Cancer beliefs, risk perceptions, and health protective behaviors among people living with diabetes in the United States: Results from the Health Information National Trends Survey

Abstract

Objective

Diabetes and cancer share common preventable risk factors, including obesity, physical inactivity, and poor diet. People with diabetes have an increased risk of developing cancer. The objective of our study was to compare cancer risk perceptions, beliefs about cancer, and cancer protective behaviors among people with and without diabetes in the United States.

Methods

We utilized data from the cross-sectional nationally representative Health Information National Trends Survey (HINTS) 6 (2022). Diabetes and cancer history were self-reported. We compared beliefs about cancer and cancer protective behaviors between individuals with and without diabetes who reported no cancer history.

Results

The sample included 3937 individuals. There was no significant difference in beliefs about cancer (worry, risk, preventability, fatality, or prevention progress) and cancer protective behaviors (smoking, drinking, physical activity, and cancer information seeking) between individuals with diabetes and those without diabetes after adjustment.

Conclusions

Our findings highlight the need for further research to enhance understanding and improve awareness of cancer risk and protective behaviors among people living with diabetes.

Highlights

• •Individuals with diabetes have similar beliefs about cancer than those without.
• •Individuals with and without diabetes engage in similar cancer preventive behaviors.
• •Diabetes alone may not drive differences in cancer beliefs and prevention behaviors.

1. Introduction

Diabetes and cancer are among the most common chronic diseases in the United States, affecting millions of individuals. The National Diabetes Statistics Report indicates that 38.4 million people (11.6 % of the United States population) were living with diabetes in 2021 (CDC, 2024a). Additionally, over 1.7 million new cases of cancer were reported in 2019, and as of January 2022, 18.1 million people living in the United States were cancer survivors (Tonorezos et al., 2024). Diabetes and cancer are complex conditions caused by similar risk factors, including genetic, environmental, and lifestyle factors (Wang et al., 2021). The concomitant presence of these risk factors further exacerbates the risk of developing both diseases (Zhang et al., 2020). For example, lifestyle factors such as lack of adequate physical activity and poor dietary habits increase the risk of developing obesity, a leading risk factor for developing diabetes and cancer (Avgerinos et al., 2019). Diabetes and cancer also disproportionately affect older individuals and those with low socioeconomic status because these groups have a higher exposure to the lifestyle risk factors that increase the risk of developing the conditions (Myers et al., 2017).

Diabetes is primarily categorized as type 1 diabetes (T1D), type 2 diabetes (T2D), and gestational diabetes, of which T2D is the most common, accounting for 90–95 % of all diabetes cases (CDC, 2024a). Diabetes management includes medications and lifestyle changes focusing on diet and weight management. Research shows that individuals living with diabetes have a significantly higher risk of living with co-occurring chronic conditions, including obesity, hypertension, cardiovascular disease, and arthritis (Almourani et al., 2019). Research also shows that individuals with diabetes have a higher risk of being diagnosed with multiple types of cancer, including liver, pancreatic, breast, and colorectal cancers (Ling et al., 2021). Hyperinsulinemia, chronic inflammation, and altered glucose metabolism are proposed mechanisms linking diabetes to cancer development (Zhu and Qu, 2022).

Perception of cancer risk, an individual's assessment of their likelihood of developing cancer, is influenced by factors such as family history, personal experiences such as smoking, exposure to media, and knowledge about cancer (Antwi et al., 2025). Constructs outlined by the Health Belief Model (HBM) (Janz and Becker, 1984), perceived susceptibility, perceived severity, perceived benefits and barriers, cues to action, and self-efficacy, predict whether individuals adopt cancer protective behaviors such as participating in cancer screening (Otto et al., 2021). Individuals who perceive that their cancer risk is low may be less likely to engage in cancer-protective behaviors (physical activity, healthy eating, and participating in cancer screening) (Noman et al., 2021). Individuals who understand factors or seek information associated with cancer prevention are more likely to engage in cancer health-protective behavior (adhering to dietary, smoking cessation, and cancer screening guidelines) (Tilburt et al., 2011; Swoboda et al., 2021). Individuals living with diabetes face unique barriers that preclude them from engaging in health-protective behaviors (Alexandre et al., 2021). Tailored interventions have been shown to improve cancer risk perception and knowledge (Calderón-Mora et al., 2020). Therefore, understanding the perception of cancer risk among individuals living with diabetes may inform effective public health interventions and communication strategies aimed at promoting cancer health protective behaviors among people living with diabetes.

