Trajectories of anxiety and depression levels in older adult patients undergoing total hip arthroplasty

Abstract

BACKGROUND

Total hip arthroplasty (THA) has become a core means of joint function enhancement among older adult patients due to global population aging. However, postoperative anxiety and depression compromise patients’ rehabilitation and quality of life.

AIM

To explore anxiety/depression trajectories among older adults who have undergone THA, identify high-risk factors, and propose stratified precision nursing strategies.

METHODS

This observational study enrolled 208 older adult patients who had undergone their first unilateral THA between June 2023 and June 2025. The Hamilton Anxiety Rating Scale and Hamilton Rating Scale for Depression-17 items assessed anxiety and depression, respectively, combined with the Visual Analog Scale (VAS), General Self-Efficacy Scale (GSES), and Social Support Rating Scale (SSRS). A latent category growth model was used to identify anxiety/depression subgroups, and multiple logistic regression was used to analyze the independent influencing factors.

RESULTS

Based on the dynamic trajectory of the patients’ anxiety and depression levels, which exhibited a “peaking at 1-week post-operation—gradual relief over the next 1 month”, patients were divided into two subgroups: The stable (42.79%; mild preoperative abnormalities and postoperative recovery after short-term fluctuations) and unstable group (57.21%; patient had moderate-to-severe negative emotions before surgery, which were relieved within three months). Old age (OR = 1.493), low education level (OR = 3.440), 1-week postoperative VAS (OR = 1.916), SSRS (OR = 0.894), and GSES (OR = 0.912) were independent factors for the high-risk group.

CONCLUSION

This study reveals the anxiety/depression heterogeneity among older adults who have undergone THA and suggests that the 1-week to 1-month post-operation is the key psychological intervention window.

Key Words: Hip replacement; Anxiety and depression; Trajectory analysis; Social support; Self-efficacy; Postoperative pain; Older adult patients

Core Tip: This study explored the anxiety/depression trajectories in older adult patients who have undergone total hip arthroplasty via a latent category growth model and identified two subgroups. Postoperative 1-week to 1-month is the key intervention window. Old age, low education level, and postoperative pain are risk factors, while social support and self-efficacy are protective factors. These results provide a basis for precise stratified nursing.

INTRODUCTION

The rapid aging of the global population has markedly increased the incidence of hip fractures and end-stage osteoarthritis worldwide. Total hip arthroplasty (THA), also known as total hip replacement, is a definitive surgical procedure that replaces the damaged acetabulum and femoral head with artificial prostheses to relieve intractable pain and restore joint function[1]. Older adults often experience significant anxiety and depression due to the decline in physiological functions, chronic pain, a long postoperative recovery period, and uncertainty regarding the surgical outcome. These factors can reduce the effectiveness of postoperative analgesia and patient compliance with rehabilitation and may delay functional recovery and increase the risk of deep vein thrombosis and surgical site infections[2,3]. Research has revealed a 30%-60% incidence of anxiety and depression among older adult patients undergoing hip replacement[4]. However, the dynamic trajectories of anxiety and depression and associated influencing factors have not been systematically examined.

Most existing studies have only examined patients’ psychological state at a single time point, before or after the operation[5], and lacked dynamic monitoring of anxiety and depression levels over time. Meanwhile, studies on influencing factors are mostly limited to a single variable[6] and fail to integrate multidimensional factors, such as social support and self-efficacy. Furthermore, existing intervention strategies are mostly based on empirical nursing[7,8], and lack precise plans tailored to the changing trajectories of anxiety and depression, which leads to the persistence or even aggravation of psychological issues in some patients. Therefore, revealing the evolution patterns of anxiety and depression in older adult patients through longitudinal research is urgent for enabling stratified intervention.

To address these gaps, this study aimed to explore the dynamic trajectories of anxiety and depression in older adults undergoing primary unilateral THA, identify independent risk factors for persistent negative emotions, and propose targeted, stratified intervention strategies. The key innovations are as follows: (1) Establishing the first dynamic trajectory model of anxiety and depression in older THA patients; and (2) Integrating psychological and social factors into the analysis, moving beyond the traditional biomedical model.

