Emily Madsen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Emily Madsen has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Jordan Wassertheil Smoller
Harvard Medical School
5 shared publications - Hyunjoon Lee
Massachusetts General Hospital
4 shared publications - Kate Ba Bentley
Center for Precision Psychiatry, Department of Psychiatry, Massachusetts General Hospital, Boston, USA
4 shared publications - Victor M. Castro
Massachusetts General Hospital
4 shared publications - Younga Heather Lee
Massachusetts General Hospital
3 shared publications - Yu Zhou
Department of Psychiatry Massachusetts General Hospital Boston Massachusetts USA
3 shared publications
Latest Publications
- Open Access
Would Lifting Versus Maintaining COVID-19 Containment Policies Have Reduced Psychological Distress in the US?
Background: Both the COVID-19 pandemic and containment policies caused widespread psychological distress, yet their independent effects remain unclear. Disentangling these effects could inform future responses that balance physical and mental health. This study sought to estimate the effect of lifting versus maintaining containment policies on psychological distress, independent of pandemic severity. Methods: We conducted a state-level longitudinal analysis using the Behavioral Risk Factor Surveillance System (BRFSS), a representative survey of US adults, restricted to the pandemic period preceding widespread vaccine availability (April 2020 to April 2021). The exposure was lifting versus maintaining containment policies (school closures, workplace closures, event cancellations, and full lockdown) from the Oxford COVID-19 Government Response Tracker. Exposure was measured during periods of low (14 days) pandemic severity. The primary outcome was prevalence of psychological distress, derived from a BRFSS survey item corresponding to a PHQ-4 score ≥6. Findings: Three causal inference approaches yielded consistent evidence of transient policy-lifting effects: (1) synthetic control analysis of Maine showed a temporary 5.5 percentage-point reduction in psychological distress lasting three months before returning to counterfactual levels; (2) within-state fixed effects found immediately after lifting full lockdown, distress decreased by 5.68 [-8.67, −2.69] percentage points, declining by 30 days (−3.24 [-6.88, 0.39]) and negligible at 60 days (−0.94 [-3.77, 1.89]); (3) target trial emulation detected no significant effects from lifting versus maintaining policies for 90 days. Interpretation: Lifting containment policies in the first year of the pandemic produced immediate but transient reductions in psychological distress. These results suggest that extended containment policies were unlikely to account for persistent increases in distress during this period.
Open AccessWould Lifting Versus Maintaining COVID-19 Containment Policies Have Reduced Psychological Distress in the US?
Background: Both the COVID-19 pandemic and containment policies caused widespread psychological distress, yet their independent effects remain unclear. Disentangling these effects could inform future responses that balance physical and mental health. This study sought to estimate the effect of lifting versus maintaining containment policies on psychological distress, independent of pandemic severity. Methods: We conducted a state-level longitudinal analysis using the Behavioral Risk Factor Surveillance System (BRFSS), a representative survey of US adults, restricted to the pandemic period preceding widespread vaccine availability (April 2020 to April 2021). The exposure was lifting versus maintaining containment policies (school closures, workplace closures, event cancellations, and full lockdown) from the Oxford COVID-19 Government Response Tracker. Exposure was measured during periods of low (14 days) pandemic severity. The primary outcome was prevalence of psychological distress, derived from a BRFSS survey item corresponding to a PHQ-4 score ≥6. Findings: Three causal inference approaches yielded consistent evidence of transient policy-lifting effects: (1) synthetic control analysis of Maine showed a temporary 5.5 percentage-point reduction in psychological distress lasting three months before returning to counterfactual levels; (2) within-state fixed effects found immediately after lifting full lockdown, distress decreased by 5.68 [-8.67, −2.69] percentage points, declining by 30 days (−3.24 [-6.88, 0.39]) and negligible at 60 days (−0.94 [-3.77, 1.89]); (3) target trial emulation detected no significant effects from lifting versus maintaining policies for 90 days. Interpretation: Lifting containment policies in the first year of the pandemic produced immediate but transient reductions in psychological distress. These results suggest that extended containment policies were unlikely to account for persistent increases in distress during this period.
