Fall and Balance Outcomes After an Intervention to Promote Leg Strength, Article Critique Example

Introduction and Research Purpose

This study in Physical Therapy fills in an important research lacuna regarding the use of therapy to increase individuals’ ability to bear weight, and assessing the potential relationship between increased exercise levels with falls. Numerous studies have shown the benefits of exercise for individuals with diabetes mellitus: Exercise can not only help improve an individual’s ability to self-regulate glycemic control, but also improve cardiovascular outcomes, including decreasing cardiovascular disease (Boule, Haddad, Kenny, et al., 2000).   Although exercise has proven to be an important therapeutic option, some patients with diabetes suffer from sensate fee (peripheral neuropathy); increased exercise, although helping various symptoms associated with the disease, can also lead to higher risk for falling (American Geriatrics Society, British Geriatrics Society, and American Academy of Orthopaedic Surgeons Panel on Falls Prevention, 2001). This study tries to combine these analytical threads in order to assess whether increased exercise levels in individuals with diabetes and peripheral neuropathy are at increased risk for falling.  This study is important because it is the first with (arguably) adequate power to assess the relationship between these two variables using the “gold standard” (randomized control trial). Although there was one previous (non-randomized study) assessing how certain exercise therapeutics affected mobility in these individuals, few studies have assessed the risk involved with increased exercise- particularly for foot ulcers and amputation.  Overall, this study’s research purpose was to assess how increased exercise in individuals would lead to greater falls and probability for disease.

Identification of Population, Subjects and Examiners

Individuals with both diabetes mellitus and neural pathology were the study’s targeted population.  Subjects for the study were found in the state of Missouri who were at least 50 years of age; they also needed to have a confirmed diagnosis of type 1 or type 2 diabetes mellitus (Kruse, LeMaster, Madsen, 2010).  Other inclusion criteria included: did not take part in weight bearing sessions of activity greater than 20 minutes more than twice a week; demo0nstrated absent sensation on at least 1 out of 10 possible sites with each point; and lost vibratory sensation in at least one foot (Kruse, LeMaster, Madsen, 2010).

This particular trial, called the “Feet First Trial” was an observer masked, randomized controlled trial of an individually adapted behavior change physical activity intervention (Kruse, LeMaster, Madsen, 2010).  For the “Feet First study”, 79 individuals with diabetes mellitus and neural pathology were recruited over 18 months and randomly allocated to intervention and control groups (Kruse, LeMaster, Madsen, 2010).  Of 260 potential participants, 105 declined participation; 76 of the remaining 155 potential participants did not meet the inclusion criteria, leaving 79 patients left for randomization.  The remaining 79 participants were then randomized and divided into two different groups: 1) Control group (N=38); 2) Treatment group (n=41).

Study’s Model and Operationalization of Certain Models

At baseline, the study’s principal investigators collected a number of variables.  Those baseline variables included: age; sex; marital status; education; ethnicity; smoking status; type and duration of diabetes mellitus, comorbid illness; and availability of health insurance foot (Kruse, LeMaster, Madsen, 2010). These were the independent variables included in the study.

Falls and data related to individual’s strength and flexibility were also included in the study.  The investigators used a number of different tests to capture an individual’s balance and fall probability.  Fall risk was disaggregated into three different variables: 1) The Berg Balance  Scale – a test consisting of 14 items including standing , reaching, turning to look over on one’s shoulder, tapping toes on  a stool, and performing progressive balance challenges; 2) The Timed “Up and Go” Test- participants rise from a seated position, walk forward 3 m turn around, walk back to their sear, turn around, and sit down without assistance.; 3) One leg static stance test- participants stand on one leg without assistance (Kruse, LeMaster, Madsen, 2010).

The principal investigators also collected variables for “motor strength.”  Motor strength was measured using the right-angle dynamometry- a test that uses a handheld dynamometer to test how strong the ankle is by using the “break” method.

Finally, investigators gathered data related to falls.  The “Falls Efficacy Test” measures an individual’s confidence in achieving 10 activities without falling.  The Foot Function Index Disability Scale is a test that measures the effect of foot pathology on function in terms of disability and restriction (Kruse, LeMaster, Madsen, 2010).

