Evaluation of a Computer-Based Decision Aid for Promoting Informed Prostate Cancer Screening Decisions Among African American Men: iDecide

Abstract

Purpose:

To evaluate the effects of iDecide on prostate cancer knowledge, informed decision-making self-efficacy, technology use self-efficacy, and intention to engage in informed decision-making among African American men.

Design:

One-group, pretest/posttest.

Setting:

Community settings in South Carolina.

Participants:

African American men, ages 40 years +, without a prior prostate cancer diagnosis (n = 354).

Intervention:

iDecide, an embodied conversational agent-led, computer-based prostate cancer screening decision aid.

Measures:

Prostate cancer knowledge, informed decision-making self-efficacy, technology use self-efficacy, and intention to engage in informed decision-making.

Analysis:

Descriptive statistics, paired t tests, general linear modeling, Spearman correlations.

Results:

On average, participants experienced significant improvements in their prostate cancer knowledge (P ≤ .001), informed decision-making self-efficacy (P ≤ .001), and technology use self-efficacy (P ≤ .001), postintervention. Additionally, 67% of participants reported an intention to engage in informed decision-making.

Conclusion:

Given the significant improvements across all measures, this research demonstrates that embodied conversational agent-led decision aids can be used to enhance the capacity for making informed prostate cancer screening decisions among African American men and increase their technology use self-efficacy. One critical limitation of this study is that most men had received prostate cancer screening prior to engaging in our intervention, so the implications of this intervention may be different for men who do not have a history of screening. Additionally, actual engagement in informed decision-making postintervention was not assessed.

Keywords

prostate cancer technology cancer screening African American decision-making

Purpose

African American (AA) men have the highest rates of prostate cancer (PrCA) in the United States, with incidence and mortality rates 2 to 3 times higher than white men. In the southeastern United States, PrCA incidence and mortality disparities among AA men are 1.5 and 2.5 times greater, respectively.^1,2 The response to these disparities, however, have been fraught with controversy.^3

-7 Although PrCA screening could lead to timely and appropriate treatment, both the digital rectal examination (DRE) and the prostate-specific antigen (PSA) test are inaccurate.⁸ The PSA test in particular can lead to overtreatment of indolent PrCA,⁸ with potentially dire consequences such as incontinence.^9
-11 The debate on the utility of the PSA^12
-14 has led organizations such as the American Cancer Society (ACS),¹⁵ American Urological Association,¹⁶ and the US Preventive Services Task Force¹⁷ to recommend informed decision-making (IDM) with a health-care provider to determine whether or not to receive PSA screening for PrCA.

Informed decision-making is characterized by a patient (1) having a clear understanding about the disease, (2) possessing knowledge of the risks, benefits, and uncertainties of screening and subsequent treatment, and (3) actively engaging at the level of decision-making desired. Research suggests men involved in IDM have numerous benefits such as improved knowledge,^18

-23 reduced decisional conflict,^19,21,23 and higher IDM self-efficacy.^19,22,23 The opportunity to make an informed decision with a health-care provider about PrCA screening could lead to the early detection and treatment of PrCA¹⁵ and, subsequently, the reduction in AA PrCA disparities.

Prostate cancer knowledge is a critical component of IDM, yet few studies focus specifically on AA men’s PrCA knowledge.^22,24

-27 These studies show AA men have low to moderate knowledge about PrCA, which is often dependent on factors such as age, education, and information-seeking behaviors.²⁸ Even with sufficient knowledge, however, racial minorities may still be less likely than whites to actively engage in IDM with their health-care provider.²⁹ Disengagement can be explained, in part, by a lack of decision self-efficacy.³⁰ Informed decision-making self-efficacy is the “self-confidence or belief in one’s ability to make decisions, including participation in informed/shared decision-making.”³¹ To improve capacity to engage in IDM (ie, increase knowledge and decision self-efficacy), the Institute of Medicine recommends the use of effective, plain-language, and culturally appropriate communication strategies which can include interactive communication technologies such as computers and mobile phones.³² Studies have also recognized these technologies as appropriate mediums for disseminating PrCA content to diverse populations, including AA men.^23,33

-36

Emerging research also establishes the feasibility and acceptance of using embodied conversational agents (ie, ECAs or animated human-like characters) within computer-based decision aids (CBDAs).^37,38 iDecide incorporates an ECA model because it is more interactive than video and voice/text-only modalities and experts have shown it to be an effective technique for promoting knowledge and behavior change^39
-41 because ECAs mimic human interaction.⁴² Embodied conversational agents have also been successfully used with older adults and those with lower levels of computer and health literacy.^43,44 Lastly, pilot research, which informed the development of iDecide, indicated AA men are open to ECAs as a means for preparing them for IDM in PrCA screening.

The goal of this study was to evaluate iDecide, an ECA-led CBDA, to determine its effect on the PrCA knowledge, IDM self-efficacy, technology use self-efficacy of AA men, and intention to engage in IDM with a health-care provider. Multiple theories informed the development and evaluation of iDecide. For effective use of iDecide as a modality for supporting IDM, the Unified Theory of Acceptance and Use of Technology (UTTAUT)⁴⁵ suggests the CBDA must meet ease of use standards to ensure adoption by the target population. Further, according to the Theory of Cognitive Multimedia Learning,⁴⁶ PrCA content within iDecide must be thoughtfully designed to not overwhelm an individuals’ cognitive load. It is posited that because of the iterative, theory-grounded, and community-driven nature of iDecide,^47
-49 coupled with its highly interactive ECA model, this CBDA will lead to an increase in PrCA knowledge and IDM self-efficacy among AA men. Based on men’s perceptions about the ease of use of iDecide in an earlier pilot study,¹⁸ we also hypothesize that iDecide will increase their technology use self-efficacy.

Methods

Design

This study employed a 1-group, pretest/posttest design to determine whether the use of iDecide leads to a positive change in (1) PrCA knowledge, (2) IDM self-efficacy, (3) technology use self-efficacy, and (4) intention to engage in IDM with a health-care provider. Consenting eligible participants completed a self-administered, paper-and-pencil, preintervention survey consisting of 79 questions about sociodemographics, computer, e-mail, and web fluency,⁵⁰ PrCA information sources, PrCA screening behaviors, PrCA knowledge,⁵¹ IDM self-efficacy,³¹ and technology use self-efficacy.⁵² Men were then invited to use iDecide in a quiet room at the location where they were recruited (eg, church fellowship hall). Participants were not provided with prior training; however, researchers were present to help participants with questions. Immediately following the intervention, a postintervention survey was administered that included only outcome variables (eg, PrCA knowledge) and items assessing iDecide’s usability (ie, effort and performance expectancy, facilitating conditions^45,46). This study was approved by the institutional review board of University of South Carolina (approval # Pro00045407).

