Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Jun 3:11:556.
doi: 10.3389/fgene.2020.00556. eCollection 2020.

Impact of Diverse Data Sources on Computational Phenotyping

Liwei Wang  1 , Janet E Olson  2   3 , Suzette J Bielinski  2 , Jennifer L St Sauver  2 , Sunyang Fu  1 , Huan He  1 , Mine S Cicek  4 , Matthew A Hathcock  5 , James R Cerhan  2 , Hongfang Liu  1
Affiliations

Impact of Diverse Data Sources on Computational Phenotyping

Liwei Wang et al. Front Genet. .

Abstract

Electronic health records (EHRs) are widely adopted with a great potential to serve as a rich, integrated source of phenotype information. Computational phenotyping, which extracts phenotypes from EHR data automatically, can accelerate the adoption and utilization of phenotype-driven efforts to advance scientific discovery and improve healthcare delivery. A list of computational phenotyping algorithms has been published but data fragmentation, i.e., incomplete data within one single data source, has been raised as an inherent limitation of computational phenotyping. In this study, we investigated the impact of diverse data sources on two published computational phenotyping algorithms, rheumatoid arthritis (RA) and type 2 diabetes mellitus (T2DM), using Mayo EHRs and Rochester Epidemiology Project (REP) which links medical records from multiple health care systems. Results showed that both RA (less prevalent) and T2DM (more prevalent) case selections were markedly impacted by data fragmentation, with positive predictive value (PPV) of 91.4 and 92.4%, false-negative rate (FNR) of 26.6 and 14% in Mayo data, respectively, PPV of 97.2 and 98.3%, FNR of 5.2 and 3.3% in REP. T2DM controls also contain biases, with PPV of 91.2% and FNR of 1.2% for Mayo. We further elaborated underlying reasons impacting the performance.

Keywords: computational phenotyping; diverse data sources; phenotyping algorithms; rheumatoid arthritis; type 2 diabetes mellitus.

PubMed Disclaimer

Figures

FIGURE 1
FIGURE 1
The eMERGE algorithm for identifying RA cases and controls. Adapted from Partners Phenotyping Group (2016).
FIGURE 2
FIGURE 2
The eMERGE algorithm for identifying T2DM cases. Adapted from Pacheco and Thompson (2012).
FIGURE 3
FIGURE 3
The eMERGE algorithm for identifying T2DM controls. Adapted from Pacheco and Thompson (2012).
FIGURE 4
FIGURE 4
Probability of all benchmark cases based on various data sources. The red line intercepts the cutoff of 0.632, probabilities above the red line are classified as RA cases.
FIGURE 5
FIGURE 5
Quantitative comparison of each step in T2DM case phenotyping among various data sources. The number of each step corresponds to Figure 2, bold numbers are derived from the combination of Mayo of REP data.
FIGURE 6
FIGURE 6
Quantitative comparison of each step in T2DM control phenotyping among various data sources. The number of each step corresponds to Figure 3, bold numbers are derived from the combination of Mayo of REP data.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Aletaha D., Neogi T., Silman A. J., Funovits J., Felson D. T., Bingham C. O., et al. (2010). 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthr. Rheum 62 2569–2581. 10.1002/art.27584 - DOI - PubMed
    1. Bullard K. M., Cowie C. C., Lessem S. E. (2018). Prevalence of diagnosed diabetes in adults by diabetes type—United States, 2016. Morb. Mortal. Wkly Rep. 67:359. 10.15585/mmwr.mm6712a2 - DOI - PMC - PubMed
    1. Carroll R. J., Eyler A. E., Denny J. C. (2015). Intelligent use and clinical benefits of electronic health records in rheumatoid arthritis. Expert Rev. Clin. Immunol. 11 329–337. 10.1586/1744666X.2015.1009895 - DOI - PMC - PubMed
    1. Carroll R. J., Thompson W. K., Eyler A. E., Mandelin A. M., Cai T., Zink R. M., et al. (2012). Portability of an algorithm to identify rheumatoid arthritis in electronic health records. J. Am. Med. Inform. Assoc. 19 e162–e169. 10.1136/amiajnl-2011-000583 - DOI - PMC - PubMed
    1. Crawford D. C., Crosslin D. R., Tromp G., Kullo I. J., Kuivaniemi H., Hayes M. G. (2014). eMERGEing progress in genomics—the first seven years. Front. Genet. 5:184. 10.3389/fgene.2014.00184 - DOI - PMC - PubMed

LinkOut - more resources