Effect of A History of Anxiety and Depression on Patient-Reported Outcomes After Hip Arthroscopy at 2-Year Follow-up

Abstract

Background:

Patient resilience and mental health, including anxiety and depression, are increasingly recognized as critical factors influencing outcomes in orthopaedic surgery. However, the relative effect of these psychosocial factors on postoperative outcomes after hip arthroscopy remains unclear.

Purpose:

To quantify the relationship between anxiety, depression, the Brief Resilience Scale (BRS), and 4 validated patient-reported outcomes (PROMs) after hip arthroscopy.

Study Design:

Cohort study; Level of evidence, 3.

Methods:

A retrospective review of prospectively collected data for patients undergoing primary hip labrum repair from January 1, 2021, to January 1, 2022. Patients completed the BRS and 4 PROMs—modified Harris Hip Score (mHHS), visual analog scale for pain (VAS), Hip Outcomes Score for Activities of Daily Living (HOS-ADL), and Hip Outcomes Score for Sports-Related Activities (HOS-SP)—preoperatively, 6 months, 1 year, and 2 years postoperatively. Patients were stratified into tertiles based on baseline BRS scores: low resilience, normal resilience, and high resilience. Outcomes were analyzed using t tests, 1-way analysis of variance, principal component analysis, and Pearson correlation coefficients to assess relationships between resilience, anxiety, depression, and PROMs.

Results:

In total, 35 patients met the inclusion criteria with a mean follow-up of 741.4 days. Resilience showed a variable effect on PROMs, with no consistent differences observed across resilience tertiles at any follow-up interval. In contrast, patients with a history of anxiety and/or depression had significantly lower mHHS scores (68.7 vs 80.5, P = .041) and higher VAS scores (8.7 vs 3.9, P = .017) at 2 years, compared with those without such a history. Principal component analysis further highlighted the stronger association of anxiety and depression with outcomes compared with resilience at 6 months (P = .041), 1 year (P = .026), and the 2-year follow-up (P = .014). Overall, 85.7% of patients achieved the minimal clinically important difference for mHHS, 85.7% for VAS, 65.7% for HOS-ADL, and 67.5% for HOS-SP.

Conclusion:

The present study suggests that a history of anxiety and/or depression is consistently associated with poorer outcomes after arthroscopic hip surgery, whereas resilience demonstrates a less consistent and variable association.

Hip arthroscopy, an increasingly used technique for treating various hip pathologies such as femoroacetabular impingement, labral tears, and hip dysplasia, has been associated with improvements in range of motion, quality of life, and activity-related pain. Previous studies characterizing hip arthroscopy outcomes have used risk factors such as age, sex, body mass index, and smoking history to stratify risk and predict outcomes.^1,2 However, understanding the way in which psychosocial factors may contribute to post–hip arthroscopy outcomes remains limited. ¹¹ To address this gap, the Brief Resilience Scale (BRS) has been increasingly studied for its role in predicting recovery and outcomes in arthroscopy patients.^4,21,22,24 Most of these investigations have shown a positive correlation between BRS and patient-reported outcome measures (PROMs).^21,24

Given the rising national prevalence of depression and anxiety, it is estimated that up to 45% of the hip arthroscopy population has a mental health disorder and that 48% of patients take at least 1 psychotropic medication.^8,14,18,28 The effects of psychosocial factors unique to the individual patient and the influence these factors have on outcomes in both operative and nonoperative orthopaedic care have been better recognized and investigated in recent literature.^27,30 Patients with symptoms of anxiety and depression undergoing hip arthroscopy have reported lower function and satisfaction at both baseline and 2-year follow-up compared with those without symptoms.^10,14 However, the extent to which a patient's resilience may or may not influence the impact of anxiety and depression on outcomes after hip arthroscopy has yet to be explored. The existing literature suggests that surgeons should continue to explore methods of evaluating and identifying mental health symptoms preoperatively and during recovery to optimize outcomes after hip arthroscopy. ¹⁴

