- Email: [email protected]
Frailty and Diabetes in Older Hospitalized Adults: The Case for Routine Frailty Assessment
Can J Diabetes xxx (2019) 1e5
Contents lists available at ScienceDirect
Canadian Journal of Diabetes journal homepage: www.canadianjournalofdiabetes.com
Original Research
Frailty and Diabetes in Older Hospitalized Adults: The Case for Routine Frailty Assessment Heather Theresa MacKenzie MD a, *; Barna Tugwell MD a; Kenneth Rockwood MD, FRCPC, FRCP b; Olga Theou PhD b a b
Division of Endocrinology & Metabolism, Dalhousie University, Halifax, Nova Scotia, Canada Division of Geriatric Medicine, Dalhousie University, Halifax, Nova Scotia, Canada
Key Messages Frailty and diabetes can coexist in older adults. Frailty was predictive of inpatient mortality and extended length of stay, but diabetes was not. Despite guidelines recommending less strict glycemic control in more frail older adults, these patients had lower admission glucose than those who were less frail.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 April 2019 Received in revised form 7 June 2019 Accepted 2 July 2019
Objectives: Diabetes is common among older hospitalized adults; however, the effect of a diabetes diagnosis, frailty and blood glucose on mortality and hospital length of stay (LOS) has not been well described, nor is frailty routinely assessed in inpatients. Methods: This study included patients 65 years of age consulted to internal medicine through the emergency department at a Canadian tertiary care hospital. An internist-geriatrician determined their frailty status using the Clinical Frailty Scale. Inpatient mortality rates and LOS were obtained from a hospital administrative database. Admission glucose (fasting or random) and hemoglobin A1C were performed within 1 and 92 days of the comprehensive geriatric assessment. Results: This study included 400 patients (mean age, 81.48.1 years), 79.3% were frail (Clinical Frailty Scale score 5) and 35.3% had diabetes. The inpatient mortality rate was 19.7%, and among those who were discharged from the hospital, mean LOS was 23.736.5 days. Patients with diabetes were more likely to be frail than patients without diabetes. Diabetes status was not associated with LOS or mortality, but frailty was associated with both outcomes in multivariate regression analysis adjusted for age, sex and admission glucose. In patients with diabetes, mean admission glucose decreased with increasing frailty. Conclusions: Frailty was more common in patients with diabetes. Frailty, not diabetes, was associated with increased mortality and LOS in multivariate analysis. In patients with diabetes, admission glucose was lower with higher frailty. Frailty should be routinely assessed in all inpatients with diabetes because it is associated with hospital outcomes. Ó 2019 Canadian Diabetes Association.
Keywords: diabetes elderly frail inpatient older
Mots clés: diabète personnes âgées fragilité patients hospitalisés âgé
r é s u m é Objectifs : Le diabète est fréquent chez les personnes âgées hospitalisées. Toutefois, on en connaît peu sur les conséquences du diagnostic du diabète, de la fragilité et de la glycémie sur la mortalité et la durée du séjour à l’hôpital (DSH), ni sur l’évaluation de la fragilité chez les patients hospitalisés.
* Address for correspondence: Heather Theresa MacKenzie MD, Division of Endocrinology & Metabolism, Dalhousie University, QEII - Victoria Building, Suite 7 North-048 Victoria Building, 1276 South Park Street, Halifax, Nova Scotia B3H 2Y9, Canada. E-mail address: [email protected] 1499-2671/Ó 2019 Canadian Diabetes Association. The Canadian Diabetes Association is the registered owner of the name Diabetes Canada. https://doi.org/10.1016/j.jcjd.2019.07.001
2
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Méthodes : La présente étude a regroupé des patients 65 ans vus en médecine interne par l’intermédiaire du service des urgences d’un hôpital de soins tertiaires du Canada. Un interniste-gériatre a déterminé leur niveau de fragilité à l’aide de l’échelle Clinical Frailty Scale. Une base de données administratives de l’hôpital a permis d’obtenir les taux de mortalité des patients hospitalisés et la DSH. Les analyses de la glycémie (à jeun ou aléatoire) et de l’hémoglobine glyquée (A1c) à l’admission ont été réalisées entre 1 et 92 jours après l’évaluation gériatrique approfondie. Résultats : La présente étude a regroupé 400 patients (âge moyen, 81,4 8,1 ans), dont 79,3 % étaient fragiles (score à l’échelle Clinical Frailty Scale 5) et 35,3 % avaient le diabète. Le taux de mortalité des patients hospitalisés était de 19,7 %. Parmi ceux qui avaient obtenu leur sortie de l’hôpital, la DSH moyenne était de 23,7 36,5 jours. Les patients diabétiques étaient plus susceptibles d’être fragiles que les patients non diabétiques. Le statut diabétique n’a pas été associé à la DSH ou à la mortalité. Toutefois, selon l’analyse de régression multivariée ajustée en fonction de l’âge, du sexe et de la glycémie à l’admission, la fragilité a été associée aux 2 issues. Chez les patients diabétiques, la glycémie moyenne à l’admission diminuait lorsque la fragilité augmentait. Conclusions : La fragilité était plus fréquente chez les patients diabétiques. Selon l’analyse multivariée, la fragilité, mais non le diabète, était associée à l’augmentation de la mortalité et de la DSH. Chez les patients diabétiques, la glycémie à l’admission était plus faible lorsque la fragilité était plus élevée. La fragilité devrait être évaluée de manière systématique chez tous les patients diabétiques en raison de son association avec les issues à l’hôpital. Ó 2019 Canadian Diabetes Association.
