Vox Sanguinis (2014) 107, 132–139 © 2014 International Society of Blood Transfusion DOI: 10.1111/vox.12143

ORIGINAL PAPER

Donors’ psychological reactions to deferral following false-positive screening test results G. Delage,1 G. Myhal,1 Y. Gr
egoire1 & G.-M. Simmons-Coley2 1

Medical Affairs, Hema-Quebec, Montreal, QC, Canada Quality Assurance, Hema-Que bec, Montreal, QC, Canada

2

Background and Objectives Being notified of a false-positive infectious disease marker result can cause psychological distress in blood donors. A new notification process, informing donors of the possibility of re-entry, was compared with the previous one in which donors were indefinitely deferred to evaluate the mitigating effect on donors’ psychological distress levels. Materials and Methods Two groups of donors, ‘deferred donors’ (DD) and ‘donors eligible for re-entry’ (DER), completed a questionnaire involving 5-point scales. Levels of psychological distress, attitude towards blood donation, desire to donate blood in the future and perception of notification process quality were assessed. Results Attitudes towards blood donation (P = 0
0276) (DD: 3
94 – 0
11 vs. DER 4
21 – 0
09) and perceived quality of communication (P = 0
0108) (DD: 2
72 – 0
12 vs. DER 3
08 – 0
10) were significantly improved with the new notification process. No significant difference was found between groups in psychological distress levels or desire to donate blood in the future.

Received: 4 October 2013, revised 5 February 2014, accepted 11 February 2014, published online 20 March 2014

Conclusion Informing donors of the possibility of re-entry appears to contribute to maintaining a positive predisposition towards future blood donation. It does not, however, appear to alleviate the distress felt after being notified of a falsepositive infectious disease marker result, nor does it increase willingness to give blood again in the future. Key words: attitudes, deferral, donor reactions, false positive, psychological distress, retention.

Introduction Donors repeatedly reactive for infectious diseases screening tests are notified of such and deferred. Studies have shown donors experience a range of negative emotional and behavioural reactions as a result, such as worry, upset, confusion, shock, disbelief, panic, fear, anger, stigmatization and a sense of loss [1–3]. Even short-term temporary deferral has a negative impact on return rates for blood donors [4]. Correspondence: Gilles Delage, 4045 C^ote-Vertu, Ville Saint-Laurent, QC H4R 2W7, Canada E-mail: [email protected] Institute where the work was conducted: H
ema-Qu
ebec.

132

Patient understanding of blood test results has also been identified as a problem. This can impact donors’ understanding of their deferral and infectious status, and whether or not they will seek care [3, 5]. However, there is no consensus as to how donors should ideally be notified of test results. Thus, different authors have explored the impact of communicating screening test results and deferral status on donors’ feelings and emotions. Feelings and emotions directly influence a donor’s attitude, which represents the individual’s favourable or unfavourable assessment towards adopting a certain behaviour [6]. This conceptual linkage is important, because attitude affects donors’ intentions to donate and intention is the best predictor of behaviour [6–8].

Donors’ reactions to deferral 133

Little research examines the direct impact of notifying donors of false-positive screening test results on attitudes towards blood donation, intention to donate and return rates (where applicable). Furthermore, we have found no specific study examining the impact of a change in notification approach on deferred donors’ psychological distress, attitude towards blood donation, willingness to donate blood in the future and donor return rates (in the case of temporary deferral). Our research seeks to fill this gap.

