Physician Insurers
Association of America
2275
Research Blvd., Suite 250, Rockville, MD 20850
Telephone:
301-947-9000 Fax: 301-947-9090
Website: www.thepiaa.org
July 8, 2003
THE WEISS RATINGS
REPORT ON MEDICAL MALPRACTICE CAPS
Propagating the
Myth That Non-Economic Damage Caps Don’t Work
On June 3, 2003, Weiss Ratings, Inc. published a report regarding the performance of the medical malpractice insurance industry entitled Medical Malpractice Caps The Impact of Non-Economic Damage Caps on Physician Premiums, Claims Payout Levels, and Availability of Coverage.[1] The major recommendation of the report is that “Legislators should put proposals involving non-economic damage caps on hold until convincing evidence can be produced to demonstrate a true benefit to doctors in the form of reduced med mal costs.”[2] Unfortunately, the Weiss report is ill conceived, and misleads the reader by falsely demonstrating that non-economic damage caps have not worked. Both of the data sources used by Weiss have gone on record disagreeing with the report’s methodology, as described herein.
The conclusions drawn by Weiss are opposite of those previously published by reputable entities, such as the Congressional Budget Office, US Department of Health and Human Services, Joint Economic Committee of the United States Congress, Standard & Poors, American Academy of Actuaries, Tillinghast, and Milliman, USA, to name a few (see Appendix A). Unlike Weiss, all of these highly respected organizations have considerable experience and acceptance by government and industry for their knowledge and analytical product.
The purpose of this document is to evaluate Weiss’ use of the data and analytical process. In short, Weiss misuses published industry data in an effort to demonstrate that non-economic damage caps enacted by several states have not been effective in reducing medical malpractice premiums in those states as compared to states without caps. Weiss underestimates the “average” claim costs for the two groups of states by employing inappropriate analytical technique to represent the burden on insurers. This is an error that is readily obvious to those who work with medical malpractice claims data, and it misleads the reader to an inappropriate conclusion.
WHAT DID WEISS DO WRONG?
Grouping the States
Weiss has grouped 19 states as having caps on non-economic damages, and 32 others (including the District of Columbia) as not having caps. Unfortunately, states with effective caps, such as California with a $250 thousand cap, are considered the same as states having various levels of caps up to and including $1 million. In fact, only 5 of the 19 states have a $250 thousand dollar cap similar to that being proposed under current legislation[3]. Eleven of the states have caps of $500 thousand or greater. No attempt has been made to evaluate the effectiveness of caps at various levels, they have simply been lumped together. The American Academy of Actuaries has testified that caps are a key element of tort reform, and must be set at a level low enough, such as $250,000, to have an effect.[4] Any comparison chosen to demonstrate the effectiveness of non-economic damage caps should be sensitive to the level of caps in the various states and to their individual effectiveness.
In addition, as clearly shown on Appendix 1 of the Weiss report, more than half of the states enacting non-economic damage caps had not done so prior to the baseline date of 1991. Weiss compares premiums and claims costs for only two years, 1991 and 2002. The caps enacted in 10 states were not in place in 1991, and thus, these states should not be included in the “cap states” category for this analysis. Two other states had only adopted their caps in 1990, and the beneficial effects of these laws may not have been recognized in the data by 1991 due to constitutional challenge and uncertainty about the ultimate effects of the caps.[5]
Measuring the Premiums
Weiss uses the annual insurance rate surveys published by Medical Liability Monitor (MLM) for three medical specialties[6] as the source of insurance premium data. He calculates median average premiums by state and then calculates a median premium for 1991 and 2002 for the two groups of states.
For example, Alabama had two insurers listed in the 2002 study, each with a premium for the three specialties. Weiss simply ranks the premiums from least to most, and then selects the middle value (or mean average of the two middle values when there is an even number of rates) as the median average value, as shown below.
MEDICAL LIABILITY MONITOR RATE SURVEY DATA
ALABAMA
Insurer Specialty 1991 Rate 2002 Rate
FPIC[7] Internal Med N/A $ 6,043
ProAssurance[8] Internal Med $ 5,008 6,806
FPIC Gen Surgery N/A 19,286
ProAssurance Gen Surgery 25,629
27,694
FPIC OB/GYN N/A 36,506
ProAssurance OB/GYN 45,368 38,873
Median 25,629 23,490*
*calculated as the mean average of $19,286 and $27,694
Alabama was selected for this discussion simply because it is alphabetically the first state. However, these data demonstrates many reasons why the use of the median is improper:
· Data for different insurers are used for the two comparison years.
· The median value is representative of only general surgery rates because general surgery rates are always higher than internal medicine and lower than OB/GYN.
· Because two carriers are represented in 2002 and only one in 1991, the median value chosen by Weiss (the average of the two general surgery rates) is actually lower than the 1991 rate. However, the actual general surgery rates for the only carrier shown for both years increased – the opposite of Weiss’ result.