Sample

From July 2015 to February 2016, a purposive sample of 354 self-identified, AA men aged 40+ years were recruited from churches and other community, fraternal, and government organizations in South Carolina. Recruitment strategies included the distribution of flyers, in-person announcements at churches, newspaper ads, and radio appearances. To be eligible for the study, men were required to (1) self-identify as AA, (2) speak and comprehend English, (3) have no personal history of PrCA, and (4) have no self-reported history of cognitive decline. A detailed narrative of recruitment strategies is published elsewhere.⁵³ Our sample size was based on a calculation using SPSS vs 15 Statistical Software, which suggests a sample of 352 participants yields 80% power to detect a difference of 0.100 in a design with 2 repeated measurements having a compound symmetry covariance structure when the standard deviation (SD) is 1.000, the correlation between observations on the same participant is 0.100, and the α level is .050.

Measures

Independent Variables

Sociodemographic variables

Year of birth, marital status, employment, income, education, and health insurance type were obtained by self-report.

Computer, e-mail, and web fluency

Fluency in computer technology use was measured via 20 items asking participants to indicate their ability to use computer technology (eg, I can turn on a computer), e-mail (eg, I can send an e-mail message), and the Internet (eg, I can use a search engine). Response options were measured on a 5-point Likert scale ranging from “not at all” to “very well.” A total score was calculated by summing the responses from the 20 items (6 items: computer fluency; 7 items: e-mail fluency; 7 items: web fluency). The total score ranged from 20 (not at all) to 120 (very well). Internal consistency reliability for the subscales was high: computer fluency (α = .85), e-mail fluency (α = .89), and web fluency (α = .92).

Sources for health and PrCA information

One item was used to determine whether and from what sources men received PrCA information. Response options included interpersonal (eg, doctor, health educator, family member), traditional (eg, books, radio), and modern (eg, cell phone applications) sources.

Prostate Cancer Screening Behaviors

Prostate cancer screening

Prostate cancer screening was measured by self-report using 2 Likert-style items asking participants to indicate how long it had been since their last screening. Participants were also asked when they received the DRE and PSA, respectively. The 3 multiple-choice response options were: “I have never had this exam,” “less than 5 years ago,” and “5 or more years ago.” These response options were chosen because the PrCA screening recommendations against routine PSA screening were published in 2011³ (ie, within the past 5 years of study implementation), which may have influenced men’s screening behaviors.

Discussion with doctor about PrCA

One binary item (ie, “yes” or “no”) was used to determine whether participants had ever discussed and made a decision about PrCA screening with their health-care provider. This question captured the IDM process as defined by the ACS.

Outcome Variables

Prostate cancer knowledge

Participants’ knowledge of PrCA was assessed at pre- and postintervention using 20 items adapted from Cormier et al’s PrCA Knowledge scale.⁵¹ Knowledge domains included prostate anatomy and function, PrCA risk factors, PrCA screening, and PrCA symptoms. Response options for each item were “true,” “false,” and “I don’t know.” A total knowledge score was calculated by summing the 20 items for purposes of analysis (ie, 1 point for each correct item with a possible total score of 20).

Informed decision-making self-efficacy

Participants’ IDM self-efficacy was assessed at pre- and postintervention using an adapted version of O’Connor’s 11-item Decision Self-Efficacy scale.³¹ Response options were measured on a 5-point Likert scale ranging from “not at all confident” to “very confident.” Based on O’Connor’s user guide, the total self-efficacy score was created by summing the 11 items, dividing by 11 and then multiplying by 25. The total score ranged from 25 (not confident) to 125 (extremely confident).

Technology use self-efficacy

Assessed at pre- and postintervention, technology use self-efficacy was measured using Compeau and Higgins’ (1995) 10-item Computer Self-Efficacy scale. Response options were measured on a 5-point Likert scale ranging from “not confident at all” to “very confident.” The total score was calculated by summing the 10 items’ responses with a possible score of 50.

Computer-based decision aid ease of use

Participants’ perceived ease of use of the CBDA was assessed using an adapted version of a 24-item scale created by Venkatesh et al.⁴⁵ Response options were measured on a 5-point Likert scale ranging from “not confident at all” to “very confident.” The total score was calculated by summing 24 items with a possible score of 120.

Intention to engage in IDM

Participants’ intention to engage in IDM with a health provider about PrCA screening was measured using one original item asking whether they would engage in IDM within the next 6 months. Response options were measured on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.”

Intervention

The conceptual framework employed in this study was adapted from the UTTAUT⁴⁵ and Theory of Cognitive Multimedia Learning.⁴⁶ According to UTTAUT, performance expectancy, effort expectancy, social influence, and facilitating conditions (ie, degree to which an individual believes organizational and technical infrastructure exists to support use of the system), moderated by factors such as age and experience, contribute to an individual’s acceptance and use of new technology. According to the Theory of Cognitive Multimedia Learning, because individuals have limited capacity for filtering, organizing, selecting, and integrating information,⁴⁵ content presented through a technology must be designed strategically to reduce cognitive load. Based on this conceptual framework (Figure 1), we posited that iDecide would lead to increases in PrCA knowledge and IDM self-efficacy. Furthermore, we posited that iDecide would increase men’s technology use self-efficacy, which may lead them to use other similar technologies to find PrCA information, further enhancing their PrCA knowledge and IDM self-efficacy.

Figure 1.

Conceptual framework based on the Unified Theory of Acceptance and Use of Technology and Theory of Cognitive Multimedia Learning. CBDA indicates computer-based decision aid.

Intervention description

iDecide was developed based on PrCA information from (1) the National Cancer Institute’s (NCI) “What You Need to Know About Prostate Cancer” booklet,⁵⁴ (2) previous cancer disparities research,^55
-57 and (3) an NCI-funded study relevant to the education of men facing PrCA screening decisions.⁵⁸ These resources included information about PrCA anatomy, risk, symptoms, and screening recommendations; the controversy about the effectiveness of PrCA screening; the results of recent PrCA screening research; and the importance of IDM.

iDecide is not prescriptive, nor does it encourage individuals to use only the information presented by iDecide to make a final decision. iDecide was designed for use on a touch screen device such as a phone or tablet computer. In this study, iDecide was administered on a 10.5-in touchscreen tablet equipped with headphones. A full description of the iterative process used to develop iDecide can be found elsewhere.⁵⁹

iDecide was divided into two 5-minute sections: education and decision-making role-play. Section 1 of iDecide was designed to increase AA men’s knowledge about PrCA and used an ECA to guide users through a list of key factors to consider when making an informed decision regarding PrCA screening (eg, the risks and benefits of screening). This section included 30-second to 1-minute clips where an ECA presented information on various topics and then prompted the user through on-screen question and answer exercises which promoted repetition of salient points or allowed the user to move to a new topic. Section 2 of iDecide engaged users in a role-playing exercise with the ECA to prepare them to make informed decisions with their health-care provider about PrCA screening with the goal of building IDM self-efficacy. In this section, users had limited question selections prompting responses from the ECA which simulated a conversation with a health-care provider (see Figure 2).

