Cognition of the warning symptoms and risk factors for cancer among Chinese college students: a cross-sectional study based on a summer social practice activity

Abstract

Background

To investigate the cognition of cancer warning symptoms and cancer risk factors among Chinese college students, analyze the influencing factors, and explain the correlations between cancer cognition and cancer symptom discrimination, cancer fear and psychological distress.

Methods

Chinese college students were recruited in this cross-sectional study funded by a summer social practice activity in Yunnan Province, China. Cognition rates of cancer warning symptoms and cancer risk factors were evaluated using Cancer Warning symptoms Cognition Questionnaire (CWSCQ) and Cancer Risk Factors Cognition Questionnaire (CRFCQ), respectively. Factors associated with cognition of cancer warning symptoms, and factors associated with cognition of cancer risk factors were evaluated using multiple linear regression analysis. Interactions between cancer cognition, cancer symptom discrimination, psychological distress, and cancer fear were evaluated by structural equation modeling.

Results

There were 846 effective samples, with an effective rate of 80.9%. The cognition rates of cancer warning symptoms were from 47.9% to 84.4%, which were affected by cancer symptom discrimination, education, attitudes towards cancer screening, living expenses, drinking history, and ways to obtain cancer knowledge (p < 0.05). The cognition rates of cancer risk factors were from 46.3% to 91.3% in participants, which were affected by education, cancer symptom discrimination, psychological distress, attitudes towards cancer screening, life satisfaction, cancer history in relatives and friends, ways to obtain cancer knowledge, smoking history, and nursing history for cancer patients (p < 0.05). Cancer cognition and cancer symptom discrimination showed intermediary effects on psychological distress and cancer fear (p < 0.001).

Conclusions

The overall cancer cognition situation among Chinese college students is not optimistic, which highlights the necessity of improving the cancer health literacy among Chinese college students. With the increasing morbidity and mortality rates of cancer, it is necessary to raise awareness of early detection, and early treatment of cancer among the general public. Health education interventions are helpful to improve cancer health literacy.

Keywords:

• Cancer warning symptoms
• cancer risk factors
• cognition
• Chinese college students

1. Background

Burden resulted from cancer is increasing in all countries and regions, with increasing cancer morbidity and mortality rates around the world year by year [1]. Recent research had revealed that the number of newly diagnosed cancer patients in China was 4.0640 million people, and about 2.4135 million people died of cancer in 2016 [2]. A lack of necessary cancer health knowledge and education in the public may lead to low awareness of cancer and cancer screening and a lack of timely and effective medical services, consequently, resulting in a further increase of morbidity and mortality of cancer [3].

Cancer cognition refers to the individual ‘s cognitive level of cancer-related knowledge including but not limited to etiology, clinical manifestations, treatment, prevention, and prognosis. It is reported that cognition of cancer among healthy people is inadequate in various countries and regions. Also, studies related to people’s cognition and knowledge of cancer have mostly been focused on cervical cancer. In a medical college in Gondar Town, Ethiopia, 59.3% of female Chinese college students surveyed had a good cognition of cervical cancer (at least 6 out of 11 questions related to cervical cancer knowledge were answered correctly), and 67.7% had a positive and active attitude towards prevention and treatment of cervical cancer [4]. However, among women of childbearing age in the same region, 81.4% of respondents did not know HPV was a trigger factor of cervical cancer; 80% of them did not know the symptoms of cervical cancer, although 65.1% of the respondents had heard of it [5]. Cognition of cervical cancer is closely related to many factors, including education, an access to knowledge and information on cervical cancer, history of contraceptive uses, income, occupation, medical histories, times of delivery [6,7]. Cognition of cervical cancer among women may also be related to their spouses. An Eswatini study found that 58.1% of male and female respondents had misconceptions about risk factors for cervical cancer, and 40.0% of female respondents said they needed their spouses’ consent before seeking medical treatment [8]. In addition, the floating population may also be a high-risk group which especially lacks cognition and knowledge of cervical cancer. A study in Shenzhen, China, found that cognition and knowledge of HPV and HPV vaccines among local permanent residents was significantly higher than those in the temporary residents and the floating population [9].

