Study design

This was a single center, pre-, and post-test quasi-experiment conducted at HFH, a five-hospital health-system with a centralized clinical microbiology laboratory located in metropolitan Detroit, MI, USA. This study received institutional review board approval with a waiver of consent.

The study was performed over two time periods: a pre-intervention (pre-nudge) period from January 1^st, 2022 to January 27^th, 2023, and the corresponding post-intervention (post-nudge) period from March 27^th, 2023 to December 31^st, 2023. Hospitalized patients were included if they were ≥18 years old, had a positive respiratory culture with S. maltophilia, and met the study definition for colonization. Patients were excluded if they met criteria for an active community/hospital/ventilator acquired pneumonia, if they had microbiological culture data from outside institutions, if they died within 72 hours of culture result, received comfort/hospice care at the time of culture, or received active S. maltophilia therapy prior to culture result. Only the index respiratory culture of S. maltophilia was included in patients with multiple encounters.

Intervention

Prior to comment implementation, the nudge comment was proposed, discussed, and voted for approval in person with members of the health system critical care council and antimicrobial stewardship subcommittee to establish stakeholder buy-in. On February 27^th, 2023, an automated microbiology comment on positive S. maltophilia respiratory cultures was implemented that reported: “S. maltophilia is a frequent colonizer of the respiratory tract. Clinical correlation for infection is required. Colonizers do not require antibiotic treatment.” Prior to the intervention, culture results stated the organism alone and without an interpretative comment. The HFH ASP developed a one-page educational handout (Supplement 1) provided to the pharmacy department in the form of a weekly email for the month prior to and after the implementation of the nudge comment from January to March 2023. This education was also shared with providers on the day of implementation through electronic communication. Target audiences included pulmonary and critical care leadership, chief medical residents, infectious diseases providers, and clinical pharmacists.

Key definitions and data

The primary outcome was the proportion of patients who did not receive treatment with S. maltophilia therapy within 72 hours of final culture result in colonized patients. Respiratory cultures included sputum, tracheal aspirate, and/or bronchoalveolar lavage samples. S. maltophilia-active therapy was defined as trimethoprim-sulfamethoxazole, minocycline, levofloxacin, ciprofloxacin, eravacycline, cefiderocol, ceftazidime-avibactam, and aztreonam in response to the culture result.

Colonization was defined as a positive respiratory culture and the absence of clinical criteria for an active pneumonia, including community-acquired (CAP), hospital-acquired (HAP), and ventilator-associated pneumonia (VAP). The criteria were defined using the National Healthcare Safety Network reporting definitions. CAP/HAP were defined as having new or worsening infiltrate on chest radiography with two of the following signs/symptoms: shortness of breath and/or hypoxia, productive cough, or tachypnea. VAP was defined as at least three of the following criteria: new or worsening infiltrates on chest radiography, maximum temperature >38.1°C, purulent and/or increased respiratory secretions, and/or increasing oxygen requirements.

Secondary outcomes included antibiotic-related adverse drug events in those receiving active S. maltophilia therapy for 72 hours or more. Adverse drug events, assessed at 72 hours of therapy, were defined as fluid overload as documented on exam, change in serum creatinine (SCr) by ≥0.3 mg/dL, serum sodium ≤135 mEq/L, serum potassium ≥5 mEq/L, and Clostridioides difficile toxin positive stool sample. A nonequivalent dependent variable of appropriate deep-vein thrombosis (DVT) prophylaxis was used to evaluate the standard of care over the study period and was defined as the use of heparin, low molecular weight heparin, fondaparinux, or sequential compression devices unless on therapeutic anticoagulation or with documented problem of bleeding.

Other key definitions included underlying lung conditions, defined as chronic obstructive pulmonary disease, asthma, lung cancer, pulmonary edema, pulmonary hypertension, bronchiectasis, and a history of airway stents. Ventilatory support was categorized as invasive and non-invasive. Invasive ventilatory support included tracheostomy and endotracheal intubation; non-invasive ventilatory support included nasal cannula, nonrebreather, high-flow nasal cannula, continuous positive airway pressure, and bilevel positive airway pressure. Immunosuppression was defined as a history of transplant, active malignancy, chemotherapy or radiotherapy within the past 90 days, CD4 cell count <200 cells/mm³, or receiving a steroid equivalent of prednisone 20 mg for at least 30 days.

Data collection and statistical analysis

Patient data were acquired for screening using Microsoft SQL Server Management Studio (Microsoft, Redford, WA, USA) based on positive S. maltophilia respiratory culture results. Patients were subsequently screened for inclusion; patents that met inclusion criteria had demographic and outcome data manually collected from the electronic health record using a standardized case report form. Data collected included patient and demographic information, microbiology culture data, antibiotic treatment, and patient outcomes.

This study was designed to detect a difference in S. maltophilia antibiotic treatment in patients with colonization. A sample size of 168 patients was calculated using a two-sided α of 0.05, β of 0.8, and an anticipated effect size of 20% decrease in antibiotic use at three days following comment implementation based on previously published nudge data. ^{Reference McBride, Schulz and Fox12}

Descriptive statistics (proportion [%], median [IQR]) were used to describe patients in the pre- and post-intervention groups. Bivariate analyses were used to compare groups; continuous data were analyzed using Mann–Whitney U test and categorical data were compared using the Pearson X ² or Fisher’s exact tests. To determine variables independently associated with treatment of S. maltophilia colonization, variables associated with the primary outcome (P <0.2) from bivariate analysis were entered into a multivariable logistic regression model using a backward, stepwise approach. Variables included in the model were selected based on clinical rationale, the absence of variable colinearity, and were restricted to an event-to-variable ratio of 10:1; model fit was performed using the Hosmer–Lemeshow goodness-of-fit test. Categorical variables were assessed for colinearity using the Pearson X ² test. For all analyses, P values <0.05 were considered statistically significant. All statistical tests were performed using SPSS Statistics, version 29 (IBM Corp., Armonk, NY, USA).