Abstract
Introduction. A pilot test of a computer-tailored intervention designed to promote blood donation among Blacks was conducted. Method. Intervention content, based on the transtheoretical model, offered participants individually and culturally tailored information on blood donation with emphasis on need specific to race (e.g., sickle-cell disease). Black adults (N = 150) with a diversity of blood donation experience were recruited from a blood center and a survey recruitment website. Posttest assessment included a 14-item evaluation and transtheoretical model questions. Results. Participants rated the program positively (81.3% to 98.7% of participants agreeing or strongly agreeing with evaluation items). For example, 98.7% of respondents reported that the program gave sound advice and that personal feedback was easily understood, and 87.3% felt the program was designed for people like themselves. Ninety-five percent of participants reported that they would recommend the program to others. There were no significant differences in ratings based on demographics. Qualitative responses support program acceptability. Furthermore, pre- and postprogram assessments indicated an increase in intention to donate, t(149) = 3.56, p = .001, d = .29. Discussion. With acceptability and feasibility confirmed, the next steps are to test efficacy and cost-effectiveness for use to increase blood donation, particularly in priority populations.
Keywords
Introduction
Our nation’s blood supply derives from volunteer blood donors. However, the demand for blood has increased and the number of blood donors has not always kept pace, periodically leading to shortages. Like other ethnic/racial minority groups, Blacks have participated in blood donation at disproportionately low levels (James, Hillyer, & Shaz, 2012; Murphy et al., 2009; Shaz, James, Hillyer, Schreiber, & Hillyer, 2011; Shaz & Hillyer, 2010) and are less likely than White or Hispanic donors to become regular blood donors (Kessler, Rebosa, France, & Shaz, 2012). Overcoming limitations to blood donation is critical for Blacks because of the need for blood to treat sickle-cell disease (SCD) patients with blood from similar genetic types. With this goal in mind, an interactive, individualized, stage-matched intervention to encourage blood donation in Blacks was developed and piloted.
SCD is an inherited blood disorder that occurs in about 1 in 500 Black births that frequently requires red blood cell transfusion therapy (Shaz et al., 2011; Shaz, Zimring, Demmons, & Hillyer, 2008). Though this therapy generally increases length and quality of life, multiple transfusions can lead to SCD patients becoming alloimmunized against blood group antigens. Alloimmunization requires future transfusions of antigen-negative blood. Additionally, to prevent alloimmunization and resulting hemolytic transfusion reactions, antigen-matched blood is used (Rosse et al., 1990). The probability of identifying matched and/or compatible red blood cell units can be increased by screening donors of the same racial/ethnic background as the SCD patient for the relevant antigens. In some cases, Black donors are the only source of compatible blood. Thus, to address the transfusion needs of SCD patients, a demand exists for blood products specifically from Black donors.
Efforts to increase blood donation need to be theory based, interactive, easily disseminated, tailored to appropriate constructs, and targeted to factors that are relevant to the population. The transtheoretical model (TTM) is a theory of behavior change that is well suited to meeting these intervention needs, particularly when coupled with computer-tailored interventions (CTIs). The TTM explains motivation and intentional behavior change based on thoughts, experiences, and behaviors and comprises four key constructs, including Stage of Change, Decisional Balance, Self-Efficacy, and Processes of Change (POC; Prochaska, 1983; Prochaska & Diclemente, 1983, 1992; Prochaska & Velicer, 1997). These dynamic variables can be used to reflect the “when” (Stage of Change), “why” (Decisional Balance and Self-Efficacy), and “how” (POC) of becoming a blood donor among Black adults. Stage of Change is typically a categorical variable and is typically defined as Precontemplation (not thinking about change), Contemplation (considering change), Preparation (planning to change in near future), Action (recently changed), and Maintenance (sustained change; Prochaska & Diclemente, 1983, 1992; Prochaska, Redding, & Evers, 2008; Prochaska & Velicer, 1997). Decisional balance refers to the weighing of the pros and the cons of changing a specific behavior. Situational Self-Efficacy reflects the level of confidence an individual has to engage in a new behavior or to maintain a behavior in a variety of challenging situations (Velicer, Diclemente, Rossi, & Prochaska, 1990).