Introduction Diabetes is a common disease among older adults and among inpatients. In Canadian adults >65 years of age, the prevalence of diabetes is 12.4%, with greater prevalence in institutionalized older adults and rising rates in the oldest population cohorts (1,2). Among inpatients, 26% have a diagnosis of diabetes on admission and an additional 12% are found to be hyperglycemic during admission (3). Inpatient hyperglycemia increases morbidity and mortality in hospitalized patients (4). Frailty, a condition often associated with aging, is defined as the accumulation of multiple deficits leading to a state of vulnerability, functional impairment and decreased ability of the organism to respond to stress (5e7). Higher frailty level is correlated with increased rates of hospitalization, inpatient mortality, new institutionalization at hospital discharge and length of stay (LOS) (8,9). It is operationally defined either by the number of health deficits accumulated, via a categorical scale, or via rules-based definitions with specific criteria (10). The prevalence of any degree of frailty in the United States population >65 years of age is 15.3%, with an additional 45.5% found to be vulnerable (9). The rate of frailty in residential care homes is higher than in the community-dwelling older adult population (9). Diabetes Canada recommends that target glycemic control for each older patient with diabetes should be based on functional status, life expectancy and level of frailty (11). Minimal cardiovascular benefit (12,13) and increased mortality (14) have been seen as a result of intensive glycemic control among older patients with a long duration of diabetes. National and international guidelines endorse looser glycemic targets for frail older adults because there is often minimal benefit to intensive glycemic control in this population, and risk of harm from hypoglycemia. Provincial long-term care guidelines in Nova Scotia suggest even more lenient glycemic control for residents in long-term care facilities, who are frail by definition. A blood glucose range of 10 to 20 mmol/L is considered most acceptable, with 7 to 9.9 mmol/L conditionally acceptable based on risk or frequency of hypoglycemia (15). Frailty and diabetes are common disorders that threaten the health of older patients, with complex interactions and management. To date, it has not been clear whether the diagnosis of diabetes itself drives the risk for poor inpatient outcomes or whether the overall frailty of patients is the predominant contributing factor. In studies across the world, type 2 diabetes is associated with an
increased LOS and inpatient mortality, particularly after cardiovascular events (16e20). However, diabetes encompasses a large clinical spectrum with varying effects on functional status for each patient, often depending on type and degree of complications. Unfortunately, frailty is not formally assessed in most inpatients. Given that our division of geriatric medicine has been formally assessing frailty in a subset of admitted patients for years, we were uniquely suited to examine the relationship between diabetes, frailty and outcomes. We sought to determine how strongly frailty and diabetes were associated with hospital outcomes. Methods We analyzed data from 400 patients 65 years of age admitted through the emergency department and consulted to internal medicine at a Canadian tertiary care hospital from April 2008 through April 2015. All patients were assessed by a senior internistgeriatrician or a member of their team (senior medical student, resident or geriatrics fellow) by conducting a comprehensive geriatric assessment (CGA) and determining their level on the Clinical Frailty Scale (CFS), a standardized validated frailty assessment instrument. Patients were considered to have diabetes if an International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Canadian version (ICD-10-CA) diagnosis code indicating any type of diabetes was present in the Discharge Abstract Database, developed by the Canadian Institute for Health Information. Assessment of frailty The CFS is a widely used 9-point clinical scale ranging from very fit to terminally ill (Supplementary Figure 1). This scale was validated initially as a 7-point scale, correlating with death within 18 months and with institutionalization (21). We defined nonfrail as CFS score 1 to 4, mildly frail as CFS score 5, moderately frail as CFS score 6 and severely frail as CFS score 7 to 9. Measures of glucose control Admission glucose (mmol/L) was the serum fasting or random glucose performed in hospital within 1 day of the CGA for 372 patients; 283 were tested the same date as the CGA. We defined hypoglycemia as a glucose level <4 mmol/L, normoglycemia as a glucose level 4 to 10 mmol/L, moderate hyperglycemia as a glucose
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Table 2 Relationship of frailty and diabetes status with in-hospital mortality and length of stay
Table 1 Demographic and clinical characteristics Characteristic Females Age, years Admission glucose, mmol/L A1C, % Admission by ambulance Admission diagnosis Respiratory Circulatory Mental and behaviour ED stay, hours CFS Nonfrail (CFS score 1e4) Mildly frail (CFS score 5) Moderately frail (CFS score 6) Severely frail (CFS score 7e9) LOS, days LOS >14 days Died
All (N¼400)
3
Diabetes (n¼141; 35.3%)
No diabetes (n¼259; 64.8%)
228 (57) 81.48.1 8.14.4 6.81.8 293 (73.3)
74 (52.5) 80.67.8 10.35.8 7.41.9 108 (76.6)
154 (59.5) 81.98.2 6.92.7* 5.60.5* 185 (71.4)
109 (27.3) 49 (12.3) 43 (10.8) 10.412.9
35 (24.8) 20 (14.2) 17 (12.1) 11.912.7
74 (28.6) 29 (11.2) 26 (10.0) 9.613.0
83 (20.8) 110 (27.5) 108 (27.0)
22 (15.6) 43 (30.5) 40 (28.4)
61 (23.6) 67 (25.9) 68 (26.3)
99 (24.8)
36 (25.5)
63 (24.3)
23.736.5 136 (42.9) 78 (19.7)
26.538.5 48 (41.7) 25 (17.9)
22.135.3 88 (43.6) 53 (20.8)
AIC, glycated hemoglobin; CFS, Clinical Frailty Scale; ED, emergency department; LOS, length of stay. Note: Values are n (%) or mean SD. * Significantly different from people with diabetes (p<0.05).
level 10.1 to 15 mmol/L and severe hyperglycemia as a glucose level >15 mmol/L. These glucose values, measured in an acute setting, are not meant to correspond to particular glycated hemoglobin (A1C) targets, which are more appropriately a reflection of longterm control. Admission A1C (%) was performed within 92 days of admission for 146 patients; 56.8% were tested within 30 days from the CGA. We considered an A1C of 7.0% to 8.5% as the usual acceptable range for older adults, with standard control being A1C <7% and high A1C defined as >8.5%, recognizing that an A1C of 8.6% to 10% is also acceptable for most long-term care facility residents in Nova Scotia (15). Ten patients from the database were misclassified on admission to not have diabetes but were found to meet A1C criteria for diabetes during admission (A1C 6.5%) and were reclassified into the diabetes group for analysis. Mortality and LOS Mortality and LOS for each patient were obtained through a hospital records database. Prolonged LOS was defined as hospital admission for 14 sequential days and did not include hospital days spent as alternative level of care awaiting an institutional care bed. Statistical analysis Baseline characteristics are expressed as mean SD. Multivariate logistic regression models were used to assess the effect of diabetes and frailty on mortality and LOS. These models were adjusted for age, sex and admission glucose. Odds ratios (ORs) and 95% confidence intervals were calculated, and p<0.05 was considered significant. Statistical analysis was performed with SPSS 24 software (IBM, Armonk, New York, United States). Results Of 400 patients in the database, the average age was 81.48.1 years, and 57% were women. In the database, 79.3% were frail (CFS score >4) and 35.3% had diabetes. The inpatient mortality
Variable
Outcome: In-hospital mortality (19.7%) OR
Diabetes 0.659 Female sex 0.471 Age 1.029 Glucose (reference: normal) Hyper (10.1e15 1.574 mmol/L) Severe elevation 0.368 (>15 mmol/L) CFS (reference: CFS score 1e4) CFS score 5 3.095 CFS score 6 3.866 CFS score 7e9 16.614
95% CI
Outcome: Length of stay >14 days (42.9%) p value
OR
95% CI
p value
0.344e1.262 0.262e0.847 0.99e1.067
0.208 0.880 0.494e1.566 0.012 1.300 0.787e2.145 0.115 1.022 0.991e1.054
0.663 0.305 0.167
0.707e3.503
0.267 0.921 0.416e2.037
0.838
0.041e3.321
0.373 1.309 0.432e3.964
0.634
0.961e9.974 0.058 2.101 1.041e4.239 0.038 1.202e12.433 0.023 3.587 1.766e7.287 <0.001 5.426e50.875 <0.001 2.484 1.102e5.599 0.028
CFS, Clinical Frailty Scale; CI, confidence interval; OR, odds ratio.
