Skip to main content
. 2018 Jan 10;18:14. doi: 10.1186/s12913-017-2824-9

Table 2.

HIV testing rates before and after EMR reminder implementation stratified by levels of patient characteristics

Patient characteristic Proportion tested before reminder Proportion tested after reminder Relative Risk and 95% Confidence Intervala P-valuea
OVERALL 15.3% 30.7% 2.02 (1.95–2.09) < 0.0001
AGE
Age ≤ 25 22.5% 34.3% 1.52 (1.38–1.69) <0.0001
Age 26–35 23.1% 36.9% 1.61 (1.53–1.71) <0.0001
Age 36–45 15.8% 31.0% 1.98 (1.83–2.13) <0.0001
Age 46–65 9.7% 27.2% 2.81 (2.65–2.99) <0.0001
GENDER
Female 14.0% 30.5% 2.19 (2.10–2.30) <0.0001
Male 17.3% 30.8% 1.82 (1.73–1.91) <0.0001
RACE
White 13.9% 27.1% 1.97 (1.88–2.07) <0.0001
Black 21.5% 41.0% 1.93 (1.81–2.06) <0.0001
Asian 14.5% 30.3% 2.11 (1.85–2.40) <0.0001
Other race 14.6% 33.4% 2.31 (2.12–2.51) <0.0001
LANGUAGE
English speaking 15.3% 30.4% 2.01 (1.94–2.08) <0.0001
Non-English speaking 17.2% 37.5% 2.20 (1.89–2.56) <0.0001
INCOME
Income < median ($67,338) 19.1% 34.9% 2.28 (2.16–2.40) <0.0001
Income > median ($67,338) 11.9% 26.9% 1.85 (1.77–1.93) <0.0001

aEstimates from generalized estimating equations models where the dependent variable is HIV testing (binary 1 or 0), and the independent variable is time period (before, after reminder). The before reminder time period serves as the reference category, and stratification is used for each level of the patient characteristic