Consumer Familiarity, Perspectives and Expected Value of Personalized Medicine With a Focus on Applications in Oncology

Susan Garfield; Michael P Douglas; Karen V MacDonald; Deborah A Marshall; Kathryn A Phillips


Personalized Medicine. 2015;12(1):13-22. 

In This Article


Sample Demographics

Between 26 February and 4 March 2013, 602 respondents completed the survey out of 1016 invited, for a response rate of 59%. The survey took an average of 14 minutes to complete. The respondents were generally middle age (mean 53 years), 52% female, nonethnic (69%), married (61%), variably educated (31% college grads) and employed (53%). The mean household income of participants was $60,550 with a median household size of 2 (Table 1). Respondents were older, more likely to be married and had higher incomes compared with the national average. This is largely due to the inclusion criteria of being aged 30 and over. Overall, respondents were relatively healthy (83% reporting good, very good, or excellent health; 16% reporting very poor, poor, or fair health). Overall, most (67%) reported having one or more medical condition diagnosed. Considering cancer specifically, 8% reported having or having had cancer (3% of which is skin cancer).

Familiarity With Personalized Medicine

In general, there is low familiarity with the term 'personalized medicine' among the respondents, with 73% of individuals indicating they have not heard the term 'personalized medicine' (Figure 1). Of those who have heard of personalized medicine (27%, n = 163), males were more likely to have heard of this term than females (33% of males vs 22% of females). Furthermore, of those who heard of personalized medicine, only 8% consider themselves very knowledgeable, compared with 41% indicating not at all or not very knowledgeable (p < 0.05%).

Figure 1.

Understanding personalized medicine term and knowledge. All responses were based on a 1–7 Likert scale, with 1 being 'not at all knowledgeable' and 7 being 'very knowledgeable'.

Once the description and definition of personalized medicine was provided, most respondents (63%) thought personalized medicine would have a very positive or positive impact, while just 5% thought personalized medicine would have a negative or very negative impact. There is variation in expectations of the timing of that impact, be it in the short, i.e., 1–3 years (19%), medium, i.e., 3–5 years (22%), or longer, i.e., more than 5 years (49%) term. At significant levels, (p < 0.05%) respondents who rated their health as very good to excellent report having a higher knowledge of personalized medicine and perceive it more positively. Conversely, respondents who rated their health as poor were less receptive to personalized medicine.

Perceived Likely Adherence to Personalized Medicine Recommendations

Respondents were asked to predict their likely action upon receiving personalized medicine test results that indicated a more efficacious treatment was not going to work for them, in the hypothetical situation they had been diagnosed with a life threatening cancer. Most (71%) consumers predicted they would get a second opinion. An additional 13% said they would want to get the treatment anyway. In total, most (84%) respondents predict no immediate behavior or treatment change from a personalized medicine approach to treatment selection if it suggests forgoing treatment. This is in contrast to a small subset of respondents (13%) who say they would not get the treatment accepting that it would not work.

Perceived Economic Impact

When we examined perceptions of the impact of personalized medicine on healthcare costs, many felt that it would increase overall healthcare costs in both the short (next 5 years) and long term (next 11+ years) (40%- short-term, 38%- longer term). However, some disagreed and saw hope for personalized medicine reducing costs in the long term (18%) (Figure 2).

Figure 2.

Consumers expected impact on healthcare costs in the USA. All responses were based on a Likert scale of personalized medicine's expected impact on healthcare costs, with 1 being 'significantly decrease' and 7 being 'significantly increase' PM: Personalized medicine.

The value of personalized medicine, for consumers, does not rest solely with perceived overall health related cost-avoidance. In contrast, personalized medicine is seen as most valuable for tailoring treatments after diagnosis (44% reported as most valuable), minimizing the impact of diseases through preventative medicine (42% reported as most valuable) and predicting what diseases they may get in the future (13%).

To understand how cost-sharing within personalized medicine might impact preferences, consumers were asked whether they would choose a treatment that has a high likelihood of success but higher rates of side effects, versus a lower likelihood of success and lower side effect rates first. Over three quarters (77%) chose the higher efficacy treatment, versus 23% choosing the lower efficacy option. When introduced to the same scenario for the more efficacious drug being 'high cost,' represented by a $100,000 cost with a $10,000 co-pay, and the lower efficacy drug having no co-pay only slightly over half (53%) chose the more efficacious high cost drug and 47% chose the lower cost, lower efficacy option. Respondents with multiple health conditions were more likely to choose the lower cost lower efficacy option than those in good health.

Personalized Medicine Testing

In the second part of the survey, questions specifically dealing with testing as a component of personalized medicine were asked. Overall, 32% of respondents were very interested, 47% moderately interested and 21% not at all interested in testing, described as:

  • Being able to predict what diseases you may get in the future and attempt to either minimize the impact of that disease or avoid it altogether through the implementation of personalized, preventive medicine;

  • Once diagnosed with a disease, tailor treatments, predicting whether a medication is likely to help or hurt you before you ever take it.

In contrast, 68% of those who had ever been diagnosed with a life threatening cancer were very interested in the concept (n = 56). Additionally, those interested in personalized medicine testing tended to be more educated, have higher interests in personalized medicine generally, live in more metropolitan areas, have higher incomes and have more access to the internet than those who were not interested in personalized medicine testing.

Cost-concerns and disease history in addition to payer type also impact receptivity to personalized medicine related testing. In general, respondents with employer sponsored health plans were most interested in testing when compared with other insured groups. This differential held true when the hypothetical test cost was $500 (Figure 3).

Figure 3.

Interest in genetic testing with and without cost prompt, by respondent's insurance type.

A proportion of respondents (43%) indicated they are willing to pay $500 for a predictive or prognostic test. The dollar amount respondents are willing to pay out-of-pocket increases along with predicted rates of cancer survival (Figure 4). Of those who have ever been diagnosed with cancer, more (62%) were very interested in testing at that cost than the general respondent pool. Many respondents (45%) did not think cost would impact their willingness to recommend testing to family members who had been diagnosed with a life-threatening cancer. Though near equal numbers of others felt it would (29%) or they were not sure (26%).

Figure 4.

Interest in and amount willing to pay out-of-pocket for blood test if diagnosed with cancer with varying survival rates. All responses were based on a 1–7 Likert scale with 1 being 'not at all Interested' and 7 being 'very interested'

However, when consumers were asked what price they would be willing to pay for testing, without information about the relationship between the test and survival improvement, the median price reported was $200. When told that survival improvement was 40%, 60%, or 80%- compared with an expected average survival without testing of 20%, median willingness to pay was $100, $200 and $400, respectively. These results suggest that respondents are willing to bear a cost for testing that positively impacts survival, and that willingness to pay increases with the expected clinical impact of the test.