Jenna C. Lester Why skin disease is often misdiagnosed in darker skin tones TED

The skin is the human body’s largest organ

and one of the most powerful
predictors of our health.

This is erythema migrans,
a hallmark feature of Lyme disease,

a tick-borne illness
present in over 80 countries

and estimated to affect 476,000 people
in the United States each year.

Dermatologists like me
are doctors of the skin

trained to diagnose
and treat skin disease.

And this is how we’re trained
to see erythema migrans,

as a bullseye-shaped rash
that ranges from red to pink.

But this is not at all
what it looks like in dark skin.

As you can see here,

there are hues of violet, of magenta,
and even dark brown.

If we were to rely
only on dermatology textbooks

to teach us how to identify skin disease,

we would frequently
misdiagnose it in patients of color.

And this is a huge problem
because Lyme disease needs to be treated.

Left untreated, Lyme disease
has significant health ramifications

including arthritis and even nerve damage.

And what’s more,

as we’ve seen an increase
in the incidence of Lyme disease,

a phenomenon attributed
in part to climate change,

as we continue to see and experience
the effects of climate change,

we may see more people
infected with Lyme disease,

making it even more important
that we’re able to accurately diagnose it.

Now, this story of erythema migrans
is emblematic of a larger issue.

In the United States, 47 percent
of graduating dermatology residents

report feeling uncomfortable diagnosing
skin disease in patients with dark skin.

47 percent. I just want that
to sink in for a second.

This is a staggering statistic,

and this means that the people
who have just undergone

their most intensive training
to become doctors of the skin

don’t feel comfortable
diagnosing and treating all patients.

And even so, they graduate from residency,

and they’re eligible to become
board-certified dermatologists,

qualified to care for all people.

Now, I wonder, could this be
why we still see and experience

health care disparities
in all aspects of medicine,

including dermatology?

I believe there’s a connection
between the fact

that almost half of dermatology residents

feel uncomfortable diagnosing
and treating certain patients

and the poorer health outcomes
of those same patients.

I speak to patients of color all the time

who express an awareness of the fact

that their dermatologist is unfamiliar

with diagnosing skin disease
in their skin tone

or uncomfortable teaching them
how to care for their hair or scalp.

And I wonder, what does this awareness
that your doctor is uncomfortable with you

due to the physician-patient relationship;

to trust in the medical establishment;

or to the likelihood that someone
returns for additional care?

A problem in dermatology is

that we’re not taught how skin disease
appears in all skin tones.

As a medical student,
my classmates and I quickly realized

that we only saw dark skin
when we were learning about syphilis.

And this observation
is supported by research that I published

in the British Journal
of Dermatology in 2019

that shows an overrepresentation
of dark skin in chapters

focused on sexually
transmitted infections,

even while those same skin tones
are underrepresented elsewhere

in the same textbook.

What does this do
to impressionable learners?

Does it make them think
that someone with dark skin

is more likely to have
a sexually transmitted infection?

Now, I know some of you may be thinking,

I know an algorithm that can solve this
or machine learning to the rescue.

And I’m here to gently disagree.

And that’s because the data
from which these algorithms learn

are the same photos
that overrepresent dark skin

in certain skin conditions,

even while underrepresenting
them in others.

In other words, these algorithms
will be as biased as we are

unless we make significant change.

I started the Skin of Color program

at the University
of California, San Francisco,

where I work with
medical students and residents

in an effort to begin to help them unlearn
some of these harmful patterns

that make it easier to see some things,

like dark skin with syphilis,

and harder to see others,
like dark skin with erythema migrans.

I teach everything from how to identify
inflammation in dark skin

to how to talk to a Black woman
about her hair care practices.

And one important fact
that I always make sure to mention

is that it’s neither good nor common

for Black women
to wash their hair every day.

And any treatment regimen

focused on taking care
of the hair and scalp

should reflect
this important understanding.

My work at the Skin of Color program,

as well as the work
of similar programs across the country,

demonstrate the importance of creating
a dedicated educational environment

for residents and medical students
to learn the full spectrum of skin disease

as they appear in all patients,
regardless of skin tone.

This is an important first step

on a long road towards eliminating
health care disparities in dermatology.

But let’s commit
to taking this journey together.

Thank you.

(Applause)

皮肤是人体最大的器官

,也是我们健康最有力的
预测指标之一。

这是游走性红斑,
莱姆病的标志性特征,

这种蜱传疾病
存在于 80 多个国家

,估计
每年在美国影响 476,000 人。

像我
这样的皮肤科医生是

接受过
皮肤病诊断和治疗培训的皮肤科医生。

这就是我们被
训练看到迁移性红斑的方式,它


一种从红色到粉红色的牛眼形皮疹。

但这根本不是
深色皮肤中的样子。

正如你在这里看到的,

有紫罗兰色、洋红色,
甚至是深棕色。

如果我们
仅仅依靠皮肤科教科书

来教我们如何识别皮肤病,

我们经常会
误诊为有色人种。

这是一个巨大的问题,
因为莱姆病需要治疗。

如果不及时治疗,莱姆病
会对健康产生重大影响,

包括关节炎甚至神经损伤。

更重要的是,

随着我们看到
莱姆病的发病率有所增加,

这种现象
部分归因于气候变化,

随着我们继续看到和体验
气候变化的影响,

我们可能会看到更多的人
感染莱姆病,

使我们能够准确诊断它变得更加重要。

现在,这个关于红斑迁移的故事
是一个更大问题的象征。

在美国,47%
的即将毕业的皮肤科住院医师

报告说,
在为深色皮肤的患者诊断皮肤病时感到不舒服。

47%。 我只是想让
它沉入一秒钟。

这是一个惊人的统计数据

,这意味着
刚刚接受

了最密集
培训成为皮肤医生的人在

诊断和治疗所有患者时感觉不舒服。

即便如此,他们从住院医师毕业

,他们有资格成为
董事会认证的皮肤科医生,

有资格照顾所有人。

现在,我想知道,这
就是为什么我们仍然

在医学的各个方面(

包括皮肤病学)看到和经历医疗保健差异的原因吗?

我相信

,几乎一半的皮肤科住院医师在

诊断
和治疗某些患者时感到不舒服

,这与这些患者的健康状况较差有关。

我一直与有色人种患者交谈,

他们表示

意识到他们的皮肤科医生不

熟悉根据
他们的肤色诊断皮肤病,

或者对教
他们如何护理头发或头皮感到不舒服。

而且我想知道,由于医患关系
,您的医生对您感到不舒服是什么意识

信任医疗机构;

或有人
返回接受额外护理的可能性?

皮肤病学的一个问题是

,我们没有被告知皮肤病是如何
出现在所有肤色中的。

作为一名医科学生,
我和我的同学很快意识到

,我们在学习梅毒时只看到了黑皮肤

这一
观察得到了我 2019 年在英国皮肤病学杂志上发表的研究的支持,该研究表明,

在关注性传播感染的章节中,深色皮肤的比例过高

即使在同一本教科书的其他地方,同样的肤色
也没有得到充分体现


对易受影响的学习者有什么影响?

这是否让他们
认为皮肤黝黑

的人更容易
感染性传播疾病?

现在,我知道你们中的一些人可能在想,

我知道一种可以解决这个问题的算法
或机器学习来救援。

我来这里是为了温和地不同意。

这是因为
这些算法从中学习的数据

是相同的照片,这些照片

在某些皮肤条件下过度代表了深色皮肤,即使在其他皮肤条件下

代表
不足。

换句话说,除非我们做出重大改变,否则这些算法
将和我们一样有偏见


加州大学旧金山分校开始了肤色皮肤计划,

在那里我与
医学生和住院医师

一起努力开始帮助他们忘记
一些有害的模式

,使他们更容易看到一些东西,

比如深色皮肤 患有梅毒

,更难看到其他人,
如皮肤黝黑伴游走性红斑。

我教授一切,从如何识别
深色皮肤的炎症

到如何与黑人女性
谈论她的头发护理实践。

我总是要提到的一个重要事实

是,

黑人女性
每天洗头既不好也不常见。

任何

专注于
头发和头皮护理的治疗方案

都应该反映
这一重要认识。

我在有色皮肤计划


工作,以及全国各地类似计划的工作,

证明了

为居民和医学生
创造一个专门的教育环境的重要性,让他们了解

所有患者身上出现的各种皮肤病,
无论肤色如何。

这是

消除
皮肤科医疗保健差异的漫长道路上重要的第一步。

但让我们承诺
一起踏上这段旅程。

谢谢你。

(掌声)