Cross-Cultural Adaptation and Validation of the Chinese Version of the Exercise Adherence Rating Scale (EARS) in Older Patients with Urinary Incontinence After Prostate Cancer Surgery

Abstract

Background

Adherence is central to the effectiveness of prescribed pelvic floor muscle training after radical prostatectomy, yet standardized tools for assessing home-exercise adherence in older Chinese patients are limited. The Exercise Adherence Rating Scale (EARS) is a brief patient-reported measure of adherence behavior. This study aimed to culturally adapt the EARS for Chinese use and evaluate its reliability and validity among older patients with post-prostatectomy urinary incontinence.

Methods

The EARS was translated and culturally adapted through forward translation, back-translation, multidisciplinary expert review, and pre-testing. Psychometric properties were evaluated in 177 older patients with post-prostatectomy urinary incontinence. Content and structural validity were assessed using content validity indices and confirmatory factor analysis for ordered categorical data, respectively. Reliability was evaluated using Cronbach’s α, odd-even split-half reliability, and two-week test–retest reliability in 30 participants. Pearson correlations examined associations between the supplementary reasons items and the six-item adherence score.

Results

The mean six-item score was 15.71 ± 5.92. The item-level content validity index ranged from 0.883 to 1.000, and the average scale-level index was 0.990. The prespecified one-factor model showed good fit: comparative fit index = 1.000, Tucker–Lewis index = 1.013, root mean square error of approximation = 0.000 (90% confidence interval, 0.000–0.101), and standardized root mean square residual = 0.036. Standardized factor loadings ranged from 0.355 to 0.872 (all P < 0.001). Internal consistency and split-half reliability were acceptable (Cronbach’s α = 0.787; Spearman–Brown coefficient = 0.811), and test–retest reliability was good (intraclass correlation coefficient = 0.879; 95% CI, 0.762–0.940; P < 0.001). Confidence, perceived health benefits, and enjoyment showed the strongest correlations with adherence (r = 0.448–0.554; all P < 0.01).

Conclusion

The EARS-C demonstrated satisfactory content validity, acceptable internal consistency, good reproducibility, and support for the original one-factor structure in this single-center sample. These findings support its use for assessing self-reported exercise adherence while independent multicenter confirmation is pursued.

Background

Prostate cancer is a major threat to men’s health worldwide and ranks among the most commonly diagnosed cancers in men.¹ Radical prostatectomy is a common treatment for localized disease, but urinary incontinence is frequent immediately after catheter removal.² This complication can cause dermatitis and infection and can substantially reduce quality of life and social functioning.³ Pelvic floor muscle training (PFMT) is recommended in clinical guidance for post-prostatectomy urinary incontinence.⁴ Regular PFMT can accelerate continence recovery, but its effectiveness depends partly on whether patients perform the prescribed exercise regimen consistently.^5–7 Previous Chinese research has reported limited adherence to PFMT after prostate surgery, underscoring the need for standardized assessment to support targeted rehabilitation.⁸

Existing approaches to PFMT adherence assessment have important limitations. Exercise diaries and other self-reports are vulnerable to recall and self-presentation bias and are not standardized across studies, whereas digital or app-based monitoring may be difficult for some older patients to use consistently.^9,10 A brief, standardized, condition-independent patient-reported instrument that is specific to prescribed home exercise is therefore clinically useful.

The Exercise Adherence Rating Scale (EARS), developed by Newman-Beinart et al and published in 2017,¹¹ was selected because it was specifically designed to measure adherence to prescribed home exercise, is brief and standardized, separates the six-item adherence construct from supplementary questions about reasons for adherence or non-adherence, and has undergone cross-cultural evaluation in several populations.^12–15 The six scored adherence items form the validated EARS measure, whereas the 10 supplementary reasons items provide single-item information about potential facilitators and barriers.¹¹ However, a Chinese version has not been established for older patients with post-prostatectomy urinary incontinence. Cross-cultural adaptation and psychometric evaluation are therefore needed before the instrument is used in this clinical context.

Materials and Methods

Scale Translation and Cross-Cultural Adaptation of the EARS

Authorization to translate and culturally adapt the EARS was obtained from the original developers. The study followed Guillemin’s guidance and Brislin’s back-translation model.^16,17 A multidisciplinary process was used across four stages: forward translation, back-translation, expert review, and pre-testing.¹⁸

Forward Translation

Two native Chinese-speaking nursing graduates independently translated the EARS from English into Chinese. A bilingual expert synthesized the two translations. The translators and members of the expert group then reviewed terminological and semantic differences and produced a preliminary Chinese version.

Back-Translation

Two bilingual experts independently performed blinded back-translations of the preliminary Chinese version. A nursing master’s graduate synthesized the back-translated versions. The expert group compared the back-translations with the original scale to assess semantic and conceptual equivalence. The synthesized version was then sent to a Duke University professor for review, and revisions were made in response to the feedback received.

Expert Review and Cultural Adaptation

A six-member multidisciplinary expert panel, comprising nurse specialists, rehabilitation medicine experts, and English professors, participated in two Delphi consultation rounds. The panel composition was intended to provide complementary clinical, rehabilitation, and linguistic perspectives relevant to cross-cultural adaptation. Experts rated item clarity, clinical applicability, and cultural appropriateness using a 4-point Likert scale and provided open-ended suggestions.¹⁹ The research team refined the translated items after each round, resulting in the pre-test version of the EARS-C.

Pre-Testing

From September to November 2024, a pre-test was conducted in 100 older patients attending the Urinary Incontinence Nursing Clinic at Shanghai Changhai Hospital. Inclusion criteria were previous radical prostatectomy, urinary incontinence requiring at least one pad daily, age 60 years or older, adequate literacy and communication ability, and provision of informed consent. Exclusion criteria were withdrawal from PFMT, recent pelvic floor rehabilitation, or inability to complete follow-up. Participants’ understanding and suggestions were recorded, and the translation team further optimized the scale to produce the version used for psychometric validation.

Validation of the EARS-C

Study Design and Participants

This validation study was conducted at the Urinary Incontinence Nursing Clinic of Shanghai Changhai Hospital and was approved by the Ethics Committee of the First Affiliated Hospital of Naval Medical University (CHEC2020-136). The study was conducted in accordance with the Declaration of Helsinki, and all participants provided informed consent. Patients were recruited by convenience sampling from September 2024 to March 2025 using the same inclusion and exclusion criteria as the pre-test. According to the COSMIN Study Design Checklist, a factor-analysis sample is considered methodologically very good when it includes at least seven participants per analyzed item and no fewer than 100 participants. As the structural analysis included the six adherence-behavior items, the final sample of 177 participants exceeded this criterion.²⁰

Measurement Instruments

Data were collected using paper-based questionnaires comprising: (a) a sociodemographic questionnaire assessing age, education, residence, income, marital status, and employment status; and (b) the Chinese adaptation of the EARS. Consistent with the original instrument, the scored EARS adherence measure comprises six items rated on a 5-point scale from 0 to 4. Items 1, 4, and 6 are reverse-scored, and the six items are summed to yield a total adherence score from 0 to 24, with higher scores indicating better adherence.¹¹ The additional 10 items assess potential reasons for adherence or non-adherence.

Data Collection Procedure

After radical prostatectomy, patients typically attended follow-up visits at 2 weeks, 1 month, 2 months, and 3 months. At the 2-week visit, a certified nurse removed the urinary catheter, provided education on urinary incontinence management, and instructed patients in home-based PFMT, recommending 150–200 contractions daily. At the 1-month visit, researchers assisted patients in recalling their exercise routine and completing the EARS-C. If an in-person follow-up was missed, questionnaire data were collected by telephone. Researchers received standardized training to improve consistency of data collection.

Statistical Analysis

Statistical analyses were performed using IBM SPSS Statistics version 27.0 and R version 4.6.1. Continuous variables are presented as mean ± standard deviation (SD) or median (Q1, Q3), as appropriate, and categorical variables as frequency and percentage. Pearson correlation coefficients were calculated between each supplementary reasons item and the six-item adherence score. All tests were two-sided, and P < 0.05 was considered statistically significant.

Content validity was evaluated using the item-level content validity index (I-CVI) and the average scale-level content validity index (S-CVI/Ave); I-CVI ≥ 0.78 and S-CVI/Ave ≥ 0.90 indicated acceptable content validity.²¹ Because the original EARS defines a one-factor structure for the six scored adherence items, confirmatory factor analysis (CFA) was used to test this prespecified structure.^11,13,22 The five-category responses were specified as ordered categorical variables, and the model was consequently fitted in the lavaan package using the WLSMV estimator with theta parameterization.²³