Materials and methods Research questions and hypotheses Donors notified of an indefinite deferral for an unconfirmed screening test result receive a mixed message [9] that summarily means to the donors: « The result is probably a false positive, with no consequence to your health, but you can no longer give blood ». This can potentially create both confusion and psychological distress. In this situation, a notification process offering donors re-entry conditional on obtaining negative results in subsequent control tests sends a more reassuring and non-contradictory message, summarily: « You had a false positive test results following your recent blood donation but you can still give blood if you undergo additional tests that turn out negative ». We hypothesized that this change would send a clearer message, reduce psychological distress and improve donors’ attitude towards blood donation and their willingness to once again become active blood donors. H
ema-Qu
ebec is the organization in charge of blood collection in the Canadian province of Qu
ebec. It holds around 3000 blood drives and collects blood from over 160,000 donors yearly. Prior to 25 May 2010, H
emaQu
ebec donors who showed positive test results for syphilis, HIV, HBV, HCV, anti-HBc and HTLV, regardless of the results of confirmatory testing, were notified that they were indefinitely deferred. After that date, having received from the Canadian regulatory authority the permission to offer donors a re-entry process for false-positive screening tests for HIV, HBV (HBsAg), HCV and syphilis, H
ema-Qu
ebec changed its notification letters accordingly. This re-entry process involves donors undergoing further screening tests at H
ema-Qu
ebec after a waiting period of 3 months for all markers except HCV, for which the waiting period is 6 months. If these test results are negative, the donor may once again give blood. This change in the notification process created an opportunity to evaluate the impact of the change in notification process on donors. Our study compared the impact of the new communication approach to that used before 25 May 2010 to © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

evaluate its influence on four variables: donors’ psychological distress, donors’ attitudes towards blood donation, donors’ willingness to once again become active blood donors and donors’ perception of the notification process.

Subjects: two groups of donors were compared Deferred donors group (DD) These donors received a letter notifying them of an indefinite deferral following false-positive screening test results, without possibility of re-entry, as was standard practice prior to 25 May 2010.

Group of donors eligible for re-entry (DER) These individuals received a letter notifying them of a deferral following a false-positive screening test result, with the possibility of re-entry if the results of further testing were negative. For each donor recruited in the DER, we recruited a donor in the DD matched for the characteristics listed in Table 1, to ensure comparability and limit selection bias [10]. In an attempt to limit recall bias, a potential threat to the internal validity of studies using self-reported data that increases with the passage of time [10, 11], we only included in our DD individuals who had been informed of their screening test results after December 2007. Recruitment began in July 2010 and ended in July 2011 for the DER. For the DD, it began in September 2010 and ended in November 2011. Donors recruited in the DD were not advised of their eligibility for re-entry until they had completed the questionnaire. Data were collected using the same method for both groups.

Recruitment Recruitment was done as follows: potential participants were sent via mail a package including a letter inviting them to participate, an informed consent form, a copy of

Table 1 Pairing characteristics and modalities Characteristics

Pairing modalities

Gender

Identical pairing – each individual from the DD is of the same gender as their DER counterpart. Approximate pairing – necessary to identify a number of potential DD participants sufficient to meet the studys recruitment needs. Each individual in the DD has a maximum age difference of plus or minus 5 years as compared with their DER counterpart. Identical pairing – each individual in the DER was matched to a DD individual according to virological marker (anti-HIV, anti-HCV, HBsAg and syphilis).

Age

Virological marker

134 G. Delage et al.

the questionnaire and a preaddressed stamped envelope in which to return the questionnaire and the signed consent form. If no reply was received after 3 weeks, participants were contacted via telephone to encourage participation; if necessary, a new package was sent. Participants received another telephone reminder after a further 3 weeks if they had not replied. Further silence was considered refusal to participate. We recruited 53 individuals per group, 106 individuals in total, which was sufficient to detect a medium or greater effect size [12, 13] between the two groups with regard to their level of psychological distress, their attitude towards donation and their willingness to donate for the tests described in the ‘statistics’ section (Power = 0
8; alpha = 0
05). (The minimum required was 52 respondents per group.) Response rates were 44% (61/139) for the DER and 37% (72/195) for the DD, which is comparable with other studies exploring similar themes [2, 6]. On average, 74 – 29 days (range: 19–199 days) were elapsed between the deferral and the reception of the completed questionnaire for the DER, while 835 – 270 days (range: 78–1266 days) were elapsed for the DD.

