Health post: Vision tests & witches afoot

Peek Vision is a health startup aimed at improving access to vision services and eye care. Their main product is the Peek Acuity mHealth solution, a smartphone app that allows anyone to conduct a vision screening in a few minutes. They have a few other products as well, all of which contribute towards their goal to perform vision screenings (particularly for schoolchildren) as well as make a real impact by providing eye care and/or glasses for those who need them.

Peek was founded by a London-based PhD candidate, piloted in Kenya, and has had a chapter in Botswana for a couple years (here is a great TED Talk by Peek’s founder). Last year, Peek partnered with the Botswana government to perform screenings in 49 schools, rural and urban, in the country’s Good Hope district.

I interviewed Maipelo, the project manager of Peek Botswana, to learn more about the screenings. She traveled to many of the schools involved throughout the screening process and personally helped train local healthcare workers so that they could use the app.

A typical visual acuity “tumbling E” board.

Since the app is free, I downloaded it myself. The app acts as a replacement for the “tumbling E” boards typically used in visual acuity tests – children are supposed to tell screeners which way the “E” is pointing (for example, an “E” in the usual orientation is pointing to the right; a backwards “E” points to the left). The typical boards can get lost or damaged, and the pattern of Es can be memorized by children (a sequence of up, right, down, etc). The Peek app addresses those problems while also keeping track of anyone who fails the test for follow-up purposes.

IMG_8867 (1).jpg
Maipelo with the Peek Acuity app.

When you first open the app, it brings you through a tutorial to show how the screening should go. The screener needs to stand exactly two meters from the student (or whoever will be screened), holding the phone so that the screen faces the student at eye level.

My favorite part about the Peek Acuity app is how the actual screening goes – the screener never needs to look at the app while the student is watching the screen. When an E is displayed on the screen, the student points in the direction of the E. The screener then swipes the phone screen in the direction that the student is pointing and never needs to look at the E. The screener doesn’t need to know if the student gave the correct answer; it is automatically recorded by the app. The Es displayed on the screen continue to change direction and size, adjusting to the student’s performance. If the student can’t see the E well enough to guess, the screener is supposed to shake the phone so that a new, slightly larger E appears.

After about two minutes, the phone plays a sound to indicate the end of the screening. The screener then looks at the phone and sees the result (for example, “0.8” for a student with quite poor vision). There’s also a built-in simulator that displays how blurry a chalkboard would look to someone with 0.8 vision, for example, so that the screener truly understands the numerical result. The simulator feature also ideally builds empathy for students who have had undetected vision impairments – students who struggle in school and often get written off as being lazy or naughty by teachers who assume that they can see perfectly fine. (This is true for hearing as well. The HearScreen people in Pretoria described hearing problems as a “silent epidemic” because kids with such impairments often go undetected and are treated like bad students when they don’t do well in school).

A screenshot from the app showing the vision simulation feature.

Maipelo told me that, for the most part, the screeners and the students responded well to the Peek screening. Everyone is excited when they see the app, she said; less so when they are told to use it and realize they have work to do. Regardless of how fast and easy the screening process is, it’s still work, especially when screeners work all day long checking hundreds of schoolchildren. Also, Maipelo said, those who were less comfortable with the phones would take longer to input data. Even if the difference is a minute and a half instead of, say, 45 seconds, that adds up with so many screenings per day – and it can get frustrating for the less tech-savvy screeners.

I also asked Maipelo about the follow-up process. When Peek Acuity indicates that a child has impaired vision, the app prompts the screener to enter their contact information. The app then automatically texts the child’s parents with the follow-up details – where they should go to meet with an eye doctor and when. That’s when the children would get glasses if they needed them.

One of the Peek Botswana employees demonstrates a screening with the Peek Acuity app.

