Improving Diabetic Management:
Why Should Our Facility Undertake This Project?
1) Improves quality of life by not restricting diets, which has been shown to have minimal if any impact on medical management. Patients will not complain that they don’t get certain foods; helps discourage “cheating” out of frustration, and will improve the dining experience.
2) Families learn that restricted diets are unnecessary and decrease quality of life
3) Reduces stress (and potentially expense) on dietary staff. Do not have to worry about the wrong patient getting the wrong diet, makes meal preparation easier, causes less staff stress explaining to persons why they can’t have certain food.
4) Reduces nursing staff time measuring accuchecks and adjusting sliding scale insulin, which is not physiologic nor designed for chronic administration (time expense example: 5 min per patient, 4xday, 12 person on SSI = 4 hours/day nursing time!).
5) Reduces medication error in administration of insulin by having set amounts of insulin for each resident every day (not having to adjust amounts given).
6) Reduces risk of hypoglycemia and associated morbidity and mortality, by setting realistic control goals which are geriatric-evidence based.
7) Can reduce polypharmacy by setting realistic goals and avoiding excessive medication.
8) Reduces need for accuchecks by improving overall diabetic control and setting realistic goals, which saves nursing time and patient expense.
Net Result: Better Resident Quality of Life, Better Medical Care, Less Errors, Reduced Nursing / Dietary Time and Cost
Date]
[Addressee]
[Street Address]
[City, State Zip]
RE: Changes in Diet at [Your NH]
Dear Family Member:
I wanted to inform you of recent changes to our diets at [your NH].
There is increasing geriatric evidence that restricted diets do not benefit older people the same way they do when they are younger. For example, diabetic diets have not been shown to offer any benefit to the overall control of blood sugars or the long-term consequences of diabetes such as heart attack or stroke in older patients. Instead, restricted diets tend to promote weight loss, malnutrition, and other undesired consequences in the advanced age of most of the patients we deal with. Geriatric experts now recommend a balanced diet with appropriate amounts of protein, carbohydrates, and vegetables, such as we already serve at this facility. Our diets are prepared under the supervision of a registered dietician and are well within the calorie requirements of generally-accepted geriatric diabetic guidelines.
Another example is sodium restriction in people with high blood pressure. Sodium restriction at an advanced age has not been shown to be of benefit in controlling high blood pressure in the vast majority of persons. The sodium in our diets is balanced and is quite acceptable to the majority of people with heart disease and congestive heart failure. Excessive restriction of sodium tends to promote loss of interest in food and weight loss.
There will always be a few exceptions. And while we can in special cases provide a restricted/special diet, we are forthwith going to utilize our regular registered-dietician-supervised and geriatric-appropriate diet for all patients with diabetes mellitus, hypertension, and congestive heart failure. Exceptions will be made on an individual basis after consulting with the physician, if a request is made for such a diet.
If you have questions regarding this diet, you can contact your physician to determine whether this diet is indeed appropriate for your loved ones. Our goal is to provide a nutritious geriatric-appropriate diet for patients without overly restricting their choices and hindering their quality of life.
Sincerely yours,
Medical Director, [your NH]
Dear Dr.
Management of diabetes is often substantially different in frail elderly persons.
It is difficult to extrapolate conclusions from studies of community-dwelling adults with type 2 diabetes to the frail elderly in long-term care facilities. It is extremely important to take patient preferences and values into account when considering approaches to diabetes management (AMDA Diabetes Clinical Practice Guidelines 2009).
As Medical Director, I am concerned about four particular circumstances:
Diabetic diets
All geriatric guidelines now note there is no improvement in overall control of blood sugars, morbidity or mortality when diabetic diets are utilized in LTC. Diets are balanced by the staff dieticians and already contain the proper amounts of nutrients and calories. If weight loss is desired, reduced portion sizes can be arranged. For those that cheat incessantly on sweets, the issue can addressed by other means – it is not their regularly ordered diet that is the issue.
