusted annual costs for the Medicare members were $327 million, $218 million, $117 million, $82 million, and $67 million, respectively). The conditions ranked highest for both per-member and total healthcare costs were hip fractures, childbirth, and non-hip, non-spine fractures. Among these, hip fractures in the Medicare member population had the highest mean cost per member (adjusted per-member cost was $21,058). Further examination specific to how pain is managed in these high-cost conditions will enable providers and payers to develop strategies to improve patient outcomes through appropriate pain management.

Allergies to antibiotics among US women with uncomplicated urinary tract infection.

Uncomplicated urinary tract infections (uUTI) are generally treated empirically with antibiotics. However, antibiotic allergies limit the available oral treatment options for some patients. We assessed the proportion of self-reported antibiotic allergies among US women with uUTI. We performed a cross-sectional survey of US women (≥18 years) with a self-reported uUTI in the previous 60 days and an oral antibiotic prescription. Participants completed an online questionnaire about their most recent uUTI episode. Descriptive self-reported allergy data were stratified into subgroups by whether the participant had recurrent UTI (≥2 uUTIs in the past 6 months or ≥3 uUTIs in past 12 months, including the index episode), the number of different antibiotics given for the index episode (1, 2, ≥3), and whether the treatment was clinically aligned according to Infectious Diseases Society of America uUTI guidelines. Overall, 375 participants completed the questionnaire. The most commonly prescribed antibiotics were trimethoprim-sulfamethoxazole (SXT; 38.7%), ciprofloxacin (22.7%), and nitrofurantoin (18.9%). Most participants (62.7%) received only 1 antibiotic for their uUTI, and most (56.5%) were classified as having a non-recurrent uUTI. No antibiotic allergies were reported for most participants (69.3%), with 24.0% reporting 1 antibiotic allergy and 6.7% reporting ≥2 antibiotic allergies. Allergies to ≥2 antibiotic types were more common among participants classified as having recurrent uUTI, or who used multiple antibiotics to treat their uUTI. The most common allergy was to SXT (15.7%), followed by amoxicillin-clavulanate (8.3%) and ciprofloxacin (5.3%). Similar allergy trends were seen across subgroups, except higher rates of ciprofloxacin allergy were seen in participants given multiple antibiotics. Antibiotic allergies were relatively frequent in this uUTI cohort and the most common allergy was to SXT, which was the most prescribed antibiotic. Allergies to antibiotics reduce the available treatment options for uUTI in some patients.

Treatment Patterns and Adherence to Guidelines for Uncomplicated Urinary Tract Infection in Germany: A Retrospective Cohort Study.

INTRODUCTION: Understanding antibiotic prescribing for uncomplicated urinary tract infection (uUTI) could help to optimize management. However, data on uUTI treatment patterns in the European Union are scarce. We used real-world data to evaluate adherence to antibiotic prescribing guidelines for femalepatients with uUTI in Germany. METHODS: This retrospective cohort study used anonymized German statutory health insurance claims data from the Wissenschaftliches Institut für Gesundheitsökonomie und Gesundheitssystemforschung from January 2013 to December 2019. Patients were female, aged ≥ 12 years, with an index uUTI diagnosis. Patient characteristics and treating physician specialties were examined overall and in sub-cohorts for recommended/non-recommended treatment (based on initial therapy adherence to German uUTI treatment guidelines) and optimal/sub-optimal outcome (based on a prescription of different antibiotics or a urinary tract infection-related episode). RESULTS: Overall, 144,645 uUTI cases in 124,971 patients were analyzed; 51,230 (35.4%) and 93,415 (64.6%) cases were assigned to the recommended/non-recommended treatment sub-cohorts, respectively. Clinically meaningful differences in age and comorbidities were observed between these sub-cohorts. Most cases had an optimal outcome (n = 122,823; 84.9%); of these, a higher proportion received antibiotics that were recommended but not as first-choice versus first-choice therapies as their initial treatment (58.6% vs. 35.3%). In the sub-optimal outcome cohort, 49.1% received antibiotics that were recommended but not as first-choice and 41.1% received first-choice therapies as their initial treatment. Most uUTIs were treated by general practitioners (GPs; 82.3%), followed by gynecologists (13.3%), and urologists (6.8%). Notably, 64.5% of initial therapy prescriptions filled by gynecologists and 32.1% by GPs were first-choice antibiotics. CONCLUSION: A high proportion of prescribed treatments for the initial uUTI episode were not recommended by German uUTI guidelines as first-choice antibiotics. Prescribing adherence varied by physician specialty; specialists showed greater adherence to treatment guidelines versus GPs. This study provides a novel and multi-dimensional picture of uUTI treatment in Germany.

