Exploratory modeling of etiological factors for cardiovascular diseases within a structured nursing follow-up program

Abstract

Background

The prevention of etiological factors for cardiovascular diseases is a critical contemporary health priority. A significant knowledge gap exists regarding robust studies that analyze cardiovascular risk factors through the lens of nursing diagnoses. Modeling these factors allows for the prioritization of variables that should be targeted in clinical care plans. This study contributes to the advancement of nursing practice and the development of structured care programs focused on cardiovascular risk. The objective was to develop exploratory models of etiological factors for cardiovascular diseases through a structured nursing follow-up program.

Methods

This clinical trial was conducted between 2023 and 2024. A total of 71 university students presenting etiological factors for NANDA-I diagnosis 00311 were recruited. The intervention consisted of nursing consultations facilitated by the e-TEORISC software compared to standard care. Data analysis was performed using regression models via the Forward Stepwise method.

Findings

The follow-up included 35 participants, primarily Brazilian, single, mixed-race, and male, with a mean age of 25 years. The resulting models elucidated the relationships between metabolic syndrome and female gender, HDL, and SBP; harmful alcohol consumption vs. fasting blood glucose; and understanding of cardiovascular risk factors vs. waist circumference (no risk), neck-to-thigh ratio (No risk), COVID-19 history (negative), LDL, and HDL.

Conclusion

The models suggest significant associations between various etiological factors and cardiovascular risk indicators among university students.

Clinical trial number

Brazilian Registry of Clinical Trials n° RBR-8y3qx39 e Universal Trial Number nº U1111-1293-7938 (date registered: 19/07/2023).

Introduction

According to the World Health Organization (2022), cardiovascular diseases (CVD) are disorders related to the heart and blood vessels, with coronary heart diseases and cerebrovascular diseases as the main representatives, responsible for up to 17.9 million deaths per year [1, 2]. However, by measuring cardiovascular risk and associated factors, it is possible to predict control strategies for the ongoing disease or the likelihood of acquiring it. Cardiovascular factors are multifactorial components, and in Portuguese-speaking countries, the main highlights are elevated systolic blood pressure (SBP), dietary risks, alcoholism, elevated body mass index (BMI), and increased low-density lipoprotein (LDL) cholesterol [3].

Cardiovascular risk predisposes individuals to vulnerability and is related to modifiable and non-modifiable risk factors. In this context, nursing plays a complex role in promotion and prevention [4]. This activity is only possible through nursing care, defined as possessing quality, safety, ethics, and collaboration, based on the best health evidence and is associated with the stages of the nursing process (NP) [5]. In the context of cardiovascular risk, the nursing diagnosis of Risk for impaired cardiovascular function (00311) is defined as the susceptibility to disturbances in substance transportation, body homeostasis, tissue metabolic waste removal, and organic function, which can compromise health [6]. Thus, care based on nursing theory strengthens the NP and favors the advancement in the scope of nursing practice evidenced by the accurate recording of the nursing diagnosis.

The Theory of Care in the Context of Cardiovascular Risk (TEORISC), the theoretical basis of this study, aims to prescribe nursing actions to promote health and reduce cardiovascular risk aligned with individual factors and phenomena, as well as to reduce morbidity and mortality from chronic non-communicable diseases in the medium and long term [7]. For this purpose, care programs with innovative potential are necessary, ensuring continuity of care for people with the nursing diagnosis of Risk for impaired cardiovascular function (00311). Consequently, actions can be accessed and implemented commonly through information technologies, among which the e-TEORISC software stands out [8]. This software has a structured and validated nursing consultation algorithm that facilitates the stratification, classification, and implementation of care for individuals with cardiovascular risk [8].

A care program should combine theory and care tools focusing on strengthening professional practice. There are CVD programs focusing on clinical parameters that faced limitations due to the lack of a professional capable of operationalizing the interventions [9]. However, when led by nurses, a reduction in cardiovascular risk in patients with CVD is observed [10]. There is a gap in knowledge regarding robust studies directed at studying the diagnosis in focus, especially concerning cardiovascular risk factors.