Figure 2.

iDecide prostate cancer computer-based decision-aid.

Analysis

Descriptive statistics were conducted to assess participant baseline sociodemographics, prior sources for PrCA information, PrCA screening behaviors, and computer, e-mail, and web fluency. Paired t tests were performed to identify significant changes (P < .05) in outcome variables pre- and postintervention. A general linear model (GLM) analyses in SAS (MIXED procedures) was used to examine the effects of age, income, computer, e-mail, and web fluency (ie, well), marital status, income, education, time, time by education interaction, and time by income interaction, across all 3 outcome variables. A Spearman correlation was performed to assess the relationship between participants’ intention to engage in IDM and all outcome variables.

Results

Summary of Characteristics

The 354 AA male participants had a mean age of 59.5 (9.61) years (Table 1). Most were married (56%, n = 194), insured (91%; n = 323), and had a regular health-care provider (87%, n = 309). Nearly half were employed (48%, n = 167), reported a household income between $20 000 and $79 999 (48.9%, n = 165), and had a high school diploma/attended some college (55%; n = 192). Participants reported moderate levels of computer, e-mail, and web fluency (M = 70.86 [29.93]).

Table 1.

Summary of African American Male Participant Characteristics.

Characteristics	N = 354
Age, in years, mean (SD)	59.5 (9.61)
Marital status, n (%)
Single/never married	73 (20.98)
Married	194 (55.75)
Employment status
Employed	167 (48.41)
Retired	94 (27.25)
Unemployed	84 (24.35)
Annual household income 2014
<$20 000	100 (29.41)
$20,000-$79 999	165 (48.53)
≥$80 000	75 (22.06)
Number of people supported by income
1	103 (33.55)
2	117 (38.11)
3+	87 (28.34)
Education
Less than high school	37 (10.69)
High school/GED or some college	192 (55.49)
Bachelor’s degree or higher	117 (33.82)
Health Insurance status
Insured	438 (93.40)
No coverage	31 (8.88)
Where have you received information about prostate cancer?
Regular doctor	216 (62.43)
Health educator	94 (27.17)
Internet	51 (14.74)
Television	42 (12.14)
Family member/friend	41 (11.85)
Magazine	34 (9.83)
Newspaper	16 (4.62)
Other	72 (22.29)
Prostate cancer screening
PSA (past 5 years)	257 (73.43)
PSA (more than 5 years)	36 (10.29)
DRE (past 5 years)	239 (68.09)
DRE (more than 5 years)	51 (14.53)
Never received screening	38 (10.73)

Abbreviations: DRE, digital rectal examination; PSA, prostate-specific antigen; SD, standard deviation.

In relation to PrCA-related factors, most participants (77%, n = 270) reported having previously received information about PrCA from a doctor or other health-care provider (62%, n = 216). The majority of men also reported being screened for PrCA in the previous 5 years (77%, n = 270) using either the PSA (73%, n = 257) and/or the DRE (68%, n = 239). Of those screened (n = 310), 67% (n = 207) reported discussing PrCA screening with their provider. Of those participants who had not been screened (n = 38), 26% (n = 10) reported speaking with their health-care provider about the disease.

Postintervention Changes in Outcomes

Overall, participants experienced significant changes in PrCA knowledge (P ≤ .0001), IDM self-efficacy (P ≤ .0001), and technology use self-efficacy (P ≤ .0001). Because our population was largely comprised of those who had a history of PrCA screening (n = 316), we ran an ad hoc analysis (t test) to determine whether differences existed between those who had a history of screening versus those who had not received screening on each of our outcomes. Through this comparison, we found that gains in knowledge and IDM self-efficacy were greater for those participants who had a history of screening than those who did not (see Table 2). However, because these 2 groups both experienced significant increases across all 3 outcomes, the following sections will focus on providing detailed results and implications of iDecide on all participants collectively.

Table 2.

Change in Outcome Variables From Pre- to Postintervention By Screening Status.

Variables	Screened					Never Screened
	Pre			Post		Pre			Post
	n	Mean	SD	Mean	SD	n	Mean	SD	Mean	SD
Total Self-Efficacy Scale	313	82.10	18.96	88.46	15.18	38	74.46	26.98	80.74	24.07
Total Technology Scale	311	38.61	9.66	41.34	9.16	38	34.82	11.43	39.47	8.75
Total Well Scale	315	72.22	29.01	–	–	38	60.32	35.38	–	–
Total Knowledge Scale	314	11.63	3.94	15.46	3.13	38	8.95	3.45	14.58	4.73
Total CBDA Scale	313	–	–	95.13	11.64	38	–	–	93.39	11.64

Abbreviations: CBDA, computer-based decision aid; SD, standard deviation.

Prostate cancer knowledge

Overall, participants’ PrCA knowledge increased from pre- (M = 11.53, SD = 4.39) to postintervention on all outcomes (M = 15.47, SD = 3.45; P ≤ .0001; Table 3). When examining PrCA knowledge across categories, there were highest changes in the percentage of participants who answered correctly on questions related to screening, anatomy, and symptoms between pre- and postintervention (which averaged increases of 43%, 34%, and 27%, respectively). There were lower average changes in the percentage of participants answering correctly on questions related to PrCA risk factors (17% increase). Our GLM analyses examining salient confounders of iDecide on PrCA knowledge showed a significant interaction effect for time and education (P = .004) on knowledge. There was also a significant main effect for income (P = .01) in relation to PrCA knowledge (see Table 4). Therefore, the amount of knowledge change between pre- and postintervention is moderated by an individual’s level of education. More specifically, the lower an individual’s education level, the greater their gains were in knowledge following the intervention. Additionally, PrCA knowledge scores were the highest among those at the highest income bracket at both pre- and postintervention.

Table 3.
Prostate Cancer Knowledge by Survey Question (Pre- and Postintervention).

