scholarly journals Natural Language Processing of Clinical Notes to Identify Mental Illness and Substance Use Among People Living with HIV: Retrospective Cohort Study (Preprint)

2020 ◽  
Author(s):  
Jessica P Ridgway ◽  
Arno Uvin ◽  
Jessica Schmitt ◽  
Tomasz Oliwa ◽  
Ellen Almirol ◽  
...  
Keyword(s):  
Mental Illness ◽  
Substance Use ◽  
Natural Language Processing ◽  
Language Processing ◽  
Predictive Value ◽  
Hiv Care ◽  
People Living With Hiv ◽  
Clinical Notes ◽  
Unstructured Text ◽  
Living With Hiv

BACKGROUND Mental illness and substance use are prevalent among people living with HIV and often lead to poor health outcomes. Electronic medical record (EMR) data are increasingly being utilized for HIV-related clinical research and care, but mental illness and substance use are often underdocumented in structured EMR fields. Natural language processing (NLP) of unstructured text of clinical notes in the EMR may more accurately identify mental illness and substance use among people living with HIV than structured EMR fields alone. OBJECTIVE The aim of this study was to utilize NLP of clinical notes to detect mental illness and substance use among people living with HIV and to determine how often these factors are documented in structured EMR fields. METHODS We collected both structured EMR data (diagnosis codes, social history, Problem List) as well as the unstructured text of clinical HIV care notes for adults living with HIV. We developed NLP algorithms to identify words and phrases associated with mental illness and substance use in the clinical notes. The algorithms were validated based on chart review. We compared numbers of patients with documentation of mental illness or substance use identified by structured EMR fields with those identified by the NLP algorithms. RESULTS The NLP algorithm for detecting mental illness had a positive predictive value (PPV) of 98% and a negative predictive value (NPV) of 98%. The NLP algorithm for detecting substance use had a PPV of 92% and an NPV of 98%. The NLP algorithm for mental illness identified 54.0% (420/778) of patients as having documentation of mental illness in the text of clinical notes. Among the patients with mental illness detected by NLP, 58.6% (246/420) had documentation of mental illness in at least one structured EMR field. Sixty-three patients had documentation of mental illness in structured EMR fields that was not detected by NLP of clinical notes. The NLP algorithm for substance use detected substance use in the text of clinical notes in 18.1% (141/778) of patients. Among patients with substance use detected by NLP, 73.8% (104/141) had documentation of substance use in at least one structured EMR field. Seventy-six patients had documentation of substance use in structured EMR fields that was not detected by NLP of clinical notes. CONCLUSIONS Among patients in an urban HIV care clinic, NLP of clinical notes identified high rates of mental illness and substance use that were often not documented in structured EMR fields. This finding has important implications for epidemiologic research and clinical care for people living with HIV.

scholarly journals Natural Language Processing of Clinical Notes to Identify Mental Illness and Substance Use Among People Living with HIV: Retrospective Cohort Study

10.2196/23456 ◽  
2021 ◽  
Vol 9 (3) ◽  
pp. e23456
Author(s):  
Jessica P Ridgway ◽  
Arno Uvin ◽  
Jessica Schmitt ◽  
Tomasz Oliwa ◽  
Ellen Almirol ◽  
...  
Keyword(s):  
Mental Illness ◽  
Substance Use ◽  
Natural Language Processing ◽  
Language Processing ◽  
Predictive Value ◽  
Hiv Care ◽  
People Living With Hiv ◽  
Clinical Notes ◽  
Unstructured Text ◽  
Living With Hiv

