Taking diabetes risk screening to the community: Feasibility and acceptability of the FINDRISC tool in vulnerable populations in Argentina

ABSTRACT

Background:

Type 2 diabetes mellitus (T2DM) is increasing in Argentina, with an adult prevalence of 14%. Early detection through risk assessment tools like the Finnish Diabetes Risk Score (FINDRISC) is recommended, but community-level implementation remains limited.

Methods:

A mixed-methods study was conducted in 40 primary care centers across San Juan, Salta, and Tandil. Adults (>18 years) without diabetes were screened using a tablet-based FINDRISC questionnaire. Quantitative data included risk scores and sociodemographic variables; qualitative data explored barriers and facilitators through interviews with healthcare providers and focus groups with patients.

Results:

Of 11,182 participants, 88.3% completed the questionnaire and 52.3% were at moderate or high risk of T2DM (FINDRISC ≥12). Completion rates were higher in primary care centers (93.3%) than in home visits (77.5%). High rates of obesity (up to 57.6%) and abdominal obesity (up to 75.5%) were observed. Key facilitators included the tool’s simplicity and community health worker (CHW) involvement, while barriers involved staff workload, initial resistance, and logistical challenges. Patients reported increased awareness and some lifestyle changes but faced economic and access constraints.

Conclusion:

FINDRISC screening is feasible and acceptable in diverse Argentine communities. Success is enhanced by CHW engagement, institutional support, and adaptation to local contexts. These findings support broader implementation of diabetes risk screening to inform prevention strategies in similar settings.

Background

With its growing global prevalence and disproportionate impact in low- and middle-income countries (LMICs), diabetes is a major public health challenge. Type 2 diabetes mellitus (T2DM) has risen sharply, driving a 5% increase in premature mortality and 1.6 million deaths worldwide in 2021.[1] In Latin America and the Caribbean (LAC), prevalence among adults aged 20–79 grew from 7.4% in 2010 to 9.7% in 2019.[2] Argentina shows an even steeper rise, from 8.4% in 2005 to 12.7% in 2018;[3] the 2024 IDF Diabetes Atlas estimates 4.3 million adults (14.0%) are living with diabetes, projected to reach 15.4% by 2050.[4]

Preventive strategies are urgently needed. Evidence shows that intensive lifestyle interventions reduce progression from prediabetes to T2DM,[5,6] and early detection is cost-effective compared to no screening.[7,8] Risk assessment tools, particularly the Finnish Diabetes Risk Score (FINDRISC), provide a practical approach to identifying individuals at high risk.[9,10] Validated across LMICs,[1,11,12] FINDRISC has been adapted for Latin America using waist circumference cut-offs (90 cm women, 94 cm men) better correlated with insulin resistance than ATP III criteria.[13,14,15,16]

Argentina’s national guidelines incorporate the FINDRISC as a useful tool for identifying individuals at increased risk of developing type 2 diabetes,[17] but uptake is still limited due to integration and awareness challenges. This approach offers accessible and noninvasive screening for people in the community, allowing them to know their risk of developing diabetes, creating a window of opportunity for care, education, and early action for diabetes prevention. We seek to provide evidence on how to integrate community screening for diabetes risk into existing primary care processes, supporting healthcare professionals in prevention and early detection. The aim of this study is to assess the feasibility and acceptability of implementing the FINDRISC tool as a community- and primary care-based strategy for type 2 diabetes risk screening in vulnerable populations in Argentina.

Methods

As part of the component of screening to detect high risk of developing diabetes in the Diabetes Prevention and Care Program, we used a mixed-methods approach to analyze barriers and facilitators of the implementation of a screening strategy to evaluate the risk of developing diabetes using the FINDRISC tool.

Population

We included adults (>18 years) without a prior diabetes diagnosis or current pregnancy.

Study setting

The study included 40 primary care centers (PCCs) and their catchment area across three districts representing Argentina’s diversity: San Juan (17 PCCs), Salta (14 PCCs), and Tandil (9 PCCs).

FINDRISC questionnaire

The FINDRISC questionnaire was administered using a tablet application that automatically calculated the total risk score for developing type 2 diabetes. The instrument includes eight variables: age, body mass index (BMI), waist circumference, physical activity, daily fruit and vegetable consumption, use of antihypertensive medication, history of high blood glucose, and family history of diabetes, in addition to the patient’s personal and sociodemographic data. According to the original classification, a score ≥12 indicates a moderate risk and ≥14 indicates a high risk of developing diabetes. In the present study, participants with a score ≥12 were considered at moderate/high risk and were included in the program.