The TTM has been applied to over 48 health behaviors among various groups, including Blacks (Amoyal et al., 2013; Burditt et al., 2009; Cunningham, 1998; Hall & Rossi, 2004, 2008; Hall, Rossi, Martel, & Koerner, 2004), with three studies showing empirical support specific to TTM constructs in a Black sample for blood donation (Amoyal et al., 2013; Burditt et al., 2009; Ferguson & Chandler, 2005). For more information on TTM constructs related to blood donation, see Burditt et al. (2009) and Amoyal et al. (2013).
One method for rapid dissemination of interventions to entire populations (e.g., via the Internet) is the utilization of a CTI. These interventions provide computer-adaptive techniques that allow for immediate and individually tailored feedback. By using relevant theory, normative databases, and empirically based decision rules, CTIs provide matching of the most appropriate information to each participant to guide the change process. This approach allows for thousands of unique combinations of individual feedback based on specific constructs within an intervention.
This article describes the development, feasibility, and initial efficacy of an innovative, individualized CTI to increase blood donation rate among Blacks. The Blood Donation Expert System (BDES) intervention is based on the TTM and uses measures of the key TTM constructs specific to blood donation in Black adults with emphasis on race-specific need for blood donation to treat SCD. The intervention focused on key decisional variables that have been found to be important in accelerating behavior change in numerous TTM-based interventions (e.g., exercise, mammography screening). The objective was to create an interactive, individualized, stage-matched intervention for blood donation that can be delivered cost-effectively to large segments of a population.
Method
Development of the BDES
Formative Research
Extensive formative work consisting of focus groups, expert interviews, and measure development studies with more than 880 Black adults aimed to understand key issues regarding blood donation and applicability of the TTM to blood donation. These data provided a rich resource to develop the content of the BDES and also provided a database for normative comparisons. Details of this formative work are described elsewhere (Amoyal et al., 2013; Burditt et al., 2009). In addition, formative work was used to gain input on the BDES look and feel (e.g., layout, graphics, photos), tone (e.g., clarity of language, level of individual tailoring), and program flow (e.g., ease of navigation, logic of content order). Once the BDES had been programmed and tested, three 60-minute, one-on-one usability interviews were conducted. Usability study participants were Black adults older than 18 years who were recruited through Craigslist and compensated $50. Usability interview feedback was integrated in the final version of the BDES for the pilot test. Procedures for the pilot study were approved by the institutional review boards at the University of Rhode Island and the New York Blood Center (NYBC).
Description of the BDES
The BDES is designed for Black adults who are eligible to donate blood. The prototype of BDES was accessed via the Internet and included one session that began with an overview of the research study, informed consent, and questions to determine eligibility. Participants interacted with the program, which alternated between assessment questions and individualized, stage-matched, behavioral feedback based on TTM constructs. Intervention components also included testimonials, images and graphics, behavior change strategies, and feedback sections, including type-in response boxes and other interactive tools. All content, including personalized strategies to increase blood donation, were written at a sixth-grade reading level. Participants were assessed on TTM constructs and received individualized feedback on how their responses compared to those of a sample of individuals who were successful at progressing toward regular blood donation (normative comparisons). Pilot participants received an intervention to increase blood donation that included distinct assessments and individualized feedback in the following four sections:
Stage of Change: Description of participant’s current readiness for being a regular blood donor (i.e., two donations per year)
Decisional Balance: Pros and cons of donating and how their scores compared to individuals who have become blood donors
Self-Efficacy: Situations that make it difficult to donate and guidance on how to increase confidence to manage these situations
POC: Use of stage-appropriate strategies and how they compare with others who were most successful in progressing to the next stage of change
Throughout the BDES, users received feedback tailored to their stage for being regular blood donors. For example, a participant in Action is given specific strategies to continue to be a regular donor and was reminded of benefits of the behavior. Feedback for a Precontemplation stage participant, however, focuses on raising awareness of the importance of blood donation in the context of SCD and provides opportunities to think about how self-image would change if he or she moved toward blood donation. Information on best practices to minimize the likelihood of physical cons (e.g., vasovagal response) was also provided. Examples of an assessment screen for Self-Efficacy to donate blood and a tailored feedback screen on the Pros of blood donation for a participant in the Contemplation stage are shown in Figure 1.
