Volume 20, Issue 3 , Pages 158-162, May 2006
Falls as a complication of diabetes mellitus in older people☆
Article Outline
- Abstract
- 1. Introduction
- 2. Methods
- 2.1. Statistical methods
- 3. Results
- 4. Discussion
- Acknowledgments
- References
- Copyright
Abstract
Objectives
The aims of this study were to determine the incidence of falls in a group of elderly patients with diabetes and to assess for the prevalence of risk factors for falls in this population.
Design
This is a population-based study with questionnaire-based interviews.
Setting
The setting for this study was the London District General Hospital outpatient department.
Participants
Seventy-seven patients with diabetes, aged over 65 years, randomly selected whilst attending for general diabetic annual review. Patients with dementia, blindness, and immobility and those who were unable to give informed consent were excluded from this study.
Measurements
The incidence of falls in the last 12 months was used. Information was collected on the incidence of hypoglycaemic episodes, the presence of other medical conditions, visual impairment, and peripheral neuropathy, the use of medications and walking aids, and HbA1C and blood pressure control.
Results
The incidence of falls was 39%. Falls occurred more frequently in female patients and patients of increasing age. Falls occurred more frequently in patients with poor diabetic control [risk ratio (RR)=7.83 (2.948–20.799), χ2 value=6.422]; patients requiring assistance with mobility: for those mobile with a stick [RR=1.839 (1.048–3.227), χ2=4.619]; and those who had previously suffered a stroke [RR=1.929 (1.143–3.257), χ2 = 4.615].
Conclusion
We provide evidence that poorly controlled diabetes and conditions associated with complications of diabetes are associated with an increased risk of falling in older people. We recommend early recognition of the multiple causes of falls in the older diabetic patient and prompt referral of this group of patients to a specialist falls clinic.
Keywords: Older people, Falls, Diabetes
1. Introduction
Falls are a major cause of disability and a preventable cause of death in older people. About 30% of people over 65 years of age fall each year; the incidence of falls in those over 75 years of age is 32–42% (Tinetti & Speechley, 1989). Diabetes mellitus is also common in older people. It has been estimated that approximately 50% of the patients with diabetes are over 65 years of age (Morley, 1998). The prevalence of diabetes in elderly individuals in the UK was estimated at between 11% and 14% (Croxson, 2002); prevalence will vary depending on the date of study, population, and method of determining diabetes. Diabetic patients over 65 years old are nearly three times more likely than nondiabetic participants to be hospitalised in a given year (Zaida & Alexander, 2001). The prevention of falls is therefore important to reduce morbidity and mortality in this age group. The importance of identifying specific risk factors and the value of a multidisciplinary assessment of patients who fall are increasingly being recognised (Chang et al., 2004, Palmer, 2001). In the National Service Framework for Older People, the importance of assessment and management of falls is strongly emphasised (National Service Framework Older People, 2001). The most efficient fall-prevention programmes target high-risk groups (Close et al., 2003). Diabetic complications lead to impairments, which would constitute recognised risk factors for falls (Bueno-Cavanillas et al., 2000). It would therefore appear that older diabetic patients would be a suitable target group for a strategy aimed at preventing falls.
We hypothesised that complications of diabetes may lead to an increased risk of falls in older diabetic patients, making falls an indirect complication of diabetes. The aim of this study was to establish the incidence of falls and to define risk factors associated with falls.
2. Methods
Seventy-seven patients aged 65 years and over with Type I or II diabetes were recruited by random selection, whilst attending for general diabetic annual review at Homerton University Hospital, a district general hospital in the east of London, between February and June 2002. Approval for the project was given by the research ethics committee of East London and The City Health Authority. Patients with dementia (documented after consultation with a psychogeriatrician), blindness, and immobility and those who were unable to give informed consent were excluded from this study. Although it is recognised that the first three conditions may contribute to falling, demented patients were unable to respond appropriately to the questionnaire, and blindness and immobility would confound the data.
An information sheet was provided to explain the purpose of the study, and written informed consent was obtained. Each patient participated in a structured questionnaire-based interview. They were asked questions relating to the duration of diabetes, monitoring, their perception of occurrence of hypoglycaemic episodes, symptoms and signs of postural hypotension, other illnesses, and medications. Patients were asked questions regarding the use of walking aids and their perception of visual impairment and peripheral neuropathy. They were also asked about the occurrence of falls within the past 12 months (“How many times, if at all, have you had a fall in the last year?”). A fall was defined as an unintentional change in body position resulting in contact with the ground or lower level, not as a result of a major intrinsic event (e.g., stroke) or overwhelming hazard (e.g., car accident; Tinetti, Speechley, Ginter, 1988); this was explained to the patient. Information was sought regarding contact with other medical specialties.
