Limits...
Measuring Mortality Information in Clinical Data Warehouses.

Jones B, Vawdrey DK - AMIA Jt Summits Transl Sci Proc (2015)

Bottom Line: Of 33,295 deaths recorded in our institution's patient registration system, 13,167 (39.5%) did not exist in the DMF.The proportion of patients still living according to Columbia's CDW who were over 100 and 120 years of age was 43.6% and 43.1%, respectively.These measures may be useful to other clinical research investigators seeking to assess the quality of mortality data (1-4).

View Article: PubMed Central - PubMed

Affiliation: Department of Statistics, Brigham Young University, Provo, UT.

ABSTRACT
The ability to track and report long-term outcomes, especially mortality, is essential for advancing clinical research. The purpose of this study was to present a framework for assessing the quality of mortality information in clinical research databases. Using the clinical data warehouse (CDW) at Columbia University Medical Center as a case study, we measured: 1) agreement in vital status between our institution's patient registration system and the U.S. Social Security Administration's Death Master File (DMF), 2) the proportion of patients marked as deceased according to the DMF records who had subsequent visits to our institution, and 3) the proportion of patients still living according to Columbia's CDW who were over 100 and 120 years of age. Of 33,295 deaths recorded in our institution's patient registration system, 13,167 (39.5%) did not exist in the DMF. Of 315,037 patients in our CDW who marked as deceased according to the DMF, 2.1% had a subsequent clinical encounter at our institution. The proportion of patients still living according to Columbia's CDW who were over 100 and 120 years of age was 43.6% and 43.1%, respectively. These measures may be useful to other clinical research investigators seeking to assess the quality of mortality data (1-4).

No MeSH data available.


Deaths recorded in Columbia’s clinical data warehouse by year (2000–2013).
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4525266&req=5

f2-2091421: Deaths recorded in Columbia’s clinical data warehouse by year (2000–2013).

Mentions: Without integrating data from outside sources, the mortality information in a healthcare delivery organization’s clinical data warehouse may be extremely incomplete. We found that agreement and accuracy of the data were low, suggesting a lack of quality in the DMF data for the population our institution serves. If the DMF had perfect agreement we would expect 100% overlap between the deaths from the DMF and deaths recorded in the hospital. We found only 60.5% agreement, meaning only 60.5% of the deaths in the hospital were recorded in the DMF. This finding may be partially explained by the hospital receiving significantly fewer records after 2011, as can be seen in the year-by-year breakdown of deaths in Figure 2. In 2012 and 2013, our data warehouse contained significantly fewer total deaths, and agreement plummeted to 17% and 13% agreement respectively. However, even if 2012 and 2013 data are excluded, we observed only 62% agreement overall. This result is lower than what has been reported in other studies. Hauser et al. found 82% agreement between the DMF and a cohort of known decedents (8). Huntington et al. found 94.7% agreement between the DMF and a random sample of Ohio deaths (9). Newman et al. found 96.5% agreement between the DMF supplemented with state death records and in hospital deaths at a California hospital. Other studies have found similar results (10–14).


Measuring Mortality Information in Clinical Data Warehouses.

Jones B, Vawdrey DK - AMIA Jt Summits Transl Sci Proc (2015)

Deaths recorded in Columbia’s clinical data warehouse by year (2000–2013).
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4525266&req=5

f2-2091421: Deaths recorded in Columbia’s clinical data warehouse by year (2000–2013).
Mentions: Without integrating data from outside sources, the mortality information in a healthcare delivery organization’s clinical data warehouse may be extremely incomplete. We found that agreement and accuracy of the data were low, suggesting a lack of quality in the DMF data for the population our institution serves. If the DMF had perfect agreement we would expect 100% overlap between the deaths from the DMF and deaths recorded in the hospital. We found only 60.5% agreement, meaning only 60.5% of the deaths in the hospital were recorded in the DMF. This finding may be partially explained by the hospital receiving significantly fewer records after 2011, as can be seen in the year-by-year breakdown of deaths in Figure 2. In 2012 and 2013, our data warehouse contained significantly fewer total deaths, and agreement plummeted to 17% and 13% agreement respectively. However, even if 2012 and 2013 data are excluded, we observed only 62% agreement overall. This result is lower than what has been reported in other studies. Hauser et al. found 82% agreement between the DMF and a cohort of known decedents (8). Huntington et al. found 94.7% agreement between the DMF and a random sample of Ohio deaths (9). Newman et al. found 96.5% agreement between the DMF supplemented with state death records and in hospital deaths at a California hospital. Other studies have found similar results (10–14).

Bottom Line: Of 33,295 deaths recorded in our institution's patient registration system, 13,167 (39.5%) did not exist in the DMF.The proportion of patients still living according to Columbia's CDW who were over 100 and 120 years of age was 43.6% and 43.1%, respectively.These measures may be useful to other clinical research investigators seeking to assess the quality of mortality data (1-4).

View Article: PubMed Central - PubMed

Affiliation: Department of Statistics, Brigham Young University, Provo, UT.

ABSTRACT
The ability to track and report long-term outcomes, especially mortality, is essential for advancing clinical research. The purpose of this study was to present a framework for assessing the quality of mortality information in clinical research databases. Using the clinical data warehouse (CDW) at Columbia University Medical Center as a case study, we measured: 1) agreement in vital status between our institution's patient registration system and the U.S. Social Security Administration's Death Master File (DMF), 2) the proportion of patients marked as deceased according to the DMF records who had subsequent visits to our institution, and 3) the proportion of patients still living according to Columbia's CDW who were over 100 and 120 years of age. Of 33,295 deaths recorded in our institution's patient registration system, 13,167 (39.5%) did not exist in the DMF. Of 315,037 patients in our CDW who marked as deceased according to the DMF, 2.1% had a subsequent clinical encounter at our institution. The proportion of patients still living according to Columbia's CDW who were over 100 and 120 years of age was 43.6% and 43.1%, respectively. These measures may be useful to other clinical research investigators seeking to assess the quality of mortality data (1-4).

No MeSH data available.