Using a social marketing framework to evaluate recruitment of a prospective study of genetic counseling and testing for the deaf community

Table 3 Logistic regression analyses: effect of participant language on communication technologies, participation location, and recruitment channel

Social marketing mix		Language^a		Education^a	Age^a	Gender^a
Concept	Outcome variable	ASL Odds ratio (95% CI^b)	ASL + English Odds ratio (95% CI)	≥ BA degree Odds ratio (95% CI)	>44 yrs Odds ratio (95% CI)	Female Odds ratio (95% CI)
Price: communication technology	Video phone	9.83	10.16	1.50	0.73	1.35
		(3.26, 29.67)	(3.06, 33.80)	(0.88, 2.56)	(0.43, 1.26)	(0.79, 2.33)
	Pager	19.12	12.91	2.28	1.12	0.84
		(5.42, 67.40)	(3.42, 48.79)	(1.31, 3.97)	(0.65, 1.93)	(0.49, 1.46)
	Voice telephone	0.12	0.09	1.17	0.93	1.12
		(0.05, 0.28)	(0.03, 0.26)	(0.64, 2.14)	(0.51, 1.72)	(0.61, 2.06)
	Sign language inclusive	35.18	21.96	1.18	3.88	0.70
		(11.96, 103.42)	(6.37, 75.65)	(0.51, 2.74)	(1.60, 9.43)	(0.31, 1.59)
Place: participation location	Community-based^c	2.72	3.39	0.96	1.42	1.08
		(1.20, 6.14)	(1.31, 8.74)	(0.57, 1.62)	(0.84, 2.37)	(0.66, 1.62)
	Community-based^d	1.49	3.01	0.83	1.43	0.90
		(0.53, 4.17)	(0.95, 9.51)	(0.41, 1.68)	(0.73, 2.81)	(0.45, 1.79)
Promotion: recruitment channel	Community event^e	23.18	19.63	--	--	--
		(3.60, 976.77)	(2.69, 879.57)	--	--	--
	Organization	0.69	0.52	1.31	0.65	1.17
		(0.27, 1.74)	(0.17, 1.61)	(0.67, 2.54)	(0.34, 1.24)	(0.62, 2.23)
	Personal media	0.51	0.64	1.75	0.84	0.93
		(0.20, 1.32)	(0.21, 1.92)	(0.86, 3.54)	(0.43, 1.64)	(0.47, 1.84)
	Mass media	0.13	0.22	0.59	1.91	1.02
		(0.04, 0.48)	(0.05, 1.04)	(0.20, 1.77)	(0.63, 5.80)	(0.34, 3.09)

^aReference group for: Language = English-users; Education = < BA degree; Age = < 44 years; Gender = Male.
^bCI = Confidence Interval.
^cOdds ratios = odds of participating at community-based location rather than medically-based location.
^dOdds ratios = odds of participating at California State University Northridge rather than University of California Los Angeles.
^eUnadjusted exact logistic regression analysis was performed due to small numbers of English-users.

ISSN: 1471-2288