Abstract
Objective
Existing studies have highlighted suboptimal diabetes management in residential aged care facilities (RACFs). However, understanding of diabetes management in Australian metropolitan RACFs has been limited. This retrospective cohort study aimed to explore the pharmacological management of diabetes in 25 RACFs in Sydney Australia and assess concordance with clinical practice guidelines (CPGs).
Methods
Data from 231 permanent RACF residents aged ≥65 years and over with type 2 diabetes mellitus over the period from 1 July 2016 to 31 December 2019 were used. Concordance was measured by assessing the medications and medical history data for each individual resident for concordance with evidence‐based CPGs. Multivariable logistic regression was used to estimate the effect of resident characteristics on concordance with CPGs.
Results
Of the 231 residents with diabetes, 87 (38%) were not taking any antidiabetic medication. Pharmacological management inconsistent with CPG recommendations was observed for 73 (32%) residents, with the most common reason for non‐concordance being the use of medications with significant adverse effects in older adults (47, 2%). Residents with hypertension or other heart diseases in addition to their diabetes had greater odds of their diabetes management being non‐concordant with CPGs (OR = 2.84 95% CI = 1.54, 5.3 and OR = 2.64, 95% CI = 1.07, 6.41, respectively).
Conclusions
Pharmacological diabetes management in metropolitan Australian RACFs is suboptimal, with a high prevalence of inconsistency with CPGs (32%) observed. Additionally, having hypertension or heart diseases significantly increased the possibility of non‐concordance among diabetic RACF residents. Further investigation into the underlying relationships with comorbidities is required to develop better strategies.
Keywords: antihypertensive agents, diabetes mellitus, residential aged care facility
Policy Impact
This study demonstrated a high proportion (32%) of Australian RACF residents with diabetes not receiving management concordant with clinical practice guidelines and a significant association between non‐concordance and having hypertension or heart diseases. These findings require increased attention from policymakers and care providers in improving diabetes management in RACF settings.
Practice Impact
There is an urgent need to provide personalised diabetes management plan concordant with clinical practice guidelines for RACF residents with diabetes. A particular focus should also be placed on diabetic RACF residents who suffer hypertension or heart diseases at the same time.
1. INTRODUCTION
Diabetes is characterised as the persistent elevation of blood glucose levels above a specified cut‐off, following the ingestion of carbohydrates. 1 Affecting more than 460 million people (7% of the world's population) in 2017, type 2 diabetes mellitus (T2DM) was the ninth highest cause of mortality across the globe. 2 The prevalence is similar in Australia and rises to 15% in those aged ≥65 years and over. 3 In Australia, 10%–20% of residents in residential aged care facilities (RACFs) have diabetes. 4 , 5 Multiple results from international and national studies analysing the management of diabetes in RACFs indicate suboptimal care, in the form of non‐concordance with guidelines and maintenance of unreasonably tight glycaemic controls. 4 , 5 , 6 , 7 , 8 , 9 , 10
Diabetes management involves non‐pharmacological methods (e.g., weight loss, physical activity, diet changes, decreased alcohol consumption and smoking cessation) and pharmacological therapy, both of which should be incorporated in resident care within RACFs. 11 Pharmacological therapy for diabetes, especially in older adults, needs to be very carefully administered and followed as they are at a high risk of complications, including hypoglycaemia. 6 , 9 Additionally, there are pharmacokinetic changes that develop as people age. Specifically with antidiabetic medications, it is important to consider the possible decline in renal function, in association with medication metabolism. 12
In ensuring that pharmacological therapy is appropriately adhered to, firstly it is important to comprehensively review the prescribed guidelines. Optimal pharmacological management for older adults with diabetes from seven clinical practice guidelines (CPGs) across the globe is summarised in Table S1. These CPGs indicate that metformin is the most commonly considered first‐line treatment unless it is contraindicated due to renal or cardiovascular impairments. Subsequent treatment is slightly different based on comorbidities and medication adverse effects. Additionally, having diabetes increases the risk of comorbidities (e.g., hypertension, hyperlipidaemia and renal impairment), and as a result, the guidelines also mention medication prescriptions for these conditions. 9 , 10
In Australia, the Report of the Royal Commission into Aged Care Quality and Safety in Australia reported information on the inappropriate administration of medications, often relating to incorrect times, doses and medications, along with staff not ensuring that residents actually take their medicines. 13 However, there has been limited research in Australia exploring the issue of diabetes management in RACFs, particularly in metropolitan setting. Existing studies are largely based on rural RACFs, only focusing on the management based on the haemoglobin A1c (HbA1c) levels and the medications prescribed. 4 , 5 For example, Haines et al. 4 identified issues with concordance to guidelines based on lower frequencies of HbA1c measurements in Australian rural RACFs, as well as longer periods of hyperglycaemia in residents. In a small study of rural RACFs, Khalil et al. 5 observed that only 41% of the residents with diabetes were managed in concordance with guidelines. This was due to most residents having HbA1c >7% and lack of antihypertensive, lipid‐lowering and antiplatelet medication in resident care. 5
Due to these findings, it is important to conduct further research and include metropolitan Australian RACFs, where the majority of RACF residents reside, to better support the clinical management of diabetes. The aim of this study was to assess pharmacological management of diabetes in Australian metropolitan RACFs, by exploring medications administered to residents with diabetes and evaluating the concordance of management with CPGs.
