Introduction

Developmental Coordination Disorder (DCD) is a specific condition classified as a neurodevelopmental disorder according to the fifth edition, text revision of the Diagnostic and Statistical Manual of Mental Disorders – DSM-5 TR1. The DSM-5 TR outlines four criteria for diagnosing DCD as follows: (A) motor coordination below what is expected for age and opportunities around; (B) compromised motor skills execution in daily, academic and leisure activities; (C) persistent lacking in motor skills since early childhood; (D) poor motor skills not attributed to physical, visual or intellectual disabilities, or caused by cerebral palsy or other specific medical condition1. DCD is a prevalent disorder, affecting approximately 6% of school-age children, and it impacts various domains, including physical, environmental, social and emotional development2. Despite its prevalence, DCD can often go unnoticed throughout childhood, adolescence, or even into adulthood3, as it is considered a less obvious disorder when compared to the more noticeable neurodevelopmental disorders, such as Autism Spectrum Disorder (ASD)4,5 or Attention Deficit Hyperactivity Disorder (ADHD)6,7.

Neglecting motor difficulties during childhood may lead to negative consequences, as children with DCD tend to avoid sports or physical activities8, due to the exposure of their motor coordination difficulties. This avoidance can lead to reduced levels of physical activity9 and lower fitness10 as well as increased sedentary behaviour11,12, which may continue to impact into adulthood13.

The persistence of DCD beyond childhood into adulthood ranges from 30% to 70%, with implications for various areas of life14,15,16, particularly as individuals with DCD are challenged by their motor difficulties when faced with new tasks over time16,17. It is hypothesized that engaging in more complex physical activities remains an ongoing challenge for individuals with DCD. Moreover, as they grow older, individuals with DCD tend to become more aware of their motor difficulties18, which may further contribute to increased sedentary behaviour over time. Previous studies also described the reduced physical activity hypothesis and the bad cycle of reduced physical activity in children19 and adolescents with DCD20.

However, the relationship between DCD, physical activity, and sedentary behaviour remains unclear, particularly when considering sociodemographic and health-related factors, as well as the scarcity of studies involving adults.

Understanding the role of sociodemographic and health-related factors in physical activity and sedentary behaviour is important due to the significant variation in health behaviours across age groups, sexes, or socioeconomic statuses21,22,23. Evidence shows that the amount of physical activity and time spent sitting can differ among individuals based on these factors23. Due to social inequities (e.g., limited access to goods and services based on race, sex or socioeconomic status), some population groups have more or fewer opportunities or motivations to be active24. A recent systematic review25 summarized that the main barriers reported by university students to being physically active included insufficient time, low motivation, and a limited number of accessible facilities. These challenges should be considering when monitoring physical activity and sedentary behaviour, particularly in chronic and heterogeneous conditions such as DCD, as it can be an additional barrier to participate in physical activity programs. In this context, exploring the role of sociodemographic and health-related factors on physical activity and sedentary behaviour among adults of DCD can help clarify how public policies might support this population in improving participation and overall health outcomes. Although scarce, the literature on DCD in adults has provided valuable insights into how individuals cope with their motor coordination challenges when performing everyday tasks in adult life or engaging in physical activities These well-documented deficits in adults with DCD suggest that they face greater challenges and difficulties both in completing their university studies and in holding jobs that require high motor skills demands26. Therefore, since there is a relationship between psychological distress, which is common in adults with DCD, and psychosocial factors27,28, a more nuanced understanding of how DCD traits across developmental periods can predict negatively or positively levels of physical activity and sedentary behaviour in adults, controlling by sociodemographic and health-related factors is warranted.

The rationale behind this proposed analysis is based on the fact that sex and age are both well-established determinants of physical activity and sedentary behaviour29,30. For example, males tend to report higher levels of physical activity than females, and activity patterns can vary significantly across age groups, even within young adults31. Furthermore, other sociodemographic factors have been investigated among university students as potential mediators of physical activity and sedentary behaviour32,33,34 For instance, academic discipline may shape daily routines and opportunities for physical activity (e.g., students in health sciences may have greater health literacy or physical education components in their curriculum)35. Marital status may influence time allocation, lifestyle habits, and social support for physical activity36. Financial assistance and employment are related to socioeconomic status, which can affect access to resources (e.g., transportation, gym memberships, time availability, and health behaviours)37,38 Students commuting from other cities may experience time constraints or environmental barriers to activity39. Academic workload or stress may vary across semesters, affecting lifestyle behaviours40. Differences in race/colour is important to account for potential sociocultural differences in physical activity patterns and access to health-promoting resources37.

Additionally, alongside sociodemographic factors, those related to health conditions or health behaviours are also relevant, as physical activity is multifactorial and social inequities may determine how healthy, including how physically active, such population groups are, particularly in low-resources countries41, where DCD is often underdiagnosed42,43. Regarding health-related factors, previous neurodevelopmental diagnosis may co-occur with or confound the expression of DCD traits44,45 and affect physical activity behaviours independently. Medical condition could limit physical activity due to physical limitations, fatigue, or treatment effects46. Tobacco and alcohol use are health behaviours often cluster, and these behaviours may be correlated with lower activity levels or higher sedentary time47.

