The Insurance Industry’s Cost-Sharing Provides Penalties for People who Act to Improve their Health, Discouraging Healthy Behaviors - David Ganick
The U.S. Insurance model is one in which health insurance is financed through multiple pieces, all paid by the consumer (employer, employee or individual). These pieces are the premium (paid on a monthly basis and making up the bulk of the payment) and the cost-sharing portions (co-pays, coinsurance and deductibles). The vast majority of U.S. workers with job-based insurance coverage face copayments when they go to the doctor; 60% are in plans with a copayment of $15 or $20, and an additional 15% are in a plan with a copayment of $25 (1). Most covered workers also face cost-sharing of prescription drugs; 74% of covered workers in plans with prescription drug copay structures have at least three different copay levels.
Anyone familiar with the U.S. healthcare industry is aware that there is a major push towards preventive care, rather than reactive medicine. The basic assumption is simple; it is better to prevent an illness than to treat it. Not only is it better for the individual to not have to go through the risks and suffering associated with any illness, but it is also more cost-effective to prevent diseases rather than paying for treatments. If preventing illness is a major goal of healthcare, why have we built a healthcare model that creates disincentives for behaving in a manner that would more effectively prevent illness?
Reinforcement Theory is the theory that the consequences of actions will modify behavior (2). Unlike the more famous Classical Conditioning (3), Reinforcement Theory is not invoking a response based on a completely independent stimulus. Reinforcement Theory experiments show that you can increase the occurrence of an action or behavior by linking it with a positive outcome related to the action. Likewise, you can decrease the occurrence of an action or behavior by linking it with a negative outcome. Having to pay money for an action is a negative outcome. On the flip side, getting something for free would be considered a positive outcome.
Now if one accepts those contentions, it is easy to see how going to get a physical and being told by the receptionist that you “don’t owe anything” represents a positive outcome. And certainly if you are told that you owe $20, this would be a negative outcome. Therefore after you have gone to the doctor several times, Reinforcement Theory suggests that you would be less inclined to go to the doctor in the future if you are paying co-pays at the time of your visit.
Now a person would logically ask, if copays prevent the moral hazard (4) associated with “free” services, how do they raise costs? Before answering that question, the more widespread affects of Reinforcement Theory should be reviewed. For example, what happens when a person fills a prescription that their doctor prescribed for cholesterol medicine and is told that they owe $30?
Reinforcement theory tells us that the individual’s behavior will change. In fact, the person may stop taking their prescribed medicine. Reinforcement Theory suggests that if the reinforcement has a large enough impact, the reinforcement will begin to change behavior, such that one may stop filling this prescription. If this person has insurance coverage, but has minimal discretionary income, will the negative outcome of owing $30 have a larger effect than for someone with less financial constraints?
Let’s take this one step further. Assume that every time a primary care physician tells a person to go see a specialist, and the person follows these instructions, the outcome is that they owe $35 (a negative outcome). Now assume that every time the doctor tells them that you need to take a specific prescription, the outcome is that you owe $15 or $30 or $45 (another negative outcome). If the behavior that we look at is listening to your doctor, Reinforcement Theory shows us that behavior will be changed in such a manner that individuals will be less inclined to listen to their doctor. It does not make sense to build a system of care that undermines itself by disengaging individuals through structural barriers. Further, the negative reinforcements are more powerful for those with less discretionary income, meaning the system more negatively influences poorer individuals.
Now I would like to view this argument through the lens of Standpoint Theory. Standpoint Theory states that the social groups to which individuals belong shape experiences, knowledge, and behaviors (5). It is clear that different socio-economic classes will put varying weights on the negative consequence, or punishment, of having to pay for following the advice of their doctor. However, Standpoint Theory tells us that the psychological impact of behavior changing negative reinforcement will have varying effects upon different social groups. If a person has a great amount of trust of their doctor, the negative outcomes of listening to this doctor may have less of an effect on their behavior. However, within minority populations, where there is a well-documented, disproportionate distrust of the doctors and hospitals, these plan structures may have a compounding effect. Therefore, the structure of this insurance system is creating increased barriers for minority populations.
