APPENDIX 1. Model Executive Summary
• A small-scale trial shall test the feasibility of the approach and the validity of our assumptions before the regional and/or nationwide introduction of the program is started. Costs and benefits from such a test trial are not included in the present financial models. The current models reflect the nationwide implementation of the program.
• Savings from implementation are derived from lower healthcare spending when “Efficient Best Practice” (EBP) algorithms are adopted, from the avoidance of “Defensive Medicine” (DM), and from a reduction of medical malpractice litigation. All three issues go hand in hand.
• Costs are generated by the need to build an infrastructure (mainly based on information technology) enabling the implementation of EBP guidelines.
• With regard to the redistribution of the program’s net savings, two different models have been calculated.
(1) Model A offers patients who join the program and, as such, forego their rights to sue physicians for medical malpractice, free life and disability insurance coverage or the equivalent value if such insurances are not applicable like in the case of children or the elderly.
(2) Model B offers participating patients guaranteed, no-fault economic damages (“commercial damage compensation”) (estimated at an average of US$300,000 [2007 dollars, to be inflated]) if they suffer from a severe, avoidable medical injury.
- In addition, both models also offer physicians a strong reduction of their medical malpractice insurance premiums and allocate some money to G&A expenses. To the extent this is affordable, net savings are also used to provide health insurance coverage to uninsured Americans.
• Model A is somewhat more costly during the first few years than Model B.
• In any case, losses during the first few years of the program are relatively limited and will be offset by enormous net savings later on.
1. Costs
Costs arise from the generation of an infrastructure that will enable the nationwide implementation of EBP guidelines in hospitals, outpatient settings, and individual or small group private practices. This infrastructure is largely based on the implementation of interactive information technology tools, including electronic medical records, computerized physician order entry, and clinical decision-making tools. Guidelines have already been developed in a number of institutions. EBP is based on experience from the Salt Lake City region, where high-quality/low-cost standards have already been implemented and practiced with great success.2
In addition to costs for implementation, maintenance expenses have to be accounted for. Guidelines and software systems will have to be updated regularly to adapt to changes and advances in diagnostics and therapeutics reflecting the continuous progress made through medical research and development. Hardware has to be upgraded or replaced from time to time. New staff has to be introduced to the system and existing staff has to get acquainted with updates and changes.
2. Savings
Savings are realized through implementation of EBP. Estimates of maximum potential savings are in the range of approximately 30% of the total healthcare costs.
Avoidance of DM will also generate substantial savings. Avoidance of DM is a special facet of EBP but will be treated separately from EBP in this report, because its realization follows a somewhat different logic than the implementation of non-DM aspects of EBP. (If properly incentivized by the conditional reduction of their liability premiums, physicians can be persuaded to comply with EBP guidelines irrespective of the number of patients participating in the system. In contrast, the extent of abandonment of DM will critically depend on the proportion of participating patients, because nonparticipating patients can and will still sue physicians directly for medical malpractice.)
The smallest component of savings stems from the reduction of medical malpractice litigation costs, which will be strongly reduced under the program.
Net Savings Redistribution Options
There are several options for redistribution of net savings from the program, including, but not limited to, the following list:
1. Reduction of medical malpractice liability insurance premiums for participating physicians
2. General and administration
3. Commercial damage compensation of participating patients for severe, avoidable injuries
4. Provision of life insurance (or the equivalent value) to every participating citizen
5. Provision of disability insurance (or the equivalent value) to every participating citizen
6. Reduction of healthcare insurance premiums (benefiting employers and employees)
7. Savings of governmental healthcare expenses (Medicaid and Medicare)
8. Provision of healthcare insurance to uninsured citizens
In modeling scenarios, two redistribution variants were run.
Both variants comprise the reduction of medical malpractice liability insurance premiums as an incentive for participating physicians (Point 1) and G&A costs (Point 2). Furthermore, as a patient benefit, Model A also includes the provision of disability and life insurance (or the equivalent value) to every participating citizen (Points 4 and 5). Because such insurances would not make sense for all citizens, the equivalent value should be transferred to the elderly, children, and potentially those already insured for life and disability to a sufficient extent. This is important to also incentivize all patients adequately.
As an alternative benefit for participating patients, Model B comprises the guaranteed provision of full commercial damage compensation for severe, avoidable injuries (Point 3). Although patients will not receive any amount for noncommercial (noneconomic) damages as may be awarded by a jury to a patient prevailing in a conventional medical malpractice suit, the advantage for patients would be threefold: (1) patients do not have to undergo the substantial inconvenience associated with a medical malpractice suit; (2) patients have no risk of losing their suit and incurring litigation costs (currently only one of 15 patients who suffer from injuries resulting from medical negligence receives compensation); and (3) patients shall receive their compensation within 1 year of the occurrence of the injury, whereas it takes on average 4 to 5 years for a medical malpractice suit to resolve with the award, if any, paid out to a prevailing patient. A few years after the inception of the program, net savings are expected to exceed the required resources for the previously described redistribution packages. In terms of defining how the remaining net savings can be reinvested, the provision of healthcare insurance to uninsured citizens (Point 8) has been included in both models to an extent this is affordable. This means that after some time, an increasing number of uninsureds could be integrated into the US healthcare system. A few years later, net savings are expected to allow the provision of health insurance to all otherwise uninsured citizens.
Financial Model and Valuation Approach and Methods
Uncertainty of input parameters is taken into account by using probability distributions for the key variables. Probability distributions are generated with @Risk (Palisade Corp, Ithaca, NY). Computer simulations determine the impact of commercial risk on costs, savings, and project values.
The calculated savings, total (net) savings, net savings after redistribution, and the overall expected net present values (eNPV) are the results of a computer simulation (Monte Carlo simulation). The simulation can be interpreted as extensive scenario modeling that provides a distribution of figures and values with some characteristic statistic measures such as mean and percentiles.
Representative Data From Modeling: Assumptions
Acceptance of the Program
Costs and savings over time will evidently depend on the acceptance of the program by both physicians and patients. Two different assumptions concerning the physician and patient participation rates were made:
1. Fast acceptance of program (optimistic assumptions, Fig 1)
In the fast program acceptance scenario (Figs 1, 3), the expected maximum 90% of physicians joining the program is reached quickly in Year 3.
The patient recruitment rate lags behind physician recruitment (50% of participating physicians in Year 1 and then 1 year behind the physician recruitment rates), because physicians must join the program first to provide a logical reason for patients to participate.
The uncertainty of these assumptions is accounted for by using a Pert distribution of ± 25% around these means or 80% to 100% when the expected maximum 90% participation will be reached.
2. Slow acceptance of program (conservative assumptions, Fig 2)
In the slow program acceptance scenario (Figs 2, 3), the expected maximum 90% of physicians joining the program is reached in Year 6.
As described for the fast scenario, patient recruitment lags behind the recruitment of physicians.
The uncertainty of these assumptions is taken into account by using a Pert distribution of ± 25% around the mean assumptions or 80% to 100% when the expected maximum 90% participation will be reached.
Figure 3 displays the increase of physician and patient participation rates over the first 8 years of the program. For all recruitment curves, it is assumed that the mean participation rates will remain constant at 90% after this maximum value is achieved.
Monte Carlo Simulations: Total Net Savings in Models Embracing Efficient Best Practices: (1) fast acceptance of the program (see Table 1, Fig 4A-B); (2) slow acceptance of the program (see Table 2, Fig 4B)