2. METHODS
2.1. Setting
We conducted a secondary retrospective cohort analysis using de‐identified electronic health record data from 25 RACFs, owned by a large not‐for‐profit aged care provider, in Sydney, New South Wales, Australia.
2.2. Ethics approval
The study was reviewed and approved by the Macquarie University Human Research Ethics Committee (No.: 52019614412614). All methods and analyses in this study were carried out in accordance with the principles of the Declaration of Helsinki.
2.3. Data
The data retrieved was dated 1 July 2016 to 31 December 2019 and included two databases: resident profile and daily medication administration. The resident profile dataset included resident demographics, stay details and medical history (free‐text fields with health conditions used to identify residents with diabetes). The medication dataset contained individual records (date and medication administered) for every single medication dose that was administered to a resident during their stay at the RACFs. Residents' medical history was retrieved to assess the associations of comorbidities on their diabetes care. Due to database restraints, medications with more than one antidiabetic agents in the same dose were also counted as one medication therapy.
2.4. Sample selection
The participant sample included residents ≥65 years old and over on entry into a RACF, had diagnosed diabetes, a minimum stay of 90 days and was a permanent resident (e.g., not respite or interim care residents). A minimum stay of 90 days was applied to ensure the resident had spent some time at the RACFs, to help iron out any initial issues with medications and comorbidities, possibly with a medication review. 14 Respite and interim care residents were excluded as their EHR medication data might not be comprehensive and differ between different facilities that they stayed at. The entry date criteria helped to ensure that the prescribing rules could be consistent for the residents assessed. A total of 231 residents met the inclusion criteria. The final dataset used for analysis consisted of a longitudinal sample since residents entered the RACFs and discharged/deceased at different times.
2.5. Concordance assessment
All the medications in the medication dataset were reviewed and antidiabetic medications used in Australia were identified using generic drug names. For each resident in the sample, the number of diabetic medications used within their stay were also identified.
For the concordance assessment, we reviewed the summarised CPGs in Table S1 and identified the following five criteria to appropriately assess within the dataset. These criteria were selected to ensure relevance to Australian prescribing.
use of second‐line medications without Metformin or Insulin;
use of medications with significant adverse effects in older adults (e.g., Sulfonylureas);
residents with hypertension not on any antihypertensives;
residents with dyslipidaemia on any lipid‐lowering medications; and
residents on SGLT2i plus other diuretics.
Based on these criteria, the medications and medical history data for each individual resident was assessed for concordance with CPGs. Additionally, it was assessed whether there was an association between the resident's demographics and medical history, with CPG concordance. All residents with diabetes were included in the concordance assessment, regardless of whether they were on antidiabetic medications. For example, a resident with non‐pharmacologically controlled diabetes would still need to be taking antihypertensives for known hypertension.
2.6. Statistical analysis
Descriptive statistics were calculated for all variables of interest. Median and interquartile range (IQR) were reported for the continuous variables and proportions for the categorical variables. The non‐parametric Wilcoxon rank sum test and the χ 2 tests were performed to compare resident characteristics of diabetic and non‐diabetic for the variables of interest.
Multivariable analysis for concordance with CPGs was also performed, focusing on the diabetic residents only. The dependent variable pertaining to compliance with CPGs was dichotomously coded as either 1 or 0, indicating that a resident was deemed to have complied with at least one concordance criterion listed above or had no instances of non‐compliance during their entire duration of stay, respectively. This means that residents' diabetes management can meet multiple concordance criteria. Multivariable logistic regression was used to estimate the effect of resident characteristics on concordance with CPGs. The resident sociodemographic details (age at entry to RACF, gender, place of birth and length of stay), comorbidities (dyslipidaemia, hypertension, heart disease, renal disease, delirium, osteoporosis, arthritis, circulatory disease, history of urinary tract infection[s] [UTI] and cognitive impairment), and complications (retinopathy, stroke and myocardial infarction) were examined as the covariates in the analysis. Backward stepwise selection method was used to select the important variables for the concordance with CPGs. Multicollinearity was also checked using variance inflation factor (VIF) criteria >3. Subgroup analysis by sex was conducted, and all results reported using odds ratio (OR) and 95% confidence intervals (95% CI) were calculated. All tests were considered statistically significant where the p‐value was <.05. Data were analysed using R version 4.2.1 in RStudio version 2022.7.2.576.