Although the variables operationalizing falls, such as fall risk and motor strength, were the putative dependent variables in this study, they were not regressed on in a traditional regression model but rather compared in the control and treatment groups.

Statistical and Research Results

In order to understand more about the study population, and potential differences between the two different study arms, the investigators conducted two types of analysis: 1) Baseline difference tests;  2) Difference and regression tests regarding the nature of falls between the two arms   Among baseline categorical variables, chi-square analyses were conducted to detect potential differences; for continuous variables, two variable t-tests were conducted; for continuous variables (with non-normal distributions), a Mantel-Haenszel chi-squared test was conducted (Kruse, LeMaster, Madsen, 2010).  The investigators found no differences at baseline between the control and treatment group.

In order to measure differences in falls between the control (non-exercise) and treatment groups (exercise), three different analyses were conducted: 1) falls were analyzed using a Mantel-Haenszel chi-squared test between the two groups; 2) Poisson regression was conducted with falls as the outcome variable and the log of days to standardize them over time; 3) Survival analysis to make sure that all falls were being reported- particularly by those who fell often (Kruse, LeMaster, Madsen, 2010).

Regarding the incidence of falls between the two groups, there was no statistically significant difference between the groups before the study; during the time of follow-up, the number of falls was not significantly different for those individuals with diabetes mellitus who received strengthening exercise and for those who didn’t receive therapy for strengthening.  Thus, this randomized trial found that the number of falls did not increase with an increase in exercise in the study population.  The study also found that the intervention had no assistance in decreasing the number of falls (Kruse, LeMaster, Madsen, 2010).

Clear Statement of Research Limitations and Internal and External Validity

The study’s investigators noted a number of limitations in the study’s methodology.  First, the study aimed at detecting differences in physical activity between groups rather than fall rates (Kruse, LeMaster, Madsen, 2010). Because the group selected for the intervention did not experience many falls before or after, it could be that the intervention given to a different group might result in a noticeable difference in falls.  That is to say, this study is not necessarily generalizable to all population and does not cover particular types of falls that may result in injury.

In addition the control group did not receive motivational calls that might have led to lower levels of exercise than those in the intervention group, which may have led to a reduced number of falls in the control group.   The study also collected data regarding falls via a phone hot line; this may have resulted in the under reporting of falls for either group.  However, no exit interviews were conducted to measure the potential level of bias.  Finally, the authors report there were no analyses done of the individuals who refused to participate in the study.  Typically, there is a difference between individuals who choose to participate in studies versus those who do not- in this study that might have led to a differential in the number of falls between groups.

Relevancy to Physical Therapy Practice 

This article has significant implications for the practice of physical therapy. There was an existing fear that among those individuals with diabetes and peripheral neuropathy that increased weight bearing exercise might lead to increased risks of falls.  This study, although only one and still not having a robust sample size, suggests that efforts to increase the strength and balance of these individuals may be helpful.  Exercise, in turn, will ameliorate some of the side effects felt by individuals with this condition.  Of concern, however, the intense physical therapy sessions did not lead to improvement except in the area of one balance measure.  In order to implement these research findings into physical therapy practice, there would need to be further studies adjusting the level and intensity of the physical therapy intervention.

Conclusion

Overall, this study is important for it provides a high level of evidence regarding the potential relationship between physical therapy, weight bearing exercise, and falls for individuals with diabetes mellitus.  Although the study did not find a significant difference in falls, these findings need to be further studied and replicated before they can be implemented.

References:

American Geriatrics Society, British Geriatrics Society, and American Academy of Orthopedic Surgeons Panel on Falls Prevention. (2001). Guideline for the prevention of falls in older persons. Journal American Geriatric Society, 49:664–672.

Boule N.G., Haddad, E, Kenny G.P, et al. (2000). Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. JAMA, 286, 1218–1227.

Kruse, R.L., LeMaster, L.W., Madsen, R.  (2010). Fall and balance outcomes after an intervention to promote leg strength, balance, and walking, with diabetic peripheral neuropathy: “Feet First” randomized controlled trial.  Physical Therapy, 90(11), 1568-1578.