Amin Adibi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Amin Adibi has a total of 157 co-authors. Here are the top co-authors by number of shared publications:
- Mohsen Sadatsafavi
Respiratory Evaluation Sciences Program, Faculty of Pharmaceutical Sciences, University of British Columbia, Vancouver, Canada
20 shared publications - Kate M. Johnson
Collaboration for Outcomes Research and Evaluation, Faculty of Pharmaceutical Sciences, University of British Columbia, Vancouver, BC, Canada
6 shared publications - Don D. Sin
Centre for Heart Lung Innovation, St. Paul's Hospital, Vancouver, BC, Canada
4 shared publications - John Mark FitzGerald
Centre for Lung Health, Vancouver Coastal Health Research Institute, University of British Columbia, Vancouver, Canada
4 shared publications - Mohsen Yaghoubi
School of Public Health, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
4 shared publications Christopher Carlsten
Division of Respiratory Medicine, Department of Medicine, and
3 shared publications
Latest Publications
- Open Access
- 8 October 2026
- Published by Springer Nature in Diagnostic and Prognostic Research
- Vol. 10 (1) , 27
ACCEPT 3.0: Recalibrating ACute COPD Exacerbation Prediction Tool (ACCEPT) for global use
Background: The ACute COPD Exacerbation Prediction Tool (ACCEPT) is an algorithm for predicting 12-month risk of chronic obstructive pulmonary disease (COPD) exacerbations. To make it suitable for international use, we externally validated and recalibrated ACCEPT 2.0 to incorporate country-specific background exacerbation risk, using multi-country data from NOVELTY (a NOVEL observational longiTudinal studY). Methods: Our study cohort included patients aged $$\:\ge\:$$ 40 years with physician-diagnosed COPD and $$\:\ge\:$$ 30 days of follow-up. We predicted the occurrence of any moderate-to-severe (primary endpoint) and severe (secondary) exacerbations at 12 months, using country-level random effects to account for between-country differences in background exacerbation risks unexplained by predictor effects. We assessed model performance via discrimination, calibration, and net benefit. Results: The analysis sample comprised 5,181 patients (59.4% male, mean age at baseline 66.5 years) from 18 countries, with a mean follow-up of 357 days. 12-month risks ranged between 0.11 and 0.46 (mean = 0.26) for moderate-to-severe and 0.03–0.21 (mean = 0.08) for severe exacerbations. Without country-level adjustment, ACCEPT 2.0’s area under the receiver operating characteristic curve (AUROC) was 0.73 (95% CI 0.72–0.75) and 0.76 (95% CI 0.73–0.78) for moderate-to-severe and severe exacerbations, respectively. However, ACCEPT 2.0 consistently overpredicted risks (observed-to-expected [O/E] risk ratio = 0.57 [95% CI 0.54–0.60] and 0.51 [95% CI 0.46–0.56]). After country-level recalibration, ACCEPT 3.0 showed similar AUROCs (0.75 and 0.76) but substantially improved calibration (O/E ratios: 0.99 [95% CI 0.94–1.05] and 0.99 [95% CI 0.90–1.09]). The net benefit of ACCEPT 3.0 exceeded ACCEPT 2.0 across treatment thresholds of 5%-20%. Conclusion: Recalibrating ACCEPT 2.0 with country-level effects significantly improved calibration of COPD exacerbation risks. The resulting ACCEPT 3.0 is expected to provide higher clinical utility across a range of treatment thresholds.- 8 October 2026
- Published by Springer Nature in Diagnostic and Prognostic Research
- Vol. 10 (1) , 27
Open AccessACCEPT 3.0: Recalibrating ACute COPD Exacerbation Prediction Tool (ACCEPT) for global use
