Race and Health Data: Too Much Information?

Patient Medical Record Icon. Flat Design.

Channel News Asia and the Institute of Policy Studies have released the results of a survey concerning race relations in Singapore.  The survey has been getting significant coverage and contains a number of interesting findings, but I’d like to focus on one particularly relevant to biomedical ethics that hasn’t been highlighted in the press: race and public information.  The survey found that, in most areas (crime, education, social problems and successes), Singaporeans are uncomfortable with “race-based public information”; the one exception was race in health information, which a (bare) majority of respondents were favourable towards.  What to make of this difference?

The data

You can see the relevant findings on Slide 19 of the CNA-IPS study slides.  Unfortunately, the precise question that was asked is not reproduced; instead, it is summarized as whether respondents “favour race based public information” in several areas.

Category of Information Yes No
The type of crime that different races are more likely to commit 30% 70%
Educational performance of the different races 35% 65%
The social problems of different races such as drug abuse, gambling, alcoholism 38% 62%
The successes of the different races 45% 55%
Health issues that affect the different races 54% 46%


Immediately, one notices that the responses are on a spectrum – from racial crime data that is very unpopular, up to health data which receives significantly more support.  But no category was wildly popular.  This is in line with other findings of the survey, particularly on the question of discussions of race: 66% responded that talking about race causes unnecessary tensions.

Explaining the difference

It makes sense that respondents would identify racial health information as the most acceptable type of racial data.  Each of the other categories can reinforce pernicious racial stereotypes: beliefs that some races are more dangerous, less trustworthy, less business-savvy, and so on.  This could lead to greater discrimination, as people feel justified (based on national data) in treating some races differently than others.  Indeed, the non-health categories focus on factors for which we typically form moral judgments, particularly blame and praise.  We condemn criminals, blame failing students for not studying hard enough and criticize drug addicts, while conversely heaping praise on those who do well in their studies and careers.  Tying race to those normative categories risks transference and the formation of moral judgments on individuals based on their race.

By contrast, propensity towards certain health conditions is not the sort of thing that would ground day-to-day discrimination.  Health information will instead primarily be used by physicians and public health officials to tailor interventions to certain patients.  In addition, falling sick does not typically give rise to a negative moral judgment.  External factors, rather than personal failings, are more typically to blame. There is thus less risk of stigmatizing certain races due to propensity towards certain diseases.

Remaining problems

We should recognize (as 46% of respondents probably did) that racial health information is not perfectly innocuous.  It can in fact be a source of discrimination – particularly by insurers and employers.  If a certain race has greater predisposition towards certain diseases, insurers may charge greater premiums for members of that race to cover the greater expected costs.  Similarly, employers may be more reluctant to employ people of a particular race due to the possible need to take more sick days.  In addition, while normative judgments are weaker for health information, some stigmatization remains.  Obesity, in particular, is often stigmatized as the result of being too lazy to exercise and too gluttonous to eat healthy.

Does this mean public agencies should not report race data for any of the areas listed, including health data?   Here, I would respectfully disagree with most of the respondents.  Identifying issues is a necessary step for those issues to be resolved.  If significant racial differences are found in, say, cholesterol levels, there may be a need for new policies or social movements to combat the disparities.  This requires both careful study of the problems, and mindful implementation of promising solutions (whether by lawmakers or community leaders).

One could counter that researchers and policymakers should have access to the racial data, but not the general public.  The researchers and policymakers could craft solutions to whatever problems were revealed by the data, while the public would be shielded from the pernicious effects on discrimination and stereotyping.  But there are at least two problems with this approach.  One, public engagement and deliberation can be an important source of policy changes or social improvements.  This is the idea behind the Our Singapore Conversation initiative launched in 2012.  Two, the shielded data would only prompt government bodies to address the problems – community organizations, patient advocates and other private bodies, which may have effective solutions to racial disparities, would be left out.

Interestingly, the CNA-IPS survey itself (and, for that matter, this blog post), is an instance of public information about and discussion of race that respondents objected to.  We will have to see whether the results do in fact stoke racial tensions, or instead prompt thoughtful dialogue and addressing of the issues raised.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s