Residents on bid to qid accuchecks who are well controlled
There is no evidence to suggest that frequent accuchecks in stable diabetic patients produces better outcomes in geriatric LTC patients. Stable nursing home residents are little different from community elders, where accuchecks performed up to several times a week, fasting and/or post-prandial, allow sufficient monitoring in those on oral agents. Basal-bolus insulin can often be monitored by bid / prn accucheck when patients are stable. Excessive testing is wasteful, diverts nursing staff from more pressing matters and is burdensome to the resident.
Residents on continual sliding scale insulin
The majority of LTC residents should not require continual sliding scale insulin. Sliding scale insulin was designed for acute situations and not for chronic maintenance. It tends to only retrospectively treat past intake and conditions; it does not mimic physiologic relationship of food to concurrent insulin production. A guide for transitioning sliding scale to the more physiologic basal-bolus treatment is given on the attached pages. While basal-bolus insulin therapy is often useful in complex patients, it too is often unnecessary and burdensome in our older chronic care patients. Basal insulin and oral agents are one example of an alternative.
Residents on tight control who run risks of hypoglycemia
Goal hemoglobin A1C values for frail elders have recently been liberalized given lack of evidence of improved outcomes but increased risk with tight control. HgbA1C values of 7 to 8 are usuallyappropriate in many of our more agedpatients (see article attached) – at times even higher goals are appropriate. Attention should be given to sulfonylureas, which while inexpensive, are especially prone to cause low sugars (more than 4-fold compared to other agents) and have been demonstrated to have equivocal improvement in outcomes.
We will be instituting a liberalized diet soon throughout the facility. We are sending a letter of explanation to families so that they are well aware of these changes, and of the rationale, safety and benefits of this diet change. We ask for your support in this effort.
We ask for your effort to reconsider the frequency of accuchecks, individualized HgbA1C goals, and significantly reducing or eliminating sliding scale insulin. At your next visit, please re-evaluate your patients’ diabetic management. Remember, Medicare will pay for the time and effort you make in coordinating medical care. We hope to jointly improve the care and quality of life for our diabetic patients.
Sincerely,
Medical Director
A Physician Guide on How to Reduce the Use of Sliding Scale Insulin
1) Request Pharmacy to provide you with a list of your patients on diabetic Rx.
2) On your nextroutine visit, review the past control of the diabetic patient in depth.
3) Start with patients who are easily controlled or on less complex regiments for better success and acceptance, then tackle those on higher dose
Example:
1) Patient is on 18 units of Lantus and only requires an average of SS coverage of 10 units a day. You would discontinue the SSI, add 80% of the sliding scale (8 units) to the Lantus for a total of 26 units long acting insulin. Then obtain qid bs (ac and hs) for 1 week, review and adjust, having nursing call if bs less than 80 or greater than 300. When control is deemed stable, accuchecks should be able to be decreased to bid and prn (or less). If not well controlled, consider adding an oral agent taking into consideration the patient’s weight, HgA1C, creatinine and GFR. Continue to adjust every 1 to 4 weeks, following accuchecks in person or via fax.
2) Patient is on 30 units of Lantus and a total of 48 units of SS coverage a day. To determine the amount of short acting insulin needed at each meal time, you need to approximately average the amount of sliding scale insulin given at thepreceding meal and the meal of interest. To be safe, use the lowest recent insulin dose given at the meal of interest. This can be expressed as: (average dose preceding meal + lowestrecent dose meal)/ 2 An example for lunch: breakfast average 16 units, lowest given for lunch 8 units, means 12 units given for lunch (1/2 of 16+8)). Monitor qid accuchecks for 1 week with orders to call you for bs <80 or >350. Then adjust the long acting if all values are above substantially above goal, and the individual values if high compared to other times. Once individual times and overall HgbA1C goals are achieved, consider decreasing accuchecks to bid (can alternate breakfast / dinner and lunch / hs to ensure all times are controlled). Remember to consider individual patient factors in setting HgbA1C goals. The following formulas may be helpful: Fasting bs: HgbA1C 6 = average fasting bs 97, each 1 point increase HgbA1C = bs 29 (so HgbA1C 7 = 127 average fasting bs (97+29) Mean bs: HgbA1C 6 = mean bs 135, each 1 point increase HgbA1C = mean bs 35 (so HgbA1C 7 = 170 mean bs (135+35)
What about diabetics controlled on diet? Consider checking an accucheck every 2 to 4 weeks and prn to ensure control is maintained.