Study and instrument design Based on the information available in the scientific literature, a questionnaire was developed. It was pretested with five potential respondents, and minor revisions were made to enhance question clarity. The questionnaire was developed in French and translated to English. It included sections exploring the four following themes: Respondents’ perception concerning aspects of the notification process. This section included 13 questions using a five-point Likert scale (where 1 = ‘completely disagree’ and 5 = ‘completely agree’), which combined into an index score. Questions were based on five facets of notification and/or patient communication identified as important in the literature: notification letter clarity, information sufficiency, tone of communication, trustworthiness of information and participants’ understanding of the information [2, 3, 14]. The psychological distress felt by respondents was measured by asking respondents to rate themselves on a fivepoint Likert scale (where 1 = ‘completely disagree’ and 5 = ‘completely agree’) using 19 different statements, such as ‘I was scared’. These statements were based on themes present in the literature, such as fear, shock and a sense of loss [1] as well as themes from another measurement instrument, the Psychological General Well-Being Index, such as anxiety, depression and stress [12, 15, 16]. These combined into an index score. We measured both the psychological distress reportedly felt at the time respondents learned of their test results and deferral sta-

tus (‘initial distress’) and that felt at the time they completed the questionnaire (‘current distress’). There was a minimum lag of 1 month between measures, as we granted potential participants 1 month postnotification before contacting them for this study. Respondents’ general attitude towards blood donation was measured by asking participants to rate themselves on a five-point Likert scale (where 1 = ‘completely disagree’ and 5 = ‘completely agree’) concerning eight different statements reflecting themes present in the literature, for example, ‘I think giving blood is good’ [6–8]. These were combined into an index score. Respondents’ willingness to once again become active blood donors was measured by asking participants to rate themselves on a five-point Likert scale (where 1 = ‘completely disagree’ and 5 = ‘completely agree’) concerning five different statements reflecting themes present in the literature, for example, ‘I want to be an active blood donor again’ [6, 7]. For donors in the DD group, statements were adapted where relevant to reflect the theoretical nature of the question posed (for example, ‘If HemaQuebec told me it was possible, I would give blood again’). These were combined into an index score. Finally, respondents were asked to indicate their education level. The project received approval from H
ema-Qu
ebec’s ethical review board.

Donor follow-up All donors deferred since 1 January 2008 who were eligible for re-entry (respondents and non-respondents) were followed up in the administrative databases for a period of twelve months after admissibility to the programme to compare their rate of participation. For those in the DD group, a letter informing them that they were now eligible for re-entry was sent after completion of the questionnaire.

Statistics Uncorrected chi-square test was used to compare the distribution of participants as well as to compare participation to the re-entry programme between groups. Mann–Whitney U-test [17] was used to compare respondents between groups according to age and number of previous blood donations. Scores indicative of each construct were obtained by establishing the average score from the different items in the questionnaire that together composed the construct. For distress, a third measure was created: the difference between initial and current distress scores was calculated (D distress). © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

Donors’ reactions to deferral 135

Cronbach’s a, commonly used to estimate the reliability of a psychometric test for a sample of examinees [18], was used to assess the internal consistency of the respondents’ initial distress, current distress, attitude towards blood donation, willingness to donate blood again and perception with regard to aspects of the notification process. Cronbach’s a scores above 0
7 indicate acceptable internal consistency [19]. Paired t-tests or Wilcoxon’s signed rank test were used to compare our two groups according to the variables of interest. A two-factor analysis of variance (ANOVA), using group (DER or DD) and gender, was used for each variable of interest. A one-way ANOVA was also used to compare each variable between the markers. HBsAg was excluded from that analysis because the sample size was too low. Robust linear regression was used to determine whether there was a relationship between initial psychological distress levels and time elapsed between the date of notification and the date of questionnaire reception for DD. Cumulative incidence rate ratios (RR) and their confidence intervals were calculated to compare the re-entry programme participation rates according to scores (low or high) for the different themes (willingness, attitudes, distress and perceived quality of the notification process) using the median as a mid-point for each theme as well as according to the confounding factors (gender, age and virological marker). We used linear logistic regression to study the association between the themes and the odds of taking part in the re-entry programme, while taking into account the possible confounding effect of other variables. All comparisons were made at the 95% confidence level (two-sided) using SAS EG version 4
1 (SAS Institute, Cary, NC, USA). All results were checked for normality using Shapiro–Wilk test.