That is where it could get complicated, Maipelo told me. Even though all the parents had a positive reaction to the idea of medical technology, she said, they never liked to hear that their kids had an impairment and needed a follow-up. People only question the technology after it illustrates a problem, she said. Even if the app just says that their child needs glasses, parents immediately respond negatively to anything they interpret as a “medical issue.” Maipelo said that some people believe such problems are curses or bewitchments. “Bewitchments?” I echoed. Yes, she said, people grow up hearing about witches.

This isn’t the first time I’ve heard about witches in Botswana. It seems to be a traditional idea that witches are afoot, causing problems or punishing people for various reasons in various ways. I think when there is a lack of awareness about these things – not knowing how common and remediable vision impairments are, for example – all medical problems could seem as serious as a witches’ curse.

Another local later told me that some people in Botswana have the misconception that glasses will actually worsen vision. If a well-sighted person looks through someone else’s prescription glasses, of course the view is distorted; this apparently leads some well-sighted people to believe that glasses are harmful. Also, people with glasses never stop needing glasses, needing stronger prescriptions as time goes on. Both glasses and crutches are medical devices, but crutches help you get to a point where you don’t need crutches any more; glasses stay forever. Apparently this, too, contributes to the misconception that glasses degrade vision. Of course, most people in Botswana do know that glasses help, but of course it would be best if everyone (especially the more skeptical parents) were on board.

Another interviewee phrased it like this: “In our culture, everything should be normal.” Everything should fit the status quo. People don’t accept the abnormal; they say it’s the work of witches, he said. (And there they are again). Unfortunately many impairments, including poor vision, aren’t normalized, so everything (even the need for glasses) gets labeled as “abnormal.” I’ve heard this in general, too – many people have told me that fitting in and maintaining the status quo is very important in Botswana, which I think makes sense with the neighborhood lifestyle here. In terms of medical problems, it all boils down to awareness and the importance of normalization. If more people wore glasses and it was seen as normal, there would be less stigma against vision impairments, and it would be easier to convince people to treat vision problems less like serious, scary medical issues.

I’ve really enjoyed getting to know Peek Vision throughout my time in Botswana. Including my interview with Maipelo, I’ve had many interactions with Peek – I’ve talked to people involved in different aspects of the company; I sat in on a government meeting where Peek pitched a budget to the Ministry of Health for a potential national rollout; and I’ve met health workers who participated in Peek screenings in very rural areas. When I started my project, almost all of my meetings were one-offs. I had hourlong chats about many different devices and technologies, definitely seeing more breadth than depth. There haven’t been so many examples of medical technology to explore in Botswana, so I’ve tried to dig deeper into the examples that are here, and it’s been cool getting to see Peek Vision from different sides. These diverse vantage points have also illustrated different challenges of getting an mHealth project underway in Botswana – such as how important hierarchy and social niceties are when dealing with government officials in the capital city, or how screeners in rural areas don’t think about how easy or difficult the app is to use if they’re not getting paid to do the screenings. I’m really grateful to Peek Vision for all that they’ve shown me here in Botswana.

This is Peek’s hardware product, Peek Retina. It wasn’t part of the school screenings, so it’s hard to talk about user responses, but I think it’s very cool. It’s a small device that can fitted over a smartphone camera for retinal screening, which can detect diabetic retinopathy and other issues.
I had my pupil dilated to be the guinea pig in a hands-on Peek Retina demonstration. Here, someone is trying to screen my retina with the Peek device and a smartphone, with an optometrist looking on.
Version 2
Always a fan of cool hardware!





As the drizzle started to fall on me in Pretoria, I thought about how neither I nor the rain was supposed to be there. I had ten days left on the Watson (six, now), and I had decided to go to Pretoria, South Africa, to meet a company there for my project. It’s winter in South Africa at this time of year, and in the northeast, where the capital of Pretoria is, that means dry season; rain is only supposed to fall there in the summer.

This is at a day care center in Mamelodi, ZA. I’m standing here with Charles, who helped HearX organize screenings for the children.