This study is original and innovative due to its complexity in studying the care provided to people with the nursing diagnosis of Risk for impaired cardiovascular function (00311), aiming at promoting cardiovascular care through a program that proposes to evaluate the impacts on health potential and morbidity reduction in university students. Regarding the prediction of cardiovascular risk in university students through models, only one study showed that knowledge had an indirect effect on health promotion [11]. However, the study did not use classification systems and nursing theories in its structure, which this study proposes to do for scientific and technological development in the field of cardiovascular nursing.

The objective of this study was to develop exploratory models of etiological factors for cardiovascular diseases based on nursing diagnoses, utilizing a structured follow-up program.

Methods

Data

To conduct the exploratory and associative modeling in this study, we used secondary data derived from a blinded randomized clinical trial (RCT) with two parallel groups, conducted at a single center and registered in the Brazilian Registry of Clinical Trials (No. RBR-8y3qx39). The study was reported in accordance with the TRIPOD guidelines.

Sample size

This was a pilot study in which the sample size calculation was based on a two-group difference model. The calculation assumed a statistical power of 80% and an anticipated 50% reduction in the proportion of risk factors (effect size), reflecting the prevalence of cardiovascular risk in the target population [12]. The calculation assumed a 30% difference in risk reduction between the two groups. This calculation determined that a sample size of 35 participants per group would provide sufficient power for the study. Any losses or withdrawals were meticulously recorded. Participants who missed appointments were discontinued from the study to maintain research fidelity.

Participants

The inclusion criteria were as follows: presenting at least two risk factors associated with the NANDA-I nursing diagnosis Risk for impaired cardiovascular function (00311), age between 18 and 59 years, and formal registration with the health service or a demonstrated interest in clinical follow-up for cardiovascular risk. For the diagnostic inference of Risk for Impaired Cardiovascular Function (00311) (NANDA-I 2021–2023) [13], definitions established by the study protocol [14] were applied. The following etiological factors were evaluated: family history of CVD, sedentary lifestyle/physical inactivity, harmful alcohol consumption, dyslipidemia, insufficient knowledge of cardiovascular risk factors, overweight, obesity, smoking, unhealthy diet, anxiety, and stress. Additionally, associated conditions such as diabetes mellitus, illicit drug use, depression, and metabolic syndrome were considered.

Each factor was assessed based on the scales and parameters detailed in the “Outcomes and Predictors” section. Following the identification of these factors, the nursing diagnostic inference was performed. This process was validated by the lead researcher and two additional nurses, who reached full internal consensus regarding the presence or absence of the diagnosis.

Data collection was structured into three distinct phases: Phase I – baseline, which consists of presenting the research and recruiting (conducted by the research team); Phase II – application (conducted by the nurse researcher for the intervention group (IG) and professors for the control group (CG)); and Phase III – outcome evaluation (conducted by the research team). For the IG, the intervention phase consisted of two in-person consultations: an initial software-mediated consultation followed by a reinforcement session 45 days later. The total duration between the initial intervention and the final outcome evaluation for both groups was 90 days, which corresponds to the minimum duration required to establish a nurse-led care program [15]. The study was conducted at the Comprehensive Health Care Center in the State of Ceará, Brazil.

The IG received a structured nursing consultation conducted by a registered nurse. The e-TEORISC software served as the primary tool for recording sociodemographic, anthropometric, and clinical-laboratory data. Furthermore, the software facilitated the identification of nursing diagnoses, the establishment of nursing outcomes, and the prescription of interventions based on the TEORISC theoretical framework.

While NANDA-I was employed as the parameter for participant selection—specifically identifying the etiological factors of the Risk for Impaired Cardiovascular Function diagnosis—the International Classification for Nursing Practice (ICNP/CIPE^®) was the classification system utilized by the software to guide the care plan [8]. For each participant in the IG, the lead researcher developed individualized care plans consisting of diagnoses, interventions, and outcomes tailored to assessed needs by selecting parameters available within the software. Data for this group were recorded directly into the e-TEORISC platform.