Remember – Medicare covers the time spent optimizing diabetic management!
Charles Crecelius MD PhD FACP CMD Washington University School of Medicine
Glycemic Goals in Elderly Patients with Type II Diabetes Mellitus
Introduction:
Guidelines for the treatment of new-onset type-2 diabetes mellitus in adults follow a fairly consistent pattern. This includes recommendations for lifestyle modification and metformin initially, followed by the administration oral insulin sensitizers and hypoglycemic agents with the intention of reducing average blood glucose levels to values that are less likely to cause the acute and chronic sequelae of the disease (1). This sequence is designed to maximizebenefits while minimizing risk and encumbrance of treatment. Goals of treatment include prevention of symptoms of hyperglycemia, quality of life improvement, reduction of complications, and extension of life. Risk-benefit analysis of treatment must be reexamined in elderly patients given higher rates of comorbidities, lower life expectancy, and special vulnerabilities to both disease and treatment.
Goals of treatment in adult DM:
Current glycemic goals of therapy as outlined by the recommendations of the American Diabetes Association set target Hemoglobin A1C at <7% with the caveat that clinical judgment must be used in when there is limited life expectancy and increased risk of causing significant hypoglycemia (1). The ADA arrived at this target mainly by way of the analysis of several large-scale clinical trials including the Diabetes Control and Complications Trial (DCCT) (2) and the Stockholm Diabetes Intervention Study (3) in type 1 diabetes, and the U.K. Prospective Diabetes Study (UKPDS) (4,5) and Kumamoto Study (6) in type 2 diabetes. None of these studies looked specifically to find optimal glycemic goals of therapy. The DCCT and UKPDS set glycemic goals at HbA1C values in the non-diabetic range, however neither study was able to achieve this goal in their intensive treatment groups, instead reaching average levels close to 7%. Regarding generalizability of these studies to the geriatric population: the DCCT looked only at patients aged 13-39 years old and the UKPDS looked at patients <65 years old.
Despite these issues, the above clinical trials did reveal significant reduction in the rates of complications associated with types 1 and 2 diabetes mellitus including retinopathy, neuropathy, and nephropathy when glycemic goals were set in the non-diabetic range. In type 1 DM, these intensive treatment goals have also been shown to reduce complications related to cardiovascular disease (7,8). The effects of intensive treatment on cardiovascular outcomes in type 2 DM are less obvious and are compounded by the effects of pre-diabetic hyperglycemia on the cardiovascular system and by comorbidities within the patient population (9,10). Looking at patients with longstanding diabetes and comparing goal A1C below 6.0% to a goal between 7.0 and 7.9%, the Action to Control Cardiovascular Risk in Diabetes Study Group (ACCORD) (11,12) revealed an increase in all-cause mortality in the intensive treatment group at 3.5 years with increased rates of hypoglycemia requiring assistance and increased rates of weight gain greater than 10 kg. Results at 5 years (17 months after cessation of intensive glycemic control) revealed reduction in non-fatal MI, but greater all-cause mortality in the intensive treatment group. In the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE) trial (15), the mean age of participants was 66 years old. This study compared the incidence of microvascular and macrovascular complications in an intensive treatment group given gliclazide (a sulfonylurea) with average A1C of 6.5% and a standard treatment group with an average A1C of 7.3%. There was a significant reduction of microvascular complications attributable to reduction of the incidence of nephropathy, however, there was no significant difference in retinopathy, macrovascular events, or overall mortality.
The randomized control trials we have to this date show greatest magnitude of benefit when reducing blood sugars from the uncontrolled range to the controlled range (<9%). Further benefit of glycemic control is significant, but smaller, and mainly due to the reduction of microvascular disease and nephropathy.