Table 2 Study participants demographic characteristics Group DD

Gender F M Age (years) 18–29 30–39 40–49 50–59 60–70 Average Education completed Primary Secondary Professional or preuniversity degree University Virological marker Anti-HCV Anti-HIV 1/2 HBsAg Syphilis Previous donations New donor 1–3 donations 4–6 donations ≥7 donations Average Total

DER

N

%

N

%

31 22

58
49 41
51

31 22

58
49 41
51

25 5 9 13 1 35
1a

47
17 9
43 16
98 24
53 1
89 – 2
0b

24 7 7 14 1 35
4a

45
28 13
21 13
21 26
42 1
89 – 2
0b

0 11 22 20

0
0 20
8 41
5 37
7

0 9 25 19

0
0 17
0 47
2 35
9

17 25 2 9

32
08 47
17 3
77 16
98

17 25 2 9

32
08 47
17 3
77 16
98

20 18 3 12 4
4a 53

37
74 33
96 5
66 22
64 –1
0b 100

21 14 2 16 8
0a 53

39
62 26
42 3
77 30
19 –2
4b 100

a

Average. Standard error.

b

Themes

Results Participant demographics Participant distribution for both cohorts was identical with regard to gender and virological marker (Table 2), as expected. Distribution according to age categories did not differ between cohorts (uncorrected chi-square test, P = 0
9584). Average age of participants did not differ between groups (Mann–Whitney U-test, P = 0
8721). Groups did not differ in levels of education (chi-square test, P = 0
8117). Finally, there were no significant differences between the two groups in the level of previous experience with blood donation (categories of previous number of blood donations: uncorrected chi-square test, P = 0
6065; participants’ average number of previous blood donations: Mann–Whitney U-test, P = 0
9662). © 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

Internal consistency was above 0
7 for respondents’ recalled initial distress (Cronbach’s a 0
92), current distress (Cronbach’s a 0
92), attitude towards blood donation (Cronbach’s a 0
86), willingness to donate blood in the future (Cronbach’s a 0
91) and perception of aspects of the notification process (Cronbach’s a 0
87). The statistical analyses showed no significant difference between groups in the scores for initial or current distress, in the differential score of distress (D distress) or in the level of willingness to give blood again (Fig. 1) except in women, which was lower in the DER group then in women in the DD (P = 0
0008; Fig. 2). The only significant differences between the two groups were found for attitudes (Wilcoxon’s signed rank test, P = 0
0276; Fig. 1) and the perception of the notification process (Wilcoxon’s signed rank test, P = 0
0108;

136 G. Delage et al.

Fig. 1 Comparison of respondents psychological distress, attitude, willingness and perception of the notification process according to the group. Individual data and the mean are presented. DD: deferred donors group; DER: group of donors eligible for reentry. *P < 0
05.

Fig. 2 Comparison of respondents psychological distress, attitude, willingness and perception of the notification process according to the group and gender. Individual data and the mean are presented. DD: deferred donors group; DER: group of donors eligible for re-entry. *P < 0
001 DER Female vs. DD Female. ‡P < 0
05 DER vs. DD

Fig. 1; ANOVA, P = 0
0181; Fig. 2): both were significantly higher in donors in the DER. Current psychological distress was significantly lower in comparison with initial distress in both groups (Wilcoxon’s signed rank test, P < 0
0001). Attitude was the only variable of interest that differed significantly among markers (one-way ANOVA, P = 0
0341), where HIV (4
21 – 0
10) was significantly higher than HCV (3
81 – 0
13). We used regression analysis to determine whether there was a relationship between initial psychological distress levels and time elapsed between the date of notification and the date at which the questionnaire was received. The relationship was not statistically significant (R2 = 0
0069, P = 0
4914).