I wasn’t supposed to be in Pretoria because South Africa is not one of my Watson project countries. Beyond that, I’m technically not supposed to go there because I’ve already spent so time in South Africa, having studied abroad in Cape Town for 5 months my junior year of college. But Pretoria is on the other side of the country, far closer to Gaborone than to Cape Town, and I figured it would be worth breaking the rules for just a few days to see something relevant to my project (especially since I’ve nearly exhausted my project opportunities in Botswana by this point).

The Voortrekker Monument of Pretoria. The monument and enclosed museum commemorate the Voortrekkers, pastoralists who traveled across South Africa in the “Great Trek” of the 19th century.
At the Voortrekker monument.

I arrived back in Gaborone last night after another 6-hour bus ride across the Botswana-South African border. Earlier in the Watson, I would have asked for permission ahead of time for this short weekend transgression. As I was visiting a monument in Pretoria enjoying the rain, weather I hadn’t felt in a long time, I realized that I had reached a new level of confidence – the confidence to make that judgement call and know, on my own, that it was still within the spirit of the Watson and still good for my project to break the rules just a little bit – a level of confidence that I could only have now, at the end of the Watson. You can only properly bend the rules once you’ve lived within them and respected their existence.

On the steps of the Voortrekker monument.
Looking down from the top floor of the monument.

Of course I have been making my own decisions all year, but always within the bounds of what had already been approved for me – going to Pretoria was a decision that I made on my own basis of what was appropriate, confident that it would be worth it. I used to think “confidence” was simply being comfortable in yourself and your abilities. But that sort of confidence is so easily confused with arrogance. There’s a deeper confidence, I’ve found, that lies within the humble acceptance that you’re making it up as you go, that there is a lot to learn, and that you can still deal with everything in life anyway. The confidence of knowing yourself and having that be enough – not needing anyone or anything else to move forward. The confidence to be able to talk to anyone and not be better than anyone else.

I wasn’t too interested by the museum in the Voortrekker monument, but I loved the architecture of the building and all these vantage points that led to geometric views.

Anyway, before this gets any sappier, I’m glad I went. My project contacts in Gaborone were the ones to suggest the trip to meet with HearX, an e-health start-up that spun out of the University of Pretoria. HearX’s main product is HearScreen, a mobile health solution that facilitates simple hearing screenings. With the HearScreen app and approved headphones, the screener plays 3 different tones in each of the listener’s ears. The listener is supposed to raise a hand when they hear a sound, and the screener notes whether or not the listener responds to all the tones played. At the end of the two-minute screening, the app alerts the screener if the listener has a hearing issue and needs to be referred to an audiologist. The audiologist can then determine why the listener failed the screening (HearX told me that the most common cause is wax blockage, a simple problem to fix) and if they need to go to the next step, such as receiving a hearing aid.

Lelanie (left) and Charles (right) at the Mamelodi day care center. Charles is holding the HearX case, which includes everything needed for a screening – mainly a smartphone with the HearScreen app along with the specific headphones.

I met the HearX people at the Innovation Hub, a set of offices for start-ups in Pretoria. From there, I went with Lelanie, a social worker at HearX, to Mamelodi, a nearby township. That’s where we visited the day care center and met with Charles, a local contact who has helped HearX do school screenings for children in the area. Charles brought in a young boy to show us how the screening worked, and he explained everything to the boy in his local language. I find that these “local ambassadors” are often key for encouraging the adoption and use of m-health and e-health products; Charles is clearly great with kids and made an effort to make the little boy feel comfortable. Lelanie also told me that the kids get more excited about the hearing screening when the screeners tell them that they have to wear the big headphones “like a DJ.”

Charles had me act as the screener for this trial run. The app was really easy to use, although I think I went through the screening a bit too quickly! 