3. RESULTS
Of the 25 RACFs, there were 908 residents who fulfilled the criteria. Among those 908 residents, 231 (25%) had a diagnosis of diabetes. The residents with diabetes were observed to have a significantly lower median age at entry to a RACF compared to their non‐diabetic counterparts (84 vs. 87 years; p < .001). Additionally, residents with diabetes were more likely to be males (46% vs. 33%; p < .001) and born outside Australia (52% vs. 64%; p < .001) (Table 1). The residents with diabetes were more likely to have circulatory disease, UTI, renal disease and osteoporosis compared to non‐diabetic residents (Table 1).
TABLE 1.
Demographic and clinical characteristics.
| Characteristic | Total residents, N = 908 a N (%) | Non‐diabetic residents, N = 677 a N (%) | Residents with diabetes, N = 231 a N (%) | p‐Value b |
|---|---|---|---|---|
| Age at entry date | 86 (80, 91) | 87 (81, 91) | 84 (78, 89) | <0.001 |
| Age group | ||||
| 65–74 | 94 (10.4%) | 59 (8.7%) | 35 (15.2%) | <0.001 |
| 75–84 | 281 (30.9%) | 194 (28.7%) | 87 (37.7%) | |
| 85–94 | 449 (49.4%) | 347 (51.3%) | 102 (44.2%) | |
| ≥95 | 84 (9.3%) | 77 (11.4%) | 7 (3.0%) | |
| Gender | ||||
| F | 581 (64.0%) | 456 (67.4%) | 125 (54.1%) | <0.001 |
| M | 327 (36.0%) | 221 (32.6%) | 106 (45.9%) | |
| Residents born in Australia | 557 (61.3%) | 436 (64.4%) | 121 (52.4%) | 0.001 |
| Length of stay | 501 (267, 790) | 503 (268, 786) | 488 (264, 808) | 0.747 |
| Comorbidities | ||||
| Hypertension | 320 (35.2%) | 238 (35.2%) | 82 (35.5%) | >0.925 |
| Heart disease | 79 (8.7%) | 52 (7.7%) | 27 (11.7%) | 0.062 |
| Retinopathy | 118 (13.0%) | 93 (13.7%) | 25 (10.8%) | 0.255 |
| Stroke | 241 (26.5%) | 172 (25.4%) | 69 (29.9%) | 0.185 |
| Myocardial infarction | 3 (0.3%) | 3 (0.4%) | 0 (0.0%) | 0.575 |
| Circulatory diseases | 817 (90.0%) | 595 (87.9%) | 222 (96.1%) | <0.001 |
| Urinary tract infection(s) | 187 (20.6%) | 150 (22.2%) | 37 (16.0%) | 0.046 |
| Delirium | 112 (12.3%) | 79 (11.7%) | 33 (14.3%) | 0.296 |
| Dyslipidaemia | 46 (5.1%) | 32 (4.7%) | 14 (6.1%) | 0.425 |
| Renal disease | 183 (20.2%) | 120 (17.7%) | 63 (27.3%) | 0.002 |
| Osteoporosis | 240 (26.4%) | 191 (28.2%) | 49 (21.2%) | 0.037 |
| Arthritis | 507 (55.8%) | 375 (55.4%) | 132 (57.1%) | 0.643 |
| Cognitive impairment | 372 (41.0%) | 282 (41.7%) | 90 (39.0%) | 0.472 |
Note: Bold values indicate p values < .05.
Median (IQR); N (%).
Wilcoxon Rank Sum Test; Pearson's Chi‐Squared Test; Fisher's Exact Test.
3.1. Diabetic medications
Out of the 231 residents with diabetes, 87 (38%) had not taken any diabetic medications during their stay at the RACFs. Of residents who were on at least one antidiabetic medication (n = 144), 64 (28%) were on one, 36 (16%) on two and 44 (19%) were on three or more antidiabetic medications.
Table 2 presents the individual diabetic medication used. Metformin was the most common oral drug administered, with 50 (22%) residents using it. Sulfonylurea class drugs were administered to 31 residents (13%). Fifty‐six residents (24%) were administered different classes of insulin (Table 2).
TABLE 2.
Number of diabetic medications used by residents.
| Antidiabetic pharmacological class | Residents with diabetes on the medication a n (%) |
|---|---|
| DPP‐4 inhibitors | 19 (8) |
| GLP‐1 agonist | 1 (0) |
| Insulin | 56 (24) |
| Combinations | 16 (7) |
| Intermediate acting | 3 (1) |
| Long acting | 17 (7) |
| Rapid acting | 20 (9) |
| Metformin | 50 (22) |
| Metformin + DPP‐4 inhibitors | 11 (5) |
| Metformin + SGLT2 inhibitors | 1 (0) |
| SGLT2 inhibitors | 3 (1) |
| Sulfonylureas | 31 (13) |
| Thiazolidinediones | 1 (0) |
Note: The sum of the insulin subtype percentages is 24.2%.
Median (IQR); n (%).