rate was 19.7%, and among those who were discharged from the hospital, the mean LOS was 23.736.5 days. Unsurprisingly, admission glucose and A1C were higher among patients with diabetes than those without diabetes (Table 1). People with diabetes were 1.96 times more likely (95% confidence interval, 1.05 to 3.67; p¼0.035) to be frail (CFS score >4) than those without diabetes in regression models adjusted for age, sex and admission glucose. In a logistic regression model including frailty, age, sex, admission glucose and diabetes status, moderate to severe frailty was associated with an increased OR for in-hospital mortality (p<0.001) (Table 2). Severe frailty was the strongest predictor of mortality in multivariate analysis (OR, 16.61; p<0.001). Female sex was protective against mortality in this population (OR, 0.47; p¼0.012), but diabetes status was not associated with mortality (OR, 0.66; p¼0.208). Moderate frailty (CFS score 6) was associated with a LOS >14 days (OR, 3.59; p<0.001), whereas diabetes status was not a significant predictor of LOS (OR, 0.88; p¼0.663) (Table 2). Admission glucose was available for 372 patients. In patients with diabetes, mean admission glucose levels decreased with increasing frailty (p¼0.003 for main effect) (Figure 1), with mean admission glucose for nonfrail patients of 10.34.6 mmol/L, mean admission glucose for patients who were mildly frail of 13.08.4 mmol/L, mean admission glucose for patients who were moderately frail of 9.03.4 mmol/L and mean admission glucose for patients who were severely frail of 8.63.4 mmol/L. This relationship remained similar when analysis was adjusted for age, and age was not correlated with admission glucose (p¼0.477). Nine patients had hypoglycemia on admission, 5 without diabetes and 4 with diabetes. Groups were too small to compare statistically, but in patients without diabetes, all had moderate to severe frailty. Hypoglycemia was more equally distributed among levels of frailty in patients with diabetes, but none with hypoglycemia were nonfrail. A1C levels were available for 146 patients. In patients with diabetes, A1C levels were 7.35%1.50% for the nonfrail patients, 8.19% 2.53% for the patients who were mildly frail, 6.87%1.18% for the patients who were moderately frail and 7.01%1.32% for the patients who were severely frail (p¼0.41). Among people with diabetes, 28 (29.8%) had A1C between 7% and 8.5%, 50 (53.2%) had A1C <7% and 16 (17%) had A1C >8.5%; no difference by frailty level was observed. In patients without diabetes, A1C levels were not statistically different (p¼0.3) across levels of frailty.
4
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Figure 1. Mean admission glucose (mmol/L) stratified by frailty category. In patients with diabetes, admission glucose decreased with increasing frailty (p¼0.003 for main effect). CFS, Clinical Frailty Scale score.
Discussion This study shows that older inpatients with diabetes are more commonly frail than those without diabetes. This is unsurprising given the contribution of diabetes to poor functional status and accumulation of health deficits through many mechanisms. In logistic regression modelling with adjustment for age, sex and glucose, frailty was a predictor of in-hospital mortality and prolonged LOS, whereas diabetes status was not. Multiple studies endorse the association between higher frailty status and mortality both in community-dwelling adults and in inpatients; therefore, our conclusions are consistent with established literature (8,22). Frailty has also been associated with increased hospital LOS in other studies (23,24). Previous studies correlating type 2 diabetes and LOS did not control for frailty (16,17), which is a confounder. We showed that it is frailty, rather than diabetes status, that drives risk for both extended LOS and inpatient mortality. By establishing the correlation of frailty with inpatient mortality and LOS, this suggests that frailty should be assessed in all older inpatients to establish their risk for death and prolonged admission. It is not sufficient to simply assess their comorbidities including diabetes; one must have a global perspective of their frailty. A recent international position statement also endorses recognition and evaluation of frailty in older inpatients through CGA, emphasizing education, understanding and appropriate use of frailty assessment instruments (25). Our findings of less hyperglycemia on admission in more frail patients with diabetes are surprising given provincial guidelines for glucose management in long-term care facilities, a common living situation for frail older adults. We found that patients with diabetes with moderate and severe frailty had lower mean admission glucose than their nonfrail and mildly frail counterparts. There may be multiple mechanisms serving to explain this, including effects of antihyperglycemic agents, undernutrition, poor detection of signs and symptoms of hypoglycemia or physiologic reasons, such as sepsis, hepatic failure or renal failure (11,15,26,27). The findings of more frail patients with less hyperglycemia on admission are consistent with those of other studies, which found that most older adults with poor health were treated to A1C <7%, often using agents that increase risk for hypoglycemia (28,29). Medications that are known to cause hypoglycemia, such as sulfonylureas and insulin, are often covered for patients under provincial drug plans, whereas other agents with lower rates of hypoglycemia are not. Older adults often depend on provincial medication coverage
because of limited income and, therefore, can be limited in accessing lower-risk agents. Interestingly, all hypoglycemia in patients with diabetes was seen in those with moderate to severe frailty. This could indicate that patients who are frail are particularly vulnerable to hypoglycemia and may be useful for hypothesis generation regarding appropriate goals of antihyperglycemic therapy when a patient with diabetes becomes moderately to severely frail. In our institution, the rates of high-risk hypoglycemic medication use among frail older adults with diabetes at the time of admission to hospital is under ongoing study. The limitations of our study include its use of a database population in which a single clinical team gathered data from patients consulted to internal medicine, and it may not be representative of the entire older adult population seen in the emergency department of the tertiary care hospital. Admission glucose values were taken within 1 day of the CFS assessment, which may have led to detection of some effects of inpatient treatment of glucose, if given. In general, results were limited by sample size, particularly in correlating A1C with outcomes and in the analysis of the number of patients with hypoglycemia. The strengths of our study include the increasing clinical recognition of frailty in our institution and routine collection of frailty data in our hospital. Specialized teams are available for inpatient consultation to optimize care choices for patients who are frail, and frailty awareness is being incorporated into multidisciplinary and interprofessional training and research. Clinical practice and training allow greater recognition and more accurate assessment of frailty in the older adult population. Frailty assessment of all older inpatients should be done routinely, particularly in those with diabetes to adjust therapy, monitor for hypoglycemia and associate with hospital outcomes. Future studies in older inpatients with diabetes might include investigation of the impact of inpatient interventions on patients with diabetes of varying frailty status, such as the use of different classes of diabetes medications and of standardized insulin orders. There is little known about the outcomes of various blood glucose targets in the frail older adult population in both inpatient and outpatient settings, and further research may allow us to establish appropriate, safe targets that minimize the risk of symptom and complication progression while maintaining safety. Whether widespread early assessment of frailty during admission leads to improvement of outcomes remains to be determined. Early involvement of geriatric medicine consultants, home support services and other allied health disciplines in the care of those identified as frail may be beneficial. Many hospitals have geriatric consultants for inpatients and geriatric medicine clinics with interdisciplinary teams and/or geriatric day care programs for outpatients. Geriatric medicine expertise can also be sought from family physicians who have completed a year-long training program in the care of the elderly. For inpatients returning to the community or to long-term care, geriatric rehabilitation wards, transitional care wards or other supports including occupational therapy, physical therapy and social work may be available to ease this transition. For older adults living in the community, home support can be accessed on a no-cost or low-cost basis and can include varied services, such as medication management, food preparation and personal care depending on the functional capacity of each older adult. It is important to be aware of the resources available in one’s own community to assist with diabetes care and maintain the overall health of frail older adults. This study shows that patients with diabetes are more likely to be frail. Frailty was a robust predictor of in-hospital mortality and prolonged LOS in our cohort of patients 65 years of age, but diabetes status was not. We also found that patients with diabetes with moderate and severe frailty had lower glucose than those with no or mild frailty, assessed by admission serum glucose values.