Background: The ACute COPD Exacerbation Prediction Tool (ACCEPT) is an algorithm for predicting 12-month risk of chronic obstructive pulmonary disease (COPD) exacerbations. To make it suitable for international use, we externally validated and recalibrated ACCEPT 2.0 to incorporate country-specific background exacerbation risk, using multi-country data from NOVELTY (a NOVEL observational longiTudinal studY). Methods: Our study cohort included patients aged $$\:\ge\:$$ 40 years with physician-diagnosed COPD and $$\:\ge\:$$ 30 days of follow-up. We predicted the occurrence of any moderate-to-severe (primary endpoint) and severe (secondary) exacerbations at 12 months, using country-level random effects to account for between-country differences in background exacerbation risks unexplained by predictor effects. We assessed model performance via discrimination, calibration, and net benefit. Results: The analysis sample comprised 5,181 patients (59.4% male, mean age at baseline 66.5 years) from 18 countries, with a mean follow-up of 357 days. 12-month risks ranged between 0.11 and 0.46 (mean = 0.26) for moderate-to-severe and 0.03–0.21 (mean = 0.08) for severe exacerbations. Without country-level adjustment, ACCEPT 2.0’s area under the receiver operating characteristic curve (AUROC) was 0.73 (95% CI 0.72–0.75) and 0.76 (95% CI 0.73–0.78) for moderate-to-severe and severe exacerbations, respectively. However, ACCEPT 2.0 consistently overpredicted risks (observed-to-expected [O/E] risk ratio = 0.57 [95% CI 0.54–0.60] and 0.51 [95% CI 0.46–0.56]). After country-level recalibration, ACCEPT 3.0 showed similar AUROCs (0.75 and 0.76) but substantially improved calibration (O/E ratios: 0.99 [95% CI 0.94–1.05] and 0.99 [95% CI 0.90–1.09]). The net benefit of ACCEPT 3.0 exceeded ACCEPT 2.0 across treatment thresholds of 5%-20%. Conclusion: Recalibrating ACCEPT 2.0 with country-level effects significantly improved calibration of COPD exacerbation risks. The resulting ACCEPT 3.0 is expected to provide higher clinical utility across a range of treatment thresholds. Performance of multivariable risk prediction algorithms in predicting COPD exacerbations: a population-based study
Introduction: Efficient preventive management of acute exacerbation of chronic obstructive pulmonary disease (COPD) is predicated on accurate risk stratification. We compared the performance of exacerbation history (current standard of care) versus a revised version of a multivariable risk scoring tool (Acute COPD Exacerbation Prediction Tool (ACCEPT)) using primary-care UK data. Methods: We used validated case definitions to identify diagnosed patients with COPD ≥40 years old from the UK Clinical Practice Research Datalink Aurum (2004–2020). For each patient, a single annual COPD visit was randomly selected as the index date. The outcome was the occurrence of ≥1 moderate/severe exacerbation(s) within a year of the index date. We conducted time-to-event analyses of the latest version of ACCEPT (ACCEPT 2.0) and developed a recalibrated version (ACCEPT 3.0-UK). Model performance was evaluated using discrimination (time-dependent area under the receiver operating characteristic curve (AUC)), calibration and net benefit. Results: The final cohort included 158 384 patients (55.0% male; mean age 71.5 years). ACCEPT 2.0 achieved an AUC of 0.77 for predicting moderate/severe exacerbations, outperforming both any and frequent exacerbator categories (AUC: 0.69 and 0.67, respectively). However, it overpredicted exacerbation events (observed-to-expected (O/E) ratio: 0.81 (95% CI 0.80 to 0.81)). Recalibration resolved this overprediction, yielding O/E ratio of 1.00 (95% CI 0.99 to 1.00), while maintaining discrimination (AUC: 0.77). ACCEPT 3.0-UK was net beneficial and superior to exacerbation history across a wide range of risk thresholds. Conclusion: ACCEPT 3.0-UK has substantially higher performance than exacerbation history, quantifies predicted risks for shared decision-making and is likely to confer clinical utility for risk stratification in primary care.