- Open Access
- 8 January 2024
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 8 (1) , e46364
Determining Distinct Suicide Attempts From Recurrent Electronic Health Record Codes: Classification Study
Background: Prior suicide attempts are a relatively strong risk factor for future suicide attempts. There is growing interest in using longitudinal electronic health record (EHR) data to derive statistical risk prediction models for future suicide attempts and other suicidal behavior outcomes. However, model performance may be inflated by a largely unrecognized form of “data leakage” during model training: diagnostic codes for suicide attempt outcomes may refer to prior attempts that are also included in the model as predictors. Objective: We aimed to develop an automated rule for determining when documented suicide attempt diagnostic codes identify distinct suicide attempt events. Methods: From a large health care system’s EHR, we randomly sampled suicide attempt codes for 300 patients with at least one pair of suicide attempt codes documented at least one but no more than 90 days apart. Supervised chart reviewers assigned the clinical settings (ie, emergency department [ED] versus non-ED), methods of suicide attempt, and intercode interval (number of days). The probability (or positive predictive value) that the second suicide attempt code in a given pair of codes referred to a distinct suicide attempt event from its preceding suicide attempt code was calculated by clinical setting, method, and intercode interval. Results: Of 1015 code pairs reviewed, 835 (82.3%) were nonindependent (ie, the 2 codes referred to the same suicide attempt event). When the second code in a pair was documented in a clinical setting other than the ED, it represented a distinct suicide attempt 3.3% of the time. The more time elapsed between codes, the more likely the second code in a pair referred to a distinct suicide attempt event from its preceding code. Code pairs in which the second suicide attempt code was assigned in an ED at least 5 days after its preceding suicide attempt code had a positive predictive value of 0.90. Conclusions: EHR-based suicide risk prediction models that include International Classification of Diseases codes for prior suicide attempts as a predictor may be highly susceptible to bias due to data leakage in model training. We derived a simple rule to distinguish codes that reflect new, independent suicide attempts: suicide attempt codes documented in an ED setting at least 5 days after a preceding suicide attempt code can be confidently treated as new events in EHR-based suicide risk prediction models. This rule has the potential to minimize upward bias in model performance when prior suicide attempts are included as predictors in EHR-based suicide risk prediction models.
- 8 January 2024
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 8 (1) , e46364
Open AccessDetermining Distinct Suicide Attempts From Recurrent Electronic Health Record Codes: Classification Study
Background: Prior suicide attempts are a relatively strong risk factor for future suicide attempts. There is growing interest in using longitudinal electronic health record (EHR) data to derive statistical risk prediction models for future suicide attempts and other suicidal behavior outcomes. However, model performance may be inflated by a largely unrecognized form of “data leakage” during model training: diagnostic codes for suicide attempt outcomes may refer to prior attempts that are also included in the model as predictors. Objective: We aimed to develop an automated rule for determining when documented suicide attempt diagnostic codes identify distinct suicide attempt events. Methods: From a large health care system’s EHR, we randomly sampled suicide attempt codes for 300 patients with at least one pair of suicide attempt codes documented at least one but no more than 90 days apart. Supervised chart reviewers assigned the clinical settings (ie, emergency department [ED] versus non-ED), methods of suicide attempt, and intercode interval (number of days). The probability (or positive predictive value) that the second suicide attempt code in a given pair of codes referred to a distinct suicide attempt event from its preceding suicide attempt code was calculated by clinical setting, method, and intercode interval. Results: Of 1015 code pairs reviewed, 835 (82.3%) were nonindependent (ie, the 2 codes referred to the same suicide attempt event). When the second code in a pair was documented in a clinical setting other than the ED, it represented a distinct suicide attempt 3.3% of the time. The more time elapsed between codes, the more likely the second code in a pair referred to a distinct suicide attempt event from its preceding code. Code pairs in which the second suicide attempt code was assigned in an ED at least 5 days after its preceding suicide attempt code had a positive predictive value of 0.90. Conclusions: EHR-based suicide risk prediction models that include International Classification of Diseases codes for prior suicide attempts as a predictor may be highly susceptible to bias due to data leakage in model training. We derived a simple rule to distinguish codes that reflect new, independent suicide attempts: suicide attempt codes documented in an ED setting at least 5 days after a preceding suicide attempt code can be confidently treated as new events in EHR-based suicide risk prediction models. This rule has the potential to minimize upward bias in model performance when prior suicide attempts are included as predictors in EHR-based suicide risk prediction models.
