Vitamin B12 deficiency can cause severe problems with the blood, nerves, brain and psychological well-being. Ironically, our modern methods for the control of diabetes mellitus can actually contribute to vitamin B12 deficiency. This is because the diabetic medication "metformin", low-cholesterol diets lacking in meats (a natural source of vitamin B12) and the use of powerful anti-gastric medication can all reduce the natural absorption of vitamin B12 from the diet, especially in elderly people with diabetes.

There is both a high prevalence of vitamin B12 deficiencies and falls among the elderly with type 2 diabetes mellitus and the investigators hypothesize that B12 deficiency contributes directly and significantly to falls in elderly diabetics through impaired muscle strength, gait and balance.

This study therefore proposes to investigate the association between vitamin B12 deficiency and fall risk among diabetic elderly patients (older than 65 years) in the polyclinic setting by assessing muscle strength, balance and walking speed. The predictors of vitamin B12, folate, homocysteine and vitamin D levels will also be explored in this study.

If the hypothesis is right, this would be of public health importance & can lead to further studies that can change the way we treat diabetes by reducing falls in our elderly diabetics through the screening for, prevention and treatment of B12 deficiency.

Singapore is a fast-aging Nation. A national survey of senior citizens in 2005 projected that the prevalence of persons >65 yrs. in Singapore will increase from 7.3% in 2000 to 18.7% in 2030. Moreover, 7.1% of adults from 65 to 74 years were at least dependent on walking aids for mobility and this prevalence increased to 22.3% (17.5% for Males & 26% for females) for adults aged 75 and above.

The Prevalence of Diabetes Mellitus is very high among the Elderly in Singapore. The National Health Survey in 2010 showed that almost 1 out of 3 (29.1%) of our elderly aged 60-69 have Diabetes mellitus.

In Marine Parade Polyclinic in 2007, 36.3% (i.e. more than 1/3) of our patients were aged 65 yrs. or older. Moreover, a recent (2011) study by the Ministry of Health (MOH) on 2,600 community-dwelling elderly people in Marine Parade showed that 78.4% suffer at least 1 chronic disease and 15% had fallen in the 12 months leading to the MOH Survey.

A prospective cohort study of 63,257 Singapore local Chinese showed "a very strong dose-dependent relationship between duration of diabetes and risk of hip fracture (P for trend < 0.0001" and that "prevention of falls should be part of the management of diabetes".

A study profile of 2847 Acute Presentations at the Tan Tock Seng Hospital A & E Dept. by the geriatric population showed that "Falls" was the most common presenting complaint (13.9%).

A general-practice-register & UK-based study showed the prevalence of B12 deficiency to be approximately 10% for the ages of 65-74 yrs. and rising to 20% for those > 75 yrs. of age.

A cross-sectional study on 203 Type 2 diabetic outpatients of an American Primary Care Clinic showed the prevalence of B12 deficiency to be 22%, with patients taking Metformin having a statistically lower level of serum B12 (314.4 pmol/l vs 389.3pmol/l; P=0.012).

A causal link between vitamin B12 deficiency and falls is hypothesised. B12 deficiency can cause neuropathy which can in turn lead to impaired balance & proprioception, amyotrophy & subsequent sarcopenia that can all contribute to falls. Vitamin B12 deficiency can also lead to depression, consequent psychomotor retardation and subsequent falls. Low vitamin B12 can lead to raised serum homocysteine and this could in turn lead to neuro-cardiovascular complications (e.g. cognitive decline, stroke) that predispose to falls. At a molecular level, homocysteine is also known to permanently degrade and impair structural integrity of collagen, elastin and proteoglycans, thus predisposing to muscular-skeletal mechanical instability, osteoporosis, falls and fractures.

However, other than a few case reports, there are no published studies to date linking Vitamin B12 deficiency to falls.

The actual degree /severity of Vitamin B12 deficiency that could put a patient at risk of falls is not known.

