Skip to main content

Table 2 Results of 14 included studies

From: Psychological, behavioral and social effects of disclosing Alzheimer’s disease biomarkers to research participants: a systematic review

Authors, year Research question/design Sample (n) Outcome measures Instruments Time points Results
Studies on psychological effects
Romero et al., 2005 [30] Participants were asked how they felt after disclosure of their APOE genotype 76   Self-developed questionnaire   ε4-negative group ε4-positive group
Feeling depressed 1 month 0/49 8/27
4 months 0/49 5/27
10 months 0/47 4/27
Feeling worried 1 month 0/49 6/27
4 months 0/49 3/27
10 months 0/47 3/27
Feeling relieved 1 month 36/49 4/27
4 months 30/49 5/27
10 months 34/47 3/27
Gooding et al., 2006 (REVEAL QRI) [25] 56 participants in the REVEAL I study and four individuals who declined participation were interviewed about their reaction to the received results (either risk estimate and genotype or risk estimate only) 60a Coping: Interviews >12 months ε4-positive group Other (ε4-, ND, not REVEAL-ptc)
Relief 4/17 (24 %) were relieved: “Even with the ε4 allele, their risk was not as high as they had previously anticipated” 25/43 felt relievedb
Worry    Six participants, all ε4+, expressed greater concern about their AD risk after disclosure, describing their results as “depressing”, “frightening” and “disappointing” 14/39 who participated in REVEAL were neither relieved nor worried
Neither    7/17 were neither relieved nor worried. They related their “lack of emotion to the lack of predictability of the APOE test and the feeling that it only confirmed what they had already come to accept as their risk for AD”  
Green et al., 2009 [23] (REVEAL I) Participants were randomly assigned to receive their APOE genotype and a risk estimate (ε4-positive and ε4-negative group) or a risk estimate only (no disclosure group) 162     ε4-positive group ε4-negative group No disclosure group
Anxiety BAI (0–63) 6 weeks 5.2 (0.7)c 4.5 (0.6) 4.4 (0.7)c
6 months 4.6 (0.6) 3.9 (0.6) 4.6 (0.6)
12 months 4.4 (0.6) 4.2 (0.6) 4.2 (0.6)
Depression CES-D (0–60) 6 weeks 9.0 (1.0) 8.5 (0.9) 9.3 (1.0)
6 months 9.6 (1.0) 8.9 (1.0) 8.7 (1.0)
12 months 8.3 (0.9) 8.5 (0.9) 8.0 (0.9)
Distress IES (0–75) 6 weeks 9.4 (1.3)d 5.2 (1.3) 6.7 (1.4)
6 months 8.6 (1.2)d 4.2 (1.2) 8.9 (1.3)e
12 months 8.5 (1.3) 5.1 (1.2) 7.7 (1.5)
Ashida et al., 2010 (REVEAL II) [31] Participants in the REVEAL II study all received their APOE genotype, and were randomly assigned to either an extended or condensed disclosure protocol 269 Effect of results communication on psychological well-being: “Have you told anyone your APOE genetic test result? If so, who?” 6 weeks 62.1 % told a family member, 52 % their spouse or significant other, 37.5 % a friend and 14.9 % a health professional
Anxiety BAI 12 months Telling the result to a friend was associated with a decrease in BAI at 12 months (regression coefficient b = –0.17, p < 0.01)
Depression CES-D   Telling the result to a healthcare professional with a decrease in CES-D at 12 months (regression coefficient b = –0.10, p < 0.05)
Distress IGT-ADf (0–60) There was no association between results communication and IGT-AD levels of distress
24 participants (9 %) reported scores > than the clinical cutoff point of 16 on the CES-D at 12 months, and 21 of these 24 had scores < the cutoff at baseline