The present study aims to compare cancer risk perceptions (a person's subjective belief about their likelihood of developing cancer (Brewer et al., 2004)) and beliefs about cancer (convictions about what is true about cancer (e.g., causes, symptoms, consequences, etc. (Lykins et al., 2008)) among people with and without diabetes. We also investigate the cancer protective behaviors of people with and without diabetes.

2. Methods

2.1. Data source

Our study used cross-sectional data from the Health Information National Trends Survey (HINTS) that was collected from March 7, 2022, to November 8, 2022, HINTS 6 (2022) (Learn More About HINTS | HINTS, 2024). HINTS is a nationally representative survey implemented by the National Cancer Institute since 2003 that recruits non-institutionalized individuals in the United States aged 18 years and older (HINTS_6_MethodologyReport.pdf, 2024). HINTS gathers population-wide cancer-related information seeking, including accessibility and utilization of health-related information (Hesse et al., 2006). HINTS 62022's overall household response rate was 28.1 % (Learn More About HINTS | HINTS, 2024). The University of Florida's Institutional Review Board deemed that our secondary analysis study was exempt from review and informed consent requirements as HINTS data is completely deidentified and publicly available.

2.2. Study population

HINTS 62022 included a sample of 6252 respondents. Participants who self-reported cancer history (n = 900) and answered yes to “Have you ever been diagnosed as having cancer?” or “I already had cancer” under Section P: Beliefs About Cancer in HINTS were excluded from the study (n = 6). From the remaining (n = 5342) participants, we then examined missing data, errors, or unreadable or nonconforming Numeric Response of “NA” on cancer perception variables. After applying listwise deletion for the missing data (n = 1409), a sample size of (n = 3937) participants was obtained. We identified (n = 725) people with diabetes, and (n = 3212) participants did not self-report with diabetes. See flowchart (Fig. 1).

Fig. 1.

Participant inclusion flowchart (CDC, 2024a). HINTS = Health Information National Trends Survey.

2.3. Measures

Fig. 2 summarizes all variables used in the study informed by the HBM (Champion and Skinner, n.d.).

Fig. 2.

Application of the Health Belief Model as a conceptual framework to examine perceptions of cancer risk and prevention.

Demographic characteristics included in HINTS and associated with the risk of diabetes or cancer in the United States include age (18–34, 35–49, 50–64, 65–74, 75+), sex (male/female), marital status (married, single, divorced), educational level (less than high school, high school graduate, some college, and college graduate or more), annual income level (less than $20,000, $20,000 to < $35,000, $35,000 to < $50,000, $50,000 to < $75,000, and $75,000 or More), race and ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other which includes non-Hispanic Asian and other) and insurance status (Yes/No) (Wang et al., 2021).

Health status variables considered were body mass index (BMI). BMI was classified into four groups according to the Centers for Disease Control and Prevention guidance (CDC, 2024b): underweight (<18.5), healthy weight (18.5–24.9), overweight (25.0–29.9), and obesity (≥ 30.0) (CDC, 2024b). Diabetes status, “Has a doctor or health professional ever told you that you had any of the following medical conditions: Diabetes or high blood sugar? (Yes/No).

For the psychosocial variables, we considered participants' general health (excellent, very good, good, fair, and poor), and their family cancer history status (Yes vs. No/Not sure).

Perceived susceptibility variables capture an individual's subjective perception of their risk of developing cancer. Cancer Worry: participants indicated agreement on a five-point Likert scale for the question, “How worried are you about getting cancer?” from “Not at all” to “Extremely.” For ease of interpretation, we grouped responses into Not at all / Slightly, Somewhat/ Moderately, Extremely. Cancer risk perception was defined using a person's perception of their chances of getting cancer based on responses to the question, “Compared to other people your age, how likely do you think you are to get cancer in your lifetime?” were dichotomized into likely (Likely /Very likely) and unlikely (Unlikely/ Neither likely nor unlikely/ Very unlikely/ I don't know).

The perceived severity variable reflects an individual's perception of the seriousness of developing cancer. We used the question about cancer fatality, “When I think of cancer, I automatically think about death,” which was dichotomized into agree (strongly agree/ Somewhat agree) and disagree (strongly disagree/ Somewhat disagree).