The precise mechanism with which B12 deficiency could increase the risk of falls is also not known. Knowledge of such a mechanism could enable better Measurement and subsequent risk stratification of fall risk. This, in turn, could help guide in screening and pre-emptive treatment for falls prevention especially in elderly diabetics. The investigators I postulate that vitamin B12 deficiency exerts its fall risk through muscle weakness (sarcopenia) and through gait, balance & mobility impairment that can be objectively measured using dynamometry and POMA scoring.

As referenced, diabetes mellitus is associated with both vitamin B12 deficiencies and increased risk of falls. The investigators postulate that vitamin B12 deficiency is not only associated with but also has a causative role in increasing the risk of falls due to its biological plausibility outlined above.

This study aims to investigate if there is an association between muscle strength and gait & balance assessments (using the POMA Scores) with serum vitamin B12 levels in elderly diabetics of Marine Parade Polyclinic. Once a correlation can be shown, an effect size can be obtained for a clinical trial to see if vitamin B12 replacement could prevent, or better still, improve muscle strength, gait and balance, thus ultimately reducing risk of falls.

The secondary aim is to investigate the magnitude of any gender disparity in such an association.

An exploratory aim of this study is to identify potential predictor variables of vitamin B12 and folate levels that could help form the basis of future clinical predictive tools for vitamin B12 and folate deficiencies respectively that could help case-find vitamin deficiencies in primary care. Vitamin D and homocysteine levels will also be similarly explored.

If the hypothesis is correct, this would be of public health importance and could change diabetic clinical practice. Healthcare providers could then better identify, risk-stratify, screen & treat for vitamin B12 deficiencies especially among elderly diabetics and in doing so, reduce the heavy socio-economic burden of falls cost-effectively.

Methodology

This was done in 2 phases. In Jan 2012 11 patients were recruited using a scholarship stipend. This provided provisional data to enable calculation of the sample size for the second phase (an additional 45 patients, see below) and to obtain a bigger grant to do so. The recruitment criteria and study methodology being the same for both except for the change in knee extension strength measuring equipment and the addition of a swaymeter to quantify balance in addition to the POMA score. Serum albumin and total protein were also measured in the 45 patients of the second phase.

Patients' details & data were be rendered anonymous & identified by alpha-numeric codes.

Data was cross-tabulated and analysed using version 13 of STATA, a statistical software.

Serum vitamin B12, folate, homocysteine and 25-OH-vitamin D levels were tested and analysed as continuous and as categorical variables (e.g. as tertiles & quartiles) with the physical measurements (i.e. muscle strength measures and POMA scores) as dependent variables. Physical measurements were also be studied as continuous (egg Torque or Kg Force, total sway in mm distance) or categorical (e.g. tertiles of muscle strength, High-Medium-Low-risk ordinal scores) variables. Correlation was looked for using scatterplots, ANOVA and multivariable regression methods. As the investigators anticipated a wide range of Vitamin B12 levels from normal to the very low in the diabetic elderly population, it would then also be possible to compare between participants with and without vitamin B12 deficiency, and patients with normal B12 levels could serve as "internal controls".

Potential confounders & effect modifiers (e.g. additive or multiplicative Interactions) were be studied via stratification and linear regression analyses.

Folate was also tested and studied because it shares many of the Biochemical Pathways of Vitamin B12 and could serve as a confounder. Excesses of Folate can also mask B12 deficiency.

Homocysteine was used as a marker for biochemically-significant vitamin B12 deficiency because B12 deficiency is known to exert a significant part of its deleterious effects directly through homocysteine mechanisms.

Many factors that predispose to vitamin B12 deficiency also predispose to vitamin D deficiency (e.g. meat-deficient diets, malabsorption and old age). In addition, vitamin D deficiency is a known cause of falls in the elderly through sarcopenia & muscle weakness. Serum 25-OH vitamin D is thus measured to study its potential as a confounder or effect modifier in Muscle weakness and poor Gait-Balance.