Lineweaver et al., 2014 [32] A group of research participants who were informed about their APOE genotype was compared with a matched group who did not receive their genotype on objective and subjective memory functioning 144 Objective memory functioning Subjective memory functioning Logical memory subtest Mean 8.2 months ε4-positive uninformed ε4-positive informed ε4-negative uninformed ε4-negative informed
immediate recall 31.9 (8.5)h 22.9 (7.8) 28.4 (7.8) 25.6 (5.9)
delayed recall Wechsler Memory Scale Revised (0–50)g 28.1 (9.3)h 17.7 (7.9) 24.3 (9.6) 20.6 (6.9)
Immediate recall delayed recall on Rey–Osterrieth Complex Figure test (0–20) 9.7 (2.7) 10.6 (2.6) 9.3 (3.7) 9.7 (3.1)
9.0 (3.1) 10.1 (2.4) 9.0 (4.0) 9.9 (2.8)
Metamemory in Adulthood Questionnaire:
Capacity subscale Change over time subscale (15 items 5-point Likert scale) 31.2 (5.2)i 28.0 (6.4) 27.8 (6.5)j 30.5 (4.7)
17.0 (5.4) 16.9 (5.5) 15.9 (4.9) 16.7 (4.1)
Memory Functioning Questionnaire:
Retrospective functioning 18.3 (5.8) 15.9 (4.7) 14.3 (4.1)k 16.7 (4.7)
Frequency of forgetting 90.5 (14.7) 83.4 (13.3) 84.5 (17.0) 88.9 (13.4)
Forgetting when reading 54.9 (8.4) 50.7 (9.3) 50.5 (13.1)k 55.5 (8.8)
Forgetting past events 19.2 (4.6) 18.3 (4.3) 18.1 (4.9) 19.9 (4.5)
Mnemonics usage (46 items with 7-point Likert scale) 23.1 (11.4) 20.3 (7.4) 22.8 (9.5) 25.1 (8.4)
Green et al., 2014 (REVEAL II) [29] Participants were randomly assigned to receive their APOE genotype in an extended protocol (SP-GC), a condensed protocol with a genetic counselor (CP-GC) or medical doctor (CP-MD). Aim was to assess whether the condensed protocols were equal (= mean score on any of the scales not more than 5 points higher) to the extended protocol 343     SP-GC CP-GC CP-MD
Anxiety BAI (0–63) 6 weeks 2.6 (0.5)c 3.6 (0.5) 4.3 (0.5)
6 months 3.2 (0.5) 3.1 (0.5) 4.4 (0.5)
12 months 3.0 (0.5) 3.7 (0.5) 3.9 (0.5)
Depression CES-D (0–60) 6 weeks 5.7 (0.7) 5.8 (0.7) 8.1 (0.7)
6 months 6.3 (0.7) 5.8 (0.7) 8.1 (0.7)
12 months 6.2 (0.6) 5.6 (0.6) 6.9 (0.6)
Distress IES (0–75) 6 weeks 2.8 (0.9) 5.1 (0.9) 8.2 (0.9)l
6 months 3.9 (0.9) 4.0 (0.9) 7.0 (0.9)l
12 months 3.4 (0.8) 3.3 (0.8) 5.5 (0.8)
Studies on behavioral effects
Zick et al., 2005 (REVEAL I) [24] REVEAL I, see Green et al., 2009 [23] 162 Insurance uptake Questionnaires on actual change in 6 weeks, 6 months, 12 months ε4-positive group ε4-negative group No disclosure group
Health insurance   12.5 5.56 6.52
Life insurance   2.08 7.41 6.52
Disability insurance   4.17 3.70 4.35
LTC insurance (%)   16.7m 1.85 4.35
Planned change in
Health insurance 25.0 13.0 23.9
Life insurance 16.67 5.56 4.35
Disability insurance 18.8 7.41 8.70
LTC insurance (%) 45.8m 22.2 32.6
OR for actual change in LTC insurance ε4+ compared with ND: 5.76 (p < 0.1)
Taylor et al., 2010 (REVEAL II) [34] REVEAL II, see Green et al., 2014 [29] 276 LTC insurance Not reported Not reported Two ε3 traits ≥ one ε4 trait At least one ε2 trait, no ε4 trait
OR of actual or planned change 1.00 2.31 (95 % CI 1.11–4.81) 1.55 (95 % CI 0.43–5.60)
Absolute probability of changing LTC insurance    0.087 0.237 0.149
Chao et al., 2008 (REVEAL I) [22] REVEAL I, see Green et al., 2009 [23] 162 Health-related behavior changes Yes/no questions about changes in: 12 months ε4-positive group ε4-negative group no disclosure group
Any behavior specific to AD prevention 52n 24 30
Medications/vitamins 40 20 28
Diet 20 11 7
Exercise (% endorsing) 8 4 5
OR of any behavior change ε4+ vs ε4– group: 2.73 (95 % CI 1.14–6.54)
Vernarelli et al., 2010 (REVEAL II) [33] REVEAL II, see Green et al., 2014 [29] 272 Changes in supplement use Yes/no questions with free-text field on changes in: overall diet use of dietary supplements exercise (ORs) 6 weeks ε4-positive group ε4-negative group