Domain Question Correct Response % Correct P Value

Pre Post

Anatomy and function 1. The prostate gland is a reproductive organ located below the bladder. True 210 (59.66) 313 (88.67) <.0001

2. The prostate gland makes some of the fluid that’s part of semen. True 190 (53.82) 325 (92.07) <.0001

Risk factors 3. Older men are more likely to get prostate cancer. True 283 (80.17) 292 (82.72) .4387

4. More AA men are diagnosed with prostate cancer than whites. True 291 (82.67) 328 93.18) <.0001

5. AA men who have fathers or brothers with prostate cancer are more likely to get prostate cancer than those who do not. True 191 (73.58) 312 (88.63) <.0001

6. Who do you think is more likely to get prostate cancer? Black Men 231 (65.62) 287 (81.53) <.0001

7. Who do you think is more likely to get prostate cancer? Man whose father has had prostate cancer 191 (54.26) 256 (72.71) <.0001

Screening 8. A PSA blood test can be done to check for prostate cancer. True 301 (85.51) 309 (87.78) .3435

9. A digital rectal examination or DRE can be done to check for prostate cancer. True 251 (71.71) 309 (88.28) <.0001

10. The only way a man can know for sure if he has prostate cancer is to have a prostate biopsy. True 206 (58.85) 277 (79.14) <.0001

11. A prostate biopsy is when a blood test is used to check for proteins in the blood. True 94 (26.86) 119 (34.00) .0013

12. Neither the PSA nor DRE is 100% accurate. True 166 (47.70) 286 (81.48) <.0001

Symptoms 13. A man can have prostate cancer and can have no symptoms. True 223 (63.53) 260 (73.65) .0040

14. The warning signs of prostate cancer are always present with prostate cancer. False 177 (50.72) 213 (60.34) .0220

15. Pain often in your lower back could be a sign of prostate cancer. True 92 (26.29) 224 (63.64) <.0001

16. Having a hard time passing urine. True 240 (68.38) 306 (86.93) <.0001

17. Passing urine often, especially at night. True 215 (61.60) 291 (83.62) <.0001

18. Blood in the urine or semen. True 216 (61.89) 306 (87.68) <.0001

19. Painful ejaculation. True 144 (41.14) 264 (75.86) <.0001

20. Pain in the stomach. True 102 (29.23) 182 (52.15) <.0001

Abbreviations: AA, African American; DRE, digital rectal examination; PSA, prostate-specific antigen.

Table 4.
P Value for Type 3 Test of Fixed Effects of Selected Variables by Knowledge, Self, and Technology.

Variables Knowledge Self-Efficacy Technology

Age .4492 .4067 .8716

Well - - .0001

Marital status .0758 .8776 .5839

Income .0112 .6937 .1377

Education .0241 .0214 .3532

Total Self-Efficacy Scale/pre .1461 - .0001

Total Technology Scale/pre .3364 .0001 -

Total Knowledge Scale/pre - .4782 .4132

Time .0001 .0001 .0001

Time × education .0039 - .0004

Time × income - .0091 -

Informed decision-making self-efficacy

There was a statistically significant difference from pre- (M = 81.24, SD = 20.06) to post- (M = 87.58, SD = 17.02) intervention for participants’ IDM self-efficacy (P < .0001; Table 2). The greatest change in an individual item was related to a participant’s confidence to figure out the best screening choice (M = 3.20 to M = 3.55). Participants’ highest scores, both at pre- and postintervention, related to their confidence in asking a medical question without feeling dumb (M = 3.44 to M = 3.63). Our GLM analyses on IDM self-efficacy showed a significant interaction effect for time and income on IDM self-efficacy (P = .004; Table 4). Therefore, the change in IDM self-efficacy between pre- and postintervention is moderated by level of income (ie, the lower an individual’s income, the greater their gains were in IDM self-efficacy from pre- to postintervention). In addition, results show a significant main effect between IDM self-efficacy and education level (P = .02). Thus, those with higher education levels had higher IDM self-efficacy scores at pre- and postintervention than those who had attained lower levels of education.

Technology use self-efficacy and CBDA ease of use

Participants experienced notable increases in technology use self-efficacy from pre- (M = 38.18, SD = 9.9) to post- (M = 41.10, SD = 9.13) intervention (P < .0001; Table 2). Participants, on average, reported at postintervention that iDecide was easy to use (M = 94.90, SD = 12.24) and that they would use iDecide in the future (M = 4.28). Our GLM analyses demonstrate a significant interaction effect for time and education in relation to technology use self-efficacy (P = .0004; Table 4). Results also show a main effect between computer, e-mail, and web fluency and technology use self-efficacy (P = .0001). Therefore, the amount of change in technology use self-efficacy between pre- and postintervention is moderated by level of education and those with the lowest levels of education had the greatest gains between pre- and postintervention. Additionally, participants’ computer, e-mail, and web fluency was significantly related to their technology use self-efficacy (P = .0001).

Intention to engage in IDM

Postintervention, 67% (n = 233/349) of respondents reported that they either agreed or strongly agreed that they planned to make an informed decision with their health-care provider about whether or not to receive PrCA screening within the next 6 months. The results from our Spearman correlation show that there is a weak, but statistically significant, positive correlation between participants’ intention to engage in IDM and IDM self-efficacy (r = 0.14, P = .02) There is also a weak, but statistically significant, negative correlation between intention to engage in IDM and age (r = 0.12, P = .009).

Conclusion

This study assessed the impact of iDecide on improving PrCA knowledge, IDM self-efficacy, and technology use self-efficacy among AA men. As mentioned, our population was largely comprised of those who had a history of PrCA screening. This population is representative of approximately one-third of AA men who receive routine PrCA screening. In addition, prior research demonstrates that AA men (particularly those 45+) in the authors’ demographic region have high rates of PrCA screening. The results of our ad hoc analyses to assess the differences that exist between those participants with a screening history and those who have never received screening revealed that those without a screening history have greater gains in PrCA knowledge, IDM self-efficacy, and technology use self-efficacy. This finding, particularly as it relates to PrCA knowledge and IDM self-efficacy, can partially be explained by the fact that those who do not have a history of PrCA screening may have been exposed to less PrCA information and may have less experience speaking with a health-care provider about PrCA screening. Despite these differences, changes in outcomes within these 2 groups were both statistically significant. Therefore, the remainder of the discussion will not focus on the dissection of the findings between groups, but rather on salient differences among all participants.

Consistent with recent literature on PrCA knowledge among AA men prior to participation in an intervention,^{18,22,24

-27} the use of iDecide led to significantly higher overall knowledge across all domains (ie, anatomy, risk factors, screening, and symptoms). Participants who had less than a high school education demonstrated the greatest gains following the intervention. This relationship can be partially explained by the fact that individuals with lower education levels lack access to plain-language health information and commonly have limited cognitive skills and problem-solving ability. These factors likely contributed to their substantially lower levels of PrCA knowledge prior to the intervention. However, having access to iDecide, a CBDA designed for individuals of varying literacy levels, greatly improved their existing PrCA knowledge.