The perceived benefits and barriers variable assessed how individuals view the effectiveness of recommendations to reduce cancer risk or the obstacles that may increase its threat. Our study used the question, “There are too many recommendations about preventing cancer, it's hard to know which ones to follow.” Responses were dichotomized into agree (strongly agree/ Somewhat agree) and disagree (strongly disagree/ Somewhat disagree). We also examined the beliefs about the scientific progress in cancer prevention among individuals with diabetes. The question, “How much progress has been made in preventing cancer?” (was recorded as “at least some progress” or “little, no progress, or Don't know “) indicates whether an individual believes that at least some progress has been made towards cancer prevention.

Cues to action represent a trigger (internal or external) that prompts an individual to engage in a health-promoting behavior or recommended health action. External cues include information from others and the media. In this case, we used an individual's response to the question, “It seems everything causes cancer.” Responses were dichotomized into agree (strongly agree/ Somewhat agree) and disagree (strongly disagree/ Somewhat disagree).

Self-efficacy reflects an individual's confidence in their ability to perform the recommended behavior successfully. Participant responses to the question “How much do you agree or disagree: There's not much you can do to lower your chances of getting cancer…” were dichotomized into agree (strongly agree/ Somewhat agree) and disagree (strongly disagree/ Somewhat disagree).

For health behaviors, we used variables related to smoking status, drinking level, physical activity, and seeking cancer information (Yes/No). Smoking status was dichotomized into yes (current & former) and no (never). Drinking level was classified as “No”, “Moderate drinking”, and “Heavy drinking” according to the definition provided by the Centers for Disease Control and Prevention (CDC, 2025). “No” means no consumption of drinks in a week. “Moderate drinking” means when consuming less than two drinks for men and one drink for women in a week. Heavy drinking was defined as consuming eight or more drinks per week for women or 15 or more drinks per week for men (CDC, 2025). Physical activity was derived from questions “In a typical week, how many days do you do any physical activity or exercise of at least moderate intensity, such as brisk walking, bicycling at a regular pace, and swimming at a regular pace?” and “On the days that you do any physical activity or exercise of at least moderate intensity, how long do you typically do these activities?” By multiplying the two questions, we derived the time to calculate moderate physical activities per week and then dichotomized into yes (at least 150 min) and no (less than 150 min), which indicates whether the participant meets the 2018 Physical Activity Guidelines for Americans (Piercy et al., 2018). The number of cancer prevention behaviors is defined by summing four health prevention behaviors: never smoking, not engaging in heavy drinking, meeting moderate-intensity physical activity guidelines, and seeking cancer-related health information.

2.4. Statistical analysis

We summarized demographic characteristics using descriptive statistics and bivariate analysis by diabetes status. We utilized multiple survey design-weighted logistic regression models to investigate whether having diabetes influences cancer beliefs or cancer protective behaviors. Two multivariable logistic regressions were conducted; the first examined the relationship between having diabetes and cancer beliefs, and the second examined the relationship between having diabetes and cancer protective behaviors. All regression models controlled for sociodemographic and health characteristics. Each cancer belief model was adjusted for age, race and ethnicity, educational level, marital status, annual income level, insurance status, BMI, and general health. These covariates were significant in the univariate analysis of baseline characteristics, examining differences between those with and without diabetes. These factors have also been reported in the literature to be significantly associated with diabetes (Zhang et al., 2020). Survey weights were applied to all analyses in our study using the jackknife replicate weights method for variance estimation (HINTS_6_MethodologyReport.pdf, 2024). We compared the characteristic distribution using the Wald chi-square tests in complex survey samples. We used R Studio software (R version 4.4.1) to conduct our analyses, and p < 0.05 was considered statistically significant.

3. Results

Table 1 summarizes the demographic characteristics of the study sample by diabetes status. Bivariate analysis comparing participants without diabetes history (85.2 %) to those with diabetes history (14.8 %) revealed statistically significant differences by age, race and ethnicity, education level, annual income level, marital status, insurance status, BMI, and general health.

Table 1.

Descriptive characteristics of the sample of individuals living with and without a diabetes diagnosis in the United States, from March 7, 2022, to November 8, 2022.