Another relevant article looked at the overall affects of using “carrots” versus “sticks” (6). This article discussed a lab-based experiment showing the different effects of rewards and punishments on an individual’s cooperation with others. One of the central findings of the experiment is that it shows that the amount of behavior change is greatest when individuals are punished for doing the wrong thing, versus when they receive rewards for doing the right thing. Applying this to the cost sharing structures used in the U.S. healthcare system, it suggests that the largest effect on individual’s behavior would be from any punishments. In the case of the healthcare system it is slightly different because the rewards are actually for doing nothing, and the punishments are for following the recommended action (e.g., filling a prescription or going to see a specialist). However, this study shows that “sticks” (punishments) play a large role in shaping individual behavior.
It is important to acknowledge that in healthcare there are clearly positive physical and psychological rewards for acting in a manner suggested by your healthcare provider. However, these rewards may be outweighed by the punishments for behaving in a manner that would bring about positive results. In addition, the rewards are not usually connected to the actions in a clear and immediate way. For example, if a person takes a maintenance prescription drug, the result may be that the person does not have any additional episodes of illness connected to that particular condition. However, the effect of their actions is not necessarily clearly linked to the action of taking the prescription drug. Likewise, if a group of employees from an insurance plan all increase their utilization of the health plan, their costs will go up. This is a negative effect but the individuals may not necessarily see the connection. That is not to suggest that they wouldn’t understand that there is a direct connection, but rather that they do not think in these terms. Unlike these rewards and punishments that are less closely linked to the actions, copays, deductibles and coinsurance at the time of care is clearly on account of the action of utilizing the health plan.
To understand the potential, and likely, implications of copays in healthcare, it is important to look beyond theory to actual practice. A study published in the Journal of the American Medicine Association in 2004 (7) shows that increasing the copays for prescription drugs lead to lower utilization of all classes of drugs. The results of this study raised concerns for the author about the negative affects on health outcomes of higher copays. Assuming that doctors prescribe prescription drugs for clinical reasons, the punishments associated with filling these prescriptions (increased costs) are having a negative effect on these individual’s health. This study does not apply any theories regarding the messages that these copays are sending.
However, it clearly shows that for a group of 528,969 privately insured individuals aged 18 to 64, there are negative impacts of paying for healthcare through these types of copays. It is important to note that the effect of the increased copays was not limited to any specific class of drugs that one might contend are ‘not as important’, or ‘less instrumental to the health of the users’. The biggest factor affecting the decline in prescription drug use was the amount the copay was increased. This shows that once the negative reinforcement reaches a certain threshold, it begins to change the individual’s behavior.
The last piece of evidence I will cite is anecdotal. Pitney Bowes implemented a program reducing copays for employees enrolled in their disease management programs. By encouraging these chronically ill individuals to heed the recommendations of their doctors, Pitney Bowes saved over $1 million dollars. By taking away the disincentives for these high-risk employees to take preventive steps (medications, doctors visits, etc.) they stayed healthier. Once the punishments were removed for acting in a productive manner, people were more inclined to behave in that way.
References
1. http://www.kff.org/insurance/7527/upload/7528.pdf
2. Skinner, B.F. (1969). Contingencies of Reinforcement: A Theoretical Analysis. Englewood Cliffs, NJ: Prentice-Hall.
http://wik.ed.uiuc.edu/index.php/Reinforcement_theory;
3. Pavlov, I. P. (1927). Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex (translated by G. V. Anrep). London: Oxford University Press. http://en.wikipedia.org/wiki/Classical_conditioning
4. Wood, J. T. (1982). Communication and relational culture: Bases for the study of human relationships. Communication Quarterly, 30, 75-82.
5. http://www.gladwell.com/2005/2005_08_29_a_hazard.html and
http://www.ingrimayne.com/econ/RiskExclusion/Risk.html
6. Andreoni et al, The American Economic Review, June 2003, Volume 93, No. 3. http://www.pitt.edu/~vester/AER2003.pdf
7. Pharmacy Benefits and the Use of Drugs by the Chronically Ill. Dana P. Goldman, PhD; Geoffrey F. Joyce, PhD; Jose J. Escarce, MD, PhD; Jennifer E. Pace, BS; Matthew D. Solomon, MA; Marianne Laouri, PhD; Pamela B. Landsman, MPH, DrPH; Steven M. Teutsch, MD, MPH. JAMA. 2004; 291:2344-2350.
