Since my time in India has come to a close and I have finally amassed the necessary internet connection, health, rest, and photo archive, I can now begin to clear my backlog of writing. I think it’s prudent to start by sharing what how I spent my time in India. Apologies in advance for the length.
As I explained in my previous blog post, I spent my summer working for the Aravind Eye Hospital in Madurai with three other co-interns from Penn. To be precise, we worked in LAICO, the outward-facing ophthalmology consultation arm of Aravind. LAICO’s mission is to help end needless blindness, and it furthers this objective by running workshops, training employees, and consulting for hundreds of aspiring independent eye care hospitals around the globe. LAICO also helps the Aravind hospitals themselves by training clinic managers, organizing research studies and spearheading innovative projects.
The structure of our internship was such that each of us focused our efforts on our own constructive project under the guidance and supervision of LAICO faculty. While very grateful to be given so much independence, I was unnerved at first by the realization that this structure implicitly assumed our competence. I distinctly remember this feeling during my second day of work in the office, which was after the brief introductions of the projects were given, each of our selections were made, and after my initial giddy excitement had calmed. As I sat at my desk in our shared office space, I yearned to support Aravind’s efforts by advancing my chosen project of creating system-wide analytics platform (comparing the different hospitals to each other), but found myself instead questioning what skills I really had to offer. Here I was, a 19 year old with nothing but a handful of college classes under my belt, trying to make a difference in a 42 year old business with dozens of case studies praising its practices. Could I really lead a impactful project? And would anyone actually listen to my ideas if I did?
Luckily I was somewhat inoculated against this potentially crippling imposter syndrome thanks to my time in Wharton. If there is anything I learned from business school last year it was that you sometimes have to “fake it ‘till you make it”. Any remaining doubts were quashed as soon as I got to know my co-workers. The tremendous kindness and respect that I was welcomed with as I began working with my team gave me the confidence I needed to move my project forward. I should acknowledge that some of this respect was probably unduly influenced by Penn’s brand name, the pervasive Indian fondness of white people, and the fact that my team members for some reason thought I was 25 until they finally asked about my age on my last day.
Before I explain my project it is important to understand the guiding premises behind it. The first premise is that Aravind’s audacious goal of curing unnecessary blindness will necessitate maximizing the efficiency and effectiveness of its limited available resources. This maximization is what has made Aravind so successful today and will undoubtedly determine Aravind’s success in the future. Additionally, improvement in any area naturally requires quantification, for how can one demonstrably improve in something without measurement.
Given these premises it follows that Aravind should put great importance in its analytic systems in order to drive the advancement of its vision. Unfortunately, this is not what one currently finds at Aravind, and my objective was to correct it.
After some initial research, I found that the main problems with Aravind’s analytics paradigm were due not to a lack of infrastructure, but to poor design. The IT department took much of the hospital administrative work digital back in 2004 and has nearly completed an impressive electronic medical record system that will serve all hospitals from the cloud. These information systems work together to put department level data at the finger tips of clinicians, managers and executives. The issue is that these metrics are analyzed in a vacuum, without the context of how well any of the other hospitals are doing. Every 6 months, each department in each of the 11 Aravind secondary and tertiary hospitals report their key statistics along with their growth from the previous year, perceived strengths and weaknesses, and goals for the next period to their respective hospital executives. This system does wonders to motivate clinicians and catalyzes some sort term improvements, but theres only so much you can do when looking at things in isolation.
Another way I thought about this deficiency was by translating it into the language of India’s favorite pastime. If I wanted to get a sense of how skilled the local Madurai cricket team is, with my limited understanding of how cricket is played and even more limited sense of what good cricket looks like, I would learn close to nothing from the isolated team statistics. Looking at the average number of runs scored by each player or the number of wickets hit per balls bowled and even comparing these numbers to the previous season will provide me with little to no insight into how good the team is. Conversely, looking at the team statistics within the context of the league they play in quickly illuminates their standing as an organization, and points out areas in need of improvement. Similarly, without context of how other hospitals of a similar class (in the same league) are doing in any given metric, it’s impossible to tell if one is doing well or what areas need improvement.
Thus our goal was thus to build a analytic platform that would provide users with accurate and well documented data for all of the hospitals in the system. This would allow Aravind to shift from treating each hospital as an island to allowing each hospital to serve as a laboratory for experimenting with different processes, policies and management strategies. Initially, we intended to work with developers in the IT department to begin development of a web application to display our analytics platform to the Aravind community. Unfortunately, the IT department was stretched too thin by its deadlines for implementing the EMR to take up the project, so we set our sights instead on solving all of the non-technical problems that prohibited this project.