The purpose of our study was to quantify the relationship between anxiety, depression, BRS, and 4 validated PROMs: the modified Harris Hip Score (mHHS), the visual analog scale for pain (VAS), the Hip Outcomes Score for Activities of Daily Living (HOS-ADL), and the Hip Outcomes Score for Sports-Related Activities (HOS-SP). ²¹ These outcomes were recorded both preoperatively and at the 2-year follow-up. We hypothesized that patients with higher resilience scores would report better outcomes than those with lower resilience scores. Additionally, we hypothesized that a history of anxiety and/or depression would correlate with lower resilience scores.

Methods

Patient Selection and Demographics

Following institutional review board approval, this retrospective review prospectively collected data for all patients undergoing primary hip labrum repair from January 1, 2021, to January 1, 2022. Patients were individually evaluated by a single fellowship-trained sports medicine orthopaedic surgeon (R.T.) and recruited for the study upon successfully meeting the indications for arthroscopic hip surgery. Surgical indications for surgery consisted of hip labral tears that failed nonoperative management, consisting of activity modification, nonsteroidal anti-inflammatory drugs, and physical therapy. Inclusion criteria consisted of age >18 years; clinical, radiographic, and magnetic resonance imaging evidence of hip pathology; no history of hip surgery; English speaking; and completion of PROM surveys at all follow-up intervals. Study exclusions were as follows: <18 years of age, failure to complete outcome surveys, previous arthroscopic hip surgery, or severe degenerative osteoarthritis.

Recording of PROMs

Following enrollment and patient consent, patient demographic data, medical history of anxiety and/or depression (recorded as a binary value), relevant imaging, and physical examination findings were reviewed and recorded. All patients completed a baseline BRS before surgery. The BRS consists of 6 items, scored on a Likert scale, and has been tested for reliability and validity according to Cronbach's α and test-retest reliability with intraclass correlation for absolute agreement, respectively. The BRS alternates between positively and negatively worded questions to assess patient self-perception of resilience. For data analysis purposes, a value of 5 was assigned to each answer representing the most resilient response and a value of 1 to the least, with a maximum total score of 30 representing the greatest resilience (Table 1). Additionally, patients completed the 4 PROMs (mHHS, VAS, HOS-ADL, and HOS-SP) to obtain baseline measures of hip functionality and associated patient perceptions, as assessed by these validated metrics. The mHHS, which asks about pain, gait, and functional activities, was scored out of a maximum of 91 points. The VAS, used to measure pain, was rated 0 to 30, with 30 indicating the worst pain. HOS-ADL and HOS-SP, used to assess self-reported functional status, consist of 17 and 9 scored items, respectively. For each item, difficulty associated with doing certain tasks was assessed, with 4/4 indicating “none at all” and 0/4 indicating “unable to do.” These were scored out of a maximum of 68 and 36 points, with higher scores indicating higher functional status. All patients completed follow-up surveys, including the 4 PROMs at subsequent postoperative visits. Patients were only included in the study if they had preoperative, 6-month, 1-year, and 2-year outcome scores.

Table 1.

Patient-Reported Outcome Measure Scores Across Resilience Groups for Each Time Point ^a

	Baseline			6 Months			1 Year			2 Years
Resilience Group	LR	NR	HR	LR	NR	HR	LR	NR	HR	LR	NR	HR
mHHS	46.0 ± 13.40	51.0 ± 9.49	53.41 ± 14.26	77.22 ± 10.57	70.64 ± 14.46	65.88 ± 27.65	58.90 ± 32.29	41.0 ± 41.15	55.26 ± 37.83	76.0 ± 14.20	80.42 ± 15.07	73.70 ± 18.19
VAS	17.09 ± 5.81	15.07 ± 5.65	16.5 ± 5.69	5.88 ± 4.43	6.00 ± 5.37	4.32 ± 4.95	4.40 ± 5.08	1.75 ± 2.67	3.46 ± 4.88	6.40 ± 5.77	3.57 ± 5.50	6.41 ± 5.70
HOS-ADL	43.64 ± 12.96	46.86 ± 9.06	45.53 ± 12.16	50.90 ± 25.05	57.21 ± 12.83	55.05 ± 21.58	42.18 ± 27.69	28.0 ± 32.36	39.82 ± 20.66	60.18 ± 10.55	63.71 ± 7.42	55.70 ± 17.50
HOS-SP	12.0 ± 8.21	12.29 ± 7.80	14.0 ± 8.97	24.33 ± 8.27	11.57 ± 11.92	23.88 ± 12.31	17.20 ± 15.49	9.16 ± 12.96	17.86 ± 13.57	25.27 ± 10.88	21.71 ± 12.10	22.58 ± 11.14

^aValues are reported as mean ± SD. There were no significant differences between resilience groups for any patient-reported outcome measure at any time point. HOS-ADL, Hip Outcomes Score for Activities of Daily Living; HOS-SP, Hip Outcomes Score for Sports-Related Activities; HR, high resilience; LR, low resilience; mHHS, modified Harris Hip Score; NR, normal resilience; VAS, visual analog scale.

Surgery

All hip arthroscopy procedures were performed by a board-certified, fellowship-trained, certificate of added qualification–certified sports medicine orthopaedic surgeon (R.T.). Patients primarily underwent hip arthroscopy for labral repair with or without osteoplasty of the proximal femur or acetabulum. The decision for debridement versus repair of the labral tear was made at the discretion of the senior author at the time of surgery based on the tear pattern and the ability to adequately repair the labrum at the chondral labral junction. All patients were given appropriate postoperative protocols for the specific procedure performed, following a standard timeline of expected full recovery by 6 months postoperation.

Statistical Analysis

Statistical analysis was performed using Python (version 3.10). A power analysis was performed using Cohen's d to estimate the minimum required sample size, assuming α = 0.05, β = 0.20, and a moderate effect size (d = 0.5). Patients were stratified into the following tertiles by baseline BRS score: low resilience (LR), normal resilience (NR), or high resilience (HR). Postoperative PROMs were compared between the groups at 6 months, 1 year, and 2 years postoperation. Change in PROMs from baseline was also calculated for each time point and compared among resilience groups. In addition, results were stratified according to history of anxiety and/or depression. One-way analysis of variance was used to analyze global differences in outcomes, t tests were used to compare outcomes between the LR and HR groups, and Pearson correlation coefficients were calculated to assess the relationship between outcomes and resilience scores (α = 0.05).

Patient outcomes were also analyzed using principal component analysis (PCA), which is a dimensionality reduction statistical method that compresses several PROMs into a single value, the first principal component (PC1). ³¹ PCA analysis was conducted at each time point, analyzing patients with and without a history of anxiety and/or depression. The PC1 was then compared using an independent t test (α = 0.05). The minimal clinically important difference (MCID) was calculated as 0.5 multiplied by the standard deviation for each PROM for each time point to determine the percentage of patients experiencing meaningful improvement.

Results

Study Demographics

Assuming α = 0.05 and a moderate effect size (d = 0.5), a sample of 34 patients was required to achieve 80% power. Forty-seven patients were identified over the study period. Five patients were excluded due to incomplete procedure details, 4 patients were excluded due to incomplete baseline data, and 3 patients were lost to follow-up. The study population consisted of a total of 35 patients, including 26 patients who underwent labral repair alone and 9 who underwent labral repair with femoroplasty or acetabuloplasty (Figure 1). There were 12 male and 23 female patients, with an average age of 36.3 ± 10.6 years. The mean 2-year follow-up time was 741.4 days. Resilience tertiles included the LR group (n = 11), composed of all patients with resilience values less than 22; the NR group (n = 7), with resilience values between 22 and 24; and the HR group (n = 17), with resilience values of 24 and greater. The mean resilience score for patients with a history of anxiety and/or depression was 21.214, compared with 23.190 for those without such a history. This difference was not statistically significant (P = .372).

Figure 1.

Flowchart representing reasoning for excluded patients and procedural breakdown of included patient cohort.

PROMs by Resilience Groups

PROMs, including the mHHS, the VAS, the HOS-ASL, and the HOS-SP, were analyzed for the LR, NR, and HR groups at 6 months, 1 year, and 2 years. Additionally, the change in PROMs compared with baseline was analyzed based on resilience tertiles and time points. There were no significant differences in any PROM based on BRS tertile alone at any follow-up interval (Table 1).

Resilience and PROMs by History of Anxiety and/or Depression

There was no difference in the mean BRS for patients with or without a history of anxiety and/or depression (21.2 vs 23.2, P = .372) (Figure 2).

Figure 2.

Resilience scores for all patients with and without a history of anxiety and/or depression. Mean resilience score for patients with a history of anxiety and/or depression was 21.214, while mean resilience score for patients without a history of anxiety and/or depression was 23.190; independent t test P = .372.

PROMs were also compared based on a history of anxiety and/or depression at 6 months, 1 year, and 2 years. Significant differences were identified in several areas. At 2 years, patients with a history of anxiety and/or depression had significantly lower mHHS scores (P = .041). Additionally, patients without a history of anxiety and/or depression demonstrated superior VAS scores at 6 months (P = .005), 1 year (P = .033), and 2 years (P = .017) postoperatively. For HOS-ADL and HOS-SP, no significant differences were noted at any time points (Figure 3).

Figure 3.

Difference in patient-reported outcome measures (PROMs) based on a history of anxiety and/or depression at each time point (6 months, 1 year, and 2 years): upper row, modified Harris Hip Score (mHHS), and lower row, visual analog scale for pain (VAS). For each PROM, t tests were conducted to compare mean scores between patients with and without a history of anxiety and/or depression (mHHS: P_2yr = .041; VAS: P_6mo = .005, P_1yr = .033, P_2yr = .017). *P < .05.

When stratified by both resilience group and a history of anxiety and/or depression, the HR group demonstrated superior mHHS at the 2-year follow-up (P = .027) and VAS scores at 6 months (P = .044), 1 year (P = .048), and the 2-year follow-up (P = .03). There were no other differences between resilience groups regardless of a history of anxiety and/or depression (Figure 4).

Figure 4.

Difference in patient-reported outcome measures (PROMs) based on resilience score tertile at each time point (6 months, 1 year, and 2 years): upper row, modified Harris Hip Score (mHHS), and lower row, visual analog scale for pain (VAS). For each PROM, t tests were conducted to compare mean scores between patients with and without a history of anxiety and/or depression within the same resilience tertile (mHHS: P_2yr;high = .027; VAS: P_6mo;high = .044, P_1yr;high = .048; P_2yr;high = .03). Open circles indicate values beyond the whiskers (1.5 × interquartile range).

PCA

A PCA was conducted to evaluate the variance explained by PC1 at 2 years, comparing resilience tertiles and a history of anxiety and/or depression. PC1 explained 87% of the variance for resilience tertiles and 78% for a history of anxiety and/or depression. Those without a history of anxiety and/or depression demonstrated superior outcomes via PC1 at 6 months (P = .041), 1 year (P = .0261), and 2 years (P = .014). At the 1- and 2-year follow-up, the high resilience group demonstrated a trend toward superior outcomes but did not reach statistical significance (P = .05) (Figure 5).

Figure 5.

Plots comparing PC1 at the 2-year time point by resilience tertile (87% of variance explained by PC1) and a history of anxiety and/or depression (78% of variance explained by PC1). PCA was conducted for each time point, and 1-way analysis of variance was used to compare mean PC1 across tertile groups (P_6mo > .05; P_1yr = .05; P_2yr = .05) or across a history of anxiety and/or depression (P_6mo = .041; P_1yr = .0261; P_2yr = .014). PC1, first principal component; PCA, principal component analysis.

MCID

The MCIDs were 7.487 for the mHHS, 3.02 for the VAS, 7.606 for the HOS-ADL, and 5.082 for the HOS-SP. At 2 years, a substantial proportion of patients achieved the MCID for various PROMs. Overall, 85.7% of patients achieved the MCID for the mHHS, 85.7% for the VAS, 65.7% for the HOS-ADL, and 67.5% for the HOS-SP. Of patients with a history of anxiety and depression, 85.7% achieved the MCID for the mHHS, 85.7% for the VAS, 64.3% for the HOS-ADL, and 71.4% for the HOS-SP.

Discussion

The results of the current study demonstrated that while self-reported resilience demonstrated a variable effect on PROMs at various time points, a history of anxiety and/or depression consistently demonstrated a detrimental effect on PROMs. Specifically, patients with a history of anxiety and/or depression consistently reported worse VAS scores at all time points in addition to a worse mHHS at the 2-year follow-up. Similarly, when comprehensively evaluating the influence of anxiety and depression on PROMs using PC1, patients with a history of anxiety and/or depression demonstrated inferior outcomes at all follow-up time intervals. Finally, at the 2-year follow-up, the percentage of patients achieving the MCID for the mHHS, VAS, HOS-ADL, and HOS-SP was 85.7%, 85.7%, 71.4%, and 65.7%, respectively. Taken together, the results of the current study suggest that while resilience may play a role in recovery after hip arthroscopy, surgeons should remain vigilant in screening for anxiety or depression in this population as these conditions may have a more substantial influence on PROMs.

Resilience, an adaptive trait, is clearly beneficial for orthopaedic patients, but the relationship may be nuanced.^7,25,30 Unfortunately, researchers have had a difficult time defining the relationship between resilience and recovery from hip arthroscopy. Studies on this topic are limited and, like many others in orthopaedics, suffer from underpowered sample sizes, a lack of standardization of resilience scales and PROMs, a lack of standardization for patient stratification, and an inconsistent collection of preoperative (baseline) subjective hip scores.^4,12,26 While some studies have suggested a positive correlation between resilience and outcomes after arthroscopic surgery, others have found no conclusive relationship.^4,21,24 A case series of 89 patients by Szukics et al ²⁴ found that lower preoperative BRS was associated with inferior PROM at the 2-year follow-up after hip arthroscopy for femoroacetabular impingement and/or labral repair. Similar findings were reported by Silverman et al ²¹ upon short-term follow-up, with higher resilience associated with improved outcomes at 6 months after hip arthroscopy. However, these findings are not consistent across the arthroscopic surgery literature. For example, a systematic review by DeFoor et al ⁴ included 887 patients from 9 studies and demonstrated inconsistent relationships between resilience and PROMs. Similarly, the findings of our study failed to show a robust relationship between resilience and PROMs at the baseline, 6-month, 1-year, and 2-year time points, with the only statistically significant result being the LR group demonstrating a greater improvement in comparison to their baseline mHHS at 2 years when compared with their NR and HR counterparts. Notably, despite not reaching statistical significance, the HR group demonstrated lower mHHS and HOS scores, as well as higher VAS scores, compared with the LR and NR cohorts. Given that these individual outcome measures may fail to synthesize a patient's overall outcome and level of function, PC1 analysis, which has been implemented across various orthopaedic subspecialties in the literature, was used to provide a more comprehensive evaluation of the relationship between resilience and outcomes after hip arthroscopy.^{3,5,13,15-17,20,31} Results of the PC1 analysis were in agreement with most results of the individual outcomes and demonstrated no difference in outcomes across any time points between resilience groups. These findings highlight the fact that contrary to the commonly proposed hypothesis that high resilience yields better PROMs, the association between self-reported resilience and outcomes after hip arthroscopy may not be as robust as previously believed.

