Akarsh Sharma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Akarsh Sharma has a total of 51 co-authors. Here are the top co-authors by number of shared publications:
- Annapoorna Kini
Icahn School of Medicine at Mount Sinai New York New York USA
3 shared publications - Samin K. Sharma
Icahn School of Medicine at Mount Sinai New York New York USA
3 shared publications - Stamatios Lerakis
Mount Sinai Fuster Heart Hospital, Icahn School of Medicine at Mount Sinai, New York, NY, USA
3 shared publications - Andreina Carbone
University of Naples Federico II, Naples, Italy
2 shared publications - Antonino Tuttolomondo
Internal Medicine and Stroke Care Ward, Department of Health Promotion, Mother and Child Care, Internal Medicine and Medical Specialities, University of Palermo, 90127 Palermo, Italy
2 shared publications - Carlo Mannina
Division of Cardiology
2 shared publications
Latest Publications
- Open Access
Mitral regurgitation trajectories after transcatheter aortic valve replacement are phenotype specific across low-flow aortic stenosis subtypes
Structured Objectives: Mitral regurgitation (MR) evolution after transcatheter aortic valve replacement (TAVR) in low-flow aortic stenosis (LFAS) is poorly characterized. We evaluated MR trajectories across LFAS phenotypes, predictors of MR worsening, and associations with clinical outcomes. Methods: We retrospectively studied 614 LFAS patients undergoing TAVR: low-flow high-gradient (LFHG; n=153, 24.9%), classical low-flow low-gradient (cLFLG; n=155, 25.2%), and paradoxical low-flow low-gradient (pLFLG; n=306, 49.8%). MR severity was abstracted from clinical echocardiography reports using a 6-level ordinal scale. MR worsening was defined as a ≥1-grade increase from baseline MR at ∼30 days or ∼1 year. Multivariable logistic models identified predictors of MR worsening. Kaplan–Meier and Cox models evaluated associations of MR trajectory and LFAS subtype with all-cause death, heart failure hospitalization (HFH), and their composite. Results: Among 614 LFAS patients, 443 had 30-day and 290 had 1-year echocardiographic follow up. At 30 days, MR trajectory differed significantly across LFAS phenotypes, with the highest rate of worsening in cLFLG and the lowest in LFHG. At 1 year, unadjusted MR trajectory distributions did not differ significantly across phenotypes. In adjusted logistic models, cLFLG remained independently associated with MR worsening at both timepoints. MR worsening was associated with worse unadjusted outcomes at 30 days but was not independently associated with the composite endpoint after multivariable adjustment. LFAS phenotype, particularly cLFLG, remained the dominant predictor of adverse clinical outcomes. Conclusions: MR evolution after TAVR is phenotype-specific: cLFLG patients have the highest risk of MR worsening and lowest event-free survival, supporting phenotype-informed post-TAVR surveillance.
Open AccessMitral regurgitation trajectories after transcatheter aortic valve replacement are phenotype specific across low-flow aortic stenosis subtypes
Structured Objectives: Mitral regurgitation (MR) evolution after transcatheter aortic valve replacement (TAVR) in low-flow aortic stenosis (LFAS) is poorly characterized. We evaluated MR trajectories across LFAS phenotypes, predictors of MR worsening, and associations with clinical outcomes. Methods: We retrospectively studied 614 LFAS patients undergoing TAVR: low-flow high-gradient (LFHG; n=153, 24.9%), classical low-flow low-gradient (cLFLG; n=155, 25.2%), and paradoxical low-flow low-gradient (pLFLG; n=306, 49.8%). MR severity was abstracted from clinical echocardiography reports using a 6-level ordinal scale. MR worsening was defined as a ≥1-grade increase from baseline MR at ∼30 days or ∼1 year. Multivariable logistic models identified predictors of MR worsening. Kaplan–Meier and Cox models evaluated associations of MR trajectory and LFAS subtype with all-cause death, heart failure hospitalization (HFH), and their composite. Results: Among 614 LFAS patients, 443 had 30-day and 290 had 1-year echocardiographic follow up. At 30 days, MR trajectory differed significantly across LFAS phenotypes, with the highest rate of worsening in cLFLG and the lowest in LFHG. At 1 year, unadjusted MR trajectory distributions did not differ significantly across phenotypes. In adjusted logistic models, cLFLG remained independently associated with MR worsening at both timepoints. MR worsening was associated with worse unadjusted outcomes at 30 days but was not independently associated with the composite endpoint after multivariable adjustment. LFAS phenotype, particularly cLFLG, remained the dominant predictor of adverse clinical outcomes. Conclusions: MR evolution after TAVR is phenotype-specific: cLFLG patients have the highest risk of MR worsening and lowest event-free survival, supporting phenotype-informed post-TAVR surveillance.
- Open Access
- 5 August 2025
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 14 (15) , e042016
Impact of Atrial Fibrillation in Patients With Severe Mitral Regurgitation Undergoing Transcatheter Edge‐to‐Edge Repair
Background: Atrial fibrillation (AF) and mitral regurgitation (MR) are closely linked, and one may worsen the other. We investigated the impact of baseline AF in patients with MR undergoing transcatheter edge‐to‐edge repair. Methods: One‐hundred‐fifty‐six consecutive patients with symptomatic heart failure (HF) undergoing mitral valve transcatheter edge‐to‐edge repair for severe MR were studied. The primary end point was the composite outcome of death or HF hospitalization. Transthoracic echocardiograms were performed at baseline and follow‐up. Results: Mean age was 80.8±8.8 years and 82 (52.6%) patients were female. MR cause was primary in 69 (44.2%) and secondary in 87 (55.8%) patients. AF or atrial flutter was present in 59 (37.8%) patients at baseline. Mitral valve transcatheter edge‐to‐edge repair was successful (≤2+ MR) in 58 (98.3%) and 94 (96.9%) patients with and without AF respectively ( P =0.59). During median 12.5 months follow‐up, the primary end point occurred in 64 patients (2‐year Kaplan–Meier estimated rate 41.0%), including death in 16 patients (10.3%) and HF hospitalization in 57 patients (36.5%). Baseline AF remained a significant independent predictor of death or HF hospitalization (adjusted hazard ratio, 2.03 [95% CI, 1.12–3.69], P =0.02). Left ventricular end‐diastolic volume, left atrial volume, and right ventricular systolic pressure decreased during follow‐up among patients in sinus rhythm but not among those in AF. AF was associated with an increased risk of severe MR recurrence (18.6% versus 8.2%, P =0.05). Conclusions: In patients with HF and severe MR treated with Mitral valve transcatheter edge‐to‐edge repair, baseline AF was associated with impaired right and left heart remodeling, more frequent MR recurrence, and more than doubling of the 2‐year risk of death or HF hospitalization.
- 5 August 2025
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 14 (15) , e042016
Open AccessImpact of Atrial Fibrillation in Patients With Severe Mitral Regurgitation Undergoing Transcatheter Edge‐to‐Edge Repair
Background: Atrial fibrillation (AF) and mitral regurgitation (MR) are closely linked, and one may worsen the other. We investigated the impact of baseline AF in patients with MR undergoing transcatheter edge‐to‐edge repair. Methods: One‐hundred‐fifty‐six consecutive patients with symptomatic heart failure (HF) undergoing mitral valve transcatheter edge‐to‐edge repair for severe MR were studied. The primary end point was the composite outcome of death or HF hospitalization. Transthoracic echocardiograms were performed at baseline and follow‐up. Results: Mean age was 80.8±8.8 years and 82 (52.6%) patients were female. MR cause was primary in 69 (44.2%) and secondary in 87 (55.8%) patients. AF or atrial flutter was present in 59 (37.8%) patients at baseline. Mitral valve transcatheter edge‐to‐edge repair was successful (≤2+ MR) in 58 (98.3%) and 94 (96.9%) patients with and without AF respectively ( P =0.59). During median 12.5 months follow‐up, the primary end point occurred in 64 patients (2‐year Kaplan–Meier estimated rate 41.0%), including death in 16 patients (10.3%) and HF hospitalization in 57 patients (36.5%). Baseline AF remained a significant independent predictor of death or HF hospitalization (adjusted hazard ratio, 2.03 [95% CI, 1.12–3.69], P =0.02). Left ventricular end‐diastolic volume, left atrial volume, and right ventricular systolic pressure decreased during follow‐up among patients in sinus rhythm but not among those in AF. AF was associated with an increased risk of severe MR recurrence (18.6% versus 8.2%, P =0.05). Conclusions: In patients with HF and severe MR treated with Mitral valve transcatheter edge‐to‐edge repair, baseline AF was associated with impaired right and left heart remodeling, more frequent MR recurrence, and more than doubling of the 2‐year risk of death or HF hospitalization.
