Designing for a Country-Specific Context

One challenge for our design is to determine a potential country for the competition. Following the advice of experts and established best practices, we recognize there is no ‘one-size-fits-all’ approach to designing frontline health worker systems. There are a variety of models, each for a different setting, and these design decisions need to be made collaboratively with a partner country’s Ministry of Health (or equivalent department).

We are currently approaching this with a quantitative and qualitative analytical approach to determine the optimal location. We would like to hear from you on your experiences:

    What have been your experiences (positive or negative) piloting digital health solutions in specific countries? What characteristics should be considered when identifying potential countries for the competition?
  • More specifically, where would you suggest is an ideal location for our competition?

Hi @pglass, @biki, @ajchenx, @Davisthedoc, @praveenraja - You might have thoughts on this discussion. Join the discussion.

Our work in China has taught us developing country like China can have very different healthcare system than USA. Particularly, financial incentives can be so different that an effective digital health solution that works here in USA will not get any traction in a developing country. But, we also confirm that the unmet medical needs are generally universal. So, my strategic is to design a digital health system that can solve common healthcare problems and at the same time easily can be adopted for the different environments in different countries. The core of the solution should be applicable in very countries in order to achieve large-scale international impact. This strategy may be valid for this xprize design as well. It is not designed for country-specific-context, rather designed for frontline healthcare anywhere with ability to adapt for country-specific context.

@ajchenx - Thanks AJ for these thoughts. These are all great points to consider. Tagging my teammates @TerryMulligan, @SevagKechichian and @HeatherSutton in case they have any thought or questions for you.

Hello @bngejane, @JohnParrishSprowl, @ymedan, @mhackett, @CHardaker, @Lizzy_2020, @Nvargas2, @preciouslunga, @anaconnav, @marschenrj - Here is our new topic on country-specific prize design; what do you think? Any items that jump out at you as key things to explore?

I believe India with a vast population, youth, geo-political importance, ease of obtaining IRB and field testing should be a site. However, there are huge disparities within the country as shown in our work with surgical unmet need. As you can see below, we have carefully defined the households for our sampling and data analysis. We published the first study arising from this proposal which consisted of urban slum, who were predominantly day laborers. Eventually, we hope to establish a digital platform to educate (courses for patients and community workers), ensure adherence with follow up and analyse data in real time.

http://rdcu.be/b3q47/10.1007/s00268-020-05502-5

Briefly, in our study of surgical unmet need, our aims are following:

  1. Surgical Volume
    a. Percentage of women and men who have undergone a surgical operation in the past.
    b. Percent distribution of women and men who have undergone a surgical operation in the past and by type of operation.

  2. Surgical Need
    a. Percentage of women and men who have been told by a doctor or another healthcare worker that they might need (another) operation.

  3. Surgical Unmet Need
    a. Percentage of women and men who have been told by a doctor or another healthcare worker that they might need an operation and were not able to access it?
    b. Percent distribution of women and men who were not able to access an operation by reason for no access.

  4. To assess the gaps in availability of surgical care at resource constraint health facilities.

Methods: There will be 3 cohorts, which have been identified and several international projects are being implemented under the auspices of the Indian Institute of Public Health, Gandhinagar, India (http://iiphg.edu.in/). This community-based panel study is nested within a larger project studying institutional childbirth in India. In Gujarat, the project was implemented in three purposely selected districts (Dahod, Sabarkantha, and Surendranagar) . The districts were selected because they cover diverse geographic and socioeconomic areas of the state . House to house detailed survey by case worker or surgical resident who has basic understanding of surgical conditions. Case worker will be trained in using the survey instruments and assessing surgical conditions. Secondary examination and confirmation by a surgeon in the local tertiary care Hospital.

Setting: Gujarat state (population of 60.4 million, 57.4% rural, 79% literate and 31.6% below poverty line, 21% tribal) is located in western India . The state is divided into 33 administrative districts, each with a population of between 1-2 million. Districts are further divided into 10–20 blocks (sub-districts) of approximately 100,000 to 200,000 people. These districts have varying human development indices and different population compositions. The population is divided into socio-economic sub groups. Government of India uses the terms ‘Schedule Tribe (ST)’ to denote these traditionally marginalized populations.

Sampling procedure: In brief, in each of the 3 districts, 3–5 blocks will be selected purposely. A list of all the villages in these blocks have already been compiled using the criteria: village population more than 1,000 and less than 2,500, greater than 40% below poverty level (BPL) population and scattered all over the block. From the list, 142 villages have been selected randomly to cover approximately 300,000 populations in 20,000 households. This selection is from the population of 6.25 million people living in the 3 districts of Gujarat described above. The sampling methodology of MATIND survey has been described elsewhere .

Analysis: Data will be double-entered and validated in REDCap. The database for this study will be imported to EpiData version v2.2.2.183 for descriptive and bivariate inferential analysis (EpiData Association, Odense, Denmark). STATA (version 12.1, copyright 1985–2011 StataCorp LP USA, serial number: 30120504773) was used for regression (Poisson) analysis (enter method). Frequencies and proportions (categorical variable) will be used to summarize utilization of surgical services. The median and interquartile ranges (IQR) will be used to summarize surgical volume, surgical need and unmet surgical need. Unadjusted analyses will be performed to assess the association (relative risk, RR) of factors with demographics and surgical volume, surgical need and unmet surgical need. Variables with p-value of <0.2 in the unadjusted analysis were included in the regression model. Adjusted RRs will be reported with 95% confidence intervals (CI).