Anyone familiar with the U.S. healthcare industry is aware that there is a major push towards preventive care, rather than reactive medicine. The basic assumption is simple; it is better to prevent an illness than to treat it. Not only is it better for the individual to not have to go through the risks and suffering associated with any illness, but it is also more cost-effective to prevent diseases rather than paying for treatments. If preventing illness is a major goal of healthcare, why have we built a healthcare model that creates disincentives for behaving in a manner that would more effectively prevent illness?
Reinforcement Theory is the theory that the consequences of actions will modify behavior (2). Unlike the more famous Classical Conditioning (3), Reinforcement Theory is not invoking a response based on a completely independent stimulus. Reinforcement Theory experiments show that you can increase the occurrence of an action or behavior by linking it with a positive outcome related to the action. Likewise, you can decrease the occurrence of an action or behavior by linking it with a negative outcome. Having to pay money for an action is a negative outcome. On the flip side, getting something for free would be considered a positive outcome.
Now if one accepts those contentions, it is easy to see how going to get a physical and being told by the receptionist that you “don’t owe anything” represents a positive outcome. And certainly if you are told that you owe $20, this would be a negative outcome. Therefore after you have gone to the doctor several times, Reinforcement Theory suggests that you would be less inclined to go to the doctor in the future if you are paying co-pays at the time of your visit.
Now a person would logically ask, if copays prevent the moral hazard (4) associated with “free” services, how do they raise costs? Before answering that question, the more widespread affects of Reinforcement Theory should be reviewed. For example, what happens when a person fills a prescription that their doctor prescribed for cholesterol medicine and is told that they owe $30?
Reinforcement theory tells us that the individual’s behavior will change. In fact, the person may stop taking their prescribed medicine. Reinforcement Theory suggests that if the reinforcement has a large enough impact, the reinforcement will begin to change behavior, such that one may stop filling this prescription. If this person has insurance coverage, but has minimal discretionary income, will the negative outcome of owing $30 have a larger effect than for someone with less financial constraints?
Let’s take this one step further. Assume that every time a primary care physician tells a person to go see a specialist, and the person follows these instructions, the outcome is that they owe $35 (a negative outcome). Now assume that every time the doctor tells them that you need to take a specific prescription, the outcome is that you owe $15 or $30 or $45 (another negative outcome). If the behavior that we look at is listening to your doctor, Reinforcement Theory shows us that behavior will be changed in such a manner that individuals will be less inclined to listen to their doctor. It does not make sense to build a system of care that undermines itself by disengaging individuals through structural barriers. Further, the negative reinforcements are more powerful for those with less discretionary income, meaning the system more negatively influences poorer individuals.
Now I would like to view this argument through the lens of Standpoint Theory. Standpoint Theory states that the social groups to which individuals belong shape experiences, knowledge, and behaviors (5). It is clear that different socio-economic classes will put varying weights on the negative consequence, or punishment, of having to pay for following the advice of their doctor. However, Standpoint Theory tells us that the psychological impact of behavior changing negative reinforcement will have varying effects upon different social groups. If a person has a great amount of trust of their doctor, the negative outcomes of listening to this doctor may have less of an effect on their behavior. However, within minority populations, where there is a well-documented, disproportionate distrust of the doctors and hospitals, these plan structures may have a compounding effect. Therefore, the structure of this insurance system is creating increased barriers for minority populations.
Another relevant article looked at the overall affects of using “carrots” versus “sticks” (6). This article discussed a lab-based experiment showing the different effects of rewards and punishments on an individual’s cooperation with others. One of the central findings of the experiment is that it shows that the amount of behavior change is greatest when individuals are punished for doing the wrong thing, versus when they receive rewards for doing the right thing. Applying this to the cost sharing structures used in the U.S. healthcare system, it suggests that the largest effect on individual’s behavior would be from any punishments. In the case of the healthcare system it is slightly different because the rewards are actually for doing nothing, and the punishments are for following the recommended action (e.g., filling a prescription or going to see a specialist). However, this study shows that “sticks” (punishments) play a large role in shaping individual behavior.