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Diabetes at www. canadianjournalofdiabetes.com. Author Disclosures B.T. reports and discloses being a research investigator or subinvestigator with the following companies: GlaxoSmithKline, Novo Nordisk Canada, AMGEN, Sanofi-Aventis Canada, Ionis, BoehringerIngelheim, Novartis, AstraZeneca, Bristol-Myers Squibb, Intarcia, Lexicon, Merck, Eli Lilly, Pfizer, Takeda, NPS Pharmaceuticals, Cerenis Pharmaceuticals, Shire, Acasti Pharma and Aegerion. K.R. reports other from president and chief science officer, DGI Clinical, other from advisory board meeting, Lundbeck, other from associate director of the Canadian Consortium on Neurodegeneration in Aging and personal fees from various organizations, outside the submitted work. No other authors have any conflicts of interest to declare. Author Contributions H.T.M. performed the literature review, drafted the manuscript and edited the manuscript. B.T. contributed to study conception and design and edited the manuscript. K.R. provided access to the database. O.T. contributed to study design and performed statistical analysis. References 1. Umpierrez G, Pasquel F. Management of inpatient hyperglycemia and diabetes in older adults. Diabetes Care 2017;40:509e17. 2. Rockwood K, Tan M, Phillips S, McDowell I. Prevalence of diabetes mellitus in elderly people in Canada: Report from the Canadian Study of Health and Aging. Age Ageing 1998;27:573e7. 3. Umpierrez G. Hyperglycemia: An independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002;87: 978e82. 4. Houlden R, Capes S, Clement M, Miller D. In-hospital management of diabetes. Can J Diabetes 2013;37:S77e81. 5. Rockwood K, Rockwood M, Mitnitski A. Physiological redundancy in older adults in relation to the change with age in the slope of a frailty index. J Am Geriatr Soc 2010;58:318e23. 6. Rockwood K, Mitnitski A. Frailty defined by deficit accumulation and geriatric medicine defined by frailty. Clin Geriatr Med 2011;27:17e26. 7. Leng S, Chen X, Mao G. Frailty syndrome: An overview. Clin Interv Aging 2014; 19:433e41. 8. Basic D, Shanley C. Frailty in an older inpatient population: Using the Clinical Frailty Scale to predict patient outcomes. J Aging Health 2014;27:670e85.
5
9. Bandeen-Roche K, Seplaki C, et al. Frailty in older adults: A nationally representative profile in the United States. J Gerontol Ser A Biol Sci Med Sci 2015;70: 1427e34. 10. Rockwood K. A global clinical measure of fitness and frailty in elderly people. Can Med Assoc J 2005;173:489e95. 11. Meneilly G, Knip A, Miller D, Sherifali D, Tessier D, Zahedi A. Diabetes in older people. Diabetes Canada 2018 clinical practice guidelines for the prevention and management of diabetes in Canada: Diabetes in older people. Can J Diabetes 2018;42(Suppl. 1):S283e95. 12. Duckworth W, Abraira C, Moritz T. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009;360:129e39. 13. Patel A, MacMahon S, Chalmers J, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008;358: 2560e72. 14. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008;358:2545e59. 15. Diabetes Care Program of Nova Scotia. Diabetes guidelines for frail elderly residents in or awaiting long-term care (LTC): Pocket reference; 2018. http://dia betescare.nshealth.ca/sites/default/files/LTCPocketRefSept142018.pdf. Accessed July 29, 2019. 16. Kubo T, Fujino Y, Murata A, et al. Prevalence of type 2 diabetes among acute inpatients and its impact on length of hospital stay in Japan. Intern Med 2011; 50:405e11. 17. Comino E, Harris M, Islam M, et al. Impact of diabetes on hospital admission and length of stay among a general population aged 45 year or more: A record linkage study. BMC Health Serv Res 2015;15:12. https://doi.org/10.1186/ s12913-014-0666-2. 18. Clement S, Braithwaite S, Magee M, et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care 2004;27:553e91. 19. Ahmed B, Davis H, Laskey W. In-hospital mortality among patients with type 2 diabetes mellitus and acute myocardial infarction: Results from the National Inpatient Sample, 2000-2010. J Am Heart Assoc 2014;3:e001090. 20. Miguel-Yanes J, Jimenez-Garcia R, Hernandez-Barrera V, Mendez-Bailon M, de Miguel-Diaz J, Lopez-de-Andres A. Impact of type 2 diabetes mellitus on inhospital mortality after major cardiovascular events in Spain (2002-2014). Cardiovasc Diabetol 2017;16:126. 21. Jones D, Song X, Mitnitski A, Rockwood K. Evaluation of a frailty index based on a comprehensive geriatric assessment in a population based study of elderly Canadians. Aging Clin Exp Res 2005;17:465e71. 22. Eeles E, White S, O’Mahony S, Bayer A, Hubbard R. The impact of frailty and delirium on mortality in older inpatients. Age Ageing 2012;41:412e6. 23. Rose M, Pan H, Levinson M, Staples M. Can frailty predict complicated care needs and length of stay? Intern Med J 2014;44:800e5. 24. Juma S, Taabazuing M, Montero-Odasso M. Clinical Frailty Scale in an acute medicine unit: A simple tool that predicts length of stay. Can Geriatr J 2016;19: 34e9. 25. Sinclair AJ, Abdelhafiz A, Dunning T, et al. An international position statement on the management of frailty in diabetes mellitus: Summary of recommendations 2017. J Frailty Aging 2018;7:10e20. 26. Abdelhafiz A, Rodríguez-Mañas L, Morley J, Sinclair A. Hypoglycemia in older people - A less well recognized risk factor for frailty. Aging Dis 2015;6:156e67. 27. Furman E. Undernutrition in older adults across the continuum of care: Nutritional assessment, barriers, and interventions. J Gerontol Nursing 2006; 32:22e7. 28. Lipska K, Ross J, Miao Y, Shah N, Lee S, Steinman M. Potential overtreatment of diabetes mellitus in older adults with tight glycemic control. JAMA Intern Med 2015;175:356. 29. Meneilly G, Berard L, Cheng A, et al. Insights into the current management of older adults with type 2 diabetes in the Ontario primary care setting. Can J Diabetes 2018;42:23e30.
5.e1
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Supplementary Figure 1. Clinical frailty scale. IADLs, instrumental activities of daily living. Adapted with permission from Dalhousie University, Division of Geriatric Medicine. Ó 2007-2009 Geriatric Medicine Research, Dalhousie University, Halifax, Canada.