Uncomplicated urinary tract infections are one of the most common infections in women. Doctors around the world use different types of antibiotics to treat people with uncomplicated urinary tract infections. We performed this study to find out more about how doctors in Germany use antibiotics to treat uncomplicated urinary tract infections. We looked at health records from female patients (aged 12+) in Germany. Overall, we examined 144,645 records. We found that around one in ten women use antibiotics to treat an uncomplicated urinary tract infection every year. We then checked to see if the doctors were giving people the right type of antibiotic, the right dose, and the right length of course. To do this, we checked against guidelines that were written by experts in Germany. We found that only one in three patients (35%) received treatment that met the guidelines. We also looked to see what differences there were between different types of doctors. For example, if general practitioners (family doctors) used different antibiotics to specialist doctors in hospitals. Four out of five patients (82%) were treated by general practitioners. We found that specialists were more likely to stick to the guidelines than general practitioners. Finally, we looked at how many patients recovered well after their first course of antibiotics. More than four out of five patients (80%) recovered well. Interestingly, more than half of the patients who had a good recovery (59%) received antibiotics that were not recommended by the guidelines.

Likelihood of Antimicrobial Resistance in Urinary E. coli Isolates Among US Female Patients with Recurrent Versus Non-Recurrent uUTI.

OBJECTIVE: To assess the relative likelihood of antimicrobial resistance (AMR) and multi-drug resistance (MDR) among E. coli isolates from outpatients with recurrent versus non-recurrent uncomplicated urinary tract infection (uUTI). METHODS: In this retrospective observational US cohort study, female outpatients (≥12 years) with uUTI, positive E. coli culture, and treated with ≥1 oral antibiotic within ±5 days of diagnosis were grouped into recurrent and non-recurrent uUTI cohorts per their UTI history (past 12 months). AMR to specific drug classes was evaluated at index. Univariable and multivariable logistic regression models estimated the likelihood of not-susceptible E. coli isolates (AMR/MDR) among patients with recurrent uUTI versus non-recurrent uUTI. RESULTS: Recurrent (N = 12,234) and non-recurrent (N = 68,033) uUTI cohorts had similar distributions (race, ethnicity, region). Patients with recurrent uUTI had a higher prevalence of E. coli resistance to trimethoprim-sulfamethoxazole (21.8% vs 18.7%) and fluoroquinolones (14.2% vs 8.6%), and more isolates were extended-spectrum ß-lactamase-producing (5.9% vs 4.1%) compared to non-recurrent uUTI patients. Patients with recurrent uUTI had a higher likelihood (odds ratio [95% confidence interval]) of any AMR (1.28 [1.22-1.34]), single drug-class resistance (1.18 [1.12-1.24]), and resistance to 2 (1.53 [1.41-1.67]) or ≥3 drug classes (1.70 [1.48-1.96]) (all P <.001). CONCLUSION: This study delineated the likelihood of AMR and MDR among E. coli isolates from patients with recurrent versus non-recurrent uUTI. While some treatment guidelines support empiric therapy in recurrent uUTI, the increased likelihood of resistance among these patients suggests that culture and susceptibility testing should be undertaken to inform recurrent uUTI treatment.

Co-resistance Among Escherichia coli and Klebsiella pneumoniae Urine Isolates from Female Outpatients with Presumed UTI: A Retrospective US Cohort Study.

