Hostname: page-component-6898cfcb55-zkb6p Total loading time: 0 Render date: 2026-11-11T03:55:38.448Z Has data issue: false hasContentIssue false
  • English
  • Français

Antibiotic prescribing trends and outcomes in outpatient low-risk diverticulitis: a retrospective analysis in Veterans Health Administration between 2016 and 2023

Published online by Cambridge University Press:  01 October 2026

Shinya HasegawaOpen the ORCID record for Shinya Hasegawa[Opens in a new window],
Elizabeth A. Chrischilles,
Tomohiro Tanaka and
Hiroyuki SuzukiOpen the ORCID record for Hiroyuki Suzuki[Opens in a new window]
Shinya Hasegawa*
Affiliation:
Division of Infectious Diseases, Department of Internal Medicine, University of Iowa, Iowa City, IA, USACenter for Access and Delivery Research and Evaluation, Iowa City Veterans Affairs Health Care System, Iowa City, IA, USA
Elizabeth A. Chrischilles
Affiliation:
Department of Epidemiology, University of Iowa, Iowa City, IA, USA
Tomohiro Tanaka
Affiliation:
Center for Access and Delivery Research and Evaluation, Iowa City Veterans Affairs Health Care System, Iowa City, IA, USADivision of Gastroenterology and Hepatology, Department of Internal Medicine, University of Iowa, Iowa City, IA, USA
Hiroyuki Suzuki
Affiliation:
Division of Infectious Diseases, Department of Internal Medicine, University of Iowa, Iowa City, IA, USACenter for Access and Delivery Research and Evaluation, Iowa City Veterans Affairs Health Care System, Iowa City, IA, USA
*
Corresponding author: Shinya Hasegawa; Email: shinya-hasegawa@uiowa.edu

Abstract

Background:

Outpatient antibiotic prescribing remains common despite evidence that selective non-antibiotic management is safe for low-risk uncomplicated diverticulitis. Although recent guidelines recommend withholding antibiotics in selected patients, real-world prescribing patterns and associated outcomes remain unclear.

Methods:

We conducted a retrospective cohort study using nationwide Veterans Health Administration data (2016–2023) to identify emergency department visits for a first case of uncomplicated diverticulitis among patients discharged without admission. Low-risk diverticulitis was defined as age <65 years, Charlson Comorbidity Index <2, and white blood cell count <15 × 103/µL. Outcomes included antibiotic prescribing trends, assessed using linear regression, and one-year hospitalization for diverticulitis. Analyses for one-year hospitalization were stratified by age and comorbidity burden to assess effect modification. Subdistribution hazard (Fine-Gray) models estimated adjusted hazard ratios, accounting for the competing risk of death.

Results:

Among 4,249 patients with low-risk diverticulitis (median age, 51 yr; 84.6% male), 4,068 (95.7%) received antibiotics. From 2016 to 2023, the proportion managed without antibiotics remained unchanged (approximately 4%), while β-lactam use increased and fluoroquinolone use declined. Antibiotic use was not significantly associated with a reduced risk of one-year hospitalization (adjusted hazard ratio, 0.99; 95% confidence interval, 0.54–1.84; P = .99). No significant differences were observed across age-comorbidity strata or antibiotic classes, although subgroup analyses may have been underpowered.

Conclusions:

Despite guideline recommendations, antibiotics were prescribed for most low-risk patients with uncomplicated diverticulitis, with no change over eight years. Clinical outcomes did not differ by antibiotic exposure, highlighting an opportunity to reduce unnecessary antibiotic use through outpatient stewardship interventions.

Information

Type
Original Article
Information
Antimicrobial Stewardship & Healthcare Epidemiology , Volume 6 , Issue 1 , 2026, e309
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2026. Published by Cambridge University Press on behalf of The Society for Healthcare Epidemiology of America

Introduction

Outpatient settings account for 85–95% of all antibiotic prescriptions, Reference Duffy, Ritchie, Metcalfe, Van Bakel and Thomas1 and the frequency of these prescriptions has increased over time. Reference Sohn, Pontefract, Dahal and Klepser2 However, approximately 30% of these prescriptions may be unnecessary or inappropriate. Reference Fleming-Dutra, Hersh and Shapiro3,Reference Hersh, King, Shapiro, Hicks and Fleming-Dutra4 Such misuse contributes to antimicrobial resistance, adverse drug events, and increased healthcare costs. Reference Butler, Brown and Durkin5 Although outpatient prescribing is a key target for antibiotic stewardship, most efforts have focused on respiratory, urinary, or skin and soft tissue infections. Reference Madaras-Kelly, Hostler and Townsend6,Reference Westerhof, Dumkow, Hanrahan, McPharlin and Egwuatu7 In contrast, less attention has been given to gastrointestinal infections, which represent an important opportunity for stewardship interventions.