- Open Access
- 8 February 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Determining Distinct Suicide Attempts From Recurrent Electronic Health Record Codes: Classification Study (Preprint)
BACKGROUND Prior suicide attempts are a relatively strong risk factor for future suicide attempts. There is growing interest in using longitudinal electronic health record (EHR) data to derive statistical risk prediction models for future suicide attempts and other suicidal behavior outcomes. However, model performance may be inflated by a largely unrecognized form of “data leakage” during model training: diagnostic codes for suicide attempt outcomes may refer to prior attempts that are also included in the model as predictors. OBJECTIVE We aimed to develop an automated rule for determining when documented suicide attempt diagnostic codes identify distinct suicide attempt events. METHODS From a large health care system’s EHR, we randomly sampled suicide attempt codes for 300 patients with at least one pair of suicide attempt codes documented at least one but no more than 90 days apart. Supervised chart reviewers assigned the clinical settings (ie, emergency department [ED] versus non-ED), methods of suicide attempt, and intercode interval (number of days). The probability (or positive predictive value) that the second suicide attempt code in a given pair of codes referred to a distinct suicide attempt event from its preceding suicide attempt code was calculated by clinical setting, method, and intercode interval. RESULTS Of 1015 code pairs reviewed, 835 (82.3%) were nonindependent (ie, the 2 codes referred to the same suicide attempt event). When the second code in a pair was documented in a clinical setting other than the ED, it represented a distinct suicide attempt 3.3% of the time. The more time elapsed between codes, the more likely the second code in a pair referred to a distinct suicide attempt event from its preceding code. Code pairs in which the second suicide attempt code was assigned in an ED at least 5 days after its preceding suicide attempt code had a positive predictive value of 0.90. CONCLUSIONS EHR-based suicide risk prediction models that include International Classification of Diseases codes for prior suicide attempts as a predictor may be highly susceptible to bias due to data leakage in model training. We derived a simple rule to distinguish codes that reflect new, independent suicide attempts: suicide attempt codes documented in an ED setting at least 5 days after a preceding suicide attempt code can be confidently treated as new events in EHR-based suicide risk prediction models. This rule has the potential to minimize upward bias in model performance when prior suicide attempts are included as predictors in EHR-based suicide risk prediction models.
- 8 February 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDetermining Distinct Suicide Attempts From Recurrent Electronic Health Record Codes: Classification Study (Preprint)
BACKGROUND Prior suicide attempts are a relatively strong risk factor for future suicide attempts. There is growing interest in using longitudinal electronic health record (EHR) data to derive statistical risk prediction models for future suicide attempts and other suicidal behavior outcomes. However, model performance may be inflated by a largely unrecognized form of “data leakage” during model training: diagnostic codes for suicide attempt outcomes may refer to prior attempts that are also included in the model as predictors. OBJECTIVE We aimed to develop an automated rule for determining when documented suicide attempt diagnostic codes identify distinct suicide attempt events. METHODS From a large health care system’s EHR, we randomly sampled suicide attempt codes for 300 patients with at least one pair of suicide attempt codes documented at least one but no more than 90 days apart. Supervised chart reviewers assigned the clinical settings (ie, emergency department [ED] versus non-ED), methods of suicide attempt, and intercode interval (number of days). The probability (or positive predictive value) that the second suicide attempt code in a given pair of codes referred to a distinct suicide attempt event from its preceding suicide attempt code was calculated by clinical setting, method, and intercode interval. RESULTS Of 1015 code pairs reviewed, 835 (82.3%) were nonindependent (ie, the 2 codes referred to the same suicide attempt event). When the second code in a pair was documented in a clinical setting other than the ED, it represented a distinct suicide attempt 3.3% of the time. The more time elapsed between codes, the more likely the second code in a pair referred to a distinct suicide attempt event from its preceding code. Code pairs in which the second suicide attempt code was assigned in an ED at least 5 days after its preceding suicide attempt code had a positive predictive value of 0.90. CONCLUSIONS EHR-based suicide risk prediction models that include International Classification of Diseases codes for prior suicide attempts as a predictor may be highly susceptible to bias due to data leakage in model training. We derived a simple rule to distinguish codes that reflect new, independent suicide attempts: suicide attempt codes documented in an ED setting at least 5 days after a preceding suicide attempt code can be confidently treated as new events in EHR-based suicide risk prediction models. This rule has the potential to minimize upward bias in model performance when prior suicide attempts are included as predictors in EHR-based suicide risk prediction models.
