A Shot in the Arm for Medicare Part D: Four Ways for the Government to Boost its Customer Communications

Jeffrey R. Kling,
Jeffrey R. Kling Former Brookings Expert, Associate Director for Economic Analysis - Congressional Budget Office
Sendhil Mullainathan,
Sendhil Mullainathan Roman Family University Professor - University of Chicago Booth
Eldar Shafir, Lee Vermeulen, and Marian V. Wrobel

November 20, 2008


A new study has found that seniors are more likely to switch Medicare healthcare plans and save substantial sums when they study personalized cost information. A typical senior with between four and six prescriptions could save about $500 by changing from their existing plan to the most cost effective plan currently available for someone taking those same drugs, according to the team of researchers at Harvard, Princeton, Brookings, Wisconsin and the National Bureau of Economic Research.

Given the overall state of the economy and recent news that the top 10 drug plans will increase their monthly premiums from 8 percent to 64 percent, the study is particularly timely since it can help inform the 24 million seniors in the program to shop around for the best price as open enrollment for Medicare Part D drug plans begins this month. Premiums are actually only one-fifth of the total cost, highlighting the importance of looking at personalized costs. The study examined plan choices and costs during 2007, and savings may be even greater for 2009 due to the large changes in prices made by plans.

Subsidized drug coverage for Americans with Medicare began with much fanfare in January 2006. Beneficiaries enroll voluntarily and choose among multiple free-standing private drug plans and multiple Medicare HMOs offering drug coverage. Most beneficiaries can only enroll or change plans during an open enrollment period, which occurs for the last six weeks of every year.

A key reason why there is so much potential savings, according to the study, is that costs depend on the match between the drugs people take and the plan they are in. Most seniors do not look at the costs for their specific drugs, since most information is derived from mailings and other sources that do not tailor the data to fit individual needs. Insurance premiums, while the most obvious part of the cost, are not as important as the cost of filling each prescription. For every $1 paid in premiums, the average senior pays another $4 in other costs for each drug purchased. These other costs differ depending on the insurance provider and plan, even for exactly the same medications.

The Ideas42 research team conducted a randomized experiment in which some seniors received a carefully designed mailing including information from the Medicare website on the potential savings from changing Medicare drug plans while others were encouraged to go to the website themselves. The purpose of the study was to understand whether seniors were already making good choices – defined as choices that they themselves would not change if they had more information – and the effect of information on choices made.

The study assessed the effect of information on several key dimensions: rates of changing plans, predicted costs in the new plan (based on information known at the time of plan selection), actual costs in the new plan (based on actual drug use in the following year), satisfaction, and other measures of plan quality. Separately, the study also examined seniors’ knowledge of various aspects of the Medicare drug plan program and the quality of information available from various sources.

The background research on knowledge indicated that the majority of seniors were not well-informed about drug plans and did not seek personalized comparative information, yet most were still satisfied with the plan they had chosen. While the Medicare help-line consistently and effectively offered personalized information about the cost differences among plans, other market players did not promote this source and its value, nor, as a general rule, did they offer comparable services.

Baseline data collected for the study in the fall of 2006 indicated that the cost differences among plans were substantial, and most seniors had access to many plans with costs lower than those in the current plan selected. For example, a typical senior with between four and six prescriptions could have saved about $500 by changing from his current plan to the lowest cost plan; if that plan did not appeal to him, there were more than 20 other plans offering some savings relative to the plan he was in.

Twenty-eight percent of seniors who received the mailing that highlighted the potential savings from changing plans elected a new plan for 2007 compared to only 17 percent of seniors who were not sent such information. On average, based on the drugs seniors were taking at the time of plan choice, the seniors who received the mailing saved $90 per person compared to seniors in the other group; this equates to 6 percent of total drug costs. Zeroing in on the seniors who made different plan choices as a result of the intervention, the study found that the average per-person projected savings were at least $200 or 13 percent of total drug costs. These finding are important because they indicate both that, under existing conditions, seniors were not making robust decisions, and that, if presented with appropriate information, they would take action to reduce their predicted costs.

In addition, data collected in 2008 indicated that not only did the informational mailing lead to greater predicted savings (based on drug utilization recorded at the time of plan choice), it also led to greater actual savings, based on the drugs seniors actually took over the next year. As one might expect, the effect on actual savings was not as dramatic as the effect on predicted savings: 4 percent for the information group as a whole, and at least $150 or 9 percent for those who made different choices because of the intervention. Importantly, there were no differences between the two study groups in terms of seniors’ reported satisfaction or access to medications or in plan quality as measured by Medicare. These second-year findings are important because they indicate that the cost information led to reductions in costs without reductions in quality.

The researchers conclude that many seniors were not fully grasping the price differences among Medicare drug plans and that additional efforts to distribute information could lead to significant reductions in Medicare beneficiaries’ costs. In the immediate term, these findings should inspire seniors to call Medicare (1-800-MEDICARE), or visit its website (, and request a personalized report on costs in different drug plans. The effort is minor and the potential gains significant.

Read the background on the study »

Read the paper »

Additional background on the research team: Ideas42 was established in June 2008 as a Social Science Research and Development laboratory at Harvard University with the goal of using scientific insights to design innovative policies and products both domestically and internationally. The origin of the name is from The Hitchhiker’s Guide to the Galaxy, where the brilliant computer Deep Thought is tasked with finding the meaning of life. “The Answer to the Great Question… Of Life, the Universe and Everything…is 42…I checked it very thoroughly,” it says, “and that quite definitely is the answer. I think the problem, to be quite honest with you, is that you’ve never actually known what the question is.” Thus, Ideas42’s mission is to ask the right questions looking at them through the lens of psychology and economics, and provide credible evidence about the answers.