Quantitative approach

Primary healthcare teams, including nurses and community health workers (CHWs), administered the FINDRISC tool both at the PCCs and in the community. Data collection was carried out in two ways, depending on the location of participant recruitment: CHWs administered the personal and family history questionnaire and performed anthropometric measurements at home, while nursing staff administered the questionnaire at PCC, following standardized procedures. All participants were assessed using tablets with a customized application that incorporated the FINDRISC questionnaire as well as a form with supplementary sociodemographic questions.

Qualitative approach

The Consolidated Framework for Implementation Research (CFIR) was used to assess contextual factors influencing implementation.[18] For data collection, in-depth interviews were conducted with healthcare providers and decision makers to explore their perceptions and experiences related to barriers and facilitators to program implementation and diabetes risk screening and prevention in their populations. Additionally, we invited patients with a high or moderate risk of developing diabetes in the next 10 years to participate in the focus group discussions as members of the community, during which the findings from the quantitative phase were presented and discussed to identify barriers and facilitators to risk detection and disease prevention. Interview and focus group sessions were audio recorded after written informed consent was obtained. Audio recordings were provided to an independent transcriber who assigned numeric identity codes to individuals for transcription. No other identifying information, including the identities of focus group discussion leaders or interviewers, was made available to the transcriber or recorded during the interviews or focus groups.

Data analysis

For the quantitative analysis, data collected using the digital application of the FINDRISC questionnaire were used. For descriptive statistics, absolute and relative frequencies were calculated for categorical variables, and measures of central tendency and dispersion (mean, standard deviation [SD]) were calculated for continuous variables. The t-test was used for continuous variables and the Chi-square test for categorical variables. The proportions of individuals in the different risk categories (low, moderate, and high, based on a FINDRISC score ≥12) were also estimated. Analyses were performed using Stata V.14 software.[19]

For the qualitative analysis, interviews and focus groups were transcribed and analyzed using a thematic analysis approach, following the dimensions proposed by the CFIR framework. Data matrices were constructed to identify patterns, facilitators, and barriers to strategy implementation. The analysis was performed using Atlas Ti.[20]

Ethical considerations

The program received approval from three institutional ethics committees in Argentina: the Institutional Review Board for Research Studies (Consejo Institucional de Revisión de Estudios de Investigación) of the Hospital Privado de Comunidad in Mar del Plata, Buenos Aires Province; the Research Ethics Committee of the Ministry of Public Health of San Juan; and the Provincial Commission for Biomedical Research of the Ministry of Public Health of Salta. Participation was voluntary, and written informed consent was obtained from all individuals before screening. Participant confidentiality was strictly safeguarded throughout the study.

Results

Quantitative results

A total of 11,182 adults were evaluated using the FINDRISC tool at the three study sites. Of the total number of people evaluated, 88.3% (9,879/11,182) completed all the questionnaire items. The sociodemographic characteristics of this population are presented in Table 1, which describes the distribution by age, sex, educational level, and type of health coverage.

Table 1.

Sociodemographic characteristics of respondents with complete vs. incomplete FINDRISC data

Characteristic	Complete FINDRISC n=9879	Incomplete FINDRISC n=1303	P
Mean age years (SD)	41.5 (14.9)	42.8 (15.4)	0.00221
≥65 years % (n)	7.65 (756/9,879)	10.05 (131/1,303)	0.0032
Female % (n)	76.19 (7,527/9,879)	74.90 (976/1,303)	0.3052
Educational level % (n)
Incomplete primary education or less	5.12 (506/9,879)	5.15 (65/1,263)	0.8372
Completed primary education	24.98 (2,468/9,879)	22.01 (278/1,263)	0.0042
Incomplete secondary education	25.12 (2,482/9,879)	27.08 (342/1,263)	0.3802
Completed secondary education	31.58 (3,120/9,879)	33.41 (422/1,263)	0.5572
Tertiary and/or university education	10.15 (1,003/9,879)	12.35 (156/1,263)	0.0432
Exclusive public health coverage % (n)	76.29 (7,537/9,879)	70.84 (923/1,303)	0.0002

SD: Standard deviation, FINDRISC: Finnish Diabetes Risk Score. *¹t-test, ²Chi²

Among the individuals evaluated, 52.3% (5,977/11,088) were found to have a moderate or high risk of developing diabetes (FINDRISC score ≥12). This total includes both those who completed the entire questionnaire and those who, without having answered all the items, achieved a score equal to or higher than 12, allowing them to be classified as high risk.

Comparable sociodemographic profiles were observed among participants who completed the FINDRISC questionnaire across the three study sites. The mean age was 42.0 years (SD: 14.8) in San Juan, 41.9 years (SD: 14.9) in Tandil, and 40.3 years (SD: 15.0) in Salta. People aged 65 and above accounted for 7.8% (425/5,450) in San Juan, 7.4% (98/1,319) in Tandil, and 7.5% (233/3,110) in Salta. The majority of participants were female: 76.9% (4,193/5,450) in San Juan, 79.4% (1,047/1,319) in Tandil, and 73.5% (2,287/3,110) in Salta. Regarding educational level, the largest proportion of participants in all three sites had either incomplete or completed secondary education. As for health coverage, 78.4% (4,270/5,450) of the people evaluated in San Juan relied exclusively on public health coverage, followed by Salta with 74.7% (2,322/3,110) and Tandil with 71.7% (945/1,319).

Questionnaires were considered incomplete when one or more items of the FINDRISC survey were left unanswered or when at least one of the required clinical measurements was missing. Among the incomplete questionnaires (n = 3,003), the most frequently missing variable was waist circumference, which was available in only 32 cases.

As can be seen in Table 1, the analysis comparing complete (n = 9,879) and incomplete (n = 1,303) questionnaires showed that participants with incomplete questionnaires had a higher proportion of individuals ≥65 years old (10.05% vs. 7.65%, P = 0.003). No significant difference was observed in the proportion of women (74.9% vs. 76.2%, P = 0.305). In terms of educational level, participants with incomplete questionnaires had a slightly lower proportion of completed primary education (22.0% vs. 25.0%, P = 0.004) and a higher proportion with tertiary/university education (12.4% vs. 10.2%, P = 0.043). Participants with complete questionnaires had a higher proportion of public health coverage (76.3% vs. 70.8%, P < 0.001).

We implemented different community screening strategies for diabetes risk detection, including home visits by healthcare workers, assessments in waiting rooms within PCC, and extramural activities in community spaces such as health posts and plazas. The proportion of incomplete FINDRISC questionnaires varied significantly by implementation setting. The highest frequency of incomplete records was observed in home visits 22.46% (743/3,308), followed by hospital screening 14.51% (27/186) and extramural activities 7.98% (130/1,631). In contrast, the lowest proportion corresponded to questionnaires administered in PCC (6.65%, 403/6,056). These results demonstrate that the implementation context influences data completeness, with its application being more consistent in institutional settings, particularly in primary care.

As shown in Figure 1, the proportion of individuals at moderate or high risk of developing diabetes (FINDRISC score ≥12) varied across the study sites. In San Juan, 64.3% (3,566/5,545) of participants fell into this category, followed by Salta with 58.2% (1,868/3,209) and Tandil with 40.7% (543/1,334). These percentages were calculated using as the denominator all participants with a complete FINDRISC assessment, as well as those with incomplete assessments whose available responses already yielded a score ≥12.

Figure 1.

Risk distribution according to FINDRISC by site. *Denominator includes participants with complete FINDRISC assessment plus those with incomplete assessments whose available responses already yielded a score ≥12. FINDRISC: Finnish Diabetes Risk Score

Comorbidities and associated risk factors

Among individuals with completed questionnaires, high prevalence rates of obesity and abdominal obesity were recorded. Obesity based on BMI was 57.6% (3,137/5,450) in San Juan, 55.9% (1,739/3.110) in Salta, and 40.0% (527/1,319) in Tandil. Abdominal obesity, measured by waist circumference, reached 75.5% (4,117/5,450) in San Juan, 72.1% (2,242/3,110) in Salta, and 60.3% (795/1,319) in Tandil. As shown in Figure 2, both general and abdominal obesity were more frequent in San Juan and Salta compared to Tandil. These differences indicate variation in the distribution of obesity indicators across the three provinces.

Figure 2.

Prevalence of obesity by site. BMI: Body mass index, WC: Waist circumference

Regarding other comorbidities, self-reported high blood pressure was found in 30.8% (1,677/5,450) of individuals in San Juan, 25.9% (341/1,319) in Tandil, and 20.4% (633/3,110) in Salta. Self-reported high cholesterol was observed in 26.3% (1,433/5,450) in San Juan, 26.7% (829/3,110) in Salta, and 18.7% (247/1,319) in Tandil.

Healthy lifestyle habits

Daily consumption of fruits and vegetables was reported by 33.1% (437/1,319) of individuals in Tandil, 20.5% (1,115/5,450) in San Juan, and 19.6% (609/3,110) in Salta. As for daily physical activity, it was recorded in 40.6% (535/1,319) of participants in Tandil, 25.5% (1,389/5,450) in San Juan, and 23.6% (733/3,110) in Salta.

Qualitative results

In-depth interviews were conducted with three health system decision makers at different hierarchical levels and ten primary healthcare providers, as well as four focus groups that included a total of 27 people at high or moderate risk of developing type 2 diabetes. The qualitative analysis, guided by the CFIR framework, identified barriers, facilitators, and conditions necessary for the sustainability of the FINDRISC screening in community settings.

Among decision makers and primary healthcare providers, key facilitators included perceived relevance of risk screening, simplicity of the tool, training received, and the use of technology that streamlined the process. These facilitators coexisted with barriers such as the overload of care-related tasks, high staff turnover, and initial resistance from some primary care teams [Table 2].

Table 2.

Focus groups with healthcare providers and decision-makers

CFIR domain	CFIR construct	Brief summary	Verbatim
Intervention characteristics	Design quality and packaging	The questionnaire was perceived as a simple and well-designed tool.	“It’s a tool that wasn’t very well known in the different health centers, but it’s simple, easily accessible, and provides a lot of information.”
Outer setting	Patient needs and resources	Anticipate potential resistance to accepting the strategy (especially regarding the use of the tablet and the request for personal information) to avoid affecting how the intervention is received and, consequently, the quality of the results obtained.	“Well, I’m familiar with this last stage. I thought it was very good, all the staff, all the researchers, who are attentive, who report every month, so that way we can also help from our place, to identify where the difficulties lie. So, the ongoing communication really. they involved people, making them protagonists, the people who collect data, who gather information, training them in communication, especially with people, because the issue of convincing someone not to be afraid to sign a consent form, I think that can be achieved with good communication.”
Inner setting	Implementation climate	The high priority given to clinical care compared to screening activities made it difficult to sustain implementation.	“As often happens in public health, priority is given to clinical care, and then others start to neglect their responsibilities.”
Characteristics of individuals	Knowledge and beliefs about the intervention	Community health workers faced challenges using technology and had to adapt gradually.	“Many of our CHWs are old; some had never used a tablet. That was a challenge, but I think we managed it well.”
Process	Engagement and participation	Initial resistance turned into acceptance over time and with practice.	“Yes, there was some initial resistance because it was an additional activity being introduced, but as days went by, it started to improve.”

CFIR: Consolidated framework for implementation research, CHW: Community health worker

In the patient focus groups, participants appreciated the close support from the healthcare team, particularly from CHWs, who encouraged participation in the screening process. Participants reported increased awareness of their risk and indicated changes in dietary and physical activity habits after being informed of their risk status. Nevertheless, they faced economic constraints, time limitations, difficulties in accessing appointments, and challenges related to trust when signing the informed consent and with the use of digital devices. Additionally, emotional support and the need to strengthen mental health emerged as relevant and/or emerging themes. Several participants emphasized psychological assistance as part of patient care, highlighting its role in helping patients adopt and maintain healthy changes in their habits. The emotional support provided by primary care staff and its positive impact on health status were also valued [Table 3].

Table 3.

Focus groups with participants with high risk of diabetes

CFIR domain	CFIR construct	Brief summary	Verbatim
Intervention characteristics	Relative advantage	The program helped to understand the risk and the importance of preventing diabetes.	“The intervention has helped increase awareness of the screening and also the disease and its consequences.”
Outer setting	Patient needs and resources	Difficulties in obtaining clinic appointments were partially mitigated by staff assistance.	“Sometimes it’s hard to get an appointment, but Community Health Workers help us.”
Inner setting	Networks and communication	The close relationship with the staff was an important facilitator.	“The support we receive here feels very much like family… we talk with the nurses, and they provide comfort.”
Characteristics of individuals	Individual stage of change	Some patients reported making significant changes in eating and physical activity habits.	“I changed a lot… I stopped eating certain things, started drinking water, eating fish. I feel really good, I feel light.”
Process	Reflection and evaluation	There is a need to receive practical guidelines on how to implement recommendations.	“While it is important to have information, it is also important to know how to apply it.”

CFIR: Consolidated framework for implementation research

Discussion

We implemented the FINDRISC tool in adults over 18 years old using a tablet app that automatically calculated risk scores. This digital tool streamlined the process, standardized data collection, reduced errors, and improved analysis efficiency. More than 88% of participants completed the questionnaire, and more than 50% of those assessed presented a moderate or high risk of developing type 2 diabetes. These results are consistent with the findings of previous studies and the 4^th National Risk Factor Survey.[3] In addition, the findings of this study are consistent with those reported by another study, which administered the FINDRISC questionnaire in a vulnerable community in northern Argentina.[21] They found that 44.8% of participants were at moderate to high risk of developing type 2 diabetes. In that study, 38.4% of participants were overweight and 30.9% were obese, with a mean BMI of 28.1 kg/m², and almost half had abdominal obesity. These results are comparable to those observed in our sample across all provinces. In contrast, another study conducted in Latin America showed that FINDRISC screening in PCCs in Paraguay reported a lower prevalence of moderate to very high risk (30.9%).[22] These differences could be explained by variations in the populations assessed, socioeconomic conditions, and local health system contexts. These studies reinforce the usefulness of FINDRISC for detecting diabetes risk in community and primary care settings in Latin America, aligning with previous calls for structured screening approaches in primary healthcare systems.[23]

Incomplete questionnaires were more common among older adults and those with private health coverage, suggesting that age and health system choice affect survey completion. Older age may limit participation due to mobility issues, while private coverage may reduce attendance at PCCs.

Most respondents were women (76.2%), likely because home visits occurred during working hours when women were more often present. This reflects women’s greater health awareness, also reported in the 4^th National Risk Factor Survey.[3] The setting influenced completion rates: home visits had the highest proportion of incomplete forms (22.5%), while PCC screenings had the lowest (6.7%). Thus, household visits improve reach but complicate data collection, whereas institutional settings ensure higher quality.

Participants valued the questionnaire’s simplicity and the role of community health workers as mediators between PCCs and the community, which facilitated acceptance. However, barriers included staff overload, limited awareness, and initial resistance to adopting the tool. Similar challenges have been reported in Argentina and Spain, where time constraints in primary care hindered integration.[24,25]

Beyond confirming previous findings, this study shows the feasibility of standardized diabetes risk screening in diverse real-world contexts in Argentina. Combining community outreach with institutional strategies may optimize both reach and data quality. These lessons are key to informing national diabetes prevention policies and replication in Latin America and beyond.

Limitations

This study has some limitations. First, it was conducted in three Argentine provinces with their own organizational, cultural, and epidemiological contexts, which may limit generalizability. However, the selected sites vary widely in geography, socioeconomic profile, and healthcare delivery practices, which enhances the relevance of findings for diverse primary care and community settings in Argentina and potentially other LMICs. Second, interviews and focus groups may have been influenced by the presence of the research team or participants’ desire to present their work or engagement positively. We minimized this risk through the use of independent interviewers, anonymized transcripts, and triangulation of qualitative findings with quantitative implementation data. Third, implementation relied on a tablet-based application, which may face challenges in areas with low digital literacy, limited connectivity, or device shortages. However, training was provided to CHWs and nurses, demonstrating that the approach can be adapted to low-resource contexts.

Strengths

The study was conducted under real-world primary care and community outreach conditions rather than in a controlled research environment, increasing the applicability of results to policy and practice. Additionally, the mixed-methods approach, combining quantitative and qualitative data, allows for a deeper understanding of feasibility, acceptability, and contextual factors influencing implementation. Of note, the large and diverse sample of over 11,000 people evaluated across three regions of the country with different sociodemographic characteristics provides robustness and relevance to the findings for various primary care and community contexts. The high completion rate of questionnaires and the active participation of key healthcare team members, such as community health workers, demonstrate the acceptability and feasibility of incorporating risk assessment through simple tools like FINDRISC in real primary care settings.

Conclusion

The implementation of the FINDRISC questionnaire as a screening tool for type 2 diabetes risk detection in primary care settings was accepted by both the healthcare team and the patients. This was mainly due to the simplicity of the instrument. The active participation of community health workers played a fundamental role as a link between the community and health centers, facilitating the application of the questionnaire. However, some barriers persist, mainly related to the workload of the staff and the consequent initial resistance to incorporating new practices within the tasks of the healthcare teams. This can be addressed through raising awareness among teams to ensure sustainable implementation. The results highlight the importance of adapting strategies to local particularities and strengthening institutional support to facilitate integration into routine practice. This study provides relevant evidence to guide future diabetes prevention initiatives in similar populations and primary care contexts.

Author contributions

Concept and design: VI, MVL, AN. Definition of intellectual content: VI, MVL. Literature search: OS, CMC. Data acquisition: AN, CMC, MVL. Data analysis/statistical analysis: LG, AC. Manuscript preparation, editing, and review: MVL, AN, CMC, AC, LG, VI. Guarantor (s): MVL.

Conflict of interest

The authors declare that they have no conflicts of interest related to this work.

Funding Statement

This project was funded by the World Diabetes Foundation (WDF), project ID WDF19-1724.