We subsequently obtained information about glycaemic control, blood pressure, and the presence of clinically determined complications of diabetes from the hospital records by looking at the documentation of the annual review. Visual impairment was defined by decreased vision on a Snellen Chart, retinopathy by findings on dilated fundoscopy, and neuropathy by abnormal monofilament score. We defined the presence of sensory impairment as either self-report on the questionnaire or by documented clinical evidence in the medical records, or both.
2.1. Statistical methods
Analysis of each category was undertaken to obtain a relative risk ratio (RR) for falling, with 95% confidence intervals. Chi-squared tests were used to determine the significance of each potential risk factor comparing participants who had fallen with those who had not. A P value of less than .05 indicates statistical significance.
3. Results
Table 1 shows the demographics and data regarding diabetes control. The average age of our cohort was 73 years (65–85 years). Twenty-seven patients (35%) were aged 75 years or older, 50 (65%) were between 65 and 74 years. There were 32 males and 45 females.
Table 1. Demographics and diabetic control
| Falls (%) [n=30] | No falls (%) [n=47] | Total (%) [n=77] | Falls/no falls ratio (95% confidence interval) | |
|---|---|---|---|---|
| Female | 23 (77) | 22 (47) | 45 (58) | 2.336 (1.144 to 4.766) |
| Male | 7 (23) | 25 (53) | 32 (42) | 0.428 (0.21 to 0.73) |
| Age >75 | 14 (47) | 13 (27) | 27 (35) | 1.62 (1.014 to 2.586) |
| Lives alone | 13 (43) | 16 (34) | 29 (38) | 1.265 (0.729 to 2.196) |
| Lives with one other | 16 (53) | 28 (60) | 44 (57) | 0.857 (0.491 to 1.495) |
| Lives in an institution | 1 (3) | 3 (6) | 4 (5) | 0.633 (0.112 to 3.521) |
| Frame to mobilize | 2 (7) | 0 (0) | 2 (3) | 2.679 (1.998 to 3.593) |
| Stick to mobilize | 17 (57) | 15 (32) | 32 (42) | 1.839 (1.048 to 3.227) |
| Mobilize independently | 11 (37) | 33 (70) | 44 (57) | 0.434 (0.241 to 0.783) |
| Smoker | 5 (17) | 5 (11) | 10 (13) | 1.34 (0.670 to 2.682) |
| >14 Units alcohol/week | 3 (10) | 0 (0) | 3 (4) | 0.670 (0.497 to 0.904) |
| Tablets only | 11 (37) | 21 (45) | 32 (42) | 0.814 (0.452 to 1.466) |
| Diet only | 0 (0) | 4 (9) | 4 (5) | 0 |
| Insulin only | 13 (43) | 14 (30) | 27 (35) | 1.699 (0.969 to 2.980) |
| Insulin and tablets | 6 (20) | 6 (13) | 12 (16) | 1.354 (0.708 to 2.591) |
| HBA1C >7% | 26 (87) | 28 (60) | 54 (70) | 7.83 (2.948 to 20.799) |
| Incidence of hypoglycaemic episode | 15 (50) | 14 (30) | 29 (38) | 1.655 (0.957 to 2.862) |
Thirty patients (39%) had suffered at least one fall in the last year. Females were at greater risk of falling, as were patients in the older group. Of the patients who had falls, none of the fallers were managed by diet alone. Poor diabetic control (HbA1C >7%) was associated with falling, with a significant χ2 value of 6.422. There was a high prevalence of reported symptoms considered by the patients to be caused by hypoglycaemic episodes within the group as a whole; this was not a significant risk factor for falls (χ2 value of 3.187). Dependency on a walking aid was another significant risk factor: Mobilizing with a stick was significantly associated with falling (χ2=4.619). It was noted that all three patients who admitted to drinking more than 14 units of alcohol a week were in the group who fell.
Table 2 gives information on sensory deficit, coexistent morbidities, and polypharmacy. Visual impairment and peripheral neuropathy were prevalent in both groups, more common among fallers, but this difference did not reach statistical significance. The only comorbidity that showed significant association with falls was previous stroke, with a χ2 value of 4.615.