Diverticulitis accounts for 450,000 diagnoses in the United States, with aggregate charges exceeding ten billion US dollars. Reference Peery, Murphy and Anderson8,Reference Francis, Sylla and Abou-Khalil9 Antibiotic treatment had long been the standard of care, but growing evidence suggests that non-antibiotic management is safe and reasonable for selected patients with uncomplicated diverticulitis. Reference Ruiz-Edo and Estrada-Ferrer10–Reference Jaung, Nisbet and Gosselink16 Recent clinical practice guidelines now recommend selective antibiotic use for low-risk patients, such as those with uncomplicated diverticulitis and without significant comorbidities. Reference Francis, Sylla and Abou-Khalil9,Reference Hall, Hardiman and Lee17–Reference Qaseem, Etxeandia-Ikobaltzeta and Lin21

Despite these updates, the majority of patients discharged from the emergency department (ED) with diverticulitis still receive antibiotics. Reference Gottlieb, Wusterbarth, Moyer and Bernard22 However, whether these prescribing patterns apply to low-risk populations remains unclear, and whether the effect of antibiotic treatment on recurrent diverticulitis outcomes differs across age-comorbidity risk subgroups is unknown. Therefore, we evaluated antibiotic prescribing patterns among low-risk patients with uncomplicated diverticulitis and assessed outcomes according to antibiotic exposure, using eight years of nationwide Veterans Health Administration (VHA) System data. Among the broader population of patients with uncomplicated diverticulitis, we further examined whether age-comorbidity risk subgroups modified the effect of antibiotic use on one-year hospitalization for recurrent diverticulitis. This expanded analysis aimed to inform clinicians about which patients across the broader spectrum of uncomplicated diverticulitis may be safely managed without antibiotics.

Methods

Study design and population

We conducted a retrospective cohort study (2016–2023) using the VHA Corporate Data Warehouse to identify ED visits for a first case of diverticulitis. Data were obtained through the Veterans Affairs Informatics and Computing Infrastructure.

We included patients who presented to the ED with an initial diagnosis of uncomplicated diverticulitis, defined by an International Classification of Diseases, 10th Revision (ICD-10) codes for diverticulitis without perforation or abscess (Supplementary Table 1), and were discharged directly from the ED without hospital admission. Patients were excluded if they were admitted to the hospital within 48 hours of the ED visit (on the assumption that admissions within this window were attributable to the index ED visit for diverticulitis), had a prior diagnosis of diverticulitis, had prior colorectal surgery, or had recent exposure (within 14 d prior to the ED visit) to antibiotics commonly used and considered appropriate for diverticulitis (β-lactams, fluoroquinolones, trimethoprim-sulfamethoxazole, and/or metronidazole). Reference Solomkin, Mazuski and Bradley23,Reference Brown, Lopez, Smith and Charles24

Definitions

Age was categorized as younger (<65 yr) or elderly (≥65 yr), as older age is associated with higher diverticulitis-related mortality and atypical presentations, Reference Brown, Lopez, Smith and Charles24,Reference Fugazzola, Ceresoli and Coccolini25 and the randomized trials supporting non-antibiotic management enrolled predominantly younger populations. Reference Balk, Adam and Bhuma26 Comorbidity burden was assessed using the original Charlson Comorbidity Index (CCI), with low comorbidity defined as CCI <2 and high comorbidity defined as CCI ≥2, to identify patients with potential immunocompromise, such as those with malignancy or moderate to severe chronic kidney or liver disease, for whom guidelines consistently recommend antibiotic treatment. Reference Peery, Shaukat and Strate20,Reference Qaseem, Etxeandia-Ikobaltzeta and Lin21,Reference Charlson, Pompei, Ales and MacKenzie27 White blood cell (WBC) count was categorized as <15 × 103/μL or ≥15 × 103/μL based on the American Gastroenterological Association. Reference Peery, Shaukat and Strate20,Reference Qaseem, Etxeandia-Ikobaltzeta and Lin21 Body mass index (BMI) was categorized as underweight (18.5 kg/m2), normal weight (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2), and obese (≥30.0 kg/m2). Reference Seidell and Flegal28

Low-risk diverticulitis was defined as patients meeting all of the following criteria: uncomplicated diverticulitis (as defined above), age <65 years, Charlson Comorbidity Index (CCI) <2, and WBC count <15 × 103 /μL, based on recently updated clinical guidelines for diverticulitis management (Supplementary Table 2). We did not include C-reactive protein (CRP) in our definition, as it was not measured in 95.1% of the low-risk population in our cohort.

Patients were categorized as the antibiotic group if antibiotics commonly used and considered appropriate for diverticulitis (β-lactams, fluoroquinolones, trimethoprim-sulfamethoxazole, and/or metronidazole) were prescribed, and otherwise they were categorized as the no-antibiotic group. Within the antibiotic group, regimens were further categorized as follows: (a) β-lactam group, if amoxicillin/clavulanic acid (monotherapy or with metronidazole) or oral cephalosporins (cefuroxime, cefaclor, cephalexin, cefadroxil, cefdinir) plus metronidazole; (b) fluoroquinolone group, if ciprofloxacin or levofloxacin plus metronidazole, or moxifloxacin (monotherapy or with metronidazole); and (c) other antibiotics group, if all other combinations or regimens, including trimethoprim-sulfamethoxazole. Antibiotics dispensed on or within two days following the ED visit were attributed to the index encounter.

