Appendix Table 6.

Multivariable Sensitivity Analysis Using Alternative Dichotomization Thresholds for Response Variable: Intention to Receive COVID-19 Within 1 Month (Rather Than 3 Months) of Regulatory Approval

Alternative DichotomizationMultivariable Odds Ratio (95% Cl, P Value)*
ConcernEduHealth§AgeSex (Men)InfluenzaDirect Interaction
1.00 (0.95-1.06, .918)2.25 (1.89-2.68, <.001)2.21 (1.57-3.18, <.001)1.16 (0.99-1.35, .064)
Alt1.00 (0.94-1.05, .938)2.25 (1.89-2.68, <.001)2.20 (1.57-3.16, <.001)1.17 (1.00–1.36, .047)
Alt1.02 (0.97-1.08, .427)2.28 (1.91–2.71, <.001)2.21 (1.57-3.17, <.001)1.14 (0.98-1.33, .090)
AltAlt1.02 (0.97-1.07, .508)2.28 (1.92–2.71, <.001)2.19 (1.56-3.15, <.001)1.15 (0.99-1.34, .070)
Alt1.01 (0.96-1.07, .706)2.28 (1.92–2.72, <.001)2.23 (1.58-3.21, <.001)1.12 (0.96-1.30, .160)
AltAlt1.01 (0.95-1.06, .830)2.29 (1.92–2.73, <.001)2.22 (1.57-3.19, <.001)1.13 (0.97-1.31, .128)
AltAlt1.03 (0.98-1.09, .269)2.31 (1.94-2.75, <.001)2.23 (1.58-3.21, <.001)1.10 (0.95-1.29, .209)
AltAltAlt1.03 (0.97-1.08, .324)2.32 (1.94-2.76, <.001)2.22 (1.57-3.19, <.001)1.11 (0.95-1.30, .174)
Alternative DichotomizationMultivariable Odds Ratio (95% Cl, P Value)*
ConcernEduHealth§ConcernEducationHealth
2.02 (1.74-2.35, <.001)1.29 (1.11–1.51, .001)0.84 (0.70–1.01, .066)
Alt2.00 (1.72–2.32, <.001)1.31 (1.13-1.53, <.001)0.82 (0.52–1.27, .386)
Alt2.03 (1.74-2.36, <.001)1.49 (1.19-1.89, <.001)0.84 (0.70–1.01, .065)
AltAlt2.00 (1.73-2.33, <.001)1.53 (1.22–1.93, <.001)0.80 (0.51–1.24, .333)
Alt2.74 (2.26-3.34, <.001)1.30 (1.12–1.52, <.001)0.86 (0.72–1.03, .101)
AltAlt2.72 (2.24-3.31, <.001)1.32 (1.13-1.54, <.001)0.84 (0.53-1.29, .429)
AltAlt2.75 (2.27-3.35, <.001)1.51 (1.20–1.91, <.001)0.86 (0.72–1.03, .100)
AltAltAlt2.73 (2.25-3.32, <.001)1.54 (1.22–1.94, <.001)0.82 (0.52–1.26, .370)
  • CI, confidence interval; Edu, educational attainment; Alt, alternative.

  • * Boldface text indicates P < .05.

  • Concern about becoming infected with COVID-19. Concern was reported on a 5-point Likert scale (extremely, moderately, somewhat, slightly, not at all). The original (–) threshold divides between moderately/somewhat. The alternative threshold divides between somewhat/slightly.

  • Educational attainment was reported on a 6-point Likert scale (some high school, high school graduate, some college, college graduate, Master's degree and Doctoral/Professional degree). The original (–) threshold divides between College graduate/Master's degree. The alternative threshold divides between some college/college graduate.

  • § Self-reported health status was reported on a 5-point Likert scale (excellent, very good, good, fair, poor). The original threshold (–) divides between very good/good. The alternative threshold divides between good/fair.