Reduced lower limb strength and grip strength are associated with frailty, falls and disability. The Lafayette hand-held dynamometer using a supine positioning offers a validated feasible, inexpensive, and portable test of quadriceps muscle strength for use in healthy older people in the clinical setting. This was used in my pilot on 11 patients. However, the Lafayette Hand-held Dynamometer for leg strength, which is a cheaper and reliable measure for knee extension strength, is limited by the strength of the tester. This could affect the reliability of the results of leg strength especially when a female nurse uses the hand to test leg strength of stronger male participants. The Knee Strength Device and Lord Sway Meter Setup has been shown to provide valid and reliable measurements that can be used for assessing falls risk and evaluation of effectiveness of interventions and a range of health care settings. These are also easy, cheap and reliable to use and have the additional potential to be used for geriatric health care provision in the local primary health care setting in the future.

The Jamar Grip Strength Dynamometer has been used reliably and extensively in studies of Grip strength & increased fall risk.

The Tinetti Performance Oriented Mobility Assessment (POMA) scoring system has also been used extensively and shown to accurately assess impairments in Gait and Balance that lead to increased fall risk. The Timed-Up-and-Go (TUG) test is incorporated as a component of the POMA test.

For this study, the abovementioned tools of measurement are chosen over other methods because of their portability, validation, reliability, simplicity of use and potential for practical future clinical application in the primary health care setting, where daily high-patient loads are the norm and quick-and-reliable risk-stratification and assessment tools are required.

Sample Size Calculation

As there was hitherto no similar study associating levels of vitamin B12 and folate with muscle strength, gait and balance measures in older people with diabetes mellitus in Singapore. Thus, to ascertain an effect size, provisional analysis of the first 11 patients showed a lower vitamin B12 level to be associated with a longer time to complete the TUG test with a Pearson correlation of 0.34. With a type I error of 0.05 and power of 0.8, a total of 66 patients should have been recruited to validate this level of correlation. However, a sample size of 56 patients was chosen, as a reasonable number for a pilot study. This number of patients would still yield significant results at a Pearson correlation of 0.37 or better.

As in any normal physical examination that involves walking, there was a small theoretical risk of falls during the POMA assessment procedure. This was be minimized by stopping the study if the patient was deemed unsafe in walking or the patient refused to proceed further with the study. Risk of falls was minimized further with adequate prompting and briefing for anticipation of manoeuvres, and adequate rehearsals and demonstration. A safety gait belt was used for increased protective grip-hold of a patient in the event of an impending fall or instability.

The muscle strength testing especially of the lower limbs could cause straining if patient overexerted. In addition to proper briefing, warm-ups, trial runs and demonstration, additional care was taken by adopting an isometric "Make" test (as opposed to a "Break" test) to test patients' strength with, as it was less likely to cause muscle straining. A "Make" test is where the patient exerts a maximal force isometrically limited by his/ her own comfort & pain thresholds. A "break" test involves full extension & locking of a limb (e.g. the knee in full extension) and forcibly overcoming and flexing the limb by the assessor.

Patients were encouraged not to be in a fasted state at recruitment. If they are in a fasted state, they would be advised to have a meal and then return 1 hour post-meal.

Inclusion Criteria:

All Diabetic Adults > 65 years old who have been on follow-up at the Polyclinic for > 1 year; with Informed Consent.

Exclusion Criteria:

• Patients not fitting the Eligibility Criteria
• Inability to walk despite assistive devices because of musculoskeletal disorders &/or
• Acute or Terminal Illness &/or
• Documented severe Disorders of the central nervous system (e.g. major stroke, Parkinson disease, dementia & severe visual impairment) however a subject with a previous history of stroke could be included provided he/she had recovered with no cognitive impairment or other residual effects that would affect the study results.
• Inability to understand spoken questions &/or commands
• Failure to give or withdrawal of Informed Consent
• Patients not fitting the Eligibility Criteria
• Patients who are part of the Family Physician Clinic (FPC), thus avoiding conflict of interest.

stratified sampling to ensure equal number of men and women in the study. So 23 men and 23 women recruited in total.

Community-living elderly patients > 65 years of age receiving primary care from the polyclinic for the treatment of type 2 diabetes mellitus for at least a year.

Plans are being made to share the de-identified data. However, the exact portal with which to share the data has not been decided upon and various options are being considered currently. Meanwhile, datasets are available form the principal investigator on reasonable request and with permission from the SingHealth Centralized Institutional Review Board.