By defining the high-risk period and key influencing factors of anxiety and depression, early warning indicators can be provided to clinical nursing staff. Thus, targeted strategies, such as cognitive behavioral intervention and family participatory support, can be implemented. This will help reduce the incidence of negative emotions among older adult patients and facilitate their postoperative functional recovery, ultimately enhancing their overall prognosis and quality of life.

MATERIALS AND METHODS

Research participants

This observational study recruited 208 older adult patients who underwent their initial unilateral THA in Henan Provincial People’s Hospital between June 2023 and June 2025. All the enrolled cases met the inclusion criteria, and those that did not meet the requirements were excluded.

Inclusion criteria: (1) Being aged ≥ 65 years; (2) Having a diagnosis of femoral neck fracture/femoral head necrosis/severe osteoarthritis; (3) Having undergone unilateral THA; (4) Having completed the Hamilton Anxiety Rating Scale (HAMA)/Hamilton Rating Scale for Depression (HAMD) evaluations before operation as well as 1 week, 1 month, and 3 months after operation; and (5) Having complete clinical data.

Exclusion criteria: (1) Having preoperative diagnosis of depression, anxiety, or other mental illness; (2) Having severe cognitive impairment; (3) Having pathological obesity or a malignant tumor; or (4) Being lost to follow-up or having > 20% missing data.

This study was approved by the Ethics Committee of Henan Provincial People’s Hospital. All participants or their legal representatives provided written informed consent prior to enrollment.

Sample size calculation

Sample size was calculated using G-Power 3.1.9.7 software based on multivariate logistic regression analysis. Key parameters were set as follows: α = 0.05 (two-tailed), 1-β = 0.80, f² = 0.15 (medium effect size, referenced from similar orthopedic psychological trajectory studies), number of independent variables = 5 [age, education level, postoperative Visual Analog Scale (VAS), General Self-Efficacy Scale (GSES), and Social Support Rating Scale (SSRS) scores], and the model inclusion probability π = 0.5. The preliminary calculated sample size was 189 cases. Considering a potential follow-up loss rate of 10%, the final sample included 208 cases.

Research methods

We reviewed the electronic medical record system and nursing records to obtain patients’ baseline data and surgical parameters for analysis.

Questionnaire surveys

All patients completed the HAMA and HAMD-17[9,10], VAS[11], and GSES[12] 3 days before operation (T0), as well as 1 week (T1), 1 month (T2), and 3 months (T3) after the procedure. Furthermore, the SSRS[13] was completed at T3.

Of these, the HAMA assessed anxiety symptom severity from somatic (e.g., muscle tension, palpitations) and psychic anxiety (e.g., anxious mood, tension, fear) domains. The total score ranges from 0 to 56 points, and higher scores suggest greater anxiety symptom severity. It is a 14-item clinician-rated scale, with each item scored from 0 (no symptoms) to 4 (severe symptoms). Cronbach’s α for this scale was 0.87 in our study.

The HAMD-17 was used to evaluate the severity of depressive symptoms and included mood, sleep disorders, anorexia, and physical symptoms. The total score ranges from 0 to 52 points, being directly proportional to the severity of depressive symptoms. It is a 17-item clinician-rated scale, with each item scored from 0 (no symptoms) to 4 (severe symptoms). Cronbach’s α for this scale was 0.89 in our study.

The VAS was used to assess pain intensity. Patients rated their own pain levels on a straight line (ranging from 0 cm to 10 cm). The scale is anchored at 0 (no pain) and 10 (worst imaginable pain).

The GSES was used to evaluate individual confidence in coping with difficult situations. It is a 10-item self-report scale, with each item scored from 1 (not at all true) to 4 (completely true). The total score ranges from 10 to 40 points, and higher scores suggest that the individual had a stronger ability to solve problems and cope with challenges. Cronbach’s α for this scale was 0.82 in our study.

