Skip to main content
. 2020 Aug 10;6(2):e17352. doi: 10.2196/17352

Table 2.

Logistic regression model of electronic personal health information use among patients who reported being diagnosed with cancer grouped by rural and urban status based from the Health Information National Trends Survey (n=1388) in the years 2008, 2011, 2013, and 2017, adjusted for sex, age, race/ethnicity, education, and income.

Characteristics Odds ratio (95% CI) P value
Residential area
.40

Urban Refa

Rural 0.78 (0.43-1.40)
Sex
.75

Female Ref

Male 0.93 (0.60-1.45)
Age in years
.52

18-34 Ref

35-49 0.50 (0.17-1.47)

50-64 0.48 (0.19-1.22)

65-74 0.42 (0.16-1.11)

≥75 0.41 (0.13-1.27)
Race/ethnicity
.27

Non-Hispanic white Ref

Hispanic 0.94 (0.46-1.93)

Non-Hispanic black 1.16 (0.51-2.64)

Non-Hispanic other 2.05 (0.98-4.31)
Education
.07

Less than high school Ref

High school graduate 1.03 (0.19-5.50)

Some college 2.01 (0.38-10.70)

College graduate 2.04 (0.41-10.03)
Income
.09

<$20,000 Ref

$20,000-<$35,000 1.50 (0.62-3.60)

$35,000-<$50,000 1.13 (0.48-2.69)

$50,000-<$75,000 1.93 (0.72-5.15)

$75,000+ 2.21 (0.90-5.40)
Survey year
<.001

2008 Ref

2011 1.57 (1.02-2.43)

2013 3.38 (1.89-6.15)

2017 13.07 (8.23-20.75)

aRef: reference.