Outcome measures

The outcomes included (i) trends in antibiotic prescriptions for patients with low-risk diverticulitis over the study period, (ii) hospitalization for recurrent diverticulitis within 365 days among patients with low-risk diverticulitis. Additionally, among all patients with uncomplicated diverticulitis, we examined whether age-comorbidity risk subgroups modified the effect of antibiotic use on one-year hospitalization for recurrent diverticulitis.

Statistical analysis

Bivariable analyses were performed using Student’s t-test or Mann-Whitney U-test for continuous variables, and χ 2 tests or Fisher’s exact tests for categorical variables.

For prescribing trends among patients with low-risk diverticulitis, we modeled the proportion of each antibiotic group among all prescriptions using linear regression, with calendar quarter (Q1 2016–Q4 2023) as the independent variable. Analyses were conducted for the proportion of patients receiving each antibiotic class (β-lactam group, fluoroquinolone group, or other antibiotics group) and the no-antibiotic group and the mean duration of a single course of antibiotic therapy. Regression coefficients representing slope estimates were reported to describe changes in prescribing patterns over time. The linearity assumption was evaluated by visual inspection of residuals plotted against fitted values and by sequential ANOVA F-tests comparing linear and quadratic polynomial models.

For the risk of one-year hospitalization among those with low-risk diverticulitis, we compared outcomes between the antibiotic and no-antibiotic groups. Missing values in baseline covariates were addressed using multiple imputation with chained equations, generating twenty imputed datasets. We fit subdistribution hazard (Fine-Gray) models to estimate hazard ratios (HRs) for one-year hospitalization due to recurrent diverticulitis, accounting for the competing risk of death. For the multivariable analysis, WBC count was categorized as normal (4.0–11.9 × 103/μL) or abnormal (<4.0 or ≥12.0 × 103/μL) according to the Sepsis-3 consensus definitions, Reference Singer, Deutschman and Seymour29 as we hypothesized that providers’ antibiotic prescribing behavior might depend on WBC count. Patient demographics, BMI, CCI, and WBC count were adjusted. Results were pooled using Rubin’s rules. As sensitivity analyses, we performed a complete-case analysis and repeated the primary analysis restricted to patients who underwent computed tomography (CT) imaging within 48 hours of the index ED visit.

Among patients with uncomplicated diverticulitis, as defined by ICD-10 codes, we examined whether age-comorbidity risk subgroups modified the effect of antibiotic use on one-year hospitalization for recurrent diverticulitis. The following groups were created based on age and comorbidity: (a) young/low comorbidity (age <65 yr and CCI <2), (b) elderly/low comorbidity (age ≥65 yr and CCI <2), (c) young/high comorbidity (age <65 yr and CCI ≥2), and (d) elderly/high comorbidity (age ≥65 yr and CCI ≥2). We tested for effect modification by including interaction terms between antibiotic use and age-comorbidity risk groups in multivariable Fine-Gray models. The overall significance of these multiplicative interaction terms was evaluated using Wald tests with robust variance estimates, pooled across imputed datasets.

In subgroup analyses, we compared individual antibiotic classes (β-lactam and fluoroquinolone groups) with the no-antibiotic group using the same analytic strategy.

All analyses were performed two-tailed, with alpha <0.05 considered statistically significant. Statistical analyses were conducted using R version 4.5.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Among 23,670 patients who visited the ED for an initial case of diverticulitis, 12,289 were discharged from the ED. Of these, 9,886 were classified as having uncomplicated diverticulitis based on the ICD-10 code (Figure 1). Among them, 4,249 (median age, 51 yr; 84.6% male) met criteria for low-risk diverticulitis (Figure 1). Only 181 (4.3%) were treated without antibiotics. Among the 4,068 patients (95.7%) who received antibiotics, 1,594 (39.2%) were classified into the β-lactam group, 2,114 (52.0%) into the fluoroquinolone group, and 360 (8.8%) into the other antibiotics group. Within the β-lactam, fluoroquinolone, and other antibiotics groups, amoxicillin/clavulanate monotherapy, ciprofloxacin or levofloxacin plus metronidazole, and trimethoprim-sulfamethoxazole plus metronidazole accounted for 1,396 (87.6%), 2,093 (99.0%), and 140 (38.9%) patients, respectively.

Figure 1.

Flow diagram of eligibility for inclusion in the study. Abbreviations: CCI, Charlson Comorbidity Index; CRP, C-reactive protein; ED, emergency department; WBC, white blood cell.

Flowchart illustrating the selection process for a study on diverticulitis patients.

Baseline characteristics of patients with low-risk diverticulitis are summarized in Table 1. As expected based on our low-risk definition (age <65 yr, CCI <2, WBC count <15 × 103 /μL), the cohort excluded patients with most major comorbidities, and the maximum age was 64 years. The mean WBC count was significantly lower in the no-antibiotic group than in the antibiotic group.

Table 1.

