Validation of LexisNexis Accurint in the Georgia Cancer Registry's Cancer Recurrence and Information Surveillance Program Öffentlichkeit
Woolpert, Kirsten (Spring 2020)
Published
Abstract
Background: LexisNexis Accurint is a database of ~45 billion public records that provides information regarding individual’s location of residence. This low-cost method can be used as a tool in prospective cohort studies to maintain high follow-up rates, but to date has not been validated to ensure accuracy in tracking individuals. This study utilized the Georgia Cancer Registry’s Cancer Recurrence and Information Surveillance Program (CRISP) of 69,494 cancer patients to validate the software and to examine predictors of patients not included or who had an inaccurate entry in LexisNexis.
Methods: Cancer patients within the Georgia Cancer Registry are routinely linked to the National Death Index (NDI), providing for decedents the US state in which the patient died. We compared the state of residence reported in Lexis Nexis with the NDI state of residence at death as the gold standard, allowing for calculations of sensitivity and specificity of state of residence information in Lexis Nexis. Additionally, multivariate logistic regression analyses were performed to examine associations between demographic information provided through the registry and three outcomes: 1. having a match between LexisNexis and NDI, 2. being missed in the LexisNexis database, and 3. moving out of the state of Georgia according to LexisNexis.
Results: Of the 69,494 patients in the CRISP cohort, 65,890 (94.8%) were found in LexisNexis, and a total of 9,597 (13.8%) had died. The sensitivity of the LexisNexis software for identifying persons who moved out of Georgia was 34.6% and the specificity was 89.3%. Unmarried individuals, blacks, Asian/Pacific Islanders, Hispanics, individuals living in high poverty neighborhoods, and younger patients were all more likely to be missed in the LexisNexis database as well as to have a discordance between LexisNexis state of residence and the National Death Index state of residence at death.
Discussion: This study showed that LexisNexis Accurint did not accurately identify state of residence at death in a large proportion of CRISP cohort members. Since achieving high follow-up rates is essential in any prospective cohort, the low validity of this software is important to note for researchers planning to use this software for follow-up. The generalizability of results to persons who had not died by end of follow-up is an important consideration.
Table of Contents
Background……………………………………………………………………….……………… 1
Methods……………………………………………………………………………………….......4
Results…………………………………………………………………………………………….. 7
Discussion………………………………………………………………………..………………. 9
Strengths and Weaknesses…….………………………………………………………....…. 11
Future Directions………………………………………………………………………..…….. 13
References……………………………………………………………………………….………. 14
Tables……………………………………………………………………………………...……… 16
Figures…………………………………………………………………………………..………... 18
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Stichwort | |
| Committee Chair / Thesis Advisor |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Validation of LexisNexis Accurint in the Georgia Cancer Registry's Cancer Recurrence and Information Surveillance Program () | 2020-04-22 14:13:57 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|