1.56 (95 % CI 0.80–3.02) 1.00
4.75 (p < 0.0001) (95 % CI 2.23–10.10) 1.00
1.85 (95 % CI 0.96–3.57) 1.00
Of 45 participants reporting a change in supplement use, 32 (71.1 %) were ε4+ (p < 0.0001). Of these 45, 38 were in the condensed protocol (84.4 %) and seven in the extended (15.5 %) (p = 0.006)
Christensen et al., 2015 (REVEAL II/III) [27] Secondary analyses were performed on data from the REVEAL II and III study. For REVEAL II, see Green et al., 2014 [29]. In the REVEAL III study, in-person and phone disclosure, and giving AD genetic info only and pleiotropic info were compared 795 Associations between recruitment status (actively recruited (ARP) or self-referred (SRP)) and behavior changes and advance planning Yes/no questions on actual and planned changes in: 6 weeks, 12 months Self-referred participants were more likely than ARPs to report changes to exercise at 12 months (35 % vs 25 %, p = 0.032). No other differences between recruitment cohorts were noted on changes or plans to change health behaviors
Health behavior: Mental activities Diet Exercise Dietary supplements Medications Secondary analyses showed that the impact of genetic risk status on certain behavior changes differed by recruitment cohort. ε4-positive participants were more likely than ε4-negative participants to report changes at 6 weeks to mental activities and diet, but only if they had self-referred to the study, although only differences in changes to mental activities persisted through the 12-month follow-up
Advance planning: LTC insurance Retirement plans No direct associations with self-referral were observed on either LTC insurance coverage or retirement plans. An interaction effect was observed (p = 0.005): self-referred ε4-positive participants were more likely than ε4-negative participants to report intentions to change LTC coverage, but no differences were noted among ARPs
No associations were noted on retirement plans, except greater intentions to change among ε4-positive participants compared with ε4-negative participants, regardless of recruitment cohort (p < 0.001)
Studies on hypothetical disclosure
Caselli et al., 2014 [28] Members of an online community for people interested in AD prevention research completed a survey on their interest in and anticipated reaction to hypothetical genetic and biomarker testing and disclosure 4036 Results communication Multiple-choice questions Not applicable If APOE ε4 positive, you would tell: (%)
 Physician 79.4
 Spouse 92.3
 Siblings 84.6
 Children 81.7
 Friends 53
 Lawyer 60
If biomarker evidence of AD, you would tell:
 Spouse 92.2
 Siblings 80.6
 Children 75.9
 Friends 46.5
 Lawyer 53.8
Behavior changes If APOE ε4 positive, you would: (%)
 Begin a healthier lifestyle 90.5
 Get LTC insurance 76.3
 Spend all your money forpleasure 18.4
 Seriously consider suicide 11.6
If biomarker evidence of AD, you would:
 Begin a healthier lifestyle 91
 Get LTC insurance 76.6
 Spend all your money for pleasure 18.7
 Seriously consider suicide 10.2
Studies on social effects
Ashida et al., 2009 (REVEAL II) [35] REVEAL II, see Green et al., 2014 [29] 271 Communication of APOE genetic test result “Have you told anyone your APOE genetic test result? If so, who?” 6 weeks Person Frequency (%)
Anyone 81.5
Family member 63.8
Spouse/significant other 50.9
Friends 34.7
Health professional 12.2
OR of results communication to health professional in condensed vs extended protocol 5.19 (95 % CI 1.50–17.89, p < 0.01)
