I. INTERNAL VIABILITY

Transcription

1 Page 1 of 8 Project Viability Belinda Chiang, Jonathan Leland, Daniel Wiesenthal, Sun Ying Designing Liberation Technologies Spring 2011

2 Page 2 of 8 I. INTERNAL VIABILITY A. PARTNER RELATIONSHIP: CDC & TABITHA CLINIC Based on our conversations with John Neatherlin of the International Emerging Infections Program at the Center for Disease Control in Kenya, diarrhea is the leading cause of death for children under the age of five in Kibera. Around the world, dehydration caused by diarrhea kills more children than AIDS, malaria and measles combined. 1 Yet it has the simple, cheap, and extremely effective 2 solution of Oral Rehydration Therapy. A concerted international effort to eradicate deaths from diarrhea was launched in the late eighties and early nineties, but afterwards, it was simply dropped as a priority for the public health community. As a result, awareness by mothers about the danger of diarrhea and effective treatments has slid back in places like Kibera. However, the CDC and international community are once again now turning their attention to this massive and incredibly preventable killer of children providing an opportune moment for them to partner with ENDesha. ENDesha was designed with substantial assistance and feedback by the CDC, from needfinding to project development. The CDCʼs primary mission in Kibera is to collect data on the disease burden of the population in Kibera, specifically diarrhea as well as pneumonia, fever, and jaundice. 3 ENDesha can contribute to that mission by providing valuable data that the CDC would not otherwise be able to capture. In particular, ENDesha would allow the CDC to better track the incidence, frequency, and severity of diarrheal disease. It would also encourage more of the users to make a clinic visit such that Tabitha (and thus the CDC) can collect greater surveillance data. As John Neatherlin wrote in an providing feedback on our idea, From CDCʼs disease surveillance perspective, we would want to capture these cases of diarrheal disease and in the case of dehydration we would want to get a stool specimen to determine the cause. Because of the CDCʼs engagement with the catchment population around Tabitha, we would benefit from partnering with the CDC to launch the service and get an initial user base. We would need to be careful, however, to avoid the same perception problem facing Tabitha at the moment, however that this is a service only available to individuals in the catchment population. Consequently, it may be wise to include a small non-catchment population in the pilot program as well. Since the only patient revenue that Tabitha 1 Oral Rehydration Salts. Salts. 2 84% of the children [at Kenyatta National Hospital] were successfully treated with ORS alone regardless of etiological agent found. 2 This means that the vast majority of the time, ORS is a sufficient treatment for diarrhea. Source: Kinoti, et al. Management of acute childhood diarrhoea with oral rehydration therapy at Kenyatta National Hospital, Nairobi, Kenya. East Afr Med J Jan;62(1): To collect this information, the CDC has partnered with NGO Carolina for Kiberaʼs Tabitha Clinic to run an intensive population-based surveillance program of 27,000 individuals (the catchment population ) surrounding the Tabitha Clinic. In return, the CDC provides the catchment population cost-free care at Tabitha. Consequently, almost all of Tabithaʼs patients come from the catchment.