Baseline characteristics of patients with low-risk diverticulitis discharged from the emergency department without admission, by whether antibiotics were prescribed

A table comparing baseline characteristics of patients with low-risk diverticulitis by antibiotic prescription status.
Table 1. Long description

A table comparing baseline characteristics of patients with low-risk diverticulitis by antibiotic prescription status. The table has 18 rows and 5 columns. Column headers are Age, Male gender, BMI, Race, Co-morbidities/past medical history. Row labels include Age, Male gender, BMI, Race, and various co-morbidities. The table compares patients who did not receive antibiotics (N=181) with those who received any antibiotics (N=4,068). Notable columns include Age, Male gender, BMI, and Race. The table also lists various co-morbidities and their prevalence in both groups. The P-value column indicates the statistical significance of the differences between the groups.

Data are presented as number (%) or median [interquartile range]. Abbreviations: AIDS, acquired immunodeficiency syndrome; BMI, body mass index; HIV, human immunodeficiency virus.

Outcomes: prescribing trends

From Q1 2016 to Q4 2023, the number of patients presenting with low-risk diverticulitis increased slightly and lineary over time (slope, 0.79 [95% CI: +0.20 to +1.38] patient per quarter; p = .011). The proportion of patients not prescribed antibiotics remained stable throughout the study period (−0.04% [95% CI: −0.11% to +0.02%] per quarter, p = .20) (Figure 2). Prescribing patterns revealed shift in β-lactam use increased steadily (slope, +2.11% [95% CI +1.90% to +2.32%]; p < .001), while fluoroquinolone use declined consistently (slope, −2.01% [95% CI −2.22% to −1.79%]; p < .001); residual diagnostics confirmed linearity for both trends. Among the 97.9% of patients for whom antibiotic duration was available, the median duration was 10 days (interquartile range, 7–10 d) across the study period, and the duration of antibiotic prescriptions declined modestly and linearly over time (slope,−0.028 [95% CI: −0.037 to −0.019] days per quarter; p < .001), corresponding to an approximately 0.9-day reduction per antibiotic course over eight years (Supplementary Figure 1).

Figure 2.

Trends in the proportion of patients prescribed each antibiotic group for low-risk diverticulitis per quarter.

Line graph showing trends in antibiotic group prescriptions over time.

Outcomes: the risk of one-year hospitalization

Missing data included race (444, 10.4%), BMI (408, 9.6%), and WBC count (313, 7.5%).

Within 365 days, 11 patients (6.1%) in the no-antibiotic group and 245 patients (6.0%) in the antibiotic group were hospitalized for diverticulitis. Antibiotic prescription was not significantly associated with a reduced hazard of one-year admission due to diverticulitis (univariable model: HR .99, 95% CI 0.54–1.81, p = .97; multivariable model: HR 0.99, 95% CI 0.54–1.84, p = .99; Table 2). A complete-case sensitivity analysis yielded consistent results (univariable model: HR 0.99, 95% CI 0.54–1.81, p = .98; multivariable model: HR 1.07, 95% CI 0.47–2.44, p = .86). Among patients with low-risk diverticulitis, 3,677 of 4,249 (86.5%) underwent CT imaging. In the sensitivity analysis restricted to these CT-performed cases, results were consistent with the primary analysis (univariable model: HR 1.13, 95% CI 0.50–2.54, p = .77; multivariable model: HR 1.13, 95% CI 0.50–2.60, p = .76). In the multivariable model, Black or African American race was associated with a higher hazard of one-year admission due to diverticulitis compared with White race (adjusted HR 1.42, 95% CI 1.05–1.92, p = .023).

Table 2.

The risk of one-year hospital admission for diverticulitis among low-risk cases comparing no-antibiotic and any antibiotic groups

A table comparing the risk of one-year hospital admission for diverticulitis among low-risk cases between no-antibiotic and any antibiotic groups.
Table 2. Long description

A table comparing the risk of one-year hospital admission for diverticulitis among low-risk cases between no-antibiotic and any antibiotic groups. The table has 18 rows and 4 columns. The columns are labeled ‘Unadjusted analysis’, ‘Hazard ratios (95% CIs)’, ‘p-value’, ‘Adjusted analysis’, ‘Hazard ratios (95% CIs)’, and ‘p-value’. The rows are labeled with different variables such as ‘No-antibiotics (N=181)’, ‘Any antibiotics (N=4,068)’, ‘Age (years)’, ‘Gender, Male’, ‘Race’, ‘White’, ‘Black or African American’, ‘Other race’, ‘BMI, kg/m^2’, ‘Normal (18.5–24.9)’, ‘Underweight (<18.5)’, ‘Overweight (25.0–29.9)’, ‘Obese (≥30.0)’, ‘CCI=1 (reference: 0)’, ‘WBC count’, ‘Normal (4.0–11.9)’, and ‘Abnormal (<4.0 or ≥12.0)’. Each row provides hazard ratios, 95% confidence intervals, and p-values for the respective variables.

Subdistribution hazard ratios from Fine-Gray models accounting for death. Multiple imputation was used for missing values in race, BMI, and WBC count; see Methods. Abbreviations: BMI, body mass index; CCI, Charlson Comorbidity Index; CI, confidence interval; WBC, white blood cell.

