Validating an algorithm for multiple myeloma based on administrative data using a SEER tumor registry and medical record review

SUMMARYThe Charlson comorbidity index (CCI) is widely used for control of confounding from comorbidities in epidemiological studies. International Classification of Diseases (ICD)-coded diagnoses from administrative hospital databases is potentially an efficient way of deriving CCI. However, no studies have evaluated its validity in infectious disease research. We aim to compare CCI derived from administrative data and medical record review in predicting mortality in patients with infections. We conducted a cross-sectional study on 199 inpatients. Correlation analyses were used to compare comorbidity scores from ICD-coded administrative databases and medical record review. Multivariable regression models were constructed and compared for discriminatory power for 30-day in-hospital mortality. Overall agreement was fair [weighted kappa 0·33, 95% confidence interval (CI) 0·23–0·43]. Kappa coefficient ranged from 0·17 (95% CI 0·01–0·36) for myocardial infarction to 0·85 (95% CI 0·59–1·00) for connective tissue disease. Administrative data-derived CCI was predictive of CCI ⩾5 from medical record review, controlling for age, gender, resident status, ward class, clinical speciality, illness severity, and infection source (C = 0·773). Using the multivariable model comprising age, gender, resident status, ward class, clinical speciality, illness severity, and infection source to predict 30-day in-hospital mortality, administrative data-derived CCI (C = 0·729) provided a similar C statistic as medical record review (C = 0·717, P = 0·8548). In conclusion, administrative data-derived CCI can be used for assessing comorbidities and confounding control in infectious disease research.

Download Full-text

Epidemiology of Kawasaki Disease in Canada 2004 to 2014: Comparison of Surveillance Using Administrative Data vs Periodic Medical Record Review

Canadian Journal of Cardiology ◽

10.1016/j.cjca.2017.12.009 ◽

2018 ◽

Vol 34 (3) ◽

pp. 303-309 ◽

Cited By ~ 17

Author(s):

Cedric Manlhiot ◽

Sunita O'Shea ◽

Bailey Bernknopf ◽

Michael LaBelle ◽

Nita Chahal ◽

...

Keyword(s):

Kawasaki Disease ◽

Medical Record ◽

Administrative Data ◽

Medical Record Review ◽

Record Review

Download Full-text

Tumor Registry Versus Physician Medical Record Review: A Direct Comparison of Patients With Pancreatic Neuroendocrine Tumors

Journal of Oncology Practice ◽

10.1200/jop.2010.000097 ◽

2011 ◽

Vol 7 (2) ◽

pp. 111-116 ◽

Cited By ~ 6

Author(s):

Elisabet E. Manasanch ◽

Jillian K. Smith ◽

Andreea Bodnari ◽

Jeannine McKinney ◽

Catherine Gray ◽

...

Keyword(s):

Medical Record ◽

Neuroendocrine Tumors ◽

Time Lag ◽

Case Finding ◽

Medical Record Review ◽

Pancreatic Neuroendocrine Tumors ◽

Tumor Registry ◽

Record Review ◽

Registry Analysis

Academic tumor registry analysis indicates many patients with pancreatic neuroendocrine tumors are not identified when compared with physician medical record review. Reasons include registry time lag and case-finding methodologies.

Download Full-text

Detection of Pelvic Inflammatory Disease: Development of an Automated Case-Finding Algorithm Using Administrative Data

Infectious Diseases in Obstetrics and Gynecology ◽

10.1155/2011/428351 ◽

2011 ◽

Vol 2011 ◽

pp. 1-7 ◽

Cited By ~ 8

Author(s):

Catherine L. Satterwhite ◽

Onchee Yu ◽

Marsha A. Raebel ◽

Stuart Berman ◽

Penelope P. Howards ◽

...

