Type 2 diabetes risk and health-promoting behaviours among health sciences students: a cross-sectional analysis using FINDRISC and HPLP-II

Abstract

Background

The global burden of type 2 diabetes mellitus has continued to increase, and early signs of behavioural risk are now being detected even in young adults. Understanding these risks in student populations is important for the development of early prevention strategies. This study aimed to evaluate type 2 diabetes risk and investigate its association with health-promoting behaviour domains, body mass index, and selected demographic characteristics among health sciences students.

Methods

A cross-sectional study was carried out during the 2023–2024 academic year with 260 students enrolled in health sciences programmes. Data were gathered through a sociodemographic form, FINDRISC, and the Health-Promoting Lifestyle Profile II (HPLP-II).

Anthropometric measurements (weight, height, waist circumference) were obtained using standard procedures. Descriptive analyses, correlation analyses, and multiple linear regression analysis were conducted to examine associations between FINDRISC scores and demographic characteristics, body mass index, and HPLP-II subdomains.

Results

The sample included 260 students (mean age: 23.63 ± 4.38 years; 68.8% female) from nutrition and dietetics, nursing, and social work programs. The mean FINDRISC score was 5.50 ± 3.56, and 70.4% of students were classified as low risk. The mean HPLP-II total score (122.52 ± 30.79) suggested a moderate level of health-promoting behaviours, with the highest scores in spiritual growth and interpersonal relations and the lowest in physical activity and stress management. Although overall HPLP-II scores were not significantly correlated with FINDRISC scores, specific subdomains, particularly physical activity, showed significant associations after adjustment for demographic characteristics and body mass index. Higher BMI was associated with increased FINDRISC scores (B = 0.431, p < 0.001), whereas higher physical activity scores were associated with lower FINDRISC scores (B = -0.159, p = 0.011).

The model explained 27.3% of the variance (adjusted R² = 0.273).

Conclusion

Although the students generally reported moderate levels of health-promoting behaviour, higher body mass index and lower physical activity levels were associated with higher diabetes risk scores in this group. These findings suggest that universities may benefit from supporting students in developing sustainable physical activity habits and maintaining healthy body weight as part of diabetes risk prevention strategies. Further longitudinal studies are needed to clarify the direction and underlying mechanisms of these associations.

Introduction

According to the World Health Organization (WHO), health encompasses physical, mental, and social well-being rather than merely the absence of disease [1]. In this context, lifestyle behaviours such as healthy eating, regular physical activity, stress management, social relationships, and personal health responsibility are considered important determinants of overall health and chronic disease prevention [2–5]. Although these behaviours are influenced by individual choices, they are also shaped by broader social and environmental factors [3].

Type 2 diabetes mellitus (T2DM) is one of the chronic diseases most strongly associated with modifiable lifestyle factors, including physical inactivity, unhealthy dietary habits, and excess body weight [6–9]. Its prevalence continues to increase worldwide and in Türkiye, where approximately 15.9% of adults aged 20 years and older are estimated to live with diabetes, the majority of whom have T2DM [10]. Despite the preventable nature of many diabetes-related risk factors, studies indicate that university students and young adults often demonstrate insufficient engagement in health-promoting behaviours and limited awareness of diabetes risk [11–14]. The university period is considered a critical stage during which long-term health behaviours may become established and persist into adulthood [12, 15–18].

Previous studies conducted among university students in Türkiye have generally reported low to moderate diabetes risk levels and moderate health-promoting behaviour scores [11–13, 15]. In particular, physical activity has frequently been identified as the lowest-scoring HPLP-II domain among health sciences students [19–21].

However, most previous studies have examined diabetes risk and health-promoting behaviours separately, and limited evidence is available regarding how specific lifestyle behaviour domains are associated with diabetes risk among health sciences students.

Addressing this gap is important, as identifying modifiable lifestyle-related factors during early adulthood may support the development of targeted prevention strategies in university settings.

The Finnish Diabetes Risk Score (FINDRISC) is a validated and non-invasive tool used to estimate the 10-year risk of developing T2DM [22]. The Health-Promoting Lifestyle Profile II (HPLP-II) assesses health-promoting behaviours across six domains, including physical activity, nutrition, stress management, and health responsibility [23]. Together, these tools enable simultaneous evaluation of diabetes risk and health-promoting behaviours.

Therefore, this study aimed to evaluate type 2 diabetes risk and examine its associations with health-promoting behaviours, body mass index, and demographic characteristics among students in the Faculty of Health Sciences using the FINDRISC and HPLP-II instruments.

Methods

Study design and participants

This cross-sectional study was conducted at the Faculty of Health Sciences, Bitlis Eren University, Türkiye, between February and March 2024 and was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [24]. The study population consisted of 913 undergraduate students enrolled in the faculty during the 2023–2024 academic year.

The required sample size was calculated using OpenEpi software [25]. Based on previous research indicating that 63.2% of university students were classified as low risk for type 2 diabetes [26], the expected proportion was set at 0.632 with a 95% confidence level and a design effect of 1. The minimum required sample size was calculated as 258 students.

Considering potential nonresponse, 260 students with complete data were included in the final analysis.

Participants were recruited using a convenience sampling method from the Nutrition and Dietetics, Nursing, and Social Work departments. Data collection was conducted in person during students’ free time in common areas such as cafeterias and dormitories. Undergraduate students who voluntarily agreed to participate and provided written informed consent were included in the study. Students with a self-reported diagnosis of type 2 diabetes mellitus, pregnant students, and those with incomplete questionnaire data were excluded from the analysis.

Students previously diagnosed with type 2 diabetes mellitus were excluded because FINDRISC is intended to estimate future diabetes risk and is not appropriate for individuals already diagnosed with the disease [22]. Pregnant students were excluded due to physiological changes during pregnancy, including increased insulin resistance, which may influence metabolic status and lifestyle behaviours [27]. Although convenience sampling facilitated participant recruitment, inclusion of students from a single university may limit the generalizability of the findings and increase the potential for selection bias.

Ethics approval and consent to participate

The research protocol received approval from the Ethics Committee of Bitlis Eren University (Decision No. 2022/15; E.76710). Participation in the study was voluntary, and each student provided written consent after being informed about the study’s purpose. The research was carried out in line with the ethical standards set forth in the Declaration of Helsinki.

Data collection instruments and procedures

Questionnaire data collection

Data were gathered through a structured questionnaire composed of three sections: sociodemographic items, the FINDRISC tool, and the HPLP-II scale. Alongside the questionnaire, trained members of the research team measured participants’ weight, height, and waist circumference using routine procedures. Data collection was conducted via interviewer-administered, face-to-face surveys.

The socio-demographic information form was developed by the research team based on variables commonly employed in national and international epidemiological research on diabetes risk and lifestyle behaviours. In addition to international guidelines [10, 28, 22], the development of the form was guided by findings from recent studies conducted with university students in Türkiye [11, 29, 30]. The sociodemographic form included age, sex, marital status, department, year of study, smoking status, and selected health-related characteristics. The full version of the socio-demographic form is presented in Appendix 1.

The FINDRISC instrument was used to estimate participants’ 10-year risk of developing type 2 diabetes [22]. The HPLP-II scale, which consists of 52 items covering six domains of health-promoting behaviours, was administered to assess students’ lifestyle practices [23].

Finnish type 2 DM risk score (FINDRISC)

FINDRISC is a widely validated screening tool designed to estimate an individual’s 10-year risk of developing type 2 diabetes mellitus. The questionnaire consists of eight items covering age, body mass index (BMI), waist circumference, daily physical activity, fruit and vegetable consumption, use of antihypertensive medication, history of high blood glucose, and family history of diabetes. Each item contributes a weighted score, and the total score ranges from 0 to 26 [22, 31].

The scale was originally developed by Tuomilehto and Lindström in the late 1980s and validated during the early 1990s, providing a practical, non-invasive method suitable for large population-based assessments. For this study, total FINDRISC scores were calculated by summing the points assigned to each item according to the original scoring protocol [22].

Risk categories were defined using the standard FINDRISC thresholds:

• < 7 points: low risk.

• 7–11 points: slightly increased risk.

• 12–14 points: moderate risk.

• 15–20 points: high risk.

• ≥ 20 points: very high risk.

These categorical cut-offs were used for descriptive purposes, whereas the total FINDRISC score (0–26) was treated as a continuous variable in correlation and regression analyses. The predictive validity of FINDRISC has been demonstrated in different populations [32, 33]. In this study, the Turkish version validated by the Turkish Society of Endocrinology and Metabolism was used [34, 35].

