mHealth for Developing Countries

Where there are no doctors, the proliferation of mobile technology offers to alleviate to some degree the deficit of expertise.

As the unit price of Android phones plummets, we can, for the first time, put decision support systems in the hands of willing but poorly-trained health workers.

We can help them do their job more thoroughly and more reliably, with better health outcomes for their clients.

no There are no doctors

In substantial parts of the world, particularly rural parts, there are, for all practical purposes, no doctors.

But the potential is greater still: use of decision support apps generates data as a by-product which gets synchroized to a server. We can then present aggregations of these data and provide vital information enabling supervisors and health system managers to make informed decisions.

It is important to note that the technology we are developing and deploying therefore are to be regarded as systems as a whole, of which the apps in use by the health worker is just one (very important) part.

In substantial parts of the world, the first line health worker is a Community Health Workermore_vert
CHWclose

In a remote village in rural Malawi, a Community Health Worker, using an interactive advice-giving application on a low-cost smartphone, gently examines a sick baby to determine whether he can provide some basic medicine and advice to the mother, or whether the child is so ill that mother needs to take the child to the health centre.

He enters information about the symptoms he observes, along with what the mother tells him about the child, and the application advises about the child's condition, and the most appropriate treatment.

We are privileged to work with D-tree International and other organisations in this complex and rewarding field.

 

dtree tz
D-tree Tanzania
dtree mw
D-tree Malawi
malawi

Care Algorithms for the Developing World

The problem we are addressing is one of higher mortality rates concentrated in poor countries.

These higher rates of mortality are identified as being primarily due not, for example, to outbreaks of Ebola, but to a systematic failure to identify and manage problems which are usually straightforward to treat, such as diahrroea.

The major causes of deaths among children under 5 are pneumonia (18%), diarrhoea (15%), malaria (8%) and neonatal infections, including sepsis (6%).

Each of these killers has, however, at least one proven effective treatment: artemisinin-based combination therapy (ACT) for malaria; low-osmolarity oral rehydration solutions (ORS) and zinc for diarrhoea; and antibiotics for pneumonia and neonatal infections.

If coverage of these interventions were universal, with community-based delivery of half of the interventions, it has been estimated that the annual number of deaths among children under 5 would fall by 63%.

Hence, the medical algorithms we need to put in the hands of health workers are specific decision-procedures which deal with these issues.

Though there are a few standardized algorithms published by the WHO, implementation of any of these is always subject to modification due to country context.

D-tree International does a lot of work in looking at national guidelines, for example for ante-natal care, and formalizing algorithms which express their logic.

IMCI

Integrated Management of Childhood Illnesses

Well before the emergence of mobile digital technology, WHO's Integrated Management of Chilhood Illnesses (IMCI) strategy was developed with an algorithm at its heart.

ETAT

Emergency Triage and Treatment

ETAT is a simplified triage algorithm developed originally for use in hospitals, which however has been shown to be useful also in busy clinics.

CCM

Community Case Management

CCM is effectively a simplified version of IMCI for use in the Community: an algorithm for determining the condition of a sick child, including treatment recommendations or referral.

malawi

Triaging in a busy clinic: ETAT in Action

Action Meningitis, D-tree International and Things Prime work together on a triaging system with the ETAT algorithm at its heart

This video illustrates this project in action in a busy clinic. A child is examined by a Community Health Worker interacting with a decision support app and is found to have worrying symptoms, making it a priority case. The CHW brings the mother and child forward in the queue to see the clinician.

malawi

If there is no ambulance ..

Exploiting community resources for Emergency Transport

Innovative use of technology can help in surprising ways. In a project funded by Vodafone Foundation, D-tree International and Things Prime, using mangologic, implemented a call-centre application which despatchers use to determine the severity of an obstetric emergency.

If needed, and if (as is often the case) no ambulance is available, the despatcher may be prompted to call drivers in the locality (stored in the app database) to find a car to take the woman to hospital - an action which may well save her life.

Once the despatcher has confirmed that the journey has been completed, and the mangologic app has synchronized with the server, the driver is automatically paid an agreed fee via the Mpesa API

Vodafone Foundation present an overview of the project in this video:

dtree mw
A typical month: the dispatchers are handling more than 7 emergencies per day.

A presentation explaining more detail concerning the project (MMH = Mobilising Maternal Health):

malawi

Immunization Campaigns need software

These are two tables published by the WHO covering routine immunizations for children
The "Routine" table is complicated, although with training and practice perhaps not so difficult to understand and apply.
However, this table alone is inadequate in situations where immunizations for any particular child are delayed or interrupted, and in the contexts in which we work such gaps in a child's immunization record are not at all uncommon.
Therefore, we need also to take into account the second "Delayed" recommendations table.
But factor in the following: a health worker has to check for invalid doses in the vaccination record - doses which do not accord with recommendations (e.g. too early after a previous dose) - such doses have to be ignored for the purposes of calculating what is due today.
All told, determining what is due to a child today is a very challenging cognitive task, particularly in the kind of high-volume, low-resourced situations which are the contexts in which we work.
The likelihood of a health worker being able to avoid mistakes when applying this complex combinations of rules in such a situation must be regarded as very low. It is the kind of task which cries out for the application of software.
For this reason MSF Suisse Innovation Unit is currently introducing mangologic within a vaccination campaign in the Central African Republic. mangologic can be configured with the rules of an immunization schedule and take the large cognitive load away from health workers.

Prescripteurs can easily and accurately determine what immunizations to give each specific child.