Table 2.

Results with regard to behaviour change, psychological responses and medical consumption

Author (year), larger project if applicable	Behaviour change	Psychological responses	Medical consumption
Bloss et al. (2011b), SGHI	Dietary fat intake (BDFS): no significant change in dietary fat intake (BL = 16.0 ± 7.9, 3-month FU = 15.2 ± 7.5, p = 0.89) Leisure time exercise (GLTEQ): no significant change in leisure time exercise level (BL = 28.6 ± 23.0, 3-month FU = 28.6 ± 22.9, p = 0.61) Change in vitamin use = 20.5% of study population indicated a changed pattern of use (18% increased or started, 2.5% decreased or quit; BL = 82.8% used vitamins) Change in alcohol use = 11.2% indicated a changed pattern of use (2.0% increased or started, 9.2% decreased or quit; BL = 80.0% used alcohol) Change in tobacco use: of pre-test smokers 17.6% had quit smoking and 21.8% of had decreased their use. Of all participants, 7.6% had increased or started smoking	State anxiety (state anxiety subscale of SSTAI, <39 considered low-anxiety state): no significant change in state anxiety at FU (BL = 35.2 ± 9.6, 3-month FU = 34.6 ± 10.0, p = 0.80) Test-related distress (IES-R, score of >8 and >23 points on avoidance and intrusion subscales indicate “some impact” and “clinically relevant distress”, respectively): No significant TRD (3-month FU = 3.2 ± 7.1, 90.3% scored <8 and 97.2% scored <23)	Discussing results with Navigenics genetic counsellor = 10.4% Sharing with physician or HCP = 26.5% Sharing was not associated with changes in test-related distress or anxiety level. Only sharing with physician was associated with lower dietary fat intake and increased exercise behaviour at follow-up
Bloss et al. (2013), SGHI	Dietary fat intake (BDFS): no significant change in dietary fat intake (BL = 15.9 ± 7.8, 1-year FU = 14.8 ± 7.3, p = 0.34) Leisure time exercise (GLTEQ): no significant change in leisure time exercise level (BL = 27.9 ± 23.1, 1-year FU = 29.8 ± 24.4, p = 0.39)	State anxiety (state anxiety subscale of SSTAI, <39 considered low-anxiety state): no significant change in state anxiety at FU (BL = 35.0 ± 9.5, 3-month FU = 34.2 ± 10.0, p = 0.50) Test-related distress (IES-R), score of >8 and >23 points on avoidance and intrusion subscales indicate “some impact” and “clinically relevant distress”, respectively): No significant TRD, significant decrease in TRD between 3-month and 1-year FU (p = 0.03) (1-year FU = 1.2 ± 3.4, 96.8% scored <8 and 99.7% scored <23)	Discussing results with Navigenics genetic counsellor = 14.1% Sharing with physician or HCP = 39.5% Discussing results with a Navigenics genetic counsellor was associated with higher anxiety, TRD and actual screening test completion, but not with fat intake, exercise behaviour or the intention to complete screening tests with greater frequency Sharing results with a physician or HCP was associated with higher exercise, actual screening test completion and intention to increase frequency of screening, but not with anxiety, TRD or fat intake. Change in health screening behaviours: the mean number of screening tests completed since receiving test results was 5.3 ± 2.8 at 1-year FU. There was no significant difference with the mean at 3-month FU (p = 0.43)
Boeldt et al. (2015), SGHI	Dietary fat intake (BDFS) and leisure time exercise (GLTEQ): no significant relationships between perceived control or genetic risk and fat intake or leisure time exercise at 3-month FU	Test-related distress (IES-R): higher genetic risk and low perceived control associated with higher distress Anxiety: participants receiving a high genetic risk and perceiving their most feared disease as controllable via lifestyle changes, experienced the lowest levels of anxiety at 3-month FU	Sharing with physician or HCP: higher genetic risk associated with greater likelihood of sharing with physician Change in self-check for breast cancer: higher perceived control and high genetic risk associated with increased likelihood of self-checking
Carere et al. (2016), Pgen study		Anxiety (GAD-2): positive GAD-2 screen for anxiety/panic disorder at 6-month FU = 14.5% (BL = 15.8%; p = 0.33) Decision regret (score 0–100) = 58.4% reported no decision regret, 97.4% reported ≤40/100	Consultation with HCP after testing = 34.9%