It is also noteworthy that participants consistently exhibited low scores on one item related to whether a biopsy is a blood test used to check for proteins (similar to the PSA), with only 34% of men answering correctly postintervention. Knowing facts, such as the process for biopsy, is highly important for IDM because the biopsy is the next step in a series of decisions that will occur should a man’s PSA results show elevated levels. Additionally, a biopsy is a much more invasive decision, with potential side effects such as pain and infection.⁶⁰ Despite having low PrCA knowledge preintervention, most participants who did report being screened also reported speaking to their health-care provider prior to screening. Most participants also reported health-care providers were their main and preferred source for PrCA information. These results are similar to earlier research conducted by Owens and colleagues¹⁸ where over 51% of the participants report making a shared decision with their physician about whether to receive PrCA screening, even with limited PrCA knowledge. One explanation for this lack of knowledge preintervention is the time that may have elapsed since men’s last receipt of PrCA education. For example, Volk and colleagues⁶¹ found that men receiving an IDM intervention reported significant increases in knowledge about PrCA compared to the control group. However, knowledge decreased substantially at the 1-year follow-up. Thus, men in our study may have had a thorough knowledge of PrCA at the time they were screened (eg, 2 years ago), but this knowledge may have since diminished.

Some of our findings related to AA men having low PrCA knowledge, but often report making an informed PrCA screening decision, may derive from AA’s common deference to a health-care provider to make health-related decisions (as opposed to engaging in IDM with a health-care provider).^62,63 For example, Peek and colleagues⁶² found that AA men were less likely to share information about disease symptoms or question the authority of a health-care provider, particularly if that provider was white. In earlier work,⁶³ they also demonstrated AAs feel health-care providers did not involve them in co-deliberation (a pillar in the IDM process), but instead followed a paternalistic model⁶⁴ by providing recommendations to which they typically agreed. African Americans’ deference to, or agreement with, a health-care provider in the manner described can be referred to as an informed consenting process, a scenario where a health-care provider simply obtains a valid, voluntary consent prior to screening.⁶⁵ The difference between IDM and processes such as informed consent is that patients engaged in IDM with a health-care provider have adequate PrCA knowledge to make a decision consistent with their values and are involved at the level they desire.⁶⁵ Therefore, it may be possible participants received screening solely based on a provider recommendation.⁶⁵

In addition to having adequate knowledge, participants’ roles in decision-making can be partly driven by their IDM self-efficacy. Self-efficacy^66
-68 has been evidenced in multiple studies as a key promoter in men’s level of involvement in their IDM process. Therefore, a man with higher IDM self-efficacy may be more likely to actively engage in IDM. Prior to our intervention, men had relatively low IDM self-efficacy scores, which is further indicative that participants may not have engaged in IDM with a health-care provider but consented to screening. Participants’ IDM self-efficacy scores following the intervention demonstrated a statistically significant increase, which suggests that users of iDecide may be more comfortable actively engaging in IDM with their health-care provider. Participants with lower incomes had the greatest gains in IDM self-efficacy following the use of iDecide, which may also be indicative of this groups’ lack of access to clear health information similar to those with lower education⁶⁹ (ie, education is a strong correlate of income).⁷⁰

iDecide also improved participants’ technology use self-efficacy, with the highest gains found among individuals who had lower levels of education. Those with higher levels of computer, e-mail, and web fluency were also more likely to have high technology use self-efficacy at pre- and postintervention. With the increasing adoption of mobile and computer-based technology, the need for culturally appropriate, user-friendly decision aids, and the success of iDecide and other interventions for enhancing IDM capacity, we believe technology will increasingly become a more ubiquitous resource for providing health information and supporting screening decisions. Men who have higher technology use self-efficacy may be more likely to use these resources to explore new information or supplement the information provided by their health-care provider. We expect that because iDecide increases technology use self-efficacy, users (particularly those who had lower technology use self-efficacy prior to use) may be more likely to use other technologies with similar features to gain additional information about PrCA and other health conditions. The receipt of this additional PrCA information may lead to maintenance of, or increase in, PrCA knowledge and additional gains in IDM and technology self-efficacy beyond our intervention.

The key goal for developing iDecide was to provide a resource to prepare AA men to engage in IDM with a health-care provider about whether or not to receive PrCA screening. Over half (67%) of the men who participated in our study reported they intend to participate in IDM with their health-care provider within the next 6 months. However, participants who were younger and had higher IDM self-efficacy had a slightly higher intent to engage in IDM. Based on our conceptual framework, we expected knowledge to have a high correlation with intention to engage in IDM, but this correlation was not significant. Therefore, IDM self-efficacy may be more important to IDM intention than PrCA knowledge. In relation to age, prior research has shown that younger AA men may be less likely to engage in IDM about PrCA screening than older AA men,²⁶ but our findings are contradictory. This may be due to the high number of AAs in our sample who have been screened in the past year.

Limitations

There were some notable limitations in this study. The population of AA participants resided in one region in South Carolina; so, findings may not be applicable to other AA men. Men were not asked to make an actual screening decision following the intervention; therefore, we neither measured screening decisions based on the intervention nor their decision quality/satisfaction. Most men in our study had received PrCA screening, so the implications of our intervention on men who do not have a history of screening are limited. Furthermore, these results cannot be applied to men of other races and ethnicities. Despite these limitations, our study has provided valuable information regarding the efficacy and cultural appropriateness of using a CBDA to facilitate PrCA education and promote IDM and technology self-efficacy in AA men. Future studies should examine how effective iDecide is in comparison to traditional education methods such as paper-based or in-person health education. These studies should also include more diverse samples of AA men, such as those who live in rural areas and/or have lower education/health literacy levels. It will also be beneficial to complete 1 (3 months to 1 year) postintervention follow-up to determine how well men retain the knowledge gained through iDecide, whether their levels of informed decision self-efficacy remain similar to those measured immediately following the intervention, and whether they used technology to find additional PrCA information. In addition, future studies should examine the effects of iDecide on informed decision-making for actual screening behaviors in a clinical setting. Lastly, research should explore if iDecide is deployed through an electronic medical record system, does it lead to benefits (eg, cost savings, time savings) for the health-care provider and/or institutions?

So What?

What is already known on this topic?

Interactive communication technologies are culturally appropriate mediums for disseminating plain-language PrCA content to diverse populations, including AAs.^23,33

-36

What does this article add?

