Thursday, May 23, 2013

Winter is Coming

I recently started watching the hit HBO series Game of Thrones. I’m watching it on DVD, starting with season 1, episode 1, and working my way forward. The series concerns the dynastic struggles of a series of noble houses, set in the fictional kingdom of Westeros, a medieval fantasy realm.


One of the noble houses, the Starks of the North, have the familial motto “Winter is Coming.” In the context of the series, the constant awareness of a cold, bleak future engenders a certain dourness of outlook, a dogged determination to perform one’s duty without joy.

But it occurs to me that as mottos go, when could do far worse. For Winter is coming! Not just the season of snow and cold. But the metaphorical winter that periodically comes to all of us. We are all subject to the unexpected illness, the unplanned breakdown. All of us occasionally face family emergencies. On the career side, very few of us escape winter’s storms. A plant closing or big layoff catches us. A market shift or technological change can doom or current career, and make our skill set obsolete.

Winter is coming, and the time to prepare is during the more verdant seasons of our lives. Spend less than you earn, and invest the difference. Keep a cash reserve, to help weather the storms. One of the most neglected ways to prepare is to invest in ones self. Education is a lifelong process, and we should all be trying to both stay current with rapidly advancing technology and acquire new skills in areas that may be far away from our current careers.

Winter is coming. But the actions we take as individuals have a lot to do with how long winter lasts. By our own efforts we can protect ourselves from the worst of winter’s ravages, and bring about an early spring.

Tuesday, May 7, 2013

Regulating Ahead of the Facts

Three weeks ago a horrific accident occurred in the US. This was the massive fire and explosion at the fertilizer plant in West, Texas that killed 14 people. Most of those killed were fire fighters and other first responders who were fighting the blaze just before the explosion occurred.


This accident is being investigated, and as of this writing a cause has not been identified. In the case of the fertilizer explosion, there has been considerable commentary indicating that at the heart of the disaster was some sort of regulatory failing.

I have to disagree with this perspective. Three weeks of investigation have not revealed the cause of either the fire or the subsequent explosion. Although there were large quantities of ammonium nitrate at the plant, ammonium nitrate alone is not explosive. Typically it is mixed with diesel fuel to create the conditions for a bomb. Even then it requires a detonator.

If you don’t know what causes a problem, it is difficult to guard against the problem before it occurs. This fertilizer plant was in operation for at least forty years before the accident, and it is one of many scattered across the country. In the last 60 years, this is only the third unintended explosion of this type to occur in the US. When you operate for tens of thousands of man-hours between incidents it is difficult to make a case that underregulation is to blame.

More government is not always to solution to every tragedy.

Thursday, May 2, 2013

Oregon Medicaid Study


A couple of years ago, Oregon decided to add people to their Medicaid rolls.  However, the state didn’t have enough money to add everyone who was uninsured, but wanted to get Medicaid coverage.  So in the interest of fairness, Oregon held a lottery.  The winners of the lottery got access to Medicaid coverage.  The losers continued on without insurance coverage.

It turns out that this is a perfect experiment to test the theory that having insurance coverage leads to better health outcomes.  The idea is that with insurance, people will have better access to physicians.  This access will increase monitoring and counseling.  Also, there will be less emergency rooms visits, leading to lower cost healthcare overall.  To test this theory, you would need a randomized group trial that would control for all the possible factors that would differentiate a population that already had insurance from a population that did not.

The lottery is what makes this a perfect experiment.  It separated the participants into two randomly selected groups: a control group with no insurance, and a test group with Medicaid insurance.  A randomized trial is the gold standard of experimental design.

At the end of three years, a study was published tracking the health results of the two groups, insured and uninsured.  The primary findings of the study:
  • Heath care utilization increased for the insured group.
  • Rates of depression were 30% lower in the insured group.
  • There were no statistically meaningful differences between groups in several measures of health: blood sugar, cholesterol, and blood pressure.
  • Hospital admissions and emergency room visits were statistically similar for both groups.

Liberal commentators are focusing on the first two bullet points: “Isn’t it great!  We’ve found a way to make poor people happier!”  “At last poor people can get the health insurance that is their birthright.”

Conservative commentators have focused on the last two bullet points.  “If your use of primary care goes up (by over a $1000 per person on average), but you don’t cut hospital and emergency room visits, aren’t you just pissing away the money spent on primary care?”  “More medical care does not lead to improved outcomes?  Then why are we expanding Medicaid?”

One of the interesting points is that even though the use of diabetes drugs increased in the insured group, their blood sugar (actually, glycolimated hemoglobin) did not drop.  This would not surprise anyone who has ever seen someone swallow a blood sugar pill, and then wash it down with a Mountain Dew.

Ultimately, I think this study shoots down the theory that the key to good health is going to the doctor regularly, which is kind of a silly idea when you think about it.  Healthy people don’t go to the doctor regularly.