Association between discharges against medical advice and readmission in patients treated for drug injection-related skin and soft tissue infections

Hazen, Amanda; Pizzicato, Lia N.; Hom, Jeffrey; Johnson, Caroline C.; Viner, Kendra

doi:10.1016/j.jsat.2021.108465

Cited by 8 publications

(3 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…27 Patient-directed discharges are associated with negative outcomes including increased 30-day readmission, cost, and mortality. 28-30 It was not possible to draw conclusions about the relationship between the patient’s decision to leave the hospital and the methadone dose received in this study based on the data we had available and the finding of no significant difference between starting and peak dose of methadone between patient groups. The length of stay for admissions ending in self-directed discharges was 2 days shorter on average than for other admissions.…”

Section: Discussionmentioning

confidence: 91%

Rapid Methadone Induction in a General Hospital Setting: A Retrospective, Observational Analysis

Casey,

Regan,

Gale

et al. 2023

Substance Abuse

View full text Add to dashboard Cite

Background: Outpatient methadone guidelines recommend starting at a low dose and titrating slowly. As fentanyl prevalence and opioid-related mortality increases, there is a need for individuals to rapidly achieve a therapeutic methadone dose. Hospitalization offers a monitored setting for methadone initiation, however dosing practices and safety are not well described. Methods: Retrospective, observational analysis of hospitalized patients with opioid use disorder seen by an inpatient addiction consult team in an academic medical center who were newly initiated on methadone between 2016 and 2022. We calculated initial daily dose, maximum daily dose, timing interval of dose escalation, whether patients were connected to an opioid treatment program (OTP) prior to discharge, whether adverse effects or safety events occurred during the hospitalization, and whether such events were definitely or probably related versus possibly related or unrelated to methadone. Results: One hundred twelve patients were included. The mean initial daily methadone dose administered was 32 mg (range: 10-90 mg). The mean maximum dose reached was 76.8 mg (range 30-165 mg). The mean number of days from initial to peak dose was 5.6 days (range 1-19 days). Overall, 30% of patients experienced a safety event, most commonly sedation. Only 4 safety events were deemed probably or definitely related to methadone. In regression analyses, there was no significant difference between starting doses among patients with or without sedation but there was a relationship between last dose and the likelihood of any possibly related event, with those ending at a dose of 100 mg or higher having a higher likelihood event, compared to those ending at lower doses (47.8% vs 12.4%, P < .001). Seventy-six percent were connected to OTP before discharge. Conclusion: Among hospitalized patients initiating methadone, rapid dose titration was infrequently associated with related safety events and most were connected to community-based methadone treatment before discharge.

show abstract

Section: Discussionmentioning

confidence: 91%

Rapid Methadone Induction in a General Hospital Setting: A Retrospective, Observational Analysis

Casey,

Regan,

Gale

et al. 2023

Substance Abuse

View full text Add to dashboard Cite

show abstract

“…[4][5][6] Barriers to optimization of health for hospitalized PWUD include under-treated pain and substance use disorder which has been linked to discharges occurring before medical optimization and increased rates of re-admission and mortality. [7][8][9] The science of identifying hospitalized PWUD, for epidemiological tracking, resource allocation, and outcomes assessment following implementation strategies aimed to improve outcomes is stymied by the inability to identify these patients. The "gold standard" to identify the hospitalization of a PWUD is human chart review, a highly-regulated, time-intensive process with potential consequences for breaches of confidentiality.…”

Section: Introductionmentioning

confidence: 99%

Natural Language Processing For Identification of Hospitalized People with Substance Use Disorder (Preprint)

