Skip to main content
. 2014 Oct 2;16(10):e217. doi: 10.2196/jmir.3239

Table 1.

Demographic and health characteristics of the study sample, by program participation level.

Demographic/characteristic Total samplea
(n=16,493)
Subsample of non-participantsa
(n=452)
PAMb only: Low incentive participation
(n=7099)
Completed 8 topics: High incentive
(n=2693)
Went beyond incentive requirements
(n=6249)
P value
Weighted percent of population - 14,609 (88.58%) 870 (5.27%) 305 (1.85%) 709 (4.30%)
Gender, n (%)

Male 8881 (53.85) 8048 (55.09) 366 (42.06) 153 (50.16) 314 (44.29)c <.001

Female 7612 (46.15) 6561 (44.91) 504 (57.93) 152 (49.84) 395 (55.71)c
Age, mean (SD)
41.3 (13.4) 41.2 (13.5) 42.8 (12.2) 39.2 (11.8) 41.9 (11.6)c <.001
Income of zip code, mean (SD) 23,967.34 (7131.31) 23,947.40 (7164.20) 24,318.16 (6899.29) 24,137.45 (7199.90) 23,874.46 (6846.15) .48
PAM score, mean (SD)
N/A N/A 70.3 (15.3) 70.5 (15.0) 71.7 (15.3)c <.001
Health Risk Score, mean (SD) 2.0 (1.3) 1.9 (1.3) 2.1 (1.2) 2.1 (1.2) 2.2 (1.1)c <.001
Chronic conditions, n (%)

Depression 2156 (13.07) 1939 (13.27) 103 (11.84) 30 (9.84) 83 (11.71) .24

Diabetes 519 (3.15) 485 (3.32) 15 (1.72) 5 (1.64) 14 (1.97)a .02

Hypertension 2307 (13.99) 2101 (14.38) 93 (10.69) 28 (9.18) 85 (11.99)a .02

aSince we examined a random sample of people who never participated in online coaching, the descriptive statistics are weighted using the inverse probability of selection for the no coaching group, and the standard errors are weighted accordingly. The numbers presented in the table for percentages reflect the weighted numbers.

bPAM: patient activation measure

cChi-square statistical tests were used for categorical demographic and health variables (gender and having specific chronic conditions), while ANOVAs were used for continuous variables (age, mean income of zip code, health risk score).