3.2. Concordance
Based on the concordance criteria, 73 (32%) residents' medication management was non‐concordant, and 158 (68%) residents' medication use was concordant with CPGs. The most common element of non‐concordance was the use of medications with significant adverse effects in older adults (47 [20%] residents). Medications included in this category were sulfonylurea and thiazolidinedione class drugs. Twenty‐two residents (10%) were not on any antihypertensives, despite being diagnosed with hypertension. None of the residents were non‐concordant with the condition of using SGLT2i with other diuretics (Table 3).
TABLE 3.
Summary of concordance to diabetes clinical practice guidelines (CPGs).
| Criteria | Number of residents n (%) |
|---|---|
| 1. Use of second‐line medications without metformin or insulin | |
| Concordant | 214 (93) |
| Non‐concordant | 17 (7) |
| 2. Use of medications with significant adverse effects in older adults | |
| Concordant | 184 (80) |
| Non‐concordant | 47 (20) |
| 3. Residents not on any antihypertensives | |
| Concordant | 209 (91) |
| Non‐concordant | 22 (10) |
| 4. Residents not on any lipid‐lowering medications | |
| Concordant | 227 (98) |
| Non‐concordant | 4 (2) |
| 5. Residents on SGLT2i plus other diuretics | |
| Concordant | 231 (100) |
We observed that there were significant differences in the demographics and comorbidities of residents receiving concordant and non‐concordant management (Table 4). A higher percentage of residents (29%) with non‐concordant management had ≥5 comorbidities, in comparison with residents with concordant management (16%).
TABLE 4.
Concordance to diabetes CPGs by demographics and co‐morbidities.
| Criteria | Total, N = 231 a | Concordant residents, N = 158 a | Non‐concordant residents, N = 73 a | p‐value b |
|---|---|---|---|---|
| Age at entry date of the first admission | 84 (78, 89) | 83 (77, 89) | 85 (78, 89) | 0.312 |
| Age group | ||||
| 65–74 | 35 (15.2%) | 26 (16.5%) | 9 (12.3%) | 0.437 |
| 75–84 | 87 (37.7%) | 63 (39.9%) | 24 (32.9%) | |
| 85–94 | 102 (44.2%) | 65 (41.1%) | 37 (50.7%) | |
| ≥95 | 7 (3.0%) | 4 (2.5%) | 3 (4.1%) | |
| Gender | ||||
| F | 125 (54.1%) | 85 (53.8%) | 40 (54.8%) | 0.888 |
| M | 106 (45.9%) | 73 (46.2%) | 33 (45.2%) | |
| Residents born in Australia | 121 (52.4%) | 77 (48.7%) | 44 (60.3%) | 0.103 |
| Length of stay (days) | 488 (264, 808) | 470 (260, 792) | 505 (304, 840) | 0.412 |
| Number of co‐morbidities | ||||
| 1 | 10 (4.3%) | 10 (6.3%) | 0 (0.0%) | 0.029 |
| 2 | 40 (17.3%) | 30 (19.0%) | 10 (13.7%) | |
| 3 | 67 (29.0%) | 49 (31.0%) | 18 (24.7%) | |
| 4 | 67 (29.0%) | 43 (27.2%) | 24 (32.9%) | |
| ≥5 | 47 (20.3%) | 26 (16.5%) | 21 (28.8%) | |
Note: Bold values indicate p values < .05.
Median (IQR); n (%).
Wilcoxon rank sum test; Fisher's exact test; Pearson's Chi‐squared test.
3.3. Demographic factors associated with concordance
The multivariable logistic regression found that concordance status was significantly associated with whether a resident had hypertension (p < .001) or other heart disease (p = .03) (Table 5). Specifically, we found that with other covariates aforementioned controlled, residents would experience a 184% (OR = 2.84, 95% CI = 1.54, 5.31) higher risk of being non‐concordant with CPGs for their diabetes management if they suffer hypertension, compared to the situations if they lived without hypertension. Similarly, residents would experience a 164% (OR = 2.64, 95% CI = 1.07, 6.41) higher risk of their diabetes management being non‐concordant with CPGs if they suffer heart diseases, compared to the situations if they were heart‐disease‐free. Specific gender subgroup analysis showed that males would be at greater risk of having non‐concordant diabetes management if they lived with hypertension or arthritis, compared to the situations if they had no such conditions (respectively hypertension OR = 3.02, 95% CI = 1.06, 8.99; and arthritis OR = 2.95, 95% CI = 1.15, 8.10). Similarly, females would have a higher risk of being non‐concordant with CPGs if they had hypertension (OR = 2.79, 95% CI = 1.09, 7.53), compared to the situations if they were free from hypertension.
TABLE 5.
Multivariate analysis for concordance based on demographics and co‐morbidities.
| Independent variables | OR | SE | p‐Value | 95% CI |
|---|---|---|---|---|
| Residents with diabetes | ||||
| Hypertension | 2.84 | 0.31 | <0.001 | (1.54, 5.31) |
| Heart disease | 2.64 | 0.45 | 0.03 | (1.07, 6.41) |
| Males | ||||
| Hypertension | 3.02 | 0.54 | 0.04 | (1.06, 8.99) |
| Arthritis | 2.95 | 0.49 | 0.03 | (1.15, 8.10) |
| Females | ||||
| Hypertension | 2.79 | 0.49 | 0.04 | (1.09, 7.53) |
Note: Myocardial Infarctions were removed as a variable from the full model, as there were no residents with diabetes who had them in their medical history. Bold values indicate p values < .05.