3 Page 3 of 8 Clinic receives is the one to two individuals from outside the catchment population who come in per day, 4 it would be additional benefit to both the CDC and Tabitha to partner with ENDesha. In addition to the CDC, Tabitha Clinic would also be a critical partner for ENDesha. Since the solution is primarily targeted at usersʼ needs during clinic afterhours, Tabitha is the main destination that users are directed to for follow-up diagnosis and treatment once it opens. Tabitha would be a valuable partner in helping to market the solution and treat patients referred by the system. In return, Tabitha will receive a greater inflow of patients who require medical attention, leading to higher and more efficient utilization of its resources. It will receive data from all hotline calls, which helps with community surveillance and patient tracking. Because Tabitha would be the first partner during the project development phase, Tabithaʼs commitment to the project would likely be substantial. We would need to develop our system to allow Tabithaʼs computers to access any data we collect from the calls, potentially rely on Tabitha workers for our call operators, find workable incentive mechanisms for non-catchment population patients, and develop a sustained dialogue about whatʼs working with the program and what is not. Additional partnerships with other hospitals and clinics will also be helpful, as it gives user more options to access care if and when they need it. B. COST OF SOLUTION The following is a preliminary budget to set up and operate ENDesha for the catchment area. Setup primarily requires engineers to develop the technology solution and a partner to test it on-the-ground and provide feedback for any necessary redesign. Technology development will be the bulk of the cost for the pilot. We budgeted $15,000 for the most expensive scenario of hiring qualified engineers at market rate. We will likely be able to hire engineers at discounted rates or to even recruit volunteers, which would bring the cost down significantly. The program would cover the entire CDC catchment area of ~4,500 households, and will require a part-time program manager and a part-time trained nurse to be on-call. Set-up cost Annual operational cost Usage Number of households Number of calls Number of calls needing nurse on-call Number of calls leading to clinic visits Technology development 2 engineers for 5 weeks $ 15,000 N/A Marketing 1 set of print marketing material per household $ 44 $ 1,969 Cost of service 3 ORS sachets/household $ 20 $ SMS per household $ 3 $ 113 Regular call 5 min voice, 3 SMS $ 17 $ 3,038 4 Each visit yields only around Ksh 200 in revenue. 5 Realistic estimate based on the statistic that every Kenyan child under five has an average of three episodes of diarrhea annually (2008 Demographic and Health Survey).

4 Nurse on-call ENDesha Medical Hotline Page 4 of FTE staffed to receive calls $ 5,000 Nurse on-call 5 min voice $ 1 $ 127 G&A: General & Administrative 0.5 FTE Program Manager/Admin $ 5,000 Overhead (food, IT, transportation, etc.) $ 2,000 $ 2,000 TOTAL $ 17,084 $ 18,145 C. SKILLS & PERSONNEL REQUIRED Our triage and assessment algorithms were developed by referencing resources from the WHO and in consulting with a medical expert. Going forward, the program will still require medical counsel, accessible from Stanford, Tabitha Clinic and the CDC. Development of the technology would require one or two skilled computer programmers. We can either hire talent at market rates, or potentially recruit volunteers over one to two months over the summer to complete the coding and testing. We estimate that the cost of this development will be $15,000. Operating ENDesha would require a part-time public health program manager, a part-time trained medical professional (to be on-call), and skills for marketing and outreach in Kibera. There may be some ability to recruit these employees from the CDC/Carolina for Kibera (CFK) community, but it is more likely that we would have to hire personnel from the community or local universities. Finally, design, management, and diplomacy skills are necessary in the implementation of this idea, especially during the pilot phase. For ENDesha to go from course concept to full-fledged hotline, a team must be able to successfully manage its internal communications and processes, improve the design and prototype it in the field, and negotiate the cooperation of organizations in Nairobi. D. TECHNOLOGY ENDesha Medical Hotline leverages basic mobile technologies that are straightforward to develop and implement, including designing the service to be free for the user, follow-up SMS messages after initial contact, and rerouting of calls to either Tabitha Clinic during operating hours or CDC-employed medicallytrained individuals who are on-call if the caller submits a severe case. All of these technologies can be easily automated. Providing a free service for the user is possible via flashing. Flashing is a common practice in East Africa, and it is analogous to calling collect in the U.S. The implementation is different, however: rather than calling a special number, the caller flashes the recipient, which means to call the recipientʼs phone number directly and end the call immediately (after one ring) before the recipient has the chance to pick up. This missed call shows up on the recipientʼs phone and is understood to mean that they should return the call at their own expense. We would like to use this common practice to our advantage in order to provide our users a free service. They would be able to flash a central hotline number, which would log the incoming call number. This central hotline would never pick up a call, but would rather wait for the caller to hang up and then immediately call them back. This is easy to implement from a