Contents lists available at ScienceDirect
Canadian Journal of Diabetes journal homepage: www.canadianjournalofdiabetes.com
Original Research
Frailty and Diabetes in Older Hospitalized Adults: The Case for Routine Frailty Assessment Heather Theresa MacKenzie MD a, *; Barna Tugwell MD a; Kenneth Rockwood MD, FRCPC, FRCP b; Olga Theou PhD b a b
Division of Endocrinology & Metabolism, Dalhousie University, Halifax, Nova Scotia, Canada Division of Geriatric Medicine, Dalhousie University, Halifax, Nova Scotia, Canada
Key Messages Frailty and diabetes can coexist in older adults. Frailty was predictive of inpatient mortality and extended length of stay, but diabetes was not. Despite guidelines recommending less strict glycemic control in more frail older adults, these patients had lower admission glucose than those who were less frail.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 April 2019 Received in revised form 7 June 2019 Accepted 2 July 2019
Objectives: Diabetes is common among older hospitalized adults; however, the effect of a diabetes diagnosis, frailty and blood glucose on mortality and hospital length of stay (LOS) has not been well described, nor is frailty routinely assessed in inpatients. Methods: This study included patients 65 years of age consulted to internal medicine through the emergency department at a Canadian tertiary care hospital. An internist-geriatrician determined their frailty status using the Clinical Frailty Scale. Inpatient mortality rates and LOS were obtained from a hospital administrative database. Admission glucose (fasting or random) and hemoglobin A1C were performed within 1 and 92 days of the comprehensive geriatric assessment. Results: This study included 400 patients (mean age, 81.48.1 years), 79.3% were frail (Clinical Frailty Scale score 5) and 35.3% had diabetes. The inpatient mortality rate was 19.7%, and among those who were discharged from the hospital, mean LOS was 23.736.5 days. Patients with diabetes were more likely to be frail than patients without diabetes. Diabetes status was not associated with LOS or mortality, but frailty was associated with both outcomes in multivariate regression analysis adjusted for age, sex and admission glucose. In patients with diabetes, mean admission glucose decreased with increasing frailty. Conclusions: Frailty was more common in patients with diabetes. Frailty, not diabetes, was associated with increased mortality and LOS in multivariate analysis. In patients with diabetes, admission glucose was lower with higher frailty. Frailty should be routinely assessed in all inpatients with diabetes because it is associated with hospital outcomes. Ó 2019 Canadian Diabetes Association.
Keywords: diabetes elderly frail inpatient older
Mots clés: diabète personnes âgées fragilité patients hospitalisés âgé
r é s u m é Objectifs : Le diabète est fréquent chez les personnes âgées hospitalisées. Toutefois, on en connaît peu sur les conséquences du diagnostic du diabète, de la fragilité et de la glycémie sur la mortalité et la durée du séjour à l’hôpital (DSH), ni sur l’évaluation de la fragilité chez les patients hospitalisés.
* Address for correspondence: Heather Theresa MacKenzie MD, Division of Endocrinology & Metabolism, Dalhousie University, QEII - Victoria Building, Suite 7 North-048 Victoria Building, 1276 South Park Street, Halifax, Nova Scotia B3H 2Y9, Canada. E-mail address: [email protected] 1499-2671/Ó 2019 Canadian Diabetes Association. The Canadian Diabetes Association is the registered owner of the name Diabetes Canada. https://doi.org/10.1016/j.jcjd.2019.07.001
2
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Méthodes : La présente étude a regroupé des patients 65 ans vus en médecine interne par l’intermédiaire du service des urgences d’un hôpital de soins tertiaires du Canada. Un interniste-gériatre a déterminé leur niveau de fragilité à l’aide de l’échelle Clinical Frailty Scale. Une base de données administratives de l’hôpital a permis d’obtenir les taux de mortalité des patients hospitalisés et la DSH. Les analyses de la glycémie (à jeun ou aléatoire) et de l’hémoglobine glyquée (A1c) à l’admission ont été réalisées entre 1 et 92 jours après l’évaluation gériatrique approfondie. Résultats : La présente étude a regroupé 400 patients (âge moyen, 81,4 8,1 ans), dont 79,3 % étaient fragiles (score à l’échelle Clinical Frailty Scale 5) et 35,3 % avaient le diabète. Le taux de mortalité des patients hospitalisés était de 19,7 %. Parmi ceux qui avaient obtenu leur sortie de l’hôpital, la DSH moyenne était de 23,7 36,5 jours. Les patients diabétiques étaient plus susceptibles d’être fragiles que les patients non diabétiques. Le statut diabétique n’a pas été associé à la DSH ou à la mortalité. Toutefois, selon l’analyse de régression multivariée ajustée en fonction de l’âge, du sexe et de la glycémie à l’admission, la fragilité a été associée aux 2 issues. Chez les patients diabétiques, la glycémie moyenne à l’admission diminuait lorsque la fragilité augmentait. Conclusions : La fragilité était plus fréquente chez les patients diabétiques. Selon l’analyse multivariée, la fragilité, mais non le diabète, était associée à l’augmentation de la mortalité et de la DSH. Chez les patients diabétiques, la glycémie à l’admission était plus faible lorsque la fragilité était plus élevée. La fragilité devrait être évaluée de manière systématique chez tous les patients diabétiques en raison de son association avec les issues à l’hôpital. Ó 2019 Canadian Diabetes Association.