4. DISCUSSION
4.1. Pharmacological management
Our study showed that 38% of residents with diabetes were not receiving antidiabetic medications. This is consistent with prior studies in rural Australian RACFs, showing that 35%–44% of older adult residents with diabetes were not being managed with medications and with diet only. 4 , 5 It was also shown that, in RACFs of New Zealand 8 and Norway, 15 28% and 26% of residents with diabetes, respectively, were managed with diet alone. However, while 38% of the residents with diabetes were not receiving antidiabetic medications, it is important to note that some of these residents might be on non‐pharmacological management (e.g., weight loss and physical activity). Given that older age is a risk factor for severe hypoglycaemia, many recent CPGs for diabetes management in older adults suggest relaxing glycaemic targets—from HbA1c of 7.0%–7.5% to 8.0%, especially if there are high levels of frailty or significant comorbidities. 11 , 16 As a result, it is possible that health‐care providers are cautious and prefer non‐pharmacological diabetes management for RACF residents, particularly if they are not experiencing significant micro‐ and macro‐vascular complications of poorly controlled diabetes.
We also found that metformin alone, followed by sulfonylureas, was the most commonly administered oral antidiabetic medication in pharmacologically managed residents; however, insulin was the most administered medication overall. These results are not consistent with the CPGs for older adults with diabetes, since metformin is the accepted first‐line medication, and sulfonylureas are not recommended. 11 , 16 , 17 , 18 , 19 It is possible that the people on insulin trialled metformin without effect, or they had suboptimal renal function for metformin prescription. 20 Sulfonylureas administration for older adults with diabetes is advised against due to the associated risk of causing hypoglycaemic events and reduced efficacy in people with age‐induced beta cell failure. 11 , 16 , 17 , 19 , 20
4.2. Concordance
Another important finding is that management for 32% of residents with diabetes was not concordant with national and international CPGs. The predominant elements of non‐concordance were firstly, the use of thiazolidinediones and sulfonylureas, which are not recommended in older people, and secondly the lack of antihypertensive medication use in residents with hypertension.
Thiazolidinediones, despite their high efficacy in maintaining glycaemic targets, contribute to increased oedema, incidence and worsening of heart failure, and specifically in females, higher risks of fractures. 16 , 17 Only one male resident in the cohort of this study was administered pioglitazone, a type of thiazolidinedione, and he did not have renal or heart disease, hence the administration could be considered reasonable. Additionally, in the cohort, two types of sulfonylureas, gliclazide and glibenclamide, were administered. While sulfonylureas as a class are generally considered high risk for older individuals with diabetes, it is also important to consider the individual drug variations within this class. For example, it has been noted that glibenclamide, which was administered to only one of the residents in the cohort, is associated with a significantly higher risk of causing hypoglycaemic events, in comparison to gliclazide. 17 , 18
In addressing the concordance issue about lack of antihypertensive use, it is important to highlight the ongoing debate about stringent hypertension control outcomes in older adults with diabetes. Among studies in literature, blood pressure target varies between <130/90 and <160/90, depending on the comorbidities, overall condition and risk of cardiovascular disease. 21 , 22 , 23 Blood pressure targets need to be lower to prevent the adverse outcomes of cardiovascular diseases. 21 , 22 However, lower targets carry the risk of causing orthostatic hypotension, making closer monitoring essential. 21 , 22 Irrespective of the blood pressure targets, it is noted that antihypertensive management is generally recommended for older adults with diabetes. 21 , 22 , 23 Refraining from deprescribing antihypertensives in older adults is suggested, unless there are significant adverse effects directly resulting from them. 24
4.3. Factors affecting concordance
We also found that non‐concordance of diabetes management to the CPGs was closely related to whether the resident also had hypertension or heart disease. There is limited research on factors affecting concordance of diabetic management with CPGs in RACFs, though a few studies on inconsistency of universal diabetes care with CPGs have found associated factors to be number of comorbidities, resource availability, staff training and quality checks, and existence of multiple guidelines. 7 As aforementioned, a significant proportion of the residents with diabetes and hypertension were not on any antihypertensives, and this was one of the concordance criteria assessed. Consequently, it is reasonable that having hypertension increased the likelihood of residents' diabetes management being non‐concordant with CPGs. Additionally, all diseases grouped under the term ‘heart disease’ as part of our study have a common risk factor of hypertension. Therefore, it is possible that in a similar way to hypertension, having heart disease also raised the likelihood of non‐concordant diabetes management for residents within our cohort. One of the implications of this finding is not to treat hypertension in insolation but in the context of the patient's overall cardiovascular risk factor. In a broader context, there can be other reasons why hypertension can increase likelihood of diabetes care inconsistent with CPGs, such as polypharmacy, lack of facility policies and medication adverse effects. 25 For example, hypertension, as with many other chronic conditions, commonly requires a multiple medication regimen. The subsequent polypharmacy is specifically found to be a risk factor in older for poor adherence to medications. 26 Evidence shows that in Australia, as well as in many other countries, 27 , 28 the utilisation of medications and the prevalence of polypharmacy in RACFs have experienced a steady increase over recent decades. This trend is particularly notable among residents with extended lengths of stay in these facilities. The rise in medication use and polypharmacy poses a heightened risk of medication interactions and adverse effects within the RACF population, presenting challenges for effective diabetes management in this demographic. This issue is further compounded by the evolving landscape of RACF residency duration in Australia, where the median length of stay has progressively extended from 16.8 months in 2012 to 22.4 months in 2021. 29
In comparing the factors that might affect the concordance of diabetes management to CPGs, by gender, it was found that hypertension was significant in both genders independently, and in males, arthritis was another significant comorbidity. Arthritis, specifically rheumatoid, is managed with medications, including glucocorticoids and disease‐modifying anti‐rheumatic drugs (DMARD), which increase the risk of iatrogenic diabetes. 30 However, further research is needed to study the relationship between arthritis and concordance to CPGs for existing diabetes.