5 Page 5 of 8 technological standpoint as it essentially involves only two steps: logging an incoming call number, and generating an outgoing call in response. In this way, we can provide a free service to the end users. In addition, if we partner with the CDC to launch the service, we will want to contribute our data into the CDC surveillance database. Determining how this integration would work would require further discussion and cooperation with the CDC. In the case of severe diarrhea, we would also provide notification of patient cases to medical providers. Because every calls coded as severe would be routed to a live operator, it is not necessary to develop a technological solution for determining the closest hospital to the patient and providing notification. Instead, we can simply have the operator talk to the caller to determine their location and closest available medical center. The operator can then notify that hospital or clinic using some kind of program or service. While notification isnʼt central to the hotline, it can provide benefits for both the caretakers (knowing the hospital is expecting you may increase the odds that you will visit) and for hospitals (increasing treatment efficiencies). Developing this kind of relationship will require low-level partnering with hospitals they would just need to express the desire for such notification and provide an avenue for text or voice communication of incoming patients. Another feature to include would be personalized interactions based on patient information logged during the course of interactions with the system. For example, when a user requests instruction on how to prepare ORS, the application would access the age and symptoms of the child previously entered (during the initial call) to provide, for example, more appropriate dosage instructions, which vary depending on age (notably for infants). This can be accomplished by leveraging caller identification technology. II. CUSTOMER ACQUISITION Our target users are mothers of infants or young children with diarrhea, who would access our service through the family or neighborʼs mobile phone. Adoption happens when our service is known and available, and when mothers whose children actually have diarrheal symptoms believe we can help their children. To encourage initial adoption, we will leverage existing outreach channels of the CDC (Tabitha, Community Interviewers, and Community Health Workers) to build awareness of the availability and benefits of our service, as well as explore other channels of communication (public areas, mobile phones). Our primary marketing mechanism, which would help ensure that caretakers have access to ORS, would be to distribute ORS packets with an ENDesha Medical Hotline sticker attached. This would ensure quick recall and access to the phone number when itʼs needed so that mothers would have access to ORS even when chemists and clinics are closed. We can work with UNICEF, 6 which distributes ORS to clinics in Kenya, to receive a large and inexpensive supply. One potential challenge to customer acquisition would be a lack of access to mobile phones. However, our needfinding suggests that almost every household has access to a mobile phone. While the men may carry the phone with them during the day, they are more available at night and on weekends, when the service is the most valuable as the clinics are closed. 6 UNICEF is the worldʼs largest provider of Oral Rehydration Salts. Source:

6 Page 6 of 8 Cost can also be a factor prohibiting adoption. To minimize the cost of using mobile phones, users can flash (calling and hanging up after one ring) the hotline number to indicate their need to use our service. They will receive a returning call almost immediately and be guided through an automated voice service. We believe that mothers would dial the ENDesha Medical Hotline given the success of other medical hotlines operational both within Kenya and in other developing countries. The Médecins Sans Frontières branch in Olympia, Kibera operates a Rape Crisis Center as well as 24/7 rape hotline ( ). During our needfinding, we found widespread knowledge of the existence of both the center and the hotline. Most people did not know the actual hotline number, however, so we aim to ensure that the number is accessible via stickers on ORS packets that will be distributed. Launched in July 2009, the Freedom from Fistula Foundation operates a 24/7 hotline in Kenya that has enjoyed tremendous success. Women can call anytime if they suspect they have a fistula. The caller will be asked a few questions and will then be given a date to come to our partner hospital for screening. If a fistula is discovered, they will get another date to return for their operation. According to their press release, In the first two weeks of operating the HOTLINE, 17 women have been referred to Jamaa Hospital and 10 of those women have already received their operations. 7 Other medical hotlines that have been successful in other developing countries include Healthline in Bangladesh, TeleDoc in Pakistan, MedicallHome in Mexico. III. GOVERNMENT The Kenyan government has been increasingly aware of the potential uses of information and communication technologies (ICTs) to support development efforts. Both the National Poverty Reduction Strategy and the National Development Plan have chapters devoted to the use of ICTs to tackle poverty and development, respectively. The 2001 Poverty Reduction Strategy Paper by the Ministry of Finance and Planning says that the develop[ment] of a regulatory/legislative regime that fosters the growth of...[and] formulates incentives for the IT industry is a priority activity. The Kenyan government has also renewed its fight against diarrheal disease. It resolved to Reduce by one-half deaths due to diarrhoea among children below five years by 2013 and to Reduce by two-thirds the mortality rate among children below five years by 2015 compared to This was articulated in Policy Guidelines on Control and Management of Diarrhoeal Diseases in Children Below Five Years in Kenya, issued by the Division of Child and Adolescent Health within the Ministry of Public Health and Sanitation. At the March 2010 launch in Nairobi, the Minister for Public Health and Sanitation, Beth Mugo, said, With 86 children dying every day, diarrhea is the third leading cause of death among under-fives in Kenya. It is unacceptable, but we can stop this! Given this rise in interest and funding to combat diarrheal disease, we believe that the government would be willing to give its support and funding to distribute ORS packets and to create an ICT4D mobile program that combats diarrheal disease. 7 Source: 17 Women Treated Since Launching of Kenya Hotline.