In pursuit of this goal, we focused our efforts in two areas. The first was to create awareness and establish the benefits of such a project. We tackled this by compiling metrics for each of the hospitals in the 2017 calendar year into a spreadsheet using data from the biannual benchmarking meetings described above. We then used this sheet to create captivating presentations that showcased the potential benefits of internal benchmarking and presented to the senior leadership in Madurai along with decision makers in other hospitals. One of the highlights of my time at Aravind was traveling with my supervisor to Salem for two days to present our findings to the leadership team at the Aravind Hospital there.
Our second area of focus was in establishing each of our chosen metrics with strong definitions. We only realized the importance of having definitions after compiling the 2017 data and noticing that none of the metrics had documentation of the patient populations included or the sources. This lack of standardization made our presentations unreliable since hospitals could be reporting completely different metrics. For example, one hospital could be including pediatric patients in their outpatient numbers while another might exclude them, leading to a difference of 60,000 visits. We thus decided to create a reference manual for our platform that would define the definitions, sources, and additional notes for each metric. Despite planning on starting this reference manual with a small number of highly desired metrics, it quickly ballooned into over 100 pages of definitions. You can find a example of one of these pages below.
I’ll skip the play-by-play of the painstaking process and instead focus on the things I learned about working at Aravind and the outcome of my project. More than just my first time working in India, I think its fair to say Aravind was my first ‘real’ job. For the past three summers I’ve interned in number of medical research laboratories, but these experiences were only a slight departure from the guidance offered in a traditional classroom setting. For this reason I can’t really compare the work cultures in the U.S. to what I experienced in India, but I can share some of the interesting realities that I came to realize in this doubly novel environment.
1. Becoming a self-starter
With no teacher to assign work and no mentor to plan experiments I found myself being especially unproductive in the first couple of weeks in the office. All my schooling and work experience had trained me to be a machine: taking in objectives and facile frameworks and spitting out easily gradable answers. At Aravind I faced something quite different. I had a long term goal, but no pre-fabricated framework to apply, no sort-term deadlines to meet and no scheduled office hours to meet with my mentor. These fairly typical transitions were then further complicated by idiosyncrasies of working at Aravind, including the fact that small meetings in LAICO are often scheduled the same day or initiated by just walking into ones office and that language limited my interactions to those who spoke good english. This In addition, my team members weren’t given any sort of break from their normal duties to help my cause, and thus any meeting required strong and persistent advocacy.
Given these conditions I was forced to diverge from my normal behavior patterns in the interest of getting anything done. Setting up a meeting with all of the stakeholders to finalize a proposal could not be done from the comforts of my desk. Instead, It would require barging in on one of my team members and remaining in their office until all the emails and phone calls were made and the meeting was set. Despite being the type of person that would avoid an interaction like this at all costs prior to this internship, I learned over the last couple of months to come to terms with the realities of office culture. The truth is that in order to get anything done one requires the help of others, and recruiting the help of others requires having some interactions that you’d probably rather avoid.
2. Finding Consensus
Another difficulty I had while working on my project was finding the best way forward when provided with a number of disparate expert opinions. One of the key components of my project was creating the reference manual for the many hospital metrics. This required meeting with a number of doctors, clinic managers, and nurses to figure out what definition would best suit our purposes. What came to surprise me from this process is that many of the opinions from the different parties disagreed (occasionally quite dramatically), and these opinions could not converge even when each of the experts were brought together for a discussion. I was thus left in a uncomfortable position as final arbiter of decisions for which I had little domain reputability. At first I avoided this issue by just using the opinion of the most senior member of the group, but this had its own problems given the uncertainty of that person fully understanding the task—which was fairly abstract and further obfuscated by the language barrier— and accurately translating their ideas. As time went on, however, I adopted a different approach. Instead of differing to the most senior opinion I took each of the inputs into account to create what I felt was the best definition. To establish the credibility of my work I included extensive notes that justified each of my decisions and touched on the concerns raised by the expert group. This was a laborious task, but as I began to build this set of definitions and share them with stakeholders I came to appreciate this rigor. Having all of the reasoning spelled out forced the experts that I was meeting with to think more structurally and provide more precise objections. Additionally, it freed me from repeating the same explanations. In the long term, the reference manual that I create is intended to be a living document with additions and edits being made by the entire Aravind community over time.