4.4. Strengths and limitations
The strengths of this study include the sample size, which was significantly larger than previous Australian studies, 4 , 5 and the method of concordance assessment. We firstly summarised the primary points from multiple national and international CPGs, and subsequently created specific concordance criteria to assess. In contrast, previous Australian studies assessed diabetes management concordance with CPGs by addressing HbA1c levels, health‐care practitioner appointments, and whether the residents were prescribed medications for hypertension, hyperlipidaemia and prophylactic antiplatelet therapy. 4 , 5
There are limitations to the present study. There are multiple ways of measuring concordance in the context of diabetes management. First, certain cardiovascular health guidelines recommend that all adults with diabetes who are 65 years and over, being high risk, should automatically be prescribed antihypertensives and lipid‐lowering medications. 31 However, our study focused on the concordance to CPGs that are specific to diabetes. Second, due to the restrictions in the initial data obtained from the RACFs, it was not possible to differentiate T1DM from T2DM. This could have affected the CPG concordance outcomes of our study since most of the guidelines we addressed focused on T2DM. However, national and international studies indicate that usually only up to 5% (generally 2%–3%) of the residents in RACFs have T1DM. 4 , 5 , 32 Additionally, it is also noted in the literature that the prevalence of T1DM decreases with age so T1DM should be less common in RACF residents. 32 Third, another limitation was the lack of data about non‐pharmacological diabetes management for the residents; therefore, the study also did not address pertinent facility level variables (e.g., RACF geographical location and staffing). Fourth, the landscape of diabetes management undergoes continual evolution with the emergence of new research, technology and treatment modalities. As our dataset only extends up to December 2019, any medications introduced thereafter are not accounted for in our analysis. This limitation may constrain the interpretation and generalisability of our findings. Fifth, while the clinical practice guidelines used offer valuable guidance for diabetes management of older adults, they are not explicitly tailored for the unique context of residential aged care settings. Residents of such facilities often present with multiple comorbidities, necessitating individualised and patient‐centred approaches to diabetes management. Consequently, certain deviations from guideline recommendations may be justified for RACF residents, given their distinct clinical needs and circumstances. For example, residents with dementia are often advised to have their diabetes medications de‐intensified as the dementia progresses. However, such adjustment might not affect residents' concordance status assessment in this study except the co‐existence of other diseases requires a full cessation of or exit from diabetes medications. Nevertheless, using a binary classification of diabetes management (concordance vs non‐concordance) might be inappropriate in the clinical context without providing detailed context for non‐concordant residents with guidelines. This might be a good opportunity to include the role of GPs to provide the tailored approach in the context of the best evidence/recommendations, their clinical judgement and patient expectations/values (e.g., Venn diagram of Evidence‐based practice https://researchguides.library.tufts.edu/ebm). Sixth, non‐use of antihypertensives may be appropriate for individuals with documented hypertension under specific circumstances, such as implementing non‐pharmacological methods for residents with well‐controlled blood pressure and low cardiovascular risk, or deprescribing antihypertensive medications due to adverse effects. However, due to the unavailability of data, we were unable to investigate these circumstances and their potential effects on concordance assessment.
5. CONCLUSIONS
This study provides new evidence about diabetes management in older adults using data from Australian metropolitan RACFs. The results demonstrated that the management of 32% of the residents with diabetes was not concordant with CPGs and having hypertension or heart diseases significantly increased the possibility of non‐concordance. Our study strengthens the literature on suboptimal pharmacological diabetes management in RACFs, based on inconsistency with CPGs. The relationship between residents' diabetes management and their chronic comorbidities requires further investigation so that relevant strategies can be developed.
