Appropriate Prescribing for Patients With Diabetes at High Risk for Hypoglycemia: National Survey of Veterans Affairs Health Care Professionals

Evidence is accumulating that older individuals with diabetes mellitus have little to gain from the treatment burdens of stringent blood glucose control.^1,2 In addition to concerns about increased mortality with tight control,¹ some older patients with diabetes may also be at risk for hypoglycemia-related harms from medications prescribed to meet standard hemoglobin A_1c (HbA_1c) targets.³ This problem has motivated patient safety campaigns that cue health care professionals to limit medications for certain older patients (eg, those with an HbA_1c level <7.5%, renal disease, or dementia) (to convert HbA_1c to a proportion of total hemoglobin, multiply by 0.01).⁴ In this study, we examined beliefs of primary care health-care professionals (PCPs) to anticipate how PCPs might receive such recommendations.

We surveyed a national random sample of practicing nontrainee Department of Veterans Affairs (VA) PCPs, including physicians, nurse practitioners, and physician assistants. The study, including a waiver of signed informed consent, was approved by the institutional review board of the Ann Arbor VA Healthcare System and was conducted from October, 6, 2014, to December, 8, 2014. Participants answered questions about practice characteristics, performance incentives, beliefs about decreasing use of inappropriate services, and demographics. They also received a scenario about a 77-year-old man with long-standing type 2 diabetes mellitus at high risk for hypoglycemia (HbA_1c level, 6.5%; severe kidney disease; and receiving glipizide, 10 mg, twice daily). Barriers to and facilitators of medication deintensification were identified using statements answered on a 4-point scale (strongly disagree to strongly agree) (Table 1). In addition, participants were asked to rate the level of difficulty they anticipated in following the Choosing Wisely recommendation to “avoid using medications other than metformin to achieve HbA_1c less than 7.5% in most older adults.”^5(p1) Data were analyzed from March 18, 2014, to April 2, 2014. We used logistic regression to identify PCP and practice setting characteristics associated with anticipated difficulty following the Choosing Wisely HbA_1c recommendation.

Of 1222 eligible PCPs, 594 returned usable surveys (48.6% response rate; numbers vary due to item nonresponse). Of these, 311 (53.0%) were women, 138 (23.4%) were nurse practitioners, 46 (7.8%) were physician assistants, and 405 (68.8%) were physicians.

A total of 217 PCPs (38.6%) thought that the 77-year-old patient at high risk for hypoglycemia would benefit if his HbA_1c level was maintained below 7.0%, and 252 participants (44.9%) reported that they would not worry about potential harm from tight control. In addition, 236 PCPs (42.1%) would worry that deintensification in this context (HbA_1c level, 6.5%) would lead to an HbA_1c level that is outside of current performance measures; 132 of the participants (23.5%) worried that deintensification could leave them vulnerable to future malpractice claims. Table 1 presents participant responses to all scenario questions.

A total of 161 of 562 PCPs (28.7%) agreed it would be somewhat or very difficult to follow the Choosing Wisely HbA_1c recommendation for older adults. The PCPs who agreed that maintaining the HbA_1c level below 7.0% would benefit the patient and who reported worrying about malpractice claims were more likely to report difficulty following the HbA_1c recommendation in the final adjusted regression model (P = .02). Conversely, PCPs who reported worrying that the patient would be harmed with tight blood glucose control were less likely to report difficulty following the HbA_1c recommendation (P = .04) (Table 2).

Almost half of VA PCPs reported that they would not worry about harms of tight control for an older patient with an HbA_1c level of 6.5% who is at high risk for hypoglycemia. A similar proportion were concerned about not meeting performance measures if they reduced the glipizide dosage, even though the VA never adopted a performance measure targeting an HbA_1c level less than 7.0%. Nearly one-quarter would even be concerned about liability with deintensification of hypoglycemic medications.

The VA has recently launched a hypoglycemia safety initiative to decrease overtreatment among veterans.⁶ Such efforts are needed outside the VA as well.⁷ To overcome provider misperceptions about the benefits of stringent blood glucose control and concerns about negative repercussions following deintensification of therapy, safety initiatives should be national and span multiple practice settings (eg, the Million Hearts campaign). In addition, national guidelines that clarify when to deintensify medication use and balanced performance measures that incentivize appropriate intensification and deintensification could improve prescribing practices and prevent many adverse events in older patients with diabetes.

Corresponding Author: Tanner J. Caverly, MD, MPH, Ann Arbor Veterans Affairs Center for Clinical Management Research, 2215 Fuller Rd, Ann Arbor, MI 48105 (tcaverly@med.umich.edu).

Published Online: October 26, 2015. doi:10.1001/jamainternmed.2015.5950.

Author Contributions: Dr Caverly had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Caverly, Fagerlin, Kirsh, Kullgren, Prenovost, Kerr.

Acquisition, analysis, or interpretation of data: Caverly, Zikmund-Fisher, Kullgren, Prenovost, Kerr.

Drafting of the manuscript: Caverly, Prenovost.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Caverly, Prenovost.

Obtained funding: Kirsh, Kerr.

Administrative, technical, or material support: Caverly.

Study supervision: Caverly, Fagerlin, Kullgren, Kerr.

Conflict of Interest Disclosures: Dr Kullgren reported having consulted for SeeChange Health and HealthMine. No other disclosures were reported.

Funding/Support: Dr Caverly was supported by the Veterans Affairs (VA) Advanced Fellowship Program in Health Services Research and Development (HSR&D). Dr Kerr was supported in part by the Veterans Health Administration’s PACT Demonstration Laboratory and by grant DIB 98-001 from the VA Diabetes Quality Enhancement Research Initiative. Dr Kullgren is the recipient of the VA HSR&D Career Development award at the VA Medical Center, Ann Arbor, Michigan.

Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data. Dr Stephan Fihn, Veteran's Health Administration PACT Demonstration Laboratory Coordinating Center, reviewed the manuscript prior to submission. The funding organizations otherwise had no role in the preparation and decision to submit the manuscript for publication.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.