Validation of an electronic coding algorithm to identify the primary indication of orthopedic surgeries from administrative data

John C Giardina; Thomas Cha; Steven J Atlas; Michael J Barry; Andrew A Freiberg; Lauren Leavitt; Felisha Marques; Karen Sepucha

doi:10.1186/s12911-020-01175-1

Validation of an electronic coding algorithm to identify the primary indication of orthopedic surgeries from administrative data

BMC Med Inform Decis Mak. 2020 Aug 12;20(1):187. doi: 10.1186/s12911-020-01175-1.

Authors

John C Giardina¹, Thomas Cha^{2

3}, Steven J Atlas^{3

4}, Michael J Barry^{3

4}, Andrew A Freiberg², Lauren Leavitt⁴, Felisha Marques⁴, Karen Sepucha^{3

4}

Affiliations

¹ Center for Health Decision Science, Harvard T.H. Chan School of Public Health, 718 Huntington Avenue, Boston, MA, 02115, USA. jgiardina@g.harvard.edu.
² Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, MA, USA.
³ Harvard Medical School, Boston, MA, USA.
⁴ Division of General Internal Medicine, Massachusetts General Hospital, Boston, MA, USA.

Abstract

Background: Determining the primary indication of a surgical procedure can be useful in identifying patients undergoing elective surgery where shared decision-making is recommended. The purpose of this study was to develop and validate an algorithm to identify patients receiving the following combinations of surgical procedure and primary indication as part of a study to promote shared decision-making: (1) knee arthroplasty to treat knee osteoarthritis (KOA); (2) hip arthroplasty to treat hip osteoarthritis (HOA); (3) spinal surgery to treat lumbar spinal stenosis (SpS); and (4) spinal surgery to treat lumbar herniated disc (HD).

Methods: Consecutive surgical procedures performed by participating spine, hip, and knee surgeons at four sites within an integrated care network were included. Study staff reviewed electronic medical records to ascertain a "gold standard" determination of the procedure and primary indication status. Electronic algorithms consisting of ICD-10 and CPT codes for each combination of procedure and indication were then applied to records for each case. The primary measures of validity for the algorithms were the sensitivity and specificity relative to the gold standard review.

Results: Participating surgeons performed 790 procedures included in this study. The sensitivity of the algorithms in determining whether a surgical case represented one of the combinations of procedure and primary indication ranged from 0.70 (HD) to 0.92 (KOA). The specificity ranged from 0.94 (SpS) to 0.99 (HOA, KOA).

Conclusion: The electronic algorithm was able to identify all four procedure/primary indication combinations of interest with high specificity. Additionally, the sensitivity for the KOA cases was reasonably high. For HOA and the spine conditions, additional work is needed to improve the sensitivity of the algorithm to identify the primary indication for each case.

Keywords: Algorithm validation; Diagnostic code; Elective surgery; Electronic medical record; Orthopedic surgery; Procedure code; Shared-decision making.

Publication types

Research Support, Non-U.S. Gov't
Research Support, U.S. Gov't, P.H.S.

MeSH terms

Algorithms*
Arthroplasty, Replacement, Hip
Arthroplasty, Replacement, Knee
Decision Making*
Electronics
Humans
Intervertebral Disc Displacement / surgery*
Orthopedic Procedures / standards*
Osteoarthritis, Hip / surgery*
Osteoarthritis, Knee / surgery*
Reproducibility of Results
Spinal Stenosis / surgery*