upstream

dan heath

So I started reading Dan Heath’s new book Upstream. In it, he describes solving problems before they occur as upstream thinking, whereas the way we typically react to problems that arise are instances of downstream thinking.

The first chapter leads with the Expedia ceo recognizing that 20 million calls from their customer department in the last year had resulted from people not remembering their password or not getting their itinerary. They said at a cost of $5/hr, they solved a $10B problem. 

He talks about the U.S. healthcare system and it’s 1:1 ratio of downstream (treatment) and upstream (preventive) dollars spent are way out of wack with what other countries spend.

He says you can prevent crime upstream all the way to making sure pregnant women have as healthy and stress free of an experience as possible. How Norway gives out pre and post government sponsored maternal leave speaks volumes about their low crime rate.

“Your victories are stories written in data, starring invisible heroes who save invisible victims.”

The second chapter talks about problem blindness. How not seeing or not valuing the existence of problems can impede upstream thinking. 

NFL changed hamstring injuries because Dr. Elliot measured muscular imbalances of athletes pre-injury to focus on proper stretching, bringing injury rates down from 22 per season to 3. Before that was less common because coaches believed injuries just happened.

The Chicago public school system doubled their graduation rate by tailoring an engagement approach to at risk teens, down to changing a student's morning schedule with their first class as an elective so them being late from taking their sibling to school beforehand was less of a big deal. The barrier was teachers thinking what happened outside of school, what happened out of their class, wasn’t their problem. It was also hard to see what issues teachers think students have when they didn’t collaborate cross-departmentally. 

C-sections were an issue in Brazil because doctors were able to charge more for less work. Mother’s rallied nurses who rallied representatives into banning the practice without explicit consent, as it was happening in 85% of births. C-sections may be safer, but babies not born vaginally are forfeited less of their mothers' gut bacteria, which makes a difference in mortality, longevity, and disease susceptibility. 

Sexual harassment was expected until the '60s, when now comso editor wrote a book Sex and the Office, which spurred journalists to editorialize and bring the issue into existence.

“The seed of improvement is dissatisfaction.”

The third chapter kicks off with the rug master, CEO of one of the biggest carpeting manufacturers, Ray Anderson of Interface. He’d grown his company over many years, and was asked to speak about sustainability to his company. Coincidentally he had just picked up the book The Ecology of Commerce which denigrated supply chain work like carpet manufacturing as a causer of pollution and waste. The book changed the way he thought about his company and the world. This chapter focuses on ownership, how we can choose to fix the problems we cause. Anderson went to talk about sustainability but ended up galvanizing a company revolution.

Interface’s core business was to sell carpet tile made from nylon yarn, and nylon is a plastic made from chemicals found in coal or petroleum. They had to burn fossil fuel to make products out of fossil fuel. After his epitome, Anderson resolved to achieve a zero environmental footprint by 2020.

While Anderson was a force, others lack accountability because they sometime feel they don’t have the legitimacy to protest. This is called psychological standing. 

In Tennessee this lack of ownership occurred with baby seat safety. It was the number one killer of infants and it was preventable. Pediatricians reasoned they had the responsibility to take the problem of child mortality upstream to pre-injury. They wrote an article and rallied others, overcoming policy makers resisting with the argument of parents' rights. 

Anderson lobbied for his company too, finding creative ways to reduce their carbon footprint, weather it was by reclaiming old carpet when they replaced it with new carpet or paying for people finding and sending them abandoned fishing nets to convert it into nylon fibers, saving oceans making carpets. 

Dan points out that even though profits continued to rise through this transition, it wasn’t as fast as the market average. The point of the is not that efforts at preventing problems always pay for themselves, or that good intentions are always rewarded, but that we should push against complacency. He asks what harms do we accept that we’re capable of changing?

The last story of this section is about a clinical psychologist turned performance coach who tells a story of conflict resolution between two female workers. The clinician started by indicating her own responsibility in the pair’s disgruntlement, how she saw signs but ignored them, then asked each of them to do the same.

“I’d like each of you to tell the story of this situation as though you’re the only one in the world responsible for where we are.”

The fourth chapter talks about tunneling, starting with examples of how using upstream thinking can help you solve easy life hacks. Tunneling, however, comes into play when considering scarcity. Researchers found when people experience a scarcity of money or time or mental bandwidth, the harm is not that the big problems crowd out the little ones, but that the little ones crowd out the big ones.

The word tunneling comes from the psychologists Eldar Shafir and Sendhil Mullainathan, who in their book Scarcity, define it as: When people are juggling a lot of problems, they give up trying to solve them all. They adopt tunnel vision. There’s no long-term planning; there’s no strategic prioritization of issues. It confines us to short-term, reactive thinking. In the tunnel, there’s only forward.