The Health-Promoting Lifestyle Profile II (HPLP-II)

The HPLP-II is a widely used instrument that assesses health-promoting lifestyle behaviours across six domains: health responsibility, physical activity, nutrition, spiritual growth, interpersonal relations, and stress management. The scale was originally developed by Pender et al. and later adapted and validated for the Turkish population [36, 37].

The scale consists of 52 positively worded items rated on a four-point Likert scale ranging from 1 (“never”) to 4 (“routinely”). Mean scores were calculated for each subdomain, while total HPLP-II scores ranged from 52 to 208. Higher scores indicate greater engagement in health-promoting behaviours. In the present study, both total scores and mean subdomain scores were analysed as continuous variables.

Previous studies have reported high internal consistency for the HPLP-II [37]. In the present study, Cronbach’s α coefficient for the overall scale was 0.97.

Anthropometric measurements

Anthropometric assessments including body weight, height, and waist circumference were performed by trained senior students from the Department of Nutrition and Dietetics. All assessors had received formal instruction and demonstrated competency in standardized measurement techniques (barefoot, light clothing). Instruments were calibrated and checked for accuracy prior to each data collection session.

Body mass index (BMI) was calculated as weight (kg) divided by height (m²) and categorised according to World Health Organization criteria:

• < 18.5: underweight.

• 18.5–24.9: normal.

• 25.0-29.9: overweight.

• ≥ 30.0: obese [38].

Waist circumference was measured using a non-elastic measuring tape at the midpoint between the lowest rib and the iliac crest, with participants standing upright and breathing out gently, in accordance with standard measurement guidelines [39]. Central obesity was classified using internationally accepted thresholds of ≥ 94 cm for males and ≥ 80 cm for females [40]. Continuous anthropometric data were used in analyses, with categorical classifications applied where relevant.

Variables

The dependent variable was the FINDRISC score. HPLP-II total and subscale scores were treated as independent variables in correlation and regression analyses.

Statistical analysis

All statistical analyses were performed using SPSS version 26.0. Descriptive statistics are presented as means ± standard deviations and frequencies with percentages. Normality was assessed using skewness and kurtosis values between − 2 and + 2, which were considered indicative of acceptable normal distribution [41]. Accordingly, parametric tests were applied.

Group comparisons were conducted using independent-samples t tests for binary variables and one-way ANOVA for variables with more than two categories. Effect sizes were calculated using Cohen’s d for t tests and eta-squared (η²) for ANOVA analyses [42].

Associations between continuous variables were evaluated using Pearson correlation coefficients.

Multiple linear regression analysis was performed to identify factors independently associated with FINDRISC scores. Variables included in the regression model were selected based on theoretical relevance, previous literature, and findings from preliminary bivariate analyses.

The model included age, sex, marital status, body mass index (BMI), and HPLP-II subscale scores, and was constructed using the Enter method.

Model fit was assessed using the F-statistic, and regression results are presented as standardized regression coefficients (β), standard errors, and p values. Multicollinearity was evaluated using variance inflation factor (VIF) values. Although some VIF values suggested moderate multicollinearity, all values remained below the commonly accepted threshold of 10 (range: 1.20–6.24), indicating acceptable levels for regression analysis. Statistical significance was defined as a two-tailed p value < 0.05.

Results

Analyses were conducted on a final sample of 260 undergraduate students, and the participant selection process is presented in Fig. 1.

Fig. 1.

Flowchart of participant selection and exclusion

The mean age of the participants was 23.63 ± 4.38 years, and most participants were female (68.8%) and single (89.6%). Nearly half of the students were enrolled in the Nutrition and Dietetics program, and the majority were classified as having normal body weight. Detailed sociodemographic and anthropometric characteristics are presented in Table 1.

Table 1.