When hospitalization risk among all uncomplicated patients was stratified by age-comorbidity burden (young/low comorbidity, N = 4,543; elderly/low comorbidity, N = 1,605; young/high comorbidity, N = 1,605; and elderly/high comorbidity, N = 2,409) to estimate effect modification by risk groups, no subgroup demonstrated a significantly decreased hazard of one-year admission due to diverticulitis associated with prescribing antibiotics (Table 3). Although point estimates suggested some numerical variation across subgroups (HRs ranging from 0.57 to 1.07), the overall test for interaction was not statistically significant (Wald test, p = .67), indicating no strong evidence that antibiotic effects differed by patient age and comorbidity status.

Table 3.

Effect of antibiotic use on one-year hospitalization for recurrent diverticulitis by age-comorbidity subgroups among patients with uncomplicated diverticulitis

A table comparing the effect of antibiotic use on hospitalization for recurrent diverticulitis across different age-comorbidity subgroups.
Table 3. Long description

A table comparing the effect of antibiotic use on one-year hospitalization for recurrent diverticulitis by age-comorbidity subgroups. The table has five columns: Subgroup, No antibiotics (N), Any antibiotics (N), Total patients (N), Events (N), Adjusted HR (95% CIs), and p-value. It includes four subgroups: Young patients with low comorbidity, Elderly patients with low comorbidity, Young patients with high comorbidity, and Elderly patients with high comorbidity. Each row provides the number of patients without antibiotics, with any antibiotics, total patients, events, adjusted hazard ratio with confidence intervals, and p-value for each subgroup. The table shows that no subgroup demonstrated a significantly decreased hazard of one-year admission due to diverticulitis associated with prescribing antibiotics.

* Reference group: no antibiotics. Adjusted for Age, Gender, Race, BMI, CCI, and WBC count. P value for multiplicative interaction (Wald test) =0.67. Abbreviations: BMI, body mass index; CCI, Charlson Comorbidity Index; CI, confidence interval; HR, hazard ratio; WBC, white blood cell.

Subgroup analyses

In subgroup analyses stratified by antibiotic class (β-lactam and fluoroquinolone groups), neither β-lactams nor fluoroquinolones were associated with a significantly altered risk of one-year hospitalization compared with the no-antibiotic group in subdistribution hazard (Fine-Gray) models (Supplementary Table 3).

Discussion

In this U.S. nationwide retrospective cohort study, we found that antibiotics were prescribed for more than 95% of low-risk patients with uncomplicated diverticulitis, despite changes in guideline recommendations, and that the proportion of patients managed without antibiotics remained stable over eight years. Among patients who received antibiotics, prescribing patterns shifted significantly, with increased use of β-lactams and decreased use of fluoroquinolones, with a modest reduction in duration of therapy. Adjusted hazard ratios for one-year hospital admission due to diverticulitis did not differ significantly between the no-antibiotic and antibiotic groups, consistent with prior randomized trials. When hospitalization risk in all uncomplicated patients was stratified by age-comorbidity burden, no obvious effect modification was observed, and no subcategory demonstrated a statistically significant association between antibiotic use and lower hazard of one-year hospitalization, supporting the potential safety of selective non-antibiotic management across diverse patient profiles.

Our findings highlight a persistent gap between updated guideline recommendations on selective antibiotic use for low-risk diverticulitis and real-world prescribing patterns. A prior survey of ED providers, general practitioners, and surgeons identified two main reasons for avoiding non-antibiotic management of uncomplicated diverticulitis: (i) concerns about treatment failure (70%) and (ii) lack of awareness of the recommendations (67%). Reference Benhamed, Blanchard, Drolet and Émond30 Another survey of surgeons showed that only 26% had incorporated selective antibiotic use for low-risk patients into their practice, and 50% disagreed that the updated guidelines were likely to change their practice. [9] These challenges may be compounded by vague or inconsistent definitions of “low-risk” across guidelines. Some guidelines rely solely on CT findings, while others incorporate symptoms or laboratory data (Supplementary Table 2). In addition, recent recommendations for selective imaging to guide the diagnosis of diverticulitis further complicate guideline implementation, Reference Francis, Sylla and Abou-Khalil9 given provider disinclination to prescribe antibiotics without a CT confirmation of uncomplicated diverticulitis. The stable rate of no-antibiotic prescribing in our U.S. multicenter cohort suggests that barriers to adopting no-antibiotic management have persisted in routine clinical practice. Further studies are needed to clarify safe thresholds for no-antibiotic management and to promote guideline-concordant practices among frontline providers.

Our cohort confirmed increased use of β-lactams and decreased use of fluoroquinolones. The reduction in fluoroquinolone use likely reflects the 2016 United States Food and Drug Administration (FDA) safety warnings, which recommended restricting fluoroquinolones due to the risk of serious adverse effects. 31,Reference Tran, Antonelli, Hincapie-Castillo and Winterstein32 Several studies have demonstrated that fluoroquinolones are not superior to β-lactams for management of diverticulitis, Reference Peery, Shaukat and Strate20,Reference Balk, Adam and Bhuma26,Reference Gaber, Kinlaw and Edwards33 consistent with our findings. While prescribing behavior appears to have shifted from fluoroquinolones to β-lactams and overall treatment durations have modestly shortened, the rate of no-antibiotic management remained low throughout the study period. Given that no-antibiotic management was not associated with significantly increased hospitalization for recurrent diverticulitis in our low-risk cohort, prospective stewardship interventions are warranted to safely reduce unnecessary antibiotic prescriptions.