- Open Access
- 19 November 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (22) , e036539
Association of Systemic Inflammatory Response Syndrome With Cardiovascular Events After Mitral Transcatheter Edge‐to‐Edge Repair
Background: Systemic inflammatory response syndrome (SIRS) following cardiovascular interventions is associated with adverse events during hospitalization and follow‐up. Mitral transcatheter edge‐to‐edge repair is increasingly utilized for treatment of mitral regurgitation (MR). We investigated whether SIRS following mitral transcatheter edge‐to‐edge repair may occur and be associated with adverse clinical outcomes. Methods and Results: A total of 158 consecutive patients with severe MR undergoing mitral transcatheter edge‐to‐edge repair were studied. SIRS was defined by leukocytosis (≥12 × 10 9 /L) and fever (≥38 °C) within 48 hours after intervention. Baseline inflammation was measured by absolute neutrophil and lymphocyte counts and neutrophil‐lymphocyte ratio. The primary end point of major cardiovascular events was the composite of nonfatal myocardial infarction, nonfatal stroke, and all‐cause death. Recurrent MR at follow‐up was also recorded. The mean patient age was 80.8±8.8 years. Forty‐four (27.9%) developed SIRS. Neutrophil‐lymphocyte ratio correlated with onset of leukocytosis and fever ( P =0.04). During a median follow‐up of 12.5 (5.4–17.4) months, the primary end point occurred in 27 (17.1%) patients (6 myocardial infarction, 5 strokes, and 16 deaths). Patients with SIRS more often had severe MR (79.5% versus 62.7%, P =0.02) at follow‐up. After adjustment for pertinent variables, SIRS (HR 2.73 [95% CI, 1.08–6.86]; P =0.03) was independently associated with major cardiovascular events. Conclusions: SIRS after mitral transcatheter edge‐to‐edge repair is a strong independent predictor of major cardiovascular events. Closer follow‐up is warranted because patients with SIRS have more severe MR at follow‐up.
- 19 November 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (22) , e036539
Open AccessAssociation of Systemic Inflammatory Response Syndrome With Cardiovascular Events After Mitral Transcatheter Edge‐to‐Edge Repair
Background: Systemic inflammatory response syndrome (SIRS) following cardiovascular interventions is associated with adverse events during hospitalization and follow‐up. Mitral transcatheter edge‐to‐edge repair is increasingly utilized for treatment of mitral regurgitation (MR). We investigated whether SIRS following mitral transcatheter edge‐to‐edge repair may occur and be associated with adverse clinical outcomes. Methods and Results: A total of 158 consecutive patients with severe MR undergoing mitral transcatheter edge‐to‐edge repair were studied. SIRS was defined by leukocytosis (≥12 × 10 9 /L) and fever (≥38 °C) within 48 hours after intervention. Baseline inflammation was measured by absolute neutrophil and lymphocyte counts and neutrophil‐lymphocyte ratio. The primary end point of major cardiovascular events was the composite of nonfatal myocardial infarction, nonfatal stroke, and all‐cause death. Recurrent MR at follow‐up was also recorded. The mean patient age was 80.8±8.8 years. Forty‐four (27.9%) developed SIRS. Neutrophil‐lymphocyte ratio correlated with onset of leukocytosis and fever ( P =0.04). During a median follow‐up of 12.5 (5.4–17.4) months, the primary end point occurred in 27 (17.1%) patients (6 myocardial infarction, 5 strokes, and 16 deaths). Patients with SIRS more often had severe MR (79.5% versus 62.7%, P =0.02) at follow‐up. After adjustment for pertinent variables, SIRS (HR 2.73 [95% CI, 1.08–6.86]; P =0.03) was independently associated with major cardiovascular events. Conclusions: SIRS after mitral transcatheter edge‐to‐edge repair is a strong independent predictor of major cardiovascular events. Closer follow‐up is warranted because patients with SIRS have more severe MR at follow‐up.
