Because Dr. Hannan is Editor of Arthritis Care & Research, review of this article was handled by the Editor of Arthritis & Rheumatism.
Racial differences in foot disorders and foot type
Version of Record online: 27 OCT 2012
Copyright © 2012 by the American College of Rheumatology
Arthritis Care & Research
Volume 64, Issue 11, pages 1756–1759, November 2012
How to Cite
Golightly, Y. M., Hannan, M. T., Dufour, A. B. and Jordan, J. M. (2012), Racial differences in foot disorders and foot type. Arthritis Care Res, 64: 1756–1759. doi: 10.1002/acr.21752
- Issue online: 27 OCT 2012
- Version of Record online: 27 OCT 2012
- Accepted manuscript online: 5 JUN 2012 10:35AM EST
- Manuscript Accepted: 22 MAY 2012
- Manuscript Received: 12 DEC 2011
- Arthritis and Immunology NIH T-32 training grant. Grant Number: AR-07416
- Arthritis Foundation Postdoctoral Fellowship
- Framingham Foot Biomechanics Study (National Institute of Arthritis and Musculoskeletal and Skin Diseases). Grant Number: R01-AR047853
- Framingham Foot Biomechanics Study (National Institute of Arthritis and Musculoskeletal and Skin Diseases). Grant Number: R01-AR047853
- CDC/Association of Schools of Public Health. Grant Numbers: S043, S3486
To describe racial differences in the frequency of structural foot disorders and pes planus and pes cavus foot types in a large cohort of African American and white men and women ages ≥50 years.
Of 1,695 Johnston County Osteoarthritis Project participants evaluated for foot disorders/types in 2006–2010, 4 with lower extremity amputation were excluded, leaving 1,691 available for analyses (mean age 69 years, mean body mass index [BMI] 31.5 kg/m2, 68% women, 31% African American). The most common foot disorders/types were identified using a validated foot examination. Each foot disorder/type was compared by race using logistic regression, controlling for age, BMI, and sex. Effect modification between race (African American versus white) and age, BMI (categorized as ≥30 kg/m2 [obese] or <30 kg/m2 [nonobese]), sex, and education was examined.
Hallux valgus (64%), hammer toes (35%), overlapping toes (34%), and pes planus (23%) were common. Compared to whites, African Americans were almost 3 times more likely to have pes planus and were nearly 5 times less likely to have Tailor's bunions or pes cavus. Among the nonobese, African Americans were more likely than whites to have hallux valgus (adjusted odds ratio [ORadj] 2.01, 95% confidence interval [95% CI] 1.39–2.92), hammer toes (ORadj 2.64, 95% CI 1.88–3.70), and overlapping toes (ORadj 1.53, 95% CI 1.09–2.13).
Foot disorders are common among adults ages ≥50 years and differ by race. Future research is needed to determine the etiology of foot problems, especially those with racial differences, in order to inform prevention approaches.
Foot disorders are highly prevalent conditions, and evidence suggests that foot problems are significant independent predictors of disability and falls (1–3). The presence of foot disorders is more likely to occur with increasing age (4), and the prevalence of hallux valgus differs by sex (5–7), possibly due to factors such as age, weight, or shoe wear with nonsupportive components or high-stress biomechanics (7, 8). Other personal characteristics, such as race, may also be associated with foot disorders or certain foot types (pes planus or pes cavus). Clinical observations suggest that hallux valgus and pes planus are more prevalent in African Americans than in whites (5). However, few studies of foot disorders and conditions in the general population have been published to confirm these clinical findings (5, 7).
The purpose of the present study was to examine racial differences in the frequency of foot disorders and foot types in African American and white men and women in a large community-based sample. The frequency of foot disorders/foot types by race was further assessed to determine whether it varied by age, body mass index (BMI), sex, and education, as a proxy for socioeconomic status.
Significance & Innovations
Structural foot disorders are an important public health concern because they are common among adults ages ≥50 years and they may contribute to more serious adverse outcomes, such as falls and disability.
Racial differences in the frequency of foot disorders and certain foot types are apparent, particularly among the nonobese.
