Showing posts with label Statistics. Show all posts
Showing posts with label Statistics. Show all posts

Wednesday, May 28, 2014

Change in Bankruptcy and Health Outcomes



Rotated Factor Matrixa

Factor
1  (52% of Variance)
2 (21% of Variance)
Infant Mortality 2007 Deaths/1000
.915
.157
Life Expectancy
-.824
-.473
% Low Birthweight Babies
.700
.341
Groups per million
.612

% change in bankruptcy
-.441

Percent Uninsured in Demographic Group for All Income Levels

.979
Percent under age 65 in 200% of Poverty
.372
.753
Extraction Method: Principal Axis Factoring.
 Rotation Method: Varimax with Kaiser Normalization.a
a. Rotation converged in 3 iterations.

I've added the change in bankruptcy rates to the factor analysis for health and hate groups.  I had to remove income from the analysis so the model would converge.  The factor structure did not change radically otherwise.  The % change in bankruptcy loaded weakly and negatively on the factor with health outcomes and hate groups.  Of all the factors entered, life expectancy correlated most strongly with the % change in bankruptcy followed by infant mortality. 


The graph for life expectancy suggests a positive relationship with the change in bankruptcy.  The increased change in bankruptcy could be due to more permissive laws in granting bankruptcy protection from creditors as can be seen in the video below.  These states may tend to have better quality of life characteristics than those who do not.  For example, Hawaii had the highest life expectancy and an increase in bankruptcies of 129% (and zero hate groups).  As can be seen in the above graph, this relationship is far from perfect accounting for only 15.2% of the variability,


There is also a significant negative correlation with infant mortality and change in bankruptcy accounting for 13.4% of the variance.  The District of Columbia is an outlier with a rate of 13.1 deaths per 1,000 live births but an increase of 26% of bankruptcies (DC also has the highest rate of hate groups)/  Removing DC gives a slightly stronger relationship with 14.9% of the variability accounted.

**Related Posts**

A Wave of Hate Groups in California? No in Washington, DC

 

How do the States Stack Up on Infant Mortality? (Cross Post with PUSH)

 

Correlation with the Number of Hate Groups per Million, Poor Health Suggests More Hate



Personal and Medical Bankruptcies: A Follow Up

 

Income and Life Expectancy. What does it Tell Us About US?

Sunday, May 11, 2014

Personal and Medical Bankruptcy Followup

Lloyd Stires has a great series of posts on the decline of mortality in Massachusetts compared to similar surrounding counties.  They are linked in related posts below.  

I've also received a nice response to my post on medical bankruptcies.  Here is a follow up.

I have received comments from Thomas M Miovas of appliedphilosophyonline.com that more exact methods are needed to determine the cause of these bankruptcies.  I agree completely.  However in the lack of these more comprehensive studies pilot work can be done to aid in planning these studies.

Assuming the prerecession estimate of 62.1% of all bankruptcies still holds (which may be a shaky assumption due to the state of the economy), we can multiply that percentage by the national increase in the rate of personal bankruptcies from '07-'11 of 62.15%.  We would find that there would be a 38.6% increase in medical bankruptcies and a 23.6% increase in the number of nonmedical bankruptcies over the same period assuming the rate of personal bankruptcies is constant.

We can also look at the rate of increase in personal bankruptcies in states expanding medicaid vs those who are not.  There is a somewhat higher increase in the rates in states expanding medicaid as can be seen in the boxplot below.  However there is no statistically significant difference between the two groups of states.


Group Statistics

Expanding medicaid
N
Mean
Std. Deviation
Std. Error Mean
% change in bankruptcy
not participating or considering not participating
22
43.0%
33.79597
7.20532
participating or leaning
29
68.4%
58.56026
10.87437

These types of analyses can provide indications of meaningful underlying patterns for future research.

**Update**

Adams County HC4ALLPA member Becky Spoon has informed me that there was a study of medical bankruptcies in Massachusetts after 2006.  There was a decrease in the rate from 59.3% to 52.9% from 2007 to 2009.  Over the same period there was a 49.5% increase in the number of total bankruptcies.  Multiplying the study rate by the % increase suggests that there is about a 25.8% increase in medical bankruptcies and a 23.7% increase in other types of personal bankruptcies.  These increases are smaller than that of the nation as a whole but the problem still persists.

 **Related Posts**

The Affordable Care Act Having an Impact in Some States but not Pennsylvania




National, State, & County Uninsured Estimates



Health Insurance and Mortality, Part 1

 

Health Insurance and Mortality, Part 2