They argue against the common notion that bad decisions lead to poverty, instead saying poverty leads to bad decision-making. People who are tunneled in poverty can’t engage in systems thinking when they’re concerned with where their next bed will be.

Outside of poverty, scarcity can be seen with jobs where workers are constantly reacting to problems. Nurses are professional problem solvers, and are not really incentivized to upstream solve, because saving the days feels great and heroism is quite an addictive drug.

Getting outside of the tunnel happens, with nurses, when hospitals can have daily safety sessions where they can strategize future prevention by discussing things that almost went wrong the day before. Taking off the blinders is necessary.

The brain is built for tunneling, and that’s why our species has survived. The brain is perfectly designed to get out of the way of danger, but it’s much more recent for it to start trying to predict when future danger would come and how to prevent it.

“When your emphasis is always forward, forward, forward, you never stop to ask whether you’re going in the right direction.”

The fifth chapter is about how you get the right people into the room. It’s mainly about getting Icelandic teens to drink less, something they solved by making natural highs in the form of sports clubs coached by professionals rather than parents. The psychologist Harvey Milkman said that the issue shouldn’t be to take away the want to get high, but to find a way to get kids high safely.

This chapter also focuses on domestic abuse in New England, and how the downstream reactivity was leading to domestic homocides. The solution was in making sure victims weren’t repeat victims through police drive bys and making sure spouses had jobs.

The last piece of this chapter was on eliminating homelessness in Rockford, Illinois. Mayor Larry Morrissey attacked the problem by providing cheap housing before people asked for it. He had a list of all the city’s homeless and where they were last seen. Through collaboration and with these tools the city was able to get to “functional zero” of homelessness within 1 year of implementation.

“There’s a paradox inherent of preventive efforts: We’ve got to create an urgent demand to fix a problem that may not happen for a while. We’ve got to make the upstream feel downstream.”

The sixth chapter is about how you’ll change the system. About a college student from Baltimore named Anthony Iton who thought the poverty he was witnessing on campus was the result of actual warfare. 

He said that fundamentally, what causes people to get sick and feel sick is a sense of a lack of control over what’s happening to them. They’re literally under siege. They’re struggling to find housing, they’re struggling to find good education, to avoid crime, to find jobs, to find healthy food, in some cases even potable drinking water. So low-income people in this country are basically juggling a whole bunch of balls simultaneously. The result of that constant juggling is stress. These communities were “incubators of chronic stress,” and that people’s neighborhoods were designed to produce premature deaths.

A Dutch bicycle company was having trouble getting their bikes shipped without being wrecked. People just didn’t feel like taking care of bike parts. So the company started plastering images of flat screen TVs on the boxes, because historically people take better care of sensitive items like TVs than they do bikes. 

“The law is just a set of rules based on inputs from power sources. If you want to change the rules, you’ve got to change the power inputs so that the outcome will be different.”

The seventh chapter is about finding leverage. He starts with Greek Polymath Archimedes’ quote, “Give me a lever long enough and a fulcrum on which to place it, and I shall move the world.” Dan breaks down that quote to really mean, “If you rig up a system that makes it easy for me to move the world, then I shall move the world,” and how it’s sort of a cop-out. 

The chapter runs through trying to bring crime down in Chicago through a program known as BAM, launched by Tony D, a bootstrap kid with an affinity for root cause solutions. 

They also focus on the issue of physicians historically being raised in silos and thrust into the fee-for-service world. They typically just focus on clinical interventions and counseling. But to actually solve someone’s health issues requires opening their refrigerator. You need to see how they’re sleeping, to understand their chronic stress. 

What Hackensack Meridian School of Medicine at Seton Hall University in New Jersey has been doing since 2018 is empowering new medical students to see the whole picture. Students are matched with a person or family in a nearby community. 

“My son has autism, he needs help getting resources,” or “I’m home alone in a wheelchair and feeling depressed, I need social outlets,” are the types of issues these students are helping their surrogate families solve. 

The point is that blossoming physicians won’t internalize how important their job is until they meet and feel connected to a real person. Another part of their curriculum is to engage with the community. They meet with leaders in local nonprofits and attend public meetings and do service work. 

Many of these students will stay in New Jersey and practice at HMH, helping to transform the health system from the inside. 

“The post-mortem for a problem can be the preamble to a solution.”

The eighth chapter is about recognizing early warnings to problems, and recognizing that sometimes those early warnings can sometimes be false negatives. 

It starts with LinkedIn’s director of Customer Success, who realized people who used the recruiting service were more likely to pay a recurring subscription if they engaged with the product as soon as they signed up.