INTRODUCTION: Urinary tract infections (UTIs) caused by antimicrobial-resistant Enterobacterales are a global health threat. There are limited surveillance data available to characterize the prevalence of antimicrobial resistance among outpatients in the United States (US). METHODS: This retrospective cohort (database) study investigated co-resistance among Escherichia coli and Klebsiella pneumoniae urinary isolates from US female outpatients aged ≥ 12 years with presumed uncomplicated UTI (uUTI), ≥ 3 months of data (2011-2019), and antimicrobial susceptibility testing results. Eligible isolates were the first urinary E. coli or K. pneumoniae isolate per patient collected within 30 days; classified as not susceptible (NS) if antimicrobial susceptibility testing results were intermediate or resistant to each antibiotic tested. Four resistance phenotypes were identified: NS to fluoroquinolones (FQ), trimethoprim/sulfamethoxazole (SXT), nitrofurantoin (NTF), and extended-spectrum ß-lactamase+/third-generation cephalosporin (ESBL+/3GC NS). Co-resistance phenotypes included all possible combinations of resistance to ≥ 2 drug classes. RESULTS: Of 1,513,882 E. coli isolates and 250,719 K. pneumoniae isolates, 856,918 and 187,459 isolates with ≥ 1 resistance phenotype were included in the analysis, respectively. The most common resistance phenotypes were SXT NS for the E. coli isolates (44.8%) and NTF NS for the K. pneumoniae isolates (75.5%), while ESBL+/3GC NS comprised 11.2 and 5.9%, respectively. Among ESBL+/3GC NS E. coli isolates, 72.4, 56.7, and 46.6% were co-resistant to FQ, SXT, and FQ + SXT, respectively. For ESBL+/3GC NS K. pneumoniae isolates, 65.7 and 45.7% were co-resistant to SXT and FQ + SXT. CONCLUSION: Both species exhibited high rates of co-resistance, emphasizing the need to raise awareness of co-resistance and of the unmet need for effective treatment options for uUTI.

Prevalence, regional distribution, and trends of antimicrobial resistance among female outpatients with urine Klebsiella spp. isolates: a multicenter evaluation in the United States between 2011 and 2019.

BACKGROUND: Antimicrobial resistance research in uncomplicated urinary tract infection typically focuses on the main causative pathogen, Escherichia coli; however, little is known about the antimicrobial resistance burden of Klebsiella species, which can also cause uncomplicated urinary tract infections. This retrospective cohort study assessed the prevalence and geographic distribution of antimicrobial resistance among Klebsiella species and antimicrobial resistance trends for K. pneumoniae in the United States (2011-2019). METHODS: K. pneumoniae and K. oxytoca urine isolates (30-day, non-duplicate) among female outpatients (aged ≥ 12 years) with presumed uUTI at 304 centers in the United States were classified by resistance phenotype(s): not susceptible to nitrofurantoin, trimethoprim/sulfamethoxazole, or fluoroquinolone, extended-spectrum ß-lactamase-positive/not susceptible; and multidrug-resistant based on ≥ 2 and ≥ 3 resistance phenotypes. Antimicrobial resistance prevalence by census division and age, as well as antimicrobial resistance trends over time for Klebsiella species, were assessed using generalized estimating equations. RESULTS: 270,552 Klebsiella species isolates were evaluated (250,719 K. pneumoniae; 19,833 K. oxytoca). The most frequent resistance phenotypes in 2019 were nitrofurantoin not susceptible (Klebsiella species: 54.0%; K. pneumoniae: 57.3%; K. oxytoca: 15.1%) and trimethoprim/sulfamethoxazole not susceptible (Klebsiella species: 10.4%; K. pneumoniae: 10.6%; K. oxytoca: 8.6%). Extended-spectrum ß-lactamase-positive/not susceptible prevalence was 5.4%, 5.3%, and 6.8%, respectively. K. pneumoniae resistance phenotype prevalence varied (p < 0.0001) geographically and by age, and increased over time (except for the nitrofurantoin not susceptible phenotype, which was stable and > 50% throughout). CONCLUSIONS: There is a high antimicrobial resistance prevalence and increasing antimicrobial resistance trends among K. pneumoniae isolates from female outpatients in the United States with presumed uncomplicated urinary tract infection. Awareness of K. pneumoniae antimicrobial resistance helps to optimize empiric uncomplicated urinary tract infection treatment.