	No Diabetes, % (95 % CI1)		Diabetes, % (95 % CI)		p-Value2
	n = 3212	Weighted % 85.2	n = 725	Weighted % 14.8
Age					0.00
18–34	711	31.9 (29.2, 34.8)	31	9.5 (5.1, 17.0)
35–49	859	30.1 (27.5, 32.7)	105	21.4 (16.5, 27.4)
50–64	880	25.2 (23.1, 27.4)	274	38.0 (31.9, 44.6)
65–74	519	8.7 (7.9, 9.5)	222	21.1 (17.5, 25.2)
75+	243	4.2 (3.5, 4.9)	93	10.0 (7.7, 12.7)
Sex					0.5
Male	1286	49.7 (47.7, 51.7)	307	51.9 (45.9, 57.9)
Female	1926	50.3 (48.3, 52.3)	418	48.1 (42.1, 54.1)
Race and ethnicity					0.00
Non-Hispanic White	1908	63.4 (60.9, 65.8)	319	53.6 (49.0, 58.1)
Non-Hispanic Black	444	9.4 (8.4, 10.6)	168	17.7 (13.6, 22.8)
Hispanic	555	16.5 (15.0, 18.0)	171	19.4 (15.1, 24.4)
Other3	305	10.7 (8.7, 13.2)	67	9.4 (6.8, 12.9)
Education level					0.00
Less than high school	153	5.1 (3.8, 6.7)	48	8.3 (3.1, 20.1)
High School Graduate	508	19.6 (17.4, 22.1)	150	26.8 (21.5, 32.8)
Some college	870	38.7 (36.7, 40.7)	228	40.3 (34.6, 46.3)
College Graduate or More	1681	36.6 (35.0, 38.3)	299	24.6 (20.0, 29.9)
Annual income level					0.00
<$20,000	419	11.7 (9.4, 14.5)	151	18.5 (14.7, 22.9)
$20,000 to < $35,000	380	9.3 (8.2, 10.6)	101	14.2 (9.0, 21.7)
$35,000 to < $50,000	392	10.9 (9.0, 13.1)	126	16.1 (12.0, 21.2)
$50,000 to < $75,000	568	17.7 (15.7, 20.0)	120	17.9 (13.6, 23.2)
≥ $75,000 or more	1453	50.4 (47.7, 53.1)	227	33.4 (27.4, 39.9)
Marital status					0.02
Married	1495	50.3 (47.8, 52.8)	334	54.8 (48.5, 61.0)
Single4	1271	44.0 (41.6, 46.5)	251	37.0 (30.8, 43.7)
Divorced	446	5.7 (5.0, 6.5)	140	8.2 (6.2, 10.6)
Insurance covered5	2907	88.2 (86.9, 89.4)	687	95.0 (91.8, 97.0)	0.00
Body Mass Index					0.00
Underweight (<18.5)	42	1.4 (0.9, 2.1)	3	2.6 (0, 86.0)
Healthy Weight (18.5–24.9)	1038	32.8 (30.0, 35.7)	105	13.2 (9.9, 17.5)
Overweight (25.0–29.9)	1088	33.6 (30.6, 36.8)	215	29.9 (24.3, 36.2)
Obesity (≥ 30.0)	1044	32.2 (29.6, 34.9)	402	54.2 (47.3, 61.0)
With a family cancer history	2184	63.7 (60.7, 66.7)	516	67.3 % (61.1, 72.9)	0.28
General health					0.00
Excellent	404	13.7 (11.6, 16.0)	24	3.7 (1.7, 7.6)
Very good	1290	40.1 (37.2, 42.9)	148	19.5 (13.7, 26.9)
Good	1152	34.6 (31.9, 37.4)	320	44.1 (37.1, 51.3)
Fair	319	10.3 (8.4, 12.6)	194	27.1 (22.3, 32.5)
Poor	47	1.3 (0.9, 2.0)	39	5.7 (3.3, 9.7)

Note: p < 0.05 indicates statistical significance.

¹CI = Confidence Interval.

²Chi-squared test adjusted Wald test.

³Other races and ethnicities include non-Hispanic Asian and non-Hispanic others.

⁴The marital status of a single includes the sum of living as married or living with a romantic partner, widowed, separated, and single, never been married.

⁵Insurance coverage included any health insurance or health care plan: employer-sponsored insurance, prepaid plans, or government plans such as Medicare, Medicaid, and TRICARE.

Table 2 summarizes the multivariable logistic regression results comparing individuals who reported having diabetes to those without diabetes on their cancer beliefs. Each cancer belief adjusted the multivariable logistic regression model controlled for age, race and ethnicity, educational level, annual income level, marital status, insurance status, BMI, and general health. People with diabetes were more likely to think there's not much you can do to lower your chances of getting cancer in the unadjusted model (Diabetes vs. not having diabetes, 34.8 % vs. 27.2 %, OR 1.43 [95 % CI, 1.03, 2.01]); however, after controlling for covariates, the model was no longer significant. We did not observe any significant difference in cancer beliefs, including cancer worry, risk of getting cancer, risk factors of cancer, lowering the chance of getting cancer, recommendation of preventing cancer, cancer fatality, and progress made to prevent cancer between individuals with and without a diabetes history.

Table 2.

Association between cancer beliefs and other factors among of the sample of individuals living with and without a diabetes diagnosis in the United States, from March 7, 2022, to November 8, 2022.

	Diabetes History, % (95 % CI)		Odds Ratio (95 % CI)
Cancer Beliefs	No (reference)2, n = 3,2124, Weighted % = 85.2	Yes, n = 7254, Weighted % = 14.8	Unadjusted	Adjusted1
Worried about getting cancer
Not at all / Slightly (reference)	52.6 % (49.9, 55.5)	47.8 (41.7, 54.0)
Somewhat / Moderately	41.6 (38.8, 44.4)	45.7 (39.9, 44.4)	1.21 (0.91, 1.61)	1.20 (0.85, 1.70)
Extremely	5.7 (4.6, 7.1)	6.5 (4.5, 9.2)	1.24 (0.77, 2.01)	1.07 (0.65, 1.77)
How likely do you think you are to get cancer in your lifetime
	23.1 (20.9, 25.4)	20.8 (16.9, 25.5)	0.88 (0.63, 1.21)	0.85 (0.58, 1.26)
It seems like everything causes cancer…
	68.3 (65.7, 70.7)	67.1 (59.5, 73.8)	0.94 (0.66, 1.36)	1.09 (0.70, 1.69)
*There's not much you can do to lower your chances of getting cancer3
	*27.2 (24.3, 30.4)	*34.8 (28.8, 41.4)	1.43 (1.02, 2.01)	1.24 (0.83, 1.86)
There are so many different recommendations for preventing cancer, it's hard to know which ones to follow
	68.8 (66.6, 71.0)	69.7 (63.3, 75.4)	1.04 (0.76, 1.44)	0.99 (0.66, 1.47)
When I think about cancer, I automatically think about death
	58.6 (55.2, 61.9)	61.5 (54.1, 68.5)	1.13 (0.82, 1.57)	1.28 (0.89, 1.83)
How much progress has been made in preventing cancer?
	60.9 (57.9, 63.8)	57.0 (50.1, 63.7)	0.85 (0.64, 1.14)	0.94 (0.64, 1.39)

¹All Multivariable logistic regressions were controlled for age, race and ethnicity, educational level, marital status, annual income level, insurance status, BMI, and general health.

²Presented with no diabetes history as the reference group.

³* means statistically significant.

⁴n means unweighted sample size.

Fig. 3 summarizes the frequency and univariate analysis results of cancer protective behaviors among the sample with diabetes and without diabetes (p-values presented at appendix table 2). People with diabetes had a higher rate of smoking history (p = 0.04), drank less (p = 0.02), and a lower percentage of engaging in moderate-intensity physical activity (p = 0.00) than people without diabetes. People with diabetes had a lower rate of engaging in three or more cancer protective behaviors (p = 0.02).

Fig. 3.

Frequency of cancer Protective Behaviors among the sample with diabetes compared to those without diabetes in the United States, from March 7, 2022, to November 8, 2022. The number of cancer prevention behaviors is defined by summing four health prevention behaviors: never smoking, not engaging in heavy drinking, meeting moderate-intensity physical activity guidelines, and seeking cancer-related health information. * Means statistically significant.

Table 3 summarizes the multivariable logistic regression results investigating the association between each cancer protective behavior among the sample with and without a diabetes history. The model controlled for cancer beliefs (cancer worry, risk of getting cancer, risk factors of cancer, lowing chance of getting cancer, recommendation of preventing cancer, cancer fatality, and progress made to prevent cancer) and age, race and ethnicity, educational level, annual income level, marital status, insurance status, BMI, and general health. The unadjusted models showed that those with diabetes had significantly higher odds of reporting being current or former smoker (OR 1.35 [95 % CI, 1.02, 1.79]), and lower odds of reporting moderate drinking (OR 0.58 [95 % CI, 0.41, 0.80]) or heavy drinking (OR 0.44 [95 % CI, 0.26, 0.76]), and lower odds of engaging in moderate intensity physical activity (OR 0.47 [95 % CI, 0.36, 0.62]). People with diabetes had higher odds of engaging in two or more cancer protective behaviors (OR 1.90 [95 % CI, 1.30, 2.78]) and three (OR 1.80 [95 % CI, 1.28, 2.52]). We did not observe any significant differences between those with and without diabetes in the adjusted multivariable models, adjusting for cancer beliefs and cancer protective behaviors.

Table 3.

Association of cancer protective behaviors of the sample of individuals living with and without a diabetes diagnosis in the United States, from March 7, 2022, to November 8, 2022.

	Univariate Model	Adjusted Multivariable Model1
	Odds Ratio (95 % CI)	Odds Ratio (95 % CI)
Smoking history
Never	1.002	1.00
Current & Former smoker	1.35 (1.02, 1.79)	0.97 (0.68, 1.38)
Drinking status
No	1.00	1.00
Moderate drinking	0.58 (0.41, 0.80)	0.79 (0.56, 1.11)
Heavy drinking	0.44 (0.26, 0.76)	0.66 (0.37, 1.18)
At least 150-min weekly moderate-intensity physical activity (No vs. Yes)
	0.47 (0.36, 0.62)	0.80 (0.55, 1.15)
Have looked for information about cancer from any source (No vs. Yes)
	0.81 (0.61, 1.09)	0.94 (0.68, 1.31)
Number of cancer protective behaviors
0	1.00	1.00
1	0.97 (0.21, 4.55)	1.10 (0.20, 5.95)
2	1.90 (1.30, 2.78)	1.50 (0.29, 7.82)
3+	1.80 (1.28, 2.52)	1.18 (0.22, 6.31)

¹All Multivariable logistic regressions were controlled for cancer beliefs (cancer worry, risk of getting cancer, risk factors of cancer, lowing chance of getting cancer, recommendation of preventing cancer, cancer fatality, and progress made to prevent cancer) and age, race and ethnicity, educational level, marital status, annual income level, insurance status, BMI, and general health.

²1.00 is the reference value.

4. Discussion

Our analysis of a nationally representative sample found that individuals with diabetes have similar beliefs, risk perceptions about cancer, and cancer protective behaviors as those without diabetes, despite having a higher cancer risk. These findings highlight a critical gap between actual and perceived cancer risk among people living with diabetes. “Teachable moment” in cancer care is the time frame after the occurrence of a health-related event during which individuals are highly amenable to modifications in their lifestyle.⁴² Detection of diabetes might be one of those teaching moments during which individuals might be sensitized to various diabetes-related issues, including cancer prevention. It is essential to highlight the preventability of cancers by early screening and its effectiveness in reducing cancer mortality among people living with diabetes.

Investigators have reported that individuals with high-risk perceptions are more likely to engage in cancer protective behaviors (Noman et al., 2021). Early detection and prompt treatment can contribute significantly to reducing cancer mortality. Cancer screening initiatives aid in detecting malignancy at an early stage, when it is amenable to treatment modalities, which can decrease cancer-related deaths (Turner et al., 2021). One of the most central factors for successfully implementing screening programs is program uptake by high-risk individuals (Patel et al., 2012). Lin and co-authors, however, found that although current smokers were more likely to be recommended for lung cancer screening, they had significantly lower interest in participating in the screening compared to non-smokers (Lin et al., 2024).

Risk perception is one of the important factors that could influence the willingness of high-risk individuals to participate in screening programs (Raz et al., 2019). The concept of risk salience, where heightened awareness of a health threat fosters proactive health behaviors, could also be utilized to improve engagement in cancer protective behaviors (Raz et al., 2019). Targeted health communication strategies could leverage cancer risk awareness among those individuals with diabetes as a tool for promoting cancer-proactive health behaviors. For example, healthcare providers may be able to leverage improved awareness about heightened cancer risk among diabetes patients to motivate them to engage in cancer health protective behaviors such as cancer screening (Miller and Pinsky, 2022). There is scarce research examining cancer risk perception among individuals living with diabetes. However, more studies have explored the relationship between diabetes and cancer screening (Miller and Pinsky, 2022; Perrodin-Njoku et al., 2024). Some findings indicate that individuals with diabetes are more adherent to cervical cancer screening (Miller and Pinsky, 2022), whereas other studies show that individuals with diabetes are less likely to receive cancer screening (Perrodin-Njoku et al., 2024). This is concerning as individuals with diabetes have higher cancer mortality compared to those without diabetes (Barone, 2008).

Ezeani and colleagues found that people with overweight or obesity who believed that there were too many recommendations on how to prevent cancer were significantly less likely to meet physical activity recommendations (Ezeani et al., 2023). BMI was not a significant factor included in our multivariable model; this may be because diabetes and obesity are typically co-occurring conditions. Family history of cancer, which has been shown to increase an individual's engagement with cancer prevention (Merten et al., 2022), was also not a statistically significant covariate.

This study has some limitations. First, HINTS data were collected from a cross-sectional sample, limiting the ability to examine trends over time or assure causal inference. We only utilized a single cycle from HINTS 6, with access to both diabetes and cancer information. HINTS relies on self-reporting conditions, and HINTS data did not specify the type of diabetes participants reported (i.e., T1D, T2D, or gestational diabetes). Thus, utilizing such self-reported survey data may introduce recall bias in reporting factors such as heavy drinking or family cancer history. The response rate of HINTS 6 (2022) was 28.1 %, which may result in sampling bias. The belief about cancer survey questions collected in HINTS 6 were not targeted to specific cancer types. HINTS also only collected data on cervical cancer screening and not other cancer locations, therefore limiting our ability to examine screening differences between those with and without diabetes.

Despite limitations, this study has several strengths. Our study sample was nationally representative, ensuring generalizability of findings to the U.S. adult population. The large sample size enhances statistical power and reliability. Our study is the first to investigate how individuals with diabetes and individuals without diabetes differ in their cancer beliefs and lays the foundation for future studies to further explore cancer risk perceptions of individuals with chronic disease who have a heightened risk of developing cancer. This study also highlights potential gaps in cancer awareness that could inform policy and educational programs. Further, our findings highlight the need to improve cancer awareness among populations with diabetes.

5. Conclusion

Despite having a higher risk of developing cancer and worse cancer outcomes, in general, individuals with diabetes have similar beliefs about cancer risk and cancer protective behaviors compared to individuals without diabetes. More research is needed to gain a better understanding of improving cancer beliefs among people with diabetes. Our results call for tailored interventions to increase awareness about cancer risk and cancer protective behaviors among people living with diabetes.

CRediT authorship contribution statement

Jiamin Hu: Writing – original draft, Visualization, Investigation, Formal analysis, Data curation, Conceptualization. Young-Rock Hong: Supervision, Methodology, Formal analysis. Daniela Rivero-Mendoza: Writing – review & editing, Project administration. William T. Donahoo: Writing – review & editing, Resources, Conceptualization. Naykky Singh Ospina: Writing – review & editing, Conceptualization. Stephanie A. Staras: Writing – review & editing, Conceptualization. Elizabeth Shenkman: Writing – review & editing, Resources. Rahma S. Mkuu: Writing – original draft, Supervision, Resources, Investigation, Conceptualization.

Funding

Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number K01CA292583. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Rahma Mkuu reports financial support was provided by National Institutes of Health National Cancer Institute. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Glossary

BMI

Body Mass Index

T1D

Type 1 Diabetes

T2D

Type 2 Diabetes

HBM

Health Belief Model

HINTS

Health Information National Trends Survey

Appendix A. Appendix

Table A1.

Differences in cancer protective behaviors among the sample of individuals living with and without a diabetes diagnosis in the United States, from March 7, 2022, to November 8, 2022.

	No Diabetes, % (95 % CI)	Diabetes, % (95 % CI)	p-Value2
Smoking status			0.04
Current & Former smoker	31.4 (28.8, 34.1)	38.2 (32.6, 44.1)
Never	68.6 (65.9, 71.2)	61.8 (55.9, 67.4)
Drinking status			0.00
No	47.7 (44.7, 50.6)	62.5 (55.3, 69.1)
Moderate drinking	40.9 (38.1, 43.8)	30.9 (25.4, 37.0)
Heavy drinking	11.4 (9.5, 13.6)	6.6 (4.4, 9.9)
At least 150-min weekly moderate-intensity physical activity			0.00
	42.8 (40.2, 45.4)	26.0 (21.2, 31.6)
Have looked for information about cancer from any source			0.16
	43.9 (40.7, 47.1)	38.9 (33.0, 45.0)
Number of cancer prevention behavior1			0.00
0	2.2 (1.2, 3.8)	1.5 (0.3, 6.3)
1	13.0 (11.1, 15.1)	17.7 (13.2, 23.3)
2	35.7 (32.7, 38.8)	45.8 (38.4, 53.4)
3+	49.1 (45.8, 52.5)	35.1 (29.5, 41.1)

¹Number of cancer prevention behavior is defined by summing four health prevention behaviors: never smoking, not engaging in heavy drinking, meeting moderate-intensity physical activity guidelines, and seeking cancer-related health information.

²Chi-squared test adjusted Wald test.

Table A2.

Differences in cancer beliefs among the sample of individuals living with and without a diabetes diagnosis in the United States, from March 7, 2022, to November 8, 2022.

	No diabetes, % (95 % CI)	Diabetes, % (95 % CI)
Worried about getting cancer
Not at all	20.8 (18.3, 23.5)	23.8 (17.3, 31.8)
Slightly	31. (29.0, 34.9)	24.0 (19.4, 29.3)
Somewhat	27.9 (25.5, 30.5)	29.8 (24.7, 35.5)
Moderately	13.7 (11.9, 15.7)	15.9 (12.2, 20.4)
Extremely	5.7 (4.6, 7.1)	6.5 (4.5, 9.2)
How likely do you think you are to get cancer in your lifetime
Very unlikely	9.6 (7.8, 11.9)	11.4 (7.0, 18.1)
Unlikely	12.8 (11.2, 14.7)	7.9 (5.7, 11.0)
Neither likely nor unlikely	33.6 (30.2, 37.1)	30.6 (24.5, 37.4)
Likely	18.2 (16.2, 20.4)	17.0 (13.4, 21.4)
Very likely	4.9 (3.8, 6.3)	3.8 (2.3, 6.2)
I don't know	20.8 (18.4, 23.5)	29.2 (23.0, 36.3)
It seems like everything causes cancer…
Strongly agree	21.7 (19.5, 24.2)	23.9 (18.4, 30.5)
Somewhat agree	46.5 (43.6, 49.5)	43.1 (36.3, 50.2)
Somewhat disagree	19.8 (17.7, 22.0)	17.6 (13.7, 22.3)
Strongly disagree	12.0 (10.1, 14.1)	15.3 (9.9, 23.0)
There's not much you can do to lower your chances of getting cancer
Strongly agree	6.0 (4.7, 7.5)	7.5 (5.1, 11.1)
Somewhat agree	21.3 (18.7, 24.1)	27.3 (21.9, 33.4)
Somewhat disagree	44.1 (41.3, 47.0)	43.5 (36.9, 50.4)
Strongly disagree	28.7 (25.8, 31.8)	21.7 (17.1, 27.0)
There are so many different recommendations for preventing cancer, it's hard to know which ones to follow
Strongly agree	18.0 (16.3, 19.9)	24.7 (18.3, 32.5)
Somewhat agree	50.8 (48.2, 53.4)	45.0 (38.8, 51.4)
Somewhat disagree	22.4 (20.4, 24.5)	23.0 (17.2, 30.1)
Strongly disagree	8.8 (7.4, 10.4)	7.3 (5.0, 10.4)
When I think about cancer, I automatically think about death
Strongly agree	19.0 (17.0, 21.2)	24.4 (19.8, 29.7)
Somewhat agree	39.6 (36.2, 43.1)	37.1 (30.0, 44.9)
Somewhat disagree	25.6 (23.0, 28.3)	24.1 (18.9, 30.2)
Strongly disagree	15.8 (13.8, 18.0)	14.3 (8.6, 22.9)
How much progress has been made in preventing cancer?
A lot	23.3 (20.9, 25.9)	22.0 (16.8, 28.3)
Some	37.6 (34.4, 40.9)	35.0 (29.8, 40.6)
A little	20.3 (17.9, 22.9)	16.9 (13.8, 20.6)
Not at all	8.4 (7.0, 10.1)	9.9 (6.4, 15.2)
Don't know	10.4 (8.6, 12.6)	16.1 (11.1, 22.8)

Data availability

Data will be made available on request.