Both the intervention and control groups were evaluated independently of their clinical setting. Each intervention was mapped to a specific nursing diagnosis, enabling longitudinal clinical follow-up. The control group (CG) received standard care, which followed a non-validated script based on the clinical experience of the nursing faculty. Consultations for the CG were performed by faculty members specializing in cardiovascular health. For this group, data and care plans were recorded using Microsoft Word documents.

The e-TEORISC software

The e-TEORISC is a clinical decision support system designed to assist nurses in identifying nursing diagnoses, interventions, and outcomes based on cardiovascular risk variables [8]. The software incorporates 169 interventions, 55 diagnoses, and 36 nursing outcomes, categorized into cardiometabolic, behavioral, psychosocial/cultural, occupational, infectious, and therapeutic factors. This structure is grounded in the TEORISC theoretical framework, which provides the conceptual basis for the platform [7]. The e-TEORISC is an institutional software registered with the Brazilian National Institute of Industrial Property (INPI) (Process No. BR512025004713-4) and was formally licensed to this study for application and clinical testing. The software interface is organized into distinct modules that enable the systematic screening of dietary habits and behavioral, psychosocial, and therapeutic factors associated with the development of cardiovascular diseases as established in the literature. By facilitating the identification of specific patient needs, the system suggests diagnoses and interventions for integration into the care plan. Consequently, following the consultation and risk assessment, the software enables the formulation of an individualized care plan with targeted goals for risk factor reduction tailored to the specific profile of each participant.

Outcome and predictors

The primary outcome was the reduction of target etiological factors or clinically significant changes in two or more factors characterizing the diagnosis Risk for impaired cardiovascular function (00311), according to specific criteria, over 45 days of clinical follow-up conducted by a nurse. The independent variables refer to the sociodemographic, clinical, and laboratory factors studied. The outcome variables were the reduction, elimination, or control of the etiological factors of nursing diagnosis 00311.

Due to the complexity of the evaluated risk factors, the following reference values were adopted for the assessment of modifiable risk factors: dyslipidemia (LDL ≥ 160 mg/dL, triglycerides ≥ 150 mg/dL, HDL < 40 mg/dL in men or < 50 mg/dL in women); overweight (BMI between 25.0 and 29.9 kg/m²); smoking (≥ 100 cigarettes lifetime or use of electronic cigarettes/vaping); sedentary lifestyle (< 150 min of physical activity per week); stress (Lipp Stress Symptoms Inventory for Adults, presence of ≥ 7 symptoms); anxiety (State-Trait Anxiety Inventory score ≥ 44); alcohol consumption; insufficient knowledge of cardiovascular risk factors (≤ 5 etiological factors identified); unhealthy diet (score ≥ 5 on an 11-point scale [18]); illicit drug use (Alcohol, Smoking and Substance Involvement Screening Test score 1–3); metabolic syndrome (National Cholesterol Education Program Adult Treatment Panel III criteria); and depression (Major Depression Inventory score ≥ 16).

These variables were assessed using validated instruments and/or appropriate biochemical parameters [14]. Once a risk factor was identified based on established thresholds, the outcome focused on the reduction of scale scores or improvement in biochemical parameters, indicating clinically meaningful changes in behavior over a relatively short follow-up period.

The following were considered clinically significant changes: reduction in BMI category (e.g., from obesity to overweight), acquisition of knowledge of at least five cardiovascular risk factors, elimination of sedentary behavior, reduction in alcohol-related risk scores, smoking cessation, and adoption of regular physical activity. Additionally, improvement or control of biochemical markers—such as total cholesterol, HDL-c, LDL-c, VLDL, triglycerides, and blood glucose—was considered, including reductions, stabilization, or increases when clinically appropriate.

Participants who achieved at least one of these changes were classified as adherent to the proposed interventions and as having a significant modification in the corresponding risk factor. This reflects the potential of the intervention to influence the evaluated predictors. However, it is acknowledged that outcomes depend on patient self-management; therefore, findings should be interpreted with caution, particularly given the pilot nature of the study.