According to the established CFA report guidelines, model fit is evaluated using multiple indicators.^24,25 Fit was assessed using the mean- and variance-adjusted chi-square (χ²) test with degrees of freedom (df) and P value, χ²/df, comparative fit index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA) with its 90% confidence interval (CI), and standardized root mean square residual (SRMR). Values of χ²/df < 3, CFI and TLI ≥ 0.95, RMSEA ≤ 0.06 for close fit or ≤ 0.08 for acceptable fit, and SRMR ≤ 0.08 were used as reference criteria.^25–27 Standardized factor loadings ≥ 0.30 were considered interpretable.^13,28 Composite reliability (CR) and average variance extracted (AVE) were also calculated; values ≥ 0.70 and ≥ 0.50, respectively, indicated adequate convergent validity.^29,30

Internal consistency was evaluated using Cronbach’s α, with values ≥ 0.70 considered acceptable.³¹ Odd-even split-half reliability was summarized using the Spearman-Brown coefficient, with values ≥ 0.70 considered acceptable.³² For test-retest reliability, 30 participants were randomly selected after the initial assessment and completed the questionnaire again after two weeks. Agreement between total scores was evaluated using a single-measurement, absolute-agreement, two-way mixed-effects intraclass correlation coefficient (ICC); an ICC of 0.75–0.90 indicated good reliability.³³

Results

Cross-Cultural Adaptation and Pre-Testing of the EARS-C

Systematic linguistic and cultural adaptation was undertaken to improve applicability and clarity for Chinese patients. The scale title was translated into Chinese as the Exercise Adherence Rating Scale, with its use contextualized for pelvic floor muscle exercise. Semantic refinements were made to several items. For example, the initial translation of item 5, “I don’t get around to doing my exercises”, could be confused with lack of time. To distinguish it from item 7, “I don’t have time to exercise”, the final Chinese wording was revised to convey failure to exercise because of delay or being sidetracked rather than simply insufficient time. All modifications were discussed and agreed upon by the expert panel. The English wording of the EARS items is presented in Table 1, and the final Chinese-language EARS-C is provided in Supplementary File 1.

Table 1.

Exercise Adherence Rating Scale Items and Participant Item Scores (n = 177)

Item	Item Score, Mean ± SD	Correlation Coefficient (a)
Exercise Adherence Rating Scale
1. I do my exercises as often as recommended	2.54 ± 1.43	0.559
2. I forget to do my exercises	3.01 ± 1.10	0.542
3. I do less exercise than recommended by my healthcare professional	2.37 ± 1.61	0.704
4. I fit my exercises into my regular routine	1.94 ± 1.62	0.476
5. I do not get around to doing my exercises	3.32 ± 0.96	0.274
6. I do most, or all, of my exercises	2.53 ± 1.63	0.686
Reasons for adherence/non-adherence
7. I do not have time to do my exercises	3.42 ± 0.82	0.360
8. Other commitments prevent me from doing my exercises	3.26 ± 1.15	0.405
9. I do not do my exercises when I am tired	1.72 ± 1.53	0.231
10. I feel confident about doing my exercises	2.82 ± 1.36	0.554
11. My family and friends encourage me to do my exercises	3.05 ± 1.15	0.270
12. I do my exercises to improve my health	3.31 ± 0.94	0.463
13. I do my exercises because I enjoy them	1.51 ± 1.55	0.448
14. I adjust the way I do my exercises to suit myself	1.85 ± 1.60	0.304
15. I stop exercising when my pain is worse	2.03 ± 1.59	0.305
16. I am not sure how to do my exercises	3.23 ± 1.14	0.211

Note: ^(a)For items 1–6, values are corrected item-total correlations. For items 7–16, values are Pearson correlations with the six-item adherence total score; all P < 0.01.

Abbreviation: SD, standard deviation.

Participant Characteristics

During the study, 225 patients were screened and 192 (85.3%) initially agreed to participate. Fifteen were subsequently excluded because of withdrawal due to time constraints (n = 8), loss to follow-up (n = 4), or receipt of electrical stimulation therapy (n = 3), leaving 177 participants in the validation analysis; 30 completed the retest. The mean age was 69.58 years. Most participants were married (89.8%), retired (84.7%), living with a spouse or partner (68.5%), and residing in urban areas (58.2%); 55.6% reported a pension as their main source of income. In terms of education, 39.5% had completed high school or vocational education, 30.5% had an associate degree or higher, and 30.0% had completed middle school or below. The mean item score of the six-item adherence scale was 2.62 ± 0.99 on the 0–4 item scale, and the mean total score was 15.71 ± 5.92 (observed range, 2–24). The 10 reasons items were not combined into a composite score; their item means ranged from 1.51 to 3.42. Item 7, “I don’t have time to do my exercises”, had the highest mean (3.42), whereas item 13, “I do my exercises because I enjoy them”, had the lowest mean (1.51). Detailed item scores are shown in Table 1.

Validity of the EARS-C

Content and Structural Validity

The Delphi expert consultation indicated satisfactory content validity, with I-CVI values ranging from 0.883 to 1.000 and an S-CVI/Ave of 0.990. CFA supported the prespecified one-factor structure of the six adherence items: mean- and variance-adjusted χ²(9) = 7.996, P = 0.535; χ²/df = 0.888; CFI = 1.000; TLI = 1.013; RMSEA = 0.000 (90% CI, 0.000–0.101); and SRMR = 0.036. Standardized factor loadings ranged from 0.355 to 0.872 (all P < 0.001), and all exceeded the prespecified criterion of 0.30 (Table 2). CR was 0.812 and AVE was 0.577. No modification index was ≥ 10, and no post hoc residual covariance or other data-driven model modification was introduced.

Table 2.

Standardized Factor Loadings for the Prespecified One-Factor EARS-C Model

Item	Standardized Loading	SE	95% CI	Z	P value	R2
Item 1	0.718	0.049	0.622–0.815	14.601	< 0.001	0.516
Item 2	0.645	0.058	0.532–0.758	11.170	< 0.001	0.416
Item 3	0.851	0.034	0.785–0.918	24.974	< 0.001	0.725
Item 4	0.558	0.064	0.432–0.684	8.680	< 0.001	0.311
Item 5	0.355	0.083	0.192–0.517	4.289	< 0.001	0.126
Item 6	0.872	0.034	0.804–0.939	25.412	< 0.001	0.760

Note: All standardized loading estimates met the prespecified criterion of ≥ 0.30.

Abbreviations: CI, confidence interval; R², coefficient of determination; SE, standard error.

Reliability

The six-item EARS-C adherence scale showed acceptable internal consistency (Cronbach’s α = 0.787). Corrected item-total correlations ranged from 0.274 to 0.704, and the odd-even Spearman-Brown coefficient was 0.811. Among the 30 participants reassessed after two weeks, the single-measurement, absolute-agreement, two-way mixed-effects ICC was 0.879 (95% CI, 0.762–0.940; P < 0.001), indicating good test-retest reliability.

Associations Between Adherence and Supplementary Reasons Items

Pearson correlation analyses showed associations between each supplementary reasons item and the six-item adherence score (all P < 0.01; Table 1). The largest coefficient was observed for confidence in performing the exercises (item 10, r = 0.554), followed by exercising to improve health (item 12, r = 0.463) and enjoyment of exercise (item 13, r = 0.448).

Discussion

This study developed a culturally adapted Chinese version of the EARS for older patients with urinary incontinence after radical prostatectomy. The six scored items retained the intended one-factor organization, and the scale showed acceptable internal consistency and good stability over two weeks. These findings provide initial evidence that the EARS-C measures a coherent self-reported exercise-adherence construct in older Chinese men undergoing continence rehabilitation. Its brevity may also facilitate repeated assessment during routine postoperative follow-up.

CFA was appropriate because this study sought to test an established structure rather than identify a new one. The original EARS development study used exploratory factor analysis to derive the one-factor six-item adherence scale and analyzed the 10 reasons items separately.¹¹ Although the Nepali adaptation explored the structure using factor analysis,¹² the Brazilian Portuguese study directly tested the established six-item model using CFA.¹³ In the present adaptation, the six scored items, response categories, and scoring direction were retained. The satisfactory fit without post hoc correlated residuals or item deletion therefore supports the cross-cultural reproducibility of the original measurement structure in this Chinese clinical sample.

This study employs multiple indicators to assess the model fitting results. χ²/df (0.888), CFI (1.000), TLI(1.013), and SRMR(0.036) were also favorable. The TLI estimate slightly exceeded 1.000, which can occur because the index is mathematically unbounded and is conventionally interpreted as 1.000. Although the upper limit of the RMSEA 90% CI was 0.101, the point estimate was 0.000. RMSEA may be imprecise in models with few degrees of freedom, so this interval was interpreted alongside the non-significant adjusted χ² test and the favorable incremental and residual indices.²⁴ All six loadings met the prespecified 0.30 criterion. Item 5 contributed less strongly than the other items but retained the intended distinction between postponing or failing to get around to exercise and simply lacking time. Its retention maintains the content coverage and scoring comparability of the original EARS.