- 4 March 2022
- Published by Oxford University Press (OUP) in Cerebral Cortex
- Vol. 33 (3) , 597-611
Altered prefrontal signaling during inhibitory control in a salient drug context in cocaine use disorder
Drug addiction is characterized by impaired response inhibition and salience attribution (iRISA), where the salience of drug cues is postulated to overpower that of other reinforcers with a concomitant decrease in self-control. However, the neural underpinnings of the interaction between the salience of drug cues and inhibitory control in drug addiction remain unclear. We developed a novel stop-signal functional magnetic resonance imaging task where the stop-signal reaction time (SSRT–a classical inhibitory control measure) was tested under different salience conditions (modulated by drug, food, threat, or neutral words) in individuals with cocaine use disorder (CUD; n = 26) versus demographically matched healthy control participants (n = 26). Despite similarities in drug cue-related SSRT and valence and arousal word ratings between groups, dorsolateral prefrontal cortex (dlPFC) activity was diminished during the successful inhibition of drug versus food cues in CUD and was correlated with lower frequency of recent use, lower craving, and longer abstinence (Z > 3.1, P < 0.05 corrected). Results suggest altered involvement of cognitive control regions (e.g. dlPFC) during inhibitory control under a drug context, relative to an alternative reinforcer, in CUD. Supporting the iRISA model, these results elucidate the direct impact of drug-related cue reactivity on the neural signature of inhibitory control in drug addiction.
- 4 March 2022
- Published by Oxford University Press (OUP) in Cerebral Cortex
- Vol. 33 (3) , 597-611
Altered prefrontal signaling during inhibitory control in a salient drug context in cocaine use disorder
Drug addiction is characterized by impaired response inhibition and salience attribution (iRISA), where the salience of drug cues is postulated to overpower that of other reinforcers with a concomitant decrease in self-control. However, the neural underpinnings of the interaction between the salience of drug cues and inhibitory control in drug addiction remain unclear. We developed a novel stop-signal functional magnetic resonance imaging task where the stop-signal reaction time (SSRT–a classical inhibitory control measure) was tested under different salience conditions (modulated by drug, food, threat, or neutral words) in individuals with cocaine use disorder (CUD; n = 26) versus demographically matched healthy control participants (n = 26). Despite similarities in drug cue-related SSRT and valence and arousal word ratings between groups, dorsolateral prefrontal cortex (dlPFC) activity was diminished during the successful inhibition of drug versus food cues in CUD and was correlated with lower frequency of recent use, lower craving, and longer abstinence (Z > 3.1, P < 0.05 corrected). Results suggest altered involvement of cognitive control regions (e.g. dlPFC) during inhibitory control under a drug context, relative to an alternative reinforcer, in CUD. Supporting the iRISA model, these results elucidate the direct impact of drug-related cue reactivity on the neural signature of inhibitory control in drug addiction.
- 1 January 2022
- Published by SAGE Publications in Substance Abuse
- Vol. 43 (1) , 564-572
Leveraging Technology to Address Unhealthy Drug use in Primary Care: Effectiveness of the Substance use Screening and Intervention Tool (SUSIT)
Background: Screening for unhealthy drug use is now recommended for adult primary care patients, but primary care providers (PCPs) generally lack the time and knowledge required to screen and deliver an intervention during the medical visit. To address these barriers, we developed a tablet computer-based ‘Substance Use Screening and Intervention Tool (SUSIT)’. Using the SUSIT, patients self-administer screening questionnaires prior to the medical visit, and results are presented to the PCP at the point of care, paired with clinical decision support (CDS) that guides them in providing a brief intervention (BI) for unhealthy drug use. Methods: PCPs and their patients with moderate-risk drug use were recruited from primary care and HIV clinics. A pre-post design compared a control ‘screening only’ (SO) period to an intervention ‘SUSIT’ period. Unique patients were enrolled in each period. In both conditions, patients completed screening and identified their drug of most concern (DOMC) before the visit, and completed a questionnaire about BI delivery by the PCP after the visit. In the SUSIT condition only, PCPs received the tablet with the patient’s screening results and CDS. Multilevel models with random intercepts and patients nested within PCPs examined the effect of the SUSIT intervention on PCP delivery of BI. Results: 20 PCPs and 79 patients (42 SO, 37 SUSIT) participated. Most patients had moderate-risk marijuana use (92.4%), and selected marijuana as the DOMC (68.4%). Moderate-risk use of drugs other than marijuana included cocaine (15.2%), hallucinogens (12.7%), and sedatives (12.7%). Compared to the SO condition, patients in SUSIT had higher odds of receiving any BI for drug use, with an adjusted odds ratio of 11.59 (95% confidence interval: 3.39, 39.25), and received more elements of BI for drug use. Conclusions: The SUSIT significantly increased delivery of BI for drug use by PCPs during routine primary care encounters.
