June 2009
|
Treatment Algorithms Enable Effective Nurse- and Pharmacist-Directed Diabetes Care
The current medical care system fails patients with diabetes because of a lack of timely, appropriate clinical decisions. Specially trained nurses or pharmacists can facilitate better diabetes care.Reviewed By Mayer B. Davidson, MD
Evidence-based treatment recommendations for the management of patients with type 2 diabetes have been put forth by the American Diabetes Association (ADA) (Table 1). When put into practice, these recommendations can go a long way toward alleviating the devastating macro and microvascular complications of the disease.1 When considering guidelines, however, clinicians must differentiate between process measures and outcome measures. Process measures are the number of tests or examinations carried out per period of time or whether an indicated treatment is given, and outcome measures are the actual results of the test or the effect of the treatment. Meeting process measure goals, however, does not always translate into improvements in outcomes.2 This means that, for example, just because a patient has frequent A1C measures, this does not necessarily lead to lowered glycemia.
Mayer B. Davidson, MD, Program Director for the Center for Clinical Research Excellence in Diabetes and Metabolism, Charles Drew University of Medicine and Science, Los Angeles, discussed the use of detailed treatment algorithms in a recent article in The Diabetes Educator (2009;35:61–71). The following is a summary of that report as well as comments from Dr. Davidson's editorial in Diabetes Care (2009;32:370–372).
DIABETES CARE GOALS
Most type 2 diabetes patients do not meet recommended goals.2 In a recent review, we found that 21% to 43% of patients had A1C levels >9.5%, only 22% to 46% met the low-density lipoprotein cholesterol (LDL-C) goal, and 29% to 33% met the blood pressure goal. With regard to the ADA's combined goals for glycemia, lipids, and blood pressure, just 2% to 10% of patients were on target. Approaches to improve care, such as appointment reminders; keeping the treating physician in the loop with regard to the patient's progress; case management approaches; physician education; as well as multifaceted quality improvement interventions in the practice setting have not been successful.
Why? There are two critical barriers to good diabetes care: (1) the lack of time physicians have to spend with patients and (2) the lack of timely, appropriate clinical decisions or clinical inertia. A primary care physician typically has 10 to 15 minutes with each patient. Studies have shown that physicians pinpoint the lack of time as a primary obstacle to meeting clinical recommendations. Most patients are asymptomatic, and their care revolves around preventing of complications by glycemic control, managing lipids and blood pressure, and carrying out other process measures of diabetes care (Table 1). Therefore, other issues, especially those related to patient symptoms, receive the physician's attention. Furthermore, when patients are only seen every 3 months, glycemia, lipids, and blood pressure can be out of control for relatively long periods of time even if over target values had begun to be treated.
Clinical inertia is a major factor for poor outcomes in diabetes care patients.2 Reports show that therapy is intensified only 20% of the time when patients' A1C levels are >8.0%. It took >3 months for changes to be made, and in one study it took 2.5 years before metformin was added in patients who were uncontrolled on a sulfonylurea. On average, A1C levels were >9.0% before the next step of drug intensification occurred. Among those patients not receiving lipid-lowering agents, only 5.6%, 8.7%, and 15.4% had started therapy when LDL-C was >100 mg/dL, >130 mg/dL, or >160 mg/dL, respectively. It's a similar story among diabetes patients with regard to undertreatment of hypertension—only 10%, 15%, and 14% had antihypertensive therapy initiated when their blood pressures were >130/80 mm Hg, >140/90 mm Hg, or >150/100 mm Hg, respectively.
I have been supervising nurses and pharmacists to use detailed treatment algorithms I developed (with the help of Anne L. Peters, MD) and refined during the past 20 years. Trained nurses and pharmacists use these treatment algorithms to make independent therapeutic decisions for diabetes patients in various clinical settings. When compared with the usual care these patients receive, algorithm-based results have been much better.3-7 Most recently in a Los Angeles County community health center,7 at enrollment 361 randomized patients had a mean A1C of 8.8%, 17% met the ADA goal of <7%, and 50% met the ADA LDL goal. After 1 year of treatment using the algorithms (Tables 2 and 3), mean A1C was 7.0%, 60% of patients met the ADA glycemic goal, and 82% met the lipid goal.
The algorithms reflect Los Angeles County's limited formulary. The outcomes were accomplished using metformin, glipizide, glyburide, regular and neutral protamine Hagedorn (NPH) insulin, gemfibrozil, and a statin. Most providers, however, do not have the drug limitations reflected in Tables 2, 3, and 4.
DRUG SELECTION CRITERIA
Four criteria should be considered when selecting an agent over its competitors: (1) effectiveness, (2) side effects, (3) adherence issues, and (4) cost. With the exception of insulin—the most effective if used appropriately—no drug or class of drugs clearly stands out as more effective. Metformin was selected as the initial drug of choice at a recent ADA consensus conference8 not because it was more effective, but because of its lack of serious side effects and weight gain and low cost.