An external researcher to the analyst team was responsible for the randomization procedure. The randomization was blocked, with an allocation ratio of 1:1 in two arms (intervention and control). We designated the number of cardiovascular etiological factors as the stratification factor of the blocks of two participants, considering the similarity of this variable for each pair.

Analysis and statistics

Excel for Windows 365 was used for data entry and Statistical Package for Social Sciences version 24 software for table and graph creation and data analysis. To summarize the data, the mean was used with its 95% confidence interval. The tests were bilateral with α = 0.05.

Statistical description, frequency distribution (mean) and standard deviation (SD) of sociodemographic, clinical, and general health profile variables were used. Then normality was identified (Kolmogorov-Smirnov test). For categorical variables, Fisher’s exact test and logistic regression technique with odds ratio, confidence interval (CI) 95%, and p-value adjusting the model to the Wald test for verification of its significance were employed. For numerical data, parametric statistical tests for mean comparison (Student’s t-test or Anova) or medians (Mann Whitney) were applied in the case of non-parametric distributions.

The five steps for developing logistic regression were adopted: Identify the dependent variable; Observe technical requirements; Estimate and adjust the model; Interpret the results; and Validate the results. The main interest is to model the probability of success, Prob(Y = 1|X), as a function of some exploratory variables (X) that may influence this probability of success.

In logistic regression, the following equation was estimated: . The objective is to estimate the probability of occurrence (success) of a given outcome. On the right-hand side of the equation are the independent variables, (categorical or continuous), along with their corresponding estimated coefficients (). The time frame used for modeling considered the period between baseline and post-intervention, corresponding to a 90-day interval between the initial assessment and the outcome.

In the application of regression models, it is common practice to use variable selection methods, particularly when a large number of predictors is available. These techniques aim to identify variables with the greatest predictive power. Accordingly, the models in this study were developed as secondary analyses based on baseline data from a RCT.

The regression technique employed was the forward stepwise method, beginning with an initial model (Model 0) that included no predictor variables. At each step, variables were added sequentially based on their statistical significance, starting with the variable presenting the lowest p-value. The procedure involved the inclusion of each variable followed by testing against a predefined significance threshold. Only coefficients that demonstrated statistical significance according to the Wald test were retained in the model. This test was used to assess the significance of each coefficient in the logistic regression equation, including the intercept.

To this end, logistic regression was applied to model multiple phenomena in the study, namely: sedentary lifestyle/physical inactivity, alcohol use/consumption, insufficient knowledge or understanding of cardiovascular risk factors, overweight, obesity, smoking, unhealthy diet, anxiety/stress, illicit drug use, depression, and metabolic syndrome.

These variables were treated as binary outcomes (yes/no), where “yes” represents the occurrence (success) of the event and “no” its absence (failure). The primary objective was to identify models capable of predicting the probability of occurrence, P(Y = Yes), given specific patient characteristics.

As this is an exploratory study, a significance level of p ≤ 0.10 was adopted as the criterion for variable inclusion in the model. Model performance was assessed using predicted versus observed values. Models with a cutoff point of ≥ 50% were considered acceptable to indicate an association between the presence of predictor variables and statistically significant independent variables [16].

Ethical considerations

The study was approved by the research ethics committee of the University Of International Integration Of Afro-Brazilian Lusophony (UNILAB, approval number 6.092.932) and developed according to the ethical principles with the Declaration of Helsinki. Written and verbal informed consent has been obtained from the client.

Results

Seventy-one students were recruited, five of whom had CVD, leaving 66 for randomization. After applying inclusion, exclusion, allocation, discontinuation, and dropout criteria for IG and CG, 35 patients remained, of whom 22 (62.8%) belonged to the IG and 13 (42.8%) to the CG, Fig. 1.

Fig. 1.