I sat behind the kid we were screening so that he wouldn’t be influenced by my actions. Lelanie and Charles told me that when the HearScreen project started, they realized that kids could just watch the screeners using the app, raising their hands when they saw the screeners tapping the phone – anticipating the tone rather than actually responding to it. Otherwise, they haven’t had any issues. HearX is planning to expand into Botswana, which I think would be great. The main challenge there, as I’ve mentioned earlier, is that they’ll have to integrate with the Botswana government to an extent that they don’t have to with the South African government.

IMG_8417 (1).jpg
The city of Pretoria from afar.

Health post: Starting with startups

This past week, I spoke with a man named Shoibal about the general state of the medtech industry in India. He studied to be a doctor but hasn’t practiced medicine, instead going straight into the pharmaceutical industry – he currently works at an Indian pharmaceutical company here in Mumbai. For a while, he was also a consultant helping healthcare startups and small companies with the projects they wanted to do.

I asked him if the startups and small companies he worked with tended towards a main topic, and he said that most of them focused on smartphone apps that aim to simplify the doctor or the patient experience. There is a need for such apps, he said, because India lacks a universal EMR system (electronic medical records). Often hospitals have no EMR system at all, so patients go to a hospital and get records created on paper. As a result, the patient has no idea what’s in their record, and they might know about parts of their record but not all, and they have a hard time going to different doctors and hospitals. A lot of these companies, therefore, focus on providing electronic data capture for patients so that they can have some control over their records and not worry about forgetting important data. This also gives patients the flexibility to move between cities and share their records with new doctors.

Shoibal mentioned that in addition to EMR-focused apps, companies were working on fitness apps, which provide support and recommendations for maintaining good health – either for an already healthy person or someone with a specific condition. These are apps that are tailored to the user, giving detailed feedback based on input data. I asked Shoibal if these companies had specific target user groups. As I mentioned in my last health post, I think that India’s large and diverse population necessitates the strict definition of intended user populations; no company, especially a start-up, can have a set of offerings wide enough to suit everyone. Shoibal replied that the companies have had to focus on urban populations since they are developing smartphone apps, and the penetration of smartphones in India is mostly in urban areas (though they are slowly spreading into rural areas, and surely the companies ultimately want to widen their focus). Also, the apps are aimed towards educated people, as the uneducated population might not even be literate, said Shoibal.

Finally, there is the issue of language. It’s very difficult to translate an app into multiple different local languages, so most of the companies make apps in English only. Even Welcome Cure, I noticed, had all of its information in English but not in Hindi, let alone any regional languages. So that means that these startups and small companies all have nearly identical target user groups: the educated, English-speaking people living in major cities who own a smartphone. In other words, the 1%. I see a potential for image-based apps here; even if they still require a phone, at least they would work for the uneducated and non-English-speaking members of the population.

I asked if any of the companies were making physical devices rather than just health apps (at the end of the day, I’m an engineer, not a computer scientist). Shoibal said that some were focused on connecting people to devices, but still via smartphone apps. Rather than these startups building and deploying their own glucose monitors, for example, they would have a licensed connection with another company (say a Chinese manufacturer) and sell that company’s device online. Shoibal pointed out that these were two different skill sets, engineering, manufacturing, and quality control vs. app development and IT, and only large companies with enough money keep both under the same roof.

He said that some of the companies that didn’t do apps would maybe do a website, such as Welcome Cure, and have a telephone number, focusing more on having some supportive back-end of people who communicate with the patients. These could be services that ease patient management for doctors, or help patients find the cheapest version of their prescribed medicine, or encourage medication adherence, for example.

Next, I asked one of my favorite big questions: what factors influenced the success of some of these medtech companies over others? Shoibal was ready with two main factors of success, the first of which is funding. Since there are so many tech companies and health startups (and it really does seem like India has a lot), “the ultimate success of these ventures is uncertain.” Even if they have a lot of success in an initial domain, such as childcare, it can be difficult to scale up the company and expand into other domains without a lot of upfront capital in addition to the money the company’s already made.

The second main factor of success is that companies have to have a good plan. They need to offer something that is actually valuable to their target group – something that the patients or the doctors need. The plan has to incorporate continuity as well, meaning that the company has to work to keep engaging its users and clients beyond the point of initial interest.

I wondered if Shoibal had any ideas of what new startups should focus on, or if there was some big need that they should address. He said that there was a lot of potential in the patient-doctor relationship – not that there was anything wrong with the relationship, but that both patients and doctors could use more support in terms of adherence, setting up meetings, keeping track of medical records, and so on. The healthcare space is quite “fragmented and disorganized in India,” said Shoibal. “Organizing it through technology really has potential.” With such a large population, it’s difficult to find doctors and the right kind of help, and it’s a big hassle to go to an overcrowded hospital just for a small health issue (I can imagine that this leads to many people avoiding care until their issue gets to be more painful or urgent, which is obviously bad for health). Shoibal also mentioned that there’s potential in rural areas, providing care by translating existing services, apps, and websites into the local languages there.

There’s clearly a large range of health issues in India (due to the large population and high socioeconomic inequality, I would say). Shoibal suggested that Indians might be a little more careless than the rest of the world when it comes to health, but even if they do want to be very careful about health, Indians are exposed to a particularly unhealthy environment. There is high pollution, the presence of environmental and biological pathogens, high levels of stress on the roads (Indian traffic!), and compounding all that, the healthcare delivery system is not well-equipped enough even to deal with normal levels of health issues. There’s a gap between what the people need and the healthcare that is available. Part of it is simply a numbers issue – that there aren’t enough doctors and healthcare professionals in India to accommodate the number of patients. I looked it up, and apparently there are 0.6 physicians per 1000 people in India, as opposed to 3.3 per 1000 people in Sweden, for example (according to Nation Master’s statistics, which has fairly old data). On top of that, the doctors are clustered in urban areas, so people in rural areas have to move whenever they have something more serious than a cold. Shoibal agreed that technology could do a great deal in terms of solving this problem, but only once the infrastructure improves. Right now, there simply isn’t a way to stream images, videos, and medical information out to the rural areas. Also, venture capitalists aren’t so eager to invest in startups that are trying to fix these particular issues. The affordability of the target population in these areas, at least to begin with, is very low (as opposed to the aforementioned 1%), and Shoibal said that many venture capitalists have too-high expectations, preferring to invest in companies with higher, faster returns.

I think, then, if startups are too small and risky, the money and drive to solve these issues will either come from big philanthropic organizations or from the government. Shoibal said that it would be great if the government invested in telemedicine in a big way, but that they’re not focusing on it at the moment (I’ll need to investigate this). They might be focused on technology, or on health, but not yet on both at once. One of the issues they are focused on is the concentration of doctors in urban areas but not in rural.

Shoibal was telling me that most doctors want to live in urban areas because that’s where they can create the life they want. Even though the need for doctors is in rural areas, they will make less money there, have worse infrastructure, have fewer things to do, have fewer options for schools for their children, and fewer options for jobs for their spouse. So even though the government has set up rural health centers, the doctors there are probably low-quality. Even then, Shoibal said that most qualified doctors will only visit the rural health centers a couple days a month, otherwise working at a private practice in an urban area. They do this to maintain a certain lifestyle and make enough money. There’s not much that the government can do in response, as it is better to have a doctor visit a rural health center for 1 day a month rather than not at all. The government could ramp up the incentives offered to these doctors, but that would require tons of money plus bettering the rural areas in multiple ways. Or, the government could try to force doctors to spend all their time in the rural health centers by making it illegal to have a private practice on the side; but then, given the choice, the doctors would all move to the private sector and not work in the public sector at all. Of course, there are some doctors that don’t care about money and really want to work where the need is. But still, the truth is that there aren’t enough doctors in rural areas, and it’s hard for the government to encourage them to move. If the government invested heavily in telemedicine, I think it could help enormously; but the first step would be building the appropriate infrastructure.