Switching to Human Insulin Not Harmful in Type 2 Diabetes

Miriam E. Tucker

January 29, 2019

Switching patients with type 2 diabetes from insulin analogs to human insulin could save a lot of money, without causing harm, new research suggests. 

Findings from the large retrospective cohort study of Medicare beneficiaries were published in this week's issue of JAMA by Jing Luo, MD, of the Program on Regulation, Therapeutics, and Law, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, and colleagues. 

The investigators examined differences in glycemic control and costs in the 3 years before (2014), during (2015), and after (2016) a health plan-based intervention in which insulin-treated patients were encouraged to switch from analog to human insulin.

Overall, there was a small, statistically — but possibly not clinically — significant increase in HbA1c, with no difference in the rate of severe hypoglycemia or hyperglycemia.

At the same time, expenditures dropped considerably, resulting in significantly fewer Medicare beneficiaries reaching the Part D coverage gap.   

Kasia J. Lipska, MD, of the Section of Endocrinology, Yale School of Medicine, New Haven, Connecticut, who has conducted similar studies, writes in an accompanying editorial that analogs have largely replaced human insulin in patients with type 2 diabetes in the United States, despite a lack of data that they provide significant benefit and have a 10-fold greater cost.

"These findings should prompt physicians and patients to reconsider which type of insulin is best. Human insulin may not be the optimal choice for everyone, but it could be a solution for many patients with diabetes," she writes.

"On the individual patient level, use of human insulin may minimize out-of-pocket spending, and, on the healthcare system level, it may allow insurers to maximize the value of diabetes care."

However, Lipska cautioned that some patients may not do well switching to human insulin. This includes those with type 1 diabetes or who are otherwise at increased risk for hypoglycemia, noting, "Any program designed to promote switching to human insulin should take individual patient needs into account and allow for flexibility with respect to the choice of insulin."

Little Impact on Glucose Control

Data for the study came from a Medicare Advantage plan and medical group, which has members in California, Arizona, Nevada, and Virginia. The participants were 14,635 plan members who filled a total of 221,866 insulin prescriptions in 2014-2016.

From February through June 2015, a program was rolled out that encouraged switching from higher-cost insulin analogs to lower-cost human insulin (either premixed human 70/30 or NPH insulin). The protocol was led by plan pharmacists and supported by nurse practitioners, physician assistants, and physicians with experience in chronic disease management. 

Participants were a mean age of 72.5 years, 51% were women, and 93.1% had type 2 diabetes.

The proportion of all insulin prescriptions that were for insulin analogs dropped from 89% in the baseline period to 30% post-intervention, while human insulin prescriptions rose from 11% to 70%.

From a mean HbA1c of 8.46% at baseline, the start of the intervention was associated with a 0.14 percentage point increase (P = .003) and a slope change of 0.02 percentage points (P < .001) per month.

But by the end of the year following the intervention, neither change in HbA1c nor slope per month differed significantly from baseline (P = .09 and P = .81, respectively).

There were no significant differences in rates of serious hypoglycemia or hyperglycemia throughout the study period.

Costs Substantially Lowered, Doughnut Hole Escaped   

Overall, expenditures for insulin decreased by more than 50%, from approximately $3.4 million per month in December 2014 to $1.4 million per month in December 2016.

While total monthly expenditures for analog insulin increased from $2,226,389 in January 2014 to a high of $3,214,437 by December 2014, those costs reached a low of $515,875 by December 2016. At the same time, human insulin expenditures increased from $160,233 in January 2014 to a high of $916,286 by December 2016.

The proportion of patients who reached the Medicare Part D coverage gap, also known as the "doughnut hole," decreased from 20.6% in 2014 to 11.1% in 2016 (hazard ratio, 0.45; P < .001).

How Did We Get Here?

In her editorial, Lipska observes that a vial of 1000 U of NPH or regular human insulin can be purchased for $25, whereas the retail price for a vial of analog insulin ranges between $178 and $320.

"It is difficult to imagine that consumers in almost any other setting would be willing to pay such a significant price difference without evidence of concomitant benefit. Yet, when it comes to insulin, patients, physicians, and insurers have been willing to spend increasing amounts on these products."

Lipska says there are several reasons this has come about, including the appeal of novelty, postmarketing studies that appeared to show benefit for the analogs, and marketing efforts that were "clearly successful."

On the other hand, she believes, if a switch-back program such as the one in the current study were widely implemented, "the potential financial savings for patients and insurers could be quite substantial."

The study was funded by the Laura and John Arnold Foundation with additional support from the Engelberg Foundation, Harvard-MIT Center for Regulatory Science, and Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital. Luo is a consultant to Alosa Health and Health Action International. Lipska works under contract for the Centers for Medicare & Medicaid Services to develop and maintain performance measures used for public reporting and receives support from the National Institute on Aging and American Federation of Aging Research through the Paul Beeson Career Development Award.

JAMA. 2019;321:350-351, 374-384. Abstract, Editorial

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....