Sulfonylurea is added when patients are not adequately controlled on metformin because it is inexpensive, it is only rarely associated with serious hypoglycemia, and it causes only mild weight gain. Thiazolidinediones (TZDs) are effective in many patients when added as a third-line drug.9 There are three reasons to initiate an oral dipeptidyl dipeptidase (DPP)-4 inhibitor as the third drug when a TZD is contraindicated: (1) a study10 found sitagliptin (Januvia, Merck & Co., Inc.) effective when added to metformin plus sulfonylurea; (2) if the oral DPP-4 inhibitor is not used the next choice is an injectable agent, which presents possible adherence issues; and (3) the maximal effect is easily and quickly ascertained (ie, fasting plasma glucose concentrations are maximally lowered by 3 weeks), therefore, patients not achieving control with the DPP-4 inhibitor will be quickly identified.
COMPLIANCE
The injectable glucagon-like peptide–1 analog exenatide (Byetta, Amylin Pharmaceuticals, Inc. and Eli Lilly and Company) is not considered earlier in the treatment algorithm because of adherence issues and cost. But it is associated with weight loss, which may make it an attractive second-line option for some patients. If A1C is not <7.5% by 4 months after initiation, however, it should be discontinued and bedtime insulin initiated. Insulin is reserved as the final drug of choice because of adherence issues. Not only does it require self-administered injections, but it also requires self-monitoring of blood glucose (SMBG) and it is associated with potential hypoglycemia. Bedtime insulin plus oral antidiabetic agents is the first insulin regimen attempted; two or more insulin injections are used last.
A1C levels of 7.0% to 7.5% are tolerated before insulin is either started or intensified because of the aforementioned adherence issues. When insulin is started, patients undergo a disruption of lifestyle because of the requirement for SMBG and the increased possibility of hypoglycemia with even further disruptions if intensification of insulin therapy occurs. Five studies in >2,000 type 111-13 and type 214,15 diabetes patients showed virtually no development or progression of retinopathy and nephropathy over 4 to 9 years when mean A1C levels were <7.0%, and the risk did not increase much with A1C between 7% and 7.5%.
HYPERTENSION TREATMENT
The hypertension algorithm (Table 4) was not developed when we evaluated nurse-directed care at the Los Angeles County Community Health Center,7 but it is now used. The ADA goals for systolic and diastolic blood pressure (Table 1) are being met in 60% and 90%, respectively, of more than 150 patients. This compares much more favorably than the 28% to 36% of patients in the literature who met ADA systolic blood pressure goals.16
Angiotensin-converting enzyme inhibitors or aldosterone receptor antagonists are the preferred first-line agents, and thiazide diuretics are the preferred second-line drug (when serum creatinine concentration is
<1.8 mg/dL or the estimated glomerular filtration rate is >50 mL/min).16 Note that beta-blockers are contraindicated in the setting of a dihydropyridine calcium channel blocker because the combination depresses cardiac function.
Algorithms improve diabetes care when providers have the time and knowledge to make appropriate, timely clinical decisions. Approved treatment algorithms when put into practice by trained and supervised nurses or pharmacists provide both the knowledge for and the timing of appropriate therapeutic decisions. This has been demonstrated in at least 19 randomized clinical trials, 14 of which used nurses and five used pharmacists, comparing their more favorable outcomes of care with usual care.2
The improvement in outcomes among diabetes patients whose care is directed by nurses and pharmacists that can make treatment decisions dictated by approved algorithms is clear.2 When this model of diabetes care is followed, the nurse or pharmacist is almost as important of a team member as the patients themselves. Because the algorithms used have been approved by the supervising physician, the patient is receiving recommended care. Treating patients on insulin is particularly challenging for busy physicians, and knowledgeable nurses and pharmacists can be especially effective in working with these patients. Diabetes education is certainly necessary for successful outcomes, and nurses and pharmacists have the opportunity to reinforce it in the specific areas required by the patient (in addition to general background information). Considering all that is involved in treating and educating people with diabetes, it is really not surprising that outcomes are so significantly improved under nurses and pharmacists who can spend more time with patients and provide the appropriate care in a timely manner.
RESPONSE FROM THE AADE
"Current literature supports a team approach for care of patients with diabetes," said Evan M. Sisson, PharmD, MSHA, CDE, from the Virginia Commonwealth University School of Pharmacy. "As the number of people with diabetes increases, so will the imperative for efficient delivery systems that match professional expertise with patient health needs."
Dr. Sisson is a member of the Board of Directors of the American Association of Diabetes Educators (AADE).
"The model described by Dr. Davidson decreases physician workload and increases patient access to care by delegating medication management authority to appropriately trained nurses and pharmacists through a diabetes treatment algorithm. Certified diabetes educators are highly specialized health professionals (including nurses, pharmacists, and dietitians) whose role on the team is to spend the necessary time to improve diabetes self-care behaviors. Because of the intense specialization, diabetes educators are also valuable resources to ensure attainment of diabetes therapy goals, especially when given authority to make appropriate treatment decisions. Nurse practitioners have at least limited prescriptive privileges in most states. Several state medical boards also promote pharmacist-initiated medication therapy changes via collaborative practice agreements between physicians and pharmacists as a means to extend the reach of physicians to the communities they serve. The AADE supports these innovative practice models that improve care and self-management education for all patients with diabetes."
Mayer B. Davidson, MD, is Program Director for the Center for Clinical Research Excellence in Diabetes and Metabolism, Charles Drew University of Medicine and Science, Los Angeles. He may be reached at mayerdavidson@cdrewu.edu.
|