Performance of multivariable risk prediction algorithms in predicting COPD exacerbations: a population-based study
Introduction: Efficient preventive management of acute exacerbation of chronic obstructive pulmonary disease (COPD) is predicated on accurate risk stratification. We compared the performance of exacerbation history (current standard of care) versus a revised version of a multivariable risk scoring tool (Acute COPD Exacerbation Prediction Tool (ACCEPT)) using primary-care UK data. Methods: We used validated case definitions to identify diagnosed patients with COPD ≥40 years old from the UK Clinical Practice Research Datalink Aurum (2004–2020). For each patient, a single annual COPD visit was randomly selected as the index date. The outcome was the occurrence of ≥1 moderate/severe exacerbation(s) within a year of the index date. We conducted time-to-event analyses of the latest version of ACCEPT (ACCEPT 2.0) and developed a recalibrated version (ACCEPT 3.0-UK). Model performance was evaluated using discrimination (time-dependent area under the receiver operating characteristic curve (AUC)), calibration and net benefit. Results: The final cohort included 158 384 patients (55.0% male; mean age 71.5 years). ACCEPT 2.0 achieved an AUC of 0.77 for predicting moderate/severe exacerbations, outperforming both any and frequent exacerbator categories (AUC: 0.69 and 0.67, respectively). However, it overpredicted exacerbation events (observed-to-expected (O/E) ratio: 0.81 (95% CI 0.80 to 0.81)). Recalibration resolved this overprediction, yielding O/E ratio of 1.00 (95% CI 0.99 to 1.00), while maintaining discrimination (AUC: 0.77). ACCEPT 3.0-UK was net beneficial and superior to exacerbation history across a wide range of risk thresholds. Conclusion: ACCEPT 3.0-UK has substantially higher performance than exacerbation history, quantifies predicted risks for shared decision-making and is likely to confer clinical utility for risk stratification in primary care.
- Open Access
Multinational Public Opinion on Race, Ethnicity, and Algorithmic Reform in Medicine
Importance: Several professional medical societies have removed race and ethnicity from widely used clinical algorithms with implications for millions. Yet, patient and public opinions regarding the tensions underlying these pivotal changes have not been systematically explored. Objective: To assess global public opinion on the use of race or ethnicity in clinical algorithms, including preferences for different approaches to algorithmic reform and perceptions of alternative predictors. Design: Cross-sectional survey study. Setting: Multinational opt-in online survey in January 2026. Participants: A volunteer convenience sample with quota sampling to achieve approximately equal participation by sex at birth and across ten categories of self-identified race and ethnicity. Main Outcomes and Measures: Self-reported comfort with demographic and social predictors in clinical calculators, with net comfort defined as percentage extremely or somewhat comfortable minus percentage extremely or somewhat uncomfortable; preferences for race-specific versus race-free algorithms; perceptions of algorithmic harm or benefit. Results: Of 1,050 responses, 994 (94.7%) met eligibility criteria. Participants resided in 43 countries with a median age of 32.0 years (IQR, 26-41). Net comfort with the use of race or ethnicity in a hypothetical cancer risk calculator was +62.4% (95% CI: +57.8% to +66.9%), compared with +14.5% (95% CI: +9.1% to +19.9%) for postal or ZIP code. Overall, 87.9% (95% CI: 85.9% to 90.0%) were comfortable with the use of race or ethnicity if a clinician explained its use, and only 12.8% agreed race and ethnicity should never be used clinically. Across spirometry, kidney function, and cardiovascular risk calculators, 40.0% to 47.6% preferred race-specific versions, whereas 16.7% to 28.2% preferred race-free alternatives. Furthermore, a substantial proportion disagreed that they were well-represented by race and ethnicity categories, ranging from 22.1% for osteoporotic fracture risk equations to 42.9% for cardiovascular risk equations. These findings were consistent across countries, self-identified race and ethnicity, and among participants reporting prior experiences of racism in healthcare. Conclusions and Relevance: In our diverse multinational survey study, respondents were comfortable with the use of race and ethnicity across application areas, but often did not feel represented by existing categories and were less comfortable with alternatives based on postal or ZIP codes.