CONFLICT OF INTEREST STATEMENT
No conflicts of interest declared.
Supporting information
Table S1
ACKNOWLEDGEMENTS
We thank the partners and collaborators of this project including Anglicare, Northern Sydney Local Health District, Sydney North Primary Health Network, the Deeble Institute for Health Policy Research, and the Australian Aged Care Quality and Safety Commission. Open access publishing facilitated by Macquarie University, as part of the Wiley ‐ Macquarie University agreement via the Council of Australian University Librarians.
Bhalla H, Huang G, Seaman K, et al. Diabetes in residential aged care: Pharmacological management and concordance with clinical guidelines. Australas J Ageing. 2024;43:792‐801. doi: 10.1111/ajag.13351
DATA AVAILABILITY STATEMENT
The data used in this study were collected by the aged care provider (Anglicare), and restrictions apply to the availability of these data, which were used under licence for the current study. According to the Data Management Plan approved by the Macquarie University Human Research Ethics Committee Medical Sciences Committee (Ref No: 52019614412614) and NSW Population and Health Services Research Ethics Committee (Ref: 2020/ETH00166/2020.10), only the named project team members approved on the ethics application will have access to the data extracted and collected during the period of the research project. Therefore, the data used in this study cannot be deposited in the public data repository to be freely accessed by the public. However, upon reasonable request and with the permission of Anglicare, data can be shared and requested from the authors (Guogui Huang, guogui.huang@mq.edu.au).
REFERENCES
- 1. Westman EC. Type 2 diabetes mellitus: a pathophysiologic perspective. Front Nutr. 2021;8:707371. doi: 10.3389/fnut.2021.707371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Khan MAB, Hashim MJ, King JK, Govender RD, Mustafa H, Al KJ. Epidemiology of type 2 diabetes – global burden of disease and forecasted trends. J Epidemiol Glob Health. 2020;10(1):107‐111. doi: 10.2991/jegh.k.191028.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Australian Institute of Health and Welfare . Older Australia at a Glance. Australian Institute of Health and Welfare; 2018. [Google Scholar]
- 4. Haines HM, Bannon‐Murphy H, Amos T, Krones R. Prevalence and management of diabetes in residential aged care facilities in north‐east Victoria, Australia. Aust Fam Physician. 2016;45(12):908‐911. [PubMed] [Google Scholar]
- 5. Khalil H, Tan L, George J. Diabetes management in Australian rural aged care facilities: a cross–sectional audit. Australas Med J. 2012;5(11):575‐580. doi: 10.4066/AMJ.2012.1411 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bouillet B, Vaillant G, Petit JM, et al. Are elderly patients with diabetes being overtreated in French long‐term‐care homes? Diabetes Metab. 2010;36(4):272‐277. doi: 10.1016/j.diabet.2010.01.009 [DOI] [PubMed] [Google Scholar]
- 7. Da Porto A, Coracina A, Fiore V, et al. Quality of care to institutionalized patients with diabetes in Italy: a national survey. Eur Geriatr Med. 2020;11(5):753‐759. doi: 10.1007/s41999-020-00329-8 [DOI] [PubMed] [Google Scholar]
- 8. Gill EA, Corwin PA, Mangin DA, Sutherland MG. Diabetes care in rest homes in Christchurch, New Zealand. Diabet Med. 2006;23(11):1252‐1256. doi: 10.1111/j.1464-5491.2006.01976.x [DOI] [PubMed] [Google Scholar]
- 9. Gomez‐Huelgas R, Perez‐Belmonte LM, Rivera‐Cabeo I, Morilla‐Herrera JC, Bellosta‐Ymbert JM, Bernal‐Lopez MR. Management of elderly patients with type 2 diabetes in long‐term care and skillednursing facilities. Pol Arch Intern Med. 2019;129(2):137‐140. doi: 10.20452/pamw.4410 [DOI] [PubMed] [Google Scholar]
- 10. Jeon SY, Shi Y, Lee AK, et al. Fingerstick glucose monitoring in veterans affairs nursing home residents with diabetes mellitus. J Am Geriatr Soc. 2021;69(2):424‐431. doi: 10.1111/jgs.16880 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. The Royal Australian College of General Practitioners . Management of Type 2 Diabetes: a Handbook for General Practice. RACGP; 2020. [Google Scholar]
- 12. Gates BJ, Walker KM. Physiological changes in older adults and their effect on diabetes treatment. Diabetes Spectr. 2014;27(1):20‐28. doi: 10.2337/diaspect.27.1.20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. National Ageing Research Institute . Royal Commission into Aged Care Quality and Safety – Final Report: care, Dignity and Respect. Vol 2. Commonwealth of Australia; 2021. [Google Scholar]
- 14. Sluggett JK, Bell JS, Lang C, et al. Residential medication management reviews in Australian residential aged care facilities. Med J Aust. 2021;214(9):432‐433. doi: 10.5694/mja2.50921 [DOI] [PubMed] [Google Scholar]