The SSRS evaluated the social support that individuals received. It is a 10-item self-report scale covering objective support (3 items), subjective support (4 items), and support utilization (3 items). The total score ranges from 12 to 66 points and is positively correlated with the perceived adequacy of social support. Cronbach’s α for this scale was 0.80 in our study.

Endpoints

Based on the combined HAMA and HAMD-17 data at multiple time points (T0-T3), a latent category growth model (LCGM) was used to identify the heterogeneous trajectories of anxiety and depression levels in older adult patients who had undergone THA to clarify the characteristics and proportions of different trajectory categories. The LCGM was specified using actual time intervals between measurements (0, 1, 4, and 12 weeks postoperatively) to accurately model linear and nonlinear changes in scores over time. Subsequently, multivariate logistic regression was used to analyze independent risk factors, such as age, education level, postoperative pain, and social support.

Statistical analysis

Data were statistically analyzed using SPSS version 28.0. Counting data were recorded as n (%), and χ² tests were used for comparisons. Measurement data that conformed to a normal distribution were recorded with mean ± SD, and an independent sample t-test was used for comparisons; those that did not conform to a normal distribution were recorded as median (P25, P75), with comparisons made by Mann-Whitney U tests. Repeated-measure analysis of variance and Bonferroni intragroup tests were used to compare multiple groups. Multiple logistic regression was used to screen independent influencing factors (α in = 0.05, α out = 0.10), with trajectory grouping as the dependent variable. Statistical significance was set at P < 0.05. Data were double-checked and verified before entry. Variables with a missing value rate of > 5% were excluded.

RESULTS

Demographics and clinical features

Participants’ age range was 65-79 years, with an average age of 69.84 ± 3.98 years. Among them, 57.21% were aged < 70 years. Furthermore, 142 were females (68.27%) and 66 were males (31.73%), with a male-to-female ratio of approximately 1:2.15. The educational level was mainly above high school (52.88%), with below high school accounting for 47.12%. Regarding living arrangements, 114 (54.81%), 73 (35.10%), and 21 (10.10%) patients lived with their spouses, children, and alone, respectively (Table 1).

Table 1 Participants’ demographic and clinical characteristics.

	n	%
Age (year; mean ± SD)	69.84 ± 3.98	-
Age (year; min-max)	65-79	-
Composition of age (year)
< 70	119	57.21
70-79	89	42.79
Sex
Female	142	68.27
Male	66	31.73
Degree of education
≤ High school	98	47.12
> High school	110	52.88
Mode of living
Live with spouse	114	54.80
Live with children	73	35.10
Live alone	21	10.10

Surgical-related indicators

The main types of surgeries were osteoarthritis (104 cases, 50.00%), followed by femoral neck fracture (73 cases, 35.10%) and avascular necrosis of the femoral head (31 cases, 14.90%). Anesthesia methods included 125 and 83 cases of spinal (60.10%) and general anesthesia (39.90%), respectively. Average intraoperative blood loss was 344.36 ± 115.17 mL, with 28 cases (13.46%) < 200 mL, 163 cases (78.37%) between 200-500 mL, and 17 cases (8.17%) > 500 mL; no cases exceeded 1000 mL. The median duration of surgery was 103 (range 44-196) minutes, and 26.44% (55 cases) had a duration of < 90 minutes (Table 2).