- 1 January 2022
- Published by SAGE Publications in Substance Abuse
- Vol. 43 (1) , 564-572
Leveraging Technology to Address Unhealthy Drug use in Primary Care: Effectiveness of the Substance use Screening and Intervention Tool (SUSIT)
Background: Screening for unhealthy drug use is now recommended for adult primary care patients, but primary care providers (PCPs) generally lack the time and knowledge required to screen and deliver an intervention during the medical visit. To address these barriers, we developed a tablet computer-based ‘Substance Use Screening and Intervention Tool (SUSIT)’. Using the SUSIT, patients self-administer screening questionnaires prior to the medical visit, and results are presented to the PCP at the point of care, paired with clinical decision support (CDS) that guides them in providing a brief intervention (BI) for unhealthy drug use. Methods: PCPs and their patients with moderate-risk drug use were recruited from primary care and HIV clinics. A pre-post design compared a control ‘screening only’ (SO) period to an intervention ‘SUSIT’ period. Unique patients were enrolled in each period. In both conditions, patients completed screening and identified their drug of most concern (DOMC) before the visit, and completed a questionnaire about BI delivery by the PCP after the visit. In the SUSIT condition only, PCPs received the tablet with the patient’s screening results and CDS. Multilevel models with random intercepts and patients nested within PCPs examined the effect of the SUSIT intervention on PCP delivery of BI. Results: 20 PCPs and 79 patients (42 SO, 37 SUSIT) participated. Most patients had moderate-risk marijuana use (92.4%), and selected marijuana as the DOMC (68.4%). Moderate-risk use of drugs other than marijuana included cocaine (15.2%), hallucinogens (12.7%), and sedatives (12.7%). Compared to the SO condition, patients in SUSIT had higher odds of receiving any BI for drug use, with an adjusted odds ratio of 11.59 (95% confidence interval: 3.39, 39.25), and received more elements of BI for drug use. Conclusions: The SUSIT significantly increased delivery of BI for drug use by PCPs during routine primary care encounters.
- 1 December 2021
- Published by Elsevier in Preventive Medicine
- Vol. 153, 106761
Hypertension, diabetes, and corresponding annual clinical testing utilization: Comparison between Asian Indians and other races/ethnicities
- 1 December 2021
- Published by Elsevier in Preventive Medicine
- Vol. 153, 106761
Hypertension, diabetes, and corresponding annual clinical testing utilization: Comparison between Asian Indians and other races/ethnicities
- Open Access
- 1 September 2021
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 9 (9) , e21990
Forecasting the Requirement for Nonelective Hospital Beds in the National Health Service of the United Kingdom: Model Development Study
Background: Over the last decade, increasing numbers of emergency department attendances and an even greater increase in emergency admissions have placed severe strain on the bed capacity of the National Health Service (NHS) of the United Kingdom. The result has been overcrowded emergency departments with patients experiencing long wait times for admission to an appropriate hospital bed. Nevertheless, scheduling issues can still result in significant underutilization of bed capacity. Bed occupancy rates may not correlate well with bed availability. More accurate and reliable long-term prediction of bed requirements will help anticipate the future needs of a hospital's catchment population, thus resulting in greater efficiencies and better patient care. Objective: This study aimed to evaluate widely used automated time-series forecasting techniques to predict short-term daily nonelective bed occupancy at all trusts in the NHS. These techniques were used to develop a simple yet accurate national health system-level forecasting framework that can be utilized at a low cost and by health care administrators who do not have statistical modeling expertise. Methods: Bed occupancy models that accounted for patterns in occupancy were created for each trust in the NHS. Daily nonelective midnight trust occupancy data from April 2011 to March 2017 for 121 NHS trusts were utilized to generate these models. Forecasts were generated using the three most widely used automated forecasting techniques: exponential smoothing; Seasonal Autoregressive Integrated Moving Average; and Trigonometric, Box-Cox transform, autoregressive moving average errors, and Trend and Seasonal components. The NHS Modernisation Agency's recommended forecasting method prior to 2020 was also replicated. Results: The accuracy of the models varied on the basis of the season during which occupancy was forecasted. For the summer season, percent root-mean-square error values for each model remained relatively stable across the 6 forecasted weeks. However, only the trend and seasonal components model (median error=2.45% for 6 weeks) outperformed the NHS Modernisation Agency's recommended method (median error=2.63% for 6 weeks). In contrast, during the winter season, the percent root-mean-square error values increased as we forecasted further into the future. Exponential smoothing generated the most accurate forecasts median error=4.91% over 4 weeks), but all models outperformed the NHS Modernisation Agency's recommended method prior to 2020 (median error=8.5% over 4 weeks). Conclusions: It is possible to create automated models, similar to those recently published by the NHS, which can be used at a hospital level for a large national health care system to predict nonelective bed admissions and thus schedule elective procedures.