Patients and methods
This cross-sectional sample was composed of participants enrolled in the Johnston County Osteoarthritis Project, an ongoing, community-based study of the occurrence of osteoarthritis in African American and white residents in a rural county in North Carolina. During 1991–1997, this study enrolled civilian, noninstitutionalized adults ages ≥45 years who resided in 6 townships in Johnston County (9). This cohort was enriched with additional Johnston County residents ages ≥45 years during 2003–2004. During a followup visit (November 2006 to November 2010), 1,695 participants (ages ≥50 years by this visit) completed clinical examinations of the foot. Four participants were excluded due to lower extremity amputation; therefore, data for 1,691 participants were available for analyses.
Foot conditions and disorders.
Foot disorders/types were assessed using a validated foot examination with visual and palpation criteria to assess specific foot disorders (10, 11). The 2 major sections to the foot examination were: 1) questions about foot pain (pain, aching, or stiffness), a history of foot fracture, heel pain in the past 3 months, and foot pain while standing; and 2) a physical examination of each foot for the specific foot disorders of interest. The foot examination was performed by a trained clinical examiner and classified structural deformities, such as hallux valgus, hammer toes or claw toes, Morton's neuroma, Tailor's bunions, overlapping toes, and hallux rigidus, and conditions, such as pes planus and pes cavus, as present or absent.
Race (African American or white) was based on self-report by the study participant. The following participant characteristics were examined as potential covariates in our analyses because they may be associated with foot conditions or disorders: age (continuous variable in years), BMI (continuous variable calculated as weight in kilograms/height in meters squared), sex, and education (categorized as <12 years versus ≥12 years of school). Height without shoes was measured using a calibrated stadiometer in inches and converted to meters, and weight was measured in pounds using a balance beam scale and converted to kilograms.
The frequency of each foot disorder/ type was calculated. Chi-square statistics for categorical variables and t-test statistics for continuous variables were used to compare age, BMI, sex, and education with each foot disorder/type by race. Logistic regression was used to compare each of these foot disorders/types by race, controlling for age, BMI, sex, and education, and odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated. Effect measure modification was evaluated between race (African American versus white) and categories of participant characteristics, as follows: age dichotomized at the mean (>69 years versus ≤69 years), BMI in groups of obese (≥30 kg/m2) versus nonobese (<30 kg/m2), sex (women versus men), and education (<12 years versus ≥12 years of school). In models that indicated interactions at P values less than 0.10, the appropriate subgroups were examined, and ORs and 95% CIs were calculated separately for each subgroup. All statistical computations were performed using SAS software, version 9.2.
Of the 1,691 participants in this study, 31.2% were African American, more than two-thirds were women, and more than three-quarters had completed ≥12 years of education (Table 1). The mean ± SD age was 69 ± 9 years (range 50–95 years). One BMI value was considered implausible, and this participant was removed from any BMI analyses, leaving a mean ± SD BMI of 31.5 ± 7.2 kg/m2 for 1,690 participants (range 12.6–78.1 kg/m2). Compared to whites, African Americans were slightly younger, with a higher BMI on average, were more likely to be women, and were less likely to have completed ≥12 years of school (Table 1).
|Total sample (n = 1,691)||African Americans (n = 528 [31.2%])||Whites (n = 1,163 [68.8%])||P*|
|Age, mean ± SD years||68.6 ± 9.1||67.3 ± 9.2||69.2 ± 9.0||< 0.01|
|Body mass index, mean ± SD kg/m2||31.5 ± 7.2||33.3 ± 8.6||30.6 ± 6.3||< 0.01|
|Women, no. (%)||1,142 (67.5)||383 (72.5)||759 (65.3)||< 0.01|
|≥12 years of education, no. (%)||1,314 (77.7)||368 (69.7)||946 (81.3)||< 0.01|
As shown in Table 2, the most common structural foot disorder was hallux valgus, followed by hammer toes, overlapping toes, Morton's neuroma, Tailor's bunions, plantar fasciitis, and claw toes. The most common foot type was pes planus. Hallux valgus, hammer toes, and pes planus were more common in African Americans than in whites, and Tailor's bunions and pes cavus were more common in whites than in African Americans (Table 2). After the regression models accounted for the effects of age, sex, BMI, and education (Table 2), pes planus was nearly 3 times more common in African Americans than in whites, with an adjusted OR of 2.94 (95% CI 2.31–3.75). African Americans were approximately 5 times less likely to have Tailor's bunions (adjusted OR 0.18, 95% CI 0.09–0.37) and 4 times less likely to have pes cavus (adjusted OR 0.28, 95% CI 0.13–0.59) than whites in similarly adjusted models.