Darst et al. (2013), SGHI			Reasons for using GC: most mentioned reasons were taking advantage of a free service, wanting more information on risk calculations, and because they were contacted by the GC. The least commonly reported reason was a perceived lack of understanding of one’s results. Reasons for not using GC: most mentioned reason was feeling to already understand the results. Outcomes of consulting a GC: improved understanding of their own results and feeling more educated about genetics in general. For 30.4% consulting a GC made them more likely to discuss their results with their physician
Darst et al. (2014), SGHI			Demographic characteristics of subjects who shared with physician or HCP: sharers were older (50.0 ± 12.5 vs. 45.4 ± 11.6; p = <0.001), had a higher annual income (150,000–199,000 vs. 100,000–149,000; p = 0.1), more likely to be married (73.2 vs. 68%; p = 0.005) and more likely to identify with a religion (80.5 vs. 74.2%; p = 0.004). Behavioural characteristics of subjects who shared with physician or HCP: sharers reported a greater amount of exercise per week (GLTEQ; 27.0 ± 24.9 vs. 22.3 ± 23.3; p = 0.003), lower dietary fat intake (BDFS; 15.0 ± 8.1 vs. 16.0 ± 7.8; p = 0.02) and more frequent physician visits per year (3.8 vs. 3.3; p = <0.001) Other characteristics of subjects who shared with physician or HCP: fewer sharers reported overall concerns related to testing (44.6 vs. 52.7% yes; p = 0.001) and privacy issues about the data (33.2 vs. 38.5% yes; p = 0.03). More sharers reported to greatly value risk information (78.2 vs. 68.8%)
Egglestone et al. (2013)a	Any change in health behaviour due to test results = 27.3% Healthier diet = 12.7% Stopped or reduced caffeine intake = 1.6% Taking vitamins or supplements = 3.2% More exercise = 6.3% Lost weight = 1.6% Other (e.g. wearing sunglasses) = 3.7% Generally reducing risk conditions = 1.1% Looking into high risk items (either research, talking with doctor or medical tests) = 1.6% Preventative checks such as eye tests = 2.1%	Any change in health anxiety due to test results = 23.8% Decreased health anxiety due to test results (out of total n) = 15% Increased health anxiety due to test results (out of total n) = 1.6% General health anxiety: no significant difference in anxiety between consumers and potential consumers (3.80 vs. 4.22; p = 0.63) Anxiety about developing a serious disease: no significant difference in anxiety between consumers and potential consumers (3.52 vs. 3.92; p = 0.30)	Preventative checks such as eye tests = 2.1%
Gordon et al. (2012), CPMC	Change in lifestyle = 33%		Shared results with a HCP = 42% (additional 23% intended to share but had not yet done so) Characteristics of subjects who shared with physician or HCP: more older than younger participants shared (53 vs. 29%) Reasons for sharing: most mentioned reason was so the provider could take some action or offer advice to reduce risk. Only 1 participant (1.6%) shared to ask the HCP to help interpret the results
Haga et al. (2014)		Distress (MICRA, range 0–30): very low levels of distress at 1-week FU (mean 2.27). Distress scores were not different between participants with and without an increased risk for T2DM. Anxiety and depression (IPQ-R emotional representations subscale, range 6–30): low levels based on T2DM results (mean 11.9)
James et al. (2011)		Worry: no significant differences between intervention and control group in percentages of participants reporting being somewhat or very worried about developing a number of tested diseases at 1-week FU and 1-yeaR FU (range = 40–70%)
Kaphingst et al. (2012), MI	Measured at 3-month FU: Information seeking about the effect of personal health habits on risk of health conditions = 65% Information seeking about the effect of family history on risk of health conditions = 36%	Measured on a scale of 1 (not at all)–7 (great deal) at 10-day FU: Nervous = 2.6 (1.7) Afraid = 1.8 (1.5) Confused = 1.7 (1.3) Regretful = 1.3 (0.9) Reactions were not associated with the number of variants with increased risk carried	Measured at 3-month FU: Discussed test results with someone = 77% Discussed test results with a HCP = 1% Discussed results with family = 18% Discussed results with spouse = 20%
