Jyotsna Bhatnagar, MD

Clinical question: Among patients discharged from the hospital, how do patient-reported outcome (PRO) measures change after discharge, and can they predict readmission or ED visit?

Background: Variables to predict 30-day rehospitalizations of discharged general medical patients have been looked into, but not many strategies have incorporated PRO measures in predictive models.

Study design: Longitudinal cohort study.

Setting: Patients discharged from an urban safety-net hospital that serves 128 municipalities in northeastern Illinois, including the city of Chicago.

Synopsis: One hundred ninety-six patients completed the initial survey; completion rates were 98%, 90%, and 88% for the 30-, 90-, and 180-day follow-up surveys, respectively. The Memorial Symptom Assessment Scale (MSAS) and the Patient Reported Outcomes Measurement Information System (PROMIS) Global Health short form assessing general self-rated health (GSRH), global physical health (GPH), and global mental health (GMH) were administered. In-hospital assessments of GMH and GSRH predicted 14-day reutilization, whereas post-hospitalization assessments of MSAS and GPH predicted subsequent utilizations. Notable limitations of the study include small sample size with high proportion of uninsured and racial/ethnic minorities and inability to count utilization at hospital(s) other than the hospital studied.

Bottom line: PRO measures are likely to be useful predictors in clinical medicine. More research is needed to improve the generalizability of PRO measures. Perhaps determination of specific measures of high predictive value may be more useful.

Citation: Hinami K, Smith J, Deamant CD, DuBeshter K, Trick WE. When do patient-reported outcome measures inform readmission risk? J Hosp Med. 2015;10(5):294-300.

Clinical question: Among patients discharged from the hospital, how do patient-reported outcome (PRO) measures change after discharge, and can they predict readmission or ED visit?

Background: Variables to predict 30-day rehospitalizations of discharged general medical patients have been looked into, but not many strategies have incorporated PRO measures in predictive models.

Study design: Longitudinal cohort study.

Setting: Patients discharged from an urban safety-net hospital that serves 128 municipalities in northeastern Illinois, including the city of Chicago.

Synopsis: One hundred ninety-six patients completed the initial survey; completion rates were 98%, 90%, and 88% for the 30-, 90-, and 180-day follow-up surveys, respectively. The Memorial Symptom Assessment Scale (MSAS) and the Patient Reported Outcomes Measurement Information System (PROMIS) Global Health short form assessing general self-rated health (GSRH), global physical health (GPH), and global mental health (GMH) were administered. In-hospital assessments of GMH and GSRH predicted 14-day reutilization, whereas post-hospitalization assessments of MSAS and GPH predicted subsequent utilizations. Notable limitations of the study include small sample size with high proportion of uninsured and racial/ethnic minorities and inability to count utilization at hospital(s) other than the hospital studied.

Bottom line: PRO measures are likely to be useful predictors in clinical medicine. More research is needed to improve the generalizability of PRO measures. Perhaps determination of specific measures of high predictive value may be more useful.

Citation: Hinami K, Smith J, Deamant CD, DuBeshter K, Trick WE. When do patient-reported outcome measures inform readmission risk? J Hosp Med. 2015;10(5):294-300.

Clinical question: Among patients discharged from the hospital, how do patient-reported outcome (PRO) measures change after discharge, and can they predict readmission or ED visit?

Background: Variables to predict 30-day rehospitalizations of discharged general medical patients have been looked into, but not many strategies have incorporated PRO measures in predictive models.

Study design: Longitudinal cohort study.

Setting: Patients discharged from an urban safety-net hospital that serves 128 municipalities in northeastern Illinois, including the city of Chicago.

Synopsis: One hundred ninety-six patients completed the initial survey; completion rates were 98%, 90%, and 88% for the 30-, 90-, and 180-day follow-up surveys, respectively. The Memorial Symptom Assessment Scale (MSAS) and the Patient Reported Outcomes Measurement Information System (PROMIS) Global Health short form assessing general self-rated health (GSRH), global physical health (GPH), and global mental health (GMH) were administered. In-hospital assessments of GMH and GSRH predicted 14-day reutilization, whereas post-hospitalization assessments of MSAS and GPH predicted subsequent utilizations. Notable limitations of the study include small sample size with high proportion of uninsured and racial/ethnic minorities and inability to count utilization at hospital(s) other than the hospital studied.

Bottom line: PRO measures are likely to be useful predictors in clinical medicine. More research is needed to improve the generalizability of PRO measures. Perhaps determination of specific measures of high predictive value may be more useful.

Citation: Hinami K, Smith J, Deamant CD, DuBeshter K, Trick WE. When do patient-reported outcome measures inform readmission risk? J Hosp Med. 2015;10(5):294-300.

Clinical question: Is there a standardized way to identify doctors at high risk of incurring repeated medicolegal events?

Background: Medicolegal agencies react to episodes of substandard care rather than intervening to prevent them due to lack of robust prediction tools at the individual practitioner level. Various studies have tried to predict complaints at the individual practitioner level accurately but had limited success.

Study design: Retrospective cohort study.

Setting: Commissions in all Australian states, except South Australia, with 70,200 practicing doctors.

Synopsis: Researchers used administrative data to analyze a national sample of 13,849 formal complaints, which were lodged by patients in Australia over a 12-year period against 8,424 doctors. Using multivariate logistic regression analysis, predictors for subsequent complaints within two years of an index complaint were estimated. These predictors were used in a simple predictive algorithm, the PRONE (Predicted Risk Of New Event), a score designed for application at the doctor level. PRONE is a 22-point scoring system that estimates a doctor’s future complaint risk based on specialty, sex, the number of previous complaints, and the time since the last complaint.

Because the scoring system has strong validity and reliability, regulators could harness such information to target quality improvement interventions and prevent substandard care and patient dissatisfaction.

Bottom line: The PRONE score appears to be a valid method for assessing individual doctors’ risks of attracting recurrent complaints.

Citation: Spittal MJ, Bismark MM, Studdert DM. The PRONE score: an algorithm for predicting doctors risks of formal patient complaints using routinely collected administrative data. BMJ Qual Saf. 2015;24(6):360-368.

Clinical question: Is there a standardized way to identify doctors at high risk of incurring repeated medicolegal events?

Background: Medicolegal agencies react to episodes of substandard care rather than intervening to prevent them due to lack of robust prediction tools at the individual practitioner level. Various studies have tried to predict complaints at the individual practitioner level accurately but had limited success.

Study design: Retrospective cohort study.

Setting: Commissions in all Australian states, except South Australia, with 70,200 practicing doctors.

Synopsis: Researchers used administrative data to analyze a national sample of 13,849 formal complaints, which were lodged by patients in Australia over a 12-year period against 8,424 doctors. Using multivariate logistic regression analysis, predictors for subsequent complaints within two years of an index complaint were estimated. These predictors were used in a simple predictive algorithm, the PRONE (Predicted Risk Of New Event), a score designed for application at the doctor level. PRONE is a 22-point scoring system that estimates a doctor’s future complaint risk based on specialty, sex, the number of previous complaints, and the time since the last complaint.

Because the scoring system has strong validity and reliability, regulators could harness such information to target quality improvement interventions and prevent substandard care and patient dissatisfaction.

Bottom line: The PRONE score appears to be a valid method for assessing individual doctors’ risks of attracting recurrent complaints.

Citation: Spittal MJ, Bismark MM, Studdert DM. The PRONE score: an algorithm for predicting doctors risks of formal patient complaints using routinely collected administrative data. BMJ Qual Saf. 2015;24(6):360-368.

Clinical question: Is there a standardized way to identify doctors at high risk of incurring repeated medicolegal events?

Background: Medicolegal agencies react to episodes of substandard care rather than intervening to prevent them due to lack of robust prediction tools at the individual practitioner level. Various studies have tried to predict complaints at the individual practitioner level accurately but had limited success.

Study design: Retrospective cohort study.

Setting: Commissions in all Australian states, except South Australia, with 70,200 practicing doctors.

Synopsis: Researchers used administrative data to analyze a national sample of 13,849 formal complaints, which were lodged by patients in Australia over a 12-year period against 8,424 doctors. Using multivariate logistic regression analysis, predictors for subsequent complaints within two years of an index complaint were estimated. These predictors were used in a simple predictive algorithm, the PRONE (Predicted Risk Of New Event), a score designed for application at the doctor level. PRONE is a 22-point scoring system that estimates a doctor’s future complaint risk based on specialty, sex, the number of previous complaints, and the time since the last complaint.

Because the scoring system has strong validity and reliability, regulators could harness such information to target quality improvement interventions and prevent substandard care and patient dissatisfaction.

Bottom line: The PRONE score appears to be a valid method for assessing individual doctors’ risks of attracting recurrent complaints.

Citation: Spittal MJ, Bismark MM, Studdert DM. The PRONE score: an algorithm for predicting doctors risks of formal patient complaints using routinely collected administrative data. BMJ Qual Saf. 2015;24(6):360-368.

Clinical question: Can novel configured rapid response teams (RRTs) improve non-ICU cardiopulmonary arrest (CPA) and overall hospital mortality rate?

Background: RRTs are primarily executed in hospital settings to avert non-ICU CPA through early detection and intervention. Prevailing evidence has not shown consistent clear benefit of RRTs in this regard.

Study design: A parallel-controlled, before-after design.

Setting: Two urban university hospitals with approximately 500 medical/surgical beds.

Synopsis: Researchers compared annual non-ICU CPA rates from two university hospitals with newly configured RRTs (implemented in November 2007) from July 2005 through June 2011 and found a decline in the incidence of non-ICU CPA to 1.1 from 2.7 per 1000 discharges (P<0.0001) while comparing pre- (2005/2006 to 2006/2007) to post- RRT implementation (2007-2011), respectively. Post-implementation, the overall hospital mortality dropped to 1.74% from 2.12% (P<0.001). With year-over-year, the RRT activation was found to be inversely related to Code Blue activations (r=-0.68, P<0.001), while the case mix index coefficients were still high.

The study lacks internal validation and may carry bias by including just one pre-implementation year (2006) data. It demonstrates that the rounding of unit manager (charge nurse) on “at risk” patients might avert decompensation; however, there was no determination of their decision-making process, with regard to RRT activation. No comparison was done with other RRT configurations.

Bottom line: Novel configured RRTs may improve non-ICU CPA and overall hospital mortality rate.

Citation: Davis DP, Aguilar SA, Graham PG, et al. A novel configuration of a traditional rapid response team decreases non-intensive care unit arrests and overall hospital mortality. J Hosp Med. 2015;10(6):352-357.

Clinical question: Can novel configured rapid response teams (RRTs) improve non-ICU cardiopulmonary arrest (CPA) and overall hospital mortality rate?

Background: RRTs are primarily executed in hospital settings to avert non-ICU CPA through early detection and intervention. Prevailing evidence has not shown consistent clear benefit of RRTs in this regard.

Study design: A parallel-controlled, before-after design.

Setting: Two urban university hospitals with approximately 500 medical/surgical beds.

Synopsis: Researchers compared annual non-ICU CPA rates from two university hospitals with newly configured RRTs (implemented in November 2007) from July 2005 through June 2011 and found a decline in the incidence of non-ICU CPA to 1.1 from 2.7 per 1000 discharges (P<0.0001) while comparing pre- (2005/2006 to 2006/2007) to post- RRT implementation (2007-2011), respectively. Post-implementation, the overall hospital mortality dropped to 1.74% from 2.12% (P<0.001). With year-over-year, the RRT activation was found to be inversely related to Code Blue activations (r=-0.68, P<0.001), while the case mix index coefficients were still high.

The study lacks internal validation and may carry bias by including just one pre-implementation year (2006) data. It demonstrates that the rounding of unit manager (charge nurse) on “at risk” patients might avert decompensation; however, there was no determination of their decision-making process, with regard to RRT activation. No comparison was done with other RRT configurations.

Bottom line: Novel configured RRTs may improve non-ICU CPA and overall hospital mortality rate.

Citation: Davis DP, Aguilar SA, Graham PG, et al. A novel configuration of a traditional rapid response team decreases non-intensive care unit arrests and overall hospital mortality. J Hosp Med. 2015;10(6):352-357.

Clinical question: Can novel configured rapid response teams (RRTs) improve non-ICU cardiopulmonary arrest (CPA) and overall hospital mortality rate?

Background: RRTs are primarily executed in hospital settings to avert non-ICU CPA through early detection and intervention. Prevailing evidence has not shown consistent clear benefit of RRTs in this regard.

Study design: A parallel-controlled, before-after design.

Setting: Two urban university hospitals with approximately 500 medical/surgical beds.

Synopsis: Researchers compared annual non-ICU CPA rates from two university hospitals with newly configured RRTs (implemented in November 2007) from July 2005 through June 2011 and found a decline in the incidence of non-ICU CPA to 1.1 from 2.7 per 1000 discharges (P<0.0001) while comparing pre- (2005/2006 to 2006/2007) to post- RRT implementation (2007-2011), respectively. Post-implementation, the overall hospital mortality dropped to 1.74% from 2.12% (P<0.001). With year-over-year, the RRT activation was found to be inversely related to Code Blue activations (r=-0.68, P<0.001), while the case mix index coefficients were still high.

The study lacks internal validation and may carry bias by including just one pre-implementation year (2006) data. It demonstrates that the rounding of unit manager (charge nurse) on “at risk” patients might avert decompensation; however, there was no determination of their decision-making process, with regard to RRT activation. No comparison was done with other RRT configurations.

Bottom line: Novel configured RRTs may improve non-ICU CPA and overall hospital mortality rate.

Citation: Davis DP, Aguilar SA, Graham PG, et al. A novel configuration of a traditional rapid response team decreases non-intensive care unit arrests and overall hospital mortality. J Hosp Med. 2015;10(6):352-357.