|
1
|
Vydiswaran VGV, Strayhorn A, Weber K, Stevens H, Mellinger J, Winder GS, Fernandez AC. Automated-detection of risky alcohol use prior to surgery using natural language processing. ALCOHOL, CLINICAL & EXPERIMENTAL RESEARCH 2024; 48:153-163. [PMID: 38189663 PMCID: PMC10783530 DOI: 10.1111/acer.15222] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/27/2023] [Revised: 10/27/2023] [Accepted: 10/30/2023] [Indexed: 01/09/2024]
Abstract
BACKGROUND Preoperative risky alcohol use is one of the most common surgical risk factors. Accurate and early identification of risky alcohol use could enhance surgical safety. Artificial Intelligence-based approaches, such as natural language processing (NLP), provide an innovative method to identify alcohol-related risks from patients' electronic health records (EHR) before surgery. METHODS Clinical notes (n = 53,629) from pre-operative patients in a tertiary care facility were analyzed for evidence of risky alcohol use and alcohol use disorder. One hundred of these records were reviewed by experts and labeled for comparison. A rule-based NLP model was built, and we assessed the clinical notes for the entire population. Additionally, we assessed each record for the presence or absence of alcohol-related International Classification of Diseases (ICD) diagnosis codes as an additional comparator. RESULTS NLP correctly identified 87% of the human-labeled patients classified with risky alcohol use. In contrast, diagnosis codes alone correctly identified only 29% of these patients. In terms of specificity, NLP correctly identified 84% of the non-risky cohort, while diagnosis codes correctly identified 90% of this cohort. In the analysis of the full dataset, the NLP-based approach identified three times more patients with risky alcohol use than ICD codes. CONCLUSIONS NLP, an artificial intelligence-based approach, efficiently and accurately identifies alcohol-related risk in patients' EHRs. This approach could supplement other alcohol screening tools to identify patients in need of intervention, treatment, and/or postoperative withdrawal prophylaxis. Alcohol-related ICD diagnosis had limited utility relative to NLP, which extracts richer information within clinical notes to classify patients.
Collapse
Affiliation(s)
- VG Vinod Vydiswaran
- Department of Learning Health Sciences, University of Michigan, MI, Ann Arbor, USA
- School of Information, University of Michigan, MI, Ann Arbor, USA
| | - Asher Strayhorn
- Department of Learning Health Sciences, University of Michigan, MI, Ann Arbor, USA
| | - Katherine Weber
- Department of Learning Health Sciences, University of Michigan, MI, Ann Arbor, USA
| | - Haley Stevens
- Department of Psychiatry, University of Michigan, MI, Ann Arbor, USA
| | - Jessica Mellinger
- Department of Psychiatry, University of Michigan, MI, Ann Arbor, USA
- Department of Internal Medicine, University of Michigan, MI, Ann Arbor, USA
| | - G Scott Winder
- Department of Psychiatry, University of Michigan, MI, Ann Arbor, USA
- Department of Surgery, University of Michigan, MI, Ann Arbor, USA
- Department of Neurology, University of Michigan, MI, Ann Arbor, USA
| | - Anne C. Fernandez
- Department of Psychiatry, University of Michigan, MI, Ann Arbor, USA
| |
Collapse
|
|
2
|
Fernandez AC, Waljee JF, Gunaseelan V, Brummett CM, Englesbe MJ, Bicket MC. Prevalence of Unhealthy Substance Use and Associated Characteristics Among Patients Presenting for Surgery. Ann Surg 2023; 278:e740-e744. [PMID: 36538617 PMCID: PMC10205913 DOI: 10.1097/sla.0000000000005767] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/24/2022]
Abstract
OBJECTIVE To assess the prevalence of and identify characteristics associated with unhealthy use before surgery. BACKGROUND Although the escalation in US drug overdose deaths is apparent, the unhealthy use of substances among patients presenting for surgery is unclear. METHODS We conducted a retrospective study of patients presenting for elective surgical procedures between December 2018 and July 2021 and prospectively recruited to 1 of 2 clinical research studies (Michigan Genomics Initiative, Prevention of Iatrogenic Opioid Dependence after Surgery Study). The primary outcome was unhealthy substance use in the past 12 months as determined using the Tobacco, Alcohol, Prescription medication, and other Substance use tool. RESULTS Among 1912 patients, unhealthy substance use was reported in 768 (40.2%). The most common substances with unhealthy use were illicit drugs [385 (20.1%)], followed by alcohol 358 (18.7%)], tobacco [262 (13.7%)], and prescription medications [86 (4.5%)]. Patients reporting unhealthy substance use were significantly more likely to be younger, male [aOR: 1.95 (95% CI, 1.58-2.42)], and have higher scores for pain [aOR: 1.07 (95% CI, 1.02-1.13)], and anxiety [aOR: 1.03 (95% CI, 1.01-1.04)]. Unhealthy substance use was more common among surgical procedures of the forearm, wrist, and hand [aOR: 2.58 (95% CI, 1.01-6.55)]. CONCLUSIONS As many as 2 in 5 patients in the preoperative period may present with unhealthy substance use before elective surgery. Given the potential impact of substance use on surgical outcomes, increased recognition of the problem by screening patients is a critical next step for surgeons and perioperative care teams.
