The potential power of patient profiling
In a recent post, We Can Do Better Than Patient Compliance A Look At Patient Segmentation As One Alternative, I made a case for patient profiling (also called patient segmentation) as a promising alternative to so-called compliance-enhancement methodologies.
* A New York Tims Magazine article, What Does Your Credit-Card Company Know About You? by Charles Duhigg (May 12, 2009) offers an example from the realm of financial services that illustrates the power of this approach
Disclaimers and Stipulations
First, however, I will, in anticipation of objections, acknowledge that (1) healthcare and credit cards are not identical or precisely congruent industries and (2) data mining can be used for nefarious, downright creepy purposes. Those disclaimers submitted, the similarities between profiling individuals along fiscal and healthcare lines are intuitively apparent and whether the results are used to rob or benefit the subject is a function of the user, not the tool.
Using Data Correlation To Assess Debt Repayment Behavior
The exercise is simple. Read these excerpts from the Times Magazine article and mentally change the context from credit cards dept repayment to healthcare behaviors.
After two decades of almost constant expansion and profitability, card companies today are in deep trouble. Monstrous losses estimated to top $395 billion over the next five years are growing as cardholders, brought low by the recession, walk away from their debts.
So credit-card firms are changing their business plans. Gone are the days of handing out cards willy-nilly and hoping that the cardholders who dutifully pay up will offset the losses from those who default. Today companies are focusing on those customers most likely to honor their debts. And they are looking for ways to convince existing cardholders that if they only have enough money to pay one bill, its wiser to pay off their credit card than, say, the phone.
Put another way, credit-card companies are becoming much more interested in understanding their customers lives and psyches, because, the theory goes, knowing what makes cardholders tick will help firms determine who is a good bet and who should be shown the door as quickly as possible.
Luckily for the industry, small groups of executives at most of the large firms have spent the last decade studying cardholders from almost every angle, and collection agencies have developed more sophisticated dunning techniques. They have sought to draw psychological and behavioral lessons from the enormous amounts of data the credit-card companies collect every day. Theyve run thousands of tests and crunched the numbers on millions of accounts.
The solution was learning to predict how different types of customers would behave. Card companies began running tens of thousands of experiments each year, testing the emotions elicited by various card colors and the appeal of different envelope sizes, for instance, or whether new immigrants were more responsible than cardholders born in this country. By understanding customers psyches, the companies hoped, they could tell who was a bad risk and either deny their application or, for those who were already cardholders, start shrinking their available credit and increasing minimum payments to squeeze out as much cash as possible before they defaulted.
The exploration into cardholders minds hit a breakthrough in 2002, when J. P. Martin, a math-loving executive at Canadian Tire, decided to analyze almost every piece of information his company had collected from credit-card transactions the previous year. Canadian Tires stores sold electronics, sporting equipment, kitchen supplies and automotive goods and issued a credit card that could be used almost anywhere. Martin could often see precisely what cardholders were purchasing, and he discovered that the brands we buy are the windows into our souls or at least into our willingness to make good on our debts. His data indicated, for instance, that people who bought cheap, generic automotive oil were much more likely to miss a credit-card payment than someone who got the expensive, name-brand stuff. People who bought carbon-monoxide monitors for their homes or those little felt pads that stop chair legs from scratching the floor almost never missed payments. Anyone who purchased a chrome-skull car accessory or a Mega Thruster Exhaust System was pretty likely to miss paying his bill eventually.
Martins measurements were so precise that he could tell you the riskiest drinking establishment in Canada Sharx Pool Bar in Montreal, where 47 percent of the patrons who used their Canadian Tire card missed four payments over 12 months. He could also tell you the safest products premium birdseed and a device called a snow roof rake that homeowners use to remove high-up snowdrifts so they dont fall on pedestrians.
Testing indicated that Martins predictions, when paired with other commonly used data like cardholders credit histories and incomes, were often much more precise than what the industry traditionally used to forecast cardholder riskiness. By the time he publicized his findings, a small industry of math fanatics many of them former credit-card executives had started consulting for the major banks that issued cards, and they began using Martins findings and other research to build psychological profiles. Why did birdseed and snow-rake buyers pay off their debts? The answer, research indicated, was that those consumers felt a sense of responsibility toward the world, manifested in their spending on birds they didnt own and pedestrians they might not know. Why were felt-pad buyers so upstanding? Because they wanted to protect their belongings, be they hardwood floors or credit scores. Why did chrome-skull owners skip out on their debts? The person who buys a skull for their car, they are like people who go to a bar named Sharx, Martin told me. Would you give them a loan?
Data-driven psychologists are now in high demand, and the industry is using them not only to screen out risky debtors but also to determine which cardholders need a phone call to persuade them to mail in a check. Most of the major credit-card companies have set up systems to comb through cardholders data for signs that someone is going to stop making payments. Are cardholders suddenly logging in at 1 in the morning? It might signal sleeplessness due to anxiety. Are they using their cards for groceries? It might mean they are trying to conserve their cash. Have they started using their cards for therapy sessions? Do they call the card company in the middle of the day, when they should be at work? What do they say when a customer-service representative asks how theyre feeling? Are their sighs long or short? Do they respond better to a comforting or bullying tone?
Its really hard to get clean insights of a cardholders state of mind, said Andy Jennings, the head of research and development at FICO, one of the biggest and oldest analytic firms. The more subtle the insight, the more cleverness finding it requires. If someone pays for a big cable television package each month with their card, are they rich? Or does it signal they dont have the sense to avoid products they cant afford? If they check their balance three times a day, are they worried or uptight? We may look at 300 different characteristics just to predict their delinquency risk.
From Cash Management To Compliance Management
If credit card usage information can predict which clients are more likely to default on loans to the level of identifying the single point of purchase most associated with low loan repayment rates and the single item, the purchase of which is most associated with high rates of loan repayment, it seems reasonable to theorize that an algorithm using prescription refill information and other available data could indicate which clients are most likely or least likely to have their prescriptions filled on schedule.
Applying data correlation tools to healthcare-pertinent behaviors, in fact, has the potential to designate if a given patient is, for example, likely to implement treatment as prescribed independent of other factors, become confused about the treatment recommendation, require ongoing reassurance from a physician to follow a medication schedule, or prove unable to commit to any treatment plan despite extensive support. Further, profiling can indicate the kind of clinician-patient relationship the subject prefers and whether a patient can benefit from or will resent reminders, further educational efforts, or extra appointments.
Not only would such information allow the derivation of optimal treatment enhancements for each patient but it could also guide the allotment of resources. Resources can be diverted, for example, from patients who routinely follows treatment recommendations without any special reminders, explanations, or incentives to those who require special efforts.
There is much, much more but the potential benefit is evident. Perhaps the next step is to put this test this hypothesis and, if it proves feasible, invest in collecting and correlating data that could improve the health and healthcare of individuals on a scale commensurate with the effort put forth by the financial institutions who want their debts repaid.
___________________________
* Again, In keeping with the principles of full disclosure, I want to alert readers to the fact that I am involved with EnrichMap, which offers a system of interventions based on patient profiling.