Examples of Assessment and Feedback Screens From the Blood Donation Expert System
Pilot Testing
Participant Recruitment
An online survey company was used to recruit 100 Black participants from the northeastern United States with emphasis on New York and New Jersey. To ensure participation in the study by regular blood donors (i.e., two or more blood donations in the past 12 months), an additional 50 participants were recruited from the NYBC donor database. The goal of recruiting from two sites was to ensure that the sample would reflect the complete range of readiness (Precontemplation to Action) and history (never to regular) of blood donation. Incentives for participants recruited via the online survey company included points for one entry into a contest for a chance to win $25,000 with an additional $3 dollars in points for those aged 18 to 24 years (to increase recruitment from that age-group). Participants recruited by the NYBC received a roughly equivalent amount of points (approximately $4), the incentive earned for one blood donation.
Measures
Measures of the TTM constructs (Stage of Change, Decisional Balance, Situational Self-Efficacy, and POC) for blood donation summarized below were previously developed with entirely Black participant samples and have been described in detail elsewhere (Amoyal et al., 2013; Burditt et al., 2009).
Stage of Change
Fourteen questions designed to assess both blood donation history and current intentions were utilized to classify participants into four stages of readiness to be a regular blood donor. Participants were classified in Precontemplation if they were not thinking about donating blood, in Contemplation if they were thinking about donating blood in the next 6 months, and in Preparation if they were planning on donating blood within the next 30 days or when next eligible. Participants were classified as being in the Action/Maintenance stage if they had donated twice in the past year and reported intent to donate in the next 30 days or when next eligible. Participants in the Action and Maintenance stages were collapsed together because these groups are both considered regular donors (donated twice per year in at least the past year).
Decisional Balance
Decisional balance is a 12-item measure designed to assess how a person values the Pros (e.g., saving someone’s life), Eligibility Cons (e.g., find out I have a disease) and Physical Cons (e.g., afraid of needles) of blood donation. Participants rate how important each item is in their decision whether or not be a regular blood donor on a 5-point scale from 1 = not at all important to 5 = extremely important. These scales have good internal reliability with Pros α = .82, Eligibility Cons α = .74, and Physical Cons α = .72 (Burditt et al., 2009).
Self-Efficacy
Six items were designed to assess individual’s confidence in their ability to donate blood in various challenging situations (e.g., when I am feeling tired). Items were rated on a 5-point scale and responses ranged from 1 = not at all confident to 5 = extremely confident and has good internal consistency (α =.81).
Processes of Change
POC for blood donation are represented by 10 four-item scales designed to assess the behaviors, thoughts, and feelings that are associated with becoming or maintaining being a regular blood donor (Amoyal et al., 2013). Each item is scored on a 5-point scale based on how often the participants think, feel, or do the items described. Responses ranged from 1 = never to 5 = repeatedly. The scales had good internal consistency ratings with α ranging from .74 to .91.
Blood Donation Intent
A single item (i.e., “How likely are you to consider donating blood?”) assessed how likely individuals are to consider donating blood prior to and after completing the BDES. The item had a rating scale ranging from 1 = not at all likely to 10 = extremely likely.
Acceptability Survey
A 14-item questionnaire with four response options (strongly disagree to strongly agree) was used to gauge acceptability. It was adapted from the National Cancer Institute’s Educational Materials Review Form (National Cancer Institute, Office of Cancer Communications, 1989) and the evaluation scale used by Rimer et al. (1994). Three open-ended questions were asked to collect participant feedback on what they liked the most and least about the BDES and, when applicable, to describe their plans to donate.
Procedure
Participants accessed the program website, provided online consent, and answered questions to determine eligibility. The intervention session lasted 20 to 25 minutes and included pretest questions interspersed with individualized computer-tailored feedback, and posttest questions with additional evaluation questions regarding acceptability.