- Open Access
- 11 March 2022
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 6 (3) , e30946
Implementing Machine Learning Models for Suicide Risk Prediction in Clinical Practice: Focus Group Study With Hospital Providers
Background: Interest in developing machine learning models that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. However, whether and how such models might be implemented and useful in clinical practice remain unknown. To ultimately make automated suicide risk–prediction models useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders, including the frontline providers who will be using such tools, at each stage of the implementation process. Objective: The aim of this focus group study is to inform ongoing and future efforts to deploy suicide risk–prediction models in clinical practice. The specific goals are to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk–prediction models in practice; and identify barriers, facilitators, recommendations, and factors to consider. Methods: We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by 2 independent study staff members. All coded text was reviewed and discrepancies were resolved in consensus meetings with doctoral-level staff. Results: Although most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk–prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the health care system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider training. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings. Conclusions: Providers were dissatisfied with current suicide risk assessment methods and were open to the use of a machine learning–based risk-prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of these new approaches in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.
- 11 March 2022
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 6 (3) , e30946
Open AccessImplementing Machine Learning Models for Suicide Risk Prediction in Clinical Practice: Focus Group Study With Hospital Providers
Background: Interest in developing machine learning models that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. However, whether and how such models might be implemented and useful in clinical practice remain unknown. To ultimately make automated suicide risk–prediction models useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders, including the frontline providers who will be using such tools, at each stage of the implementation process. Objective: The aim of this focus group study is to inform ongoing and future efforts to deploy suicide risk–prediction models in clinical practice. The specific goals are to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk–prediction models in practice; and identify barriers, facilitators, recommendations, and factors to consider. Methods: We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by 2 independent study staff members. All coded text was reviewed and discrepancies were resolved in consensus meetings with doctoral-level staff. Results: Although most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk–prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the health care system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider training. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings. Conclusions: Providers were dissatisfied with current suicide risk assessment methods and were open to the use of a machine learning–based risk-prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of these new approaches in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.
- Open Access
- 3 June 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Implementing Machine Learning Models for Suicide Risk Prediction in Clinical Practice: Focus Group Study With Hospital Providers (Preprint)
BACKGROUND Interest in developing machine learning models that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. However, whether and how such models might be implemented and useful in clinical practice remain unknown. To ultimately make automated suicide risk–prediction models useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders, including the frontline providers who will be using such tools, at each stage of the implementation process. OBJECTIVE The aim of this focus group study is to inform ongoing and future efforts to deploy suicide risk–prediction models in clinical practice. The specific goals are to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk–prediction models in practice; and identify barriers, facilitators, recommendations, and factors to consider. METHODS We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by 2 independent study staff members. All coded text was reviewed and discrepancies were resolved in consensus meetings with doctoral-level staff. RESULTS Although most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk–prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the health care system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider training. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings. CONCLUSIONS Providers were dissatisfied with current suicide risk assessment methods and were open to the use of a machine learning–based risk-prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of these new approaches in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.
- 3 June 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessImplementing Machine Learning Models for Suicide Risk Prediction in Clinical Practice: Focus Group Study With Hospital Providers (Preprint)
BACKGROUND Interest in developing machine learning models that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. However, whether and how such models might be implemented and useful in clinical practice remain unknown. To ultimately make automated suicide risk–prediction models useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders, including the frontline providers who will be using such tools, at each stage of the implementation process. OBJECTIVE The aim of this focus group study is to inform ongoing and future efforts to deploy suicide risk–prediction models in clinical practice. The specific goals are to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk–prediction models in practice; and identify barriers, facilitators, recommendations, and factors to consider. METHODS We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by 2 independent study staff members. All coded text was reviewed and discrepancies were resolved in consensus meetings with doctoral-level staff. RESULTS Although most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk–prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the health care system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider training. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings. CONCLUSIONS Providers were dissatisfied with current suicide risk assessment methods and were open to the use of a machine learning–based risk-prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of these new approaches in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.