|Total sample (n = 1,691), no./total (%)||African Americans (n = 528 [31.2%]), no./total (%)||Whites (n = 1,163 [68.8%]), no./total (%)||Unadjusted OR (95% CI)||Adjusted OR (95% CI)*|
|Hallux valgus||1,088/1,691 (64.3)||371/528 (70.3)||717/1,163 (61.7)||1.47 (1.18–1.83)||–|
|Nonobese||535/818 (65.4)||149/198 (75.3)||386/620 (62.3)||–||2.01 (1.39–2.92)†|
|Obese||553/872 (63.4)||222/330 (67.3)||331/542 (61.1)||–||1.33 (0.98–1.80)†|
|Hammer toes||597/1,691 (35.3)||219/528 (41.5)||378/1,163 (32.5)||1.47 (1.19–1.82)||–|
|Nonobese||292/818 (35.7)||103/198 (52.0)||189/620 (30.5)||–||2.64 (1.88–3.70)†|
|Obese||305/872 (34.9)||116/330 (35.2)||189/542 (34.9)||–||1.12 (0.83–1.51)†|
|Overlapping toes||578/1,691 (34.2)||190/528 (36.0)||388/1,163 (33.4)||1.12 (0.91–1.39)||–|
|Nonobese||296/818 (36.2)||85/198 (42.9)||211/620 (34.0)||–||1.53 (1.09–2.13)†|
|Obese||282/872 (32.3)||105/330 (31.8)||177/542 (32.7)||–||0.94 (0.70–1.28)†|
|Pes planus||391/1,691 (23.1)||202/528 (38.3)||189/1,163 (16.3)||3.19 (2.53–4.04)||2.94 (2.31–3.75)|
|Morton's neuroma||160/1,691 (9.5)||55/528 (10.4)||105/1,163 (9.0)||1.17 (0.83–1.65)||1.03 (0.72–1.47)|
|Tailor's bunions||106/1,691 (6.3)||8/528 (1.5)||98/1,163 (8.4)||0.17 (0.08–0.35)||0.18 (0.09–0.37)|
|Plantar fasciitis||89/1,691 (5.3)||31/528 (5.9)||58/1,163 (5.0)||1.19 (0.76–1.89)||0.98 (0.62–1.57)|
|Pes cavus||79/1,691 (4.7)||8/528 (1.5)||71/1,163 (6.1)||0.24 (0.11–0.50)||0.28 (0.13–0.59)|
|Claw toes||55/1,691 (3.3)||21/528 (4.0)||34/1,163 (2.9)||1.38 (0.79–2.39)||1.57 (0.89–2.78)|
Statistically significant interactions were observed between race and BMI for hallux valgus, hammer toes, and overlapping toes. Among participants who were not obese (BMI <30 kg/m2), African Americans were approximately 2 times more likely than whites to have hallux valgus (adjusted OR 2.01, 95% CI 1.39–2.92), 2.6 times more likely to have hammer toes (adjusted OR 2.64, 95% CI 1.88–3.70), and 1.5 times more likely to have overlapping toes (adjusted OR 1.53, 95% CI 1.09–2.13). Among obese (BMI ≥30 kg/m2) participants, the frequency of hallux valgus (adjusted OR 1.33, 95% CI 0.98–1.80), hammer toes (adjusted OR 1.12, 95% CI 0.83–1.51), and overlapping toes (adjusted OR 0.94, 95% CI 0.70–1.28) did not differ significantly by race.
Structural foot disorders and pes planus were common in this community-based study of adults ages ≥50 years. Compared to the other 3 community- or population-based studies of structural foot disorders/types of which we are aware, the frequencies of hammer toes (5, 12), claw toes (5), pes planus (5), pes cavus (5), and plantar fasciitis (5) were similar to those observed in our study. In the present study, hallux valgus (64%) was more common compared to older samples from the Feet First Study (37%) (5) and the population-based MOBILIZE Boston Study of older adults (46%) (7). In addition, overlapping toes (34%) were twice as common compared to the Feet First Study (16%) (5). Tailor's bunions (6%) were less common in the present study compared to the Feet First Study (13%) (5). Variation in the frequency of these foot disorders across cohorts may be related to differences in population characteristics, such as age (e.g., Feet First Study  participants were ages ≥65 years and MOBILIZE Boston Study  participants were ages ≥70 years compared to our cohort of persons ages ≥50 years), BMI, race, or other extrinsic factors such as shoe wear or occupational history. Since the relationship of these variables to foot disorders is not known except in clinical samples, possible differences in cohorts should be considered and these factors further evaluated. The high frequency of foot disorders in these cohorts suggests that they are of important public health concern, especially since they may contribute to more serious adverse outcomes common for older adults (i.e., falls, decreased physical activity, functional limitations, and decreased quality of life) (1, 13, 14).