Background Mental illness and substance use are prevalent among people living with HIV and often lead to poor health outcomes. Electronic medical record (EMR) data are increasingly being utilized for HIV-related clinical research and care, but mental illness and substance use are often underdocumented in structured EMR fields. Natural language processing (NLP) of unstructured text of clinical notes in the EMR may more accurately identify mental illness and substance use among people living with HIV than structured EMR fields alone. Objective The aim of this study was to utilize NLP of clinical notes to detect mental illness and substance use among people living with HIV and to determine how often these factors are documented in structured EMR fields. Methods We collected both structured EMR data (diagnosis codes, social history, Problem List) as well as the unstructured text of clinical HIV care notes for adults living with HIV. We developed NLP algorithms to identify words and phrases associated with mental illness and substance use in the clinical notes. The algorithms were validated based on chart review. We compared numbers of patients with documentation of mental illness or substance use identified by structured EMR fields with those identified by the NLP algorithms. Results The NLP algorithm for detecting mental illness had a positive predictive value (PPV) of 98% and a negative predictive value (NPV) of 98%. The NLP algorithm for detecting substance use had a PPV of 92% and an NPV of 98%. The NLP algorithm for mental illness identified 54.0% (420/778) of patients as having documentation of mental illness in the text of clinical notes. Among the patients with mental illness detected by NLP, 58.6% (246/420) had documentation of mental illness in at least one structured EMR field. Sixty-three patients had documentation of mental illness in structured EMR fields that was not detected by NLP of clinical notes. The NLP algorithm for substance use detected substance use in the text of clinical notes in 18.1% (141/778) of patients. Among patients with substance use detected by NLP, 73.8% (104/141) had documentation of substance use in at least one structured EMR field. Seventy-six patients had documentation of substance use in structured EMR fields that was not detected by NLP of clinical notes. Conclusions Among patients in an urban HIV care clinic, NLP of clinical notes identified high rates of mental illness and substance use that were often not documented in structured EMR fields. This finding has important implications for epidemiologic research and clinical care for people living with HIV.

The Validity of the Substance Abuse and Mental Illness Symptom Screener (SAMISS) in People Living with HIV/AIDS in Primary HIV Care in Cape Town, South Africa

2014 ◽  
Vol 18 (6) ◽  
pp. 1133-1141 ◽  
Author(s):  
Erica Breuer ◽  
Kevin Stoloff ◽  
Landon Myer ◽  
Soraya Seedat ◽  
Dan J. Stein ◽  
...  
Keyword(s):  
South Africa ◽  
Substance Abuse ◽  
Mental Illness ◽  
Cape Town ◽  
Hiv Care ◽  
People Living With Hiv ◽  
Living With Hiv ◽  
Hiv Aids

Substance Use and Motivational Interviewing

2019 ◽  
pp. 105-114 ◽  
Author(s):  
Liu yi Lin ◽  
Linda R. Frank ◽  
Antoine Douaihy
Keyword(s):  
Substance Use ◽  
Motivational Interviewing ◽  
Substance Use Disorder ◽  
Integrated Approach ◽  
Hiv Care ◽  
People Living With Hiv ◽  
Co Morbidity ◽  
Important Health ◽  
Living With Hiv ◽  
The Impact

People living with HIV (PLWH) who use drugs and alcohol are particularly likely to experience gaps across the HIV care continuum. People with co-occurring HIV and a substance use disorder face significant challenges in treatment. Substance use is well-known to be linked to important health behaviors and outcomes including adherence to antiretroviral and treatment, immunosuppression, and sexual risk behaviors. This chapter provides a review of the impact of substance use in PLWH and the role of motivational interviewing as part of an integrated approach to care of PLWH with co-occurring substance use disorders. The chapter concludes with a case example to illustrate the role that motivational interviewing can play the care of PLWH with a co-morbidity of substance use disorder.

Comparison of Clinical Outcomes of Persons Living With HIV by Enrollment Status in Washington, DC: Evaluation of a Large Longitudinal HIV Cohort Study (Preprint)

2019 ◽  
Author(s):  
Jenevieve Opoku ◽  
Rupali K Doshi ◽  
Amanda D Castel ◽  
Ian Sorensen ◽  
Michael Horberg ◽  
...  
Keyword(s):  
Cohort Study ◽  
Health Outcomes ◽  
Odds Ratio ◽  
Adjusted Odds Ratio ◽  
Hiv Care ◽  
Disease Stage ◽  
People Living With Hiv ◽  
Hiv Disease ◽  
Persons Living With Hiv ◽  
Living With Hiv