Collapse
Affiliation(s)
- Anne C Fernandez
- Department of Psychiatry, Addiction Center, University of Michigan, Ann Arbor, MI
| | - Jennifer F Waljee
- Department of Surgery, University of Michigan, Ann Arbor, MI
- Opioid Prescribing Engagement Network, Institute for Healthcare Policy and Innovation, University of Michigan School of Public Health, University of Michigan, Ann Arbor, MI
| | - Vidhya Gunaseelan
- Opioid Prescribing Engagement Network, Institute for Healthcare Policy and Innovation, University of Michigan School of Public Health, University of Michigan, Ann Arbor, MI
| | - Chad M Brummett
- Opioid Prescribing Engagement Network, Institute for Healthcare Policy and Innovation, University of Michigan School of Public Health, University of Michigan, Ann Arbor, MI
- Department of Anesthesiology, University of Michigan, Ann Arbor, MI
| | - Michael J Englesbe
- Department of Surgery, University of Michigan, Ann Arbor, MI
- Opioid Prescribing Engagement Network, Institute for Healthcare Policy and Innovation, University of Michigan School of Public Health, University of Michigan, Ann Arbor, MI
| | - Mark C Bicket
- Opioid Prescribing Engagement Network, Institute for Healthcare Policy and Innovation, University of Michigan School of Public Health, University of Michigan, Ann Arbor, MI
- Department of Anesthesiology, University of Michigan, Ann Arbor, MI
| |
Collapse
|
|
3
|
Chapman L, Ren T, Solka J, Bazzi AR, Borsari B, Mello MJ, Fernandez AC. Reducing Alcohol Use Before and After Surgery: Qualitative Study of Two Treatment Approaches. JMIR Perioper Med 2023; 6:e42532. [PMID: 37494103 PMCID: PMC10413235 DOI: 10.2196/42532] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/14/2022] [Revised: 04/14/2023] [Accepted: 05/31/2023] [Indexed: 07/27/2023] Open
Abstract
BACKGROUND High-risk alcohol use is a common preventable risk factor for postoperative complications, admission to intensive care, and longer hospital stays. Short-term abstinence from alcohol use (2 to 4 weeks) prior to surgery is linked to a lower likelihood of postoperative complications. OBJECTIVE The study aimed to explore the acceptability and feasibility of 2 brief counseling approaches to reduce alcohol use in elective surgical patients with high-risk alcohol use in the perioperative period. METHODS A semistructured interview study was conducted with a group of "high responders" (who reduced alcohol use ≥50% postbaseline) and "low responders" (who reduced alcohol use by ≤25% postbaseline) after their completion of a pilot trial to explore the acceptability and perceived impacts on drinking behaviors of the 2 counseling interventions delivered remotely by phone or video call. Interview transcripts were analyzed using thematic analysis. RESULTS In total, 19 participants (10 high responders and 9 low responders) from the parent trial took part in interviews. Three main themes were identified: (1) the intervention content was novel and impactful, (2) the choice of intervention modality enhanced participant engagement in the intervention, and (3) factors external to the interventions also influenced alcohol use. CONCLUSIONS The findings support the acceptability of both high- and low-intensity brief counseling approaches. Elective surgical patients are interested in receiving alcohol-focused education, and further research is needed to test the effectiveness of these interventions in reducing drinking before and after surgery. TRIAL REGISTRATION ClinicalTrials.gov NCT03929562; https://clinicaltrials.gov/ct2/show/NCT03929562.