7 Page 7 of 8 IV. COMPETITION While our partners on the ground would be more knowledgeable about programs and services combating childhood diarrheal diseases in Kibera, we have been unable to discover organizations that focused on the particularly pressing and prevalent issue of diarrheal diseases through our needfinding. Neither have we been able to find any organizations within Kibera. The clinics and some chemists have supplies of Oral Rehydration Salts, but simple availability has not lead to widespread adoption. After a big push (and much success) to stop child deaths from diarrheal disease in the late 1980ʼs and 1990ʼs, diarrheal disease largely disappeared from as a priority of the international health community. Governments and health organizations seem to be turning back to the issue, recognizing it as one of the largest killers of children in the developing world but also one of the easiest to combat. There have, however, been great strides made in rural, western Kenya. There, PATH is an organization that has spearheaded much of the work treating and preventing diarrheal diseases. It has been able to establish 22 Oral Rehydration Treatment Corners and trained over 200 health care workers on the protocol for administering ORS and zinc treatment. PATH is working closely with the Kenyan government to expand their program to other parts of the country. There is reason to believe that the PATH model may not immediately translate to the Kiberan context, since it was developed in rural communities where there is not the same access to self-medicating pharmacies and the culture around clinic use may be substantially different. V. OTHER PARTNERS AND STAKEHOLDERS Partnership with a mobile operator is not critical but would be beneficial. A mobile operator may provide funding or in-kind donations (handsets, unlimited free SMS and voice) as part of its CSR activity and/or strategic investment in mhealth. If we can convince one or more carriers of the potential market for mobile health services and see this as a low-cost pilot program, we may be able to find a partner. However, we should not assume that we will be able to arrange such a partnership. The mobile companies are large profit-centered companies that are probably pursued as partners by almost every mobile ICT development project. Until we can build a compelling case of why we represent an opportunity to develop a market for their products, we are unlikely to get much traction.

8 Page 8 of 8 Appendix: Dialogue Path and Medical Algorithm Step 1: Screen Initial screen to make sure the person is calling about a case of childhood diarrhea Step 2: Triage Quick check for symptoms that definitely require hospitalization: the child (1) is unconscious or losing consciousness; (2) has not urinated today; or (3) has blood in his or her stool Yes: Live Operator The call is immediately directed to a live operator who will recommend immediate ORS treatment and hospitalization. No: Voice Automated Medical Assessment Algorithm based on 5 data points >2 days x Fever defer x >2 days Fever Vomit Low UOP Worse Vomit defer hospital x Low UOP defer defer defer x Worse defer hospital hospital defer x >2 days Fever Vomit Low UOP Worse Fever, >2 x x hospital defer hospital Vomit, >2 x hospital x defer defer Low UOP, >2 x defer defer x hospital Worse, >2 x hospital defer hospital x Low UOP, Fever defer x hospital x hospital Low UOP, Vomit defer hospital x x hospital Worse, Low UOP defer hospital hospital x x >2 days: has diarrhea for more than 2 days Fever: Has fever Vomit: Has been vomiting Low UOP: Low Urine Output Worse: Child is getting worse If immediate hospitalization is not necessary: Recommend ORS for patient and educate about ORS. Recommend clinic visit when the clinic opens.