SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

To investigate associations between self-reported knee confidence and pain, self-reported knee instability, muscle strength, and dynamic varus–valgus joint motion during walking.

Methods

We performed a cross-sectional analysis of baseline data from 100 participants with symptomatic and radiographic medial tibiofemoral compartment osteoarthritis (OA) and varus malalignment recruited for a randomized controlled trial. The extent of knee confidence, assessed using a 5-point Likert scale item from the Knee Injury and Osteoarthritis Outcome Score, was set as the dependent variable in univariable and multivariable ordinal regression, with pain during walking, self-reported knee instability, quadriceps strength, and dynamic varus–valgus joint motion during walking as independent variables.

Results

One percent of the participants were not troubled with lack of knee confidence, 17% were mildly troubled, 50% were moderately troubled, 26% were severely troubled, and 6% were extremely troubled. Significant associations were found between worse knee confidence and higher pain intensity, worse self-reported knee instability, lower quadriceps strength, and greater dynamic varus–valgus joint motion. The multivariable model consisting of the same variables significantly accounted for 24% of the variance in knee confidence (P < 0.001).

Conclusion

Worse knee confidence is associated with higher pain, worse self-reported knee instability, lower quadriceps muscle strength, and greater dynamic varus–valgus joint motion during walking. Since previous research has shown that worse knee confidence is predictive of functional decline in knee OA, addressing lack of knee confidence by treating these modifiable impairments could represent a new therapeutic target.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Knee osteoarthritis (OA), predominantly affecting the medial tibiofemoral joint compartment, is a major public health problem ([1]) and one of the leading causes of chronic functional disability in the elderly population ([2]). People with knee OA report symptoms of pain, stiffness, and instability ([3, 4]). Recently, lack of knee confidence has been highlighted in the literature as a problem, with more than half of those with or at high risk of knee OA reporting that they are troubled by this feeling ([5]).

Lack of knee confidence appears to be related to functional outcomes. In a recent longitudinal study, worse knee confidence at baseline was shown to predict functional decline in a large cohort of people with or at increased risk of having knee OA ([5]). The authors defined knee confidence using a question from the Knee Injury and Osteoarthritis Outcome Score (KOOS) ([6]) asking how much the individual is troubled by lack of confidence in his or her knee(s). The study reported that 54% of people were troubled by lack of knee confidence ([5]), indicating that it is an important issue in knee OA. Therefore, lack of knee confidence represents a potential therapeutic target. In order to design and evaluate the effects of interventions directed at improving knee confidence, a better understanding of contributors to knee confidence is needed.

A person's knee confidence may be associated with several activity-related factors. While it is likely that self-reported variables such as knee instability and pain during walking are associated with worse knee confidence, objective mechanical and neuromuscular factors may also be important. Muscle forces are integral for locomotion and general functional activities and in stabilizing the knee against external forces (e.g., those forcing the knee into a varus direction during walking [7]) during weight-bearing activities ([8-10]). Varus–valgus joint motion represents a direct measure of knee joint motion during weight-bearing activities. During the stance phase of normal walking, very little varus–valgus knee joint motion occurs in healthy individuals when tracked directly by bone pins ([11]), whereas mean ± SD varus–valgus joint motion of 3.24° ± 1.47° has been reported by marker-based methods in patients with knee OA ([12]), suggesting greater varus–valgus joint motion. A previous study showed that muscle weakness had a stronger impact on functional ability in knee OA patients with greater varus–valgus joint motion than in those with less motion, suggesting that both strength and varus–valgus joint motion are important for the functional ability of the patient ([12]). Since the knee plays a central role in all functional, weight-bearing activities, dynamic measures of varus–valgus joint motion during walking and muscle strength could possibly be associated with knee confidence.

To our knowledge, the association between knee confidence and pain, self-reported knee instability, dynamic varus–valgus joint motion, and strength in knee OA has not yet been established. We hypothesized that worse knee confidence would be associated with greater pain, the presence of self-reported knee instability, lower muscle strength, and greater dynamic varus–valgus motion during walking in people with medial knee OA, and that a combination of these measures will predict the degree of knee confidence.

Box 1. Significance & Innovations

  • Worse knee confidence is associated with higher pain intensity, worse self-reported knee instability, lower quadriceps strength, and greater dynamic varus–valgus joint motion during walking.
  • Since worse knee confidence previously has been shown to predict functional decline in knee osteoarthritis, treatment to address this symptom could target pain, knee instability, muscle strength, and dynamic varus–valgus joint motion.
  • Further studies investigating other determinants of knee confidence as well as the effectiveness of treatment to restore knee confidence are needed.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Participants

Baseline data from 100 participants with medial knee pain, radiographic medial tibiofemoral compartment OA, and varus malalignment recruited for a randomized controlled trial ([13]) were analyzed in the present study. Participants were recruited through advertisements in local clubs, university web sites, university staff newsletters, community centers, newspapers, Arthritis Australia, radio, and Facebook. Furthermore, brochures and study posters were placed in medical and physiotherapy clinics, presentations about knee OA were conducted in the local community, and we used our database of people who had previously participated in studies and gave consent for future contact.

The inclusion criteria for the randomized controlled trial were 1) average knee pain over the past week of ≥25 on a 100-mm visual analog scale (VAS) with the medial knee region as the site predominantly affected by pain/tenderness, and 2) radiographic OA in the medial tibiofemoral joint with varus knee alignment with the following specific criteria: Kellgren/Lawrence grade ≥2 ([14]), anatomic axis angle of <183° for men or <181° for women indicating varus alignment, medial tibiofemoral joint narrowing grade greater than lateral tibiofemoral joint narrowing grade ([15]), and medial compartment osteophyte grade greater than or equal to lateral compartment osteophyte grade ([15]).

The exclusion criteria for the randomized controlled trial were any of the following: 1) intraarticular corticosteroid injection or knee surgery within the past 6 months; 2) current or past (within 4 weeks) oral corticosteroid use; 3) systemic arthritic conditions; 4) history of tibial osteotomy surgery or hip or knee joint replacement; 5) other condition affecting lower extremity function; 6) participation in a strengthening or neuromuscular exercise program or other nonpharmacologic treatment for their knee pain in the past 6 months, including physiotherapy, acupuncture, and massage therapy; or 7) unable to ambulate without a gait aid. Participants who had been taking glucosamine, chondroitin, and/or nonsteroidal antiinflammatory drugs were not excluded.

Only the most painful eligible knee was included. If both knees were equally painful, the dominant knee was evaluated. All data reported in the present study were obtained at baseline prior to any intervention. The study was approved by the University of Melbourne Human Research Ethics Committee and participants provided written informed consent.

Radiographic protocol and knee alignment

Semiflexed, weight-bearing, short-extremity, posteroanterior radiographs were obtained with a caudal angle of 10° to achieve a superimposed tibial plateau. Focus-to-film distance was set to 100 cm, and at least 10 cm of the femur and 10 cm of the tibia were visible on the film in order to determine the anatomic axis. The anatomic axis was determined from the knee radiograph based upon the methods of Moreland et al ([16]). Radiographic OA severity was assessed using the Kellgren/Lawrence grading system ([14]), where higher grades indicate greater OA severity.

Self-reported knee confidence

Knee confidence was evaluated using a 5-point Likert scale (not at all, mildly, moderately, severely, and extremely) in response to the question: “How troubled are you with lack of confidence in your knee?” which is an item from the knee-related quality of life subscale in the KOOS. The KOOS is a valid, reliable, and responsive patient-reported instrument that has been used for short-term and long-term followup of knee OA ([6]).

Walking pain

Average walking pain intensity in the past week was assessed on a 100-mm VAS with terminal descriptors of “no pain” and “worst pain possible.”

Self-reported knee instability

Self-reported knee instability was assessed using the question: “Does your knee give way, buckle, or shift with…?” The participants responded using a 5-point Likert scale, with possible responses of 0 (My knee does not give way, buckle, or shift), 1 (Strenuous activities like heavy physical work, skiing, or tennis), 2 (Moderate activities like moderate physical work, running, or jogging), 3 (Activities of daily living/light activities like walking or housework), or 4 (Unable to perform any of the above activities due to giving way of the knee).

Muscle strength

Maximum, isometric strength (Nm/kg; normalized for body mass) was obtained for the quadriceps muscle at 60° of knee flexion in a sitting position using an isokinetic dynamometer (KinCom 125-AP, Chattecx). The participants performed several practice trials of increasing intensity until they understood the requirements of the test contraction. Participants performed 3 maximal contractions, each lasting for 5 seconds, separated by 40-second rest periods. Standardized, strong, verbal encouragement was given during the measurements. The best of the 3 contractions was applied in the analysis.

Dynamic varus–valgus joint motion

Participants underwent 3-dimensional gait analysis to assess dynamic varus–valgus joint motion (°) during walking. Using a 12-camera motion analysis system (Vicon MX) and force plates (AMTI), the movement of the participants was recorded at 120 Hz while they walked barefoot at a self-selected speed along a 10-meter level walkway, with speed monitored by 2 photoelectric beams 4 meters apart. Five successful trials were obtained from each participant (equal to complete foot strike on a force plate from the foot of the affected leg). Reflective markers were placed, according to the Plug-in-Gait lower limb model (Vicon), over the anterior and posterior superior iliac spines, lateral thigh, lateral femoral condyle, lateral shank, and lateral malleolus. Additional markers placed over the medial knees and ankles were included during the initial static standing trial to determine relative positioning of joint centers. The marker trajectories were filtered using the Woltring cross-validated quintic spline routine (mean square error 15). Vicon Plug-in-Gait software (version 2) was used to calculate segment anatomic coordinate systems according to Davis et al ([17]). A Cardan angle sequence with the knee varus–valgus angle calculated first was used to derive dynamic knee frontal plane joint motion. The difference between the maximum and minimum angles over 20–80% of stance phase was extracted to quantify the amount of dynamic varus–valgus motion during the main period when the knee is under load. The maximum and minimum angles were defined as the greatest (maximum; varus = positive) and smallest (minimum; valgus = negative) angles reached. The initial and final 20% of stance were avoided because these sometimes show artifacts associated with initial contact and the beginning of swing phase. Therefore, the reported difference reflects the amount of angular motion, in either direction, between 20% and 80% of stance.

Statistical analysis

Associations between pain, dynamic varus–valgus joint motion, and strength were assessed using Pearson's product moment correlations, while the associations between these parameters and self-reported knee instability were assessed using Spearman's rho. To correct for multiple correlations, P values were Bonferroni corrected. Ordinal logistic regression analysis was used to investigate associations between the dependent variable, self-reported knee confidence, and the independent variables (pain, self-reported knee instability, strength, and dynamic varus–valgus joint motion).

The construction of the ordinal logistic regression model followed the construction proposed by Bursac et al ([18]) and Hosmer et al ([19]): variables with a P value less than 0.10 in the univariable analysis were included in the multivariable regression analysis because more traditional levels such as 0.05 can fail to identify important variables ([20, 21]). Variables included in the initial model were removed if they were nonsignificant (P ≥ 0.10) and nonconfounders (a change in the estimate of the other variables of >20%). After this, any variable not selected for the initial model due to nonsignificance in the univariable analysis was reentered into the model one at a time to identify variables contributing to the model in the presence of the other variables. If significant (P < 0.10), the variable was kept in the model, and the process described above was repeated for the additionally added variables until the final model was obtained.

Age and sex were included as covariates in both the univariable and multivariable analyses. The significance level of the final model was set at P less than 0.05. All analyses were performed using IBM SPSS Statistics, version 20.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Demographics and clinical characteristics of the participants are shown in Table 1. Ninety-nine percent (n = 99) and 76% (n = 76) of the participants reported being troubled with lack of knee confidence and self-reported knee instability, respectively. The distribution of answers is shown in Tables 2 and 3. The associations between the independent variables of pain, self-reported knee instability, muscle strength, and dynamic varus–valgus joint motion were nonsignificant (Table 4).

Table 1. Demographics and clinical characteristics of the participants (n = 100)
 Value
Age, mean ± SD years62.4 ± 7.3
Female sex, no. (%)52 (52)
Duration of knee symptoms, mean ± SD years7.1 ± 6.1
Mass, mean ± SD kg82.7 ± 14.3
Body mass index, mean ± SD kg/m229.6 ± 4.1
Kellgren/Lawrence, no. (%) 
Grade 222 (22)
Grade 343 (43)
Grade 435 (35)
Anatomic axis, mean ± SD °176.8 ± 3.5
Walking pain, mm57 ± 19
Self-reported knee instability, no. (%) with instability76 (76)
Quadriceps strength, mean ± SD Nm/kg1.5 ± 0.4
Varus–valgus motion, mean ± SD °3.2 ± 1.8
Table 2. Self-reported knee confidence in knee osteoarthritis (n = 100)
 No. (%)
How troubled are you with lack of confidence in your knee? 
Not at all1 (1)
Mildly17 (17)
Moderately50 (50)
Severely26 (26)
Extremely6 (6)
Table 3. Self-reported knee instability in knee osteoarthritis (n = 100)
 No. (%)
Does your knee give way, buckle, or shift with…? 
My knee does not give way, buckle, or shift24 (24)
Strenuous activities like heavy physical work, skiing, or tennis9 (9)
Moderate activities like moderate physical work, running, or jogging18 (18)
Activities of daily living/light activities like walking or housework47 (47)
Unable to perform any of the above activities due to giving way of the knee2 (2)
Table 4. Associations between independent variables*
 Walking painKnee instabilityMuscle strength
  1. The P values shown are Bonferroni corrected. Walking pain = average walking pain intensity in the past week; knee instability = self-reported knee instability; muscle strength = maximum, isometric strength in the quadriceps muscle relative to body mass; VAS = visual analog scale; NS = nonsignificant; varus–valgus motion = maximum minus minimum varus–valgus angles during 20–80% stance.

Walking pain (0–100 VAS)   
Knee instability   
R0.14  
PNS  
Muscle strength, Nm/kg   
R0.003−0.21 
PNSNS 
Varus–valgus motion, °   
R−0.130.14−0.04
PNSNSNS

For the ordinal regression model, the assumption of proportional odds was met and tested in SPSS with the test of parallel lines, and the observed data were found to be consistent with the fitted model using goodness-of-fit statistics (P > 0.05).

Univariable regression analysis showed associations (P < 0.10) between worse knee confidence and higher pain intensity (odds ratio [OR] 1.02, 95% confidence interval [95% CI] 1.00–1.05), worse self-reported knee instability (OR 1.59, 95% CI 1.15–2.20), lower muscle strength (OR 0.37, 95% CI 0.14–0.98), and greater dynamic varus–valgus joint motion (OR 1.25, 95% CI 1.01–1.54) (Table 5). The multivariable model consisted of pain (OR 1.03, 95% CI 1.01–1.05), knee instability (OR 1.48, 95% CI 1.06–2.06), muscle strength (OR 0.305, 95% CI 0.11–0.84), and dynamic varus–valgus joint motion (OR 1.32, 95% CI 1.06–1.65), with each variable being a significant (P < 0.05) independent predictor of worse knee confidence (Table 5). The model significantly accounted for 24% of the variance in knee confidence (Nagelkerke's R2 = 0.243, P < 0.001). Neither age nor sex was a significant covariate in the univariable or the multivariable analysis (P > 0.20).

Table 5. Ordinal regression models of associations between self-reported knee confidence and the independent variables*
 Univariable analysisMultivariable analysis
OR95% CIPOR95% CIP
  1. Odds ratios (ORs) are only indicated when P < 0.10; see Table 4 for explanations. Age and sex were included as covariates in the univariable and multivariable analysis. 95% CI = 95% confidence interval.

Walking pain1.021.00–1.050.061.031.01–1.050.02
Knee instability1.591.15–2.200.011.481.06–2.060.02
Muscle strength0.370.14–0.980.060.300.11–0.840.02
Varus–valgus motion1.251.01–1.540.051.321.06–1.650.01

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

This study found that higher pain intensity during walking, worse self-reported knee instability, lower quadriceps muscle strength, and greater dynamic varus–valgus joint motion were significantly associated with worse knee confidence in participants with medial knee OA. Furthermore, a combination of the 4 measures accounted for 24% of the variance in knee confidence.

In our study, 99% of the participants were troubled with a lack of knee confidence, 67% were mildly or moderately troubled, and 32% were severely or extremely troubled. In comparison, the single previous study investigating knee confidence found that only 54% of the participants were troubled with a lack of confidence, with the majority being mildly troubled (31%) and only 7% being severely or extremely troubled ([5]). The difference in prevalence might relate to the characteristics of the samples. The previous study included both symptomatic, radiographic knee OA participants as well as asymptomatic, nonradiographic participants at high risk of knee OA. Approximately 45% did not have radiographic joint space narrowing and the sample had mean ± SD knee pain of only 3.3 ± 3.6 on a 0–20 scale. In our study, the majority of participants had moderate to severe radiographic OA disease (78% with a Kellgren/Lawrence grade of 3 or 4) and mean ± SD knee pain during walking of 57.4 ± 19.1 on a 0–100 scale. Given that disease severity (pain) and knee confidence seem to be related, as demonstrated in our study, the inclusion of asymptomatic, nonradiographic participants would likely lower the prevalence of participants troubled with a lack of knee confidence. This is supported by a trend toward increased pain and joint space narrowing with worse knee confidence in the previous study on knee confidence ([5]). Our results suggest that the majority of participants with symptomatic knee OA report a lack of knee confidence, further substantiating it as an important issue in knee OA.

Our study is the first to show associations between knee confidence and both self-reported and objective parameters. Given the central role of the knee in all weight-bearing activities, knee confidence may be affected by (and/or affect) several factors related to activity. Therefore, the associations are not surprising, since walking pain, self-reported knee instability, muscle strength, and dynamic varus–valgus joint motion are all related to functional activities. Whether this relationship is affected by other parameters, e.g., knee joint loading, should be explored in future studies. Nevertheless, the findings give interesting implications for the treatment of knee OA, and lack of knee confidence could represent a potential therapeutic target. Our study warrants further research on lack of knee confidence and other items from patient-reported outcomes such as the KOOS.

It has been suggested that confidence is related to self-efficacy ([5]), defined as an individual's belief that he or she is capable of successfully performing a behavior or exercising control over one's health habits ([22, 23]). A previous prospective study showed that the combination of low self-efficacy and low strength at baseline resulted in a larger 30-month decline in both self-reported and objective outcomes in men and women ages ≥65 years with knee pain ([24]). Similarly, greater strength and self-efficacy have been suggested to be protective against poor functional outcome over 3 years ([25]). Our study supports a relationship between pain, self-reported knee instability, muscle strength, varus–valgus joint motion, and knee confidence in people with medial knee OA. Pain relief in knee OA has been found to increase the relative activation of the quadriceps muscle ([26]). This suggests that pain perhaps inhibits voluntary muscle activation ([27]) about the knee, leading to a lack of knee confidence during activities where the muscles are required to provide joint stability. Greater dynamic varus–valgus joint motion may be associated with inefficient use of muscle strength/control during gait, and the combination of greater varus–valgus joint motion and lower muscle strength is related to functional ability in knee OA ([12]). This suggests that worse knee confidence, a predictor of functional decline in knee OA ([5]), is also related to both greater varus–valgus joint motion and lower muscle strength, as found in the present study. A downward spiral of pain, worse self-reported knee instability, muscle weakness, greater varus–valgus joint motion, and worse knee confidence may further exacerbate decreased activity, additionally increasing pain and reducing muscle strength. In support of this notion, patients with OA-related knee pain demonstrate reduced muscle strength compared to controls without knee pain ([28]), and both muscle strength and pain ([29]), strength and dynamic varus–valgus joint motion ([12]), as well as strength and self-efficacy ([24]), have been shown to interact when evaluating the effects on physical function.

The prevalence of self-reported knee instability in this study (76%) is consistent with previous studies showing that between 60% and 81% of patients with knee OA report sensations of knee instability ([4, 30-33]). It has been speculated that knee confidence could be related to self-reported knee instability ([5]), but our study is the first to demonstrate this relationship. Since both parameters are subjective measures reported by the patient, it is likely that they are closely related and that the knee confidence of a given patient is affected by feelings of knee instability/giving way in activities of daily living.

Although the mechanisms of knee instability/giving way in knee OA remain unclear, muscles acting on the knee joint (i.e., the quadriceps muscle) are important to sustain knee stability in participants with knee OA ([27, 30]). Quadriceps weakness is well documented in patients with knee OA ([27]), with evidence relating quadriceps weakness to disease onset ([34-37]), while the relationship with progression remains inconclusive ([38, 39]). However, our study further underlines the importance of involving muscle strengthening in the treatment of knee OA, since strength deficits ([25]) and worse knee confidence ([5]) have been shown to be related to poor functional outcome over a 3-year period.

Greater dynamic varus–valgus joint motion is suggested to be associated not only with soft tissue laxity and joint erosion, but also with inefficient muscle function in the loading phase ([12]). Our findings support this, by associating both muscle strength and dynamic varus–valgus joint motion to knee confidence. Since the presence of both greater dynamic varus–valgus joint motion and muscle weakness decreases functional ability ([12]), perhaps interventions targeting both are needed to effectively treat knee OA in participants with varus malalignment. While quadriceps strength training has been shown to be effective at improving pain and physical function in knee OA ([40, 41]), it has not been shown to be effective at reducing pain in the subgroup with varus malalignment ([42]), indicating that it may not affect the biomechanical contributors to medial compartment knee load in a beneficial way. One way of targeting the treatment toward these aspects could be neuromuscular exercises. Neuromuscular exercise, performed in weight-bearing positions, focuses on the quality of the movement and correct knee alignment relative to the hip and foot to control lateral knee movement ([13]), thereby possibly reducing the varus–valgus motion during functional activities.

Based on the findings of our study, neuromuscular exercise could potentially improve knee confidence and thereby prevent decline in function in knee OA. Recent international evidence-based guidelines in knee OA recommend a multimodal treatment approach with patient education as one of the cornerstones ([43, 44]). Therefore, including patient education aimed at improving self-management could also improve outcomes, due to an improved confidence in knee function.

The model consisting of pain during walking, self-reported knee instability, muscle strength, and dynamic varus–valgus joint motion during walking accounted for 24% of the variance in knee confidence. The variables applied in this study were a combination of objective (muscle strength and dynamic varus–valgus joint motion) and self-reported (pain and knee instability) measures. Since none of the independent variables was associated with one another, they could serve as independent predictors of self-reported knee confidence and be combined without being significant confounders to the regression model. However, since a substantial proportion of the variance in knee confidence remains unaccounted for, the incorporation of other independent variables could improve the understanding of knee confidence. Future studies should address these aspects by including other objective and self-reported variables.

There are several limitations to our study. The cross-sectional design means that it is not possible to determine the cause and effect relationship between self-reported knee confidence and the independent variables. Furthermore, although self-reported knee confidence, pain, and dynamic varus–valgus joint motion were assessed in relation to weight-bearing functional activities, muscle strength was assessed maximally and in sitting. Maximal strength is rarely called upon during functional activities, although it may indicate the extent to which a patient has “reserves” of strength, allowing them to perform such tasks without great difficulty.

Varus–valgus joint motion was derived from external markers placed on the skin and therefore may have been affected by skin motion artifact. It is possible that markers attached to bone pins would have elicited different varus–valgus motion, particularly if the early stance phase of loaded knee motion was accurately captured.

In conclusion, higher pain intensity during walking, worse self-reported knee instability, lower muscle strength, and greater dynamic varus–valgus joint motion are associated with worse self-reported knee confidence in medial knee OA. This suggests that treatment should target pain, knee instability, muscle strength, and dynamic varus–valgus joint motion in medial knee OA patients with varus malalignment, and include patient education to address the psychological component of knee confidence. Additionally, further studies investigating self-reported knee confidence are needed, including both objective and self-reported variables.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

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. Mr. Skou 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. Skou, Wrigley, Hinman, Bennell.

Acquisition of data. Wrigley, Metcalf.

Analysis and interpretation of data. Skou, Wrigley.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES