A comparison between physicians and computer algorithms for form CMS‐2728 data reporting

Malas, Mohammed Said; Wish, Jay B.; Moorthi, Ranjani N.; Grannis, Shaun J.; Dexter, Paul; Duke, Jon; Moe, Sharon M.

doi:10.1111/hdi.12445

Cited by 7 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although NLP can process notes in an efficient manner, few studies in nephrology have utilized NLP and included symptoms or patient-centered outcomes. [9][10][11][12][13][14] Symptom prevalence identified in the BioMe cohort by NLP is similar to prior published survey data on symptoms. 2,8,15 Symptoms such as itching and cramps were less frequent, whereas other symptoms such as N/V were found more commonly.…”

Section: Discussionsupporting

confidence: 78%

Natural language processing of electronic health records is superior to billing codes to identify symptom burden in hemodialysis patients

Chan

Beers

Yau

et al. 2020

Kidney International

View full text Add to dashboard Cite

Section: Discussionsupporting

confidence: 78%

Natural language processing of electronic health records is superior to billing codes to identify symptom burden in hemodialysis patients

Chan

Beers

Yau

et al. 2020

Kidney International

View full text Add to dashboard Cite

“…These could serve as good future research questions. 7,8,10,38,39 The limitations of our study should be noted. First, in the simulation study, we only considered simple scenarios with limited configurations; for example, misclassifications and size constant across providers, non-differential ME, and 2 covariates.…”

Section: Discussionmentioning

confidence: 85%

Assessing the Impacts of Misclassified Case-Mix Factors on Health Care Provider Profiling: Performance of Dialysis Facilities

Chin

Kshirsagar

et al. 2020

INQUIRY

View full text Add to dashboard Cite

Quantitative metrics are used to develop profiles of health care institutions, including hospitals, nursing homes, and dialysis clinics. These profiles serve as measures of quality of care, which are used to compare institutions and determine reimbursement, as a part of a national effort led by the Center for Medicare and Medicaid Services in the United States. However, there is some concern about how misclassification in case-mix factors, which are typically accounted for in profiling, impacts results. We evaluated the potential effect of misclassification on profiling results, using 20 744 patients from 2740 dialysis facilities in the US Renal Data System. In this case study, we compared 30-day readmission as the profiling outcome measure, using comorbidity data from either the Center for Medicare and Medicaid Services Medical Evidence Report (error-prone) or Medicare claims (more accurate). Although the regression coefficient of the error-prone covariate demonstrated notable bias in simulation, the outcome measure—standardized readmission ratio—and profiling results were quite robust; for example, correlation coefficient of 0.99 in standardized readmission ratio estimates. Thus, we conclude that misclassification on case-mix did not meaningfully impact overall profiling results. We also identified both extreme degree of case-mix factor misclassification and magnitude of between-provider variability as 2 factors that can potentially exert enough influence on profile status to move a clinic from one performance category to another (eg, normal to worse performer).

show abstract

“…It is important to note that some of the differences observed in patient comorbidities between the US population and our Canadian population may have been due to the US study capturing comorbidities using the CMS 2728 Medical Evidence Report which has been found to undercapture comorbidities. 36 , 37 Fourth, mortality risk estimates assume that all treatment options are immediately available to a patient. However, we know that the average wait time for a deceased donor kidney transplant in Ontario is 5 years.…”

Section: Discussionmentioning

confidence: 99%

Predicting 3-Year Survival in Patients Receiving Maintenance Dialysis: An External Validation of iChoose Kidney in Ontario, Canada

Tan

Garg

McArthur

et al. 2018

Can J Kidney Health Dis

View full text Add to dashboard Cite

Background:Many patients with end-stage kidney disease (ESKD) do not appreciate how their survival may differ if treated with a kidney transplant compared with dialysis. A risk calculator (iChoose Kidney) developed and validated in the United States provides individualized mortality estimates for different treatment options (dialysis vs living or deceased donor kidney transplantation). The calculator can be used with patients and families to help patients make more educated treatment decisions.Objective:To validate the iChoose Kidney risk calculator in Ontario, Canada.Design:External validation study.Setting:We used several linked administrative health care databases from Ontario, Canada.Patients:We included 22 520 maintenance dialysis patients and 4505 kidney transplant recipients. Patients entered the cohort between 2004 and 2014.Measurements:Three-year all-cause mortality.Methods:We assessed model discrimination using the C-statistic. We assessed model calibration by comparing the observed versus predicted mortality risk and by using smoothed calibration plots. We used multivariable logistic regression modeling to recalibrate model intercepts using a correction factor, when appropriate.Results:In our final version of the iChoose Kidney model, we included the following variables: age (18-80 years), sex (male, female), race (white, black, other), time on dialysis (<6 months, 6-12 months, >12 months), and patient comorbidities (hypertension, diabetes, and/or cardiovascular disease). Over the 3-year follow-up period, 33.3% of dialysis patients and 6.2% of kidney transplant recipients died. The discriminatory ability was moderate (C-statistic for dialysis: 0.70, 95% confidence interval [CI]: 0.69-0.70, and C-statistic for transplant: 0.72, 95% CI: 0.69-0.75). The 3-year observed and predicted mortality estimates were comparable and even more so after we recalibrated the intercepts in 2 of our models (dialysis and deceased donor kidney transplantation). As done in the United States, we developed a Canadian Web site and an iOS application called Dialysis vs. Kidney Transplant- Estimated Survival in Ontario.Limitations:Missing data in our databases precluded the inclusion of all variables that were in the original iChoose Kidney (ie, patient ethnicity and low albumin). We were unable to perform all preplanned analyses due to the limited sample size.Conclusions:The original iChoose Kidney risk calculator was able to adequately predict mortality in this Canadian (Ontario) cohort of ESKD patients. After minor modifications, the predictive accuracy improved. The Dialysis vs. Kidney Transplant- Estimated Survival in Ontario risk calculator may be a valuable resource to help ESKD patients make an informed decision on pursuing kidney transplantation.

show abstract

A comparison between physicians and computer algorithms for form CMS‐2728 data reporting

Abstract: Using computer algorithms showed superior identification of comorbidities for form CMS-2728 and altered standardized mortality ratios. Adapting similar algorithms in available EMR systems may offer more thorough evaluation of comorbidities and improve quality reporting.

Cited by 7 publications

References 20 publications

Natural language processing of electronic health records is superior to billing codes to identify symptom burden in hemodialysis patients

Natural language processing of electronic health records is superior to billing codes to identify symptom burden in hemodialysis patients

Assessing the Impacts of Misclassified Case-Mix Factors on Health Care Provider Profiling: Performance of Dialysis Facilities

Predicting 3-Year Survival in Patients Receiving Maintenance Dialysis: An External Validation of iChoose Kidney in Ontario, Canada

Contact Info

Product

Resources

About