The reliability findings were broadly consistent with previous language versions. Cronbach’s α for the EARS-C was 0.787, comparable to the Japanese version (0.77)¹⁵ and close to the original EARS (0.81),¹¹ although lower than the Brazilian Portuguese version (0.88).¹³ For a six-item instrument, this value indicates adequate internal coherence without suggesting unnecessary item redundancy. The split-half coefficient and two-week ICC provided complementary evidence that the score was internally consistent and reproducible over time. Differences in reliability across adaptations may reflect variation in clinical conditions, prescribed exercises, and cultural interpretation rather than a change in the intended construct.

The mean adherence score of 15.71 describes this sample but should not be treated as a clinical cut-off. The 17-point threshold proposed for EARS-Br was derived in patients with chronic low back pain using responsiveness and external outcome data,¹³ and it cannot be assumed to apply to post-prostatectomy continence rehabilitation. Future longitudinal studies should determine score interpretability, responsiveness, and meaningful change in this population.

The 10 supplementary reasons items should likewise remain separate rather than be combined into a psychometric subscale. Their associations with adherence may help clinicians explore confidence, perceived health benefit, enjoyment, fatigue, pain, time, and social encouragement. The pattern of item responses may suggest clinically relevant targets for discussion during rehabilitation. For example, the association between confidence and adherence is compatible with a possible role for exercise self-efficacy, whereas responses concerning time, fatigue, pain, enjoyment, and social encouragement may help clinicians identify patient-specific difficulties or supports.^34,35 However, because the present analyses were correlational, statements about family support, intrinsic motivation, or other mechanisms remain hypotheses rather than confirmed explanations. Longitudinal studies are required to determine whether these factors prospectively influence PFMT adherence.

The EARS-C may also have multidisciplinary value. A common adherence measure could support communication among urologists, nurses, rehabilitation physicians, physiotherapists, continence specialists, and other professionals involved in post-prostatectomy recovery. The six-item adherence score can provide a standardized description of self-reported exercise behavior, while the supplementary reasons items can structure conversations about potentially modifiable barriers and facilitators. In research, a common instrument may facilitate interdisciplinary studies of rehabilitation implementation, behavioral support, and patient-centered continence care. These potential applications should be evaluated beyond the present single-center population before routine broad use.

Strengths and Limitations

This study provides the first Chinese cross-cultural adaptation and psychometric evaluation of the EARS in older patients with urinary incontinence after radical prostatectomy. Strengths include systematic translation, multidisciplinary review, pre-testing, and a prespecified CFA. Several limitations should be considered. First, participants were recruited by convenience sampling from a single tertiary medical center, which creates potential selection bias and limits external validity. Second, the population was relatively homogeneous: participants were older, predominantly married and retired, and many lived with a spouse or partner and in urban areas. The findings therefore may not generalize to younger patients, rural populations, unmarried individuals, women with urinary incontinence, other cultural or socioeconomic groups, or patients with other chronic conditions. Third, the one-factor model was tested in a single sample without independent cross-validation or assessment of measurement invariance across relevant subgroups. Fourth, external convergent, discriminant, and criterion validity were not examined. Fifth, as a self-report instrument, the EARS-C remains susceptible to recall bias. Future multicenter studies should recruit more diverse populations.

Conclusion

The Chinese adaptation of the EARS showed satisfactory content validity, acceptable reliability, and support for the original one-factor structure among older Chinese patients with post-prostatectomy urinary incontinence. The six-item scale provides a brief measure of self-reported exercise adherence, while the supplementary reasons items can support individualized discussion of barriers and facilitators. These findings support further clinical and research evaluation of the EARS-C, including multidisciplinary rehabilitation applications, but do not establish generalizability beyond the studied population. Multicenter validation in more diverse samples is warranted.

Funding Statement

This study was supported by the National Natural Science Foundation of China (Project Number: 81903182), the Deep Blue Nursing Research Fund of Naval Medical University (Project Number: 2022KYD09), and the Changfeng Talent Program of the First Affiliated Hospital of Naval Medical University. The funding providers provided financial support but did not participate in the research process.

Data Sharing Statement

The datasets used and/or analyzed in this study may be requested from the corresponding author or first author upon reasonable request.

Ethics Approval and Consent to Participate

This study has been approved by the Ethics Committee of the First Affiliated Hospital of Naval Medical University (Ethics Review Number: CHEC2020-136) and was conducted in strict compliance with the ethical principles of the Helsinki Declaration. All patients were informed of their relevant rights and signed an informed consent form.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no competing interests.