- 15. Andreassen LM, Sandberg S, Kristensen GBB, Sølvik UØ, Kjome RLS. Nursing home patients with diabetes: prevalence, drug treatment and glycemic control. Diabetes Res Clin Pract. 2014;105(1):102‐109. doi: 10.1016/j.diabres.2014.04.012 [DOI] [PubMed] [Google Scholar]
- 16. American Diabetes Association . 12. Older adults: standards of medical care in diabetes—2022. Diabetes Care. 2022;45(Supplement_1):S195‐S207. doi: 10.2337/dc22-SINT [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee . 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:S283‐S295. [DOI] [PubMed] [Google Scholar]
- 18. IDF Working Group . Managing Older People with Type 2 Diabetes [Internet]. International Diabetes Federation; 2012. [Google Scholar]
- 19. Strain WD, Hope SV, Green A, Kar P, Valabhji J, Sinclair AJ. Type 2 diabetes mellitus in older people: a brief statement of key principles of modern day management including the assessment of frailty. A national collaborative stakeholder initiative. Diabet Med. 2018;35(7):838‐845. doi: 10.1111/dme.13644 [DOI] [PubMed] [Google Scholar]
- 20. Diabetes UK . Good Clinical Practice Guidelines for Care Home Residents with Diabetes. Diabetes UK; 2010. [Google Scholar]
- 21. Solini A, Grossman E. What should Be the target blood pressure in elderly patients with diabetes? Diabetes Care. 2016;39(Supplement_2):S234‐S243. doi: 10.2337/dcS15-3027 [DOI] [PubMed] [Google Scholar]
- 22. LeRoith D, Biessels GJ, Braithwaite SS, et al. Treatment of diabetes in older adults: an Endocrine Society* clinical practice guideline. J Clin Endocrinol Metab. 2019;104(5):1520‐1574. doi: 10.1210/jc.2019-00198 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Grossman A, Grossman E. Blood pressure control in type 2 diabetic patients. Cardiovasc Diabetol. 2017;16(1):3. doi: 10.1186/s12933-016-0485-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Sheppard JP, Benetos A, McManus RJ. Antihypertensive deprescribing in older adults: a practical guide. Curr Hypertens Rep. 2022;24(11):571‐580. doi: 10.1007/s11906-022-01215-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Blackburn DF, Wilson TW. Antihypertensive medications and blood sugar: theories and implications. Can J Cardiol. 2006;22(3):229‐233. doi: 10.1016/s0828-282x(06)70902-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Lloyd CE, Mughal S, Roy T, et al. What factors influence concordance with medications? Findings from the UK Asian diabetes study. Diabet Med. 2014;31(12):1600‐1609. doi: 10.1111/dme.12554 [DOI] [PubMed] [Google Scholar]
- 27. Page AT, Potter K, Lee G, et al. Medicines regimens charted for older people living in residential aged care: a repeated cross‐sectional study describing the number of medicines, regimen complexity, high‐risk prescribing, and potential underprescribing. J Am Med Dir Assoc. 2024;25:104944. [DOI] [PubMed] [Google Scholar]
- 28. Jokanovic N, Tan EC, Dooley MJ, Kirkpatrick CM, Bell JS. Prevalence and factors associated with polypharmacy in long‐term care facilities: a systematic review. J Am Med Dir Assoc. 2015;16(6):535.e1‐12. [DOI] [PubMed] [Google Scholar]
- 29. Gen Aged Care Data . People leaving aged care. Website. https://www.gen‐agedcaredata.gov.au/topics/people‐leaving‐aged‐care. Published 2023. Accessed May 2, 2024.
- 30. Lillegraven S, Greenberg JD, Reed GW, et al. Immunosuppressive treatment and the risk of diabetes in rheumatoid arthritis. PLoS One. 2019;14(1):e0210459. doi: 10.1371/journal.pone.0210459 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. National Vascular Disease Prevention Alliance . Guidelines for the Management of Absolute Cardiovascular Disease Risk. National Vascular Disease Prevention Alliance; 2012. [Google Scholar]
- 32. Menke A, Orchard TJ, Imperatore G, Bullard KM, Mayer‐Davis E, Cowie CC. The prevalence of type 1 diabetes in the United States. Epidemiology. 2013;24(5):773‐774. doi: 10.1097/EDE.0b013e31829ef01a [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1
Data Availability Statement
The data used in this study were collected by the aged care provider (Anglicare), and restrictions apply to the availability of these data, which were used under licence for the current study. According to the Data Management Plan approved by the Macquarie University Human Research Ethics Committee Medical Sciences Committee (Ref No: 52019614412614) and NSW Population and Health Services Research Ethics Committee (Ref: 2020/ETH00166/2020.10), only the named project team members approved on the ethics application will have access to the data extracted and collected during the period of the research project. Therefore, the data used in this study cannot be deposited in the public data repository to be freely accessed by the public. However, upon reasonable request and with the permission of Anglicare, data can be shared and requested from the authors (Guogui Huang, guogui.huang@mq.edu.au).