Table 2 Participants surgical indicators.

	n	%
Surgical types
Osteoarthritis	104	50.00
Femoral neck fracture	73	35.10
Avascular necrosis of the femoral head	31	14.90
Anesthesia types
Epidural anesthesia	125	60.10
General anesthesia	83	39.90
Intraoperative blood loss (mL; mean ± SD)	344.36 ± 115.17	-
Intraoperative blood loss (mL; min-max)	65-598	-
Composition of intraoperative blood loss (mL)
< 200	28	13.46
200-500	163	78.37
> 500	17	8.17
Surgical duration (minute; mean ± SD)	103.58 ± 23.01	-
Surgical duration (min-max)	44-196	-
Composition of surgical duration (minute)
< 90	55	26.44
90-120	110	52.89
> 120	43	20.67

Pain and self-efficacy

The VAS scores were T0: 3.08 ± 1.13 points; T1: 6.39 ± 1.05 points (peak); T2: 2.51 ± 1.14 points; and T3: 1.75 ± 0.80 points (P < 0.001). The GSES total scores were T0: 22.08 ± 4.94 points (moderately low); T1: 24.12 ± 5.58 points; T2: 26.95 ± 3.98 points; and T3: 29.01 ± 4.57 points (moderate level), gradually increasing (P < 0.001; Table 3).

Table 3 Participants’ Visual Analog Scale and General Self-Efficacy Scale.

	Time	Score
VAS	T0	3.08 ± 1.13
	T1	6.39 ± 1.05a
	T2	2.51 ± 1.14a,b
	T3	1.75 ± 0.80a,b,c
	F	808.024
	P value	< 0.001
GSES	T0	22.08 ± 4.94
	T1	24.12 ± 5.58a
	T2	26.95 ± 3.98a,b
	T3	29.01 ± 4.57a,b,c
	F	84.203
	P value	< 0.001

^aP < 0.05 vs 3 days before operation.

^bP < 0.05 vs 1 week after the procedure.

^cP < 0.05 vs 1 month after the procedure.

T0: 3 days before operation; T1: 1 week after the procedure; T2: 1 month after the procedure; T3: 3 months after the procedure. VAS: Visual Analog Scale; GSES: General Self-Efficacy Scale.

Social support and coping strategies (SSRS)

SSRS scores revealed that the objective (material/service support) and subjective support dimensions (emotional/respect support) had average scores of 13.83 ± 4.13 points (moderate support) and 18.01 ± 6.14 points (high support), respectively. Furthermore, the support utilization dimension (tendency to seek help actively) had an average score of 10.31 ± 6.93 points (moderate utilization). The total score was 42.16 ± 10.32 points, and the overall support status was acceptable (Table 4).

Table 4 Participants’ Social Support Rating Scale.

SSRS	Score
Objective support score	13.83 ± 4.13
Subjective support dimension	18.01 ± 6.14
Utilization support dimension	10.31 ± 6.93
Total score	42.16 ± 10.32

SSRS: Social Support Rating Scale.

Dynamic changes in mental state

Mean changes in the HAMA and HAMD-17 scores exhibited a “rise-followed-by-fall” pattern, with an increase at T1 compared with T0, reaching a peak, followed by continuous decreases at T2 and T3, reaching the lowest value at T3. The effect of time was significant (F = 205.716 and 393.407 respectively, P < 0.001; Table 5).

Table 5 Participants’ Hamilton Anxiety Rating Scale and Hamilton Rating Scale for Depression-17 items scores, mean ± SD or n (%).

	Time	Score	Mild depression/anxiety rate	Moderate depression/anxiety rate
HAMA	T0	10.41 ± 3.14	180 (86.54)	7 (3.37)
	T1	14.95 ± 4.31a	125 (60.10)	78 (37.50)
	T2	10.04 ± 2.85b,c	181 (87.02)	4 (1.92)
	T3	7.28 ± 2.07a,b,c	136 (65.38)	0 (0.00)
	F	205.716
	P value	< 0.001
HAMD-17	T0	11.55 ± 2.53	202 (97.12)	2 (0.96)
	T1	16.00 ± 3.19a	121 (58.17)	87 (41.83)
	T2	10.88 ± 3.06a,b	190 (91.35)	4 (1.92)
	T3	6.83 ± 1.97a,b,c	111 (53.37)	0 (0.00)
	F	393.407
	P value	< 0.001

^aP < 0.05 compared with 3 days before operation.

^bP < 0.05 compared with 1 week after the procedure.