Beyond the direct impact of psychiatric comorbidities, it is important to recognize that anxiety and depression are characterized by cognitive distortions and negative bias in self-perception. These processes may influence how patients perceive their pain, function, and recovery trajectory, potentially amplifying negative symptoms even in the setting of objective improvement.^6,9,10,23 The relationship between pain, anxiety, and depression is complex, and untangling those relationships to infer causality or correlation has proved immensely difficult for researchers. In light of this, our findings that patients with anxiety and/or depression reported inferior PROMs despite many achieving MCID are consistent with the role of cognitive bias. The hip surgeon is often left wondering if, in the setting of pain and psychiatric illness comorbidity, their proposed surgical intervention may help alleviate or potentially exacerbate the patient's experience of pain and affect their overall function. Currently, there is literature to support both of these potential outcomes. In a large retrospective database study, almost half of hip arthroscopy patients discontinued their preoperative psychotropic medications after surgery. ²⁹ In a prospective propensity-matched cohort study of 153 patients, psychological distress was associated with lower baseline patient-reported outcomes but meaningful clinical improvement. ¹⁹ Conversely, literature suggests that a history of anxiety and/or depression is predictive of poor outcomes, increased health care costs, and greater opioid use compared with individuals without anxiety and/or depression.^6,9,10,23 The results of the current study are in agreement with much of the existing literature in that anxiety and depression were associated with lower overall mHHS scores and higher overall VAS. Furthermore, these findings were further supported by statistically significant superior PC1 scores across all time intervals for those without a history of anxiety and/or depression. Given the findings of the current study in the context of the existing literature, the authors believe it is imperative that surgeons view not only low resilience but also psychiatric comorbidities, such as anxiety and/or depression, as important prognostic indicators and not significant barriers that preclude surgery. Additionally, these findings underscore the critical importance of preoperative and postoperative clinical visits as these variables should be transparently discussed with appropriate counseling provided for patients.

Limitations

This study has several limitations. First, the relatively small sample size and all procedures being completed by a single surgeon affect statistical power and generalizability, respectively. Although an a priori power analysis demonstrated that our final cohort of 35 patients was sufficient to detect moderate effect sizes with 80% power (β = 0.20, α = 0.05), all studies remain subject to some risk of type II error. Follow-up studies of larger cohorts will be needed to better understand the relative influence of resilience, anxiety, and depression. Second, our cohort of pooled patients underwent labral repair with or without concomitant femoroplasty or acetabuloplasty. While this methodological approach has precedent in the hip arthroscopy literature, pooling of procedures introduces heterogeneity and therefore may affect the outcomes of the current study. ²⁴ Similarly, we did not adjust for other variables such as age, sex, dysplasia, deformity size, articular cartilage status, or the quality of bony resection. Although the primary focus of the current article was on the influence of resilience, anxiety, and depression on outcomes after hip arthroscopy, it is important to acknowledge that those additional variables may influence outcomes and therefore confound interpretation. Another limitation is the binary recording of anxiety and depression, which lacked details on diagnosis, severity, and treatment. Furthermore, the reliance on self-reported measures, such as the BRS and PROMs, introduces the possibility of response bias. Patients’ perceptions of their resilience and function could be influenced by various factors unrelated to their actual physical condition, such as mood or social support, potentially affecting the study's findings. Despite these limitations, this study showed that while anxiety and/or depression and low resilience can negatively affect PROMs, there is also significant upside for those patients when taking into account their baseline status. This highlights the need for larger, more robust studies to further explore these relationships.

Conclusion

The present study suggests that a history of anxiety and/or depression is consistently associated with poorer outcomes after arthroscopic hip surgery, whereas resilience demonstrates a less consistent and variable association.