CONSORT flow diagram

It is noteworthy that the participants in this study are young university students, on average, 25 years old, Brazilian (n = 32, 91.4%), male (n = 21, 60%) and self-identified as brown (n = 16, 45.7%), cisgender (n = 34, 97.1%), single (n = 31, 58.6%), childless (n = 32, 91.4%) and predominantly Catholic (n = 13, 38.3%). It can be observed that the average weight is around 71 kg and height is 1.67 m2. The BMI reported an average overweight of 25.9 kg/m2. The waist-height ratio was presented as an important marker (n = 22, 62.8%). Regarding the etiological factors of nursing diagnosis, in the IG the average was seven factors per person, while in the CG eight factors per person. Table 1 presents the relationships between a comparison of the intervention and control arms for the etiological factors targeted by the clinical trial.

Table 1.

Pre- and post-intervention data from nursing consultations in both groups

Etiological factors for diagnosis 00311		Pre				Post
		Groups				Groups
		IG (n = 22)		CG (n = 13)		IG (n = 22)		CG (n = 13)
		n	%	n	%	n	%	n	%	p-valuea
Sedentary lifestyle	Yes	15	68.18	5	38.46	11	50.00	8	61.54	0.018
	No	7	31.82	8	61.54	11	50.00	5	38.46
Alcohol consumption	Yes	2	9.09	7	53.85	3	13.64	5	38.46	0.012
	No	20	90.91	6	46.15	19	86.36	8	61.54
Dyslipidemia	Yes	1	7.69	0	0.00	10	47.62	2	20.00	-
	No	12	92.31	2	100.00	11	52.38	8	80.00
Insufficient knowledge	Yes	17	77.27	11	84.62	8	36.36	9	69.23	0.009
	No	5	22.73	2	15.38	14	63.64	4	30.77
Excess weight	Yes	14	63.64	7	53.85	13	59.09	6	46.15	0.499
	No	8	36.36	6	46.15	9	40.91	7	53.85
Obesity	Yes	5	22.73	1	7.69	4	18.18	2	15.38	0.444
	No	17	77.27	12	92.31	18	81.82	11	84.62
Smoking	Yes	2	9.09	1	7.69	2	9.09	1	7.69	0.887
	No	20	90.91	12	92.31	20	90.91	12	92.31
Unhealthy diet	Yes	12	54.55	10	76.92	6	27.27	9	69.23	0.035
	No	10	45.45	3	23.08	16	72.73	4	30.77
Anxiety	Yes	21	95.45	12	92.31	19	86.36	11	84.62	0.712
	No	1	4.55	1	7.69	3	13.64	2	15.38
Stress	Yes	16	72.73	10	76.92	8	36.36	8	61.54	0.009
	No	6	27.27	3	23.08	14	63.64	5	38.46
Diabetes	Yes	-	-	-	-	-	-	-	-	-
	No	13	100.00	2	100.00	21	100.00	10	100.00
Illicit drug use	Yes	13	59.09	10	76.92	9	40.91	9	69.23	0.122
	No	9	40.91	3	23.08	13	59.09	4	30.77
Depression	Yes	9	40.91	10	83.33	6	27.27	6	46.15	0.051
	No	13	59.09	2	16.67	16	72.73	7	53.85
Metabolic syndrome	Yes	1	7.69	1	50.00	4	19.05	2	16.67
	No	12	92.31	1	50.00	17	80.95	10	83.33	0.302

Note: IG – Intervention group, CG – Control group. ^aANOVA

The data on etiological factors from the RCT served as the basis for constructing logistic regression models, thus allowing us to predict and explain values of a binary variable based on known values of other variables, which may or may not be categorical. Thus, we modeled several phenomena in the study, namely: sedentary lifestyle, harmful alcohol consumption, understanding of cardiovascular risk factors, overweight, obesity, smoking, unhealthy eating, anxiety stress, illicit drug use, depression, and metabolic syndrome. The dependent variables were binary outcomes, representing the presence or absence of each etiological factor. Accordingly, in the model, P(Y) denotes the probability that the patient presents the outcome (“Yes”). To estimate this probability, sociodemographic and clinical variables were used as predictors.

The objective of the logistic regression models was to estimate the probability of the presence of each etiological factor based on selected predictors, given specific patient characteristics. Potential predictors were initially screened through bivariate analyses, and variables demonstrating statistical association were subsequently considered for inclusion in the regression models. Table 2 presents the regression analyses with prediction models adjusted according to the best predictability/outcome for etiological factors of the diagnosis Risk for impaired cardiovascular function (00311).

Table 2.

Estimated coefficients for the etiological factors of the nursing diagnosis Risk for impaired cardiovascular function (00311)

Predictive variables	Independent variable	OR a	CI b	p-value c	Observed model
Understanding of cardiovascular risk factors	Waist circumference (no risk)	0.038	(0.002–0.735)	0.030	86.9%
	Neck-to-thigh ratio (no risk)	0.177	(0.030–1.042)	0.056
	History of COVID-19 (no)	0.016	(0.001–0.253)	0.003
	LDL cholesterol	0.460	(0.187–1.131)	-
	HDL cholesterol	0.238	(0.068–0.828)	0.024
Metabolic syndrome	Sex (male)	0.051	(0.002–1.130)	-	93.2%
	HDL cholesterol	0.107	(0.011–1.083)	0.058
	Systolic blood pressure	7.540	(1.958-110.961)	-
Illicit drug use	Fasting blood glucose	2.278	(1.101–4.712)	0.026	75.4%
	Weight	0.506	(0.260–0.982)	0.044
	Coffee consumption (no)	0.355	(0.104–1.212)	-
Unhealthy eating	History of COVID-19 (no)	0.203	(0.044–0.946)	0.042	82.0%
	VLDL cholesterol	4.774	(1.539–14.804)	0.007
	Fasting blood glucose	2.030	(0.876–4.705)
Sedentary lifestyle	Family history of CVD (no)	0.210	(0.045–0.970)	0.046c	70.5%
	Height	0.502	(0.250–1.007)	0.052
	Diastolic blood pressure	2.588	(1.200–5.580)	0.015
Harmful alcohol consumption	Fasting blood glucose	6.904	(2.088–22.833)	0.002	88.5%
Stress	VLDL cholesterol	0.518	(0.283–0.948)	0.033	67.2%

^a Odds Ratio; ^b CI: Confidence Interval; ^c p-value of the predictor variables from the logistic regression models with the best predictive performance for the outcome under analysis; values above 0.05 were not included in the table

Regression models for overweight, obesity, smoking, and anxiety did not converge and were therefore excluded from further analysis. For the sedentary lifestyle outcome, family history of cardiovascular disease (no), height, and systolic blood pressure were identified as the variables that best explained the probability of a “Yes” response. Sedentary behavior may be associated with anthropometric variables such as body weight and height, as well as with the absence of a family history of CVD.

Additionally, higher SBP was associated with an increased likelihood of sedentary lifestyle, particularly when considered alongside height and absence of family history of CVD, suggesting a combined effect of these predictors on the outcome.

The estimated odds ratios indicate associations between the presence of the outcome and specific predictors, including: sedentary lifestyle and systolic blood pressure; harmful alcohol consumption and fasting blood glucose; unhealthy diet and both VLDL and fasting blood glucose; illicit drug use and fasting blood glucose; and metabolic syndrome and SBP.

The exploratory models also suggested potential associations between metabolic syndrome and variables such as sex, HDL cholesterol levels, and SBP. Additionally, understanding of cardiovascular risk factors was associated with waist circumference (no risk), neck-to-thigh ratio (no risk), absence of prior COVID-19 infection, LDL cholesterol, and HDL cholesterol.

As shown in Table 2, model performance was assessed by comparing predicted values with those observed in the empirical data. The results indicate that the model—including the predictors family history of cardiovascular disease (no), height, and diastolic blood pressure—correctly classified “Yes” or “No” outcomes in 70.5% of cases.

Regarding the TEORISC framework, the significant findings align with its theoretical propositions related to psychosocial, cultural, behavioral, and affective factors. These results reinforce the need for targeted, nurse-led interventions for this population.

Discussion

This research enabled the development of models for etiological factors of the nursing diagnosis of impaired cardiovascular function risk (00311) based on a clinical trial. A study that evaluated 21 regions of the world highlighted the global variation in estimating and predicting cardiovascular risk, with South America being an important development hub for CVD, up to 20% concerning the globe [17], reinforcing the need for more studies like this, especially using theoretical foundations such as TEORISC.

An important cohort study, ATTICA, corroborates the data of the present study, highlighting that risk factors are multivariate and their burden is associated with lifestyle, with sex and age increasing the chances of developing a cardiac event by up to 56%. Added to the non-modifiable factors are hypercholesterolemia, hypertension, diabetes, and increased abdominal circumference leading to heart disease within up to 20 years [18], necessitating that the population be monitored by care programs that can reduce these indicators in the medium and long term.

Family history is associated with dyslipidemia, which in turn is linked to smoking, both cardiovascular risk factors (OR 1.18, 95% CI 1.02–1.36) [19]. In patients whose relatives have atherosclerotic disease, there is a possibility of developing a heart attack, stroke, or cardiac death within one year after age 55 for men and 65 for women [HR, 1.22; 95% CI, 1.05–1.42] [20]. In this study, family history was associated with sedentary lifestyle, risk factors, and conditions, as presented in TEORISC.

According to researchers, hypertension, obesity, and sedentary lifestyle present the greatest potential risks for coronary artery disease [21]. People diagnosed with heart disease in Brazil have sedentary lifestyle as one of the main risk factors (OR = 1.5; 95% CI: 1.02–2.1) [22]. This exploratory research suggests that a sedentary lifestyle may be associated with elevated diastolic blood pressure. In controlled situations, interrupting a sedentary lifestyle helps reduce overweight and influence cardiovascular function, which can be carried out with individuals diagnosed with impaired cardiovascular function risk accompanied by the e-TEORISC software in primary care.

Insufficient knowledge is a negative predictive marker and one of the associated risk factors according to TEORISC [8] and can be strengthened through a nursing-led care program. A systematic review on the knowledge of cardiovascular risk factors demonstrated the public’s inability to identify clinical symptoms, with up to 98.2% only knowing one risk factor [23]. The perception of risk factors is crucial for devising prevention and control strategies; without self-awareness, effective treatment is unfeasible [24], reinforcing the power of studies focusing on cardiovascular health education.

In this study, university students were overweight, and unhealthy diet relates to predictors of the nursing diagnosis 00311. Being seated for long periods and being overweight and obese should be targeted by health practices to reduce cardiometabolic risk [25]. Generally, being overweight is associated with dietary patterns. Global studies on the prevalence of unhealthy habits demonstrate that 255 million years of life are lost due to disability in youth because of poor dietary patterns [26].

The utilization of illicit substances constitutes a concerning element not only among university students but also across diverse healthcare settings. In individuals with cardiovascular diseases (CVD), the use of illicit drugs such as marijuana (53.14%), methamphetamine/cocaine/heroin (23.82%), and injectable drug use (4.67%), both with a significance level of p < 0.0001, represent potential factors for acute events [27]. Within this study, when tobacco use or illicit drug consumption was reported, participants primarily indicated the use of marijuana. Tobacco use is a behavioral factor integrated into the TEORISC framework [8], possessing the potential for reduction through the implementation of preventive measures such as clinical programs.

Stress is prevalent among university students and, in this research, demonstrates a correlation with VLDL levels. In a study involving 400 health science university students, 23.4% exhibited severe or extremely severe scores according to the Depression, Anxiety and Stress Scale-21, while 42.37% presented moderate symptoms, particularly within the psychological domain [28]. Another study involving 500 Pakistani university students using the same scale revealed that 75% had depression, with 35.8% experiencing moderate and 14.6% severe levels [29]. Within the TEORISC framework, these factors are categorized under psychosocial, cultural, and affective factors, guiding care in health promotion categories, as implemented in the program under examination [8].

Metabolic syndrome emerged as etiological factor in the nursing diagnosis 00311. A study conducted with 5,026 individuals in Canada corroborates the data from this research, wherein metabolic syndrome was associated with 42.1% (95% CI 40.7–43.5) of the population, with women being more affected [30]. Notably, a cumulative relationship was observed between hip circumference and an increased likelihood of metabolic syndrome (OR = 4.69; 95% CI 3.45–6.36 in women and OR = 8.25; 95% CI 5.38–12.64 in men) [30]. The TEORISC framework [8] underscores that metabolic syndrome is the central phenomenon within the care context of individuals at cardiovascular risk.

The study indicates that nursing consultations enable the development of exploratory models for eight risk factors: sedentary lifestyle, alcohol consumption, insufficient knowledge, unhealthy diet, stress, drug use, depression, and metabolic syndrome. However, the data should be interpreted with caution, particularly regarding inverse risk associations for blood pressure, glucose, and VLDL levels. For the intervention group, the e-TEORISC tool facilitated a data collection process that may offer new approaches for robust models, potentially surpassing the exploratory ones described here for cardiovascular risk based on the diagnosis of Risk for impaired cardiovascular function (00311).

Study limitations

The data described in this study should be interpreted with caution and lack broad generalizability, as they were estimated from a specific population of university students with cardiovascular risk factors. It is important to emphasize that the sample size is only sufficient for an exploratory pilot study. The research faced a high attrition rate due to the specific characteristics of the target population regarding follow-up interventions and the low prevalence of certain cardiovascular risk factors. Furthermore, there was no explicit validation of the models due to the limited sample size, which may have introduced potential selection bias. For future research, validation should ideally be performed using sample splitting. Additionally, as this is an exploratory study, the data are subject to unstable regression coefficients; odds ratios were organized according to data behavior, which may be associated with overestimation or sampling bias. Analytically, the models carry a risk of overfitting and may be unable to provide more accurate measures given the small initial data source.

Contributions to the field of nursing and health

The findings of this study provide preliminary insights and exploratory correlation models based on a pilot RCT. It is important to note that the associations described herein can guide nursing interventions in larger populations, which will enable more robust correlations in future studies. Consequently, new approaches should be implemented through nursing-led interventions based on a program utilizing TEORISC and nursing classification systems. Specifically, these should focus on the NANDA-I diagnosis of Risk for impaired cardiovascular function (00311) within primary care settings and among populations underserved by health services. The e-TEORISC is considered to have innovative potential to address the vulnerabilities and needs of the population, with the possibility of reducing associated cardiovascular risk factors and conditions identified by TEORISC in the medium and long term. This demands more dissemination, investment, and strengthening to achieve technological transfer to the healthcare system.

Conclusion

The study presents exploratory models estimating the presence of etiological factors of the nursing diagnosis Risk for impaired cardiovascular function (00311) and clinical associations with negative repercussions on the cardiovascular health of university students, particularly regarding insufficient knowledge, unhealthy diet, and depression in the control group, as well as sedentary behavior, drug use, and metabolic syndrome in both the intervention and control groups. These can be reversed through nurse-led actions with precise care plans. The study proposes an innovative alternative in the field of health and nursing care through the implementation of a cardiovascular risk care program based on TEORISC, e-TEORISC, and the nursing diagnosis 00311, which allows for the advancement of the discipline and new studies in the area.

Author contributions

Conception and design or analysis and interpretation of data: J.C.N, T.F.C, N.D.C.F. Drafting of the manuscript or relevant critical revision of the intellectual content: J.C.N, T.F.C, N.D.C.F., M.V.O.L. Final approval of the version to be published: J.C.N, T.F.C, N.D.C.F., M.V.O.L.

Funding

No funding was provided.

Data availability

The datasets generated and analysed during the current study are not publicly available due to prior commitments to not openly share the data, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the research ethics committee of the University Of International Integration Of Afro-Brazilian Lusophony (UNILAB, approval number 6.092.932) and developed according to the ethical principles with the Declaration of Helsinki. Written and verbal informed consent has been obtained from the client.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.