^cP < 0.05 compared with 1 month after the procedure.

T0: 3 days before operation; T1: 1 week after the procedure; T2: 1 month after the procedure; T3: 3 months after the procedure. Hamilton Anxiety Rating Scale (HAMA) 7-16 points: Mild anxiety rate; HAMA 17-24 points: Moderate anxiety rate; Hamilton Rating Scale for Depression (HAMD) 7-17 points: Mild depression rate; HAMD-17 17-24 points: Moderate depression. HAMA: Hamilton Anxiety Rating Scale; HAMD: Hamilton Rating Scale for Depression.

Classification of anxiety and depression trajectories

The HAMA and HAMD-17 scores from T0 to T3 were merged to construct an “anxiety-depression joint score” (a higher score suggested heavier negative emotions). Starting from model 1 (assuming that all patients have the same trajectory), the number of categories (2, 3, and 4) was gradually increased, and the optimal number of categories was determined by comparing the model fitting indicators. Model 2 was confirmed to have the best fit (lowest Akaike information criterion/Bayesian information criterion, entropy 0.85 > 0.8, Lo-Mendell-Rubin likelihood ratio test P = 0.04); therefore, two trajectories were determined (Table 6). The score trajectory plots were constructed. Names and characteristics of the trajectories were defined as follows: (1) Stable group (42.8%, 89 cases): Combined score of the HAMA and HAMD-17 remained at a relatively low level throughout; even though it increased at T1, it quickly dropped below the T0 level at T2 and continued to decrease at T3; and (2) Unstable group (57.21%, 119 cases): Combined HAMA and HAMD-17 scores at T0 were higher than those of the stable group; although it decreased at T2, no significant difference was observed from baseline (Figure 1).

Figure 1 Trajectories of the Hamilton Anxiety Rating Scale + Hamilton Rating Scale for Depression-17 items scores in the two groups. T0: 3 days before operation; T1: 1 week after the procedure; T2: 1 month after the procedure; T3: 3 months after the procedure. HAMA: Hamilton Anxiety Rating Scale; HAMD: Hamilton Rating Scale for Depression.

Table 6 Comparison of the fitting results.

Categories	AIC	BIC	Entropy	LMR-LRT (P value)	Rationality of classification
1	3256.8	3289.2	-	-	A single column cannot highlight heterogeneity
2	2891.2	2952.9	0.85	0.04	2 columns: Stable group (89 cases), unstable group (119 cases)
3	2984.6	3034.6	0.82	0.08	3 columns: Low-risk group (89 cases), medium-risk group (97 cases), high-risk group (22 cases)
4	2875.6	2952.0	0.76	0.12	4 columns: Only 7 cases of 1 class, and stability was poor

AIC: Akaike information criterion; BIC: Bayesian information criterion; LMR-LRT: Lo-Mendell-Rubin likelihood ratio test.

Factors influencing the anxiety and depression trajectory grouping

We used the two types of trajectories identified by the LCGM as dependent variables and analyzed the independent effects of baseline characteristics, surgical parameters, psychological state, and social support on trajectory grouping via a multivariate logistic regression to screen for high-risk factors. Significant differences were observed in age, education level, the VAS, GSES, and SSRS between the two groups (P < 0.05), and further regression analysis was performed (in which the VAS and GESE were analyzed at T1; Table 7). Results revealed that advanced age, lower education level of high school, and VAS at 1 week after surgery were independent risk factors for the trajectory of anxiety and depression. The SSRS and GSES were protective factors (Table 8).

Table 7 Univariate analysis of factors influencing the trajectory of anxiety and depression, mean ± SD or n (%).