Table 2. Sensory deficit, coexistent morbidities, and polypharmacy
| Falls (%) [n=30] | No falls (%) [n=47] | Total (%) [n=77] | Falls/no falls ratio (95% confidence interval) | |
|---|---|---|---|---|
| Visual impairment | 26 (87) | 32 (68) | 58 (75) | 2.129 (0.852 to 5.317) |
| Peripheral neuropathy | 15 (50) | 15 (32) | 30 (39) | 1.567 (0.904 to 2.716) |
| SBP >135 mm Hg | 21 (70) | 36 (77) | 57 (74) | 0.819 (0.453 to 1.480) |
| DBP >85 mm Hg | 17 (57) | 22 (47) | 39 (51) | 1.275 (0.723 to 2.249) |
| Ischaemic heart disease | 9 (30) | 14 (30) | 23 (30) | 1.006 (0.547 to 1.851) |
| Osteoarthritis | 13 (43) | 19 (40) | 32 (42) | 1.075 (0.613 to 1.886) |
| Stroke | 9 (30) | 5 (11) | 14 (18) | 1.929 (1.143 to 3.257) |
| Pacemaker | 1 (3) | 2 (4) | 3 (4) | 0.851 (0.167 to 4.323) |
| >4 Medications | 21 (70) | 23 (49) | 44 (57) | 1.75 (0.925 to 3.300) |
4. Discussion
This study examined the prevalence of being a faller within a group of older diabetic patients and sought to investigate the association of being a faller with a range of putative risk factors within this population. This is the first study to prove that a significant relationship exists between poor diabetic control and falls.
This study has several limitations. The study is retrospective. We did not enquire into the number of falls in those who had fallen, and a longer period of recall may have yielded a higher percentage of fallers. We studied a relatively small number of participants, which may not represent the average population of older diabetic patients, limiting generalisability.
The results confirm that falls are prevalent among older diabetic patients. Our figure of 39% is in line with other studies into the incidence of falls in all older people (Masud & Morris, 2001). However, we would expect the figure to be higher, as we are proposing that diabetes increases the risk of falls. It would therefore appear that estimation of falls by patient self-reporting is too low. We identified an increased risk in the older subset of diabetic patients, and we have shown that women are at greater risk of falling than men are. It is not clear why women should be predisposed to falls, but this finding has previously been documented (Schwartz et al., 2002).
The majority of patients were managing their diabetes with insulin, tablets, or a combination of both. Glycaemic control was not good, as evidenced by HbA1C values; a fear of hypoglycaemic episodes, which was prevalent in both groups of patients, may lead to less strict control, or it may be that older people have poorer control of their diabetes generally, which would act as a confounding variable. The results of the UKPDS trial have shown that treatment to improve overall glucose control significantly reduces the risk of microvascular and macrovascular complications (UKPDS 38, 1998). We demonstrated the presence of both microvascular and macrovascular complications in our population. Previous studies have shown that visual impairment and peripheral neuropathy are important risk factors for falls (UKPDS 35, 2000, UKPDS 36, 2000), consistent with the data presented here. Visual impairment is particularly widespread within our population, and although this factor did not show statistical association with falls, many patients reported that they felt that it contributed to their fall. It might similarly be expected that peripheral neuropathy could lead to impaired balance and, therefore, falls, and 50% of those who fell in our study reported suffering from this complication of diabetes.
Those who required walking aids were at greater risk of falls. Peripheral neuropathy may contribute as outlined above. We demonstrated that suffering a previous stroke was a significant risk factor for falls. Blood pressure control in the group was suboptimal; three quarters of patients had readings over the recommended 135/85 (UKPDS 38, 1998). All but 1 of the 14 people who had suffered a previous stroke had blood pressure readings above the recommended limits (data not shown). Hypertension is a risk factor for stroke and, therefore, may indirectly contribute to falls in this population; however, postural hypotension has also been associated with hypertension (Krolewski et al., 1985).
Diabetes often coexists with other chronic disease in older people. As a consequence of this, these patients are prescribed many tablets (polypharmacy), the side effects of which may increase their risk of falling. A frequently observed side effect of antihypertensive medication is postural hypotension, a recognised risk factor for falls. Oral or injectable hypoglycaemic agents may induce a level of hypoglycaemia that may precipitate collapse. We have acknowledged the existence of these potential confounding factors in our questionnaire; that there may be more than one cause for a fall is one of the challenges pertinent to elderly care medicine, a focus of the falls prevention clinic.
The association with alcohol and falls should be noted. Although only three patients admitted to drinking over the 14 units that we set as the weekly limit, it is possible this figure is an underrepresentation, as some patients may be unwilling to disclose this information or underestimate their consumption. All three patients had fallen. If their alcohol consumption was excessive, it could have contributed to hypoglycaemia in the presence of diabetes.
Only four patients surveyed came from an institution (residential or nursing home). This is a small number and may be because diabetic monitoring is the responsibility of a visiting general practitioner in many institutions. Our sample population may therefore underrepresent such patients. The participants in our study were relatively young (three quarters of them aged less than 75 years) and we did not see many frail or very old patients in the outpatient department. This group of patients is likely to be at high risk of falls; they may have osteoporotic bones and, as such, be at greater risk of injury (Wei, Hu, Wang, & Hwang, 2001).