Donor follow-up Of the donors in the study who were invited to participate in the re-entry process (53/53 in the DER and 31/53 in the DD), 45
2% [38/84; 32
3% (10/31) in DD and 52
8% (28/53) in DER] took part. There was no difference between groups in the participation rate (uncorrected chisquare test, P = 0
0676). Among all donors who were invited to participate in the re-entry process, including the study participants, 29
7% [289/971; 21
9% (106/483) in DD and 37
5% (183/488) in DER] took part. Participation was significantly higher in the DER group (uncorrected chi-square test, P < 0
0001). Table 3 shows that the rates of participation in the programme for the two groups combined were not affected

© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

Donors’ reactions to deferral 137

Table 3 Comparison of the percentage of respondents who took part in the re-entry programme in the 12 months following their admissibility to the programme according to demographics and scores for each theme. The median was used as a mid-point for each theme to set apart low and high values Donor took part in the re-entry processa Gender F 20/48 (41
7) M 18/36 (50
0) Age (years) 18–29 17/40 (42
5) 30–39 3/10 (30
0) 40–49 7/11 (63
6) 50–70 11/23 (47
8) Virological marker Anti-HCV 12/25 (48
0) Anti-HIV 19/41 (46
3) HBsAg 1/4 (25
0) Syphilis 6/14 (42
9) Group DD 10/31 (32
3) DER 28/53 (52
8) Attitude Low 13/40 (32
5) High 25/44 (56
8) Willingness Low 12/42 (28
6) High 26/42 (61
9) Initial distress Low 21/42 (50
0) High 17/42 (40
5) Current distress Low 18/42 (42
9) High 20/42 (47
6) D distress Low 22/41 (53
7) High 16/43 (37
2) Perception of notification process Low 19/41 (46
3) High 19/43 (44
2) All 38/84 (45
2)

Unadjusted rate ratio

Adjusted rate ratiob

Ref. 1
40 (0
59–3
34) Ref. 0
58 (0
13–2
58) 2
37 (0
60–9
40) 1
24 (0
44–3
48) Ref. 0
94 (0
35–2
53) 0
36 (0
03–3
96) 0
81 (0
22–3
03) Ref. 2
35 (0
93–5
94) Ref. 2
73 (1
12–6
66)

Ref. 2
76 (1
06–7
21)

Ref. 4
06 (1
63–10
14)

Ref. 8
49 (2
53–28
44)

Ref. 0
68 (0
29–1
61)

Ref. 0
59 (0
23–1
53)

Ref. 1
21 (0
51–2
87)

Ref. 1
16 (0
42–3
20)

Ref. 0
51 (0
21–1
22)

Ref. 0
52 (0
21–1
29)

Ref. 0
92 (0
39–2
17)

Ref. 0
78 (0
29–2
11)

a

Data reported as number/total number (%). Adjusted for gender, age, virological marker and group.

b

by age, gender, virological marker or group. Attitude (crude RR 2
73, 95% CI, 1
12 to 6
66; adjusted RR 2
76, 95% CI, 1
06 to 7
21) and willingness to donate again (crude RR 4
06, 95% CI, 1
63–10
14; adjusted RR 8
49, 95% CI, 2
53–28
44) are the only themes that had a significant impact on the rates of participation to the re-entry programme, both crude and adjusted for age, gender, virological marker and group.

© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

Discussion The lack of significant differences between the two respondent groups with regard to gender, age, education completed and number of previous donations suggests that they were equivalent, thus limiting potential sources of selection bias linked to these factors [10]. Contrary to what was expected, no significant difference in recalled psychological distress was found between the DD and the DER at the time of notification or at the time of questionnaire completion. This suggests that the level of psychological distress engendered is similar whether the donor is offered the possibility of re-entry or not. However, the scores obtained did not appear to indicate high levels of distress to begin with. Psychological distress levels appear to fade progressively in the time elapsed between notification and completion of the questionnaire; we can speculate that donors become reconciled with their deferral status and test results over time. A similar lowering of psychological distress in time has also been observed by others [3]. In contrast, Whittaker and his colleagues [1] found that 3 months after having been informed of a deferral, whether for confirmed positive or false-positive test results, study participants were not reconciled with their deferral status. Their results are in contradiction with ours as DER participants seemed reconciled with the test results and the deferral status within a similar time period. Finally, with regard to the causes of psychological distress more specifically, previous research showed participants who did not feel sufficiently informed reported higher levels of preoccupation [2]. Other authors report that a majority of donors excluded for a false-positive result had difficulties understanding the information they were given [3]. In our study, it seems that the donors’ perception of the notification process is more positive with the new approach. However, the mean score for that variable (3
08 on a 5-point scale in the DER) suggests there is still room for improvement. We have subsequently modified our notification letter to attain that goal. Scores representing donors’ general attitudes towards blood donation were relatively high among both groups, but higher in the DER group. Others have indicated that attitudes towards blood donation play an important role in determining the intention to repeat donation among experienced donors [7]. Though attitude towards blood donation was more positive with the new approach, willingness to once again become an active blood donor did not appear to be affected. One possible explanation could be that the level of willingness to once again give blood was already high among DD, leaving little room for improvement. In fact,

138 G. Delage et al.

this is the variable for which the highest scores were noted in our study. Surprisingly, willingness to donate blood again in the future was lower among women advised of the possibility of re-entry ( x = 4
17) than among those not advised of that possibility ( x =4
61), a finding that cannot be explained in the context of this study. Attitudes towards blood donation and willingness to once again give blood were important determinants of participation in the re-entry programme, as reflected in Table 3 when combining data from both groups. Donors with high scores on attitudes were 2
76 times more likely to take part in the re-entry programme, and those with high scores for willingness to give blood again were 8
49 times more likely to take part, when compared with donors having low scores for these variables. This is consistent with previous studies showing that intention to donate blood in the future is one of the best predictors of donor return [6, 8]. A potential option is to use a double testing algorithm where donors who have previously had a false-positive result with one assay are tested with another of similar sensitivity, and if negative are considered acceptable. This approach, which is carried out in the United Kingdom [20], raises two operational issues. It creates logistical difficulties for blood centres, who have to maintain two testing platforms approved by regulatory authorities for the sole purpose of continuing to accept donation for testpositive donors. It also implies being able to configure the computer systems in order for those donors to be appropriately identified for testing on the alternate platform. The absence of significant difference between DER and DD among study participants in terms of participation in the re-entry programme despite a 20% intergroup difference is most likely explained by the low sample size. Comparison of participation rate with the re-entry programme for all eligible DER and DD donors listed in H
ema-Quebec’s database for the study time period, whether they took part in the study or not, showed a significantly higher rate in DER (15
5% higher).

A weakness of our study is that it was not a prospective randomized trial but a comparison study of two cohorts. A randomized trial would have been preferable but could have raised major ethical issues. We attempted to minimize recall bias in our study by carefully selecting the donors in the DD group [10]. Regression analysis results suggest that we were successful in that endeavour. In conclusion, even though the quality of the new notification process is perceived more positively than the previous one, the notification letter did not have the expected positive effects on psychological distress levels nor on the willingness to give blood again. This may be due to the fact that distress scores were low and willingness scores were high in both groups to begin with, leaving little room for improvement. However, the attitudes towards blood donation were more positive with the new notification process. Finally, it was interesting to observe that attitudes towards blood donation and willingness to give blood again were good predictors of participation in the re-entry process.