Special Considerations in the Elderly:
When performing a cost-benefit analysis of intensive treatment in elderly patients, one must take into consideration the special risks associated with both disease and treatment in this population, as well as the potential benefits. Vulnerabilities in this population include risk of falling, risk of cognitive impairment, and risk of polypharmacy. Many of these vulnerabilities interact with one another.
In a cohort study looking specifically at falls in community-dwelling patients older than 65, diabetes was associated with an increased risk of falls with HR=1.67 (95% CI: 1.11-2.51) (13). Cox-regression revealed factors that most explained increased risk to be number of medications, level of pain, self-perceived health, lower levels of physical activity, limitations in ADLs, lower extremity strength, and cognitive impairment. The symptoms of hyperglycemia including nocturia, dizziness, weakness, confusion, and UTIs may contribute to falls. These symptoms can largely be avoided by maintaining A1C below 9% (~212 mg/dl average glucose) (14). By setting a higher goal, one would reduce the risk of falls due to both hyperglycemia and hypoglycemia with the possibility of reducing medication burden on patients. However, other interacting factors such as the development of peripheral neuropathy could eventually contribute to an increased risk of falls.
Very tight glycemic control was found to result in higher mortality in the ACCORD trial, with increases in hypoglycemic events. Increases in hypoglycemic events were also observed in the tight-control arm of the ADVANCE study (HR=1.86 95% CI: 1.42-2.40). This has special implications in the elderly as these episodes increase the risk of falls and may contribute to the hastening of cognitive impairment (16). An additional study looking at 2 years of strict glycemic control in veterans, the Veterans Affairs Cooperative Study on Diabetes Mellitus (VACSDM), intensive insulin therapy (mean A1C 6.9%) was also associated with an increase in incidence of hypoglycemia and a non-significant increase in cardiovascular events over standard therapy (mean A1C 8.4%). Given the special vulnerabilities in the geriatric population, negative consequences of intensive treatment may be exacerbated in the elderly while benefits may be blunted.
Life expectancy is an important factor in determining the potential benefits of tight glycemic control. The effects of diabetes on the microvascular system take time, even years to develop. In order to see benefit of treatment, a patient would have to not already be burdened with these complications and would have to live long enough for the difference to manifest. As expected, with decreasing life expectancy, there are decreasing returns on tight glycemic control (17). Microvascular benefits of tight glycemic control are not significant when life expectancy falls below 5 years.
Recommendations For Glycemic Goals:
Current recommendations of various national organizations are shown below.
Organization / Goal HgbA1C / ConsiderationsAmerican Diabetic Association / < 7 / Re-evaluate if reduced life expectancy, co-morbidities, CAD, silent hypoglycemia
American Geriatric Society / <8 / For frail, life expectancy < 5 yr, risk of treatment complications
Department of Veteran Affairs / <7 / Only for life expectancy >10 yr, no major co-morbidities
<8 / Life expectancy 5-10 yrs, only mild co-morbidities
<9 / Life expectancy <5 yrs, major co-morbidities
These recommendations are made with most attention paid to the dwindling benefit gained from tight glycemic control in those with reduced life expectancy and greater co-morbidity. There is less attention paid to those factors that may make tight glycemic control more dangerous in certain patients including level of cognitive impairment, level of polypharmacy, and risk of falls. Perhaps these factors should be included in future recommendations for glycemic goals. There remains much room for research in determining the optimal glycemic goals in elderly patients. Quality of life at various glycemic targets has not been examined sufficiently in randomized control trials. Both the burden of the disease and the burden of treatment will weigh in when determining an optimal treatment strategy in an individual patient.
Charles Crecelius MD PhD FACP CMD
Assistant Clinical Professor of Medicine
Washington University School of Medicine
Diabetic Control in the Elderly References
1. David M. Nathan, MD,John B. Buse, MD, PHD,Mayer B. Davidson, MD,Robert J. Heine, MD,Rury R. Holman, FRCP,Robert Sherwin, MDand Bernard Zinman, MD:Medical Management of Hyperglycemia in Type 2 Diabetes: A Consensus Algorithm for the Initiation and Adjustment of Therapy.Diabetes Care. 2009 January; 32(1): 193–203.