Introduction Diabetes is a common disease among older adults and among inpatients. In Canadian adults >65 years of age, the prevalence of diabetes is 12.4%, with greater prevalence in institutionalized older adults and rising rates in the oldest population cohorts (1,2). Among inpatients, 26% have a diagnosis of diabetes on admission and an additional 12% are found to be hyperglycemic during admission (3). Inpatient hyperglycemia increases morbidity and mortality in hospitalized patients (4). Frailty, a condition often associated with aging, is defined as the accumulation of multiple deficits leading to a state of vulnerability, functional impairment and decreased ability of the organism to respond to stress (5e7). Higher frailty level is correlated with increased rates of hospitalization, inpatient mortality, new institutionalization at hospital discharge and length of stay (LOS) (8,9). It is operationally defined either by the number of health deficits accumulated, via a categorical scale, or via rules-based definitions with specific criteria (10). The prevalence of any degree of frailty in the United States population >65 years of age is 15.3%, with an additional 45.5% found to be vulnerable (9). The rate of frailty in residential care homes is higher than in the community-dwelling older adult population (9). Diabetes Canada recommends that target glycemic control for each older patient with diabetes should be based on functional status, life expectancy and level of frailty (11). Minimal cardiovascular benefit (12,13) and increased mortality (14) have been seen as a result of intensive glycemic control among older patients with a long duration of diabetes. National and international guidelines endorse looser glycemic targets for frail older adults because there is often minimal benefit to intensive glycemic control in this population, and risk of harm from hypoglycemia. Provincial long-term care guidelines in Nova Scotia suggest even more lenient glycemic control for residents in long-term care facilities, who are frail by definition. A blood glucose range of 10 to 20 mmol/L is considered most acceptable, with 7 to 9.9 mmol/L conditionally acceptable based on risk or frequency of hypoglycemia (15). Frailty and diabetes are common disorders that threaten the health of older patients, with complex interactions and management. To date, it has not been clear whether the diagnosis of diabetes itself drives the risk for poor inpatient outcomes or whether the overall frailty of patients is the predominant contributing factor. In studies across the world, type 2 diabetes is associated with an
increased LOS and inpatient mortality, particularly after cardiovascular events (16e20). However, diabetes encompasses a large clinical spectrum with varying effects on functional status for each patient, often depending on type and degree of complications. Unfortunately, frailty is not formally assessed in most inpatients. Given that our division of geriatric medicine has been formally assessing frailty in a subset of admitted patients for years, we were uniquely suited to examine the relationship between diabetes, frailty and outcomes. We sought to determine how strongly frailty and diabetes were associated with hospital outcomes. Methods We analyzed data from 400 patients 65 years of age admitted through the emergency department and consulted to internal medicine at a Canadian tertiary care hospital from April 2008 through April 2015. All patients were assessed by a senior internistgeriatrician or a member of their team (senior medical student, resident or geriatrics fellow) by conducting a comprehensive geriatric assessment (CGA) and determining their level on the Clinical Frailty Scale (CFS), a standardized validated frailty assessment instrument. Patients were considered to have diabetes if an International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Canadian version (ICD-10-CA) diagnosis code indicating any type of diabetes was present in the Discharge Abstract Database, developed by the Canadian Institute for Health Information. Assessment of frailty The CFS is a widely used 9-point clinical scale ranging from very fit to terminally ill (Supplementary Figure 1). This scale was validated initially as a 7-point scale, correlating with death within 18 months and with institutionalization (21). We defined nonfrail as CFS score 1 to 4, mildly frail as CFS score 5, moderately frail as CFS score 6 and severely frail as CFS score 7 to 9. Measures of glucose control Admission glucose (mmol/L) was the serum fasting or random glucose performed in hospital within 1 day of the CGA for 372 patients; 283 were tested the same date as the CGA. We defined hypoglycemia as a glucose level <4 mmol/L, normoglycemia as a glucose level 4 to 10 mmol/L, moderate hyperglycemia as a glucose
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Table 2 Relationship of frailty and diabetes status with in-hospital mortality and length of stay
Table 1 Demographic and clinical characteristics Characteristic Females Age, years Admission glucose, mmol/L A1C, % Admission by ambulance Admission diagnosis Respiratory Circulatory Mental and behaviour ED stay, hours CFS Nonfrail (CFS score 1e4) Mildly frail (CFS score 5) Moderately frail (CFS score 6) Severely frail (CFS score 7e9) LOS, days LOS >14 days Died
All (N¼400)
3
Diabetes (n¼141; 35.3%)
No diabetes (n¼259; 64.8%)
228 (57) 81.48.1 8.14.4 6.81.8 293 (73.3)
74 (52.5) 80.67.8 10.35.8 7.41.9 108 (76.6)
154 (59.5) 81.98.2 6.92.7* 5.60.5* 185 (71.4)
109 (27.3) 49 (12.3) 43 (10.8) 10.412.9
35 (24.8) 20 (14.2) 17 (12.1) 11.912.7
74 (28.6) 29 (11.2) 26 (10.0) 9.613.0
83 (20.8) 110 (27.5) 108 (27.0)
22 (15.6) 43 (30.5) 40 (28.4)
61 (23.6) 67 (25.9) 68 (26.3)
99 (24.8)
36 (25.5)
63 (24.3)
23.736.5 136 (42.9) 78 (19.7)
26.538.5 48 (41.7) 25 (17.9)
22.135.3 88 (43.6) 53 (20.8)
AIC, glycated hemoglobin; CFS, Clinical Frailty Scale; ED, emergency department; LOS, length of stay. Note: Values are n (%) or mean SD. * Significantly different from people with diabetes (p<0.05).
level 10.1 to 15 mmol/L and severe hyperglycemia as a glucose level >15 mmol/L. These glucose values, measured in an acute setting, are not meant to correspond to particular glycated hemoglobin (A1C) targets, which are more appropriately a reflection of longterm control. Admission A1C (%) was performed within 92 days of admission for 146 patients; 56.8% were tested within 30 days from the CGA. We considered an A1C of 7.0% to 8.5% as the usual acceptable range for older adults, with standard control being A1C <7% and high A1C defined as >8.5%, recognizing that an A1C of 8.6% to 10% is also acceptable for most long-term care facility residents in Nova Scotia (15). Ten patients from the database were misclassified on admission to not have diabetes but were found to meet A1C criteria for diabetes during admission (A1C 6.5%) and were reclassified into the diabetes group for analysis. Mortality and LOS Mortality and LOS for each patient were obtained through a hospital records database. Prolonged LOS was defined as hospital admission for 14 sequential days and did not include hospital days spent as alternative level of care awaiting an institutional care bed. Statistical analysis Baseline characteristics are expressed as mean SD. Multivariate logistic regression models were used to assess the effect of diabetes and frailty on mortality and LOS. These models were adjusted for age, sex and admission glucose. Odds ratios (ORs) and 95% confidence intervals were calculated, and p<0.05 was considered significant. Statistical analysis was performed with SPSS 24 software (IBM, Armonk, New York, United States). Results Of 400 patients in the database, the average age was 81.48.1 years, and 57% were women. In the database, 79.3% were frail (CFS score >4) and 35.3% had diabetes. The inpatient mortality
Variable
Outcome: In-hospital mortality (19.7%) OR
Diabetes 0.659 Female sex 0.471 Age 1.029 Glucose (reference: normal) Hyper (10.1e15 1.574 mmol/L) Severe elevation 0.368 (>15 mmol/L) CFS (reference: CFS score 1e4) CFS score 5 3.095 CFS score 6 3.866 CFS score 7e9 16.614
95% CI
Outcome: Length of stay >14 days (42.9%) p value
OR
95% CI
p value
0.344e1.262 0.262e0.847 0.99e1.067
0.208 0.880 0.494e1.566 0.012 1.300 0.787e2.145 0.115 1.022 0.991e1.054
0.663 0.305 0.167
0.707e3.503
0.267 0.921 0.416e2.037
0.838
0.041e3.321
0.373 1.309 0.432e3.964
0.634
0.961e9.974 0.058 2.101 1.041e4.239 0.038 1.202e12.433 0.023 3.587 1.766e7.287 <0.001 5.426e50.875 <0.001 2.484 1.102e5.599 0.028
CFS, Clinical Frailty Scale; CI, confidence interval; OR, odds ratio.