Open AccessMultinational Public Opinion on Race, Ethnicity, and Algorithmic Reform in Medicine
Importance: Several professional medical societies have removed race and ethnicity from widely used clinical algorithms with implications for millions. Yet, patient and public opinions regarding the tensions underlying these pivotal changes have not been systematically explored. Objective: To assess global public opinion on the use of race or ethnicity in clinical algorithms, including preferences for different approaches to algorithmic reform and perceptions of alternative predictors. Design: Cross-sectional survey study. Setting: Multinational opt-in online survey in January 2026. Participants: A volunteer convenience sample with quota sampling to achieve approximately equal participation by sex at birth and across ten categories of self-identified race and ethnicity. Main Outcomes and Measures: Self-reported comfort with demographic and social predictors in clinical calculators, with net comfort defined as percentage extremely or somewhat comfortable minus percentage extremely or somewhat uncomfortable; preferences for race-specific versus race-free algorithms; perceptions of algorithmic harm or benefit. Results: Of 1,050 responses, 994 (94.7%) met eligibility criteria. Participants resided in 43 countries with a median age of 32.0 years (IQR, 26-41). Net comfort with the use of race or ethnicity in a hypothetical cancer risk calculator was +62.4% (95% CI: +57.8% to +66.9%), compared with +14.5% (95% CI: +9.1% to +19.9%) for postal or ZIP code. Overall, 87.9% (95% CI: 85.9% to 90.0%) were comfortable with the use of race or ethnicity if a clinician explained its use, and only 12.8% agreed race and ethnicity should never be used clinically. Across spirometry, kidney function, and cardiovascular risk calculators, 40.0% to 47.6% preferred race-specific versions, whereas 16.7% to 28.2% preferred race-free alternatives. Furthermore, a substantial proportion disagreed that they were well-represented by race and ethnicity categories, ranging from 22.1% for osteoporotic fracture risk equations to 42.9% for cardiovascular risk equations. These findings were consistent across countries, self-identified race and ethnicity, and among participants reporting prior experiences of racism in healthcare. Conclusions and Relevance: In our diverse multinational survey study, respondents were comfortable with the use of race and ethnicity across application areas, but often did not feel represented by existing categories and were less comfortable with alternatives based on postal or ZIP codes.
- Open Access
Boards-style benchmarks overestimate prior-chat bias in large language models: a factorial evaluation study
Background: Large language models (LLMs) are increasingly piloted as chat interfaces for chart review and clinical decision support. Although leading models achieve and even exceed physician-level accuracy on exam-style benchmarks such as MedQA, recent perturbation studies show large drops in accuracy after small changes to prompts, distractor content, or answer format. Prior work has not systematically examined how these vulnerabilities unintentionally manifest in clinically realistic settings, including multi-turn chatbot interactions, free-text response formats, and tasks involving patient medical records. Methods: We evaluated susceptibility to bias from prior chat messages across 14 LLMs (10 closed-source, 4 open-source) on two medical question-answering tasks: a boards-style benchmark (1000 MedQA test questions) and an electronic health record (EHR) information retrieval task (962 EHRNoteQA questions about real patient discharge summaries). Using a factorial design, we independently varied the presence and type of prior-chat distractors and response format across these two tasks. Distractors ranged from simple statements of incorrect answers to more realistic conversational exchanges between user and model, including interactions referencing a different patient. Findings: Prior-chat distractors produced large and consistent accuracy decrements in the MedQA multiple-choice setting, particularly when the prior message stated an incorrect answer. In this setting, insertion of this user message led to significant accuracy decreases in 13 of 14 models, with drops averaging 15.0 percentage points across models. Effects were smaller for more plausible, conversational distractors and in free-response formats. In contrast, prior-chat bias in the discharge summary-based task was modest and inconsistent. Average accuracy decreases were under 2 percentage points across all distractor types and response formats assessed, with significant effects observed in a minority of models. Interpretation: LLM performance can be biased toward incorrect answers by plausible prior-chat distractors, but these effects are highly context-dependent. We find that distraction effects are common and often substantial in the boards-style multiple-choice task, particularly when the distractor is an explicit (and unrealistic) prior message containing an incorrect answer. In contrast, these effects are markedly attenuated when the same questions are posed in free-response format and the distractor is incorporated into a clinically-realistic user-model exchange in the chat history, or when the task is switched from a boards-style vignette to a question about a real (de-identified) patient record. Taken together, these results suggest that evaluations based solely on single-turn, boards-style multiple-choice questions with unrealistic distractors may overstate the impact of prior-chat bias. These findings highlight the need to assess LLM behavior in multi-turn settings involving realistic clinical use cases, rather than relying on boards-style benchmarks for assessment of safety risks.
