7 / 40
SA JOURNAL OF DIABETES & VASCULAR DISEASE
RESEARCH ARTICLE
VOLUME 16 NUMBER 2 • NOVEMBER 2019
49
methodological limitations include choice of participants in terms
of age, and the use of insensitive outcome measures. For example,
studies investigate fall risk mainly in older adults with diabetes
and therefore age-related deterioration is a confounder.
3
In other
studies, there is a lack of interrogation of how multiple sensory
deficits (neuropathy and retinopathy), which are highly prevalent
in diabetes patients, impact on fall risk.
4
And lastly, there is limited
research from developing countries such as South Africa, where
there is a high prevalence of diabetes and socio-economic factors
that further increase the risk of fall-related injuries.
7,8
It is important to evaluate and diagnose risk for falls early, to
enable prevention and rehabilitation of injury and associated
debilitating effects thereof. This study therefore aimed to add to
the knowledge base in fall-risk identification and prevention in
young to middle-aged adults with diabetes in a low- to middle-
income country such as South Africa.
Methods
The study aims were to determine (1) the fall risk in young to middle-
aged adults (20–55 years of age) with diabetes and compare to
those without diabetes; (2) the associations between fall risk and
characteristics of diabetes including type, duration (in years), control
(glycaemic status in g/mol documented in participants’ files), age,
gender and diabetes co-morbidities (peripheral neuropathy and
vision difficulties).
The study utilised an observational, cross-sectional, matched-
groups design with a cohort (patients with diabetes) and control
(volunteers without diabetes) group of participants. The sample size
required for this study was determined using a G-power analysis
calculator with a power of 0.95 and an error probability of 0.05,
and the targeted sample size for this study was 222 individuals; 111
participants per group.
9
All participants (cohort and controls) were selected to participate
in this study based on the following inclusion criteria: clinically
confirmed diagnosis of diabetes of either type (cohort group);
and above the age of 18 and below 55 years. The upper limit of
55 years was selected in order to avoid the impact of age-related
deterioration (> 55 years).
10
The exclusion criteria (established through case history) were:
(1) no prior use of ototoxic drugs to avoid vestibulotoxicity, which
may influence fall risk.
11
A history of all medication taken was
documented in the participants’ files. (2) No history of head injury,
radiotherapy to the head or ear surgery, as these may influence
fall risk;
12
(3) no clinical diagnoses or reports of neurological
impairments that may influence fall-risk findings, such as multiple
sclerosis, cerebrovascular accidents (ischaemic and haemorrhagic
strokes), Parkinson’s disease and ataxia.
13
Participants were recruited from a primary healthcare clinic in
Limpopo province, Polokwane, which has the highest poverty rate
in comparison to other South African provinces; 78.9% of the
population live below the national poverty line and it accounts
for 2.8% of the national diabetes prevalence numbers in people
aged 25 years and older. Sampling was carried out using purposive
and convenience sampling for the cohort and control group,
respectively. Participants in both groups were matched for age and
gender to allow for comparison between the two groups.
Ethics clearance was obtained from the University of Cape Town’s
Faculty of Health Sciences Human Research Ethics Committee
(HREC134/2015) prior to the commencement of the study. This
study adhered to the ethical principles outlined in the Declaration
of Helsinki (2013) throughout, which included transparency and
integrity of data.
14
Data were obtained through a participant case-history
interview, a medical-folder review (to obtain information related
to participants’ diabetes type and status), as well as assessment
for fall risk through static and dynamic balance tests with the
Dynamic Gait Index (DGI) and the Modified Clinical Test of Sensory
Integration (M-CTSIB).
15-17
Participants were screened for possible
diabetes complications, including peripheral neuropathy, with the
diabetic neuropathy symptoms (DNS) score, and vision screening
was done with the Snellen E visual screening chart.
18,19.
In this study there were two main data-collection tools used to
assess static and dynamic balance to quantify fall risk as the main
variable of interest using the DGI and M-CTSIB assessments.
15-17
The
DGI is commonly used clinically and in research, and can predict
dynamic balance disorders and fall risk. It is reported to have
adequate discriminative ability with 0.84 for sensitivity and 0.89
for specificity.
16
The M-CTSIB assesses balance function in terms
of sensory integration and static balance with a 0.99 test–retest
reliability and 0.68–1.0 inter-rater reliability.
17
Both the DGI and
M-CTSIB were administered as described in the literature to ensure
instrument validity.
15,17
Statistical analysis
Proportions (%) were used to report participant and diabetes
characteristics, information on diabetes complications, and fall-
risk prevalence data from both DGI and M-CTSIB assessments.
Pearson’s correlation coefficient determined the strength of
associations between the presence of fall risk and characteristics
of diabetes (control and duration), patient age and presence
of co-morbidities. Independent
t
-tests were used to determine
significance of differences in prevalence between the participant
groups (
α
= 0.05).
Results
A total of 192 individuals participated in this study; 110 in the cohort
(participants with diabetes) and 82 in the control (without diabetes)
group. There were similar distributions of gender in both groups,
with the following age distributions for each group: cohort: median
46 years, range 20–55; control: median 43 years, range 21–55.
Overall, the majority (49%) of participants were 49 years or younger.
The difference in distribution of the ages between the cohort and
control groups was not statistically significant across all age bands.
The majority (92%) of the cohort participants presented with type
2 diabetes and only 8%had type 1 diabetes. Most (74%) participants
had an uncontrolled glycaemic status (diabetes control, defined as
7 g/mol, was determined by a medical doctor and documented in
the patient file). The duration of disease ranged from a month up to
33 years post-diabetes diagnosis, with most living with diabetes for
less than five years. In terms of complications, more than half of the
participants in the cohort group had a possible complication with
their diabetes, where 51% screened positive for diabetic neuropathy,
while 56% failed the vision screening.
The DGI assessment consisted of eight tasks with varying
demands, with each item scored on a four-level ordinal scale, with
a maximum possible score of 24. A score of 19 or less indicated
decreased dynamic balance and therefore an increased risk of
falling.
15,16