Exploring Medicare Advantage Star Ratings Over Time

I used the Medicare Advantage (MA) star rating data published from 2019 to 2023 to examine the dynamics of the MA plan ratings.

A quick overview — MA plans are offered in the US to those aged 65 and older. These plans are evaluated annually by the Centers for Medicare & Medicaid Services (CMS). CMS then publishes the assessments using a star rating system, which ranges from 1 to 5 stars, with a rating of five stars denoting the highest performance. Importantly, these ratings carry significant financial implications as they influence the funding received by each plan.

Only plans rated for the past five years were included (N = 354 plans). When I first began playing with this data, I had the assumption of a relatively bi-modal market, where specific plans consistently achieved high ratings across the years (i.e., 4+ stars) and others consistently low (i.e., equal or below 3 stars), with few plans in-between. First, I plotted the scores of all plans together, expecting nothing more than an aesthetic presentation (see Figure 1). However, an immediate pattern emerged — no plan scored 2.5 stars for more than two consecutive years. For example, plans that scored 2.5 stars in 2020 and 2021, either improved or dropped out. This suggested that plans with consistently low scores might be a smaller portion of the population than I initially expected.

Figure 1. Star rating over the years, 2020–2024

To quantify this, I calculated the fraction of plans that consistently received 4 or more stars over five consecutive years, in contrast to those that consistently held a rating of 3 stars or lower during the same period, as well as all other plans. While about 30% of plans maintained a rating of 4 stars or higher over the last five years, only about 1% consistently received low ratings (see Figure 2). The majority of plans, approximately 70%, exhibited fluctuating star ratings over the same time period.

Figure 2. Proportion of plans with consistent high star rating across five years (>=4 stars), consistent low star rating (<= 3 stars) and dynamic rating.

This observation indicates that most plans experience variable ratings, potentially leading to significant year-to-year changes in their funding. To further explore the ratings stability, I next investigated the consistency of their performance, namely how many of these plans had high ratings (4 or more stars) in x years out of the past five years.

Figure 3. Proportion of plans that scores >= 4 stars for x years

The results, depicted in Figure 3, reveal that 50% of plans experienced variability in their ratings, alternating between years of qualifying for additional funding and years of not meeting the threshold. About 30% of the plans consistently achieved more than 4 stars over the five years (aligned with the results shown in Figure 2), and about 19% of plans never reached a 4-star rating.

Overall, these results highlight the dynamic and sometimes unpredictable nature of MA plan performance, and especially underscores the opportunities available for plans to adapt and excel.