Reported Interest in Notification Regarding Use of Health Information and Biospecimens

Abstract

This study describes reported interest in notification regarding use of personal health information and biospecimens for research and preference-associated factors among a sample of the US population in 2019.

Research with deidentified health information or biospecimens does not currently require notification to, or consent from, participants. Federal regulators recently contemplated changing research regulations such that all biospecimens would be considered “inherently identifiable,” meaning that biospecimens without any additional identifying information (eg, name, birthday) would still require, in some cases, notification or consent.¹ Regulators argued that this change was warranted because failing to recognize the autonomy interests of participants in research using their biospecimens (vs health information) could “diminish public support” for research.¹

Treating all biospecimens as inherently identifiable would significantly limit their use.² Most who commented on this proposal opposed it,³ and the final 2018 regulations did not immediately adopt it—although they require agencies to continually reassess this decision.⁴

We explored whether a sample of the US population reported interest in notification regarding use of their health information and biospecimens for research and preference-associated factors.

Methods

We conducted an online survey (or by telephone if respondents lacked online access) in spring 2019 using the National Opinion Research Center’s (NORC) nationally representative AmeriSpeak Panel with English-speaking US adults. Black and Hispanic individuals and respondents with annual household incomes below 200% of the federal poverty level were oversampled (see Table 1 footnote a for information on race and ethnicity).⁵ Poststratification survey weights were applied to account for oversampling and enable population-level inferences. The University of Michigan institutional review board approved this protocol; NORC obtained oral or written consent.

Table 1. Demographic Characteristics of Survey Respondents: Weighted Descriptive Statistics (N = 2054).

Characteristics	No. (weighted %)
Sex
Female	1046 (51.7)
Male	1008 (48.3)
Age, mean (SD), y	49 (16.6)
Race and ethnicitya
Black, non-Hispanic	321 (11.6)
Hispanic	400 (15.6)
White, non-Hispanic	1194 (64.2)
Otherb	139 (8.6)
Education
No high school diploma	81 (8.5)
High school graduate or equivalent	352 (29.2)
Some college	939 (27.4)
Bachelor’s degree or above	682 (34.9)
Household income, $
<30 000	626 (24.0)
30 000-84 999	907 (44.5)
≥85 000	521 (31.5)

^a The combined race and ethnicity variable was provided by the National Opinion Research Center, University of Chicago. Respondents could self-select (1) White, non-Hispanic, (2) Black, non-Hispanic, (3) other, non-Hispanic, (4) Hispanic, (5) Asian, non-Hispanic, or (6) 2 or more other categories, non-Hispanic. Race and ethnicity were assessed in this study due to a priori hypothesis that several measures, including one assessing experiences of discrimination, would be associated.

^b Includes other, non-Hispanic; Asian, non-Hispanic; and 2 or more of other categories, non-Hispanic.

Respondents watched a short video describing current policy related to health information, sharing, and protections. Respondents were asked “how true” the following statement was on a 4-point Likert scale (not true [1]; somewhat true; fairly true; or very true [4]): “I would like to be notified if [user type] use my [identifiability type] [material type].” Attributes were defined, changing the user (commercial vs university researcher), identifiability (identifiable vs deidentified), and material (health information vs biospecimen), for a total of 8 questions (eAppendix in the Supplement), presented in randomized order.

Weighted descriptive statistics and paired t tests compared the means of single questions and combined indexes (eg, all 4 questions regarding university use vs all 4 questions regarding commercial use) to examine interest in notification across user type, identifiability, and type of material (Stata version 17.0 SE; StataCorp). Statistical significance was defined as P < .05 in 2-tailed tests.

Results

Of 3253 surveys sent, 2054 individuals (63.1%) responded to all relevant measures (51.7% female; mean age, 48 [SD, 16.6] years; 64.2% non-Hispanic White) (Table 1).