- 1 September 2021
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 9 (9) , e21990
Open AccessForecasting the Requirement for Nonelective Hospital Beds in the National Health Service of the United Kingdom: Model Development Study
Background: Over the last decade, increasing numbers of emergency department attendances and an even greater increase in emergency admissions have placed severe strain on the bed capacity of the National Health Service (NHS) of the United Kingdom. The result has been overcrowded emergency departments with patients experiencing long wait times for admission to an appropriate hospital bed. Nevertheless, scheduling issues can still result in significant underutilization of bed capacity. Bed occupancy rates may not correlate well with bed availability. More accurate and reliable long-term prediction of bed requirements will help anticipate the future needs of a hospital's catchment population, thus resulting in greater efficiencies and better patient care. Objective: This study aimed to evaluate widely used automated time-series forecasting techniques to predict short-term daily nonelective bed occupancy at all trusts in the NHS. These techniques were used to develop a simple yet accurate national health system-level forecasting framework that can be utilized at a low cost and by health care administrators who do not have statistical modeling expertise. Methods: Bed occupancy models that accounted for patterns in occupancy were created for each trust in the NHS. Daily nonelective midnight trust occupancy data from April 2011 to March 2017 for 121 NHS trusts were utilized to generate these models. Forecasts were generated using the three most widely used automated forecasting techniques: exponential smoothing; Seasonal Autoregressive Integrated Moving Average; and Trigonometric, Box-Cox transform, autoregressive moving average errors, and Trend and Seasonal components. The NHS Modernisation Agency's recommended forecasting method prior to 2020 was also replicated. Results: The accuracy of the models varied on the basis of the season during which occupancy was forecasted. For the summer season, percent root-mean-square error values for each model remained relatively stable across the 6 forecasted weeks. However, only the trend and seasonal components model (median error=2.45% for 6 weeks) outperformed the NHS Modernisation Agency's recommended method (median error=2.63% for 6 weeks). In contrast, during the winter season, the percent root-mean-square error values increased as we forecasted further into the future. Exponential smoothing generated the most accurate forecasts median error=4.91% over 4 weeks), but all models outperformed the NHS Modernisation Agency's recommended method prior to 2020 (median error=8.5% over 4 weeks). Conclusions: It is possible to create automated models, similar to those recently published by the NHS, which can be used at a hospital level for a large national health care system to predict nonelective bed admissions and thus schedule elective procedures.
