What influences respondents to behave consistently when asked to consent to health record linkage on repeat occasions?

Figures & data

Table 1. Consent to link CM’s health records.

Wave	Content of consent request
MCS1 age 9 months	Consent for linking information on pregnancy and birth and for following the baby’s National Health Service (NHS) registration
MCS2 age 3 years	Consent for linking health records (hospital admissions and records held by the NHS) from birth to age 7
MCS4 age 7 years	Consent for linking health records (hospital admissions and records held by the NHS) from birth to age 14

Figure 1. Consent rates for health record linkage in the three waves.

Note: CM stands for cohort member and W denotes the wave of data collection.

Table 2. Consent patterns in waves 1, 2, and 4 of MCS.

Patterns of consent	Freq.	Percent
000	52	.5
001	95	1.0
010	68	.7
011	279	2.6
100	174	1.4
101	1690	14.0
110	471	3.9
111	8916	75.8
Total	11,745	100

Figure 2. Interviewers’ success rates in obtaining health record consent for waves 1, 2, and 4.

Figure 3. Cross-equation correlations based on a multivariate probit model.

Note: All correlations are significant at level of p < .01. Standard errors are in parentheses.

Table 3. Cross-sectional influences of socio-demographic characteristics on the willingness to consent using a multivariate probit model.

The CM is a boy, reference: girl	Wave 1		Wave 2		Wave 4
	−.031	(.047)	−.026	(.031)	−.048	(.039)
Highest socio-economic status, reference: managerial and professional
Intermediate	.024	(.083)	−.0051	(.047)	.093	(.068)
Small employers and self-employed	.14	(.094)	−.069	(.067)	.091	(.079)
Lower supervisory and technical	.13	(.090)	−.18^***	(.056)	.014	(.092)
Semi-routine and routine	−.0046	(.090)	−.16^***	(.052)	.072	(.068)
MR living in a couple, reference: single	.064	(.097)	−.072	(.059)	−.082	(.074)
Combined labour market status, reference: both in work
Only one in work	−.024	(.059)	−.013	(.037)	−.10^*	(.055)
Both not in work	−.18	(.130)	−.0036	(.090)	.056	(.115)
Housing tenure, reference: own
Rent	−.015	(.073)	.035	(.046)	−.0029	(.064)
Other	−.065	(.108)	.18^*	(.097)	−.026	(.156)
MR is Non-White, reference: white	−.44^***	(.102)	−.14	(.106)	−.39^***	(.096)
MR’s religion, reference: Christian
Non-Christian	−.2	(.166)	.24^**	(.096)	.24	(.147)
None	.069	(.056)	−.022	(.034)	−.0058	(.047)
Language spoken at home, reference: english
English and other languages	−.053	(.161)	−.024	(.076)	−.025	(.129)
Only other	−.11	(.226)	−.057	(.146)	.15	(.163)
The interview was translated, reference: no	−.28^**	(.123)	−.11	(.149)	.16	(.179)
MR’s health status, reference: excellent
Very good, good	−.055	(.064)	.011	(.037)	−.016	(.049)
Fair, poor	−.13	(.080)	.061	(.049)	.16^*	(.082)
CM’s health status, reference: some problems
No problems	−.15^***	(.056)	−.031	(.046)	.12^*	(.061)
MR: I am a very private person, reference: strongly agree
Agree	.23^*	(.122)	.12^*	(.065)	−.023	(.096)
Neither	.13	(.119)	.100	(.069)	.10	(.094)
Disagree	.17	(.125)	.057	(.065)	.090	(.098)
Strongly disagree	.16	(.167)	.16	(.098)	.29^**	(.140)
Can’t say	−.21	(.237)	−.0015	(.144)	−.25	(.243)
Other	.008	(.173)	−.020	(.116)	−.72^***	(.136)
Response history, reference: participated in all waves
Absent in at least one wave	.16	(.130)	−.11	(.089)	−.24^**	(.103)
Log OECD adjusted income	.015	(.057)	.078^***	(.030)	−.096^*	(.055)
MR’s age	.001	(.005)	−.0026	(.003)	.0064	(.004)
Constant	1.61^***	(.430)	.67^***	(.239)	1.85^***	(.366)
N	11,745		11,745		11,745

Notes:

Standard errors in parentheses.

