Footwear Style and Risk of Falls in Older Adults Below are the link for the abov
ID: 3301396 • Letter: F
Question
Footwear Style and Risk of Falls in Older Adults
Below are the link for the above article.
https://www.ncbi.nlm.nih.gov/pubmed/15341551
DOI:10.1111/j.1532-5415.2004.52412.x
OBJECTIVES: To determine how the risk of a fall in an older adult varies in relation to style of footwear worn.
DESIGN: Nested case-control study.
SETTING: Group Health Cooperative, a large health maintenance organization in Washington state.
PARTICIPANTS: A total of 1,371 adults aged 65 and older were monitored for falls over a 2-year period; 327 qualifying fall cases were compared with 327 controls matched on age and sex.
MEASUREMENTS: Standardized in-person examinations before fall occurrence, interviews about fall risk factors after the fall occurred, and direct examination of footwear were conducted. Questions for controls referred to the last time they engaged in an activity broadly similar to what the case was doing at the time of the fall.
RESULTS: Athletic and canvas shoes (sneakers) were the styles of footwear associated with lowest risk of a fall. Going barefoot or in stocking feet was associated with sharply increased risk, even after controlling for measures of health status (adjusted odds ratio 5 11.2, 95% confidence interval (CI) 5 2.4–51.8). Relative to athletic/canvas shoes, other footwear was associated with a 1.3-fold increase in the risk of a fall (95% CI 5 0.9–1.9), varying somewhat by style.
CONCLUSION: Contrary to findings from gait-laboratory studies, athletic shoes were associated with relatively low risk of a fall in older adults during everyday activities. Fall risk was markedly increased when participants were not wearing shoes. A variety of intrinsic and extrinsic risk factors for falls have been identified. Footwear is a potentially modifiable factor that has long been thought to play a contributing role in some falls, by any of several mechanisms. A shoe’s sole material and tread design can affect the coefficient of friction on the walking surface, which may in- fluence the risk of slipping. Heel height and width may affect a shoe’s tendency to tip sideways on an uneven surface, as well as gait and posture. Sole thickness and shoe collar height may affect proprioception. Two kinds of studies account for much of the research on footwear and falls in older adults. First, laboratorybased studies have examined gait, balance, and tactile perception in relation to features of shoe design, using volunteers under controlled conditions. For example, relatively thick, soft midsoles have been found to interfere with position sense and contribute to instability, as assessed by falls off a balance beam, raising questions about the safety of athletic shoes. One study found that volunteers performed better on a balance beam in tennis shoes than in cowboy boots. High-heeled shoes have been found to lead to lateral instability and to reduce stride length in a gait laboratory. Nonetheless, the extent to which the results of laboratory studies on healthy volunteers can be validly extrapolated to typical older adults and real-world walking environments remains unclear. Second, case reports and case series have described the footwear worn at the time of a fall and the faller’s belief about whether shoes played a causal role, but case series may simply reveal which shoe styles older adults most commonly wear. To determine whether the distribution of shoe styles worn by fallers is atypical, information is needed about shoe styles worn by controls who have not fallen. The present study was undertaken to investigate the degree to which the style and biomechanical characteristics of footwear were associated with risk of falling in a cohort of community-dwelling older adults. This article focuses on fall risk in relation to shoe style worn. METHODS The study was conducted in a predefined cohort of older adult enrollees in Group Health Cooperative, a large consumer-owned health maintenance organization in Washington state. A nested case-control study design was used. Study Subjects Study subjects were recruited from among 2,581 adults who were already enrolled in the Adult Changes in Thought (ACT) study, an ongoing prospective study of incidence and risk factors for Alzheimer’s disease.40 Participants in the ACT study had originally been identified by random sampling of Group Health enrollees aged 65 and older. To qualify for the ACT study cohort, subjects had to score 86 or greater on the Cognitive Abilities Screening Instrument (CASI) or be found not to meet clinical criteria for dementia after further diagnostic evaluation despite a low CASI score, as described in detail elsewhere.40 Some 1,371 ACT study participants who were ambulatory, noninstitutionalized, and willing to take part in the present study became the fall surveillance cohort. Fall occurrence among them was monitored prospectively from July 1998 through June 2000. All cases and controls were drawn from this cohort. A qualifying fall was defined as an unintentional fall to the ground, not preceded by loss of consciousness and not resulting from an external force (such as being pushed or hit), in a cohort member who was ambulatory at the time of the fall. Falls in persons who depended on a walker at all times were excluded, as were falls involving equipment that supports body weight (such as sports equipment or ladders). Falls qualified for study only if they occurred within King County, so that a visit to the fall site by the study team was practical. Fallers were asked to call a telephone hotline at the study center as soon as practicable after falling. All cohort members were also supplied with study calendars with attached postage-paid postcards and asked to return one postcard at the end of every month, indicating on it whether they had fallen during the month. Each