Sociodemographic characteristics of the participants (n = 260)

Variable	Category	n	%
Age group	18–20	40	15.4
	21–23	137	52.7
	≥ 24	83	31.9
Age Mean ± SD (Min-Max)	23.63 ± 4.38 (18–45)
Sex	Male	81	31.2
	Female	179	68.8
Marital status	Single	233	89.6
	Married	27	10.4
Department	Nutrition and dietetics	109	42.0
	Nursing	89	34.2
	Social work	62	23.8
Year of study	1st year	44	16.9
	2nd year	55	21.2
	3rd year	59	22.7
	4th year	102	39.2
Living arrangement	Dormitory	100	38.5
	With family	126	48.5
	Own home (alone)	21	8.0
	Own home (with friends)	13	5.0
BMI Category	Underweight	22	8.5
	Normal weight	175	67.3
	Overweight	51	19.6
	Obese	12	4.6
BMI Mean ± SD (Min-Max)	23.07 ± 3.86 (17.17–37.29)

The mean total FINDRISC score of the participants was 5.47 ± 3.54, with most students classified as having low diabetes risk. Detailed FINDRISC category distributions are presented in Table 2.

Table 2.

FINDRISC and 10-year type 2 diabetes risk levels (n = 260)

FINDRISC	n	%	Mean ± SD	Risk level	10-year diabetes risk
< 7	183	70.4	3.64 ± 1.93	Low	1%
7–11	63	24.2	8.83 ± 1.40	Slightly elevated	4%
12–14	9	3.5	13.22 ± 0.67	Moderate	16%
15–20	5	1.9	16.20 ± 1.30	High	33%
> 20	-	-	-	Very high	50%

FINDRISC Finnish Diabetes Risk Score, Total FINDRISC (mean ± SD): 5.47 ± 3.54 (Min–Max: 0–18), FINDRISC scores indicate higher diabetes risk

The mean total HPLP-II score was 122.52 ± 30.79, indicating a moderate level of health-promoting behaviours among the participants. Among the subdomains, the highest mean scores were observed for spiritual growth and interpersonal relations, whereas physical activity and stress management had the lowest scores (Table 3).

Table 3.

HPLP-II total and subscale scores (n = 260)

Scale/Subscale	Number of items	Observed range	Mean ± SD
HPLP-II total	52	52–208	122.52 ± 30.79
Spiritual growth	9	9–36	23.75 ± 6.35
Nutrition	9	9–36	19.80 ± 5.41
Physical activity	8	8–32	17.02 ± 5.80
Health responsibility	9	9–36	20.27 ± 5.83
Interpersonal relations	9	9–36	23.18 ± 5.86
Stress management	8	8–32	18.50 ± 4.93

SD  Standard deviation, HPLP-II Health-Promoting Lifestyle Profile II, Higher HPLP-II subscale scores indicate stronger engagement in health-promoting behaviours

In terms of the distribution of mean FINDRISC scores across variables, participants aged ≥ 24 years had significantly higher scores than those aged 21–23 years (p < 0.001), with a medium effect size (η² = 0.065). Male students also had significantly higher FINDRISC scores compared with female students (p = 0.042), although the effect size was small (Cohen’s d = 0.275). No statistically significant differences in FINDRISC scores were observed across departments (Table 4).

Table 4.

Comparison of the mean FINDRISC across selected variables (n = 260)

Variable	Category	FINDRISC Mean ± SD	Test p value	Effect size
Age group	18–20	5.30 ± 2.90	F: 8.945 p < 0.001* (a-b)	η2: 0.065
	21-23a	4.74 ± 3.09
	≥ 24b	6.76 ± 4.13
Sex	Female	5.17 ± 3.50	t: 2.043 p = 0.042	Cohen’s d: 0.275
	Male	6.14 ± 3.57
Department	Nutrition and dietetics	5.15 ± 3.44	F: 0.774 p = 0.462	-
	Nursing	5.65 ± 3.39
	Social work	5.77 ± 3.90

t  independent-samples t test, F = one-way ANOVA

* Statistically significant values are presented in bold (p < 0.05), Post hoc group differences indicated by superscripts (a–b), Effect sizes: Cohen’s d = 0.2 (small), 0.5 (medium), and 0.8 (large); η² = 0.01 (small), 0.06 (medium), and 0.14 (large)

FINDRISC Finnish Diabetes Risk Score. FINDRISC scores indicate higher diabetes risk

HPLP-II total scores differed significantly across age groups (F(2,257) = 3.367, p = 0.037, η² = 0.025), with students aged 18–20 years demonstrating higher scores than those aged 21–23 years. Significant age-related differences were also observed for the nutrition (F(2,257) = 3.954, p = 0.021, η² = 0.030) and interpersonal relations subscales (F(2,257) = 5.448, p = 0.005, η² = 0.041). In both subdomains, students aged 21–23 years had lower scores compared with the other age groups (Table 6).

Table 6.

Correlations between HPLP-II and FINDRISC (n = 260)

Variable		FINDRISC
HPLP-II total	r p	− 0.106 0.087
Spiritual growth	r p	− 0.083 0.182
Nutrition	r p	− 0.082 0.186
Physical activity	r p	− 0.153 0.013*
Health responsibility	r p	− 0.076 0.222
Interpersonal relations	r p	− 0.072 0.247
Stress management	r p	− 0.111 0.074

r: Pearson correlation coefficient, Statistically significant values are presented in bold (*p < 0.05)

FINDRISC Finnish Diabetes Risk Score, HPLP-II Health-Promoting Lifestyle Profile II

Significant departmental differences were identified for the nutrition (F(2,257) = 4.585, p = 0.011, η² = 0.034) and health responsibility subscales (F(2,257) = 3.764, p = 0.024, η² = 0.028). Nutrition and dietetics students demonstrated higher scores in these domains compared with nursing students, and higher health responsibility scores compared with social work students (Table 5).

Table 5.

HPLP-II total and subscale mean scores across selected variables (n = 260)

Grouping variable	Category	HPLP-II total	Spiritual growth	Nutrition	Physical activity	Health responsibility	Interpersonal relations	Stress management
		Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD
Age Group	18–20	129.20 ± 33.39a	25.32 ± 6.07	20.72 ± 5.84	17.30 ± 6.29	21.72 ± 6.46	25.05 ± 5.82a	19.07 ± 5.49
	21–23	117.94 ± 29.25b	22.87 ± 6.27	18.92 ± 4.50a	16.51 ± 5.41	19.54 ± 5.59	22.10 ± 5.73b	17.98 ± 4.69
	≥ 24	126.86 ± 30.79c	24.42 ± 6.45	20.81 ± 5.67b	17.72 ± 6.16	20.78 ± 5.76	24.04 ± 5.77c	19.07 ± 4.99
		F: 3.343	F: 3.036	F: 3.924 p: 0.021* (a-b)	F: 1.184	F: 2.678	F: 5.448 p: 0.005* (a-b, b-c)	F: 1.585
		p: 0.037 (a-b, b-c)	p: 0.050		p: 0.308	p: 0.071		p: 0.207
		η2: 0.025		η2: 0.030			η2: 0.041
Sex	Female	123.42 ± 28.24	24.14 ± 5.86	19.78 ± 5.01	16.62 ± 5.65	20.46 ± 5.58	23.71 ± 5.20	18.69 ± 4.61
	Male	120.54 ± 35.90	22.86 ± 7.29	19.86 ± 6.23	17.90 ± 6.07	19.85 ± 6.34	22.00 ± 6.98	18.06 ± 5.57
		t: -0.637	t: -1.391	t: 0.104	t: 1.654	t: -0.784	t: -1.970	t: -0.964
		p: 0.525	p: 0.167	p: 0.917	p: 0.099	p: 0.434	p: 0.051	p: 0.336
Department	Nutrition and dietetics	127.87 ± 30.10	24.69 ± 5.97	20.95 ± 5.90a	17.54 ± 5.77	21.42 ± 5.92 a	24.11 ± 5.52	19.14 ± 4.99
	Nursing	118.84 ± 28.42	23.22 ± 6.17	18.71 ± 4.93b	16.84 ± 5.58	19.55 ± 5.30 b	22.34 ± 5.72	18.15 ± 4.47
	Social work	118.40 ± 34.16	22.82 ± 7.08	19.35 ± 5.95	16.35 ± 6.16	19.29 ± 6.11 c	22.72 ± 6.45	17.85 ± 5.37
		F: 2.876	F: 2.196	F: 4.585	F: 0.888	F: 3.764	F: 2.485	F: 1.694
		p: 0.058	p: 0.113	p: 0.011* (a-b)	p: 0.413	p: 0.024* (a-b, a-c)	p: 0.085	p: 0.186
				η2: 0.034		η2: 0.028

t = results from the independent-samples t test, F = results from one-way analysis of variance (ANOVA)

* Statistically significant values are presented in bold (p < 0.05), Effect size benchmarks for η ² (eta squared): 0.01 = small, 0.06 = medium, and ≥ 0.14 = large

Post hoc comparisons are denoted by the superscripts a, b, and c

HPLP-II   Health-Promoting Lifestyle Profile II. Higher HPLP-II subscale scores indicate stronger engagement in health-promoting behaviours

No significant associations were found between sex and HPLP-II total or subscale scores (p > 0.05).

No significant association was observed between the overall HPLP-II score and FINDRISC (p = 0.087). On the other hand, the physical activity subdimension of HPLP-II showed a modest but statistically significant inverse association with FINDRISC (r = -0.153, p = 0.013) (Table 6).

Multiple linear regression analysis was performed to identify variables independently associated with FINDRISC scores (Table 7). The model included age, sex, marital status, body mass index, and HPLP-II subscale scores. The regression model was statistically significant (F(10,249) = 10.731, p < 0.001) and explained 27.3% of the variance in FINDRISC scores (adjusted R² = 0.273). Among the examined variables, only BMI (β = 0.470, p < 0.001) and physical activity (β = −0.260, p = 0.011) showed significant associations with FINDRISC scores. Higher BMI was associated with higher diabetes risk scores, whereas higher physical activity scores were associated with lower FINDRISC scores (Fig. 2).

Table 7.

Multiple linear regression predicting FINDRISC (n = 260)

Predictor	B (Unstandardized Coefficients)	Std. Error	β (Beta, Unstandardized Coefficient)	t	p	95% CI Lower	95% CI Upper	Tolerance	VIF
Age	0.052	0.052	0.064	0.989	0.324	-0.05	0.15	,667	1,500
Sex	-0.306	0.442	-0.040	-0.693	0.489	-1.17	0.56	,836	1,196
Marital status	-0.635	0.661	-0.061	-0.96	0.338	-1.93	0.66	,691	1,447
BMI	0.431	0.054	0.470	8.043	0.000	0.33	0.54	,823	1,215
Spiritual growth	-0.046	0.068	-0.082	-0.676	0.499	-0.18	0.09	,190	5,276
Nutrition	0.012	0.071	0.018	0.17	0.865	-0.13	0.15	,236	4,229
Physical activity	-0.159	0.062	-0.260	-2.551	0.011	-0.28	-0.04	,270	3,704
Health responsibility	0.103	0.073	0.169	1.414	0.159	-0.04	0.25	,197	5,080
Interpersonal relations	0.023	0.071	0.038	0.322	0.748	-0.12	0.16	,202	4,957
Stress management	-0.036	0.095	-0.050	-0.379	0.705	-0.22	0.15	,160	6,236
R² = 0.301; Adjusted R2 = 0.273
Model fit: F(10, 249) = 10.731, p < 0.001

t: t-statistic for regression coefficients, Dependent variable: FINDRISC, Statistically significant values are presented in bold (p < 0.05)

BMI Body mass index, FINDRISC Finnish Diabetes Risk Score, VIF Variance inflation factor

Fig. 2.

Association of FINDRISC scores with (a) body mass index (BMI) and (b) physical activity scores among health sciences students. Higher BMI values were positively associated with FINDRISC, whereas higher physical activity scores were inversely related to FINDRISC (p < 0.05)

A total of 260 students were included in the analysis. The multiple linear regression model examining factors associated with diabetes risk (FINDRISC score) was statistically significant (F = 10.731, p < 0.001) and explained 30.1% of the variance in FINDRISC scores (R² = 0.301; adjusted R² = 0.273).

Among the variables included in the model, BMI was positively associated with diabetes risk (B = 0.431, 95% CI: 0.33 to 0.54, p < 0.001). In contrast, physical activity was negatively associated with diabetes risk (B = − 0.159, 95% CI: −0.28 to − 0.04, p = 0.011), indicating that higher physical activity scores were associated with lower FINDRISC scores.

No statistically significant associations were observed between FINDRISC scores and age, sex, marital status, spiritual growth, nutrition, health responsibility, interpersonal relations, or stress management (p > 0.05 for all). To reduce multicollinearity, the overall HPLP-II total score was excluded from the regression model because of its high correlations with the HPLP-II subscale scores.

Discussion

The findings of this study provide insight into the relationship between type 2 diabetes risk and health-promoting behaviours among health sciences students. Most participants were classified as having a low 10-year risk of type 2 diabetes, which is consistent with previous studies conducted among university students in Türkiye and other countries [11, 15, 26, 43–45].

The generally low diabetes risk observed in this population is likely related to the relatively young age of the participants, as diabetes prevalence and metabolic risk tend to increase with age [46]. Although students enrolled in health-related programs may have greater exposure to health information, the present findings do not allow conclusions regarding the influence of academic training on diabetes risk or lifestyle behaviours.

In the present study, students aged 24 years and older demonstrated higher FINDRISC scores compared with younger age groups. This finding is consistent with evidence indicating that diabetes risk increases progressively across adulthood [1, 45, 46]. Although the mechanisms underlying this difference cannot be determined in a cross-sectional design, behavioural factors such as reduced physical activity and increased sedentary behaviours may partially contribute to higher risk scores among older students [47, 48]. However, these interpretations should be considered cautiously, as lifestyle-related variables were not examined longitudinally in the present study.

Male students demonstrated slightly higher FINDRISC scores than female students, although the effect size was small. This finding is generally consistent with epidemiological studies reporting higher diabetes prevalence among men in some populations [49]. However, the present study was not designed to examine the biological or behavioural mechanisms underlying these differences, and the observed sex-related variation should therefore be interpreted cautiously. Social and cultural factors influencing health-related behaviours may also contribute to these differences [50], although such factors were not directly assessed in the current study.

The mean HPLP-II score indicated a moderate level of engagement in health-promoting behaviours among the participants. Similar findings have been reported in previous studies conducted among university students [11, 13, 51–53], although both lower and higher scores have also been documented in the literature [54–57]. These variations may reflect differences in cultural context, educational environments, and study populations. The moderate scores observed in the present study may suggest that awareness of healthy lifestyle practices does not always translate into consistent behavioural change. Previous research has similarly indicated that university students may experience barriers to maintaining health-promoting behaviours, including time constraints, limited access to healthy food options, and low self-efficacy [58–60].

Subscale analyses showed that spiritual growth and interpersonal relations had the highest scores, whereas physical activity and stress management were the lowest-scoring domains.

Similar patterns have been reported in previous studies among university students and health sciences populations [60, 61]. The consistently lower scores observed for physical activity and stress management may reflect common barriers experienced during university life, including academic workload, limited time, and difficulties maintaining regular health behaviours.

However, because contextual and behavioural determinants were not directly assessed in the present study, these interpretations should be considered cautiously.

Significant differences were observed across age groups and academic departments for selected HPLP-II subdomains, although the effect sizes were generally small. Nutrition and Dietetics students demonstrated higher nutrition and health responsibility scores than Nursing and Social Work students, which may reflect greater exposure to nutrition-related content within their academic curriculum [11, 62]. In contrast, no significant sex-related differences were identified for total HPLP-II scores or subdomains. Although previous studies have reported mixed findings regarding sex differences in lifestyle behaviours [51, 63–65], the relatively similar educational and social environments of the participants in the present study may partially explain the lack of substantial differences observed between male and female students. Nevertheless, these findings should be interpreted cautiously given the unequal distribution of participants by sex.

No significant association was observed between total HPLP-II scores and FINDRISC scores in this study. Similar findings have been reported in previous research suggesting that composite lifestyle measures may not always adequately reflect the influence of individual behavioural components on diabetes risk [14, 22, 31, 66]. In the present study, demographic variables and several HPLP-II subdomains were also not independently associated with diabetes risk in the regression analysis. The relatively homogeneous characteristics of the study population, consisting primarily of young adults with similar educational backgrounds and lifestyle patterns, may have reduced the variability required to detect significant associations [16, 67]. In addition, the effects of broader lifestyle indices may have been overshadowed by stronger predictors such as BMI and physical activity.

Higher BMI was associated with increased diabetes risk, whereas higher physical activity scores were associated with lower FINDRISC scores. These findings are consistent with previous studies identifying obesity and physical inactivity as important determinants of diabetes risk across different populations [32, 67–71]. The regression model explained 27.3% of the variance, indicating a moderate level of explanatory capacity. From a physiological perspective, excess body weight is associated with insulin resistance and metabolic dysregulation, whereas regular physical activity contributes to improved glucose metabolism and insulin sensitivity [22–31]. These mechanisms may partially explain why BMI and physical activity emerged as significant factors in the present study, while other lifestyle domains did not demonstrate independent associations.

Nevertheless, the findings should be interpreted cautiously because of the cross-sectional design and the limited range of variables included in the analysis. Other behavioural, environmental, and biological factors not assessed in this study may also contribute to diabetes risk.

Limitations and strengths

A major strength of this study is the use of validated instruments (FINDRISC and HPLP-II), together with the reporting of effect size statistics, which supports a clearer interpretation of the findings. The study also contributes additional data on diabetes risk and health-promoting behaviours among health sciences students, a population that remains relatively underrepresented in the literature in Türkiye.

Several limitations should be considered. First, the cross-sectional design precludes causal interpretation of the observed associations. Second, behavioural data were based on self-report and may therefore be subject to recall bias or socially desirable responses. Third, although anthropometric measurements were performed by trained senior Nutrition and Dietetics students, minor measurement variability cannot be completely excluded.

The study sample included a higher proportion of Nutrition and Dietetics students and was predominantly female, which may limit the representativeness and generalizability of the findings. In addition, convenience sampling from a single faculty may have increased the potential for selection bias. Group comparisons, particularly those based on age and sex, should therefore be interpreted cautiously because unequal group sizes may have influenced statistical significance. Furthermore, the regression model explained a moderate proportion of the variance in diabetes risk, suggesting that additional behavioural, environmental, and socioeconomic factors not assessed in this study may also contribute to diabetes risk.

Future longitudinal and multi-centre studies are needed to better clarify temporal relationships and provide a more comprehensive understanding of lifestyle-related diabetes risk patterns among young adults. Further research examining contextual, environmental, and cultural factors may also help explain why certain lifestyle domains, such as nutrition and stress management, were not independently associated with diabetes risk in this population.

Conclusion and recommendations

Most health sciences students in this study were classified as having a low 10-year risk of developing type 2 diabetes mellitus, although nearly one-third showed slightly or moderately elevated risk levels. Higher BMI was associated with increased diabetes risk, whereas higher physical activity scores were associated with lower risk. Although overall engagement in health-promoting behaviours was moderate, physical activity and stress management emerged as the lowest-scoring domains. These findings suggest that greater emphasis on healthy weight management and regular physical activity may be beneficial in student health promotion programs. Future longitudinal studies are needed to better understand how lifestyle behaviours influence diabetes risk over time.

Abbreviations

ADA

American Diabetes Association

ANOVA

Analysis of Variance

BMI

Body Mass Index

CI

Confidence Interval

FINDRISC

Finnish Diabetes Risk Score

HPLP-II

Health-Promoting Lifestyle Profile II

IDF

International Diabetes Federation

SPSS

Statistical Package for the Social Sciences

SD

Standard Deviation

T2DM

Type 2 Diabetes Mellitus

WHO

World Health Organization

Authors' contributions

E.A.K. conceptualized the study, developed the research protocol, coordinated and conducted data collection, and prepared the initial manuscript draft. K.K. assisted in designing the study framework and performed all statistical analyses. F.S. contributed to the planning stage, supported the interpretation of findings, and assisted with manuscript revisions. R.D. and D.A. participated in data acquisition. All authors critically reviewed the manuscript and approved the final version.

Funding

This research did not receive any financial support from governmental, commercial, or nonprofit organizations.

Data availability

The datasets generated and analysed during the study are available from the corresponding author upon reasonable request.

Declarations

This manuscript has not been published previously and is not under consideration elsewhere. Participant information was collected with the explicit condition that it would be used solely for scientific purposes and not shared publicly. Accordingly, the dataset is not openly accessible; any sharing would require collective consent from participants. All data relevant to the findings are presented within the article.

Artificial intelligence (ChatGPT, OpenAI) was used only for language editing. It did not contribute to study design, statistical analysis, data interpretation, or the drawing of conclusions.

Ethics approval and consent to participate

The research protocol received approval from the Ethics Committee of Bitlis Eren University (Decision No. 2022/15; E.76710). Participation in the study was voluntary, and each student provided written consent after being informed about the study’s purpose. The research was carried out in line with the ethical standards set forth in the Declaration of Helsinki. All participants were informed in detail about the purpose and procedures of the study, and written informed consent was obtained prior to participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.