^* p < .10

^** p < .05

^*** p < .01.

Table 4. Conditional probit regressions of changing consent decisions over consecutive waves.

	W1 to W2		W1 to W2		W2 to W4		W2 to W4
	(Yes to Yes) = 0		(No to No) = 0		(Yes to Yes) = 0		(No to No) = 0
	(Yes to No) = 1		(No to Yes) = 1		(Yes to No) = 1		(No to Yes) = 1
The CM is a boy, reference: Girl	.025	(.033)	−.079	(.137)	.084^*	(.047)	.059	(.086)
Highest socio-economic status, reference: managerial and professional
Intermediate	.026	(.051)	.14	(.238)	−.14^**	(.068)	−.044	(.133)
Small employers and self-employed	.18^***	(.064)	.48	(.313)	−.05	(.085)	.28	(.174)
Lower supervisory and technical	.17^***	(.065)	−.06	(.283)	−.092	(.096)	.31^*	(.189)
Semi-routine and routine	.19^***	(.049)	.12	(.196)	−.069	(.082)	−.011	(.153)
MR living in a couple, reference: Single	.061	(.065)	−.093	(.224)	−.075	(.087)	.078	(.127)
Combined labour market status, reference: both in work
Only one in work	−.08^**	(.036)	−.21	(.168)	.07	(.065)	−.049	(.106)
Both not in work	−.088	(.090)	−.24	(.328)	.15	(.111)	−.27	(.224)
Housing tenure, reference: own
Rent	.038	(.048)	−.26	(.188)	−.025	(.065)	−.51^***	(.133)
Other	−.07	(.089)	−.58^*	(.299)	.19	(.137)	−.082	(.328)
MR is Non-White, reference: White	.15	(.102)	.55^*	(.310)	.32^***	(.119)	−.25	(.170)
MR’s religion, reference: Christian
Non-Christian	−.3^***	(.113)	.04	(.289)	−.27	(.177)	−.31	(.242)
None	.0091	(.037)	.074	(.160)	−.028	(.049)	.33^***	(.095)
Language spoken at home, reference: english
English and other languages	.086	(.086)	−.15	(.226)	.043	(.122)	−.0085	(.197)
Only other	−.11	(.149)	−.51^*	(.300)	−.19	(.167)	−.24	(.327)
The interview was translated, reference: No	.21	(.145)	−.3	(.310)	−.16	(.190)	.72^**	(.300)
MR’s health status, reference: excellent
Very good, good	.032	(.041)	−.051	(.181)	.062	(.060)	−.038	(.092)
Fair, poor	−.025	(.055)	.42^*	(.233)	.0082	(.081)	.12	(.130)
CM’s health status, reference: some problems
No problems	.057	(.037)	.33^**	(.153)	−.093	(.063)	.022	(.120)
MR: I am a very private person, reference: strongly agree
Agree	−.13^**	(.067)	−.42	(.262)	.073	(.113)	.12	(.186)
Neither	−.13^*	(.068)	−.6^**	(.295)	−.14	(.116)	−.053	(.184)
Disagree	−.068	(.067)	−.43	(.281)	−.053	(.120)	.17	(.173)
Strongly disagree	−.2^*	(.104)	−.73^*	(.394)	−.28^*	(.167)	.22	(.273)
Can’t say	−.0077	(.156)	−.15	(.461)	.3	(.280)	.28	(.436)
Other	.044	(.116)	−.0059	(.397)	.79^***	(.150)	−.39	(.272)
Response history, reference: participated in all waves
Absent in at least one wave	.1	(.091)	−.27	(.321)	.17	(.123)	−.18	(.208)
Log OECD adjusted income	−.023	(.037)	.16	(.158)	.091^*	(.048)	−.48^***	(.089)
MR’s age	.00043	(.003)	−.038^**	(.015)	−.0075	(.005)	.0077	(.009)
Constant	−.94^***	(.249)	1.14	(1.092)	−1.82^***	(.375)	3.59^***	(.646)
N	11,251		494		9734		2011

Notes:

Standard errors in parentheses.