month, those from whom no postcard was received were contacted by telephone to determine fall status. Cohort members who died, chose to stop participating, or were lost to follow-up were considered no longer under surveillance after their last postcard or follow-up telephone contact. Once a new fall case was identified, a matched control was selected at random from among all cohort members still under surveillance at the time of the fall who were similar to the faller with regard to age (3 years) and sex. To limit respondent burden, a person was exempt from further study after he or she had participated once as a case or a Data Collection After a fall case had been found eligible in a screening telephone interview, a two-person field team visited the faller by appointment at the location of the fall, if feasible, or at the faller’s home. Some 93.4% of eligible cases agreed to such a visit, which took place a median of 22 days after the fall. The case was asked to have available the shoes worn at the time of the fall. During the visit, one field worker interviewed the faller about circumstances of the fall and about other risk factors. Meanwhile, the second field worker took biomechanical measurements of the case’s right foot, the right shoe worn at the time of the fall, and the environment where the fall had occurred. The activity in which the case was engaged at the time of the fall was classified into one of five categories: walking indoors, walking outdoors, climbing stairs, moderate activity (such as vacuuming floors), or vigorous activity (such as playing tennis). When a potential control was then contacted by telephone, the control was asked to recall the most recent occasion on which he or she had been engaged in an activity in the category that included what the matched case was doing at the time of the fall. That occasion for the control then determined a reference date, time, and location. The footwear worn by the control on that occasion was then characterized in the same manner as for the case. Ultimately, 97.3% of selected controls agreed to be visited by the field team, and they were interviewed a median of 8 days after the reference date. Except for questions about the fall itself, the same standardized questions were asked of cases and controls. Topics included sociodemographics, self-reported health status, mobility, health behaviors, footwear wearing habits, and social/recreational activities in the previous month. Most questions in the interview were drawn from ACT study instruments or from previous surveys involving older adults. Participants had also previously undergone a standardized in-person interview and examination every 2 years as part of the ACT study. Data on grip strength, difficulty rising from bed or chair, and presence or absence of any gait abnormality were available from the most recent such visit for use in the present study. Data Analysis The incidence rate of falls was estimated as the number of fall cases divided by total person-time at risk in cohort members. For the main analyses, odds ratio (OR) estimates and 95% confidence intervals (CIs) were obtained from conditional logistic regression. This analytic method yielded an estimate of the adjusted relative odds of a fall among wearers of each shoe type, relative to wearers of a reference shoe type, after statistically adjusting for potential confounding factors such as measures of health status. For convenience of presentation, most tables show the prevalence of exposures in all cases and controls, ignoring matching. Because of the wide variety of footwear worn by cases and controls, three separate schemes were used to categorize footwear styles, involving 12, five, or three footwear categories. The 12-category scheme required fewest assumptions about which factors should be used to combine styles into broader categories, but because of the small number of cases and controls in any given category, CIs around estimates of effect were wide. For the five- and three-category schemes, styles were combined based on structural shoe design features, the daily activities for which shoes would typically be used, and observed frequency of use. All analyses were performed using Stata 7.0 (Stata Corporation, College Station, TX). Institutional review boards at the University of Washington and Group Health Cooperative reviewed and approved the study. RESULTS A total of 327 qualifying falls occurred in 1,900.6 personyears of follow-up, for an overall rate of 17.2 per 100 person-years at risk. Another 345 falls reported by participants were excluded: 88 occurred outside the study area, 59 occurred in subjects who were not ambulatory at the time, 52 were not falls to the ground, 46 occurred after a loss of consciousness, 41 were caused by an external force, 31 occurred on supportive equipment, 15 occurred in subjects who used mechanical aids, and 25 were excluded for other reasons. Most falls occurred in or around the home, 61% of them outdoors. About 62% took place while walking on a level surface, 13% on a slope, and 23% where there was a change in surface level, such as stairs, steps, or a curb. About 30% occurred between 6:00 p.m. and noon, 48% between noon and 6:00 p.m., 20% between 6:00 p.m. and midnight, and fewer than 3% between midnight and 6:00 a.m. The most common activity at the time of the fall was walking outdoors (43%), followed by walking indoors (31%), and climbing or descending stairs (15%). Only about 10% of falls occurred during moderate physical activity and fewer than 2% during vigorous exercise. About 36% of falls involved tripping on something, 23% involved an elevation change, 13% involved slipping, and the rest involved various other mechanisms. Most (65%) fall-related injuries were contusions or lacerations, but there were 15 fractures (4 hip, 5 hand/wrist, 6 other) and 15 head injuries. Overall, 25% of falls resulted in a