Collapse
Affiliation(s)
- Lyndsay Chapman
- Department of Psychiatry, University of Michigan Medical School, Ann Arbor, MI, United States
| | - Tom Ren
- College of Medicine, Central Michigan University, Mount Pleasant, MI, United States
| | - Jake Solka
- Department of Psychiatry, University of Michigan Medical School, Ann Arbor, MI, United States
| | - Angela R Bazzi
- Herbert Wertheim School of Public Health, University of California, San Diego, San Diego, CA, United States
- Department of Community Health Sciences, Boston University School of Public Health, Boston, MA, United States
| | - Brian Borsari
- Mental Health Service, San Francisco VA Medical Center, San Francisco, CA, United States
- Department of Psychiatry and Behavioral Sciences, University of California, San Francisco, San Francisco, CA, United States
| | - Michael J Mello
- Department of Emergency Medicine, Rhode Island Hospital, Providence, RI, United States
- Department of Health Services, Policy and Practice, Brown University School of Public Health, Providence, RI, United States
| | - Anne C Fernandez
- Department of Psychiatry, University of Michigan Medical School, Ann Arbor, MI, United States
| |
Collapse
|
|
4
|
Fernandez AC, Aslesen H, Golmirzaie G, Stanton S, Gunaseelan V, Waljee J, Brummett CM, Englesbe M, Bicket MC. Patient Responses to Surgery-relevant Screening for Opioid and Other Risky Substance Use Before Surgery: A Pretest-posttest Study. PAIN MEDICINE (MALDEN, MASS.) 2023; 24:896-899. [PMID: 36478099 PMCID: PMC10321761 DOI: 10.1093/pm/pnac190] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/21/2022] [Revised: 11/02/2022] [Accepted: 12/01/2022] [Indexed: 07/20/2023]
Affiliation(s)
- Anne C Fernandez
- Department of Psychiatry, University of Michigan, Ann Arbor, Michigan, USA
- Institute for Healthcare Policy and Innovation, University of Michigan, Ann Arbor, Michigan, USA
| | - Heidi Aslesen
- Michigan Opioid Prescribing Engagement Network, University of Michigan, Ann Arbor, Michigan, USA
- Department of Surgery, University of Michigan, Ann Arbor, Michigan, USA
| | - Goodardz Golmirzaie
- Department of Anesthesiology, University of Michigan, Ann Arbor, Michigan, USA
| | - Sofea Stanton
- Michigan Opioid Prescribing Engagement Network, University of Michigan, Ann Arbor, Michigan, USA
| | - Vidhya Gunaseelan
- Michigan Opioid Prescribing Engagement Network, University of Michigan, Ann Arbor, Michigan, USA
- Department of Anesthesiology, University of Michigan, Ann Arbor, Michigan, USA
| | - Jennifer Waljee
- Michigan Opioid Prescribing Engagement Network, University of Michigan, Ann Arbor, Michigan, USA
- Department of Surgery, University of Michigan, Ann Arbor, Michigan, USA
| | - Chad M Brummett
- Institute for Healthcare Policy and Innovation, University of Michigan, Ann Arbor, Michigan, USA
- Michigan Opioid Prescribing Engagement Network, University of Michigan, Ann Arbor, Michigan, USA
- Department of Anesthesiology, University of Michigan, Ann Arbor, Michigan, USA
| | - Michael Englesbe
- Institute for Healthcare Policy and Innovation, University of Michigan, Ann Arbor, Michigan, USA
- Michigan Opioid Prescribing Engagement Network, University of Michigan, Ann Arbor, Michigan, USA
- Department of Surgery, University of Michigan, Ann Arbor, Michigan, USA
| | - Mark C Bicket
- Correspondence to: Mark Bicket, MD, PhD, Department of Anesthesiology, University of Michigan School of Medicine, 1500 E Medical Center Drive, Ann Arbor, MI 48109, USA. E-mail:
| |
Collapse
|