To our knowledge, iDecide is the only documented ECA-led, CBDA developed for AA men using a community-driven methodology.^57,59 Our findings show iDecide is an effective resource for increasing the PrCA knowledge, IDM self-efficacy, technology use self-efficacy, and intention to engage in IDM among AA men. Further, our findings show that this tool is appropriate for individuals with lower socioeconomic statuses.

What are the implications for health promotion practice or research?

The preparation provided to AA men through iDecide may also help health-care providers optimize the short time allotted for a patient visit.⁷¹ Furthermore, the incorporation of iDecide into clinic settings through their electronic medical record systems can also contribute to additional financial incentives for clinics offered through the Health Information Technology for Economic and Clinical Health Act, which rewards medical institutions improving patients’ access to their health information.^72,73 However, we acknowledge that there are additional considerations prior to implementing iDecide in a clinic setting, such as (1) the compatibility of iDecide with an existing electronic medical record system, (2) the training and technical support needed for staff and users, and (3) the need for an implementation plan or protocol detailing when and how patients will interact with iDecide.⁷⁴

Domain	Question	Correct Response	% Correct	P Value
Anatomy and function	1. The prostate gland is a reproductive organ located below the bladder.	True	210 (59.66)	313 (88.67)	<.0001
2. The prostate gland makes some of the fluid that’s part of semen.	True	190 (53.82)	325 (92.07)	<.0001
Risk factors	3. Older men are more likely to get prostate cancer.	True	283 (80.17)	292 (82.72)	.4387
4. More AA men are diagnosed with prostate cancer than whites.	True	291 (82.67)	328 93.18)	<.0001
5. AA men who have fathers or brothers with prostate cancer are more likely to get prostate cancer than those who do not.	True	191 (73.58)	312 (88.63)	<.0001
6. Who do you think is more likely to get prostate cancer?	Black Men	231 (65.62)	287 (81.53)	<.0001
7. Who do you think is more likely to get prostate cancer?	Man whose father has had prostate cancer	191 (54.26)	256 (72.71)	<.0001
Screening	8. A PSA blood test can be done to check for prostate cancer.	True	301 (85.51)	309 (87.78)	.3435
9. A digital rectal examination or DRE can be done to check for prostate cancer.	True	251 (71.71)	309 (88.28)	<.0001
10. The only way a man can know for sure if he has prostate cancer is to have a prostate biopsy.	True	206 (58.85)	277 (79.14)	<.0001
11. A prostate biopsy is when a blood test is used to check for proteins in the blood.	True	94 (26.86)	119 (34.00)	.0013
12. Neither the PSA nor DRE is 100% accurate.	True	166 (47.70)	286 (81.48)	<.0001
Symptoms	13. A man can have prostate cancer and can have no symptoms.	True	223 (63.53)	260 (73.65)	.0040
14. The warning signs of prostate cancer are always present with prostate cancer.	False	177 (50.72)	213 (60.34)	.0220
15. Pain often in your lower back could be a sign of prostate cancer.	True	92 (26.29)	224 (63.64)	<.0001
16. Having a hard time passing urine.	True	240 (68.38)	306 (86.93)	<.0001
17. Passing urine often, especially at night.	True	215 (61.60)	291 (83.62)	<.0001
18. Blood in the urine or semen.	True	216 (61.89)	306 (87.68)	<.0001
19. Painful ejaculation.	True	144 (41.14)	264 (75.86)	<.0001
20. Pain in the stomach.	True	102 (29.23)	182 (52.15)	<.0001

Variables	Knowledge	Self-Efficacy	Technology
Age	.4492	.4067	.8716
Well	-	-	.0001
Marital status	.0758	.8776	.5839
Income	.0112	.6937	.1377
Education	.0241	.0214	.3532
Total Self-Efficacy Scale/pre	.1461	-	.0001
Total Technology Scale/pre	.3364	.0001	-
Total Knowledge Scale/pre	-	.4782	.4132
Time	.0001	.0001	.0001
Time × education	.0039	-	.0004
Time × income	-	.0091	-

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the University of South Carolina’s Office of the Vice President for Research. Funding was also received through the University of South Carolina’s School of Pharmacy through an American Cancer Society Institutional Research Grant and 2 grants from the US Department of Health and Human Services’ National Cancer Institute (U54 CA15346; PI: Hébert and U54-MD010706; PI: Hughes-Halbert).

References

Siegel

Miller

Jemal

. Cancer statistics, 2016. CA Cancer J Clin. 2016;66(1):7–30. doi:10.3322/caac.21332.

Hebert

Daguise

Hurley

. Mapping cancer mortality-to-incidence ratios to illustrate racial and gender disparities in a high-risk population. Cancer. 2009;115(11):2539–2552. doi:10.1002/cncr.24270.

Moyer

; US Preventive Services Task Force. Screening for prostate cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2012;157(2):120–134. doi:10.7326/0003-4819-157-2-201207170-00459.

Carlsson

Vickers

Roobol

. Prostate cancer screening: facts, statistics, and interpretation in response to the US Preventive Services Task Force review. J Clin Oncol. 2012;30(21):2581–2584. doi:10.1200/JCO.2011.40.4327.

Schröder

. Stratifying risk—The U.S. Preventive Services Task Force and prostate-cancer screening. N Engl J Med. 2011;365(21):1953–1955. doi:10.1056/NEJMp1112140.

Brett

Ablin

. Prostate-cancer screening—what the US Preventive Services Task Force left out. N Engl J Med. 2011;365(21):1949–1951. doi:10.1056/NEJMp1112191.

Catalona

. The United States preventive services task force recommendation against prostate-specific antigen screening—counterpoint. Cancer Epidemiol Biomarkers Prev. 2012;21(3):395–397. doi:10.1158/1055-9965.

Prostate Cancer Prevention and Early Detection. American Cancer Society. Web site. http://www.cancer.org/acs/groups/cid/documents/webcontent/003182-pdf.pdf. Updated 2016. Accessed June 2, 2016.

US Department of Health and Human Services. Prostate Cancer Screening: A Decision Guide for African American Men. Washington, DC: US Department of Health and Human Services; 2003.

10.

Welch

Albertsen

. Prostate cancer diagnosis and treatment after the introduction of prostate-specific antigen screening: 1986-2005. J Natl Cancer Inst. 2009;101(9):1325–1329. PMCID:2758309.

11.

Sandhu

Andriole

. Overdiagnosis of prostate cancer. J Natl Cancer Inst Monogr. 2012;2012(45):146–151.

12.