Our results may suggest that selective no-antibiotic management may be safe even for older patients or those with higher comorbidity burdens. Analyses stratified by age and comorbidity demonstrated no clear association between antibiotic prescription and the risk of one-year admission, supporting the potential safety of selective no-antibiotic management across diverse patient profiles. However, the precise thresholds for safely withholding antibiotics remain unclear. In fact, point estimates of adjusted hazard ratios among patients treated with antibiotics were lower in the young/high comorbidity and elderly/low comorbidity groups relative to the elderly/high comorbidity group. This pattern suggests that certain older or more comorbid populations could potentially benefit from antibiotic treatment, although the boundary between those who may benefit and those who may not remains unclear. These findings may also reflect limitations of age or CCI as risk indicators for recurrent diverticulitis or potential misclassification from ICD coding. Further studies are needed to better define safe boundaries for no-antibiotic management across varying risk profiles.

Black or African American patients demonstrated an increased hazard of one-year hospital admission for diverticulitis compared to White patients. This finding aligns with exisiting literature documenting racial disparities in diverticulitis outcomes. A recent systematic review found that the majority of studies reported worse outcomes in Black compared to White patients, including higher readmission and mortality rates. Reference Ueland, Vimalathas, Sweeting, Shroder, Younan and Hawkins34 However, insurance status and socioeconomic factors appear to be stronger predictors than race alone, suggesting these disparities reflect systemic barriers to care rather than biological differences. Reference Ueland, Vimalathas, Sweeting, Shroder, Younan and Hawkins34

This study has several limitations. Residual confounding from unmeasured variables cannot be excluded; providers likely prescribed antibiotics to sicker patients based on factors not captured in administrative data, which would most plausibly bias the estimate toward the null. Additionally, the small no-antibiotic group yielded wide confidence intervals, though the point estimate is consistent with prior randomized trials. Reference Ruiz-Edo and Estrada-Ferrer10–Reference Daniels and de Korte12,Reference Chabok, Hjern, Haapaniemi and Smedh15 Our cohort was predominantly male, potentially limiting generalizability, although prior studies have not demonstrated sex-specific differences in outcomes. Reference Al-Masrouri, Garfinkle and Al-Rashid35 We relied on ICD codes without access to CT imaging data, though a sensitivity analysis restricted to CT-performed cases revealed results. Incorporating CT findings via natural language processing would further improve diagnostic accuracy, as CT remains the diagnostic gold standard. Reference Peery, Shaukat and Strate20,Reference Qaseem, Etxeandia-Ikobaltzeta and Lin21 When stratified by age-comorbidity burden, the non-significant hazard ratios may be attributed to small sample sizes within each group. The number of events per variable in our multivariable model (n = 37) was adequate for covariates with moderate prevalence but may have limited precision in estimating effects of covariates with sparse categories. Reference Austin, Allignol and Fine36 Adherence to prescribed antibiotics after the ED visit was also unknown.

In conclusion, antibiotics were prescribed for most low-risk patients with uncomplicated diverticulitis in the VHA, with increasing use of β-lactams and decreasing use of fluoroquinolones over time. Many antibiotic prescriptions in this population may be unnecessary, as clinical outcomes did not differ significantly between patients treated with or without antibiotics, consistent with prior randomized trials, though the small no-antibiotic group limits the precision of this comparison in our cohort. These findings may also be relevant to older patients and those with higher comorbidity burdens, but further studies are needed to refine definitions of low-risk diverticulitis and to develop strategies that safely reduce unnecessary antibiotic use.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/ash.2026.10884.

Acknowledgments

None.

Financial support

None.

Competing interests

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs of the United States Government. The authors report no conflicts of interest.

