|
100
90
|
FIGURE 1 |
Non-PWID-IE PWID-IE |
Cases that did not meet the criteria for opioid use were described as “ non-PWID- IE ” ( n = 172 ). Cases that met the criteria for opioid use were described as “ PWID- IE ” ( n = 272 ). CI = confidence interval . |
Total Number
80 70 60 50 40 30 20 10 0
Admit Year
2010 2011 2012 2013 2014 2015
Analysis of Annual Infective Endocarditis ( IE ) Admissions at Cabell-Huntington Hospital Based on Opioid Use from 2010-2019 ( n = 444 ).
TABLE 1 : Odds of Opioid Use in 444 Patients with Infective Endocarditis by Year of Hospital Admission .
Non-PWID-IE ( N = 172 )
PWID-IE ( N = 272 )
Odds Ratio of Opioid Use ( 95 % CI )
P Value
2010 8 3 1.0 ( Referent ) Referent 2011 14 5 0.95 ( 0.18-5.1 ) 0.95 2012 14 14 2.7 ( 0.58-12.2 ) 0.21 2013 18 19 2.8 ( 0.64-12.3 ) 0.17 2014 16 15 2.5 ( 0.56-11.2 ) 0.23 2015 18 24 3.6 ( 0.82 – 15.3 ) 0.09 2016 29 26 2.4 ( 0.57 – 10.0 ) 0.23 2017 20 58 7.7 ( 1.9 – 32.0 ) 0.005 2018 27 52 5.1 ( 1.2-20.9 ) 0.02 2019 8 56 18.7 ( 4.1-85.3 ) < 0.001
the criteria for opioid use ( Non-PWID- IE ) ( n = 172 ).
As displayed in Table 1 , there was a significant difference ( P for trend across years < 0.001 ) in the number of annual admissions for PWID-IE versus non-PWID-IE between 2010 and 2019 . The number of
Admit Year
2016 2017 2018 2019
PWID-IE admissions and the proportion of PWID-IE comprising total IE cases steadily increased from 2010 to 2019 . As compared to the baseline year of 2010 , the odds of PWID-IE were significantly higher in 2017 ( OR 7.7 ), 2018 ( OR 5.1 ), and 2019 ( 18.7 ) ( All p values < 0.05 ) relative to non-PWID-IE .
Baseline characteristics , including patient demographics , payer status , length of stay , hospital charges , readmissions , and comorbid conditions are reported in Table 2 .
Cases that did not meet the criteria for opioid use were described as “ non-PWID- IE ” ( n = 172 ). Cases that met the criteria for opioid use were described as “ PWID-IE ” ( n = 272 ). Numbers in parentheses represent the total N . Numbers in brackets represent the interquartile range ( IQR ).
Significant differences among PWID-IE and non-PWID-IE were reported for several demographic variables . The median age on admission for PWID-IE ( 34 years ; IQR 14 years ) was significantly lower ( P < 0.001 ) than non-PWID-IE ( 61 years ; IQR 30 ). In both groups , females comprised a greater proportion of IE cases . However , there was no significant difference in gender between PWID-IE and non-PWID-IE .
Differences in length of stay were significant ( P = 0.004 ), with non-PWID-IE ( 8 days ; IQR 9.5 days ) reporting a median length of stay that was three days longer than PWID-IE ( 5 days ; IQR 8.5 days ). No significant difference was found for hospital charges between the two groups ( P = 0.36 ), but taking into account hospital charges per day , charges were higher for PWID-IE ( Median $ 10,400 vs . $ 9,189 ; P = 0.02 ).
Data regarding comorbid conditions were also significantly different between the study groups . There were more non- PWID-IE patients with diabetes ( 48.8 % vs . 13.6 %), hypertension ( 77.3 % vs . 34.9 %), congestive heart failure ( 41.9 % vs . 15.4 %), and chronic kidney disease ( 36.0 % vs . 9.9 %) compared to PWID-IE patients ( all P < 0.01 ). Differences in smoking status were also significant between the two groups ( P < 0.001 ). A greater percentage of PWID-IE were current smokers ( 81.2 %) at the time of admission than non-PWID-IE ( 33.7 %).
16 • www . wvsma . org