	Stable group (n = 89)	Unstable group (n = 119)	t/χ2	P value
Age (year)	67.57 ± 1.86	71.54 ± 4.29	8.158	< 0.001
Sex			0.952	0.329
Female	64 (71.91)	78 (65.55)
Male	25 (28.09)	41 (34.45)
Degree of education			15.603	< 0.001
High school and below	56 (62.92)	42 (35.29)
High school above	33 (37.08)	77 (64.71)
Mode of living			0.288	0.866
Live with spouse	47 (52.81)	67 (56.30)
Live with children	33 (37.08)	40 (33.61)
Live alone	9 (10.11)	12 (10.08)
Surgical types			0.675	0.714
Osteoarthritis	42 (47.19)	62 (52.10)
Femoral neck fracture	32 (35.96)	41 (34.45)
Avascular necrosis of the femoral head	15 (16.85)	16 (13.45)
Anesthesia types			1.012	0.315
Epidural anesthesia	57 (64.04)	68 (57.14)
General anesthesia	32 (35.96)	51 (42.86)
Intraoperative blood loss (mL)	355.60 ± 112.11	335.96 ± 117.17	1.218	0.225
Surgical duration (minutes)	100.66 ± 24.60	105.76 ± 21.60	1.588	0.114
VAS	5.98 ± 0.77	6.71 ± 1.12	5.266	< 0.001
GSES	26.47 ± 4.61	22.35 ± 4.63	5.649	< 0.001
SSRS	46.48 ± 4.66	41.11 ± 7.65	5.862	< 0.001

VAS: Visual Analog Scale; GSES: General Self-Efficacy Scale; SSRS: Social Support Rating Scale.

Table 8 Multivariate analysis of influencing anxiety and depression trajectory.

	β	SE	Wald χ2	P value	Exp (B)	95%CI
						Lower limit	Upper limit
Age	0.401	0.079	25.787	< 0.001	1.493	1.279	1.743
VAS	0.65	0.222	8.591	0.003	1.916	1.24	2.959
GSES	-0.092	0.04	5.276	0.022	0.912	0.843	0.987
SSRS	-0.112	0.033	11.707	0.001	0.894	0.839	0.953
Degree of education	1.236	0.428	8.33	0.004	3.440	1.487	7.962

VAS: Visual Analog Scale; GSES: General Self-Efficacy Scale; SSRS: Social Support Rating Scale.

DISCUSSION

This study used an LCGM to conduct a longitudinal analysis of the anxiety and depression levels of 208 older adult patients who underwent THA. Results revealed that changes in patients’ psychological states exhibited significant heterogeneity in trajectories. Specifically, their anxiety and depression levels peaked at T1 and gradually decreased thereafter. However, some patients experienced moderate-to-severe negative emotions with no improvement. Multivariate analysis confirmed that old age (OR = 1.493), low education level (OR = 3.440), and severe postoperative pain (VAS, OR = 1.916) were independent risk factors for the high-risk groups, whereas social support (OR = 0.894) and general self-efficacy (OR = 0.912) were independent protective factors. To the best of our knowledge, this is the first study to reveal the dynamic evolution patterns of anxiety and depression while providing a quantifiable classification basis for stratified interventions. Notably, the two-trajectory classification provides a more nuanced understanding of psychological outcomes compared with traditional single-time-point assessments, as it captures the dynamic nature of emotional changes over the perioperative period.

In this study, 57.21% of patients were allocated to the unstable group, and their anxiety and depression levels were always in a high state, which was consistent with previous studies[14-16]. Lin et al[14] reported that approximately one-fourth of patients suffered from moderate and severe depressive symptoms three months after THA. Furthermore, although both HAMA and HAMD-17 scores peaked at T1, they began to decrease at T2, suggesting that the one-week to one-month post-operation period is a crucial window for psychological intervention. During this stage, patients may trigger a stress response due to increased pain, limited mobility, and concerns about rehabilitation, which could lead to temporary deterioration in their emotions[17]. This peak at 1 week postoperatively coincides with maximum surgical tissue inflammation and acute pain, which activate the hypothalamic-pituitary-adrenal axis and trigger systemic stress responses. Additionally, significant limitations in activities of daily living at this stage can lead to feelings of helplessness, thereby further exacerbating negative emotions.