It is important to acknowledge that in healthcare there are clearly positive physical and psychological rewards for acting in a manner suggested by your healthcare provider. However, these rewards may be outweighed by the punishments for behaving in a manner that would bring about positive results. In addition, the rewards are not usually connected to the actions in a clear and immediate way. For example, if a person takes a maintenance prescription drug, the result may be that the person does not have any additional episodes of illness connected to that particular condition. However, the effect of their actions is not necessarily clearly linked to the action of taking the prescription drug. Likewise, if a group of employees from an insurance plan all increase their utilization of the health plan, their costs will go up. This is a negative effect but the individuals may not necessarily see the connection. That is not to suggest that they wouldn’t understand that there is a direct connection, but rather that they do not think in these terms. Unlike these rewards and punishments that are less closely linked to the actions, copays, deductibles and coinsurance at the time of care is clearly on account of the action of utilizing the health plan.
To understand the potential, and likely, implications of copays in healthcare, it is important to look beyond theory to actual practice. A study published in the Journal of the American Medicine Association in 2004 (7) shows that increasing the copays for prescription drugs lead to lower utilization of all classes of drugs. The results of this study raised concerns for the author about the negative affects on health outcomes of higher copays. Assuming that doctors prescribe prescription drugs for clinical reasons, the punishments associated with filling these prescriptions (increased costs) are having a negative effect on these individual’s health. This study does not apply any theories regarding the messages that these copays are sending.
However, it clearly shows that for a group of 528,969 privately insured individuals aged 18 to 64, there are negative impacts of paying for healthcare through these types of copays. It is important to note that the effect of the increased copays was not limited to any specific class of drugs that one might contend are ‘not as important’, or ‘less instrumental to the health of the users’. The biggest factor affecting the decline in prescription drug use was the amount the copay was increased. This shows that once the negative reinforcement reaches a certain threshold, it begins to change the individual’s behavior.
The last piece of evidence I will cite is anecdotal. Pitney Bowes implemented a program reducing copays for employees enrolled in their disease management programs. By encouraging these chronically ill individuals to heed the recommendations of their doctors, Pitney Bowes saved over $1 million dollars. By taking away the disincentives for these high-risk employees to take preventive steps (medications, doctors visits, etc.) they stayed healthier. Once the punishments were removed for acting in a productive manner, people were more inclined to behave in that way.
References
1. http://www.kff.org/insurance/7527/upload/7528.pdf
2. Skinner, B.F. (1969). Contingencies of Reinforcement: A Theoretical Analysis. Englewood Cliffs, NJ: Prentice-Hall.
http://wik.ed.uiuc.edu/index.php/Reinforcement_theory;
3. Pavlov, I. P. (1927). Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex (translated by G. V. Anrep). London: Oxford University Press. http://en.wikipedia.org/wiki/Classical_conditioning
4. Wood, J. T. (1982). Communication and relational culture: Bases for the study of human relationships. Communication Quarterly, 30, 75-82.
5. http://www.gladwell.com/2005/2005_08_29_a_hazard.html and
http://www.ingrimayne.com/econ/RiskExclusion/Risk.html
6. Andreoni et al, The American Economic Review, June 2003, Volume 93, No. 3. http://www.pitt.edu/~vester/AER2003.pdf
7. Pharmacy Benefits and the Use of Drugs by the Chronically Ill. Dana P. Goldman, PhD; Geoffrey F. Joyce, PhD; Jose J. Escarce, MD, PhD; Jennifer E. Pace, BS; Matthew D. Solomon, MA; Marianne Laouri, PhD; Pamela B. Landsman, MPH, DrPH; Steven M. Teutsch, MD, MPH. JAMA. 2004; 291:2344-2350.
posted by Elli Resnick at 4:59 PM
0 Comments:
Post a Comment
<< Home