In this study, the racial differences for the frequency of structural foot disorders and certain foot types, specifically with hallux valgus, hammer toes, and pes planus, occurred more commonly among African Americans than whites, whereas Tailor's bunions and pes cavus presented more frequently among whites than African Americans. The Feet First Study also reported that hallux valgus and pes planus were more common in African Americans than in whites (5). We did not observe any significant interaction between race and sex in the present study; however, Nguyen et al (7) reported that nonwhite race was associated with hallux valgus among men only in the MOBILIZE Boston Study of older adults. In our study, the higher odds of hallux valgus, hammer toes, and overlapping toes among African Americans compared to whites was statistically significant among participants whose BMI was <30 kg/m2, but was less apparent among obese participants (BMI ≥30 kg/m2). The effects of obesity may mask a racial discrepancy in the frequency of hallux valgus, hammer toes, and overlapping toes; therefore, the results among the nonobese participants may provide a more accurate picture of the racial difference. The important interaction between race and obesity groups suggests that the effect of obesity may supersede the effects of unknown factors that may be responsible for the racial disparities in having these foot disorders, thereby equalizing the frequency of these disorders in obese African Americans and whites. Interestingly, nonobese African Americans had the highest frequency of hallux valgus, hammer toes, and overlapping toes compared to all other subgroups (obese African Americans and nonobese and obese whites), as shown in Table 2. The reason for the higher frequency of these foot disorders in nonobese African Americans is not known and warrants further exploration, particularly examining occupational factors and shoe wear history.
An important strength of this study is that the foot examinations were conducted using a validated foot evaluation tool for specific, common foot disorders. Additionally, the study is community based and includes large numbers of African American and white men and women, therefore allowing for racial comparisons in a sample that may be more representative of adults ages ≥50 years than those in clinic-based studies. To date and to our knowledge, this is the largest study that has assessed the relationship between common foot disorders or types and race. Our study is limited by several factors, however. Foot radiographs, which are useful for confirming the presence of structural disorders such as hallux valgus, were not obtained in this study. Foot biomechanics data, which may confirm pes planus/cavus foot types and pronatory/supinatory foot movement patterns, were collected but not fully analyzed at the time of this report and will be examined in future studies. Information on possible genetic influences, effects from various shoe wear worn at different ages across the lifespan, multijoint osteoarthritis, and occupational factors was not considered in these analyses. However, these aspects that might affect the relationship between foot disorders and race are planned to be the topics of future studies as these data become available, since they may help explain racial differences that we observed in this study.
In conclusion, the frequency of structural foot disorders and pes planus among adults ages ≥50 years is high, and in our community-based sample, certain foot disorders/types differed by race. Future research is needed to determine the etiology of these structural foot disorders/types in order to help explain the racial differences observed in this study. Understanding the origin of these foot disorders/types and their differences by race may inform prevention efforts as well as early intervention approaches, especially in important population subgroups.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Golightly had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Golightly, Hannan, Dufour, Jordan.
Acquisition of data. Jordan.
Analysis and interpretation of data. Golightly, Hannan, Dufour, Jordan.
- 10The epidemiology of foot disorders and foot pain in men and women: the Framingham Study [abstract]. Arthritis Rheum 2003; 48 Suppl: S672., , , , , .
- 11Physical limitations and foot disorders in elders [abstract]. J Am Geriatr Soc 2001; 49 Suppl: S22., , , .
- 12Third National Health and Nutrition Examination Survey, 1988-1994 (NHANES III examination data file). Hyattsville (MD): Centers for Disease Control and Prevention; 1996.
- 13Foot and ankle characteristics associated with impaired balance and functional ability in older people. J Gerontol Med Sci 2005; 60A: 1546–52., , .