Regarding high-risk factors, we believe that older adult patients are more likely to experience a negative emotional cycle due to the decline in physiological reserve and higher comorbidity rate of chronic pain, exemplified by notably higher preoperative pain scores in patients with osteoarthritis compared with patients with fractures[18]. Low education level may reduce patients’ coping ability by limiting their understanding of disease and rehabilitation knowledge[19]. Notably, a low education level emerged as the strongest risk factor (OR = 3.440) in our study, consistent with previous reports linking limited health literacy to catastrophic thinking and higher psychological distress. Postoperative pain is the core driver of emotional deterioration. In this study, patients with a VAS score > 6 at 1-week post-operation exhibited an increased risk of developing persistent high-level anxiety and depression, which aligned with Florencio’s theory of the “pain-anxiety vicious cycle”[20]. Previous studies have repeatedly verified that pain directly causes discomfort and indirectly increases psychological burden by interfering with sleep and limiting social activities[21,22]. Finally, lack of social support weakens patients’ psychological resilience. The increased high-risk level observed in patients with a total SSRS score < 33 points supports the notion that social support plays a role in alleviating stress by offering emotional comfort and resource assistance[23]. Notably, the contribution of spousal support accounted for 54.80%, highlighting the core role of the family system in the rehabilitation of older adult patients. General self-efficacy, reflecting a belief in one’s ability to cope with challenges, was also protective. Patients with higher self-efficacy are more likely to adhere to rehabilitation and set realistic recovery goals, thereby improving both functional and psychological outcomes.

These findings have direct clinical implications for optimizing perioperative psychological care strategies. Thus, we suggest a “three-stage stratified intervention program” for older adult patients undergoing THA: (1) Perioperative early warning stage: Before surgery, high-risk groups (e.g., older adults, those with low education, and those living alone) should be screened, and the baseline level of anxiety should be reduced through measures such as cognitive behavioral therapy; (2) Critical postoperative window period (T1): For patients with a pain VAS score > 6 points, multimodal analgesia (medication + physical therapy) can be implemented while strengthening social support interventions (such as home visits and peer support groups); and (3) Long-term follow-up stage (T2-T3): Individualized psychological counseling can be provided to patients with persistently high adverse mood. Furthermore, progressive functional training can be formulated in conjunction with the rehabilitation department to improve their confidence in prognosis.

However, despite the prospective design and standardized scales, this study has some limitations. First, data collection may have been affected by participants’ subjective reporting biases. Future multicenter prospective studies should utilize larger samples to further verify the findings. Second, all the 208 patients enrolled were from the same grade 3, class A hospital without accounting for geographical and cultural characteristics, which may affect the universality and generalizability of the results. Additionally, the study population was limited to patients aged 65-79 years, and the higher proportion of female patients (68.27%) reflects the higher incidence of hip diseases in older women but may limit the generalizability of findings to male patients and adults aged 60-64 years. Multicenter and large-scale studies can enhance the reliability of our conclusions. Third, although the LCGM model effectively identified the three types of trajectories, it did not differentiate the influence of potential confounding factors (e.g., the number of comorbidities, differences in anesthesia methods) on the trajectories. Thus, mixed effect models or machine learning techniques can be employed to improve the classification accuracy in subsequent research. Finally, although we proposed a hierarchical intervention framework, its effectiveness has not been verified in randomized controlled trials. Future intervention experiments should evaluate the health and economic benefits of these strategies.

CONCLUSION

This study pioneered a systematic exploration of the dynamic evolutionary patterns of anxiety and depression in older adult patients who had undergone THA. Their emotional stress peaked within one week after surgery, and 57.21% presented a persistent psychological risk. Old age, low education level, postoperative pain, and insufficient social support were key risk factors. These results provide a scientific basis for accurately identifying high-risk groups and formulating stratified intervention strategies with profound clinical implications for optimizing the perioperative management of geriatric orthopedics and improving patients’ quality of life.