3. Dealing with new customs
Even after working for 10 weeks in at Aravind I still could not keep up with all of the different office norms. From standing whenever a doctor or senior executive enters a room to completing every sentence with sir or ma’am in a conversation with a anyone higher up the ladder and adapting to a 6-day work week, it was difficult to act properly at all times. Although everyone was kind in pointing out our errors, I couldn’t help but think we were just living up to the stereotype of the arrogant American. One of the most jarring differences, which ended up helping me better understand myself, was presentation etiquette. I learned first hand that if your slides have any mistake—small or large, it will get exposed openly, without any remorse. This happened to me twice, once with a mix up of ‘principal’ and ‘principle’ in a small meeting, and a second time when I was interrupted during a presentation to 50+ people by someone to (wrongly) accusing me of misusing ‘biannual’. In both of these situations I was completely caught off guard. Filled with embarrassment and a bit of anger, I fumbled for words for a few seconds before apologizing for my mistake as gracefully as my emotions would allow. As I reflected on these experiences after my emotions had faded away I realized how egotistical my reactions were. There was no harm done by either of these comments, except to my perceived reputation in the eyes of those in attendance. In fact, one could argue that these comments were in fact productive since they could flesh out confusions shared by others. I thus resolved to react more intelligently in the future, but was this even possible? How could I circumvent my emotions and act as I actually desired? (More on this in my final post… )
One of my biggest apprehensions during this internship is that my work would culminate in little actual impact. Thankfully, when my project concluded with a presentation to the chairman and other senior executives I was told that they were fully committed to seeing this project move forward. For continuation of the project I provided my team members with the presentations I made such that they can continue building support for the endeavor, along with the extensive reference manual (will soon be circulated through the hospital system for edits).
In addition to the benchmarking project, I spent a fair amount of time at Aravind working with my Co-Intern Liz to try to help quantify compliance of Diabetic Retinopathy patients to their treatment regimens using data from the electronic medical record (EMR) system for her project. Since these data sets contained 70,000+ patients , there was no way to extract information from an excel sheet, and instead had to use a more powerful scripting approach. Although there were times when either the unintuitive data structure of the EMR or issues with R (the programming language we used) made me want to pull my hair out, I enjoyed the puzzling and was grateful for the opportunity to practice my computational thinking since my own project ended up not requiring any.
I have also been making full use of my time here outside of the office
Just about ever Monday Wednesday and Friday evening for the last two months I’ve been full heartedly embarrassing myself in front of my group mates and local teenagers on the dance floor. Whether it was Bollywood, Zumba, contemporary (if that’s even a genre), or salsa dancing was always up to the whims of master Gopi, our lively instructor, but regardless of the style I would spend each hour or so flailing my body awkwardly around the room—and having a great time doing doing it. Over the weeks I undoubtedly improved, but if any of you have seen me break out my moves before this trip you would know how little this actually says. Perceived dancing ability aside, I leaned a great deal in this class about rhythm, self confidence, and the art of taking oneself less seriously. For this, I owe thanks to Gopi for always bringing his infectious passion and my co-interns both for taking the initiative to find the class and for suppressing some of their laughter as I learned to stop moving like a robot.
On weekday evenings without dance class I usually took a quick auto-rickshaw to the nearby gym to let some energy out. There’s not much more to say about this activity other than its striking similarities between gym culture in Madurai to the U.S. In addition to having almost all the same equipment of a typical gym in the US, grunting, mirror flexing, and an obsession with bicep curls have also seems to have transcended cultural boundaries. To complete the western vibe, the gym also outfitted all of its non-mirrored walls with posters of white body builders. Not sure what exactly to conclude from this experience but I do appreciate the resource for allowing me to avoid complete muscular atrophy.
Despite our six day work week, we were able to appreciate the beauty of Tamil Nadu and beyond by way of several short Sunday excursions and few longer trips. With the help of our LAICO coworkers in the planning of our trips and the generosity our supervisor to give us 2.5 days off, we were able to explore the cities of Rameswaram, Kodaikanal, Thanjavur, Puducherry, Munnar (Kerala), and Delhi/Agra. On the days that we stayed in Madurai, we checked out the local attractions and even joined our co-workers on a couple nearby nature treks. I have come to a number of realizations about travel through these trips and my journey this summer as a whole, but I’m going to save this for a later post in order to do it justice.
More content to come in the next few days!