Results
Black adults (N = 150) between the age of 18 and 75 years (M = 39.7, SD = 13.3) from eight states in the northeast region of the United States participated in this study. Participants were predominantly female (54.0 %) and 76.7% reported a history of blood donation (Table 1). Participants were classified into four stages of change for blood donation (Table 2). Thirty-five percent of participants reported that they were not thinking about donating blood within the next 6 months (Precontemplation) and 34.7% of participants had donated twice in the past year and reported intent to donate in the next 30 days or when next eligible (Action/Maintenance). The majority of participants (76.7%) had donated blood in the past. The most common place to donate blood was at a mobile drive (36.0%). A third of participants (33.3%) donated in response to blood center phone calls.
Participant Demographics
Participant Responses to Blood Donation History Items for Total Sample
Acceptability Findings
Responses to evaluation items are presented in Table 3. Accessibility of the program was one of its most valuable characteristics, with a majority of participants strongly agreeing or agreeing that the program was easy to use (96.0%). Furthermore, the questions (88.6%) and feedback (98.7%) were easy to understand. Most important, 95.3% of the participants indicated that they would recommend the program to others. No differences in average evaluation score were found by Stage of Change or gender.
Means, Standard Deviations, and Response Scores of Evaluation Items
NOTE: N = 150 for all items.
Qualitative Participant Feedback
Two open-ended questions assessed what participants liked most and least about the question and feedback parts of the BDES. Responses were sorted into general themes and tallied. For the “most liked” question, the most commonly reported responses were that the BDES was clear, direct, and/or easy to use (n = 32) or that they learned new information (n = 26). Other responses included favorable ratings of the visuals (pictures and stories), length of program, tailored aspect of feedback, and the educational and encouraging design of the questions. For the “least liked” question, the most commonly reported responses were thinking that the program was too long (n = 12); that the questions were repetitive (n = 11); the program, content, or questions were confusing (n = 11); and feeling guilty for not donating (n = 2). Sixty-four respondents reported that they would make no changes to the program.
Indicators of Preliminary Behavior Change
Although the primary goals of this pilot test were to test acceptability and feasibility, additional efforts were made to assess preliminary evidence of behavior change. As the pretest tailored feedback and posttest were completed within one session, pre-Action participants (Precontemplation, Contemplation, or Preparation) could not progress to Action during the session. Transitions between the Stages of Change among those who were in pre-Action stages at pretest were analyzed as indicators of initial behavior change. Stage progression was examined by comparing the pretest Stage of Change to the posttest Stage of Change for pre-Action participants only. Participants could progress (e.g., move from Precontemplation to Contemplation), regress (e.g., move from Contemplation to Precontemplation), or remain stable within their Stage of Change. Results indicate that 46.9% of those in a pre-Action stage at pretest progressed at least one stage at posttest assessment.
Change in blood donation intent was assessed with the Blood Donation Intent item that was designed to capture attitudinal changes that may occur due to the intervention. Though some participants are not able to progress in stage movement, their attitudes may have changed. Participants reported that they were more likely to consider donating blood after completing the program, t(149) = 3.56, p = .001, d = .29. Furthermore, change in Pros and Cons was examined from pretest to posttest, with participants reporting a significant increase in Physical Cons from pre- to posttreatment, t(149) = 2.41, p = .017, d = .20. No significant differences were found for Pros or Eligibility Cons. Results also indicated that participants in pre-Action stages were significantly more likely to consider donating blood after completing the intervention program, t(97) = 3.36, p = .001, d = .34 and showed a similar significant increase in Physical Cons, t(97) = 2.20, p = .030, d = .22.
Discussion
This study demonstrated the feasibility and acceptability of a tailored expert system intervention based on the TTM to increase blood donation among Black adults. The BDES intervention provides tailored empirically driven feedback for Blacks in all stages of readiness to donate blood. One of the major goals was to create an intervention that would receive favorable ratings on evaluation of acceptability. This was achieved, and results indicated that all items received favorability ratings of agree or strongly agree, ranging from 81.3% to 98.7% endorsement. Furthermore, acceptability ratings did not differ by stage or gender, suggesting that the program can appeal to a broad range of potential blood donors, even for those not thinking about donating blood. Though 35.3% of participants were not planning to donate, the BDES was able to engage these Precontemplators to consider blood donation and provided support for dissemination to promote blood donation behavior in a future sample.