Andriole

Crawford

Grubb

III . Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360(13):1310–1319. doi:10.1056/NEJMoa0810696.

13.

Barry

. Screening for prostate cancer—the controversy that refuses to die. N Engl J Med. 2009;360(13):1351–1354. doi:10.1056/NEJMe090116614.

14.

Schröder

Hugosson

Roobol

; ERSPC Investigators. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360(13):1320–1328.

15.

American Cancer Society. Prostate Cancer: Early Detection. Atlanta, GA: American Cancer Society; 2013.

16.

Carter

Albertsen

Barry

. Early detection of prostate cancer: AUA Guideline. J Urol. 2013;190(2):419–426. doi:10.1016/j.juro.2013.04.119.

17.

Bibbins-Domingo

Grossman

Curry

. The US Preventive Services Task Force 2017 draft recommendation statement on screening for prostate cancer: an invitation to review and comment. JAMA. 2017;317(19):1949–1950. doi:10.1001/jama.2017.4413.

18.

Owens

Jackson

Thomas

. Prostate cancer knowledge and decision making among African-American men and women in the southeastern United States. Int J Mens Health. 2015;14:55–70. PMCID:PMC4505933.

19.

Lepore

Wolf

Basch

. Informed decision making about prostate cancer testing in predominantly immigrant black men: a randomized controlled trial. Ann Behav Med. 2012;44(3):320–330. doi:10.1007/s12160-012-9392-3.

20.

Sheridan

Golin

Bunton

. Shared decision making for prostate cancer screening: the results of a combined analysis of two practice-based randomized controlled trials. BMC Med Inform Decis Mak. 2012;12(1):130. doi:10.1186/1472-6947-12-130.

21.

Williams

Davis

Luta

. Fostering informed decisions: a randomized controlled trial assessing the impact of a decision aid among men registered to undergo mass screening for prostate cancer. Patient Educ Couns. 2013;91(3):329–336. doi:10.1016/j.pec.2012.12.013.

22.

Drake

Shelton

Gilligan

Allen

. A church-based intervention to promote informed decision making for prostate cancer screening among African American men. J Natl Med Assoc. 2010;102(3):164–171.

23.

Allen

Mohllajee

Shelton

Drake

Mars

. A computer-tailored intervention to promote informed decision making for prostate cancer screening among African American men. Am J Mens Health. 2009;3(4):340–351. doi:10.1177/1557988308325460.

24.

Dhillon

Grivas

Raska

. Informed decision making (IDM) for prostate cancer (PCa) screening in a high-risk population. J Clin Oncol. 2017;35(suppl 6):112–112. doi:10.1200/JCO.2017.35.6_suppl.112.

25.

Ukoli

Patel

Hargreaves

. A tailored prostate cancer education intervention for low-income African Americans: impact on knowledge and screening. J Health Care Poor Underserved. 2013;24(1):311–331. doi:10.1353/hpu.2013.0033.

26.

Jackson

Owens

Friedman

Dubose-Morris

. Innovative and community-guided evaluation and dissemination of a prostate cancer education program for African-American men and women. J Cancer Educ. 2015;30(4):779–785. doi:10.1007/s13187-014-0774-z.

27.

Jackson

Owens

Friedman

Hebert

. An intergenerational approach to prostate cancer education: findings from a pilot project in the southeastern USA. J Cancer Educ. 2014;29(4):649–656. doi:10.1007/s13187-014-0618-x.

28.

Ross

Dark

Orom

. Patterns of information behavior and prostate cancer knowledge among African-American men. J Cancer Educ. 2011;26(4):708–716. doi:10.1007/s13187-011-0241-z.

29.

Hawley

Morris

. Cultural challenges to engaging patients in shared decision making. Patient Educ Couns. 2017;100(1):18–24. doi:10.1016/j.pec.2016.07.008.

30.

Carpenter

Godley

Clark

. Racial differences in trust and regular source of patient care and the implications for prostate cancer screening use. Cancer. 2009;115(21):5048–5059. doi:10.1002/cncr.2453

31.

O’Connor

. User Manual—Decision Self-Efficacy Scale. Ottawa, Ontario, Canada: Ottawa Hospital Research Institute; 1995.

32.

Institute of Medicine, Committee on Health Literacy, Board of Neuroscience and Behavioral Health. Health Literacy: A Prescription to End Confusion. Washington, DC: The National Academies Press; 2004.

33.

Ellison

Weinrich

Lou

Powell

Baquet

. A randomized trial comparing web-based decision aids on prostate cancer knowledge for African-American men. J Natl Med Assoc. 2008;100(10):1139–1145.

34.

Evans

Joseph-Williams

Edwards

. Supporting informed decision making for prostate specific antigen (PSA) testing on the web: an online randomized controlled trial. J Med Internet Res. 2010;12(3):e27. doi:10.2196/jmir.1305.

35.

Kassan

Williams

Kelly

. Men’s use of an internet-based decision aid for prostate cancer screening. J Health Commun. 2012;17(6):677–697. doi:10.1080/10810730.2011.579688.

36.

Sultan

Rivers

Osongo

. Affecting African American men’s prostate cancer screening decision making through a mobile tablet-mediated intervention. J Health Care Poor Underserved. 2014;25(3):1262–1277. doi:10.1353/hpu.2014.0148.

37.

Lisetti

Yasavur

de Leon

Amini

Rishe

Visser

. Building an on-demand avatar-based health intervention for behavior change. Paper presented at: Proceedings of the 25th International FLAIRS Conference; May 23-25, 2012; Marco Island, FL.

38.

Schulman

Bickmore

Sidner

. An intelligent conversational agent for promoting long-term health behavior change using motivational interviewing. Palo Alto, CA: AAAI Spring Symposium on AI and Health Communication; 2011.

39.

Baker

Wentz

Woods

. Using virtual worlds in education: Second Life® as an educational tool. Teach Psychol. 2009;36(1):59–64.

40.

Beard

Wilson

Morra

Keelan

. A survey of health-related activities on second life. J Med Internet Res. 2009;11(2):e17.

41.

Wood

. Using emerging technologies to enhance learning. Nurs Sci Q. 2010;23(2):173–179.

42.

van den Brekel

AJP

. Get Your Consumer Health Information From an Avatar!: Health and Medical Related Activities in a Virtual Environment. Krakow, Poland: European Association for Health Information & Libraries Workshop; 2008.

43.

King

Bickmore

Campero

Pruitt

Yin

. Employing virtual advisors in preventive care for underserved communities: results from the COMPASS study. J Health Commun. 2013;18(2):1449–1464.

44.