medical care visit. As shown in Table 1, matching produced case and control groups of similar age and sex. Cases and controls also proved to be similar on self-reported health status and on physical activity during the 4 weeks before the reference date, but cases appeared to be less healthy than controls before the reference date on several interview and examination measures. Cases had weaker grip strength in the dominant hand, were more likely to report difficulty with rising from a bed or chair or with walking around the home, were more likely to have been observed to have some form of gait abnormality at their most recent ACT study visit, and were more likely to use a mechanical aid (e.g., cane or walker) at least part of the time. The most commonly worn style was athletic shoes, which was used as the reference exposure group under the 12-category scheme (Table 2). Relative to those wearing athletic shoes, risk of a fall was higher for wearers of nearly all other shoe styles under the 12-category scheme, although CIs around the ORs were quite wide because of the small number of subjects in each footwear style category. Nonetheless, fall risk was markedly elevated for persons who were barefoot or in stocking feet at the reference time. Like athletic shoes, canvas shoes (sneakers) are used for casual activities and have relatively wide rubber or crepe soles and fabric upper material and low heel height. On this basis, and because they appeared empirically to be similar to athletic shoes in their association with fall risk under the 12-category scheme, they were combined with athletic shoes to form the reference category in the five- and threecategory schemes. Bare feet and stocking feet were combined into a single shoeless category. For the five-category scheme, lace-up oxfords and loafers or flats were sufficiently common to warrant keeping them as separate categories, whereas both were combined with all other shoe styles in the three-category scheme. As shown in Table 2, athletic/ canvas shoes were the lowest-risk category under all three schemes, whereas shoeless was the highest-risk category. Health status was regarded as an important potential confounding factor. Table 1 shows associations between several measures of health status and increased risk of a fall. Also, the style of shoe most commonly worn could plausibly differ in relation to health status. Table 3 shows the observed associations between shoe style use and several measures of health status. Cases and controls are combined in Table 3, but results were similar when stratified by casecontrol status. Using the three-category scheme, the distribution of footwear styles differed most markedly in relation to measures of health directly related to walking: namely, presence of any gait abnormality and use of a mechanical aid when walking. Both factors were associated with higher likelihood of being shoeless and lower likelihood of wearing athletic/canvas shoes.
Please answer the questions below in realtion to the above article.
a ) What is the hypothesis being tested?
b) What is the disease or health outcome that is being studied?
C) How is the disease or health outcome assessed?
d) What is the exposure of primary interest in this study?
e) How is exposure measured in this study?
f) What do the authors say the study design is?
g) What do you say the study design is?
h) List the details of the methodology that support your answer to question 3.
i) Explain any reasons why this study design would or would not be appropriate to address this research question.
j) How does this study design compare to other designs that might be used to address this research question?
k) List all inclusion and exclusion criteria for study subjects. [HINT: include each in only 1 list - i.e. if you put "age 20-25" in the inclusion list don't repeat it as "not age 20-25" in the exclusion list. [HINT: if they used matching list the variables matched on]
L) What is the target population?
m) How were the study subjects identified or located?
n) How were subjects invited to participate in the study?
o) What measures of association and indicators of statistical significance do the authors present? Choose all that apply.
p) Select ONE potential source of bias in this study.
q) How does the bias you selected in question part O, operate in this study?
r) Identify cells in the 2x2 that will have more people in them than they should because of the bias you selected in question part O. [If the bias could cause different cells to increase under different scenarios, choose each of them.]
s) Identify cells in the 2x2 that will have fewer people in them than they should because of the bias you selected in question part O. Choose all that apply. [If the bias could cause different cells to decrease under different scenarios, choose each of them.]
t) How will the bias named in question part O, affect the measure of association (the relative risk or odds ratio)?
u)Do you have any comments on or explanations of your answers to questions part Q,R,S ?
v)What do the authors say about the bias you chose in question O?
w) What were the results of the study?
X) List the strengths of the study.
y) List the weaknesses of the study.
Z) Do you agree with the authors' conclusions?
Explanation / Answer
Dear student,
According to Chegg guidelines , only first four subparts are answered.
a, "The risk of a fall in an older adult doesnt not vary in relation to style of footwear worn." is the null hypothesis.
b, The risk of a fall is the health outcome under study.
c,A total of 1,371 adults aged 65 and older were monitored for falls over a 2-year period; 327 qualifying fall cases were compared with 327 controls matched on age and sex.
d,Standardized in-person examinations before fall occurrence, interviews about fall risk factors after the fall occurred, and direct examination of footwear were conducted.
Please give a THUMBS UP!!