Open AccessBoards-style benchmarks overestimate prior-chat bias in large language models: a factorial evaluation study
Background: Large language models (LLMs) are increasingly piloted as chat interfaces for chart review and clinical decision support. Although leading models achieve and even exceed physician-level accuracy on exam-style benchmarks such as MedQA, recent perturbation studies show large drops in accuracy after small changes to prompts, distractor content, or answer format. Prior work has not systematically examined how these vulnerabilities unintentionally manifest in clinically realistic settings, including multi-turn chatbot interactions, free-text response formats, and tasks involving patient medical records. Methods: We evaluated susceptibility to bias from prior chat messages across 14 LLMs (10 closed-source, 4 open-source) on two medical question-answering tasks: a boards-style benchmark (1000 MedQA test questions) and an electronic health record (EHR) information retrieval task (962 EHRNoteQA questions about real patient discharge summaries). Using a factorial design, we independently varied the presence and type of prior-chat distractors and response format across these two tasks. Distractors ranged from simple statements of incorrect answers to more realistic conversational exchanges between user and model, including interactions referencing a different patient. Findings: Prior-chat distractors produced large and consistent accuracy decrements in the MedQA multiple-choice setting, particularly when the prior message stated an incorrect answer. In this setting, insertion of this user message led to significant accuracy decreases in 13 of 14 models, with drops averaging 15.0 percentage points across models. Effects were smaller for more plausible, conversational distractors and in free-response formats. In contrast, prior-chat bias in the discharge summary-based task was modest and inconsistent. Average accuracy decreases were under 2 percentage points across all distractor types and response formats assessed, with significant effects observed in a minority of models. Interpretation: LLM performance can be biased toward incorrect answers by plausible prior-chat distractors, but these effects are highly context-dependent. We find that distraction effects are common and often substantial in the boards-style multiple-choice task, particularly when the distractor is an explicit (and unrealistic) prior message containing an incorrect answer. In contrast, these effects are markedly attenuated when the same questions are posed in free-response format and the distractor is incorporated into a clinically-realistic user-model exchange in the chat history, or when the task is switched from a boards-style vignette to a question about a real (de-identified) patient record. Taken together, these results suggest that evaluations based solely on single-turn, boards-style multiple-choice questions with unrealistic distractors may overstate the impact of prior-chat bias. These findings highlight the need to assess LLM behavior in multi-turn settings involving realistic clinical use cases, rather than relying on boards-style benchmarks for assessment of safety risks.
- 1 November 2025
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 211 (11) , 2117-2126
Social Determinants of Health and Racial Disparities in Lung Function: Findings from the National Health and Nutrition Examination Survey, 2007–2012
Rationale: We hypothesized that the disproportionate impact of social determinants of health (SDoH) captured in survey data could help explain a larger proportion of racial gaps in lung function than previously reported. Objective: To estimate the contribution of SDoH to racial gaps in lung function. Methods: We defined a series of nested, increasingly healthy reference populations using data from the National Health and Nutrition Examination Survey, 2007–2012. Starting with nonsmokers without respiratory symptoms or diagnoses, we sequentially excluded those with confirmed occupational exposure to dust and/or fumes, physical inactivity, maternal or secondhand tobacco use, obesity, no home ownership, no insurance, lower education, and self-reported unhealthy diet. Across successive populations, we compared average age-, sex-, and height-adjusted differences in FEV1 and FVC between racial and ethnic minority groups and non-Hispanic White participants for adults (aged ≥20 yr) and children (aged 6–19 yr). Measurements and Main Results: In successively healthier reference populations, the proportion of represented participants declined for non-Hispanic Black, Mexican American, and other Hispanic participants; increased for non-Hispanic White participants; and remained stable for non-Hispanic Asian participants. At baseline, adjusted FEV1 and FVC were similar for Mexican American and non-Hispanic White American participants but lower for other racial and ethnic minority groups. After excluding individuals with unfavorable SDoH, racial disparities in FEV1 and FVC decreased for non-Hispanic Black children (24.8% and 26.2%) and adults (26.3% and 19.4%), other Hispanic children (15.2% and 19.3%) and adults (85.9% and 12.4%), and non-Hispanic Asian children (6.6% and 12.5%), but they increased for non-Hispanic Asian adults (14.1% and 11.1%). Conclusions: Unfavorable SDoH disproportionately affected non-Hispanic Black, Mexican American, and other Hispanic populations and explained a higher proportion of racial disparities in lung function than previously reported.