Bickmore

Pfeifer

Paasche-Orlow

. Using computer agents to explain medical documents to patients with low health literacy. Patient Educ Couns. 2009;75(3):315–320.

45.

Venkatesh

Morris

Davis

. User acceptance of information technology: toward a unified view. MIS Quart. 2003;27(3):425–478. doi:10.2307/30036540.

46.

Mayer

Moreno

. Nine ways to reduce cognitive load in multimedia learning. Educ Psychol. 2003;38(1):43–52. doi:10.1207/S15326985EP3801_6.

47.

Israel

Schulz

Parker

Becker

Allen

III Guzman

. Critical issues in developing and following community based participatory research principles. In: Minkler

ed. Community-Based Participatory Research for Health. San Francisco, CA: Jossey-Bass; 2003:53–79.

48.

Israel

Schulz

Parker

Becker

; Community-Campus Partnerships for Health. Community-based participatory research: policy recommendations for promoting a partnership approach in health research. Educ Health (Abingdon). 2001;14(2):182–197. doi:10.1080/13576280110051055.

49.

Israel

Schulz

Parker

Becker

. Review of community-based research: assessing partnership approaches to improve public health. Annu Rev Public Health. 1998;19:173–202. doi:10.1146/annurev.publhealth.19.1.173.

50.

Bunz

. The computer-email-web (CEW) fluency scale—development and validation. Int J Hum Comput Interact. 2004;17(4):479–506.

51.

Cormier

Kwan

Reid

Litwin

. Knowledge and beliefs among brothers and sons of men with prostate cancer. Urology. 2002;59(6):895–900. doi:10.1016/S0090-4295(01)01657-0.

52.

Compeau

Higgins

. Computer self-efficacy: development of a measure and initial test. MIS Quart. 1995;19(2):189–211. doi:10.2307/249688.

53.

Owens

James

Friedman

. Overcoming the challenges of African-American recruitment in health sciences research: strategies and recommendations. Urol Nurs. 2017;37(6):293–315.

54.

National Cancer Institute. What You Need to Know About Prostate Cancer. Bethesda, MD: National Cancer Institute; 2008.

55.

Friedman

Corwin

Dominick

Rose

. African American men’s understanding and perceptions about prostate cancer: why multiple dimensions of health literacy are important in cancer communication. J Community Health. 2009;34(5):449–460. doi:10.1007/s10900-009-9167-3.

56.

Friedman

Thomas

Owens

Hébert

. It takes two to talk about prostate cancer a qualitative assessment of African American men’s and women’s cancer communication practices and recommendations. Am J Mens Health. 2012;6:472–84 doi:10.1177/1557988312453478.

57.

Owens

Friedman

Brandt

Bernhardt

Hébert

. Digital solutions for informed decision making: an academic–community partnership for the development of a prostate cancer decision aid for African American men. Am J Mens Health. 2016;10(3):207–19. doi:10.1177/1557988314564178.

58.

Chan

McFall

Byrd

. A community-based intervention to promote informed decision making for prostate cancer screening among Hispanic American men changed knowledge and role preferences: a cluster RCT. Patient Educ Couns. 2011;84(2):e44–e51. doi:10.1016/j.pec.2010.07.033.

59.

Owens

Friedman

Brandt

Bernhardt

Hebert

. An iterative process for developing and evaluating a computer-based prostate cancer decision aid for African American men. Health Promot Pract. 2015;16(5):642–655. doi:10.1177/1524839915585737.

60.

Loeb

Vellekoop

Ahmed

. Systematic review of complications of prostate biopsy. Eur Urol. 2013;64(6):876–892. doi:10.1016/j.eururo.2013.05.049.

61.

Volk

Spann

Cass

Hawley

. Patient education for informed decision making about prostate cancer screening: a randomized controlled trial with 1-year follow-up. Ann Fam Med. 2003;1(1):22–28. doi:10.1370/afm.7.

62.

Peek

Odoms-Young

Quinn

Gorawara-Bhat

Wilson

Chin

. Race and shared decision making: perspectives of African-Americans with diabetes. Soc Sci Med. 2010;71(1):1–9. doi:10.1016/j.socscimed.2010.03.014.

63.

Peek

Wilson

Gorawara-Bhat

Odoms-Young

Quinn

Chin

. Barriers and facilitators to shared decision making among African-Americans with diabetes. J Gen Intern Med. 2009;24(10):1135. doi:10.1007/s11606-009-1047-0.

64.

Smith

Manassaram-Baptiste

Brooks

. Cancer screening in the United States, 2014: a review of current American Cancer Society guidelines and current issues in cancer screening. CA Cancer J Clin. 2014;64(1):30–51. doi:10.3322/caac.21212.

65.

Rimer

Briss

Zeller

Chan

Woolf

. Informed decision making: what is its role in cancer screening? Cancer. 2004;101(suppl 5):1214–1228. doi:10.1002/cncr.20512.

66.

Purnell

. Key concepts relevant to quality of complex and shared decision making in health care: a literature review. Soc Sci Med. 2012;74(4):582–587. doi:10.1016/j.socscimed.2011.11.015.

67.

Légaré

Thompson-Leduc

. Twelve myths about shared decision making. Patient Educ Couns. 2014;96(3):281–286. doi:10.1016/j.pec.2014.06.014.

68.

Durand

Carpenter

Dolan

. Do interventions designed to support shared decision making reduce health inequalities? A systematic review and meta-analysis. PLoS One. 2014;9(4):e94670. doi:10.1371/journal.pone.0094670.

69.

Zimmerman

Woolf

Haley

. Understanding the Relationship Between Education and Health. Rockville, MD: Agency for Healthcare Research and Quality; 2015.

70.

Tamborini

Kim

Sakamoto

. Education and lifetime earnings in the United States. Demography. 2015;52(4):1383–1407.

71.

Schroy

Mylvaganam

Davidson

. Provider perspectives on the utility of a colorectal cancer screening decision aid for facilitating shared decision making. Health Expect. 2014;17(1):27–35. doi:10.1111/j.1369-7625.2011.00730.x.

72.

Blumenthal

. Launching HITECH. N Engl J Med. 2010;362(5):382–385. doi:10.1056/NEJMp0912825.

73.

Step 5: Achieve Meaningful Use Stage 2. HealthIT.gov . Web site. https://www.healthit.gov/providers-professionals/achieve-meaningful-use/core-measures-2/patient-specific-education-resources. Updated February 24, 2014. Accessed November 11, 2017.

74.

Lenert

Dunlea

Del Fiol

Hall

. A model to support shared decision making in electronic health records systems. Med Decis Making. 2014;34(8):987–995. doi:10.1177/0272989X14550102.