Accuracy and complexities of using automated clinical data for capturing chemotherapy administrations: implications for future research - PubMed (original) (raw)

Multicenter Study

Accuracy and complexities of using automated clinical data for capturing chemotherapy administrations: implications for future research

Erin J Aiello Bowles et al. Med Care. 2009 Oct.

Abstract

Background: Chemotherapy data are important to almost any study on cancer prognosis and outcomes. However, chemotherapy data obtained from tumor registries may be incomplete, and abstracting chemotherapy directly from medical records can be expensive and time consuming.

Methods: We evaluated the accuracy of using automated clinical data to capture chemotherapy administrations in a cohort of 757 ovarian cancer patients enrolled in 7 health plans in the HMO Cancer Research Network. We calculated sensitivity and specificity with 95% confidence intervals of chemotherapy administrations extracted from 3 automated clinical data sources (Health Care Procedure Coding System, National Drug Codes, and International Classification of Diseases) compared with tumor registry data and medical chart data.

Results: Sensitivity of all 3 data sources varied across health plans from 79.4% to 95.2% when compared with tumor registries, and 75.0% to 100.0% when compared with medical charts. The sensitivities using a combination of 3 data sources were 88.6% (95% confidence intervals: 85.7-91.1) compared with tumor registries and 89.5% (78.5-96.0) compared with medical records; specificities were 91.5% (86.4-95.2) and 90.0% (55.5-99.7), respectively. There was no difference in accuracy between women aged < 65 and > or = 65 years. Using one set of codes alone (eg, Health Care Procedure Coding System alone) was insufficient for capturing chemotherapy data at most health plans.

Conclusions: While automated data systems are not without limitations, clinical codes used in combination are useful in capturing chemotherapy more comprehensively than tumor registry and without the need for costly medical record abstraction. Key Words: validation of automated clinical data, chemotherapy, medical chart, tumor registry, ovarian cancer.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Sensitivity of automated clinical data for capturing chemotherapy administrations compared to tumor registries by health plan

Each bar represents a single health plan indicated by a color and number (1-7). Health plans are ordered from highest to lowest sensitivity for each type of automated clinical data. For example, health plan 1 has the highest sensitivity for HCPCS codes alone (90.2%), second highest for ICD-9-CM codes alone (65.9%), and fourth highest for NDCs alone (26.8%). When using all three types of codes, the overall sensitivity for health plan 1 is 92.7%.

Figure 2

Figure 2. Sensitivity of automated clinical data for capturing chemotherapy administrations compared to medical records by health plan

Each bar represents a single health plan indicated by a color and number (1-7). Health plans are ordered from highest to lowest sensitivity for each type of automated clinical data. For example, health plan 1 has the highest sensitivity for HCPCS codes alone (100%), third highest for ICD-9-CM codes alone (66.7%), and fourth highest for NDCs alone (16.7%). When using all three types of codes, the sensitivity for health plan 1 is 100%.

Figure 3

Figure 3. Sensitivity of automated clinical data for capturing chemotherapy administrations compared to tumor registries and medical records by age at diagnosis

Each bar represents the sensitivity for an age group (<65 or ≥65) indicated by a color with the sensitivity value at the top of the bar. For example, the sensitivity of HCPCS + NDC + ICD-9-CM codes combined for women <65 years was 88.7% compared to tumor registries (3a) and 87.1% compared to medical records (3b).

Similar articles

Cited by

References

    1. Ries LAG MD, Krapcho M, Mariotto A, Miller BA, Feuer EJ, Clegg L, Horner MJ, Howlader N, Eisner MP, Reichman M, Edwards BK, editors. SEER Cancer Statistics Review, 1975−2004. National Cancer Institute; Bethesda, MD:
    1. NCI Plans and Priorities Cancer Survivorship. [July 23, 2008]. Available at: http://plan2004.cancer.gov/public/survivor.htm.
    1. Geiger AM, Buist DS, Greene SM, et al. Survivorship research based in integrated healthcare delivery systems: the Cancer Research Network. Cancer. 2008;112:2617–2626. - PubMed
    1. Warren JL, Harlan LC, Fahey A, et al. Utility of the SEER-Medicare data to identify chemotherapy use. Med Care. 2002;40:IV, 55–61. - PubMed
    1. Du XL, Key CR, Dickie L, et al. Information on chemotherapy and hormone therapy from tumor registry had moderate agreement with chart reviews. J Clin Epidemiol. 2006;59:53–60. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources