In recent posts, I have been focusing on Ebola crisis as a case study about data.   The Ebola example provides a good example because so much information is only recently discovered and even more information is missing.  As the information emerges in popular news reporting media, there is an opportunity to observe patterns and to identify new areas for improving data collection or interpretation.   My focus was to look at the problem from a data perspective instead of a medical perspective.

Recently, I encountered an article on LinkedIn that described the challenges hospitals encounter in order to handle contagious diseases, where the biggest fear is an influenza outbreak.  Of the many messages in this article, there is one that I want to focus on, and that is the prospect of seeking alternatives to hospitals for treating highly contagious diseases.    He makes a good point that the biggest bottleneck for handing contagious epidemics may be the housekeeping that have to follow very rigorous procedures to prepare a room for a new patient.   Another point is that even in the case of influenza, the immunization shots may only be 50% effective so that half of the hospital staff treating such patients may contract the illness.   If the disease is as dangerous as the flu of 1918-1919 (unfairly named Spanish flu) that killed a significant number of patients who were previously very healthy, work in the hospital can become a dangerous occupation.

In this article, he suggests that a better alternative would be to keep the sick out of the hospitals and treat them at home in particular.  He alludes to the quote “A Hospital is no place to be sick.” by Samuel Goldwyn, but applies it to the specific case of contagious diseases.  The hospitals are best suited for treatable conditions like cancers, heart attacks, or strokes, and for addressing traumas.   These conditions have specific remedies that can be optimized in the hospital setting.   Also, these are not contagious although there is a need to protect from spreading coincidental diseases that these patients may have.

In contrast, contagious diseases like the flu and Ebola (that initially presents itself with flu symptoms), there really is no cure for the disease outside of managing the disease such as maintaining hydration while the body eventually cures itself.   The hospital really offers nothing more than isolation and comfort for patients, where these objectives are nearly as effectively available by keeping the patients in their homes.  We could effectively treat contagious diseases by building an infrastructure to make the home the first choice for treating flu symptoms.

Coincidentally, there was a recent demonstration by the disruptive transportation company Uber to deliver Flu shots at home. This service was a short trial and involved a simple one time visit to administer flu shots by certified health providers, but the service was delivered to individual’s homes.   This is an example that shows that there is at least a thought of using modern mobile-technologies to dispatch health providers to homes instead of requiring them to come to high-traffic health centers.   In this case, the immunization is low risk simple procedure, but it does demonstrate the concept of delivering health care to homes by using scheduling technologies to stage resources in places that are more likely to need it.

The LinkedIn article and the Uber experiment suggests the possibility of applying ride-sharing concepts to the management of actual flu cases (and perhaps even Ebola cases) at the patient’s home and thus keep these cases out of the hospital.   This medical house-call service would require better equipped mobile units to handle carrying blood samples, deliver substantial protective garments and hydration fluids to the house, and a staging area to don and doff PPE for the provider.  This will require some investment in specially equipped vehicles but it is easy to imagine that this concept can work with similar opportunities for optimization that Uber has shown with their ride-sharing service.

A flu outbreak will result in mobile calls for service (like a taxi calls).  Because the disease is contagious, there are likely to be hot spots in the city where the disease is more prevalent.  These hot spots can be readily identified using the call data to allow for optimal allocation of the vehicles and staff to be in the vicinity of the hot spots before the call for a new service arrives.   The health provider’s time can be optimized to keep specific house calls a short commute distance because they will be staged near where their services are needed.

In terms of labor optimization, such a mobile technology approach may not be as efficient as hospital settings where a single nurse can handle many more patients in an day because the patients are all nearby.

However, as noted in the LinkedIn article, these diseases are highly contagious.   Bringing these diseases into the hospital risks infecting other patients and increases the labor to keep things disinfected.  There is a risk that the contagious disease can infect the health provider staff.  When that happens, the whole hospital suffers directly from the lost labor and indirectly due to the fear that could cause some staff to stop showing up to work from justifiable fear of contracting the disease in the hospital.   That second problem of fear only has to start with the lowest-paid housekeeping staff to bring the hospital operations to a halt.

There is really nothing unique that hospitals can offer for treating Flu (and Ebola) that can not be provided in private homes.  For such diseases, the critical objectives are to keep the patient hydrated and comfortable and to isolate the patient to avoid the spread of the disease.   These objectives can be met at homes when the homes are provided adequate supplies for hydration or pain-relief and for protecting the rest of the family from infection (with gloves, masks, etc).  There would also have to be special transport for disposal of hazardous materials (such as used protective items).

Although the LinkedIn article stresses the challenges of handling contagious diseases in hospitals, there is the additional problem of contact tracing that becomes essential for managing diseases like Ebola (and perhaps some very dangerous strains of Flu like the 1918-1919 flu).   This contact tracing is complicated by the transport of a patient to the hospital and the initial processing in the emergency room full of others seeking care for non-flu symptoms.   This transport of a pre-diagnosed contagious-disease patient will likely occur when the patient is very contagious.  The contacts will include the transport staff and vehicle to get the patient to the emergency room, all the other patients in the emergency room (including those who arrive later but before the room is completely sanitized), and the staff in the emergency room who lack full protective gear.

From the perspective of optimizing contact tracing, it seems the best option is to keep the patient at home when they start to show flu symptoms.   The home can be quarantined to limit the spread of the disease.  Modern technologies and infrastructure exists to provide efficient home delivery of all relevant supplies include household goods such as groceries and drugstore needs.   On a per-patient basis, this home delivery is more costly than asking the patient to go to distribution centers like stores or clinics.    But on a social scale, such delivery will be more cost effective by limiting the spread of the disease to other healthy people and in particular to protect the limited number of healthcare providers.

A contagious disease is contagious.  Bringing such diseases to the hospital puts the health care providers at risk of catching this disease.  Even with the best protocols to prevent spread of protection, there is a high risk of spreading a disease within the hospital as illustrated with recent experience with healthcare associated MRSA that are firmly established within hospitals and other clinical settings.

The example trial by Uber demonstrates a willingness for private industry to explore possibilities for home health care.  For infectious diseases, there could be great economy for having these services for more extensive delivery of services for home treatment.

The first problem is that that economy only exists during an outbreak.   It is difficult to fund the service year-round when infectious diseases tend to be seasonal, and large outbreaks are very rare.   Unfortunately, an infectious outbreak can occur very quickly (within a few weeks of the first patient) so there is not enough time to start building up such a service after the outbreak is recognized.  There is a need to have the infrastructure in place before the outbreak occurs.

One approach would be to expand the service for more routine care so the service could be used year-round.   Routine health care visits (doctors appointments and annual physicals) also can be performed in homes.  The necessary equipment can be hand-carried by the physician making the call.   This is how medicine used to work with local family doctors making house calls and arriving with a doctor’s bag.   That doctor’s bag can be more sophisticated today and may involve a portable cart.   The health-delivery vehicle may have heavier equipment like EKG machines and ultrasound equipment as necessary.

If we can deliver all routine health care to the home, then we would have the infrastructure in place to provide at-home delivery of health care in the event of contagious disease outbreaks.  The cost advantage will be realized by quickly containing the disease to prevent its wider spread that otherwise would increase the cost to the community.

We once had an infrastructure for house call delivery of routine healthcare and at-home treatment of many diseases.  We dismantled these because of the cost-savings opportunities of concentrating the services in dense clinic settings like group doctor’s offices and hospitals.   We pay health providers by their time and we can optimize their time spent on individual patients by bringing the patients to them.  Over time, both public policies and health insurance coverage discouraged (or denied) access to home delivery of health services.

The old model for house-call medical services were simply too expensive.  A doctor’s day will involve visiting his patients according to which ones have made an appointment that day.  The doctor would need to travel between widely separated houses to accommodate specific scheduling needs for specific patients.  As a result, a doctor may only be able to see a couple patients per day and those patient’s would have to collectively cover the costs of that doctor for that day.   This is expensive.

Modern technologies of high-bandwidth mobile communications and big-data optimization of scheduling can offer new opportunities to make home-delivery of healthcare more cost-competitive.

The model introduced by ride-sharing applications uses big-data analytics predict where services are needed so that providers can position themselves close to where the next service will be needed.  This may work in health care with specially equipped vehicles and advanced nurses who can perform their tasks with immediate access to video conferencing with the physician (who may be in an office or in a different mobile unit delivering in-person examinations).

This model may required a cultural change for patients to accept doctors who service the area of their homes.  I suspect many patients will adapt to change in providers especially since the provider serves a particular geographically-defined community.  They will get access to the same provider (or at least the same pool of providers) for each house-call appointment.  Even if they get a visit from an unfamiliar provider, that provider can have quick mobile-communication access to the preferred provider as needed.

Some patients may continue to insist on specific doctors across town that will not be able to provide home delivery of service.  Those patients will need to continue to go to distant office.   The doctor’s office or clinic setting will continue to offer cost-savings advantage over the home-delivery option as long as we are not experiencing an outbreak of a contagious epidemic.   When there is a contagious disease outbreak, even these patients may be discouraged by the risk of visiting a waiting room with other patients who may include someone who is infected.

Despite the cost-savings of the central location for a provider, we should seek ways to discourage these setting in favor of directing people to obtain their health care at home.  Then we will have a system that will be much more robust when facing an epidemic and perhaps even other localized calamities such as terrorist attacks or natural disasters where minor traumas and burns may be treated at home.

In the long run, home-delivery of common health services may be very cost-effective.   We will still need hospitals for intensive care scenarios as well as treatments that require very heavy and expensive equipment that must be placed in a fixed facility.   Most of these scenarios do not involve infectious diseases, but instead traumas and chronic conditions such as cancers or heart disease.   By moving routine care and infectious-disease care to homes, we can optimize the hospitals for the services that are uniquely available in hospitals.  In particular, there can be less of a need to plan and to train for sudden overflow capacity due to infectious disease outbreak.  The planning an training will still occur but through the separately operating home-healthcare system.   This type of treatment does not need to be part of the hospital’s mission.

Also, I have little doubt that once such services start to be used, the industry will find ways to optimize the costs for home-delivery of services instead of fixed-site delivery.  The scheduling algorithms will improve to place the right resources near the current day’s patients and this will allow providers to see more patients in a day.   There will be new medical technologies scaled for consumer-level operation and affordability just as we today have consumer affordable technologies for taking blood pressure and blood glucose measurements.

There may be new consumer-level health equipment.  For example, some lab-tests may be reinvented as at-home test kits.  In contrast to high-quality versions that must be durable for long periods (and repeated cleanings) to support clinical settings, the consumer versions may be cheaper as a result of the expectation of single use or short term use by only a single patient.

It is impossible to predict what kinds of innovations will come when we redirect healthcare to a home-delivery model for routine healthcare (including treating infectious diseases).   Given recent advancements in technologies, I have a high confidence that these innovations will occur rapidly and cost-savings benefits will follow.  It may not be unreasonable to expect that eventually such at home health care will become cheaper to deliver routine health-care at home instead of in clinics.  We have to start when it is clearly not cheaper.

Unfortunately, we may never have the opportunity to move toward a house call model because it is effectively outlawed by government and not covered by health insurance (themselves constrained by laws).   Any initial experiments with disruptive house-call medical care will have to be paid out of pocket by the individual, if the service is legally permitted at all.  The above mentioned demonstration by Uber may be one of the few areas where home care is permitted: the low risk administration of flu vaccines.   I can imagine something as simple as collecting blood, urine, or stool samples at home will be legally impractical due to requirements for transporting potential bio-hazardous material.

To get started with re-introducing house-call option for health-care delivery, we need to change laws governing health care and allow health insurance to cover house call medical services.   Here I am pessimistic about the political possibility of making this happen.  Even if we can start to discuss the option, the initial period will involve more costly health care by not-yet optimized house-call medical services.

The population has a deeply held sense of a right to their choice of doctors and hospitals (even very distant ones), and they have a right to receiving care in clinical settings.   The population perceives some curative advantage of such settings that go beyond the medical services provided.   The physical space itself within a hospital has some inherent curative magic despite clear evidence of the exact opposite presented by the HA-MRSA infections.

It has been over 100 years since the catastrophic influenza outbreak of 1918-1919.  There is a wide expectation that we are due for another outbreak of common flu with equal infectiousness and lethal consequences.   100 years ago, we had a model where most health care, particularly treatment of flu, was delivered in homes.  The infection did spread within the household but this can be better managed with today’s more affordable access to disinfectants and protective equipment.

Today, we do not have the option of home health care with house call visits by health providers.  Today, we treat flu by transporting the patient to the emergency room or health clinic for examination and monitoring.   A repeat of the 1918-1919 today will probably be far worse for us as a consequence.  The hospitals will be quickly overwhelmed, the disease will result in loss of health providers catching the disease, and the disease will spread to others sharing the transportation and the waiting rooms.

The hospitals are not the right place to treat such infections.

Perhaps recent events of the Ebola crisis provide an opportunity to open a debate about alternatives to hospitals for treating contagious diseases.   Already with the few Ebola cases in US, we observe that the treatment of the disease puts health-care providers at risk not just of catching the disease but of lengthy quarantines that will prevent them doing their jobs.   We also observe the challenges of contact-tracing involving just the initial transport and admission of the patient to the hospital, putting at risk ambulance, ambulance providers, and everyone in emergency rooms.   Even when the risk for catching the disease is low, there is a high risk of contract-tracing putting someone under quarantine orders that could cause major disruptions in lives.

The threat of quarantine, voluntary isolation, or even the stigma of being identified as possible contact could encourage a debate to consider some changes in approach for handling infectious diseases.  We should consider options other than immediate transport to emergency rooms.   One option may be to revive the abandoned house-call medical service with our access to better technologies that can make this more affordable.  The house-call model can serve all routine medical attention and preserve hospitals for treating only those conditions requiring expensive and fixed assets.   Treatment of contagious diseases do not require that kind of equipment.