Skip to main content
. Author manuscript; available in PMC: 2012 Oct 1.
Published in final edited form as: J Clin Epidemiol. 2010 Dec 15;64(10):1144–1151. doi: 10.1016/j.jclinepi.2010.08.007

Table 2.

Predictors of time to participant contact

Interim model Final model
HR (95% CI) p-value HR (95% CI) p-value
Patient factors
Enrolled from the primary study site 1.45 (0.84-2.50) (0.20) 1.32 (0.76-2.28) 0.32
Employed 0.53 (0.19-1.50) (0.27)
Type of work
 Non-physical labor 1.54 (0.46-5.12) (0.48)
 Physical labor, not working or unemployed Ref
Patient-reported chronic fatigue* 1.55 (0.91-2.62) (0.10) 1.85 (1.12-3.06) 0.02
Requires assistance at home 1.28 (0.67-2.41) (0.44)
Lives in a health care facility 0.50 (0.20-1.27) (0.15) 0.57 (0.23-1.43) 0.23

Research visit factors
Number of contact phone numbers available 1.04 (0.91-1.18) (0.51)
Visit in clinic 1.05 (0.60-1.84) (0.88)
Last visit not conducted in clinical due to
patient illness/condition
0.45 (0.20-1.01) (0.05) 0.40 (0.19-0.84) 0.02
Incomplete data collection due to personal
factors£
0.88 (0.48-1.61) (0.71)
Missed last follow-up visit 1.22 (0.33-4.45) (0.47)
Missed visit due to personal factor$ 0.67 (0.16-2.71) (0.38)
Ever missed any follow-up visit 0.73 (0.31-1.72) (0.26)

Abbreviations: HR: hazard ratio, >1 indicates a shorter time to contact with the participant; CI confidence interval

*

Missing data for 1 patient.

£

Patient incarcerated, on ventilator, excessively fatigued, cognitively, mentally or, physically incapable.

$

Patient incarcerated, refusal-not interested, no time or too much stress, drug or alcohol influence.

Measured at last follow-up visit; due to collinearity “physical labor” category was combined with “not working or unemployed”