Saving lives, one text message at a time

“The idea of the Real-Time Bio-Surveillance Program (RTBP) was to streamline the current paper-based system in India and Sri Lanka – which was established in the Nineteenth Century – by using mobile phones to submit patient data from rural communities to epidemiological centres for automated data mining.” – Nuwan Waidyanatha, Project Director at LIRNEasia

Researchers in Sri Lanka and India are developing a system that can help save lives by quickly identifying potential outbreaks of H1N1, chickenpox or malaria.

This may sound like science fiction, but such an early warning system is being piloted in India and Sri Lanka by a partnership that includes IDRC, LIRNEasia and Sarvodaya.

The Real-Time Bio-Surveillance system is a modern alternative to the paper-based process introduced by the British in 1897. Infectious disease information currently takes three to four weeks to be processed using the old routine. In 2003, the Sri Lanka’s Central Province was faced with a fever-like disease – the disease went unnoticed until it claimed three lives.

Before the pilot system was introduced, a survey conducted by the technological think tank LIRNEasia, highlighted that many healthcare workers only learned of outbreaks via the media, “word of mouth” or from peers.

“We need to be able to view cases in real-time for detecting outbreaks swiftly, which otherwise would take several days before the hospitals send the notification paper forms, by which time the patient may be dead or discharged,” stated one inspector.

Streamlining surveillance

Nuwan Waidyanatha, a LIRNEasia Project Director, stated that mobile phones are the perfect medium for streamlining the current process.

“Mobile phones are the most affordable technology with the widest reach in India and Sri Lanka. Data is submitted instantaneously compared with the current ‘snail mail’ system, and is also much cheaper. A text message costs two cents, whereas sending forms via regular mail costs 50 cents,” he said.

The concept is simple, but the technology behind this new real-time system is cutting edge.

Community healthcare workers record patient’s symptoms using software installed on their mobile phone, and submit the data directly to the national epidemiology centres in Colombo and Chennai. This data is analyzed daily by data mining software, allowing epidemiologists to visualize potential epidemics via mapping tools.

The system can return text messages to the healthcare workers, alerting them to potential dangers. These messages can be translated into the local dialect, relayed to the communities, and placed on bulletin boards in the village centres.

The new system has been tested for the past year in 28 facilities in the southern state of Tamil Nadu in India, and 12 hospitals in the northwestern province of Wayamba in Sri Lanka.

Greater efficiency and cost-saving

The system has already proved instrumental by identifying a recent chickenpox outbreak in Kurunegala district in Sri Lanka.

“The platform was able to detect this outbreak much faster than the paper-based system, which would have taken 3-4 days for the public health inspector to discover. In this case, the inspector found out about this the next day,” said Waidyanatha.

On average health inspectors are located 40km from the regional epidemiology centre in the cities and approximately 10-15km from their respective divisional centres.

A November LIRNEasia report indicated that the new system can reduce operational and archiving costs by 30-50%. The new process also means clerks no longer need to decipher the handwriting on the paper forms, nor does patient information have to be manually copied into logs, eliminating the risk of clerical error.

By limiting potential outbreaks, the new platform lessens the financial strain on the Indian and Sri Lankan health systems. The new system no longer requires nurses to travel to town once a week to compile their data; thus eliminating the need for travel subsidies.

The new system can be used to even identify everyday issues in local communities. Waidyanatha stated it had “identified that men were complaining of pain during harvest season – this highlighted how farmers just needed better tools.”

Challenges ahead

However, November’s report also underlined challenges still facing the new system in India and Sri Lanka.

There was no problem training the healthcare workers to enter the patient data, the quality of that digital data became an issue. Analysis showed that submission error rates fluctuated between 23-45%. These error rates affected the complex statistical analysis that the system performs, resulting in many false predictions.

Some errors were attributed to different spellings of medical terms such as tuberculosis. Other errors were caused by the misuse of synonyms like ‘dementia’ and ‘memory loss’, as well as workers mistaking treatment for diagnosis like ‘remove catheter’ or ‘vaccination’.

Some healthcare workers in India saw the new platform as a bureaucratic hindrance, whereas others feared the system would take away their jobs.

A sustainable solution?

Waidyanatha calls the new system a “useable solution”, but one that requires further enhancement.

However, the future of the life-saving project looks promising.

The Sri Lankan Ministry of Health plans to continue to operate the system and expand it to 50 hospitals in a second district called Puttalam, also in the province of Wayamba. A telecommunications company has even donated 50 mobile phones to the new system. The health ministry in India has yet to expand the system into further states, but an extensive media campaign targeting papers like the Hindu Times has stoked government interest.