A large number of patients had regular appointments with an ophthalmologist and chiropodist, but very few had contact with the physiotherapist, occupational therapist, or elderly care medicine specialist (see Fig. 1; “involvement” indicates that the patient has seen this health care professional at least once within the last year). This reflects the medical profession's vigilance towards diabetes care, but lack of recognition of risk for falling in the same population. Of the 30 patients who had fallen, 11 required hospital treatment for an injury sustained during a fall; although drawn from a small sample, this figure is higher than reported for the population as a whole. Previous studies have estimated the number of patients seeking medical help following a fall to be between 10% and 25% (Berg et al., 1997, Campbell et al., 1990). Only three had been referred for physiotherapy, one for occupational therapy and one to the falls clinic (data not shown).
Fig. 1.
Involvement of healthcare professionals—percent of patients. Ophth—ophthalmology; physio—physiotherapy; OT—occupational therapy; f/u—follow.
Falls in older people place heavy demands on healthcare systems (Alexander, Rivara, & Wolf, 1992), and interventions that target the risk factors contributing to falls are highly recommended by the government (NSF Older People, 2001). The British Diabetic Association guidelines on the structure of diabetes services recommend a specialist diabetes geriatrician in each district (Alexander, 1999). Reducing the number of falls in older people depends on identifying those most at risk of falling and coordinating appropriate preventative action. Many people who fall do not seek medical help, but they may be identified as being at risk through the presence of risk factors.
In conclusion, reporting of falls was more common among those with poor glycaemic control. Prospective study is needed to establish if better control reduces subsequent falls. The presence of diabetes in an older person should raise the possibility of increased risk for falls, and the annual diabetic clinic review provides an opportunity for screening. Early recognition of specific risk factors, patient education, and intervention, including referral for specific falls prevention programmes such as strength and balance training, may reduce the incidence of falls in this population (Chang et al., 2004).
Acknowledgments
We would like to thank all the patients that participated in this study. We also thank the department of Metabolic Medicine at Homerton University Hospital for their support. We are grateful to all the staff in the diabetes clinic and medical records for their assistance in carrying out the study in a busy diabetic clinic.
References
- . The cost and frequency of hospitalization for fall related injuries in older adults. American Journal of Public Health. 1992, Jul;82(7):1020–1023
- Alexander, W., for the BDA. (1999). Recommendations for the structure of specialist diabetes care services; a British Diabetic Association report. London.
- . Circumstances and consequences of falls in independent community dwelling older adults. Age and Ageing. 1997, Jul;26(4):261–268
- Risk factors in falls among the elderly according to intrinsic and extrinsic precipitating causes. European Journal of Control. 2000;16(9):849–859
- Circumstances and consequences of falls experienced by a community population aged 70 years and over. Age and Ageing. 1990, Mar;19(2):136–141
- Interventions for the prevention of falls in older adults: Systematic review and meta-analysis of randomised clinical trials. British Medical Journal. 2004, Mar 20;328(7441):680
- Predictors of falls in a high risk population: Results for the prevention of falls in the elderly trial (PROFET). Emergency Medicine Journal. 2003, Sep;20(5):421–425
- . Diabetes in the elderly: Problems of care and service provision. Diabetic Medicine. 2002;19(Suppl. 4):66–72
- Hypertension, orthostatic hypotension and the microvascular complications of diabetes. Journal of Chronic Diseases. 1985;38(4):319–326
- . Epidemiology of falls. Age and Ageing. 2001, Nov;30(Suppl. 4):3–7
- . The elderly Type 2 diabetic patient; special considerations. Diabetic Medicine. 1998;15(Suppl. 4):541–546
- National service framework for older people. London, UK: Department of Health; 2001;www.doh.gov.uk/nsf/olderpeople
- . Falls in elderly patients: Predictable and preventable. Cleveland Clinic Journal of Medicine. 2001, Apr;68(4):303–306
- Older women with diabetes have a higher risk of falls: A prospective study. Diabetes Care. 2002, Oct;25(10):1749–1754
- . Prevention of falls among the elderly. New England Journal of Medicine. 1989, Apr 20;320(16):1055–1059
- . Risk factors for falls among elderly people living in the community. New England Journal of Medicine. 1988, Dec 29;319(26):1701–1707
- Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes prospective observational study. British Medical Journal. 2000;321(7258):405–412
- Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (prospective observational study). British Medical Journal. 2000;321(7258):412–419
- . Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes. UK Prospective Diabetes Study Group. British Medical Journal. 1998;317(7160):703–713
- . Fall characteristics, functional mobility and bone mineral density as risk factors of hip fracture in the community-dwelling ambulatory elderly. Osteoporosis International. 2001, Dec;12(12):1050–1055
- . Falls in the elderly: Identifying and managing peripheral neuropathy. Nurse Practitioner. 2001;26(3):86–88
☆ No funding was sought for this study.
PII: S1056-8727(05)00057-7
doi:10.1016/j.jdiacomp.2005.06.004
© 2006 Elsevier Inc. All rights reserved.
Volume 20, Issue 3 , Pages 158-162, May 2006