Keyword(s):

Pelvic Inflammatory Disease ◽

Medical Record ◽

Administrative Data ◽

Inflammatory Disease ◽

Regression Tree ◽

Case Finding ◽

Medical Record Review ◽

Classification And Regression Tree ◽

Case Identification ◽

Record Review

ICD-9 codes are conventionally used to identify pelvic inflammatory disease (PID) from administrative data for surveillance purposes. This approach may include non-PID cases. To refine PID case identification among women with ICD-9 codes suggestive of PID, a case-finding algorithm was developed using additional variables. Potential PID cases were identified among women aged 15–44 years at Group Health (GH) and Kaiser Permanente Colorado (KPCO) and verified by medical record review. A classification and regression tree analysis was used to develop the algorithm at GH; validation occurred at KPCO. The positive predictive value (PPV) for using ICD-9 codes alone to identify clinical PID cases was 79%. The algorithm identified PID appropriate treatment and age 15–25 years as predictors. Algorithm sensitivity (GH=96.4%;KPCO=90.3%) and PPV (GH=86.9%;KPCO=84.5%) were high, but specificity was poor (GH=45.9%;KPCO=37.0%). In GH, the algorithm offered a practical alternative to medical record review to further improve PID case identification.

Download Full-text

Comparison of the Use of Administrative Data and an Active System for Surveillance of Invasive Aspergillosis

Infection Control and Hospital Epidemiology ◽

10.1086/524324 ◽

2008 ◽

Vol 29 (1) ◽

pp. 25-30 ◽

Cited By ~ 30

Author(s):

Douglas C. Chang ◽

Lauren A. Burwell ◽

G. Marshall Lyon ◽

Peter G. Pappas ◽

Tom M. Chiller ◽

...

Keyword(s):

Invasive Aspergillosis ◽

Quality Assessment ◽

Medical Record ◽

Administrative Data ◽

Surveillance System ◽

Medical Record Review ◽

Solid Organ ◽

Cell Transplant ◽

Transplant Recipients ◽

Record Review

Background.Administrative data, such as International Classification of Diseases, Ninth Revision (ICD-9) codes, are readily available and are an attractive option for surveillance and quality assessment within a single institution or for interinstitutional comparisons. To understand the usefulness of administrative data for the surveillance of invasive aspergillosis, we compared information obtained from a system based on ICD-9 codes with information obtained from an active, prospective surveillance system, which used more extensive case-finding methods (Transplant Associated Infection Surveillance Network).Methods.Patients with suspected inyasive aspergillosis were identified by aspergillosis-related ICD-9 codes assigned to hematopoietic stem cell transplant recipients and solid organ transplant recipients at a single hospital from April 1, 2001, through January 31, 2005. Suspected cases were classified as proven or probable invasive aspergillosis by medical record review using standard definitions. We calculated the sensitivity and positive predictive value (PPV) of identifying invasive aspergillosis by individual ICD-9 codes and by combinations of codes.Results.The sensitivity of code 117.3 was modest (63% [95% confidence interval {CI}, 38%-84%]), as was the PPV (71% [95% CI, 44%-90%]); the sensitivity of code 117.9 was poor (32% [95% CI, 13%-57%]), as was the PPV (15% [95% CI, 6%-31%]). The sensitivity of codes 117.3 and 117.9 combined was 84% (95% CI, 60%-97%); the PPV of the combined codes was 30% (95% CI, 18%-44%). Overall, ICD-9 codes triggered a review of medical records for 64 medical patients, only 16 (25%) of whom had proven or probable invasive aspergillosis.Conclusions.A surveillance system that involved multiple ICD-9 codes was sufficiently sensitive to identify most cases of invasive aspergillosis; however, the poor PPV of ICD-9 codes means that this approach is not adequate as the sole tool used to classify cases. Screening ICD-9 codes to trigger a medical record review might be a useful method of surveillance for invasive aspergillosis and quality assessment, although more investigation is needed.

Download Full-text

Real-World Treatment Patterns, Outcomes, and Healthcare Resource Utilization in Relapsed or Refractory Multiple Myeloma: Evidence from a Medical Record Review in France

Advances in Hematology ◽

10.1155/2019/4625787 ◽

2019 ◽

Vol 2019 ◽

pp. 1-12 ◽

Cited By ~ 2

Author(s):

Huamao Mark Lin ◽

Keith L. Davis ◽

James A. Kaye ◽

Katarina Luptakova ◽

Saurabh P. Nagar ◽

...