· The premiums shown are not adjusted for various discounts or surcharges, and do not reflect any dividends which may have been paid back to policyholders, thus reducing their total outlay. Medical malpractice insurers paid substantial dividends in the 1991 era, which had been largely reduced by 2002 due to industry losses.
Using the product of this calculation to represent insurance industry revenues is flawed for many additional reasons. First, there is no certainty that any of the table rates listed in MLM are actually charged. Carriers may have a premium filed in a given state (or in multiple territories in states), but may not write much business there. Weiss’ analysis gives no weight to the actual amount of insurance sold by the various companies in any state, nor does it reflect discounts or surcharges which are routinely applied to standard premiums. In addition, many insurers pay policyholder dividends, which in effect reduce the annual premiums paid.
MLM has objected to Weiss’ misuse of its data. In a July 7, 2003 email to Senate Majority Leader Frist, MLM Editor Barbara Dillard states “We believe it is misleading to use median annual premiums compiled with data from Medical Liability Monitor to demonstrate the effect of non-economic damage limits on liability rates.”
The Weiss analysis only includes premium data for three medical specialties, thus ignoring the experience for all of the rest. Even more glaring is the fact that the MLM data does not exist for seven of the capped states and five of the non-capped states for 1991. But, this did not stop Weiss from irresponsibly including these states in the analysis (see Weiss’s Appendix 1 and 2).
An analysis using actual premiums as reported to the National Association of Insurance Commissioners (not medians) is helpful in evaluating differences between states having effective damage caps throughout the period of Weiss’ analysis and those without. Such premiums include surcharges and discounts which may have been applied to standard rates.
The four states having a $250,000 cap prior to 1991 (CA, CO, IN, KS) saw their total premiums increase by 28.0% between 1991 and 2001 (2002 data not available yet). States not having the $250,000 non-economic damage cap experienced a collective 47.7% increase in premiums, over 70% greater. See Appendix B for details. This wide gap in premiums actually collected compares inversely to Weiss’ faulty conclusion that annual premiums in states with caps increased by 48.2% as compared to 35.9% in states without caps.
Measuring Claim Costs
In order to evaluate the difference in claim costs between the two groups of states, Weiss analyzes median claim payments by state for 1991 and 2002 as reported to the National Practitioner Data Bank (NPDB). The NPDB provides the only readily available source of medical malpractice insurance indemnity payments by state. However, in order to use these data effectively, one must understand the nature of the claim payment values reported, and the shortcomings from that which might be normally expected (see Appendix C for a discussion of the NPDB claim payment data).
The use of the median claim payment value greatly compromises the accuracy of Weiss’ analysis. While the median (or middle value of the claim payment distribution) might be an effective descriptor of what a plaintiff might receive as payment (before paying almost half to his/her lawyer), it cannot be used to measure the claim payment burden on insurers. The use of total claim payments reported by state shows a much larger differential result than Weiss’ reported payout increase of 83.3% for capped states as compared to 127.9% for non-capped states.
The increase in total claim payments for the four states having a $250,000 non-economic damage cap during the period of the Weiss analysis is 52.8%, compared to 100.1% for all other states – an 89.6% difference (See Appendix D). Thus the experience in the capped states is almost twice as good as that for states without effective non-economic damage caps prior to 1991. Using his faulty median calculation, Weiss would have us believe that the difference is only 53.5% (127.9/83.3).
The NPDB has gone on record opposing Mr. Weiss’ methodology, saying that “Although the statistical median is usually the best measure of the “average” malpractice payment received by claimants, it does not show the ‘burden on insurers.’ The ‘burden on insurers’ is the total amount of dollars paid, not the ‘average’ or median payment.” (see Appendix E for NPDB statement).
Investment Performance
In addition to inappropriate analysis of premium and claims data, the Weiss report comments on the investment performance of medical malpractice insurers. Being a long tail line of insurance, medical malpractice insurers routinely utilize the investment income generated by the premiums they collect and hold for the payment of claims in the future.[9] It is no secret that bond yields have declined over the past decade, and are now at historically low levels.
In spite of the fact that medical malpractice insurers are 80% invested in bonds and have less than 10% invested in the stock market[10], Weiss still concludes that stock market losses are responsible for insurers’ poor performance. While the fall in interest rates has reduced the interest income available to offset premiums, Weiss fails to mention that when rates go down, bond values go up, and insurers have been able to book capital gains to bolster their investment income.
As shown in the exhibit which
follows, the total return on investments for the industry has remained fairly
stable, and does not explain why rates are rising. Rates are rising because of increasing claim costs. 
CONCLUSION
The Weiss report recommends that “…legislators must immediately put on hold all proposals involving non-economic damage caps until convincing evidence can be produced to demonstrate a true benefit to doctors in the form of reduced med mal cost.” This information exists, as reported herein and by many other reputable sources, and now is the time for the enactment of effective federal health care liability reform.
VIEW APPENDICIES Appendix A