It is important to note that almost 60% of the study participants reported a history of blood donation within the past 2 years and were more ready to donate than participants who had no history of blood donation. The BDES can be targeted to this subgroup of “lapsed” donors to encourage these individuals to become regular donors (i.e., two donations per year). NYBC had 19,646 blood donors who self-identified as Black in a 12-month period from 2011 to 2012 and who completed an average of 1.6 donations in that period. However, of the subgroup of 8,214 first-time blood donors during that same period, approximately 75% did not donate again in a year after their first donation (Kessler et al., 2012). So retention of new Black donors is poor and using the BDES to convert “lapsed” donors to regular donors is likely the quickest route to increasing the donation rate.
There were no major obstacles to delivering the BDES via the web through Survey Sampling or through the NYBC listserv, and recruitment goals were achieved within a few weeks. Additionally, through online delivery, participants were able to respond to items and received individualized feedback in a private, convenient, and nonthreatening format. These results support the potential to disseminate the BDES broadly and easily through various distribution channels, including local blood centers, colleges, employers, and health organizations.
In addition to gathering acceptability data, this study aimed to measure preliminary behavior change. Readiness and intention to donate were assessed before and after each session and showed general increases with 46% of participants in pre-Action progressing at least one stage at posttest. Participants also reported a significant increase in their likelihood to consider donating blood after completing the program. Interestingly, though neither Pros nor Eligibility Cons significantly changed from pre- to postassessment, an increase in Physical Cons was reported. The BDES was designed to be educational, and thus it is reasonable to consider that increasing awareness about the potential physical effects (e.g., feeling faint) may in turn increase participants’ ratings of the importance of the Physical Cons of donating blood. These results demonstrate that the BDES could influence blood donation attitudes using readiness and intention as markers for behavior change and support a longitudinal assessment of its efficacy.
The BDES makes several major contributions to the current state of science on blood donation intervention development. First, this program includes theoretical and empirical dynamic tailoring. Second, the program uses a population-based approach, offering tailored feedback to participants in all stages of readiness. Additionally, this is the first blood donation CTI to offer tailored feedback on all relevant TTM constructs for a particular stage of change. The next step is to develop a full-length CTI that can include measurement to provide tailored feedback at multiple time points. Consequently, the BDES could be the first blood donation CTI to provide tailored feedback utilizing baseline and follow-up sessions to track progress over time. Additionally, by using multiple time points, any changes can be tracked and compared to earlier time points, which allows for “ipsative” feedback (i.e., providing comparisons to a participant’s previous responses) in addition to normative feedback, which provides comparisons to peers.
The BDES was designed to appeal to the needs of an underrepresented group in order to promote blood donation. All of the measures used in the BDES were developed with Black participants. In addition, focus groups and expert interviews were elicited to provide feedback on images, content, and items. This extensive development process helped ensure that the BDES would be acceptable to the priority population and the usability ratings suggest that this program was received very favorably. With appropriate cultural and measure development, the BDES could also be expanded to address blood donation rates for all racial and ethnic groups. The BDES is a low-cost intervention that with further development could be delivered to potential donors via any Internet-connected device.
The generalizability of these results may be limited when comparing the current sample to Blacks in other regions, as participants were recruited from the northeastern United States. Participants also tended to be more educated than the general population of Black adults. Additionally, the study is cross-sectional, thus limiting the impact of the BDES to attitudinal change favoring blood donation. Future research will aim to recruit a more representative sample using a longitudinal design and track actual blood donation as a primary outcome.
Conclusions
Blacks have higher rates of SCD among other diseases that require blood transfusions for treatment. Given that a disproportionately small number of Blacks donate blood, it is imperative to develop innovative interventions that are empirically based and culturally tailored. The BDES offers a convenient, private, and efficient means of providing individually and culturally tailored information about blood donation to Black adults. With favorable reviews of this pilot program and improvements based on participant feedback, the development of an enhanced program with multiple time points can ensue. If the enhanced program is deemed effective, the BDES can be an inexpensive and quick way to disseminate information and encourage behavior change that can address an important public health need for blood donation among Blacks.
Footnotes
This research was supported by Grant # 1R21HL092390 to Dr. Robbins from the National Heart, Lung and Blood Institute.