Kaufman et al. (2012)	More careful about diet after viewing results = 33.3% (participants with a positive family history were more likely to be more careful about their diet (p = 0.03)) Changed at least 1 medication or supplement regimen = 16% Changed a dietary supplement = 10% (participants with poorer self-perceived health were more likely to change their supplement regimen (p = 0.007)) Exercising more = 14% Sought additional information about at least 1 health condition covered in their test = 43%		Shared results with at least 1 HCP = 28% Shared results with GP = 20% Shared results with genetic counsellor = 1% Shared results with other HCP = 19% Shared results with more than 1 provider = 11% Characteristics of sharers with HCP: participants reporting getting regular physical exams and those with reported poorer self-perceived health were significantly more likely to share with a HCP (p = 0.001 and p = 0.03, respectively). Result of sharing: people who shared with their HCP were significantly more likely to get a follow-up test, change a prescription or supplement regimen and be more careful about their diet. Of the people who shared with a HCP, 49% said they learned something new and useful from their results compared to 27% among non-sharers (p < 0.0001). Follow-up with additional laboratory tests as a result of receiving their data = 10% (26% of people who shared with a HCP compared to 2% of non-sharers; participants with a positive family history were more likely to have gotten follow-up tests (p = 0.001))
Lee et al. (2013)			Shared results with a HCP = 56.6% Shared results with family = 98.3% Shared results with friends = 81.7% Sharing results online (23andMe website, Facebook, other platforms) = 66.7% Sources consulted for help: mostly internet websites (70.4%) and 23andMe Help (53.7%), compared to 22.7% from a HCP
McGrath et al. (2016)			Shared results with anyone = 82.2% Shared results with a medical professional = 10.7% Shared results with family = 40.9% Shared results with spouse = 38.5%
McGuire et al. (2009)			Shared results with physician = 53%
Olfson et al. (2016), Pgen	Quit smoking = 22% (of current smokers) Started smoking = 1% (of never and former smokers)
O’Neill et al. (2015), MI		Psychological responses to testing Neutral = 53.9% (most frequently were “not surprised” at 14.2% and “surprised” at 10.8%) Positive = 26.9% (most common was “interesting” at 9.2%) Negative = 19.2% (most common was “nervous” at 6.5%) No significant differences by gender, race and education were found with regard to frequency of reported responses within the three categories
Reid et al. (2012), MI			Difference 12 months prior to and 12 months post testing, comparing people who completed genetic testing and those who were lost or excluded before testing: Visited HCP (% of participants): no significant difference Number of physician visits: no significant difference Difference 12 months prior to and 12 months post testing, comparing people who completed genetic testing and those who were lost or excluded before testing: Laboratory tests or procedures (% of participants): no significant difference
van der Wouden et al. (2016), Pgen			Discussed results with at least 1 HCP = 34.7% Discussed results with GP = 27.1% Discussed results with genetic specialist = 1% Characteristics of sharers: participants who discussed with a HCP were more frequently women, did not have a college degree (for GP only), were parents and had a positive screen for baseline anxiety

BDFS Block Dietary Fat Screener, GAD-2 two-item Generalized Anxiety Disorder screener, GLTEQ Godin Leisure-Time Exercise Questionnaire, IES-R Impact of Events Scale-Revised, IPQ-R Illness Perception Questionnaire Revised, MICRA Multidimensional Impact of Cancer Risk Assessment, SSTAI Spielberger State–Trait Anxiety Inventory, T2DM type 2 diabetes mellitus, TRD test-related distress, FU follow-up, BL baseline

^aSome values have been recalculated with “actual consumers” only (n = 189)