References

Duffy, E, Ritchie, S, Metcalfe, S, Van Bakel, B, Thomas, MG. Antibacterials dispensed in the community comprise 85%-95% of total human antibacterial consumption. J Clin Pharm Ther 2018; 43:59–64.10.1111/jcpt.12610CrossRefGoogle Scholar
Sohn, M, Pontefract, B, Dahal, K, Klepser, M. Outpatient antibiotic stewardship during the COVID-19 era: analysis of prescribing trends and guideline compliance. Antimicrob Steward Healthc Epidemiol 2025; 5:e168.10.1017/ash.2025.10081CrossRefGoogle ScholarPubMed
Fleming-Dutra, KE, Hersh, AL, Shapiro, DJ, et al. Prevalence of inappropriate antibiotic prescriptions among US ambulatory care visits, 2010-2011. Jama 2016; 315:1864–1873.10.1001/jama.2016.4151CrossRefGoogle ScholarPubMed
Hersh, AL, King, LM, Shapiro, DJ, Hicks, LA, Fleming-Dutra, KE. Unnecessary antibiotic prescribing in US ambulatory care settings, 2010-2015. Clin Infect Dis 2021; 72:133–137.7.10.1093/cid/ciaa667CrossRefGoogle ScholarPubMed
Butler, AM, Brown, DS, Durkin, MJ, et al. Association of inappropriate outpatient pediatric antibiotic prescriptions with adverse drug events and health care expenditures. JAMA Netw Open 2022; 5:e2214153.10.1001/jamanetworkopen.2022.14153CrossRefGoogle ScholarPubMed
Madaras-Kelly, K, Hostler, C, Townsend, M, et al. Impact of implementation of the core elements of outpatient antibiotic stewardship within veterans health administration emergency departments and primary care clinics on antibiotic prescribing and patient outcomes. Clin Infect Dis 2021; 73:e1126–e1134.10.1093/cid/ciaa1831CrossRefGoogle ScholarPubMed
Westerhof, LR, Dumkow, LE, Hanrahan, TL, McPharlin, SV, Egwuatu, NE. Outcomes of an ambulatory care pharmacist-led antimicrobial stewardship program within a family medicine resident clinic. Infect Control Hosp Epidemiol 2021; 42:715–721.10.1017/ice.2020.1275CrossRefGoogle ScholarPubMed
Peery, AF, Murphy, CC, Anderson, C, et al. Burden and cost of gastrointestinal, liver, and pancreatic diseases in the United States: update 2024. Gastroenterology 2025; 168:1000–1024.10.1053/j.gastro.2024.12.029CrossRefGoogle ScholarPubMed
Francis, NK, Sylla, P, Abou-Khalil, M, et al. EAES and SAGES 2018 consensus conference on acute diverticulitis management: evidence-based recommendations for clinical practice. Surg Endosc 2019; 33:2726–41 10.1007/s00464-019-06882-zCrossRefGoogle ScholarPubMed
Mora-López L, Ruiz-Edo, N, Estrada-Ferrer, O, et al. Efficacy and safety of nonantibiotic outpatient treatment in mild acute diverticulitis (DINAMO-study): a multicentre, randomised open-label, noninferiority trial. Ann Surg 2021; 274:e435–e442.Google Scholar
Isacson, D, Smedh, K, Nikberg, M, Chabok, A. Long-term follow-up of the AVOD randomized trial of antibiotic avoidance in uncomplicated diverticulitis. Br J Surg 2019; 106:1542–1548.10.1002/bjs.11239CrossRefGoogle ScholarPubMed
Daniels, L, Ünlü Ç, de Korte, N, et al. Randomized clinical trial of observational versus antibiotic treatment for a first episode of CT-proven uncomplicated acute diverticulitis. Br J Surg 2017; 104:52–61.10.1002/bjs.10309CrossRefGoogle ScholarPubMed
Garfinkle, R, Salama, E, Amar-Zifkin, A, et al. Observational versus antibiotic therapy for acute uncomplicated diverticulitis: a non-inferiority meta-analysis based on a Delphi consensus. Surgery 2022; 171:328–335.10.1016/j.surg.2021.07.012CrossRefGoogle ScholarPubMed
Dichman, ML, Rosenstock, SJ, Shabanzadeh, DM. Antibiotics for uncomplicated diverticulitis. Cochrane Database Syst Rev 2022; 6:Cd009092.Google ScholarPubMed
Chabok, A, Påhlman L, Hjern, F, Haapaniemi, S, Smedh, K. Randomized clinical trial of antibiotics in acute uncomplicated diverticulitis. Br J Surg 2012; 99:532–539.Google ScholarPubMed
Jaung, R, Nisbet, S, Gosselink, MP, et al. Antibiotics do not reduce length of hospital stay for uncomplicated diverticulitis in a pragmatic double-blind randomized trial. Clin Gastroenterol Hepatol 2021; 19:503–10.e1.10.1016/j.cgh.2020.03.049CrossRefGoogle ScholarPubMed
Hall, J, Hardiman, K, Lee, S, et al. The American society of colon and rectal surgeons clinical practice guidelines for the treatment of left-sided colonic diverticulitis. Dis Colon Rectum 2020; 63:728–747.10.1097/DCR.0000000000001679CrossRefGoogle ScholarPubMed
Schultz, JK, Azhar, N, Binda, GA, et al. European Society of Coloproctology: guidelines for the management of diverticular disease of the colon. Colorectal Dis 2020; 22 Suppl 2:5–28.10.1111/codi.15140CrossRefGoogle Scholar
Sartelli, M, Weber, DG, Kluger, Y, et al. Update of the WSES guidelines for the management of acute colonic diverticulitis in the emergency setting. World J Emerg Surg 2020; 15:32.10.1186/s13017-020-00313-4CrossRefGoogle ScholarPubMed
Peery, AF, Shaukat, A, Strate, LL. AGA clinical practice update on medical management of colonic diverticulitis: Expert Review. Gastroenterology 2021; 160:906–11.e1.10.1053/j.gastro.2020.09.059CrossRefGoogle ScholarPubMed