Chilibeck et al., 2011 (REVEAL I) [26] Interviews were conducted with participants from REVEAL I and with FDRs of AD patients who did not undergo genetic testing 79 (REVEAL participants) and 40 (non-REVEAL) Effect of personalized genetic information on conceptualization of personal risk, family health and familial relationships Open-ended interviews Not reported Drawn from the report: when genetic information corresponds with previous beliefs about risk and inheritance, people emphasize how the information provided by genetic testing is “not new” to them but only confirms what they already knew or at least suspected. Yet risk predictions generated by genetic technologies sometimes conflict with those rooted in everyday beliefs about heredity. The visible evidence of risk provided by family history is often more compelling than that based on a genetic test
Conclusion: “The genetic risk information given to subjects in the REVEAL trial is interpreted through a process of ‘familiarization’ in which risk estimates are absorbed into and embedded within pre-existing beliefs about who in the family will succumb to AD. These narratives resemble those of individuals from AD families who have not been genetically tested, strongly suggesting that ideas about embodied risk for AD in families are not dramatically changed as a result of genetic testing.”
  1. Of the 14 studies included in the systematic review, two report on two types of effects. In this table they are taken up only once. Ashida et al. [31] report on both psychological and social effects; results from this study are to be found among the studies on psychological effects in this table. Caselli et al. [28] report on both behavioral and social effects; results from this study are to be found among the studies on behavioral effects in this table
  2. AD Alzheimer’s disease, APOE apolipoprotein E, ARP actively recruited participants, BAI Beck Anxiety Inventory, CES-D Center for Epidemiological Studies Depression Scale, CI confidence interval, CP-GC condensed protocol genetic counselor, CP-MD condensed protocol medical doctor, FDR first-degree relative, I informed, IES Impact of Event Scale, IGT-AD Impact of Genetic Testing for Alzheimer’s Disease, LTC long-term care, ND no-disclosure group, OR odds ratio, RCT randomized controlled trial, REVEAL Risk Evaluation and Education for Alzheimer’s Disease Study, SP-GC standard protocol genetic counselor, SRP self-referred participants, UI uninformed
  3. aAlthough the question of one’s response to the results is not applicable to those who declined participation in REVEAL (n = 4), the original article does not leave those individuals out of this part of the results section
  4. bFifteen participants were in the no-disclosure group in REVEAL and received a lifetime risk estimate only. This group is not analyzed separately in their reaction to the results
  5. cScores are mean values ± SE
  6. dDifference ε4+ vs ε4– is significant at 6 weeks and 6 months (p < 0.05)
  7. eDifference ND vs ε4– is significant at 6 months (p < 0.05)
  8. fIGT-AD based on the Multidimensional Impact of Cancer Risk Assessment Questionnaire
  9. gOn all memory scales, higher scores indicate better objective/subjective memory functioning. Scores are mean values ± SD
  10. h p ≤ 0.001 ε4+ UI vs ε4+ I
  11. i p < 0.05 ε4+ UI vs ε4+ I
  12. j p < 0.05 ε4–UI vs ε4– I
  13. k p < 0.05 ε4–UI vs ε4– I
  14. lNoninferiority of CP-MD vs SP-GC could not be confirmed for test-related distress at 6 weeks and 6 months
  15. m p < 0.05 ε4+ vs ε4–
  16. nε4+ vs ε4–, p = 0.003; ε4+ vs ND, p = 0.03