- Open Access
- 16 June 2021
- Published by Levy Library Press in ISMMS Journal of Science and Medicine
- Vol. 1 (2)
Seasonal Variation in Presentation and Time to Revascularization Therapy for Acute Ischemic Stroke
- 16 June 2021
- Published by Levy Library Press in ISMMS Journal of Science and Medicine
- Vol. 1 (2)
Open AccessSeasonal Variation in Presentation and Time to Revascularization Therapy for Acute Ischemic Stroke
- Open Access
- 30 June 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Forecasting the Requirement for Nonelective Hospital Beds in the National Health Service of the United Kingdom: Model Development Study (Preprint)
BACKGROUND Over the last decade, increasing numbers of emergency department attendances and an even greater increase in emergency admissions have placed severe strain on the bed capacity of the National Health Service (NHS) of the United Kingdom. The result has been overcrowded emergency departments with patients experiencing long wait times for admission to an appropriate hospital bed. Nevertheless, scheduling issues can still result in significant underutilization of bed capacity. Bed occupancy rates may not correlate well with bed availability. More accurate and reliable long-term prediction of bed requirements will help anticipate the future needs of a hospital’s catchment population, thus resulting in greater efficiencies and better patient care. OBJECTIVE This study aimed to evaluate widely used automated time-series forecasting techniques to predict short-term daily nonelective bed occupancy at all trusts in the NHS. These techniques were used to develop a simple yet accurate national health system–level forecasting framework that can be utilized at a low cost and by health care administrators who do not have statistical modeling expertise. METHODS Bed occupancy models that accounted for patterns in occupancy were created for each trust in the NHS. Daily nonelective midnight trust occupancy data from April 2011 to March 2017 for 121 NHS trusts were utilized to generate these models. Forecasts were generated using the three most widely used automated forecasting techniques: exponential smoothing; Seasonal Autoregressive Integrated Moving Average; and Trigonometric, Box-Cox transform, autoregressive moving average errors, and Trend and Seasonal components. The NHS Modernisation Agency’s recommended forecasting method prior to 2020 was also replicated. RESULTS The accuracy of the models varied on the basis of the season during which occupancy was forecasted. For the summer season, percent root-mean-square error values for each model remained relatively stable across the 6 forecasted weeks. However, only the trend and seasonal components model (median error=2.45% for 6 weeks) outperformed the NHS Modernisation Agency’s recommended method (median error=2.63% for 6 weeks). In contrast, during the winter season, the percent root-mean-square error values increased as we forecasted further into the future. Exponential smoothing generated the most accurate forecasts (median error=4.91% over 4 weeks), but all models outperformed the NHS Modernisation Agency’s recommended method prior to 2020 (median error=8.5% over 4 weeks). CONCLUSIONS It is possible to create automated models, similar to those recently published by the NHS, which can be used at a hospital level for a large national health care system to predict nonelective bed admissions and thus schedule elective procedures.
- 30 June 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessForecasting the Requirement for Nonelective Hospital Beds in the National Health Service of the United Kingdom: Model Development Study (Preprint)
BACKGROUND Over the last decade, increasing numbers of emergency department attendances and an even greater increase in emergency admissions have placed severe strain on the bed capacity of the National Health Service (NHS) of the United Kingdom. The result has been overcrowded emergency departments with patients experiencing long wait times for admission to an appropriate hospital bed. Nevertheless, scheduling issues can still result in significant underutilization of bed capacity. Bed occupancy rates may not correlate well with bed availability. More accurate and reliable long-term prediction of bed requirements will help anticipate the future needs of a hospital’s catchment population, thus resulting in greater efficiencies and better patient care. OBJECTIVE This study aimed to evaluate widely used automated time-series forecasting techniques to predict short-term daily nonelective bed occupancy at all trusts in the NHS. These techniques were used to develop a simple yet accurate national health system–level forecasting framework that can be utilized at a low cost and by health care administrators who do not have statistical modeling expertise. METHODS Bed occupancy models that accounted for patterns in occupancy were created for each trust in the NHS. Daily nonelective midnight trust occupancy data from April 2011 to March 2017 for 121 NHS trusts were utilized to generate these models. Forecasts were generated using the three most widely used automated forecasting techniques: exponential smoothing; Seasonal Autoregressive Integrated Moving Average; and Trigonometric, Box-Cox transform, autoregressive moving average errors, and Trend and Seasonal components. The NHS Modernisation Agency’s recommended forecasting method prior to 2020 was also replicated. RESULTS The accuracy of the models varied on the basis of the season during which occupancy was forecasted. For the summer season, percent root-mean-square error values for each model remained relatively stable across the 6 forecasted weeks. However, only the trend and seasonal components model (median error=2.45% for 6 weeks) outperformed the NHS Modernisation Agency’s recommended method (median error=2.63% for 6 weeks). In contrast, during the winter season, the percent root-mean-square error values increased as we forecasted further into the future. Exponential smoothing generated the most accurate forecasts (median error=4.91% over 4 weeks), but all models outperformed the NHS Modernisation Agency’s recommended method prior to 2020 (median error=8.5% over 4 weeks). CONCLUSIONS It is possible to create automated models, similar to those recently published by the NHS, which can be used at a hospital level for a large national health care system to predict nonelective bed admissions and thus schedule elective procedures.