The informatics of direct-to-consumer genetic tests

Introduction

Direct-to-consumer (DTC) genetic tests are personalized genetic risk assessments sold directly to consumers over the internet. After purchasing a DTC genetic test, consumers send the company a sample of their saliva, which uses it to scan their DNA for single nucleotide polymorphisms (SNPs) which have been associated with various different complex diseases, such as diabetes or cancers. Using this method, companies claim to provide an assessment of the consumer’s genetic risks for these diseases, with the intention of encouraging them to mitigate any high risks by improving their health behaviour (e.g. increasing their level of exercise or improving their diet). However, there have been many criticisms of DTC genetic tests in the literature, including accusations of inaccuracy and unreliability.

One area of concern is the information provided to the consumer by the companies that sell the tests, with worries that the quantity and quality of information provided on company websites is frequently poor, with important information often left out and misleading information often included. For example, one content analysis found that there was a wide variation in the amount and usability of information provided on company websites, and stated that it was “apparent that most users would struggle to find and understand the important information on most sites”.¹ Another study found that a majority of websites did not “meet a minimum set of quality criteria” for the information they should provide, leading to “fundamental information deficits”.²  These findings raise questions about the ability of consumers to give fully informed consent to the DNA analysis, and to have access to the necessary information to correctly interpret and understand their results.  However, to the authors’ knowledge, no study has actually examined these information issues from the consumers’ point of view.

A second area of concern is the effect that the tests may have on the health behaviour and health anxiety of consumers. There has been much concern in the literature that (contrary to claims made by the companies) results showing an increased risk of disease may not cause an improvement in health behaviours, that results showing a decreased risk may actually encourage people to curtail their healthy behaviour efforts and that the tests may increase health anxiety. However, to the author’s knowledge, only three studies to date have examined the effect of the tests on groups of users.  The first of these was a survey of 1048 customers of three of the main DTC genetic testing companies, which found that a third had improved their diet and 14% had increased their exercise levels.³ In the second study, which consisted of interviews with 60 participants who had been given a free test, approximately one third had changed their behaviour and another third were planning to.⁴ In contrast the third study, in which 2037 were provided with a subsidised test, found no significant changes in participants’ level of exercise, proportion of fat in diet or health anxiety after three months.⁵    

Objectives

The study had two broad objectives.

Firstly, to investigate the informational aspects of DTC genetic tests from the consumers’ point of view. This included the information need of consumers, where and how they searched for information about the tests and their opinion of the information provided to them by the companies.

Secondly, to assess the effect of DTC genetic tests on consumers’ health behavior and health anxiety.

Methods

The study consisted of an online survey.  Respondents were identified and contacted through social media: namely Twitter, Facebook, internet forums and blogs, using a convenience sampling method. The study sample included both actual consumers of DTC genetic tests, and individuals who were considering purchasing a test, but had not yet done so (potential consumers).  This allowed a comparison between the two groups, who were asked questions about a variety of issues, including their information needs, information seeking behaviour and current health behaviour and health anxiety.

Results

A total of 275 usable responses were obtained. When respondents were asked where they had heard about DTC genetic tests, the most common response was blogs (35.7%), followed by friends, family or other social connection (29.2%) and the website of a company that sells the tests (24.4%).

The most common information that consumers wished to know before purchasing a test was to do with the coverage of the tests (32.3%), followed by details about the analyses used (21.1%), the cost of the tests (19.3%) and privacy issues (16.8%). The most common information source used by consumers was blogs (22.7%), followed by Google (18.0%), general non-specific websites (18.0%) and articles (11.0%). In total, 13.4% of consumers looked at no information source other than the websites of the companies that sell the tests, 40.1% looked at one other source and 27.3% looked at two other sources.  When information sources were divided into suitable and non-suitable sources (erring to the generous side in regards to suitable sources), 52.3% of consumers were found to have not used any suitable sources.

A change in health behavior was reported by 27.3% of consumers, all positive from the consumer’s point of view. Consumers were significantly more likely than potential consumers (p=0.022) to follow a combination of six common health behaviours (intake of salt, fat, fibre and fruit and vegetables, sufficient exercise and lack of smoking). A change in health anxiety was reported by 24.6% of consumers, with 85.3% of these changes a decrease in anxiety, 8.8% an increase and 5.9% a mixture of the two.  However, there was no significant difference between the current health anxiety levels of consumers and potential consumers.

Discussion

The results show a large variability in the information seeking behaviour of consumers of DTC genetic tests, with many reporting a detailed search for information and others only a little; although almost all participants looked for some information before purchasing a test.  If appropriate information was not provided on the company website, it is likely that those who did little searching were not exposed to it elsewhere, backing up the informational concerns raised in the literature and underlining the importance of good quality information provision by the companies.  Given the wide range of topics on which consumers wished information before purchasing, the importance of providing information on all aspects of the tests is also highlighted.

Despite concerns raised in the literature, no negative effects on health behaviour were identified in the study.  Also, although a survey of this type can never establish causation (and the convenience sampling method prevents generalisation to the population as a whole), the results appear to show a positive effect on health behaviour in a minority of consumers.  Taken together with the previously mentioned studies, these results should somewhat allay concerns about this aspect of the tests.  However, it should be noted that some respondents, if only a small minority, reported an increase in health anxiety.