- 1 November 2025
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 211 (11) , 2117-2126
Social Determinants of Health and Racial Disparities in Lung Function: Findings from the National Health and Nutrition Examination Survey, 2007–2012
Rationale: We hypothesized that the disproportionate impact of social determinants of health (SDoH) captured in survey data could help explain a larger proportion of racial gaps in lung function than previously reported. Objective: To estimate the contribution of SDoH to racial gaps in lung function. Methods: We defined a series of nested, increasingly healthy reference populations using data from the National Health and Nutrition Examination Survey, 2007–2012. Starting with nonsmokers without respiratory symptoms or diagnoses, we sequentially excluded those with confirmed occupational exposure to dust and/or fumes, physical inactivity, maternal or secondhand tobacco use, obesity, no home ownership, no insurance, lower education, and self-reported unhealthy diet. Across successive populations, we compared average age-, sex-, and height-adjusted differences in FEV1 and FVC between racial and ethnic minority groups and non-Hispanic White participants for adults (aged ≥20 yr) and children (aged 6–19 yr). Measurements and Main Results: In successively healthier reference populations, the proportion of represented participants declined for non-Hispanic Black, Mexican American, and other Hispanic participants; increased for non-Hispanic White participants; and remained stable for non-Hispanic Asian participants. At baseline, adjusted FEV1 and FVC were similar for Mexican American and non-Hispanic White American participants but lower for other racial and ethnic minority groups. After excluding individuals with unfavorable SDoH, racial disparities in FEV1 and FVC decreased for non-Hispanic Black children (24.8% and 26.2%) and adults (26.3% and 19.4%), other Hispanic children (15.2% and 19.3%) and adults (85.9% and 12.4%), and non-Hispanic Asian children (6.6% and 12.5%), but they increased for non-Hispanic Asian adults (14.1% and 11.1%). Conclusions: Unfavorable SDoH disproportionately affected non-Hispanic Black, Mexican American, and other Hispanic populations and explained a higher proportion of racial disparities in lung function than previously reported.
- 28 October 2025
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 66 (6) , 2501752
Discrepancies between observed and self-identified race in spirometry interpretation: insights from the Canadian Longitudinal Study on Aging
- 28 October 2025
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 66 (6) , 2501752
Discrepancies between observed and self-identified race in spirometry interpretation: insights from the Canadian Longitudinal Study on Aging
- Open Access
Cost-effectiveness of a government rebate program for air cleaners in preventing asthma and related adverse health outcomes
Background: Wildfire-derived fine particulate matter (PM 2.5 ) is an increasing contributor to air pollution in British Columbia (BC), Canada, and is linked to asthma development and asthma-related adverse outcomes including exacerbations. Portable high-efficiency particulate air (HEPA) cleaners can reduce indoor PM 2.5 , but evidence on their long-term, population-level cost-effectiveness is limited. Methods: We developed a monthly, time-varying Markov cohort model (2010–2035; healthcare payer perspective; 1.5% annual discounting; $50,000/QALY willingness-to-pay) for two BC cohorts (age 5 and age 25 at baseline) across 16 Health Service Delivery Areas (HSDAs). The model simulated monthly transitions between health states (well-controlled asthma, not well-controlled asthma, exacerbations, and death) over a 25-year time horizon (2010–2035). The target population was children age 5 and adults aged 25 (separate cohorts). We combined historical total PM 2.5 (2010–2022) from provincial data sources with projected PM 2.5 (2023–2035) combining anthropogenic emissions with wildfire-derived PM 2.5 . Wildfire PM 2.5 projections were calculated by multiplying monthly PM 2.5 averages from 2018–2022 by 0%, 5.5%, and 11% cumulative PM 2.5 increase scenarios, informed by climate modeling. The base-case rebate was $150, and varied rebates between $50 to $200. We assumed continuous use with unit replacement every 5 years and filter replacement every 9 months. We calculated asthma incidence attributable to wildfire PM 2.5 and incremental cost-effectiveness ratios (ICERs) across BC’s 16 Health Service Delivery Areas (HSDAs). We conducted the analysis from a health payer perspective with a 1.5% discount rate and $50,000/QALY willingness to pay threshold. Costs are expressed in 2024 CAD. Results: From 2023–2036, wildfire-derived PM 2.5 was associated with 13–14 incident asthma cases per 100,000 person-years annually across BC. Over 25 years, air cleaners prevented 444 moderate exacerbations, 55 emergency department visits, and 42 hospitalizations (combined cohorts), but the base-case program was not cost-effective in any HSDA (ICER range: $149,408–$154,749/QALY). A $50 rebate was cost-effective province-wide and $100 was cost-effective in three HSDAs. Results were most sensitive to concentration-response functions for PM2.5 (incidence, control, exacerbations) and to HEPA air cleaner and asthma care costs. Cost-effectiveness was most sensitive to air cleaner costs and the concentration-response function between PM 2.5 and asthma incidence. Conclusions: Wildfire-related PM 2.5 contributes meaningfully to asthma incidence in BC. A universal $150 HEPA rebate program was not cost-effective for primary prevention under base-case assumptions, whereas lower rebates ($50 province-wide, $100 in three HSDAs) may offer better value. Future evaluations should co-benefits across multiple disease outcomes to better support policies to reduce the health impacts of increasing wildfires.