rate was 19.7%, and among those who were discharged from the hospital, the mean LOS was 23.736.5 days. Unsurprisingly, admission glucose and A1C were higher among patients with diabetes than those without diabetes (Table 1). People with diabetes were 1.96 times more likely (95% confidence interval, 1.05 to 3.67; p¼0.035) to be frail (CFS score >4) than those without diabetes in regression models adjusted for age, sex and admission glucose. In a logistic regression model including frailty, age, sex, admission glucose and diabetes status, moderate to severe frailty was associated with an increased OR for in-hospital mortality (p<0.001) (Table 2). Severe frailty was the strongest predictor of mortality in multivariate analysis (OR, 16.61; p<0.001). Female sex was protective against mortality in this population (OR, 0.47; p¼0.012), but diabetes status was not associated with mortality (OR, 0.66; p¼0.208). Moderate frailty (CFS score 6) was associated with a LOS >14 days (OR, 3.59; p<0.001), whereas diabetes status was not a significant predictor of LOS (OR, 0.88; p¼0.663) (Table 2). Admission glucose was available for 372 patients. In patients with diabetes, mean admission glucose levels decreased with increasing frailty (p¼0.003 for main effect) (Figure 1), with mean admission glucose for nonfrail patients of 10.34.6 mmol/L, mean admission glucose for patients who were mildly frail of 13.08.4 mmol/L, mean admission glucose for patients who were moderately frail of 9.03.4 mmol/L and mean admission glucose for patients who were severely frail of 8.63.4 mmol/L. This relationship remained similar when analysis was adjusted for age, and age was not correlated with admission glucose (p¼0.477). Nine patients had hypoglycemia on admission, 5 without diabetes and 4 with diabetes. Groups were too small to compare statistically, but in patients without diabetes, all had moderate to severe frailty. Hypoglycemia was more equally distributed among levels of frailty in patients with diabetes, but none with hypoglycemia were nonfrail. A1C levels were available for 146 patients. In patients with diabetes, A1C levels were 7.35%1.50% for the nonfrail patients, 8.19% 2.53% for the patients who were mildly frail, 6.87%1.18% for the patients who were moderately frail and 7.01%1.32% for the patients who were severely frail (p¼0.41). Among people with diabetes, 28 (29.8%) had A1C between 7% and 8.5%, 50 (53.2%) had A1C <7% and 16 (17%) had A1C >8.5%; no difference by frailty level was observed. In patients without diabetes, A1C levels were not statistically different (p¼0.3) across levels of frailty.
4
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Figure 1. Mean admission glucose (mmol/L) stratified by frailty category. In patients with diabetes, admission glucose decreased with increasing frailty (p¼0.003 for main effect). CFS, Clinical Frailty Scale score.
Discussion This study shows that older inpatients with diabetes are more commonly frail than those without diabetes. This is unsurprising given the contribution of diabetes to poor functional status and accumulation of health deficits through many mechanisms. In logistic regression modelling with adjustment for age, sex and glucose, frailty was a predictor of in-hospital mortality and prolonged LOS, whereas diabetes status was not. Multiple studies endorse the association between higher frailty status and mortality both in community-dwelling adults and in inpatients; therefore, our conclusions are consistent with established literature (8,22). Frailty has also been associated with increased hospital LOS in other studies (23,24). Previous studies correlating type 2 diabetes and LOS did not control for frailty (16,17), which is a confounder. We showed that it is frailty, rather than diabetes status, that drives risk for both extended LOS and inpatient mortality. By establishing the correlation of frailty with inpatient mortality and LOS, this suggests that frailty should be assessed in all older inpatients to establish their risk for death and prolonged admission. It is not sufficient to simply assess their comorbidities including diabetes; one must have a global perspective of their frailty. A recent international position statement also endorses recognition and evaluation of frailty in older inpatients through CGA, emphasizing education, understanding and appropriate use of frailty assessment instruments (25). Our findings of less hyperglycemia on admission in more frail patients with diabetes are surprising given provincial guidelines for glucose management in long-term care facilities, a common living situation for frail older adults. We found that patients with diabetes with moderate and severe frailty had lower mean admission glucose than their nonfrail and mildly frail counterparts. There may be multiple mechanisms serving to explain this, including effects of antihyperglycemic agents, undernutrition, poor detection of signs and symptoms of hypoglycemia or physiologic reasons, such as sepsis, hepatic failure or renal failure (11,15,26,27). The findings of more frail patients with less hyperglycemia on admission are consistent with those of other studies, which found that most older adults with poor health were treated to A1C <7%, often using agents that increase risk for hypoglycemia (28,29). Medications that are known to cause hypoglycemia, such as sulfonylureas and insulin, are often covered for patients under provincial drug plans, whereas other agents with lower rates of hypoglycemia are not. Older adults often depend on provincial medication coverage
because of limited income and, therefore, can be limited in accessing lower-risk agents. Interestingly, all hypoglycemia in patients with diabetes was seen in those with moderate to severe frailty. This could indicate that patients who are frail are particularly vulnerable to hypoglycemia and may be useful for hypothesis generation regarding appropriate goals of antihyperglycemic therapy when a patient with diabetes becomes moderately to severely frail. In our institution, the rates of high-risk hypoglycemic medication use among frail older adults with diabetes at the time of admission to hospital is under ongoing study. The limitations of our study include its use of a database population in which a single clinical team gathered data from patients consulted to internal medicine, and it may not be representative of the entire older adult population seen in the emergency department of the tertiary care hospital. Admission glucose values were taken within 1 day of the CFS assessment, which may have led to detection of some effects of inpatient treatment of glucose, if given. In general, results were limited by sample size, particularly in correlating A1C with outcomes and in the analysis of the number of patients with hypoglycemia. The strengths of our study include the increasing clinical recognition of frailty in our institution and routine collection of frailty data in our hospital. Specialized teams are available for inpatient consultation to optimize care choices for patients who are frail, and frailty awareness is being incorporated into multidisciplinary and interprofessional training and research. Clinical practice and training allow greater recognition and more accurate assessment of frailty in the older adult population. Frailty assessment of all older inpatients should be done routinely, particularly in those with diabetes to adjust therapy, monitor for hypoglycemia and associate with hospital outcomes. Future studies in older inpatients with diabetes might include investigation of the impact of inpatient interventions on patients with diabetes of varying frailty status, such as the use of different classes of diabetes medications and of standardized insulin orders. There is little known about the outcomes of various blood glucose targets in the frail older adult population in both inpatient and outpatient settings, and further research may allow us to establish appropriate, safe targets that minimize the risk of symptom and complication progression while maintaining safety. Whether widespread early assessment of frailty during admission leads to improvement of outcomes remains to be determined. Early involvement of geriatric medicine consultants, home support services and other allied health disciplines in the care of those identified as frail may be beneficial. Many hospitals have geriatric consultants for inpatients and geriatric medicine clinics with interdisciplinary teams and/or geriatric day care programs for outpatients. Geriatric medicine expertise can also be sought from family physicians who have completed a year-long training program in the care of the elderly. For inpatients returning to the community or to long-term care, geriatric rehabilitation wards, transitional care wards or other supports including occupational therapy, physical therapy and social work may be available to ease this transition. For older adults living in the community, home support can be accessed on a no-cost or low-cost basis and can include varied services, such as medication management, food preparation and personal care depending on the functional capacity of each older adult. It is important to be aware of the resources available in one’s own community to assist with diabetes care and maintain the overall health of frail older adults. This study shows that patients with diabetes are more likely to be frail. Frailty was a robust predictor of in-hospital mortality and prolonged LOS in our cohort of patients 65 years of age, but diabetes status was not. We also found that patients with diabetes with moderate and severe frailty had lower glucose than those with no or mild frailty, assessed by admission serum glucose values.
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Diabetes at www. canadianjournalofdiabetes.com. Author Disclosures B.T. reports and discloses being a research investigator or subinvestigator with the following companies: GlaxoSmithKline, Novo Nordisk Canada, AMGEN, Sanofi-Aventis Canada, Ionis, BoehringerIngelheim, Novartis, AstraZeneca, Bristol-Myers Squibb, Intarcia, Lexicon, Merck, Eli Lilly, Pfizer, Takeda, NPS Pharmaceuticals, Cerenis Pharmaceuticals, Shire, Acasti Pharma and Aegerion. K.R. reports other from president and chief science officer, DGI Clinical, other from advisory board meeting, Lundbeck, other from associate director of the Canadian Consortium on Neurodegeneration in Aging and personal fees from various organizations, outside the submitted work. No other authors have any conflicts of interest to declare. Author Contributions H.T.M. performed the literature review, drafted the manuscript and edited the manuscript. B.T. contributed to study conception and design and edited the manuscript. K.R. provided access to the database. O.T. contributed to study design and performed statistical analysis. References 1. Umpierrez G, Pasquel F. Management of inpatient hyperglycemia and diabetes in older adults. Diabetes Care 2017;40:509e17. 2. Rockwood K, Tan M, Phillips S, McDowell I. Prevalence of diabetes mellitus in elderly people in Canada: Report from the Canadian Study of Health and Aging. Age Ageing 1998;27:573e7. 3. Umpierrez G. Hyperglycemia: An independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002;87: 978e82. 4. Houlden R, Capes S, Clement M, Miller D. In-hospital management of diabetes. Can J Diabetes 2013;37:S77e81. 5. Rockwood K, Rockwood M, Mitnitski A. Physiological redundancy in older adults in relation to the change with age in the slope of a frailty index. J Am Geriatr Soc 2010;58:318e23. 6. Rockwood K, Mitnitski A. Frailty defined by deficit accumulation and geriatric medicine defined by frailty. Clin Geriatr Med 2011;27:17e26. 7. Leng S, Chen X, Mao G. Frailty syndrome: An overview. Clin Interv Aging 2014; 19:433e41. 8. Basic D, Shanley C. Frailty in an older inpatient population: Using the Clinical Frailty Scale to predict patient outcomes. J Aging Health 2014;27:670e85.
5
9. Bandeen-Roche K, Seplaki C, et al. Frailty in older adults: A nationally representative profile in the United States. J Gerontol Ser A Biol Sci Med Sci 2015;70: 1427e34. 10. Rockwood K. A global clinical measure of fitness and frailty in elderly people. Can Med Assoc J 2005;173:489e95. 11. Meneilly G, Knip A, Miller D, Sherifali D, Tessier D, Zahedi A. Diabetes in older people. Diabetes Canada 2018 clinical practice guidelines for the prevention and management of diabetes in Canada: Diabetes in older people. Can J Diabetes 2018;42(Suppl. 1):S283e95. 12. Duckworth W, Abraira C, Moritz T. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009;360:129e39. 13. Patel A, MacMahon S, Chalmers J, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008;358: 2560e72. 14. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008;358:2545e59. 15. Diabetes Care Program of Nova Scotia. Diabetes guidelines for frail elderly residents in or awaiting long-term care (LTC): Pocket reference; 2018. http://dia betescare.nshealth.ca/sites/default/files/LTCPocketRefSept142018.pdf. Accessed July 29, 2019. 16. Kubo T, Fujino Y, Murata A, et al. Prevalence of type 2 diabetes among acute inpatients and its impact on length of hospital stay in Japan. Intern Med 2011; 50:405e11. 17. Comino E, Harris M, Islam M, et al. Impact of diabetes on hospital admission and length of stay among a general population aged 45 year or more: A record linkage study. BMC Health Serv Res 2015;15:12. https://doi.org/10.1186/ s12913-014-0666-2. 18. Clement S, Braithwaite S, Magee M, et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care 2004;27:553e91. 19. Ahmed B, Davis H, Laskey W. In-hospital mortality among patients with type 2 diabetes mellitus and acute myocardial infarction: Results from the National Inpatient Sample, 2000-2010. J Am Heart Assoc 2014;3:e001090. 20. Miguel-Yanes J, Jimenez-Garcia R, Hernandez-Barrera V, Mendez-Bailon M, de Miguel-Diaz J, Lopez-de-Andres A. Impact of type 2 diabetes mellitus on inhospital mortality after major cardiovascular events in Spain (2002-2014). Cardiovasc Diabetol 2017;16:126. 21. Jones D, Song X, Mitnitski A, Rockwood K. Evaluation of a frailty index based on a comprehensive geriatric assessment in a population based study of elderly Canadians. Aging Clin Exp Res 2005;17:465e71. 22. Eeles E, White S, O’Mahony S, Bayer A, Hubbard R. The impact of frailty and delirium on mortality in older inpatients. Age Ageing 2012;41:412e6. 23. Rose M, Pan H, Levinson M, Staples M. Can frailty predict complicated care needs and length of stay? Intern Med J 2014;44:800e5. 24. Juma S, Taabazuing M, Montero-Odasso M. Clinical Frailty Scale in an acute medicine unit: A simple tool that predicts length of stay. Can Geriatr J 2016;19: 34e9. 25. Sinclair AJ, Abdelhafiz A, Dunning T, et al. An international position statement on the management of frailty in diabetes mellitus: Summary of recommendations 2017. J Frailty Aging 2018;7:10e20. 26. Abdelhafiz A, Rodríguez-Mañas L, Morley J, Sinclair A. Hypoglycemia in older people - A less well recognized risk factor for frailty. Aging Dis 2015;6:156e67. 27. Furman E. Undernutrition in older adults across the continuum of care: Nutritional assessment, barriers, and interventions. J Gerontol Nursing 2006; 32:22e7. 28. Lipska K, Ross J, Miao Y, Shah N, Lee S, Steinman M. Potential overtreatment of diabetes mellitus in older adults with tight glycemic control. JAMA Intern Med 2015;175:356. 29. Meneilly G, Berard L, Cheng A, et al. Insights into the current management of older adults with type 2 diabetes in the Ontario primary care setting. Can J Diabetes 2018;42:23e30.
5.e1
H.T. MacKenzie et al. / Can J Diabetes xxx (2019) 1e5
Supplementary Figure 1. Clinical frailty scale. IADLs, instrumental activities of daily living. Adapted with permission from Dalhousie University, Division of Geriatric Medicine. Ó 2007-2009 Geriatric Medicine Research, Dalhousie University, Halifax, Canada.