[Peripherals] Regarding the TeleHealth Module

[Benjamin M. Schwartz]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

DyDisMe asked me to share my opinion of the TeleHealth Module.

In short, I feel that the proposed TeleHealth peripheral is useless, but the
Health Info Database is very promising.

Regarding the hardware:
The proposed hardware, as I understand it, is an infrared LED and photocell,
designed to be clipped onto the patient's finger (or perhaps toe, for children).
 This setup is designed to provide information on pulse, blood oximetry, and
(speculatively) other blood chemistry information, like blood sugar.  There is
also the possibility of providing EKG functionality.

I grew up in the US, and like most US children I was sent to a pediatrician as a
child for regular checkups.  The doctor did a number of things: he used a
lighted magnifying glass to peer into my ears and mouth, with a tongue
depressor; he held a stethoscope to my chest and asked me to cough; he tested my
reflexes with a rubber mallet; he used a sphygmomanometer (pressure cuff) to
measure my blood pressure; he tested my mouth temperature; he checked my hair
for lice; he tested my hearing and eyesight; he checked for hernia, or any
externally visible problems; he drew my blood and sent it to the lab for
testing; and perhaps most importantly, he interviewed me about how I felt.  Each
of these tests is a screen for specific ailments common in children.

He did not check my blood oxygen concentration, or perform an EKG.  Blood oxygen
is only measured in hospitals, in order to ensure that unconscious patients are
breathing properly.  EKG is sometimes performed on healthy older adults, but
generally restricted to those with a known heart condition.  Neither is an
appropriate basic health screen for children.

The situation is even more dramatic in poor areas, where children typically
suffer from one of a handful of diseases like bacterial diarrhea, malaria, HIV,
tuberculosis, or simple malnutrition.  These diseases are easily diagnosed by a
simple interview with the patient (and possibly a parent), and perhaps physical
inspection.  The TeleHealth peripheral is entirely irrelevant to these diseases,
even more than it would be in a first-world setting.  Essentially none of the
elements of a standard checkup can be delivered by telemetry.  With some
difficulty, it might be possible to make a working remote stethoscope and
pressure cuff, but I think this would be difficult, expensive, and still not
especially useful.  Remote thermometers could work, but would probably be more
expensive and cumbersome than the usual thermometers, which do not require an
expert's help to use.

The idea of a low-cost USB EKG has merit, but not in the context of OLPC.  It is
worth remembering that acceptable EKG results require special conductive gel and
probably electrode stickers, as well as an experienced technician, possibly
armed with a razor and shaving cream.

Regarding the software:
Reliable self-diagnosis is not achievable.  The variety of diseases is so vast,
and medicine is so poorly understood, that even the world's best doctors,
equipped with the world's best equipment, cannot always get it right.  Medicine
is _not_ a science.  However, self-diagnosis can still be useful.  A carefully
designed, simple questionnaire might be very helpful just by letting people know
what they _may_ have.

One of the strongest constraints on self-diagnosis is hypochondria.  People are
easily convinced that they have some disease; it is a well-known phenomenon that
upon reading about a condition, many people will begin to suspect that they have
it.  Hospitals are routinely flooded with patients whenever the local news media
reports on a disease.  Any self-diagnosis engine must come with some very
clearly worded disclaimers.

What I find most promising is the idea of a simple digital medicine textbook.
This needn't be much more than the slice of wikipedia devoted to diseases and
drugs.  Ideally, it should be extensively indexed and searchable by prevalence,
symptoms, and similar criteria.  It must be written in clear language that
anyone can understand.  Self-diagnosis could then be nothing more than an
interface for sequentially refined search.

This has been called the TeleHealth Module Database, but that name suggests the
wrong purpose to me.  It would be most valuable as a digital textbook.  Basic
medical education is universal, and universally valuable.  A simple medical
reference would be greatly beneficial, both for the children and their parents.
 By placing such a reference into the base distribution, we would make it
omnipresent in the lives of our users. It would be a fundamental element of the
computer, to be called upon at any time, whenever a health question arises.
- From age 10 onwards, it would be an excellent base for classes, or just for
reading by medically curious kids.  The potential benefit to public health is
enormous.

The most valuable contribution we can make for TeleHealth is videoconferencing.
 Allowing a doctor to speak with a patient is extremely valuable, and is
sufficient to diagnose many basic diseases.  This only makes sense in cases
where the patient is far from the nearest doctor, but both have internet access.
 I do not expect that to be a common situation.  However, in those rare cases,
we can make a big difference.  In that case, the primary hurdle is
organizational.  Someone must locate doctors in each country, provide them with
appropriate videoconferencing hardware (presumably an XO), and convince them to
conduct these interviews for free.  There must also be a system for scheduling
appointments.

- --Ben Schwartz
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHQ7VMUJT6e6HFtqQRAp7CAKCictHDlW90PYOY6VNklPm2sLk4fgCfSXmg
B0kRDv7nuyQXxsrmxyks7mE=
=JkSq
-----END PGP SIGNATURE-----