Open AccessCost-effectiveness of a government rebate program for air cleaners in preventing asthma and related adverse health outcomes
Background: Wildfire-derived fine particulate matter (PM 2.5 ) is an increasing contributor to air pollution in British Columbia (BC), Canada, and is linked to asthma development and asthma-related adverse outcomes including exacerbations. Portable high-efficiency particulate air (HEPA) cleaners can reduce indoor PM 2.5 , but evidence on their long-term, population-level cost-effectiveness is limited. Methods: We developed a monthly, time-varying Markov cohort model (2010–2035; healthcare payer perspective; 1.5% annual discounting; $50,000/QALY willingness-to-pay) for two BC cohorts (age 5 and age 25 at baseline) across 16 Health Service Delivery Areas (HSDAs). The model simulated monthly transitions between health states (well-controlled asthma, not well-controlled asthma, exacerbations, and death) over a 25-year time horizon (2010–2035). The target population was children age 5 and adults aged 25 (separate cohorts). We combined historical total PM 2.5 (2010–2022) from provincial data sources with projected PM 2.5 (2023–2035) combining anthropogenic emissions with wildfire-derived PM 2.5 . Wildfire PM 2.5 projections were calculated by multiplying monthly PM 2.5 averages from 2018–2022 by 0%, 5.5%, and 11% cumulative PM 2.5 increase scenarios, informed by climate modeling. The base-case rebate was $150, and varied rebates between $50 to $200. We assumed continuous use with unit replacement every 5 years and filter replacement every 9 months. We calculated asthma incidence attributable to wildfire PM 2.5 and incremental cost-effectiveness ratios (ICERs) across BC’s 16 Health Service Delivery Areas (HSDAs). We conducted the analysis from a health payer perspective with a 1.5% discount rate and $50,000/QALY willingness to pay threshold. Costs are expressed in 2024 CAD. Results: From 2023–2036, wildfire-derived PM 2.5 was associated with 13–14 incident asthma cases per 100,000 person-years annually across BC. Over 25 years, air cleaners prevented 444 moderate exacerbations, 55 emergency department visits, and 42 hospitalizations (combined cohorts), but the base-case program was not cost-effective in any HSDA (ICER range: $149,408–$154,749/QALY). A $50 rebate was cost-effective province-wide and $100 was cost-effective in three HSDAs. Results were most sensitive to concentration-response functions for PM2.5 (incidence, control, exacerbations) and to HEPA air cleaner and asthma care costs. Cost-effectiveness was most sensitive to air cleaner costs and the concentration-response function between PM 2.5 and asthma incidence. Conclusions: Wildfire-related PM 2.5 contributes meaningfully to asthma incidence in BC. A universal $150 HEPA rebate program was not cost-effective for primary prevention under base-case assumptions, whereas lower rebates ($50 province-wide, $100 in three HSDAs) may offer better value. Future evaluations should co-benefits across multiple disease outcomes to better support policies to reduce the health impacts of increasing wildfires.