We chose Germany for our trial. The health system was advanced so we needed to show improved efficacy to get a hearing. The medical insurance environment is not so brutal as the US and there are “wellness” incentive that encourage public health insurance to push for better outcomes. One of the biggest incentives was the absolute transparency which mercantile we could get both outcome and financial data that was complete and accurate. This allows you to model for other country’s with only a few variables (assuming a lack of corruption)

I am not sure if I understand the last question but some ideas come to mind. Let me rephrase the question just to make sure that we are on the same page. What place or country would be an ideal location for a project-based competition with a digital health component for improving frontline health worker systems?

In low- and middle-income countries, many public health programs and political measures aimed to improve the general population’s health are based on successful programs that are already implemented and documented in other countries. Although, as @TerryMulligan mentioned above, there is no “one-size-fits-all”. In Mexico, just last week, the state legislature of Oaxaca voted a new policy that prohibits selling sugar-sweetened beverages (SSBs) to children, a behavioral measure that places SSBs in the same category as tobacco and alcohol. The rise in diabetes mellitus and obesity since the 1990s explains and justifies the measure. Nearly one in ten adults have diabetes. In addition, because Oaxaca is located in the region with the highest poverty level, the prevalence of food insecurity is high. In the capital city, Oaxaca de Juárez, you can see young mothers bottle-feeding SSBs to infants, a cheap and quick energy source, instead of milk. The question here is, What needs to change?

To find out if a state-wide policy will be successful by first implementing it, without prior testing is bold. Therefore, based on my experiences, I would suggest to promote the use of in silico models, like Agent-Based Modeling. In these computational environments, a modeler can set specifications on the population’s characteristics and behaviors, as well as characteristics and expected behaviors of environmental variables. This computational tool is powerful and useful to observe what works and what does not and to find out what to expect under the proposed, pre-specified conditions before implementing the program in a real-life setting.

There are a limited number of research groups and scholarly institutions with expertise in agent-based computational modeling. The Santa Fe Institute in New Mexico through their e-learning environment “Complexity Explorer” offers an online course on ABM that is helpful as an introduction for beginners. It would be interesting to see more of these educational groups and environments available in LMICs working on better frontline worker systems.

India is a good place to start, given the characteristics of its healthcare system.
Israel is another good option with totally different ecosystem characteristics.

incidendtly, both were freed from British occupation in the same year (1948).

Thank you @RahulJindal, @CHardaker, @anaconnav and @ymedan for all the valuable insights! These are all key to consider and actually align perfectly with this prize design. We will begin unpacking these.

Hi @joshnesbit, @SArora, @rajpanda, @a1m2r3h4, @addy_kulkarni, @DrAhimsa, @Mellie64 - Curious if you might have any input, And let us know if you have any questions.

In general, when sociologists are planning a study and are faced with a diverse population it seems best to choose not one case but at least a few quite different ones using theoretical sampling. Looking over the many insightful comments offered by others on this topic I am impressed that India must be included if for no other reason than a highly sophisticated group of researchers have already begun working there and have what appears to be a good plan for moving forward. But the external validity of a study even of a country as big and diverse culturally linguistically etc. as India is not clear. So, a small country might be quite useful and if the Peace Corps experience (from what I have read and heard) with Botswana is any guide, I would nominate that nation as a possible second site. A third country I would pick based solely on my reading would be the United Kingdom. Its economy to a great extent is a service economy (more so than other developed European nations) where personal relationships are important to do business. It has a very difficult situation right now with COVID-19 in part because of that service economy model. So I think you have with these three countries the necessary diversity, the necessary likelihood that they would benefit from the assistance English speaking professionals can offer, and for varying reasons might be the easiest to gain the necessary cooperation of the official health agencies such as the ministries of health.

Thanks @boblf029 for sharing your perspective.
@shamakarkal, @dzera, @Lizzy_2020, @elekaja, @PHall and @Neal_Lesh - What would be your ideal country to test an ehealth system? Any items that jump out at you as key things to explore?

Thanks @Shashi! I would echo the suggestion that limiting it to one country may be insufficient. Countries (whether India or Mexico or China) are diverse and testing it at a country level would be limiting unless there is a functional ehealth system at a national level. I’d recommend multiple-sites. Countries such as Nigeria, Botswana, Zambia along with India (the opportunity to integrate is significant) where parts of the health system (community interface, supply chain etc) are using technology tools should be considered. Also important to define 3-4 key necessary elements for testing - such as government partnership, multiple solutions which work across the system and the outcome measure.

Hello @Shashi, @TerryMulligan and the rest of the colleagues.

Just like in any field, it’s important in frontline health to learn from positive outliers: developing countries and systems that have outperformed their peers. I would like to put the question back to our team, which would make the best setting, the outliers or those outperformed?

To rephrase the question:
Would the frontline technology prove more valuable in an emerging country that has performed well in public health in the past few years, which means the Government, NGO’s and other stakeholders have the willingness and ability to see greater change?

Or being an XPRIZE you want to opt for an emerging country that is dismally failing to implement its public health programs and could do with a lot of help? So if you choose this option you might have to address other stakeholder issues outside of the competition process itself.

This makes a huge difference in how you design the prize, as it determines the level of difficulty, as not all developing countries are equal.

Happy to contribute once we have made a call on this.

Regards

Thanks @shamakarkal and @bngejane for sharing your insights.
@bngejane - What do you think, which type of emerging countries we should be looking at for this competition. It would be nice to hear few countries you feel appropriate for this competition and why.

Hi @Debbie_Rogers @dykki, @jenyxp, @saebipour - What are your thoughts on an ideal country to test an eHealth solution.