Qaseem, A, Etxeandia-Ikobaltzeta, I, Lin, JS, et al. Diagnosis and management of acute left-sided colonic diverticulitis: a clinical guideline from the American College of Physicians. Ann Intern Med 2022; 175:399–415.10.7326/M21-2710CrossRefGoogle ScholarPubMed
Gottlieb, M, Wusterbarth, E, Moyer, E, Bernard, K. Diverticulitis evaluation and management among United States emergency departments over an eight-year period. Am J Emerg Med 2024; 86: 83–86.10.1016/j.ajem.2024.10.002CrossRefGoogle ScholarPubMed
Solomkin, JS, Mazuski, JE, Bradley, JS, et al. Diagnosis and management of complicated intra-abdominal infection in adults and children: guidelines by the Surgical Infection Society and the Infectious Diseases Society of America. Clin Infect Dis 2010; 50:133–164.10.1086/649554CrossRefGoogle ScholarPubMed
Brown, RF, Lopez, K, Smith, CB, Charles, A. Diverticulitis: A Review. JAMA 2025. doi:10.1001/jama.2025.10234.Google ScholarPubMed
Fugazzola, P, Ceresoli, M, Coccolini, F, et al. The WSES/SICG/ACOI/SICUT/AcEMC/SIFIPAC guidelines for diagnosis and treatment of acute left colonic diverticulitis in the elderly. World J Emerg Surg 2022; 17:5.10.1186/s13017-022-00408-0CrossRefGoogle ScholarPubMed
Balk, EM, Adam, GP, Bhuma, MR, et al. Diagnostic imaging and medical management of acute left-sided colonic diverticulitis : a systematic review. Ann Intern Med 2022; 175:379–387.10.7326/M21-1645CrossRefGoogle ScholarPubMed
Charlson, ME, Pompei, P, Ales, KL, MacKenzie, CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40:373–383.10.1016/0021-9681(87)90171-8CrossRefGoogle ScholarPubMed
Seidell, JC, Flegal, KM. Assessing obesity: classification and epidemiology. Br Med Bull 1997; 53:238–252.10.1093/oxfordjournals.bmb.a011611CrossRefGoogle ScholarPubMed
Singer, M, Deutschman, CS, Seymour, CW, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). Jama 2016; 315:801–810.10.1001/jama.2016.0287CrossRefGoogle ScholarPubMed
Benhamed, A, Blanchard, PG, Drolet, S, Émond, M. Evaluation of physicians current practices and awareness regarding the treatment of acute uncomplicated diverticulitis: results of a provincial survey. Cjem 2023; 25:968–975.Google Scholar
U.S. Food and Drug Administration. FDA Drug Safety Communication: FDA updates warnings for oral and injectable fluoroquinolone antibiotics due to disabling side effects (2016) Available at: https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-updates-warnings-oral-and-injectable-fluoroquinolone-antibiotics. Accessed March 5, 2026.Google Scholar
Tran, PT, Antonelli, PJ, Hincapie-Castillo, JM, Winterstein, AG. Association of US food and drug administration removal of indications for use of oral quinolones with prescribing trends. JAMA Intern Med 2021; 181:808–816.10.1001/jamainternmed.2021.1154CrossRefGoogle ScholarPubMed
Gaber, CE, Kinlaw, AC, Edwards, JK, et al. Comparative effectiveness and harms of antibiotics for outpatient diverticulitis : two nationwide cohort studies. Ann Intern Med 2021; 174:737–746.10.7326/M20-6315CrossRefGoogle ScholarPubMed
Ueland, TE, Vimalathas, P, Sweeting, RS, Shroder, MM, Younan, SA, Hawkins, AT. Social determinants of health in diverticulitis: a systematic review. Dis Colon Rectum 2024; 67:1515–1526.10.1097/DCR.0000000000003425CrossRefGoogle ScholarPubMed
Al-Masrouri, S, Garfinkle, R, Al-Rashid, F, et al. Readmission for treatment failure after nonoperative management of acute diverticulitis: a nationwide readmissions database analysis. Dis Colon Rectum 2020; 63:217–225.10.1097/DCR.0000000000001542CrossRefGoogle ScholarPubMed
Austin, PC, Allignol, A, Fine, JP. The number of primary events per variable affects estimation of the subdistribution hazard competing risks model. J Clin Epidemiol 2017; 83: 75–84.10.1016/j.jclinepi.2016.11.017CrossRefGoogle ScholarPubMed
Figure 0

Figure 1.Flow diagram of eligibility for inclusion in the study. Abbreviations: CCI, Charlson Comorbidity Index; CRP, C-reactive protein; ED, emergency department; WBC, white blood cell.

Figure 1

Table 1.Baseline characteristics of patients with low-risk diverticulitis discharged from the emergency department without admission, by whether antibiotics were prescribedTable 1 long description.

Figure 2

Figure 2.Trends in the proportion of patients prescribed each antibiotic group for low-risk diverticulitis per quarter.

Figure 3

Table 2.The risk of one-year hospital admission for diverticulitis among low-risk cases comparing no-antibiotic and any antibiotic groupsTable 2 long description.

Figure 4

Table 3.Effect of antibiotic use on one-year hospitalization for recurrent diverticulitis by age-comorbidity subgroups among patients with uncomplicated diverticulitisTable 3 long description.

Supplementary material: File

Hasegawa et al. supplementary material 1

Hasegawa et al. supplementary material
Download Hasegawa et al. supplementary material 1(File)
File 137.1 KB
Supplementary material: File

Hasegawa et al. supplementary material 2

Hasegawa et al. supplementary material
Download Hasegawa et al. supplementary material 2(File)
File 31 KB