Keyword(s):

Multiple Myeloma ◽

High Risk ◽

Medical Record ◽

Real World ◽

Treatment Patterns ◽

Medical Record Review ◽

Second Line ◽

Risk Patients ◽

Record Review ◽

Standard Risk

Background. Limited data are available from real-world practices in Europe describing prevailing treatment patterns and outcomes in relapsed/refractory multiple myeloma (RRMM), particularly by cytogenetic risk. Methods. A retrospective medical record review was conducted in 200 RRMM patients in France. From first relapse, patients were assessed on second-/third-line treatments, progression-free survival (PFS), overall survival (OS), and healthcare utilization. Results. Fifty-five high risk and 113 standard risk patients were identified. Overall, 192 patients (96%) received second-line therapy after relapse. Lenalidomide-based regimens were most common (>50%) in second line. Hospitalization incidence in high risk patients was approximately twice that of standard risk patients. From Kaplan-Meier estimation, median (95% CI) second-line PFS was 21.4 (17.5, 25.0) months (by high versus standard risk: 10.6 [6.4, 17.0] versus 28.7 [22.1, 37.3] months). Among second-line recipients, 47.4% were deceased at data collection. Median second-line OS was 59.4 (38.8, NE) months (by high versus standard risk: 36.5 [17.4, 50.6] versus 73.6 [66.5, NE] months). Conclusions. The prognostic importance of cytogenetic risk in RRMM was apparent, whereby high (versus standard) risk patients had decidedly shorter PFS and OS. Frequent hospitalizations indicated potentially high costs associated with RRMM, particularly for high risk patients. These findings may inform economic evaluations of RRMM therapies.

Download Full-text

Tumor registry versus physician medical record review: A head-to-head comparison of malignant pancreatic neuroendocrine tumor (mPNET) cases

Journal of Clinical Oncology ◽

10.1200/jco.2009.27.15_suppl.6600 ◽

2009 ◽

Vol 27 (15_suppl) ◽

pp. 6600-6600

Author(s):

E. E. Manasanch ◽

T. P. McDade ◽

A. Bodnari ◽

J. McKinney ◽

M. E. Sullivan ◽

...

Keyword(s):

Medical Record ◽

Medical Center ◽

Academic Medical Center ◽

National Level ◽

Pancreatic Neuroendocrine Tumor ◽

Medical Record Review ◽

Tumor Registry ◽

Identification Rate ◽

Diagnosis Codes ◽

Record Review

6600 Background: Researchers have underutilized single-institution tumor registry (TR) data, instead using researcher medical record review and/or personal, departmental or institutional databases to identify patients with particular malignancies. However, TR data is becoming increasingly prominent on a national level through broader use of the National Cancer Database (NCDB). We selected mPNETs as example to compare the accuracy of TR identification of these tumors with physician medical record review (MD Review). Methods: For MD Review, the health information services department of a single academic medical center was queried for all patients with pancreatic ICD9 codes (157.0–157.9; 211.6–211.7) January 2000-August 2008. A single physician reviewer analyzed computerized and paper medical records and identified mPNET cases. For TR data, mPNET patients were identified using the TR database with assistance of TR staff by two separate strategies. From January 2000-December 2006, patients were identified by using diagnosis codes from manual review of admission, discharge, clinic and pathology reports. From January 2007-August 2008 the TR used an automated case finding program (CAL by C/NET, California, USA) that downloaded cases with terms and codes related to malignancy. All MD Review- and TR-identified mPNET cases were reviewed by a second investigator blinded to identification strategy to assure consistency of mPNET definitions. Results: Using MD Review, 1194 pts with pancreatic ICD9 codes were identified. After MD Review, 42 mPNET patients were identified and confirmed. In comparison, TR identified 17 patients, of whom 5 were not identified by MD Review. Of the 47 patients identified by either strategy, TR identified 17/47 (32.6%) patients, whereas MD Review identified 42/47 (89.4%). TR identification rate in time periods 1 and 2 were 30% and 40%, respectively. Conclusions: Analysis of an academic tumor registry demonstrates that a substantial proportion of mPNET cases are missed when compared to ICD-9 identification and physician medical record review. Since MD review is imperfect, the TR may be even less effective at identifying PNETs than our data suggest. This may be applicable to other tumor registries and TR-based national studies. No significant financial relationships to disclose.

Download Full-text