Sato,

Grussing,

Patel

et al. 2024

Preprint

View full text Add to dashboard Cite

BACKGROUND People Who Use Drugs (PWUD) are at heightened risk for severe injection-related infections. Current clinical practices and research mostly rely on biomarkers, medication records, ICD codes, and self-screening forms for patients to identify PWUD; the combination of these tools still often fails to identify hospitalized SUD patients, missing crucial intervention opportunities for Serious Injection Related Infections (SIRI). OBJECTIVE This study explores using Natural Language Processing (NLP) to enhance the equitable and comprehensive identification of PWUD in electronic medical records (EMR). METHODS We retrospectively compiled a cohort of hospitalizations that involved PWUD at Tufts Medical Center (2020-2022). Criteria for entering the cohort included ICD10 codes for SUD, positive drug toxicology, SUD treatment prescriptions, and specific NLP keywords. We conducted human review of clinical notes in Electronic Health Records (EHR) to calculate the positive and negative predictive value of two subcohorts: admissions associated with a diagnosis code of substance use disorder only (D-only) and admissions associated with NLP identification of drug use only (N-only). We also conducted a regression analysis to evaluate the impact of race, ethnicity, and Social Vulnerability Index (SVI) on the outcomes of highly documented drug use versus drug use only documented with NLP. RESULTS The study identified 4548 hospitalizations with broad heterogeneity in how people entered the cohort and subcohorts. 288 hospitalizations entered the cohort through NLP presence alone. NLP demonstrated a 54% positive predictive value (PPV), outperforming biomarkers, medication records, and ICD codes in identifying hospitalizations of PWUD. Additionally, NLP significantly enhanced these methods when integrated into the identification algorithm. The study also found that people from racially and ethnically minoritized communities and lower socioeconomic patients were significantly more likely to have SUD not documented in EMRs. CONCLUSIONS NLP proved effective in identifying hospitalizations of PWUD, surpassing traditional methods. While further refinement is needed, NLP shows a promising capability in minimizing healthcare disparities, particularly in infectious disease care for SUD patients, highlighting a crucial step towards more equitable healthcare.

show abstract

“…There are many barriers to care for PWID, including difficulty navigating complex and fragmented systems of care and clinicians that are ill-equipped to treat-or do not prioritize treating substance use disorders (SUDs) [13][14][15]. About 1 in 6 PWID hospitalized with a SIRI leave the hospital early under "patient-direct discharge", also known as "against medical advice" (AMA) [3,14,[16][17][18], leading to incomplete antibiotic treatments and readmission. There is a critical need for novel approaches to help this vulnerable population [19].…”

Section: Introductionmentioning

confidence: 99%

Implementation of an integrated infectious disease and substance use disorder team for injection drug use-associated infections: a qualitative study

Hervera

Seo

Bartholomew

et al. 2023

Addict Sci Clin Pract

View full text Add to dashboard Cite

Background Hospitalizations for severe injection drug use-related infections (SIRIs) are characterized by high costs, frequent patient-directed discharge, and high readmission rates. Beyond the health system impacts, these admissions can be traumatizing to people who inject drugs (PWID), who often receive inadequate treatment for their substance use disorders (SUD). The Jackson SIRI team was developed as an integrated infectious disease/SUD treatment intervention for patients hospitalized at a public safety-net hospital in Miami, Florida in 2020. We conducted a qualitative study to identify patient- and clinician-level perceived implementation barriers and facilitators to the SIRI team intervention. Methods Participants were patients with history of SIRIs (n = 7) and healthcare clinicians (n = 8) at one implementing hospital (Jackson Memorial Hospital). Semi-structured qualitative interviews were performed with a guide created using the Consolidated Framework for Implementation Research (CFIR). Interviews were transcribed, double coded, and categorized by study team members using CFIR constructs. Results Implementation barriers to the SIRI team intervention identified by participants included: (1) complexity of the SIRI team intervention; (2) lack of resources for PWID experiencing homelessness, financial insecurity, and uninsured status; (3) clinician-level stigma and lack of knowledge around addiction and medications for opioid use disorder (OUD); and (4) concerns about underinvestment in the intervention. Implementation facilitators of the intervention included: (1) a non-judgmental, harm reduction-oriented approach; (2) the team’s advocacy for PWID as a means of institutional culture change; (3) provision of close post-hospital follow-up that is often inaccessible for PWID; (4) strong communication with patients and their hospital physicians; and (5) addressing diverse needs such as housing, insurance, and psychological wellbeing. Conclusion Integration of infectious disease and SUD treatment is a promising approach to managing patients with SIRIs. Implementation success depends on institutional buy-in, holistic care beyond the medical domain, and an ethos rooted in harm reduction across multilevel (inner and outer) implementation contexts.

show abstract

Association between discharges against medical advice and readmission in patients treated for drug injection-related skin and soft tissue infections

Cited by 8 publications

References 43 publications

Rapid Methadone Induction in a General Hospital Setting: A Retrospective, Observational Analysis

Rapid Methadone Induction in a General Hospital Setting: A Retrospective, Observational Analysis

Natural Language Processing For Identification of Hospitalized People with Substance Use Disorder (Preprint)

Implementation of an integrated infectious disease and substance use disorder team for injection drug use-associated infections: a qualitative study

Contact Info

Product

Resources

About