Acknowledgements The authors wish to thank H
elene Grenier, Elvira Lebedeva, Sylvie Lebel, Na€ıma Marzoug, Marisa Mocchiutti, Katherine-Anne Mondou and Sylvie Tremblay for their help with recruitment. We also thank Richard Renaud for his support, and Dr Marc Germain and Professor Gaston Godin for their review of this article. All authors have contributed to either research design, data acquisition, data analysis or interpretation; all were involved in drafting and revision and have approved the submitted version.

Competing interests The authors have no competing interests.

Sources of research support None.

References 1 Whittaker S, Carter N, Arnold E, et al.: Understanding the meaning of permanent deferral for blood donors. Transfusion 2008; 48:64–72 2 Tynell E, Norda R, Ekermo B, et al.: False-reactive microbiologic screening test results in Swedish blood donors – how big is the problem? A survey

among blood centers and deferred donors. Transfusion 2007; 47:80–89 3 Kleinman S, Wang B, Wu Y, et al.: The donor notification process from the donor’s perspective. Transfusion 2004; 44:658–666 4 Halperin D, Baetens J, Newman B: The effect of short-term, temporary defer-

ral on future blood donation. Transfusion 1998; 38:181–183 5 Moyer LA, Shapiro G, Shulman PD, et al.: A survey of hepatitis B surface antigen-positive blood donors: degree of understanding and action taken after notification. Transfusion 1992; 32:702–706

© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

Donors’ reactions to deferral 139

6 Godin G, Conner M, Sheeran P, et al.: Determinants of repeated blood donation among new and experienced blood donors. Transfusion 2007; 47:1607–1615 7 France JL, France CR, Himawan LK: A path analysis of intention to redonate among experienced blood donors: an extension of the theory of planned behavior. Transfusion 2007; 47:1006– 1013 8 Masser BM, White KM, Hyde MK, et al.: Predicting blood donation intentions and behavior among Australian blood donors: testing an extended theory of planned behavior model. Transfusion 2009; 49:320–329 9 Bianco C, Kessler D: Donor notification and counseling management of blood donors with positive test results. Vox Sang 1994; 67(Suppl. 3):S255– S259 10 Grimes DA, Schulz KF: Bias and causal associations in observational

11

12

13

14

15

research. The Lancet 2002; 359:248– 252 Margetts BM, Vorster HH, Venter CS: Evidence-based nutrition – the impact of information and selection bias on the interpretation of individual studies. South Afr J Clin Nutr 2003; 16:78–87 McDowell I: Measuring Health: A Guide to Rating Scales and Questionnaires, 3rd edn. Oxford, Oxford University Press, 2006 Cohen J: Statistical Power Analysis for the Behavioral Sciences, revised edn. New York, Academic Press, 1977 Canadian Medical Protective Association: Physician-Patient Communication: Making it Better. December 2010, [4p]. Available at: http://www.cmpaacpm.ca/cmpapd04/docs/resource_files /perspective/2010/04/com_p1004_3-e. cfm (Last accessed 13 September 2013) Dupuy HJ: The Psychological General Well-Being (PGWB) index; in Wenger NK, Mattson ME, Furberg CD, Ellinson

© 2014 International Society of Blood Transfusion Vox Sanguinis (2014) 107, 132–139

16

17 18

19 20

J (eds): Assessment of Quality of Life in Clinical Trials of Cardiovascular Therapies. New York, Le Jacq Publishing Inc., 1984:170–183 Dupuy HJ. Appendix I: Selected test instruments; in: Wenger NK, Mattson ME, Furberg CD, Ellinson J (eds): Assessment of Quality of Life in Clinical Trials of Cardiovascular Therapies. New York, Le Jacq Publishing Inc.; 1984:353–356 Rosner B: Fundamentals of Biostatistics, 4th edn. Belmont, Wadsworth, 1985 Kline P: The Handbook of Psychological Testing. Florence, Psychology Press, 2000 Nunnally JC: Psychometric Theory. New York, McGraw–Hill, 1978 UK Blood Transfusion & Tissue Transplant Services: Guidelines for the Blood Transfusion Services in the UK, 8th edn. The Stationery Office, Norwich, 2013