The churn rate, or the rate at which people didn’t renew their subscriptions, was about 30% before the directors intervention. What she did with her knowledge was outreach to new customers, going so far as to draft recruitment emails they figured out those specific recruiters were using to hunt. It cut the churn rate in half. Knowing why people stayed let them develop a strategy to help keep even more customers, valued at over 10 millions bucks in annual revenue. 

The next case is about an ambulance service in NYC, which places its ambulances in places it’s identified to be high volume (more calls during the day than night, more calls in areas with older citizens than younger ones, higher volume on NYE and July 4th, less on Christmas and Thanksgiving, high during flu season, spikes in nursing homes during feeding time (when nurses visits rooms and find issues) and more heart attacks in winter from shoveling snow). It’s a recognized tactic called forward-deployment. 

In areas where EMS is embedded in a fire department, proximity to a firehouse can sometimes mean life or death (a selling point for real estate agents: First floor master - and 3 minutes from the firehouse).

Japan has a similar detection system with earthquakes. A primary wave, which precedes a quake and is usually imperceptible to human senses, can save lives. Japan’s seismographs can detect a primary wave, and within 3 seconds send a message to businesses, railway operators, factories, hospitals, schools, nuclear plants, and the general public. Everyone is ready half a minute before the world starts to shake.

The flip side to this is that sometimes these early signs can be misleading. In South Korea, thyroid cancer seemingly went up 15-fold in as many years. The problem was how cancer’s definition changed. It was being identified as any benign tumor, and so everyone was getting a surgery to have it removed, some with irreversible side effects, with some even dying. Before this was known, though, the expertise was so regaled that South Korea promoted medical tourism to offer aliens to come just for a screening and potential surgery. 

The reason we should not ignore warning signs, though, is because they’re there for a reason. Currently, we’re experiencing a global flu pandemic, and some people are not taking the social distancing and self-sheltering seriously. This happens because of alarm fatigue. The media sensationalizes catastrophe, and when everything is cause for alarm, nothing is. 

The example they use to drive this home is Sandy Hook. An Atlantic writer once wrote, “There is one developed country—and only one—in which it is not only legal, but easy and convenient, to amass a private arsenal of mass slaughter,” wrote David Frum in the Atlantic. “That country also happens to be the one—and the only one—regularly afflicted by mass slaughters perpetrated by aggrieved individuals.”

The point was that the warning signs are there, if we’re not too wrapped up in what doesn’t matter to notice. 

The ninth chapter is about how to know when we’re winning and what ghost victories we need to stay on the lookout for. 

Police in one decade had low crime rates, so after retirement all the chiefs of that decade had consulting firms. Crime in another decade was worse, and none of those chiefs have the same retirement setup.

The first example was about sidewalk repair in Boston. The chief engineer of the public works department was tasked with fixing sidewalks, and how the city seemed to win a lot based on the way they were measuring what was broken, what was being asked to be fixed, and what was actually being fixed. What was actually happening was that the areas that were being fixed were in the rich white areas, because those are the types of people who were asking for repairs.

The city had organically created two self-fulfilling prophecies, because the rich white community thought their calls would be answered, so they called. The poor, black communities thought they would be ignored, so they didn’t call. 

Recognizing their fake wins, the system has been restructured to fix the places that matter most. 

“Our brains, when confronted with complexity, will often perform an invisible substitution, trading a hard question for an easy one.”

“We look for noisy measures, and choosing the wrong short-term measures can doom upstream work.”

The worst type of ghost victory occurs when the measures become the mission. In England the DOH was concerned with long wait times in hospitals. As a result of a policy to cut down on wait times, wait times shrunk. Upon investigation, however, it was revealed EMS was just waiting in ambulances outside the ER longer to wait for an open slot. 

NYC had CompStats to adhere to, so they would downgrade issues like rape to theft so that their stats would look good.

In the mindless pursuit to hit the numbers, people will do anything that’s legal without the slightest remorse, even if it grossly violates the spirit of the mission, and they will find ways to look more favorably upon what’s illegal. 

The 10th chapter was about knowing you’re not doing harm. 

This island south of Australia was a haven until hunters brought in rats and cats from Tasmania and they fucked up the whole ecosystem to the point that they thought they could fix it by killing all the rats and cats that had infiltrated the island. Problem was, once they did that, they created other issues, which had other issues.

Upstream interventions tinker with complex systems, and as such, we should expect reactions and consequences beyond the immediate scope of our work. In shaping the water, we will create ripple effects.

The city of NY had a ton of suits related to Central Park branches falling on peoples heads. The pruning budget for the city had been cut by 100k, which led to 11m in suits. 

Open floor plans at work hinder communication, not as expected. When people are crammed in together, they find ways to retain some privacy through headphones or books or deeply unwelcoming glances.

"It comes down to the point not being that we don’t succeed by foreseeing the future accurately, but by ensuring that we’ll have the feedback we need to navigate."

