Skip to main content
. Author manuscript; available in PMC: 2021 Sep 1.
Published in final edited form as: Am J Manag Care. 2021 Mar 1;27(3):e80–e88. doi: 10.37765/ajmc.2021.88602

Table 4.

Choice of Second Medication after Starting Metformin or a Sulfonylurea in Traditional Medicare and Medicare Advantagea

Metformin First SU/Metformin Second TZD Second Insulin Second Other Second

TM MA Diff TM MA Diff TM MA Diff TM MA Diff
Metformin First 53.7% 61.5% −7.8% 5.4% 5.6% −0.2% 11.2% 10.8% 0.4% 29.7% 22.0% 7.7%

Region + + +

Northeast 49.1% 56.9% −7.8% 3.5% 3.6% −0.1% 9.3% 10.0% −0.7% 38.1% 29.5% 8.6%
Midwest 61.4% 62.8% −1.4% 4.6% 4.9% −0.3% 11.6% 12.5% −0.8% 22.4% 19.9% 2.5%
South 54.8% 62.7% −8.0% 6.0% 5.4% 0.6% 10.7% 10.8% −0.1% 28.5% 21.0% 7.5%
West 50.2% 62.4% −12.2% 6.6% 8.2% −1.7% 13.3% 10.4% 2.9% 30.0% 19.0% 11.0%

Plan Type

Established 54.4% 60.5% −6.1% 5.3% 5.7% −0.5% 11.6% 11.0% 0.7% 28.7% 22.8% 5.9%
New 53.3% 62.1% −8.7% 5.5% 5.6% −0.1% 11.0% 10.7% 0.2% 30.2% 21.6% 8.6%

Plan Size + +

Small 52.9% 63.4% −10.5% 5.4% 5.3% 0.1% 11.3% 10.2% 1.1% 30.3% 21.1% 9.2%
Large 54.4% 59.7% −5.3% 5.4% 5.9% −0.6% 11.0% 11.4% −0.4% 29.2% 22.9% 6.3%

Sulfonylurea First 58.6% 62.5% −3.9% 5.5% 6.0% −0.5% 14.3% 14.3% 0.0% 21.6% 17.2% 4.4%

Region + +

Northeast 59.1% 62.2% −3.1% 3.5% 3.9% −0.4% 12.7% 12.6% 0.2% 24.6% 21.4% 3.3%
Midwest 60.3% 60.3% 0.0% 5.8% 5.8% 0.0% 16.2% 17.6% −1.4% 17.7% 16.3% 1.4%
South 56.9% 61.7% −4.8% 5.6% 6.4% −0.8% 14.7% 14.5% 0.2% 22.8% 17.4% 5.4%
West 60.7% 66.4% −5.7% 6.8% 7.1% −0.4% 13.4% 12.9% 0.5% 19.1% 13.5% 5.6%

Plan Type

Established 60.7% 62.3% −1.5% 5.2% 5.7% −0.5% 14.5% 14.6% −0.2% 19.7% 17.5% 2.2%
New 57.6% 62.7% −5.0% 5.7% 6.2% −0.5% 14.2% 14.2% 0.1% 22.5% 17.0% 5.5%

Plan Size + +

Small 57.2% 62.3% −5.1% 5.4% 6.1% −0.6% 14.8% 14.4% 0.4% 22.6% 17.3% 5.4%
Large 60.1% 62.8% −2.7% 5.6% 5.9% −0.4% 13.9% 14.2% −0.3% 20.5% 17.1% 3.4%
a.

TM estimates are weighted to match MA distributions by age category, sex, race/ethnicity categories, and geography. Overall MA-TM differences (first rows) are statistically significant at p<.0001 for SU/metformin and other. For those who started an SU first, there were no significant differences for starting a TZD or insulin second. For those who started metformin first, the difference for those starting insulin second was significant (p<.01). Significant differences by region, plan age, and plan size are noted by (+), signifying p≤.01).

Other includes GLP-1 agonists, DPP-4 inhibitors, SGLT-2 inhibitors, Meglitinides, and D-Phenylalanine Derivatives.