When it comes to digestive tract malignancies, a study in Hunan Province, China, found that the most recognized warning symptoms of gastric cancer in a descending order were digestive tract bleeding (49.8%), abdominal lumps (45.1%), and abdominal pain (43.7%); the lowest perceived warning symptoms were weight loss (37.7%) and frequent nausea and vomiting (34.4%); the most recognized risk factors for gastric cancer were irregular diets (59.8%), pickled food (57.5%), alcohol consumption (56.8%), smoked food (55.8%) and salty food (50.9%); the least recognized risk factors for gastric cancer were being a male (25.3%), having a history of gastric surgery (23.8%), and regularly eating late night snacks and meals (18.8%) [10]. Among people at high risk of colorectal cancer, the most recognized risk factors for colon cancer included inflammatory bowel diseases (63.2%), followed by colon polyps (62.1%) and excessive alcohol consumption (60.4%); The risk factors with low cognition levels included a family history of colorectal cancer (49.6%), smoking (48.1%), older age (45.3%), overweight or obesity (44.0%), genetic syndrome (40.8%) and diabetes (11.5%); The warning symptoms with the highest cognition levels for colon cancer were hematochezia (76.0%), rectal bleeding (74.0%) and changes in defecation habits (66.4%), while the warning symptoms with the lowest cognition levels included unexplained weight loss (57.3%), bowel doesn’t empty(55.3%) and fatigue/anemia (29.5%); Only 40.9% of respondents knew that colorectal cancer could be asymptomatic [11].

In terms of breast cancer, the rates of breast self-examination and active health check-ups are closely related to the cognition levels of breast cancer. Raising women’s awareness and cognition of the risk factors can help women look for and accept early diagnoses for breast cancer [12]. A lack of breast cancer awareness and cognition may lead to lower breast self-examination rates [13]. 30.7% of the middle-aged women surveyed in Macao, China, had never undergone any breast self-examination (BSE). The risk factors for a low rate of breast self-examination are low education, not acquainted with any breast cancer patients, no history of breastfeeding, having a negative attitude towards health check-ups, limited knowledge of breast cancer, and unwillingness to take a mammography screening; women who do not know anyone with breast cancer are more likely to have an insufficient knowledge of breast cancer; low education and not knowing anyone with breast cancer were identified as predictors for perceived barriers to mammography in women [14].

To sum up, studies from both China and other countries suggest that cognition and knowledge of cancer in the general public is insufficient, neither is the related health education for cancer. At present, there are only a few studies on cancer cognition in China and other countries, and these studies are mostly limited to the cognition status and influencing factors of a specific cancer. There is a lack of research on the overall cognition of common warning symptoms and risk factors for cancer, and systematic research in China is especially inadequate. In this study, Chinese college students were recruited as the research subjects in a summer social practice activity (SSPA) in Yunnan Province, in 2021. This study investigated the cognition of common warning symptoms and risk factors for cancer among Chinese college students, analyzed the influencing factors, and explained the correlations between cancer cognition and cancer symptom discrimination, cancer fear and psychological distress.

2. Methods

2.1. Study design

This was a cross-sectional study. Chinese college students were recruited to participate in a survey funded by a summer social practice activity (SSPA) in Yunnan Province, China (from August 6 to August 31, 2021). Judgment sampling was used as the sampling method, and its inclusion criteria were: People who were ① Chinese college students, ② voluntary participants in the survey, ③ ≥16 years old, and who ④ filled the correct name of his/her university, ⑤ denied a history of malignant tumor in the past and at present (including non-hematopoietic tumors and hematopoietic tumors).

The survey was completely voluntary, anonymous, confidential and there were no commercial interests involved. The respondents were not induced to participate in the survey by any incentive/punitive measures. To enable the college students to communicate better in this summer project, the event organizers have established WeChat groups or QQ groups for them. With the help of the event organizers, the QR code and link of an electronic questionnaire were sent to the students in the WeChat groups and QQ groups, and the target population could scan the QR code or click the link to read the questionnaire manual. The respondents were reviewed and selected for their qualification to participate in the survey. The program automatically loaded the questionnaire if the respondent met the inclusion criteria, and would exit automatically if s/he did not. It was required to complete each and every item in the questionnaire so that there was no missing value, otherwise the questionnaire would not be submitted successfully.

According to the sample estimation formula, the confidence degree of 99%, and the health literacy level of tumor prevention and control of urban residents in China of 72.3% were used as the references [15]. The ratio of cognition level of cancer warning symptoms π was expected to be 75%, allowable error δ < 5%, and α = 0.05 was taken. According to the following formula, the minimum sample size of this study was calculated to be 500 people, but considering a 20% invalid response rate, the minimum sample size should be set at least 625 people. $$N={\mu }_{\alpha ∕2}^2\times \pi \left(1-\pi \right)/{\delta }^2$$

2.2. Study tools

In order to investigate the cognition level of respondents on common cancer warning symptoms and risk factors, we invited 5 experts from oncology, public health, and nursing to conduct expert consultation, and designed Cancer Warning symptoms Cognition Questionnaire (CWSCQ) and Cancer Risk Factors Cognition Questionnaire (CRFCQ) for investigation. The questionnaire entries used non-medical wording which are synonymous with warning symptoms and risk factors. After that, we invited 11 college students to conduct cognitive interviews to ensure that each item was easy to understand and its understanding by the respondents was consistent with the actual meaning represented by the language.

2.2.1. CWSCQ

A self-compiled questionnaire, including 20 common cancer warning symptoms. The participants were required to identify whether each item was a cancer warning symptom. They were all single choice questions. Once a ‘Yes’ was chosen, 1 point was counted. Choosing a ‘No’ or ‘Don’t know’ meant 0 point. The score range was from 0 to 20. Higher scores indicated higher cognition levels of cancer warning symptoms in the participants.

2.2.2. CRFCQ

A self-compiled questionnaire, including 14 cancer risk factors. The participants were required to identify whether each item was a risk factor for cancer. They were all single choice questions. Once a ‘Yes’ was chosen, 1 point was counted. Choosing a ‘No’ or ‘Don’t know’ meant 0 point. The score range was from 0 to 14. Higher scores indicated higher cognition levels of cancer risk factors in the participants.

2.2.3. A form for demographic sociological characteristics

There are 30 variables, including gender, age, place of residence, nationality, major, educational background, grade, guardian’s educational background, monthly living expense, the number of family members, self-perceived family economic status, self-perceived family relationship, personality characteristics, self-perceived health status, self-perceived learning pressure, self-perceived economic pressure, self-perceived social pressure, life satisfaction, whether s/he is an only child, history of serious diseases, history of cancer in relatives and friends, nursing history for cancer patients, attitudes towards cancer screening, smoking history, drinking history, the frequency of staying up late, the frequency of eating junk food, the frequency of being furious, the frequency of crying secretly, and the way to obtain cancer knowledge.

2.2.4. Fear of cancer scale (FOCS)

A 17-item scale from our previous research was used to measure the fear of cancer in healthy people. It is divided into two dimensions: direct fear and indirect fear, with a score range from 0 to 68 points. Higher scores indicate higher fear levels towards cancer in the participants. This scale had good reliability and validity in Chinese college students [16].

2.2.5. Cancer symptoms discrimination scale (CSDS)

A 21-item scale from our previous research was used to measure the level of symptom discrimination of the public against cancer patients. CSDS is divided into three dimensions: common clinical manifestations, physical appearance defects and drainage tube(s) wearing. Higher scores indicate higher levels of cancer symptom discrimination in the participants. The scale had good reliability and validity in Chinese college students [17].

2.2.6. Generalized anxiety disorder-7 (GAD-7)

A classic scale for screening and measuring psychological distress and generalized anxiety disorder in different groups, which contains 7 items. The score range is from 0 to 21 points. Higher scores indicate higher levels of psychological distress. Several studies had confirmed that GAD-7 has good reliability and validity in college students [18–20].

2.3. Statistical analysis

The database was established by Microsoft Excel 365. IBM SPSS Statistics 19.0 and SPSS AMOS 24.0 were used for data processing and statistical analysis. Confirmatory factor analysis was used to test the construct validity of FOCS and CSDS, and the model fit was calculated. Cronbach α and Spearman-Brown coefficient were used to evaluate the internal consistency reliability and split-half reliability of the CWSCQ, CRFCQ, FOCS, CSDS, and GAD-7, respectively. Descriptive statistics were used to analyze the current cognitive status of cancer warning symptoms and risk factors among college students. The scores of CWSCQ, CRFCQ, GSES, FOCS, CSDS, and GAD-7 were normally distributed by normality test.

Kruskal-Wallis Test was used to compare the differences in cognition of cancer warning symptoms and cancer risk factors among Chinese college students with different characteristics. The independent variables with p < 0.10 were substituted into a multiple linear regression model to analyze the influencing factors for cognition of cancer warning symptoms and cancer risk factors among Chinese college students. We used AMOS 24.0 to construct a structural equation model to further explain the interaction among psychological distress, cancer cognition, cancer symptom discrimination, and cancer fear.

The significant level was α = 0.05, and all P values represented bilateral probabilities.

3. Results

3.1. The social demographic characteristics of the participants

There were 1046 respondents in this study, among whom 16 refused to participate in the survey, 84 were not Chinese college students, 82 claimed a history of malignant tumors, 12 did not fill in the correct name of their universities, and 6 studied in universities in other countries. They were excluded. There were 846 effective samples, with an effective rate of 80.9%, including 530 female students (62.6%), 231 medical students (27.3%), and 340 rural students (40.2%). See Table 1 for details.

Table 1. The social demographic characteristics of the participants (N = 864).

Characteristics	N (%)
Gender
Female	530(62.6%)
Male	316(37.4%)
Age
16 ∼ 19 years	383(45.3%)
≥20 years	463(54.7%)
Place of residence
Rural areas	340(40.2%)
Urban areas	506(59.8%)
Nationality
Han	657(77.7%)
Ethnic minority	189(22.3%)
Major
Medical sciences	231(27.3%)
Non- Medical sciences	615(72.7%)
Education
Junior college	115(13.6%)
Undergraduate	667(78.8%)
Postgraduate	64(7.6%)
Grade
Grade 1	263(31.1%)
Grade 2	302(35.7%)
Grade 3 and above	281(33.2%)
Guardian’s educational background
Middle school and below	447(52.8%)
High school and above	399(47.2%)
Monthly living expenses
≤￥1200 yuan	365(43.1%)
>￥1200 yuan	481(56.9%)
Number of family members
≤ 4 people	551(65.1%)
≥ 5 people	295(34.9%)
Self-perceived family economic status
Average	233(27.5%)
Good	613(72.5%)
Self-perceived family relationship
Average	279(33.0%)
Harmonious	567(67.0%)
Personality characteristics
Non-extroversive	736(87.0%)
Extroversive	110(13.0%)
Self-perceived health status
Poor	30(3.5%)
Fair	489(57.8%)
Good	327(38.7%)
Self-perceived learning pressure
Less	694(82.0%)
Much	152(18.0%)
Self-perceived economic pressure
Less	140(16.5%)
Much	706(83.5%)
Self-perceived social pressure
Less	687(81.2%)
Much	159(18.8%)
Life satisfaction
Average	62(7.3%)
Higher	784(92.7%)
Whether s/he is an only child
Yes	239(28.3%)
No	607(71.7%)
History of serious diseases
Yes	43(5.1%)
No	803(94.9%)
History of cancer in relatives and friends
No	660(78.0%)
Yes	186(22.0%)
Nursing history for cancer patients
No	761(90.0%)
Yes	85(10.0%)
Attitudes towards cancer screening
Negative	68(8.0%)
Positive	778(92.0%)
Smoking history
No	715(84.5%)
Yes	131(15.5%)
Drinking history
No	329(38.9%)
Yes	517(61.1%)
Frequency of staying up late
Occasionally	427(50.5%)
Frequently	419(49.5%)
Frequency of eating junk food
Occasionally	672(79.4%)
Frequently	174(20.6%)
Frequency of being furious
Occasionally	804(95.0%)
Frequently	42(5.0%)
Frequency of crying secretly
Never	180(21.3%)
Occasionally	666(78.7%)
The way to obtain cancer knowledge
≤3	402(47.5%)
≥4	444(52.5%)

3.2. Reliability of study tools

The Cronbach coefficient of CWSCQ, CRFCQ, FOCS, CSDS and GAD-7 were 0.944, 0.896, 0.933, 0.942 and 0.951, respectively. The results showed good internal consistency reliability of these tools.

The split-half reliability of the scale was verified by dividing the items into two halves based on odd and even numbers. Spearman-Brown Coefficient of CWSCQ, CRFCQ, FOCS, CSDS and GAD-7 were 0.962, 0.896, 0.950, 0.955 and 0.960, respectively. The results showed good split-half reliability of these tools.

3.3. Cognition levels of cancer warning symptoms in Chinese college students

The cognition rates of cancer warning symptoms were from 47.9% to 84.4%. The cancer warning symptoms with higher cognition levels among Chinese college students were: abnormal mass (84.4%), cervical lesions (78.9%), repeated hemoptysis or vomiting (75.8%), liver cirrhosis (73.4%), unexplained weight loss (72.9%), pulmonary nodules (72.7%), and continuous aggravation of abdominal pain (71.5%). The cancer warning symptoms with lower cognition levels among Chinese college students were: oral leukoplakia (59.7%), dysphagia (59.3%), repeated cough (53.9%), fatigue that cannot be relieved after rest (51.5%), and persistent dyspepsia (47.9%). See Figure 1 for details.

Figure 1. The cognition levels of cancer warning symptoms among Chinese college students (N = 846).

The multiple linear regression analysis showed that the influencing factors for cognition of cancer warning symptoms among Chinese college students were cancer symptom discrimination, education, attitudes towards cancer screening and health check-ups, living expenses, drinking history and ways to obtain cancer knowledge (p < 0.05). The collinearity diagnosis showed that the tolerance and variance expansion factor of 6 independent variables were close to 1, indicating that there was no collinearity problem among the independent variables. See Table 2 for details.

Table 2. The influencing factors for the cognition levels of cancer warning symptoms among Chinese college students .

Variable	β	S.E.	Standardized regression coefficient	t	P	VIF
CSDS score	0.177	0.030	0.193	5.850	<0.001	1.058
Education	2.406	0.476	0.170	5.049	<0.001	1.105
Attitudes towards cancer screening	2.806	0.769	0.118	3.649	<0.001	1.023
Monthly living expenses	1.503	0.440	0.116	3.417	0.001	1.111
Drinking history	−1.371	0.428	−0.104	−3.202	0.001	1.020
The way to obtain cancer knowledge	0.839	0.424	0.065	1.978	0.048	1.051

Dependent variable:

CWSCQ score: 0–20 points.