The eleventh chapter was about who pays for what doesn’t happen. It was a crucial chapter in upstream efforts because it addressed changing payment models for healthcare. 

The chapter starts with a discussion on the misrepresentation of life expectancy. How 120 years ago the average life expectancy was 47 and now its in the 70s. Dan says most people interpret that as modern society helping people live 30 years longer than they did a century ago, when really we’re just stopping more child mortality. We have less dead 0 year olds pushing our number down, about 3% to the 19th century’s 30%.

The chapter goes on to discuss how public health isn’t necessarily set up to fit in with upstream preventative spending. He uses the analogy of a car, just like I do in my Health Plan 101 class at work. 

“Imagine if a car that took more hours to build was more expensive. That would make no sense. Cars would not become better and cheaper if you paid for them that way”

Reactive efforts succeed when problems happen that are then fixed. Proactive efforts succeed when nothing happens, so who pays for what doesn’t happen.

For decades nursing homes were getting the short stick of this preventative facelift. Nurses and other staff suffered many lower back injuries due to moving patients to or from their bed. This cost the nursing home industry a lot of money in lost work and worker claims, but leaders were reluctant to invest in the solution - transport machines - because they were slower, they cost a lot, and they created a whole new set of procedures and training which would slow work. It’s easier to stay in the tunnel and accept that, every now and then, someone’s going to get hurt. 

The argument for entry was that the cost of these machines would pay for themselves in all the work that wouldn’t be lost and the claims that wouldn’t need to be paid out. Industries like this, largely due to their leadership, need proof before investment. It was practiced and studied in a small sample, eventually revealing a 35% reduction in claims. The inertia eventually pushed against the prevention.

The situation was similar with the Nurse-Family Partnership (NFP) which was a program that paired nurses with pregnant women to lower infant mortality and poor pregnancy practice. A guy who worked in childcare designed it to confront the negative behaviors in daycare he wouldn’t affect because the damage was done during pregnancy. 

“We can pay to fix problems once they happen, or we can pay in advance to prevent them”

Paying for these upstream efforts ultimately boils down to three questions: Where are there costly problems? Who is in the best position to prevent those problems? And, how do you create incentives for them to do so?

The chapter rounds out with an alternative payment model to fee for service in healthcare. It’s the accountable care organization that Lumeris is founded on, and the idea of capitation, where doctors are paid a flat fee to care for a member, a stipend to spend against based on their risk rather than to charge the insurer per member claim. It reduces cost, and when quality measures are met alongside them, the physicians share in the savings they mutually create.

The twelfth chapter of this bad boy was about going really far upstream. The chapter describes the Y2K czar and his position in the US government to quell the panic induced by the public at the fear of all computers crashing when the century turned over.

He was assigned to prevent the worst from happening, a thankless job considering if he did it well people would question why his job was necessary in the first place. 

Part of that expected downturn was that the Fed ordered 50B in new currency put into circulation. 

All that went out when the century changed was that the Low-level Windshear Alert Systems went down in a few cities and had to be rebooted. 

FEMA was featured doing a hurricane simulation program in New Orleans in 2004 with a fictional storm named “Pam.” It was practiced not even a year before Katrina hit, and the lives lost were incredible amid the many complaints that the response was poor. The city engaged in an emergency protocol called contraflow that turns the highways all into one direction so that people can evacuate the city. 

"The problem with isolated events is that it’s hard to convince people to collaborate when hardship hasn’t forced them to."

Another example is with hacking when so many people are the weakest link in their own security system, despite education. 

There’s something to be said in this chapter for the prophet's dilemma, where a prediction of events is what prevents those events from happening. 

Another example was with NASA’s first exobiologist, now known as an Astro biologist, who is responsible for making sure we neither bring foreign viruses from off planet but for also ensuring we don’t bring our viruses to foreign soil. The office is quoted as saying, “So far as I can tell, planetary protection is the first time in human history that humans as a global species decided to prevent damage before we were capable of doing something.” 

The thirteenth and final chapter in this wonderful book is about going personally upstream.

The chapter leads with hero dolls given to children whose parents were deployed with the military. Likenesses were sewed onto dolls so kids would have something to remember and compare against when their parents returned home. 

“The best time to plant a tree is 20 years ago. The second best time is now.”

“Be impatient for action but patient for results.”

The last example used to demonstrate the cascading effect of upstream effort, comes, ironically, from CMS. 

CMS actuaries revealed that they couldn’t, in fact, certify that a diabetes prevention program was a cost-saving program because it helped people live longer, which means their life overall costs the healthcare system more. 

The ending message of the book is to bristle against the urge to be an immediate hero. How the real heroes are in the shadows, thankless, because they prevented the problems requiring heroic action in the first place.