- 1 March 2025
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 211 (3) , 432-435
Is Achieving a Fully Race-Neutral Approach to Lung Function Classification Even Possible?
- 1 March 2025
- Published by Oxford University Press (OUP) in American Journal of Respiratory and Critical Care Medicine
- Vol. 211 (3) , 432-435
Is Achieving a Fully Race-Neutral Approach to Lung Function Classification Even Possible?
- Open Access
Recent Advances, Applications and Open Challenges in Machine Learning for Health: Reflections from Research Roundtables at ML4H 2024 Symposium
The fourth Machine Learning for Health (ML4H) symposium was held in person on December 15th and 16th, 2024, in the traditional, ancestral, and unceded territories of the Musqueam, Squamish, and Tsleil-Waututh Nations in Vancouver, British Columbia, Canada. The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for the ML4H community. The organization of the research roundtables at the conference involved 13 senior and 27 junior chairs across 13 tables. Each roundtable session included an invited senior chair (with substantial experience in the field), junior chairs (responsible for facilitating the discussion), and attendees from diverse backgrounds with an interest in the session's topic.
Open AccessRecent Advances, Applications and Open Challenges in Machine Learning for Health: Reflections from Research Roundtables at ML4H 2024 Symposium
The fourth Machine Learning for Health (ML4H) symposium was held in person on December 15th and 16th, 2024, in the traditional, ancestral, and unceded territories of the Musqueam, Squamish, and Tsleil-Waututh Nations in Vancouver, British Columbia, Canada. The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for the ML4H community. The organization of the research roundtables at the conference involved 13 senior and 27 junior chairs across 13 tables. Each roundtable session included an invited senior chair (with substantial experience in the field), junior chairs (responsible for facilitating the discussion), and attendees from diverse backgrounds with an interest in the session's topic.
- 14 September 2024
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 64 (suppl 68) , OA5572
The effect of health disparities on racial gaps in lung function
Background: Racialized populations disproportionately experience exposures that impair lung function. Race-neutral reference equations mitigate bias due to disparities at the cost of reduced precision. However, it is unclear what proportion of racial gaps in lung function is due to known disparities. Hypothesis: Measurable disparities will help explain racial gaps in lung function. Methods: We defined reference populations from 20,050 Black and White NHANES 2007-2012 participants. Starting with non-smokers without respiratory symptoms or diagnoses, we sequentially excluded those with occupational exposure to dust/fumes, maternal tobacco use, second-hand smoke exposure, obesity, low physical activity, unhealthy diet, self-assessed poor health, and no insurance. Across populations, we compared average age-, sex-, and height-adjusted differences in FEV1 and FVC between Black and White adults (≥20) and youth (1 and from 0.48 L (95%CI 0.44-0.51) to 0.36 (95%CI 0.30-0.42) in FVC in youth but increased in adults. Conclusions: The disparities investigated herein explain up to 25% of the racial gap in lung function among American youth. Further research on adults is required. erj;64/suppl_68/OA5572/F1 F1 F1
- 14 September 2024
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 64 (suppl 68) , OA5572
The effect of health disparities on racial gaps in lung function
Background: Racialized populations disproportionately experience exposures that impair lung function. Race-neutral reference equations mitigate bias due to disparities at the cost of reduced precision. However, it is unclear what proportion of racial gaps in lung function is due to known disparities. Hypothesis: Measurable disparities will help explain racial gaps in lung function. Methods: We defined reference populations from 20,050 Black and White NHANES 2007-2012 participants. Starting with non-smokers without respiratory symptoms or diagnoses, we sequentially excluded those with occupational exposure to dust/fumes, maternal tobacco use, second-hand smoke exposure, obesity, low physical activity, unhealthy diet, self-assessed poor health, and no insurance. Across populations, we compared average age-, sex-, and height-adjusted differences in FEV1 and FVC between Black and White adults (≥20) and youth (1 and from 0.48 L (95%CI 0.44-0.51) to 0.36 (95%CI 0.30-0.42) in FVC in youth but increased in adults. Conclusions: The disparities investigated herein explain up to 25% of the racial gap in lung function among American youth. Further research on adults is required. erj;64/suppl_68/OA5572/F1 F1 F1