^* p < .10

^** p < .05

^*** p < .01.

Table 5. Interviewers’ effects for health record consent outcomes.

Consent outcomes	Base model	Area effects	Interviewer FE	N interviewers	N respondents
	Adjusted R^2
CM health W1	.032	.036	.077	328	11,745
CM health W2	.01	.015	.156	334	11,745
CM health W4	.020	.022	.120	443	11,745

Table

Wave		Wave 1, age 9 months		Wave 2, age 3 years		Wave 4, age 7 years
		Freq.	%	Freq.	%	Freq.	%
Marital status	Single	1650	14.0	1832	15.6	2644	22.5
	Couple	10,095	86.0	9913	84.4	9101	77.5
SES	Managerial and professional	5271	44.9	5385	45.8	5505	46.9
	Intermediate	1607	13.7	1669	14.2	1606	13.7
	Small employers and self-employed	823	7.0	927	7.9	1151	9.8
	Lower supervisory and technical	1291	11.0	1183	10.1	1048	8.9
	Semi-routine and routine	2753	23.4	2581	22.0	2435	20.7
Employment status	Both in work	5477	46.6	5536	47.1	6138	52.3
	One in work	5527	47.1	5593	47.6	5059	43.1
	Both not in work	741	6.3	616	5.2	548	4.7
Housing tenure	Own	7597	64.7	7946	67.7	8177	69.6
	Rent	3517	29.9	3424	29.2	3356	28.6
	Other	631	5.4	375	3.2	212	1.8
Religion	Christian	5637	48.0	5,643	48.0	4717	40.2
	Non-Christian	1124	9.6	1122	9.6	1119	9.5
	None	4984	42.4	4980	42.4	5909	50.3
Language spoken at home	English	10,296	87.7	10,185	86.7	10,945	93.2
	Half english	1098	9.3	1264	10.8	414	3.5
	Other	351	3.0	296	2.5	386	3.3
Interview translated	No	11,426	97.3	11,521	98.1	11,473	97.7
	Yes	319	2.7	224	1.9	272	2.3
MR health	Excellent	3684	31.4	3601	30.7	2723	23.2
	Good	6122	52.1	6107	52.0	7623	64.9
	Poor	1939	16.5	2037	17.3	1399	11.9
CM health	Some problems	4922	41.9	1892	16.1	1512	12.9
	No health problems	6823	58.1	9,853	83.9	10,233	87.1
Total		11,745		11,745		11,745

Table

Wave	W1 (Row %)	W2 (Row %)	W4 (Row %)
	Consenters	Consenters	Consenters
Living in a couple in wave 1
Single	93.9	81.9	93.6
Couple	95.2	83.3	93.2
Socio-economic status in wave 1
Managerial and professional	95.9	85.0	93.0
Intermediate	95.7	84.4	94.2
Small employers and self-employed	94.2	82.5	94.0
Lower supervisory and technical	94.5	79.6	92.9
Semi-routine and routine	93.4	80.0	93.2
Employment status in wave 1
Both in work	96.3	84.1	93.9
Only one in work	94.3	82.4	92.6
Both not in work	91.2	79.6	93.8
Housing tenure in wave 1
Own	95.7	83.7	93.4
Rent	94.1	81.4	93.2
Other	92.6	85.5	91.8
Ethnic group
White	96.4	83.3	94.0
Non-White	84.2	80.7	88.4
Religion in wave 1
Christian	95.9	83.4	93.5
Non-Christian	82.1	82.3	89.9
None	96.5	82.8	93.7
Language spoken at home in wave 1
English	96.2	83.3	93.5
Half english half other	86.2	81.6	89.9
Other	80.1	79.0	89.1
Whether the interview was translated in wave 1
No	95.5	83.1	93.4
Yes	75.7	77.4	87.4
Main respondent’s health status in wave 1
Excellent	95.8	83.3	93.4
Good	95.0	82.7	93.0
Poor	93.6	83.5	94.6
Cohort member’s health status in wave 1
Some problems	95.9	83.5	90.9
No problems	94.3	82.9	93.7
N	11,745