CLINICAL ORTHOPAEDICS AND RELATED RESEARCH Number 431, pp.

150156 2005 Lippincott Williams & Wilkins

Osteoarthritis of the Knees Increases the Propensity to Trip on an Obstacle

Nirav K. Pandya, BA; Louis F. Draganich, PhD; Andreas Mauer, BA; Gary A. Piotrowski, BS; and Lawrence Pottenger, MD, PhD

Tripping on an object is the most frequent cause of falls. We examined the effects of painful osteoarthritis of the knee on obstacle avoidance success rates in older adults. Obstacle avoidance success rates, pain, body mass index, visual acuity, contrast sensitivity, depth perception, and single-leg stance duration were evaluated in 17 patients with painful osteoarthritis of the knees (age range, 59.6 8.1 years) and 14 age-matched healthy control subjects (age range, 61.1 10.0 years). The patients with osteoarthritis of the knees had a 37% lower obstacle avoidance success rate, a 54% lower single-leg stance duration, and a 24% greater body mass index than the control subjects. Age, visual acuity, contrast sensitivity, and depth perception were not different between the two groups. Obstacle avoidance success rates and singleleg stance durations decreased linearly as pain increased in the patients with osteoarthritis of the knees. Obstacle avoidance success rates also decreased linearly as single-leg stance duration decreased in the patients with osteoarthritis of the knees. Osteoarthritis of the knees reduced obstacle avoidance success rates, supporting epidemiologic studies that have found osteoarthritis to be a risk factor for falls. This study showed that pain associated with osteoarthritis of the knees increased the propensity to trip on an obstacle (the greater the pain the greater the propensity to trip and fall) and

underscores the importance of treating pain associated with osteoarthritis.

Received: February 5, 2004 Revised: July 9, 2004 Accepted: September 1, 2004 From the Motion Analysis Laboratory, Section of Orthopaedic Surgery and Rehabilitation Medicine, Department of Surgery, The University of Chicago, Chicago, IL. Each author certifies that he or she has no commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article. Each author certifies that his or her institution has approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research, and that informed consent was obtained. Correspondence to: Louis F. Draganich, PhD, The University of Chicago, Section of Orthopaedic Surgery and Rehabilitation Medicine, Department of Surgery, MC 3079, 5841 South Maryland Ave., Chicago, IL 60637. Phone: 773-702-6839; Fax: 773-702-0076; E-mail: ldragani@surgery.bsd.uchicago. edu. DOI: 10.1097/01.blo.0000150316.97009.f2

In the elderly population, falls are a public health problem, and is a leading cause of fatalities for people 65 years and older in the United States, accounting for 10,000 deaths in 1999.27,28 Falls are the most common reason for traumarelated hospital admissions and are responsible for 87% of all fractures in the elderly.20 Tripping on an object has been cited as the most frequent cause of falls.6,8,43,46,50,51 Numerous factors have been found to be associated with falls or to affect the ability of older people to avoid contacting an obstacle. These include delayed reaction time,10,37,44 divided attention,11 inadequate depth perception,36 low-contrast visual acuity,36,38 low-contrast sensitivity,36,38 reduced lower-limb strength,38 and impaired balance.38 Numerous investigators have examined the ability of subjects to cross obstacles that are recognized well in advance of reaching the obstacle.11,12,44 However, this affords the subject time to adjust his or her stride to safely clear the obstacle. Not knowing that an obstacle is present, because of factors such as hurrying, not looking where one is going, or not looking at the surface on which one is walking, has been reported to account for 66% of all falls among the elderly.5 Often, people do not see an obstacle until the last moment. In such cases, a subjects ability to avoid contacting an obstacle, and subsequent fall risk, may be more accurately tested using a suddenly appearing obstacle.10,11,45 The suddenly appearing obstacle may be a particularly important tool in evaluating the propensity to fall in patients with predisposing risk factors. One such subset of patients is patients with osteoarthritis (OA) of the knees, which has been cited in epidemiologic studies as a risk factor for falls.35,47 Osteoarthritis of the knees affects 14.4% of men and 28.4% of women in the United States older than 45 years,15 and is the most common cause of
150

Number 431 February 2005

Knee Osteoarthritis Causes Tripping

151

chronic disability among older people in the this country.39 The pain associated with OA of the knees is a major symptom of the condition,16 and is correlated with disability.15,39 Furthermore, pain from OA of the knees causes multiple gait abnormalities.1,30,31,40,42,49 These abnormalities include decreased loading of the medial compartment of the knee,30 increased stance time,42 reduced angular velocity of the knee,40 reduced hip and knee range of motion,1 and shorter stride length.1 These gait abnormalities may prevent persons with OA of the knees from successfully stepping over a suddenly appearing obstacle. For example, reduced ranges of hip and knee motion may reduce ones ability to avoid an obstacle. Yet, we were unable to find any studies regarding the effects of OA of the knees on obstacle avoidance. We hypothesized that patients with OA of the knees would have decreased ability to avoid contacting a suddenly appearing obstacle, as measured by an obstacle avoidance success score, because of increasing levels of knee pain as compared with healthy control subjects. MATERIALS AND METHODS
We tested 17 patients (two men, 15 women) with painful unilateral or bilateral OA of the knees, and 14 age-matched healthy control subjects (10 men, four women). Eight patients had unilateral OA (five in the right knee, three in the left knee); seven patients had bilateral OA, with pain greater in one limb (four in the right knee, three in the left knee), and two patients had bilateral OA with equal pain in both limbs. Exclusion criteria were: use of antidepressants or anxiolytics; history of heart disease or stroke; active cancer; diabetes; lower-extremity fractures during the last year; use of an assistive device, or prosthetic implants in the lower extremities. Also, patients with OA of the knees were excluded if symptomatic OA was present in joints other than the knee. Participants had to have 20/40 vision or better. The Institutional Review Board of our institution approved the protocol for this study. The purpose and the methods of the study were explained to the participants in the study and informed consent was obtained. Patients with OA of the knees indicated the pain in their afflicted knee (patients with bilateral OA indicated pain in the more painful knee) using a visual analog scale (VAS) of 100 mm. On the left end of the VAS was the phrase, no pain, and on the right end was the phrase, pain as bad as it can be. Patients with bilateral OA who stated they could not differentiate between the pain in their knees made one mark on the VAS for both knees. For one of the 17 patients with OA of the knee, the VAS scores were excluded from the analysis because, although the patient stated he had little pain, he indicated his pain level was 85 mm on the VAS, suggestive of near maximal pain. Therefore, there was a clear discrepancy between the level of pain stated verbally and that indicated numerically on the VAS. Body mass index (BMI) was computed for each participant as the ratio of body mass divided by the square of body height

(kg/m2). Visual acuity was tested using a standard Snellen eye chart. Contrast sensitivity was measured using the Melbourne edge test, which contained 20 circular patches with a series of edges of variable orientation and reducing contrast.52 Participants were asked to identify the highest patch number (lowest contrast patch) at which they still could see the edge. The Howard-Dolman depth perception apparatus (Lafayette Instrument, Lafayette, IN) was used to test the participants depth perception.24 The device consisted of an illuminated box with two vertical rods to which strings were attached. The participant moved both rods simultaneously in opposite directions by pulling on the two strings, one string with each hand. The participant sat 3 m away from the viewing aperture and was directed to align the rods. The rods were set 15 cm apart, and the order of the rods (front or back) was alternated from trial to trial, for a total of six trials. The resulting separation between the rods was recorded to the nearest millimeter for each trial, and then averaged across all six trials to obtain one mean score. The participants ability to stand on only one limb was used as a measure of balance.26,32 Healthy control subjects stood on their left limbs, and the patients with OA stood on their afflicted limbs (more painful limb in patients with bilateral OA). Patients with bilateral OA who stated they could not differentiate which of their knees was more painful stood on their left limbs. Participants were asked to stand barefoot on their limb for a maximum of 30 seconds or until the contralateral foot contacted the floor. The duration of single-limb stance was averaged for three trials. To test the participants ability to avoid stepping on a suddenly appearing light beam, or virtual obstacle, participants were asked to walk down a walkway 9.5 m long and 0.94 m wide, on which there were 600 conductive aluminum strips, each 8 mm wide, and with consecutive strips spaced 2 mm apart. An aluminum strip was attached to the sole of the heel of the participants shoes. The participants stood with their shoes on two aluminum plates at the beginning of the walkway. These were used to trigger sampling of the strips when the participants stepped off either plate. The complex of strips was sampled at a rate of 100 Hz using a data-acquisition board (Intelligent Instrumentation Inc., Tucson, AZ), which was controlled with a personal computer (Aberdeen Inc., Santa Fe Springs, CA). The system computed walking speed and step length, and predicted upcoming foot-fall locations. The personal computer also controlled a servomotor (QuickSilver Controls, Covina, CA), which was used to flash a virtual obstacle (band of light) onto the floor in a random order at predicted right and left foot falls. The light band was 10 cm wide and 85 cm long, and projected transversely across the walkway over a 2-m range near the far end of the walkway. The obstacle position was controlled within 0.5 cm and measured within 1 cm. Times were measured within 10 milliseconds. The conducting strips detected foot longitudinal placement within 1 cm. This apparatus was similar to that reported by Chen et al.10 Walking speed was controlled. This was done because we wanted to prevent variations in self-selected speed from being a confounding variable, and because a faster than self-selected speed in patients with OA of the knees was reported to increase

152

Pandya et al

Clinical Orthopaedics and Related Research

the reliability with which to evaluate gait and interventions that may affect gait.21 Therefore, because it has been reported that the speed of self-selected gait in healthy control subjects with ages similar to the ages of the control subjects in our study ranged from 1.001.10 m/second,2,18 and the self-selected gait of patients with OA of ages similar to ages of patients in our study ranged from 0.881.01 m/second,21 walking speed was required to be within the range of 1.00 1.07 m/second for all trials as a compromise. The participants wore their own comfortable low-heeled shoes. Lighting in the room was dimmed so that the participant could easily see the virtual obstacle. The light was flashed on the walkway, and the participant was asked if he or she could see the light clearly. Participants were asked to stand on the trigger plates, were informed that the light would not flash, and were asked to walk onto the aluminum strips at their self-selected walking speed. This was repeated four times. During these four trials, the participant was instructed to increase or decrease walking speed as necessary to remain within the specified range of walking speeds. The participant was then asked to walk on the walkway at the same speed as that just attained, was informed that the obstacle would appear on the walkway during the upcoming trials, and was asked to avoid stepping on the virtual obstacle. Trials outside the range of the walking speeds required were repeated. Participants were not instructed to direct their gaze to any particular point while walking. There were six trials for the right limb and six for the left limb, done in a random order. The virtual obstacle was flashed on the floor 350 milliseconds before the predicted foot fall. In each set of six trials, there were three possible foot falls greater than the 2-m range in which the light could be projected. The obstacle was presented twice for each of these foot falls in a random order. During the 12 trials, participants were reminded to maintain their speed. If the participant deviated from this range, the participant was asked to slow down or speed up as necessary, and the trial was repeated. For each of the 12 trials in which the light flashed, a score of 1.0 was given for successful avoidance, and a score of 0.0 was given for stepping on the obstacle. Successful avoidance was defined as not stepping on the virtual obstacle with any part of the shoe. Some participants were able to avoid the obstacle by rotating the forefoot laterally at the last instance (initially breaking the plane of the light), or dorsiflexing the ankle with only the toes elevated above the band of light. In these circumstances, 0.5 was given as the score. A similar method was used by Chen et al.10 The scores from the 12 trials were pooled and divided by 12 to obtain an obstacle avoidance success rate for each participant. To do the statistical analysis, SPSS (SPSS Science, Chicago, IL) was used. The one-tailed Students t test was used to determine differences in pain, BMI, single-leg stance, and obstacle avoidance success rates between the patients with OA of the knees and the control subjects, because patients with OA of the knees have greater pain, BMI,13 and decreased balance29 than healthy control subjects, and because OA of the knees is a risk factor for falling.35,47 The two-tailed t test was used to determine if there were differences in age, contrast sensitivity, depth perception, height, and weight between the patients with OA of the

knees and the control subjects, and to determine the differences in obstacle avoidance success rates between patients who had unilateral disease and patients who had bilateral disease. A paired t test was used to determine differences in obstacle avoidance success rates between limbs. Patients with bilateral OA who could not indicate which knee was more painful were excluded from this latter analysis. Linear regression analysis was done to determine if pain, BMI, and single-leg stance duration, had a significant predictive capability for the dependent variable (obstacle avoidance success rate) in the group of patients with OA of the knees, and for BMI and single-leg stance duration in the control subjects. An additional regression to determine if pain had a significant predictive capability for the dependent variable (single-leg stance) in the patients with OA of the knees also was done. A multiple linear regression analysis also was done with data entered backward to determine if pain and single-leg stance duration together had statistically significant predictive capability for the dependent variable (obstacle avoidance success rate) in the patients with OA of the knees. Variables for each simple linear regression analysis were entered one at a time, and were not entered together. Statistical significance was defined as p 0.05, and 95% confidence intervals for the difference in means were used.

RESULTS The patients with OA of the knees had a 37% lower (p < 0.0001) obstacle avoidance success rate than the control subjects (Table 1). There was no difference (p 0.64) between the mean obstacle avoidance success rates of patients with unilateral OA (0.49) of the knee and patients with bilateral OA (0.52) of the knees. In addition, there were no differences in obstacle avoidance success rates between limbs among all patients with OA of the knees, among patients with unilateral OA, or among patients with bilateral OA (Table 2). The mean pain score for the patients with OA of the knees (41.5 mm) was greater (p < 0.0001) than that of the control subjects (0.0 mm). The mean single-leg stance duration of the patients with OA of the knees was 54% lower (p < 0.0001) than that of the control subjects. Also, the BMI of the patients with OA of the knees was 24% greater (p 0.0005) than that of the control subjects. The increased BMI of the patients with OA of the knees was attributable to a combination of their decreased (p 0.01) heights and increased (p 0.02) weights as compared with the control subjects. Age, contrast sensitivity, and depth perception were not different (p 0.33) between the patients with OA of the knees and the control subjects. Obstacle avoidance success rates decreased (p 0.03; 0.31) linearly as pain increased in the group of par2 tients with OA of the knees (Fig 1). Obstacle avoidance success rates also decreased (p 0.02; r2 0.33) linearly as single-leg stance duration decreased in patients with OA of the knees (Fig 2). In addition, single-leg stance duration

Number 431 February 2005

Knee Osteoarthritis Causes Tripping

153

TABLE 1.

Mean (SD) Values

Patients with OA of Knees
0.51 (0.10) 41.5 (25.6) 11.7 (9.91) 1.63 (0.07) 96.5 (28.4) 34.3 (7.71) 59.6 (8.10) 17.4 (1.20) 1.00 (0.92)

Variable
Obstacle avoidance success rate Pain (mm) Single-leg stance (seconds) Height (m) Weight (kg) BMI Age (years) Contrast sensitivity (decibels) Depth perception (cm)

Control Subjects
0.81 (0.15) 0.0 (0.0) 25.6 (5.71) 1.72 (0.10) 76.5 (12.4) 26.1 (4.43) 61.1 (10.0) 17.5 (1.29) 1.46 (1.60)

Confidence Interval
0.390.21 27.955.2 19.78.00 0.020.15 36.73.31 3.4212.94 8.105.23 0.970.85 1.420.46

decreased (p 0.02; r2 0.32) linearly as pain increased in patients with OA of the knees (Fig 3). Multiple linear regression analysis showed that there was little improve0.34) in predicting obstacle avoidment (p 0.063, r2 ance success rates for patients with OA of the knees when pain and single-leg stance duration were included, as compared with pain or single-leg stance duration alone. Therefore, there was collinearity (tolerance value 0.677) between the two independent variables (pain and single-leg stance duration) in the multiple regression analysis. In addition, no linear relationships were seen between obstacle avoidance success rates and BMI for control subjects, BMI for patients with OA of the knees, and single-leg stance duration for control subjects (Table 3). DISCUSSION Tripping on an object is the most frequent cause of falls,6,8,43,46,50,51 and epidemiologic studies have associated OA of the knees with falls.35,47 However, we were unable to locate any studies about the effects of painful OA of the knees on obstacle avoidance. We hypothesized that patients with painful osteoarthritis of the knees would have decreased ability to avoid contacting a suddenly appearing obstacle as a result of increasing levels of knee pain when compared with the ability of a healthy control

subject. We found that patients with painful OA of the knees were less successful in avoiding the virtual obstacle than the age-matched healthy control subjects. Our results showed a 37% lower obstacle avoidance success rate in patients with OA. Furthermore, there was a linear relationship between pain and obstacle avoidance success rates for patients with OA of the knees, indicating that increased pain could account for 31% of the decrease in obstacle avoidance success rates. A linear relationship between single-leg stance duration and obstacle avoidance success rates in the patients with OA of the knees also was seen, indicating that decreased balance could account for 33% of the decreased obstacle avoidance success rates. In addition, a linear relationship between pain and single-leg stance duration was found in the patients with OA of the knees, indicating that increased pain could account for 32% of the decreased single-leg stance duration. Differences in obstacle avoidance success rates were not seen between patients with unilateral and bilateral OA of the knees or between limbs in patients with OA of the knees. Therefore, the obstacle avoidance success results were not affected by disease state (unilateral versus bilateral) or by limb condition (affected versus nonaffected limb in patients with unilateral OA, or more affected versus least affected limb in patients with bilateral OA). This was not unexpected because previous investigators have

TABLE 2.

Mean (SD) Obstacle Avoidance Success Rates

OAS Rate Affected or More Painful Limb
0.52 (0.18) 0.51 (0.16) 0.54 (0.22)

Group
Patients with OA of knees Patients with unilateral OA Patients with bilateral OA

OAS Rate Unaffected or Less Painful Limb

0.48 (0.18) 0.49 (0.19) 0.48 (0.19)

p Value*
0.61 0.83 0.65

Confidence Interval
0.120.20 0.220.27 0.240.36

OAS = obstacle avoidance success; *p values for OAS rate were obtained from paired t tests.

154

Pandya et al

Clinical Orthopaedics and Related Research

Fig 1. Increased pain leads to decreased obstacle avoidance success. Pain (VAS score) in patients with OA of the knees was linearly (p = 0.025; r2 = 0.31; adjusted r2 = 0.26; SE of estimate, 0.07) related to the obstacle avoidance success rates. OAS = obstacle avoidance success

Fig 3. Increased pain leads to decreased single-leg stance duration. Pain (VAS score) in patients with OA of the knees was linearly (p = 0.02; r2 = 0.32; adjusted r2 = 0.28; SE of estimate, 8.13) related to single-leg stance duration.

reported a bilateral adaptation in gait when only one lower limb has an abnormality.17,53 Age and visual outcomes also were not different between patients with OA of the knees and control subjects, excluding age-related performance or observation of the virtual obstacle from accounting for the differences in obstacle avoidance success rates between groups. There was a limitation in this study that may have effected interpretation of the results. The limitation involves the cognitive demand of maintaining the speed of 1.00 1.07 m/second rather than allowing participants to walk at self-selected speeds. Because the range of walking speeds we used was on the low end of walking speeds normally used by healthy control subjects,2,18 and on the high end of walking speeds normally used by patients with OA of the knees,21 a cognitive demand likely was placed on participants in both groups. This may have affected their obstacle

Fig 2. Decreased single-leg stance duration leads to decreased obstacle avoidance success. Single-leg stance duration was linearly (p = 0.02; r2 = 0.33; adjusted r2 = 0.28; SE of estimate, 0.09) related to the obstacle avoidance success rates. OAS = obstacle avoidance success

avoidance success scores. Furthermore, it is unknown whether the difference in cognitive demands (ie, maintaining a slower as compared with maintaining a faster walking speed) would have had similar effects on obstacle avoidance. Chen et al10 used a similar virtual obstacle apparatus and the same available response time as we used in our study (350 milliseconds). They reported that healthy, older subjects with a mean age of 73 years had obstacle avoidance success rates of 0.76 to 0.74. This was similar to the obstacle avoidance success rate of the healthy control subjects in the current study (0.81). The slightly higher obstacle avoidance success rates in our study may be attributable to the younger age of our control subjects (61 years), consistent with the findings of Chen et al10 that younger subjects had higher obstacle avoidance success rates. It also is possible that the walking speeds of the subjects in the current study were slower than those of the subjects in the study by Chen et al (1.40 m/second), and that the slower walking speeds allowed the subjects more time to avoid contacting the obstacle, resulting in a higher obstacle avoidance success rate. Nevertheless, our virtual obstacle method seems to be similar to that used by Chen et al.10 Pain associated with OA of the knees has been reported to affect gait,30,40 and to be associated with an increased risk for falling.34 To be successful in avoiding a suddenly appearing obstacle, subjects must be able to react quickly, using movements of gait that are impaired in patients with OA of the knees. These impairments of gait likely include reduced angular velocity of the knee,40 reduced hip and knee ROM,1 shorter stride length,1 and prolonged stance phase.1 Pain associated with OA of the knees also has been reported to result in quadriceps inhibition.4,7 Therefore, quadriceps reflex inhibition, as a result of knee pain, may be responsible for reducing the participants ability to

Number 431 February 2005

Knee Osteoarthritis Causes Tripping

155

TABLE 3.

Linear Regression Analysis Results

Dependent Variable
OAS rate OAS rate OAS rate

Independent Variable
Single-leg stance BMI BMI

Group
Control OA of knees Control

p Value*
0.90 0.64 0.78

r2 Value/ Adjusted r2 Value

0.002/0.08 0.01/0.05 0.01/0.08

SE of the Estimate
0.16 0.10 0.16

OAS = obstacle avoidance success rate; *p value for the regression analysis used to determine if the independent variables had a significant predictive capability for the dependent variables. Variables were entered in the regression analysis one at a time.

move quickly when the obstacle flashed in midstep. In addition, Chen et al11 found that dividing the attention of a healthy person affects obstacle avoidance success rate. Therefore, the pain may have acted to divide the attention of the patients with OA because their attention was focused not only on avoiding the obstacle but also in modifying their gait to limit pain.30 It was not surprising that the patients with OA of the knees had a reduced single-leg stance duration. This is because impaired balance is a characteristic of patients with OA29 and has been cited as a risk factor for falling.38,51 Therefore, the reduction in single-limb stance duration may have contributed to a decreased obstacle avoidance success rate in the patients with OA of the knees, shown by the linear regression associating decreased single-leg stance duration with decreased obstacle avoidance success rates. In addition, increased pain was found to be related to decreased single-leg stance in patients with OA of the knees. This also is not surprising because it has been reported that decreased pain leads to increased loading of the joint in patients with OA of the knees.18,30 The pain experienced by the patients with OA of the knees while loading the joint may have made them less willing to maintain prolonged single-leg stance, and is a possible mechanism through which their balance was decreased compared with the control subjects who were pain free. The patients with OA of the knees also had a greater BMI, a risk factor for OA13 and associated with functional limitations,19 than the control subjects. However, the regression analysis showed no relationship between BMI and obstacle avoidance success rates. Therefore, greater BMI apparently was a characteristic of the patients with OA of the knees rather than a mechanism that decreased their obstacle avoidance success rates. Despite the importance of the disease, conservative management of OA frequently is inadequate, and many patients have extended pain and disability, with a considerable socioeconomic impact.3,22 The results of our study suggest that patients with painful OA of the knees have an increased risk for tripping on an obstacle. Furthermore, the linear relationship between pain and obstacle avoidance

success rates for patients with OA of the knees suggests the importance of treating knee pain. Additional study is necessary to examine the effects of pain relief, weight loss, balance training, and radiographic disease severity33 on obstacle avoidance success rates in patients with OA of the knees. Additional factors that may be expected to affect the obstacle avoidance success in patients with OA of the knees include proprioception (joint position sense),23 abnormal knee laxity,48 lower-extremity weakness,35,38 and varus deformity.14 References
1. Al-Zahrani KS, Bakheit AM: A study of the gait characteristics of patients with chronic osteoarthritis of the knee. Disabil Rehabil 24:275280, 2002. 2. Andriacchi TP, Galante JO, Fermier RW: The influence of total knee-replacement design on walking and stair-climbing. J Bone Joint Surg 64A:13281335, 1982. 3. Badley EM, Rasooly I, Webster GK: Relative importance of musculoskeletal disorders as a cause of chronic health problems, disability, and health care utilization: Findings from the 1990 Ontario Health Survey. J Rheumatol 21:505514, 1994. 4. Barton R: Feinberg J: The Role of Occupational Therapy. In Brandt KD, Lohmander LS, Doherty M (eds). Osteoarthritis. New York: Oxford University Press 217239, 1998. 5. Berg WP, Alessio HM, Mills EM, Tong C: Circumstances of falls in independent community-dwelling older adults. Age Ageing 26:261268, 1997. 6. Blake AJ, Morgan K, Bendall MJ, et al: Falls by elderly people at home: Prevalence and associated factors. Age Ageing 17:365372, 1988. 7. Brucini M, Duranti R, Galletti R, Pantaleo T, Zucchi PL: Pain thresholds and electromyographic features of periarticular muscles in patients with osteoarthritis of the knee. Pain 10:5766, 1981. 8. Campbell AJ, Borrie MJ, Spears GF, et al: Circumstances and consequences of falls experienced by a community population 70 years and over during a prospective study. Age Ageing 19:136141, 1990. 9. Chen HC, Ashton-Miller JA, Alexander NB, Schultz AB: Stepping over obstacles: Gait patterns of healthy young and old adults. J Gerontol 46:M196M203, 1991. 10. Chen HC, Ashton-Miller JA, Alexander NB, Schultz AB: Effects of age and available response time on ability to step over an obstacle. J Gerontol 49:M227M233, 1994. 11. Chen HC, Schultz AB, Ashton-Miller JA, et al: Stepping over obstacles: Dividing attention impairs performance of old more than young adults. J Gerontol A Biol Sci Med Sci 51:M116M122, 1996. 12. Chou LS, Draganich LF: Placing the trailing foot closer to an obstacle reduces flexion of the hip, knee, and ankle to increase the risk of tripping. J Biomech 31:685691, 1998.

156

Pandya et al

Clinical Orthopaedics and Related Research

13. Coggon D, Reading I, Croft P, et al: Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord 6:622627, 2001. 14. Cooke TD, Pichora D, Siu D, Scudamore RA, Bryant JT: Surgical implications of varus deformity of the knee with obliquity of joint surfaces. J Bone Joint Surg 71B:560565, 1989. 15. Davis MA, Ettinger WH, Neuhaus JM, Mallon KP: Knee osteoarthritis and physical functioning: Evidence from the NHANES I Epidemiological Followup Study. J Rheumatol 18:591598, 1991. 16. Dekker J, Boot B, van der Woude LH, Bijlsma JW: Pain and disability in osteoarthritis: A review of biobehavioral mechanisms. J Behav Med 15:189214, 1992. 17. Draganich LF, Nicholas RW, Shuster JK, et al: The effects of resection of the proximal fibula on knee stability and gait. J Bone Joint Surg 73A:575583, 1991. 18. Draganich LF, Shrader MW, Pottenger LA, Piotrowski GA: The effects of pain-relieving knee injections on the gait of patients with osteoarthritis. Clin Orthop 421:188193, 2004. 19. Evers Larsson U, Mattsson E: Functional limitations linked to high body mass index, age, and current pain in obese women. Int J Obes Relat Metab Disord 25:893899, 2001. 20. Fife D, Barancik JI: Northeastern Ohio Trauma Study III: Incidence of fractures. Ann Emerg Med 14:244248, 1985. 21. Fransen M, Crosbie J, Edmonds J: Reliability of gait measurements in people with osteoarthritis of the knee. Phys Ther 77:944953, 1997. 22. Gabriel SE, Crowson CS, Campion ME, OFallon WM: Indirect and nonmedical costs among people with rheumatoid arthritis and osteoarthritis compared with nonarthritic controls. J Rheumatol 24:4348, 1997. 23. Gauchard GC, Gangloff P, Jeandel C, Perrin PP: Influence of regular proprioceptive and bioenergetic physical activities on balance control in elderly women. J Gerontol A Biol Sci Med Sci 58:M846 M850, 2003. 24. Graham CH: Vision and Visual Perception. Sydney, Australia Balance Systems 504547, 1965. 25. Guccione AA, Felson DT, Anderson JJ, et al: The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health 84:351358, 1994. 26. Hasan SS, Lichtenstein MJ, Shiavi RG: Effect of loss of balance on biomechanics platform measures of sway: Influence of stance and a method for adjustment. J Biomech 23:783789, 1990. 27. Hoyert DL, Arias E, Smith BL, Murphy SL, Kochanek KD: Deaths: Final data for 1999. Natl Vital Stat Rep 49:1113, 2001. 28. Hoyert DL, Kochanek KD, Murphy SL: Deaths: Final data for 1997. Natl Vital Stat Rep 47:1104, 1999. 29. Hurley MV, Scott DL, Rees J, Newham DJ: Sensorimotor changes and functional performance in patients with knee osteoarthritis. Ann Rheum Dis 56:641648, 1997. 30. Hurwitz DE, Ryals AR, Block JA, et al: Knee pain and joint loading in subjects with osteoarthritis of the knee. J Orthop Res 18:572579, 2000. 31. Hurwitz DE, Sharma L, Andriacchi TP: Effect of knee pain on joint loading in patients with osteoarthritis. Curr Opin Rheumatol 11:422426, 1999. 32. Judge JO, Lindsey C, Underwood M, Winsemius D: Balance improvements in older women: Effects of exercise training. Phys Ther 73:254262, 1993. 33. Kellgren JH, Lawrence JS: Atlas of Standard Radiographs: The

34. 35. 36. 37. 38. 39. 40.

41. 42.

43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54.

Epidemiology of Chronic Rheumatism. Oxford: Blackwell Scientific Publications, 1963. Leveille SG, Bean J, Bandeen-Roche K, et al: Musculoskeletal pain and risk for falls in older disabled women living in the community. J Am Geriatr Soc 50:671678, 2002. Lipsitz L, Jonsson P, Kelley M, Koestner J: Causes and correlates of recurrent falls in ambulatory frail elderly. J Gerontol 46:M114 M122, 1991. Lord SR, Dayhew J: Visual risk factors for falls in older people. J Am Geriatr Soc 49:508515, 2001. Lord SR, Fitzpatrick RC: Choice stepping reaction time: A composite measure of falls risk in older people. J Gerontol A Biol Sci Med Sci 56:M627M632, 2001. Lord SR, Ward JA, Williams P, Anstey K: Physiological factors associated with falls in older community-dwelling women. J Am Geriatr Soc 42:11101117, 1994. McAlindon TE, Cooper C, Kirwan JR, Dieppe PA: Determinants of disability in osteoarthritis of the knee. Ann Rheum Dis 52:258262, 1993. Messier SP, Loeser RF, Hoover JL, Semble EL, Wise CM: Osteoarthritis of the knee: Effects on gait, strength, and flexibility. Arch Phys Med Rehabil 73:2936, 1992. [published erratum appears in Arch Phys Med Rehabil 73:252, 1992] Myles PS, Troedel S, Boquest M, Reeves M: The pain visual analog scale: Is it linear or nonlinear? Anesth Analg 89:15171520, 1999. Otsuki T, Nawata K, Okuno M: Quantitative evaluation of gait pattern in patients with osteoarthritis of the knee before and after total knee arthroplasty: Gait analysis using a pressure measuring system. J Orthop Sci 4:99105, 1999. Overstall PW, Exton-Smith AN, Imms FJ, Johnson AL: Falls in the elderly related to postural imbalance. BMJ 1:261264, 1977. Patla AE, Prentice SD, Robinson C, Neufeld J: Visual control of locomotion: Strategies for changing direction and for going over obstacles. J Exp Psychol Hum Percept Perform 17:603634, 1991. Persad CC, Giordani B, Chen HC, et al: Neuropsychological predictors of complex obstacle avoidance in healthy older adults. J Gerontol B Psychol Sci Soc Sci 50:P272P277, 1995. Prudham D, Evans JG: Factors associated with falls in the elderly: A community study. Age Ageing 10:141146, 1981. Robbins A, Rubenstein L, Josephson K, et al: Predictors of falls among elderly people: Results of two population-based studies. Arch Intern Med 149:16281633, 1989. Sharma L, Lou C, Felson DT, et al: Laxity in healthy and osteoarthritic knees. Arthritis Rheum 42:861870, 1999. Stauffer RN, Chao EY, Gyory AN: Biomechanical gait analysis of the diseased knee joint. Clin Orthop 126:246255, 1977. Stevens M, Holman CD, Bennett N: Preventing falls in older people: Impact of an intervention to reduce environmental hazards in the home. J Am Geriatr Soc 49:14421447, 2001. Tinetti ME, Speechley M: Prevention of falls among the elderly. N Engl J Med 320:10551059, 1989. Verbaken JH, Johnston AW: Population norms for edge contrast sensitivity. Am J Optom Physiol Opt 63:724732, 1986. Youdas JW, Wood MB, Cahalan TD, Chao EY: A quantitative analysis of donor site morbiditiy after vascularized fibula transfer. J Orthop Res 6:621629, 1988. Young A, Stokes M, Iles JF: Effects of joint pathology on muscle. Clin Orthop 219:2127, 1987.