Independent variable:

CSDS score: 0–21 points; Education: 1 = junior college, 2 = undergraduate, 3 = postgraduate; Attitudes towards cancer screening and health check-ups: 1 = negative, 2= positive; Monthly living expenses: 1= ≤￥1200 yuan, 2= >￥1200 yuan; Drinking history: 1 = no, 2 = yes; The way to obtain cancer knowledge: 1= ≤3, 2= ≥4.

Chinese college students with obvious cancer symptom discrimination (β = 0.177, p < 0.001), higher education(β = 2.406, p < 0.001), positive attitudes towards cancer screening (β = 2.806, p < 0.001), higher monthly living expenses (β = 1.503, p = 0.001), more ways to obtain cancer knowledge (β = 0.839, p = 0.048) showed higher cognition levels of cancer warning symptoms. A drinking history (β=–1.371, p = 0.001) was a risk factor for cognition of cancer warning symptoms in Chinese college students. See Table 2 for details.

3.4. Cognition levels of cancer risk factors in Chinese college students

The cognition rates of cancer risk factors were from 46.3% to 91.3%. Chinese college students showed higher cognition levels for the following risk factors: smoking (91.3%), excessive alcohol consumption (90.3%), fried food/barbecue (88.7%), spoiled/expired food (87.9%), decoration materials (85.7%) and pickled food (85.3%). Chinese college students showed lower cognition levels for the following risk factors: obesity (66.2%), excessive exposure to the sun (59.7%), drinking hot water (46.3%). See Figure 2 for details.

Figure 2. The cognition levels of cancer risk factors among Chinese college students (N = 846).

The multiple linear regression analysis showed that the influencing factors for cognition of cancer risk factors among Chinese college students were education, cancer symptom discrimination, psychological distress, attitudes towards cancer screening, life satisfaction, cancer history in relatives and friends, ways to obtain cancer knowledge, smoking history, and nursing history for cancer patients (p < 0.05). The collinearity diagnosis showed that the tolerance and variance expansion factor of 9 independent variables were also close to 1, indicating that there was no collinearity problem among the independent variables. See Table 3 for details.

Table 3. The influencing factors for the cognition levels of cancer risk factors among Chinese college students.

Variable	β	S.E.	Standardized regression coefficient	t	P	VIF
Education	2.106	0.247	0.267	8.523	<0.001	1.063
CSDS score	0.094	0.016	0.183	5.801	<0.001	1.077
GAD-7 score	−0.112	0.022	−0.160	−5.141	<0.001	1.053
Attitudes towards cancer screening	1.243	0.410	0.094	3.032	0.003	1.040
Life satisfaction	1.366	0.427	0.099	3.199	0.001	1.037
History of cancer in relatives and friends	0.603	0.286	0.069	2.107	0.035	1.177
The way to obtain cancer knowledge	0.560	0.225	0.078	2.490	0.013	1.057
Smoking history	−0.709	0.306	−0.071	−2.316	0.021	1.026
Nursing history for cancer patients	0.802	0.388	0.067	2.069	0.039	1.139

Dependent variable:

CRFCQ score: 0–14 points.

Independent variable:

Education: 1 = junior college, 2 = undergraduate, 3 = postgraduate; CSDS score: 0–21 points; GAD-7 score: 0–21 points; Attitudes towards cancer screening and health check-ups: 1 = negative, 2= positive; life satisfaction: 1 = average, 2= higher; History of cancer in relatives and friends: 1 = no, 2 = yes; The way to obtain cancer knowledge: 1= ≤3, 2= ≥4; Smoking history: 1 = no, 2 = yes; Nursing history for cancer patients: 1 = no, 2 = yes.

Chinese college students with a higher education (β = 2.106, p < 0.001), obvious cancer symptom discrimination (β = 0.094, p < 0.001), positive attitudes towards cancer screening and health check-ups (β = 1.243, p = 0.003), higher life satisfaction (β = 1.366, p = 0.001), relatives and/or friends with a history of cancer (β = 0.603, p = 0.035), more ways to obtain cancer knowledge (β = 0.560, p = 0.013), and a history of nursing for cancer patients (β = 0.802, p = 0.039) showed higher cognition levels for cancer risk factors. Chinese college students with obvious psychological distress (β=-0.112, p < 0.001) and a history of smoking (β=-0.709, p = 0.021) showed lower cognition levels for cancer risk factors. See Table 3 for details.

3.5. The mediating effects of cancer cognition and cancer symptom discrimination on psychological distress and cancer fear

The structural equation model with cancer cognition and cancer symptom discrimination as intermediary variables fitted well, and the parameter estimates of each latent variable were statistically significant (P < 0.001). Psychological distress and cancer symptom discrimination had direct impacts on cancer fear. Psychological distress indirectly affects cancer symptom discrimination through cancer cognition. Cancer cognition indirectly affects cancer fear through cancer symptom discrimination. Therefore, cancer cognition and cancer symptom discrimination showed intermediary effects on psychological distress and cancer fear among Chinese college students. See Figure 3, Tables 4 and 5 for details [21,22].

Figure 3. Interactions between cancer cognition, cancer symptom discrimination, psychological distress, and cancer fear in SEM.

Table 4. Fitting degrees of the SEM.

	χ^2/df	GFI	AGFI	RMSEA	CFI	NFI	TLI	RFI	PNFI
Fitting degrees	6.439	0.969	0.935	0.080	0.967	0.961	0.946	0.936	0.584
Threshold	—	>0.90	>0.90	<0.08	>0.90	>0.90	>0.90	>0.90	>0.50

Table 5. Load coefficient estimations of the interactions between cancer cognition, cancer symptom discrimination, psychological distress, and cancer fear in SEM.

Variable		Estimate	S.E.	C.R.	P
Cancer cognition	Psychological distress	−0.188	0.037	−5.100	<0.001
Cancer symptom discrimination	Cancer cognition	0.221	0.030	7.353	<0.001
Cancer fear	Psychological distress	0.656	0.054	12.218	<0.001
Cancer fear	Cancer symptom discrimination	0.466	0.083	5.617	<0.001

4. Discussion

In this study, Chinese college students surveyed had higher levels of cognition on abnormal mass, cervical lesions, repeated hemoptysis or vomiting, liver cirrhosis, unexplained weight loss and pulmonary nodules, considering them as cancer warning symptoms. They also showed higher levels of cognition on smoking, excessive alcohol consumption, fried food/barbecue, spoiled/expired food, decoration materials and pickled food, considering them as cancer risk factors. Among them, the cancer symptom and risk factor with the highest cognition level were ‘abnormal mass’ (84.4%) and ‘smoking’ (91.3%). A study on residents over 40 years old in Malaysia also found that the most recognized cancer symptom and risk factor were ‘unexplained mass or swelling’ (74.5%) and ‘smoking’ (88.7%) [23]. Abnormal mass is a common first symptom of a variety of cancers. We believe that healthy people are more likely to subjectively associate ‘abnormal mass’ with words like ‘tumor’ and ‘cancer’. From 2010 to 2018, the prevalence and intensity of smoking in China showed a downward trend. Active and effective tobacco control measures can significantly reduce the economic burden of lung cancer caused by smoking [24]. The high cognition rate that considering smoking as a cancer risk factor among Chinese college students may be related to China’s active tobacco control measures [25].

In terms of cancer warning symptoms, dyspepsia and fatigue that cannot be relieved after rest received the lowest cognition rates among the surveyed Chinese college students. Dyspepsia is one of the most common complaints of digestive diseases. Persistent dyspepsia may also be an early warning symptom of digestive malignant tumors or precancerous lesions (especially gastric cancer) [26]. However, persistent dyspepsia (47.9%) had the lowest cognition rate as a cancer warning symptom among the surveyed Chinese college students, which may be related to the fact that dyspepsia is common in a variety of benign digestive diseases and easy to be ignored. Fatigue is a persistent symptom associated with the development of various stages of cancer [27]. However, fatigue occurs not only in daily life, work, and learning [28,29], but also in chronic diseases such as diabetes [30], chronic obstructive pulmonary disease [31], stroke [32], autoimmune diseases [33], inflammatory bowel diseases [34], chronic kidney disease and in the process of blood purification treatment [35,36]. This may also be the reason for the low cognition level of considering fatigue as a cancer early warning symptom (51.5%). Therefore, the symptoms common in both benign diseases and cancer may be one of the causes for healthy people to fail to consider them as cancer warning symptoms, and may also be one of the causes of delayed diagnosis and treatment among cancer patients. These warning symptoms should become the focuses of cancer health education for healthy people and the focuses of clinical cancer screening [37].

In terms of cancer risk factor cognition, the most easily overlooked risk factors are the habit of drinking hot water and exposure to the sun. In some areas of China, tea, soup, and porridge are often heated to very high temperatures, and high-temperature cooking methods are also popular. Studies in China and other countries have found that the habits of drinking hot water and eating hot food led to a significant increase in the risk of esophageal cancer [38], especially the increased risk of esophageal squamous cell carcinoma [39], and the risk is positively correlated with the increase of temperatures [40]. This may be related to the stem cell division induced by thermal injury of esophageal tissue and the accumulation of the divided cells in the esophagus, which increase the risk of esophageal cancer [41]. Health education activities based on the Theory of Planned Behavior (TPB) can significantly improve female students’ perceived behavioral control and willingness to refuse high-temperature tea, and the consumption of high-temperature tea can be significantly reduced [42]. Frequent exposure to solar ultraviolet radiation is an important inducement of non-melanoma skin cancer [43]. Chinese college students are a group with a high sunburn rate and high risk of solar ultraviolet exposure. However, we found that more than 40% of Chinese college students do not know that exposure to the sun is a risk factor for cancer, indicating that the knowledge and cognition of skin cancer and sunscreen among Chinese college students need to be improved [44].

This study found that students with higher education, more ways to obtain cancer knowledge and positive attitudes towards cancer screening and health check-ups had higher levels of cognition on cancer warning symptoms and risk factors. Students with a history of interactions with cancer patients (such as knowing a cancer patient among their relatives and friends, or those who had taken care of a cancer patient) also had good cognition on cancer risk factors, suggesting that positive attitudes towards cancer may be positively related to their levels of education. Meanwhile, those who were familiar with cancer symptoms or signs, those who knew a cancer patient, and those who paid attention to cancer risk factors were more likely to respond positively to cancer related information [45].

We found that students with a history of alcohol consumption had a lower level of cognition on cancer warning symptoms, and students with a history of smoking had a lower level of cognition on cancer risk factors, suggesting that people with bad health behaviors such as smoking and alcohol abuse should be the key population of cancer health education, and active cancer health education may help to improve bad health behaviors. In addition, students with higher monthly living expenses and high life satisfaction had relatively high cognition on cancer warning symptoms and risk factors, suggesting that cancer health education should also focus on groups whose life and family economic background are underprivileged [46].

Our previous research found that some Chinese college students have discrimination against and fear of cancer [47]. Another research by our team found that cancer symptom discrimination (CSDS score) and psychological distress (CAD-7 score) were influencing factors for cancer fear (FOCS score), but we did not elaborate further on the interaction of the three at that time [16]. From previous studies, we hypothesized that cancer cognition in healthy people may play a moderating role in cancer-related negative psychological events. Therefore, in this study we further analyzed the interrelation and effects of cancer cognition and cancer symptom discrimination on psychological distress and cancer fear. In this study, multiple linear regression models showed that cognition of cancer warning symptoms and risk factors was positively correlated with cancer symptom discrimination among Chinese college students, but not directly related to cancer fear; cognition of cancer risk factors is negatively correlated with psychological distress among Chinese college students. Based on the above conclusions, we hypothesized that cancer fear has a direct effect on psychological distress and cancer symptom discrimination, and cancer cognition also has a direct effect on psychological distress and cancer symptom discrimination, while there may be an indirect effect between cancer fear and cancer cognition. So, we constructed a structural equation model to verify our hypothesis in this study.

The structural equation modelling (SEM) confirmed that students’ psychological distress can directly lead to fear of cancer, and can also lead to fear of cancer through the intermediary effects of cancer cognition and cancer symptom discrimination. Therefore, there is an indirect relationship between cancer cognition and cancer fear. Through the mediation of cancer symptom discrimination, cancer cognition may lead to cancer fear among Chinese college students. We believe that higher levels of cancer awareness and cognition in healthy people may result in more prejudice or discrimination against cancer symptoms. Therefore, the purpose of public cancer health education is not only to teach and popularize cancer health knowledge, but also, to instruct the public how to treat cancer and cancer patients rationally, objectively, and correctly, to reduce the negative psychology and behaviors, such as discrimination, psychological distress, and fear of cancer, in healthy people.

Health education interventions may be helpful to improve cancer health literacy, which has been confirmed by many studies. Besides, this study found that health education can play a positive and effective role in improving teachers’ knowledge level, attitude and practical ability of breast cancer and cervical cancer, and different health education models can play different roles [48–50]. Even short-term cancer health education can have a positive effect [51]. The ‘knowledge-attitude/belief-practice’ theory in behavior science also suggests that to promote cancer prevention and health behavior in the public, it is important to improve cancer cognition in the public by providing cancer cognition knowledge, and innovating ways of passing cancer-related information and knowledge, such as using accessible channels, initiating interesting subjects and producing understandable contents. This way, the public’s attitude and behavior can be changed timely and effectively [52].

For non-medical students in colleges and universities, we suggest that Education on diseases and health should be taken as a compulsory course and a general course, so that college students can have a preliminary understanding of common diseases, common causes, clinical manifestations, prevention and health care and other knowledge. In addition, we suggest that education on cancer health should be developed in rich and various forms (such as making short videos, establishing WeChat official accounts on this issue) in both primary schools, secondary schools, and universities. Universities should set up certain compulsory or optional courses in health education for students in all majors. These courses would enhance their overall awareness of cancer, correct their mistaken ideas, and negative attitudes, and reduce their fear, discrimination, or psychological distress [53–55].

Before the study, we thought that medical students should have higher levels of cancer cognition than non-medical students did. However, after the analysis of Kruskal-Wallis Test, we found that there was no statistically significant difference between medical students and non-medical students in the cognition levels of cancer warning symptoms (p = 0.462) and cancer risk factors (p = 0.289), then the independent variable of ‘major’ was not included in the multiple linear regression equation. We believed that there may be two reasons for this phenomenon. First, medical students in this study included clinical medicine, public health, nursing, pharmacology, forensic science, laboratory medicine, basic medicine, and other related majors. Some majors may have a relatively low level of cancer cognition for they don’t deal with cancer or cancer patients in their practices. Second, in the early stage of medical higher education, colleges/universities in some countries and regions (such as medical colleges/universities in China) mainly carried out basic medical courses, which may lead to the lack of understanding of common and frequent diseases among junior medical students, resulting in the low level of disease and health literacy of students. Therefore, for medical students, it is also recommended to carry out regular courses related to education on diseases and health and humanities medicine at an early stage of their university life. This will help medical students focus on cultivating and improving their disease and health literacy before receiving standardized clinical medicine education. It also helps to fully mobilize students’ desire to explore and seek professional clinical knowledge, and plays a role in linking the past and the future between basic medical education and clinical medical education.

5. Limitation

The study also has some limitations. First, the survey was conducted during the period of the novel coronavirus epidemic. In order to cooperate with local epidemic control policies, we could not conduct offline surveys or random sampling studies, which may lead to systematic non-sampling error caused by sample bias. The advantages of online questionnaires include: the process of questionnaire delivery and recovery is simple and fast, shortening the data collection process. It also saves paper, reduces financial investment and labor costs, helps to obtain a larger sample size, effectively avoids human data entry errors, and significantly increases the efficiency of constructing databases. The limitations of online questionnaires, on the other hand, include: lack of face-to-face communication and interaction between the researchers and the respondents; relatively low recovery and validity rates; and possible compromises in respondent honesty.

In the present study sample, Cronbach’s coefficients of CWSCQ, CRFCQ, FOCS, CSDS and GAD-7 were examined to confirm the reliability of the above questionnaires/scales. It was generally accepted that the acceptable range of Cronbach’s coefficients is between 0.75 and 0.95, and that high Cronbach’s coefficients may indicate that some items may be redundant. Therefore, there was a need to be aware of the negative effects of high Cronbach’s coefficients [56]. In addition, there are many warning symptoms and risk factors for cancer, it is not possible for us to list all the warning symptoms and risk factors in the questionnaire. CWSCQ and CRFCQ are compiled based on the most common, evidence-based, known warning symptoms and risk factors, using them to judge the basic cognition level of respondents. So, it is not possible to evaluate the cancer knowledge level of the respondents comprehensively and completely. We will continue to update and modify the items in CWSCQ and CRFCQ in our subsequent studies, and continue to study the current status and influencing factors of cancer cognition among different groups (such as teachers, industrial workers, community residents, etc.), as well as the interaction between cancer cognition and psychological and behavioral factors.

6. Conclusion

With the increasing morbidity and mortality rates of cancer, it is necessary to raise the awareness of early detection, and early treatment of cancer in the general public. Chinese college students are a group with higher education, strong information acquisition abilities and a wide range of knowledge. However, the overall cancer cognition situation of Chinese college students is not optimistic, which highlights the necessity of improving the cancer health literacy among Chinese college students and even other groups. School-based health education can improve students’ cognition on diseases and promote students’ healthy behaviors [57,58].

Authors contributions

LF and ZD contributed to the writing and statistical analyses of this article, including putting forward this study and carrying out the study. QL, QY, YZ, FY, QW and WZ contributed to performing the investigations and collecting all data. They had contributed equally to the manuscript. XW and RY contributed to performing the investigations and collecting partial data.

Competing interests

The authors declare that they have no competing interests.

Ethics approval and consent to participate

This was an exploratory, cross-sectional study conducted in China which was approved by The Ethics Committee of The Sixth Affiliated Hospital of Kunming Medical University (Grant number: 2020kmykdx6h03). The study was conducted in accordance with the Declaration of Helsinki. All methods were performed in accordance with the relevant guidelines and regulations.

Informed consents were obtained from all participants. Participants read an informed consent on the front page of the questionnaire. The participants were required to confirm their participation or declare their refusal, and to verify whether they met the inclusion criteria. The body of the questionnaire would be loaded after they confirmation; Otherwise, the program would exit automatically.

Acknowledgments

We would like to extend our heartfelt gratitude to all interviewees for their participation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data used and analyzed in the current study are available from the corresponding author on reasonable request.

Additional information

Funding

This study was supported by a special program for Yuxi City approved by National Summer Social Practice Activities for Chinese College Students, Program of Philosophy and Social Science Foundation of Yunnan Province (Grant number: QN202214), Yunnan Provincial Department of Education Research Fund (Grant number: 2022J0140), Teaching Research and Reform Foundation of Kunming Medical University (Grant number: 2022-JY-Y-151), and National Training Program of Innovation and Entrepreneurship for Undergraduates (Grant number: 202210678007).