Activity impairment, health-related quality of life, productivity, and self-reported resource use and associated costs of uncomplicated urinary tract infection among women in the United States.

BACKGROUND: Uncomplicated urinary tract infections (uUTIs) are among the most common infections in the US. Only a few studies, however, describe the impact of uUTIs from the patient perspective. METHODS: A cross-sectional online survey of US women aged ≥18 years was performed assessing uUTI burden regarding activity impairment, health-related quality of life (HRQoL), workplace productivity, healthcare resource use (HRU), and costs. Participants who self-reported a uUTI in the prior 60 days treated with ≥1 oral antibiotic were included. Activity impairment was assessed with the Activity Impairment Assessment scale. HRQoL was assessed using a modified Short Form 36 (SF-36). Direct costs were sum of out-of-pocket expenditures and monetized HRU; indirect costs were calculated using Work Productivity and Activity Impairment (WPAI). Participants were stratified by uUTI recurrence, number of prescribed antibiotics for recent uUTI and therapy appropriateness (1 first-line/1 second-line/multiple antibiotics). Multivariable regression analysis assessed the relationship between stratifications and outcomes while controlling for demographic/clinical characteristics. Propensity score matching was used to compare participants to a matched population from the 2020 National Health and Wellness Survey (NHWS), to control for any impact of COVID-19 on responses. RESULTS: Among 375 participants, impaired activities included sexual intercourse (66.9%), sleep (60.8%) and exercise (52.3%). HRQoL was worse (p<0.0001) than the NHWS population (46.4 vs. 51.3 [physical component score]; 40.0 vs. 46.9 [mental component score]; 0.63 vs. 0.72 [health utility index]). All included WPAI assessments were worse for uUTI cohort vs. NHWS (p<0.0001). Adjusted direct costs were higher for participants receiving 2 vs. 1 antibiotic ($2090 vs. $776; p<0.0001) and receiving multiple antibiotics vs. 1 first-line ($1642 vs. $875; p = 0.002). Recurrent uUTI was associated with increased activity impairment, worse HRQoL, and costs vs. non-recurrent. CONCLUSIONS: uUTIs were associated with increased activity impairment, worse productivity, and reduced HRQoL. Higher costs were found vs. a matched population.

Escherichia coli resistance, treatment patterns and clinical outcomes among females with uUTI in Germany: a retrospective physician-based chart review study.

Real-world data were collected to examine antimicrobial resistance (AMR) prevalence, treatment patterns, and clinical outcomes among female patients with uncomplicated urinary tract infection (uUTI) in Germany. Data were from a retrospective physician-based chart review completed by physicians treating patients with uUTI. Non-pregnant women aged ≥ 12 years, with a uUTI diagnosis, an E. coli-positive urine culture between January 2017-December 2019, and susceptibility test results for ≥ 4 drug classes were eligible. Patients were stratified into three cohorts by drug class susceptibility: susceptible to all (SUS), resistant to one or two drug classes (DR1/2), and resistant to ≥ 3 (MDR) drug classes tested. Among 386 eligible patients [SUS (67.1%); DR1/2 (29.0%); MDR (3.9%)], AMR prevalence was highest for FMIs (18.3%) and lowest for fluoroquinolones (5.2%). The most prescribed drugs were fosfomycin in SUS (44.0%), DR1/2 (41.4%), and fluoroquinolones in MDR (40.0%). Treatment for uUTI failed for 8.8% of patients; failure was more likely in MDR versus SUS [adjusted odds ratio [95% CI] = 4.21 [1.14-1.50]; P = 0.031); incidence of recurrent infection in the 6-months post-index period was higher in DR1/2 versus SUS. These findings may have implications for empiric prescribing, suggesting an unmet need for new treatments.

Older patient age and prior antimicrobial use strongly predict antimicrobial resistance in Escherichia coli isolates recovered from urinary tract infections among female outpatients.

BACKGROUND: Increasing prevalence of antimicrobial resistance (AMR), including multidrug resistance (MDR), among Escherichia coli (E. coli) makes treatment of uncomplicated urinary tract infection (uUTI) difficult. We assessed risk factors for fluoroquinolone (FQ)-not-susceptible (NS) and MDR E. coli among US female outpatients. METHODS: This retrospective cohort study utilized data from female outpatients aged ≥ 12 years with E. coli positive urine culture and oral antimicrobial prescription ± 1 day from index. We assessed patient-level factors within 90 and 91-360 days prior to index as predictors of FQ NS (intermediate/resistant) and MDR (NS to ≥ 1 drug across ≥ 3 classes) E. coli: age, prior oral antimicrobial dispensing, prior AMR phenotypes, prior urine culture, and prior hospitalization. RESULTS: Among 1,858 outpatients with urine-isolated E. coli, 369 (19.9%) had FQ NS and 59 (3.2%) had MDR isolates. After multivariable adjustment, independent risk factors (p < 0.03) for FQ NS E. coli were older age, prior FQ NS isolates, prior dispensing of FQ, and dispensing of any oral antibiotic. Independent risk factors (p < 0.02) for MDR were prior extended-spectrum ß-lactamase-producing isolates (ESBL+), prior FQ dispensing, and prior oral antibiotic dispensing. CONCLUSIONS: In women with uUTI due to E. coli, prior dispensing of FQ or any oral antibiotic within 90 days predicted FQ NS and MDR urine E. coli. Prior urine culture with FQ NS isolates and older age were predictive of FQ NS E. coli. Prior ESBL+ was predictive of MDR E. coli. These data could help identify patients at risk for AMR E. coli and inform empiric prescribing.

Health care costs in patients with painful diabetic peripheral neuropathy prescribed pregabalin or duloxetine.

BACKGROUND: Pregabalin and duloxetine are two FDA-approved medications for the treatment of pain associated with diabetic peripheral neuropathy (pDPN). The objective of this study was to compare changes in all-cause and pDPN-related health care costs in patients with pDPN initiated on pregabalin or duloxetine. METHODS: Patients at least 18 years of age initiating pregabalin or duloxetine between March 1, 2006 and December 31, 2008 were identified from a large U.S. managed care plan database. The date of the first pregabalin or duloxetine prescription was defined as the index date. Patients with claims-based evidence of pDPN and who had continuous enrollment for 6-month pre- and post-index periods were selected for study inclusion. Duloxetine patients with depression or generalized anxiety disorder (GAD) were excluded. All-cause and pDPN-related total health care costs (over 6 month pre-index and post-index periods) were analyzed with difference-in-differences (DiD) models. RESULTS: A total of 2,136 patients (1,785 pregabalin and 351 duloxetine) were identified. No significant differences in gender, age, or pre-index Quan-Charlson comorbidity score were observed between the two cohorts. No significant differences (pregabalin vs. duloxetine) in pre-index to post-index change in mean all-cause health care costs ($1,411 vs. $1,560, P = 0.93) or mean pDPN-related health care costs ($704 vs. -$240, P = 0.22) were found. The DiD models showed no significant difference in all-cause (mean) costs attributable to pregabalin vs. duloxetine therapy between pre-index and post-index periods (mean cost ratio = 0.97, 95% CI: 0.75 to 1.26), but showed that patients receiving pregabalin had a significantly higher increase in pDPN-related costs compared with patients receiving duloxetine (mean cost ratio = 2.35, 95% CI: 1.01 to 5.46). However, the difference (pre- to post-index) in pDPN-related costs attributable to pregabalin vs. duloxetine therapy was nonsignificant (mean cost ratio = 2.30, 95% CI: 0.93 to 5.68) in a sensitivity analysis in which patients with depression and GAD were excluded from both cohorts. CONCLUSION: No differences were noted in all-cause costs attributable to pregabalin or duloxetine. Although patients receiving pregabalin had a significantly greater pre- to post-index increase in pDPN-related health care costs compared with patients receiving duloxetine, this may have been due to an imbalance in patient exclusion criteria between cohorts.