BACKGROUND HIV cohort studies have been used to assess health outcomes and inform the care and treatment of people living with HIV disease. However, there may be similarities and differences between cohort participants and the general population from which they are drawn. OBJECTIVE The objective of this analysis was to compare people living with HIV who have and have not been enrolled in the DC Cohort study and assess whether participants are a representative citywide sample of people living with HIV in the District of Columbia (DC). METHODS Data from the DC Health (DCDOH) HIV surveillance system and the DC Cohort study were matched to identify people living with HIV who were DC residents and had consented for the study by the end of 2016. Analysis was performed to identify differences between DC Cohort and noncohort participants by demographics and comorbid conditions. HIV disease stage, receipt of care, and viral suppression were evaluated. Adjusted logistic regression assessed correlates of health outcomes between the two groups. RESULTS There were 12,964 known people living with HIV in DC at the end of 2016, of which 40.1% were DC Cohort participants. Compared with nonparticipants, participants were less likely to be male (68.0% vs 74.9%, <i>P</i>&lt;.001) but more likely to be black (82.3% vs 69.5%, <i>P</i>&lt;.001) and have a heterosexual contact HIV transmission risk (30.3% vs 25.9%, <i>P</i>&lt;.001). DC Cohort participants were also more likely to have ever been diagnosed with stage 3 HIV disease (59.6% vs 47.0%, <i>P</i>&lt;.001), have a CD4 &lt;200 cells/µL in 2017 (6.2% vs 4.6%, <i>P</i>&lt;.001), be retained in any HIV care in 2017 (72.9% vs 59.4%, <i>P</i>&lt;.001), and be virally suppressed in 2017. After adjusting for demographics, DC Cohort participants were significantly more likely to have received care in 2017 (adjusted odds ratio 1.8, 95% CI 1.70-2.00) and to have ever been virally suppressed (adjusted odds ratio 1.3, 95% CI 1.20-1.40). CONCLUSIONS These data have important implications when assessing the representativeness of patients enrolled in clinic-based cohorts compared with the DC-area general HIV population. As participants continue to enroll in the DC Cohort study, ongoing assessment of representativeness will be required.

scholarly journals Substance Use Among Older People Living With HIV: Challenges for Health Care Providers

2019 ◽  
Vol 7 ◽  
Author(s):  
Sherry Deren ◽  
Tara Cortes ◽  
Victoria Vaughan Dickson ◽  
Vincent Guilamo-Ramos ◽  
Benjamin H. Han ◽  
...  
Keyword(s):  
Health Care ◽  
Substance Use ◽  
Older People ◽  
Health Care Providers ◽  
People Living With Hiv ◽  
Care Providers ◽  
Living With Hiv

Depressive Symptoms, Gender, Disclosure, and HIV Care Stage Among People Living with HIV in Cameroon

2021 ◽  
Author(s):  
Angela M. Parcesepe ◽  
Molly Remch ◽  
Anastase Dzudie ◽  
Rogers Ajeh ◽  
Denis Nash ◽  
...  
Keyword(s):  
Depressive Symptoms ◽  
Hiv Care ◽  
People Living With Hiv ◽  
Living With Hiv

scholarly journals The Utilisation of Payment Models Across the HIV Continuum of Care: Systematic Review of Evidence

2021 ◽  
Author(s):  
Tiago Rua ◽  
Daniela Brandão ◽  
Vanessa Nicolau ◽  
Ana Escoval
Keyword(s):  
Positive Impact ◽  
Cost Effective ◽  
Relative Magnitude ◽  
Hiv Care ◽  
People Living With Hiv ◽  
Hiv Continuum Of Care ◽  
Clinical Conditions ◽  
Study Designs ◽  
Living With Hiv ◽  
Payment Models

AbstractThe increasing chronicity and multimorbidities associated with people living with HIV have posed important challenges to health systems across the world. In this context, payment models hold the potential to improve care across a spectrum of clinical conditions. This study aims to systematically review the evidence of HIV performance-based payments models. Literature searches were conducted in March 2020 using multiple databases and manual searches of relevant papers. Papers were limited to any study design that considers the real-world utilisation of performance-based payment models applied to the HIV domain. A total of 23 full-text papers were included. Due to the heterogeneity of study designs, the multiple types of interventions and its implementation across distinct areas of HIV care, direct comparisons between studies were deemed unsuitable. Most evidence focused on healthcare users (83%), seeking to directly affect patients' behaviour based on principles of behavioural economics. Despite the variability between interventions, the implementation of performance-based payment models led to either a neutral or positive impact throughout the HIV care continuum. Moreover, this improvement was likely to be cost-effective or, at least, did not compromise the healthcare system’s financial sustainability. However, more research is needed to assess the durability of incentives and its appropriate relative magnitude.

Sign in / Sign up

Export Citation Format

Share Document