Most respondents reported that it was “very true” they were interested in notification regarding commercial use of their health information (both identified [70.5%; 95% CI, 67.4%-73.3%] and deidentified [50.4%; 95% CI, 47.4%-53.5%) and specimens (both identified [72.1%; 95% CI, 69.3%-74.8%] and deidentified [50.4%; 95% CI, 47.4%-53.4%]). Most also reported that it was “very true” they were interested in notification regarding university use of their identified health information (70.5%; 95% CI, 67.6%-73.2%) and specimens (64.9%; 95% CI, 61.8%-67.9%); however, 40.1% (95% CI, 37.1%-43.1%) were interested in notification for university use of deidentified health information and 40.1% (95% CI, 37.2%-43.1%) for specimens.

Comparing combined indexes, respondents were significantly more likely to want notice regarding all commercial use (mean, 3.20; 95% CI, 3.16-3.24) vs university use (mean, 3.09; 95% CI, 3.05-3.12; P < .001) as well as all use of identified (mean, 3.43; 95% CI, 3.39-3.46) vs deidentified (mean, 2.86; 95% CI, 2.81-2.90) health information and specimens (P < .001). Respondents were also more likely to want notice regarding use of their health information (mean, 3.15; 95% CI, 3.12-3.19) vs biospecimens (mean, 3.13; 95% CI, 3.10-3.17; P = .01) (Table 2).

Table 2. Participant Preferences for Notification About Use of Identified or Deidentified Health Information and Biospecimens by Commercial Companies and University Researchers (N = 2054).

Question	For you, how true are the following statements: I would like to be notified about [I] using my [II] [III].			Mean (95% CI)a	P value
	I	II	III
A	Commercial companies	Identified	Biospecimens	3.45 (3.41-3.50)
B	Commercial companies	Deidentified	Biospecimens	2.94 (2.89-2.30)
C	Commercial companies	Identified	Health information	3.45 (3.41-3.49)
D	Commercial companies	Deidentified	Health information	2.96 (2.91-3.01)
E	University researchers	Identified	Biospecimens	3.37 (3.33-3.41)
F	University researchers	Deidentified	Biospecimens	2.76 (2.71-2.81)
G	University researchers	Identified	Health information	3.44 (3.40-3.48)
H	University researchers	Deidentified	Health information	2.77 (2.72-2.83)
Comparison of composite indexes of preference for notificationb
Index 1		Index 2		Mean difference (95% CI)
Questions used in index	Mean (95% CI)a	Questions used in index	Mean (95% CI)a
All commercial companies (A, B, C, D)	3.20 (3.16-3.24)	All university researchers (E, F, G, H)	3.09 (3.05-3.12)	0.12 (0.09-0.14)	<.001
All identified materials (A, C, E, G)	3.43 (3.39-3.46)	All de-identified materials (B, D, F, H)	2.86 (2.80-2.90)	0.57 (0.52-0.61)	<.001
All health information (C, D, G, H)	3.15 (3.12-3.19)	All biospecimens (A, B, E, F)	3.13 (3.10-3.17)	0.02 (0.00-0.04)	.01

^a Mean responses on the following Likert scale: 1 = not true; 2 = somewhat true; 3 = fairly true; 4 = very true.

^b Indexes are generated as the sum of participants’ answers to questions A through H divided by the number of questions answered.

Discussion

A national sample of US adults reported interest in notification of research using their health information and specimens. Interest was associated with user, identifiability, and material type. Limitations include that the survey did not ask whether respondents had participated in previous research and did not assess comprehension. There are arguments both for and against requiring notification to or consent from individuals for research use of their biospecimens.^2,6 Understanding attributes associated with patients’ preferences for notification may assist in designing such policy.

Section Editors: Jody W. Zylke, MD, Deputy Editor; Kristin Walter, MD, Senior Editor.

Supplement.

eAppendix. Reported Desire for Notice Regarding Use of Health Information and Biospecimens