Investigation and analysis of rumination level and its associated factors in patients with thyroid cancer

Abstract

This study aimed to understand the status of rumination in patients with thyroid cancer and analyze its associated factors, so as to provide reference for medical staff to implement personalized intervention according to the adjustable associated factors of cognitive processing after illness. A total of 159 patients with thyroid cancer were selected by convenient sampling method. The general information questionnaire, event-related Rumination Scale and social support rating scale were used to investigate the patients. SPSS 21.0 software was used for statistical analysis. The total score of rumination was (34.98 ± 7.72), invasive rumination was (16.84 ± 3.99) and purposeful rumination was (18.14 ± 5.00). Multiple linear regression analysis showed that gender, educational level and family per capita monthly income were the main factors associated with invasive rumination; educational level, marital status, family per capita monthly income and social support were the main factors associated with purposeful rumination. The level of rumination in patients with thyroid cancer is relatively high. In clinical work, we should pay more attention to female patients, those with low educational level, unmarried patients, those with low family per capita monthly income and those with poor social support.

1. Introduction

Thyroid cancer is the most common malignant tumor in the endocrine system, accounting for approximately 1% of all malignant tumors.^[1] In recent years, its incidence has been increasing annually worldwide, making it one of the fastest-growing malignancies.^[2–4] This disease predominantly affects young and middle-aged individuals, particularly females. When confronted with traumatic experiences such as diagnosis, surgery, and ¹³¹I therapy, patients may not only experience negative psychological states like anxiety, depression, and pessimism, but also develop positive psychological experiences such as post-traumatic growth.^[5] Rumination is the core process of post-traumatic growth and an important indicator of post-traumatic cognitive processing.^[6,7] The concept of rumination was proposed by Tedeschi et al in 2004, referring to the cognitive processing that facilitates personal growth after experiencing a traumatic event.^[8–10] It has become a research hotspot in the field of positive psychology in recent years. Studies^[11,12] have shown that the type and level of rumination directly affect psychological adjustment after trauma and play a role in coping with stressful events and achieving psychological adaptation. Therefore, this study aimed to evaluate the rumination level and its associated factors in patients with thyroid cancer, so as to provide empirical evidence for medical staff to implement effective psychological interventions for these patients.

2. Materials and methods

2.1. Participants

A cross-sectional survey was conducted on patients with thyroid cancer recruited from the inpatient departments of Endocrinology and General Surgery (or Oncology) at 2 Grade A tertiary hospitals in Sichuan Province from March 2024 to June 2025, using the convenient sampling method. All participants were enrolled at the following specific treatment stages: within 1 week after receiving postoperative pathological confirmation of thyroid cancer, or within 1 week before or after the initiation of the first course of ¹³¹I radiotherapy. This targeted recruitment strategy was designed to capture patients’ cognitive processing during the early, high-stress phases of their cancer journey. The inclusion criteria were as follows: patients diagnosed with thyroid cancer by pathology; aged ≥18 years; patients with clear consciousness, normal ability to communicate verbally, and literacy; patients who voluntarily participated in this study and signed the informed consent form. The exclusion criteria were: a history of mental illness; patients with serious diseases in other systems. A total of 180 questionnaires were distributed in this study, and 159 valid questionnaires were recovered, with an effective response rate of 88.3%. This study was approved by the Ethics Committee of Deyang People’s Hospital.

2.2. Demographic information

It included gender, age, educational level, marital status, family per capita monthly income, number of surgeries, etc.

2.3. Study tool

2.3.1. Event-related rumination inventory

This scale was developed by Cann et al, consisting of 20 items divided into 2 dimensions: intrusive rumination and purposeful rumination, with 10 items in each dimension. All items were scored using a 4-point Likert scale, ranging from 0 points (never having such thoughts after experiencing a traumatic event) to 3 points (often having such thoughts). The total score is 60 points, and a higher score indicates a higher level of cognitive processing in the individual. The Cronbach α coefficients of the Chinese version of the intrusive rumination subscale and the purposeful rumination subscale were 0.94 and 0.88, respectively.^[12]

2.3.2. Social support rating scale

This scale was used to assess an individual’s social support level, consisting of 10 items divided into 3 dimensions: subjective support, objective support, and support utilization. The score range is 12 to 66 points, and a higher score indicates a higher level of social support. The Cronbach α coefficient of the total scale is 0.880.^[13]

2.4. Methods for data collection

A unified guideline was adopted for the questionnaire survey. Information including patients’ pathological type, treatment modality and number of surgeries were completed by the researchers based on medical records, while the remaining items were completed independently by the patients. For patients with reading difficulties, assistance was provided in reading, but no guiding information was given when reading the questions. The questionnaires were distributed and collected on the spot, and their integrity was checked in a timely manner to ensure the validity of the data. The time required to complete the questionnaire was 10 to 15 minutes.

2.5. Statistical analysis

After double-checking, all data were entered, and statistical analysis was performed using SPSS 21.0 software (IBM Corporation). The mean ± standard deviation $\$\overline{x}\pm \mathrm{s}\$$ and the number of cases were used for data description. The t test and analysis of variance were used to compare the rumination of thyroid cancer patients with different characteristics. Pearson correlation analysis was used to explore the relationship between the rumination level and social support of the patients. Multiple linear regression analysis was used for the multifactor analysis of rumination in thyroid cancer patients, with the stepwise method used for model selection. The stepwise regression approach was chosen because it is an exploratory variable selection technique suitable for studies where the relative importance of multiple potential influencing factors is not yet well established. This method automatically selects the most statistically significant variables from a set of candidate predictors, helping to build a more parsimonious and interpretable model by excluding variables with limited explanatory power while controlling for the effects of other variables. The test level α was set at 0.05.

2.6. Strategies to minimize bias

To reduce the risk of common biases in cross-sectional questionnaire surveys, the following strategies were implemented:

1. Selection bias: Although convenience sampling was used, we explicitly defined the recruitment setting (inpatient wards) and timing (within 1 week after diagnosis or peri-¹³¹I therapy) to ensure a clear target population. All eligible patients admitted during the study period were consecutively invited to participate, rather than selectively approaching certain patients.

2. Information bias (measurement bias): Standardized, validated instruments (the Chinese version of event-related rumination inventory and social support rating scale) with established reliability were used. A unified instruction guideline was followed for all participants. For patients with reading difficulties, researchers provided assistance only by reading the questions aloud without any guiding or explanatory modifications to the wording.

3. Social desirability bias: Participants were assured of the anonymity and confidentiality of their responses. They were informed that their answers would not affect their medical care in any way. Questionnaires were completed independently by patients in a quiet, private setting whenever possible.

4. Nonresponse bias: The effective response rate (88.3%) is reported transparently. No significant differences in age or gender were found between responders and nonresponders (data not shown), suggesting that nonresponse bias may be limited.

3. Results

3.1. Rumination status of patients with thyroid cancer

The rumination status of the 159 thyroid cancer patients was assessed, and the detailed scores are presented in Table 1. The total rumination score had a possible range of 0 to 60 (based on 20 items), with the actual mean score for the cohort being 34.98 ± 7.72. The scores ranged from a minimum of 10 to a maximum of 58. When examining the subscales, the score for intrusive rumination (10 items, possible range 0–30) was 16.84 ± 3.99, with actual scores ranging from 0 to 29. In comparison, the score for purposeful rumination (10 items, possible range 0–30) was slightly higher, with a mean of 18.14 ± 5.00 and actual scores ranging from 0 to 30.

Table 1.

Rumination status of patients with thyroid cancer (n = 159).

Item	Number of items	Score ($\$\overline{\mathit{x}}\pm \mathbf{s}\$$)	Maximum score	Minimum score
Total score of rumination	20	34.98 ± 7.72	58	10
Intrusive rumination	10	16.84 ± 3.99	29	0
Purposeful rumination	10	18.14 ± 5.00	30	0

3.2. Univariate analysis of rumination in patients with thyroid cancer

Independent sample t tests or one-way analysis of variance showed statistically significant differences in the scores of intrusive rumination among thyroid cancer patients with different genders, educational levels, marital statuses, work statuses, and family per capita monthly incomes (P < .05). There were also statistically significant differences in the scores of purposeful rumination among thyroid cancer patients with different genders, educational levels, marital statuses, and family per capita monthly incomes (P < .05), as shown in Table 2.

Table 2.

Univariate analysis of rumination in patients with thyroid cancer (n = 159).

Item	Category	Number of Cases	Intrusive rumination ($\$\overline{\mathit{x}}\pm \mathbf{s}\$$)	Purposeful rumination ($\$\overline{\mathit{x}}\pm \mathbf{s}\$$)
Gender	Male	47	13.79 ± 2.61	20.15 ± 2.96
	Female	112	18.12 ± 3.78*	17.30 ± 5.43*
Age (y)	< 40	55	16.93 ± 4.69	18.40 ± 4.26
	40–60	98	16.77 ± 4.66	17.99 ± 5.46
	> 60	6	17.17 ± 2.14	18.33 ± 3.90
Educational level	Primary school and below	31	18.77 ± 4.46*	15.16 ± 7.91*
	Junior high school	47	16.89 ± 3.81	17.43 ± 3.61
	Senior high/technical secondary	15	16.40 ± 3.46	19.20 ± 2.93
	Junior college	35	15.43 ± 2.96	19.29 ± 2.92
	Bachelor’s degree and above	31	16.61 ± 4.36	20.42 ± 4.23
Marital status	Married	134	16.45 ± 3.81*	18.87 ± 3.92*
	Unmarried	25	18.92 ± 4.39	14.24 ± 7.78
Work status	Employed	88	16.26 ± 4.25*	18.43 ± 4.34
	Unemployed	71	17.55 ± 3.55	17.79 ± 5.72
Place of residence	Urban	68	17.09 ± 4.46	18.09 ± 4.74
	Rural	91	16.65 ± 3.62	18.19 ± 5.21
Family per capita monthly income	<2000 Yuan	46	19.41 ± 4.77*	17.50 ± 7.47*
	2000–5000 Yuan	45	16.11 ± 3.11	16.49 ± 3.09
	>5000 Yuan	68	15.57 ± 3.06	19.68 ± 3.27
Pathological type	Papillary carcinoma	131	17.06 ± 3.81	18.40 ± 5.14
	Follicular carcinoma	26	15.58 ± 4.80	16.73 ± 4.20
	Medullary carcinoma	2	18.50 ± 2.12	20.00 ± 1.41
Adjuvant treatment	131I radiotherapy + LT4	44	16.36 ± 3.93	18.30 ± 5.10
	LT4 alone	78	17.31 ± 4.46	18.05 ± 4.88
	Targeted therapy	37	16.41 ± 2.84	18.08 ± 5.24
Number of surgeries	1 time	109	17.16 ± 3.56	18.64 ± 5.09
	≥2 times	50	16.14 ± 4.77	17.06 ± 4.65

LT4 = levothyroxine.

^*P < .05.

3.3. Correlation analysis between social support level and rumination in patients with thyroid cancer

Pearson correlation analysis showed that intrusive rumination was negatively correlated with the total score of social support (P < .05); purposeful rumination was positively correlated with the total score of social support (P < .05), as shown in Table 3.

Table 3.

Correlation analysis between social support score and rumination in patients with thyroid cancer (n = 159).

Item	Total score of social support
	r	P
Intrusive rumination	‐0.173	<.05
Purposeful rumination	0.311	<.05
Total score of rumination	0.148	.063

3.4. Multivariate analysis of rumination in patients with thyroid cancer

Coding of variables in regression analysis: categorical variables were dummy-coded before being entered into the regression models. Specifically: gender (0 = male, 1 = female); marital status (0 = married, 1 = unmarried); work status (0 = employed, 1 = unemployed); place of residence (0 = urban, 1 = rural). Educational level was treated as an ordinal variable and coded as follows: 1 = primary school and below, 2 = junior high school, 3 = senior high school or technical secondary school, 4 = junior college, 5 = bachelor‘s degree and above. Family per capita monthly income was also treated as an ordinal variable: 1 = <2000 Yuan, 2 = 2000–5000 Yuan, 3 = >5000 Yuan. Age (years), number of surgeries (times), and the total score of social support were entered as continuous variables. The dependent variables, intrusive rumination score and purposeful rumination score, were entered as continuous variables. Statistically significant variables from univariate analysis and the total social support score were regarded as independent variables. With intrusive rumination and purposeful rumination scores taken as dependent variables, respectively, multiple linear regression analyses were performed. For intrusive rumination, gender, educational level, and family per capita monthly income entered the regression equation in sequence (F = 11.927, P < .01; R² = 0.320, adjusted R² = 0.293). For purposeful rumination, educational level, marital status, family per capita monthly income, and social support entered the regression equation in sequence (F = 11.411, P < .01; R² = 0.272, adjusted R² = 0.248), as shown in Table 4.

Table 4.

Multivariate analysis of rumination in patients with thyroid cancer (n = 159).

Variable		B	SE	t	P
Intrusive rumination	Constant term		2.399	4.251	<.05
	Gender	3.906	0.694	5.631	<.05
	Educational level	0.567	0.271	2.090	<.05
	Family per capita monthly income	‐1.774	0.486	‐3.646	<.05
Purposeful rumination	Constant term		2.960	6.270	<.05
	Educational level	1.376	0.346	3.979	<.05
	Marital status	‐3.139	1.049	‐2.993	<.05
	Family per capita monthly income	‐1.678	0.628	‐2.674	<.05
	Social support	0.136	0.041	3.302	<.05

Coding of categorical variables: gender (0 = male, 1 = female); marital status (0 = married, 1 = unmarried); educational level (1 = primary school and below, 2 = junior high school, 3 = senior high school or technical secondary school, 4 = junior college, 5 = bachelor’s degree and above); family per capita monthly income (1 = < 2000 Yuan, 2 = 2000–5000 Yuan, 3 = >5000 Yuan).

B = unstandardized regression coefficient; SE = standard error.

4. Discussion

4.1. Rumination level of patients with thyroid cancer

The results of this survey showed that the scores for intrusive rumination and purposeful rumination in patients with thyroid cancer were (16.84 ± 3.99) and (18.14 ± 5.00), respectively. Each subscale had a theoretical median score of 15, and both subscale scores were higher than this midpoint. The total rumination score was (34.98 ± 7.72). Given that the theoretical median score for the total scale was 30, the overall rumination level of patients with thyroid cancer was relatively high. The rumination scores in this study were higher than those of patients with accidental trauma reported by Gong Shumei et al.^[14] The reason might be that patients with accidental trauma gradually return to society as their condition improves, while patients with thyroid cancer may face radiotherapy, lifelong medication, and disease recurrence after treatment. Such long-term disease-related stress contributes to persistent rumination among these patients.

4.2. Main factors associated with rumination in patients with thyroid cancer

4.2.1. Gender

The results of this study showed that there was a statistically significant difference in the rumination scores among patients of different genders (P < .05), and female patients had a higher level of intrusive rumination. The reason might be that women tend to be more sensitive to external events, more delicate in emotions, and poorer in psychological resilience. Their emotions are more likely to be affected by stressful events, making them prone to repetitive and negative thinking about events.^[15,16] Therefore, during the treatment process, medical staff should actively encourage female patients with thyroid cancer to express their emotions and prevent them from indulging in negative emotions.

4.2.2. Educational level

The results of this study showed that there was a statistically significant difference in the rumination scores among patients with different educational levels (P < .05), and patients with higher educational levels had a higher level of purposeful rumination, which was consistent with the research results of Chen Jianhui et al^[15] and Taku et al.^[17] The reason was that patients with higher educational levels had stronger abilities to acquire and understand knowledge and could view disease treatment, rehabilitation, and prognosis more rationally.^[18] In addition, patients with higher educational levels often master more psychological adjustment skills and can vent their emotions through various channels, thereby promoting them to think and self-regulate stressful events purposefully and find benefits from them.^[19] It can be seen that clinical medical staff should focus on patients with lower educational levels, introduce disease and treatment-related knowledge to them in an easy-to-understand way to help them face their own diseases correctly. At the same time, medical staff should strengthen psychological counseling for patients to promote their positive psychological adjustment.

4.2.3. Marital status

The results of this study showed that there was a statistically significant difference in the rumination scores among patients with different marital statuses (P < .05), and unmarried patients had a higher level of intrusive rumination. Studies have shown that marriage is one of the important factors associated with post-traumatic growth, and married individuals can obtain more post-traumatic growth.^[20] As the main source of social support for patients, spouses not only cope with the disease together with them but also provide understanding and support to patients through listening, counseling, encouragement, and other methods. This kind of care and support from spouses can reduce the negative emotions caused by the disease to a certain extent and promote them to cope with the disease positively. Therefore, marriage plays a “protective umbrella” role in the process of patients’ illness, buffers the impact of emergency events, and thus reduces the level of intrusive rumination of patients. This suggests that medical staff should pay attention to unmarried patients, strengthen care and attention to such patients, and thus promote their positive psychological adjustment.

4.2.4. Family per capita monthly income

The results of this study showed that there was a statistically significant difference in the rumination scores among patients with different family economic levels (P < .05), and patients with a lower family per capita monthly income had a higher level of intrusive rumination, which was consistent with the research results of Chen Jianhui et al^[15] and Yang Jun et al.^[21] Thyroid cancer usually requires a long and frequent course of radiotherapy. Moreover, this disease mainly occurs in young and middle-aged people, who are usually the primary economic contributors to their families. After illness, hospitalization leads to the partial interruption of the main family economic source. In addition, the expensive medical expenses increase the family economic burden to a certain extent. Therefore, for patients with poor family economic conditions, in addition to bearing the physical pain caused by the disease and its treatment, they also have to face considerable economic pressure, which makes them prone to negative psychology such as pessimism, guilt, and anxiety, and then fall into repetitive and negative thinking. This suggests that clinical medical staff should implement personalized interventions according to the economic status of patients, and jointly determine the optimal treatment plan with patients under the full consideration of their economic conditions, so as to reduce their economic pressure to a certain extent. At the same time, medical staff should strengthen mental health education for patients, guide them to change their cognitive attitudes, and gradually transform to a constructive cognitive processing method, so that they can better adjust their psychology after illness and alleviate the psychological distress caused by family economic difficulties.

4.2.5. Social support

The results of this study showed a statistically significant difference in rumination scores among patients with different social support levels (P < .05). The higher the level of social support, the lower the level of intrusive rumination and the higher the level of purposeful rumination, which was consistent with the research results of Chen Jianhui et al.^[15] Social support is an important resource for patients to cope with disease and can provide a positive force.^[22] Good social support can reduce the fear, anxiety, and loneliness caused by cancer, a serious traumatic event, alleviate negative emotions, and help patients establish confidence in overcoming the disease.^[23,24] Patients with good social support can get more care and support when facing stressful events, which helps them form a positive cognitive processing form.^[25] Therefore, clinical medical staff should take the initiative to care for patients in daily work, encourage family members to provide more behavioral and emotional support to patients, strengthen their family support, and at the same time guide patients to actively seek effective social support to improve their social support level, create good conditions for the generation of positive cognitive processing methods, and promote their rehabilitation.

5. Limitations

This study has several limitations that should be acknowledged. First, the cross-sectional design employed only allows for the assessment of associations between variables at a single point in time. Consequently, it is impossible to infer causal relationships between the identified associated factors (such as social support) and the levels of intrusive and purposeful rumination. Second, the study participants were recruited solely from inpatient departments of 2 tertiary hospitals within a single province in China using a convenience sampling method. This sampling approach and setting may introduce selection bias, as patients recruited from inpatient wards during specific treatment windows (e.g., early post-diagnosis or peri-¹³¹I therapy) might have different levels of rumination compared to outpatients, long-term survivors, or those in other treatment phases (e.g., chronic hormone suppression therapy). Consequently, the sample may not fully represent the entire spectrum of the thyroid cancer population. These factors limit the geographical and clinical generalizability of the findings to broader patient populations in other regions or different healthcare settings. Future research should aim to employ more rigorous sampling strategies, such as stratified random sampling across multiple centers and diverse clinical settings (including outpatient clinics and cancer rehabilitation centers), and across various treatment stages to enhance the representativeness of the sample. Third, all data, including rumination levels and social support, were collected through patient self-reporting instruments. This method is susceptible to response biases, such as social desirability bias or recall bias, which may affect the accuracy and objectivity of the data. Fifth, the present study did not analyze rumination levels using the clinically significant age cutoff of 55 years, which is a critical threshold for the staging of most histological types of thyroid cancer. The primary reason is that the study was originally designed to focus on broad age group comparisons (<40 years, 40–60 years, and > 60 years) based on previous literature on psychological adaptation in cancer patients. Additionally, the sample size in the >60 years group was very small (n = 6), and reclassifying participants using a 55-year cutoff would result in highly uneven subgroup sizes (e.g., ≤55 years: n = approximately 126; >55 years: n = approximately 33), which could compromise the statistical power and stability of the analysis. Nevertheless, we fully acknowledge the clinical importance of the 55-year threshold as suggested by the reviewer. Future studies with larger sample sizes and more balanced age distributions should specifically examine rumination levels using this clinically meaningful cutoff to better integrate psychological outcomes with cancer staging systems. Finally, this study did not systematically collect or control for key clinical and psychological variables such as cancer stage, time since diagnosis, specific treatment phase, or other psychological states like depression and anxiety. These unmeasured variables could confound the observed associations. Otherwise, despite the implementation of multiple bias-reduction strategies as described in Section 2.6, certain biases cannot be completely eliminated. For instance, the use of convenience sampling inherently limits the representativeness of the sample, and although we attempted to consecutively invite all eligible patients, selection bias may still exist. Additionally, self-report instruments are subject to recall bias and current mood states, which we could not fully control for.

Finally, despite the implementation of multiple bias-reduction strategies as described in Section 2.6, certain biases cannot be completely eliminated. For instance, the use of convenience sampling inherently limits the representativeness of the sample, and although we attempted to consecutively invite all eligible patients, selection bias may still exist. Additionally, self-report instruments are subject to recall bias and current mood states, which we could not fully control for.

Based on these limitations, future research could be improved by conducting longitudinal or nationwide multicenter studies. Such designs would help track the dynamic changes in rumination over time, establish causality, and incorporate more representative and diverse samples. Furthermore, supplementing self-report data with objective clinical measures or qualitative interviews could provide a more comprehensive understanding of the cognitive processing in thyroid cancer patients. In addition, future studies should adequately control for potential confounding clinical and psychological variables to refine the robustness of analytical findings.

6. Implications

This study offers significant theoretical and practical implications. Theoretically, it shifts the focus beyond the prevalent psychological issues of anxiety and depression in thyroid cancer patients, delving into the core cognitive process of post-traumatic growth (rumination). By quantitatively describing the levels of both intrusive and purposeful rumination and elucidating their distinct associated factors, this research enriches the theoretical framework of cognitive processing in cancer survivors and provides a valuable empirical foundation for understanding the psychological adaptation mechanisms in this population.

Practically, the findings provide actionable evidence for clinical healthcare providers. The identification of key associated factors (such as gender, education, marital status, income, and social support) allows clinicians to quickly identify patient subgroups at higher risk for maladaptive cognitive processing (e.g., females, those with lower education or income, and unmarried individuals). This enables the implementation of targeted, personalized psychological interventions. For instance, medical staff can prioritize psychological counseling for female and unmarried patients, provide easy-to-understand health education for those with lower education levels, and collaborate with patients facing financial difficulties to develop economically feasible treatment plans. Most importantly, the strong link between social support and higher purposeful rumination highlights a critical intervention pathway. Clinicians should proactively engage family members to strengthen emotional and behavioral support for patients and guide patients themselves to actively seek and utilize social resources. These measures are crucial for facilitating a transition from intrusive to purposeful rumination, ultimately fostering post-traumatic growth and promoting comprehensive patient rehabilitation.

7. Conclusion

In this study, the rumination level of patients with thyroid cancer was relatively high. Gender, educational level, and family per capita monthly income were the main factors associated with intrusive rumination, while educational level, marital status, family per capita monthly income, and social support were the main factors associated with purposeful rumination. For such patients, clinical medical staff should carry out targeted psychological interventions and, at the same time, fully mobilize the social support system of patients to promote the transformation of patients from intrusive rumination to purposeful rumination, so as to help patients better adjust their psychology.

Author contributions

Conceptualization: Jie Liu, Jinlan Dai.

Data curation: Jie Liu.

Formal analysis: Jie Liu, Li Tao.

Investigation: Li Tao, Mengting Zhu, Siyu Ling, Huanhuan Deng, Jinlan Dai.

Methodology: Jie Liu, Li Tao, Siyu Ling, Huanhuan Deng, Jing Gu.

Project administration: Li Tao.

Resources: Huanhuan Deng, Jing Gu.

Software: Li Tao, Siyu Ling.

Supervision: Mengting Zhu.

Writing – original draft: Mengting Zhu, Jing Gu.