Previous literature48 has also shown that 68.6% of university students were not physically active, while the prevalence of tobacco smoking and alcohol consumption was approximately 16% and 84%, respectively. Some medications may have side effects (e.g., fatigue, motor impairments) that influence physical activity49. Although DCD does not primarily require the use of controlled medication, common co-occurring conditions such as ADHD or ASD may necessitate specific prescription medications. In addition, emotional comorbidities are frequently observed in adults with DCD28, and pharmacological treatment for these symptoms may also mediate the relationship between physical activity and sedentary behaviour in individuals with this condition. Further worsening this scenario, self-medication was highly prevalent among university students, with rates exceeding 70%50.

Controlling for these sociodemographic and health-related factors allows the researchers to isolate the specific effect of DCD traits on physical activity and sedentary behaviour, minimizing the risk of confounding. Without controlling for them, observed associations could be due to these other variables rather than DCD traits per se. This approach improves the internal validity of the study and strengthens causal inference, especially important when examining longitudinal DCD traits into adulthood. Additionally, because DCD traits can be classified into different subgroups to potentially capture the severity of symptoms, an exploratory sensitivity analysis was also conducted. Similar analyses have been performed in previous studies51,52,53 to better understand complex relationships among variables, particularly those related to human behaviours. As hypothesized, we believe that DCD traits significantly reduce physical activity levels and increase sedentary behaviour in university students, regardless of sociodemographic and health-related variables. Therefore, the aim of this study was to investigate whether longitudinal DCD traits are associated with physical activity and sedentary behaviour in a sample of Brazilian young adult university students, controlling for sociodemographic and health-related variables. Additionally, we tested the severity of DCD in the relationship with physical activity and sedentary behaviour in a sensitivity analysis considering pDCD across developmental periods.

Methods

Study design and sample

This cross-sectional study design was conducted with a convenient sample of 124 university students from a multi-campus State University, in Bahia, Brazil. All methods were performed in accordance with the Declaration of Helsinki, the project was approved by the State University of Bahia Human Research Ethics Committee (n. 6.814.433, CAEE n. 78718024.5.0000.0057) and all the volunteers signed the informed consent form. Participants were recruited by advertisement through social media (Instagram, WhatsApp and Facebook) and email, containing a link and a QR-code to Google forms, where the participants accessed the electronic consent form and questionnaires. After consenting to participate, participants completed three questionnaires used in this study.

The university where this research was conducted is a multicampus public institution that covers approximately the entire state of Bahia. Bahia, located in the Northeast of Brazil, is a very large state, and the university’s logistics are highly multicultural and diverse. Reaching all students in person would have been extremely difficult; for this reason, the questionnaires were administered as an online survey. We made every effort to recruit the maximum number of participants, with the support of coordinators and directors. Recruitment was carried out through social media, emails, and WhatsApp messages using a link or QR code. Flyers containing research information and the QR code were also distributed at the facilities of Campus I, Campus IV, Campus X, and Campus XII of UNEB. These campuses were selected based on the research team’s availability within the data collection schedule. In addition, a snowball sampling strategy was employed to increase participation, whereby students were encouraged to disseminate information about the research through their own social media, WhatsApp, or email contacts. A post hoc power analysis was conducted to evaluate whether the final sample size was adequate for the outcomes investigated. The study included 129 participants, of which five were excluded due to incomplete or duplicate information, resulting in a final sample of 124 individuals. Considering the observed prevalence of each outcome and using a 95% confidence level (α = 0.05), the post hoc power estimates indicated a statistical power of 80% for DCD traits and 99.8% for sedentary behavior. These values demonstrate that the achieved sample size was sufficient to detect meaningful differences for the larger effects. Specifically, for DCD traits, the sample had a good chance (≈ 80%) of detecting large effects ( ~ ≥ 24% points); for sedentary behavior, with the observed large difference (~ 45% points) and the given group sizes, the study had very high power to detect this difference, practically certain detection, assuming the validity of the estimates. Although the sample size was sufficient to detect significant differences, considering the total number of students enrolled in the university’s various courses, we expected to achieve a larger sample size. For this reason, the limited sample should be acknowledged as a limitation of the study.

Measures

The levels of physical activity and sedentary behaviour were collected through the short-form International Physical Activity Questionnaire (IPAQ-short)54, the DCD traits were assessed through the Adult Developmental Coordination Disorder Checklist (ADC)55, while the sociodemographic and health-related factors through a sociodemographic questionnaire developed by the research team.

The IPAQ-short form was used to evaluate the levels of physical activity and sedentary behaviour among participants56. It is considered a valid and cost-effective method to estimate the levels of physical activities in adults57. Levels of physical activity are in accordance of activities performed over the last seven days and classified in vigorous-intensity, moderate-intensity and walking54. The last part of the IPAQ-short assessed the sitting time during week and weekend days. The online version of the questionnaire was attested as reliable (rho > 0.30) to estimate physical activity in university students in Brazil58.

Levels of physical activity were categorized into sufficiently active (> 1.368 MET-minutes/week) and insufficiently active (≤ 1.368 MET-minutes/week), based on the median value of the metabolic equivalents (METs), corresponding to 1.368 MET-minutes/week. This value is in accordance to the international guidelines for levels of physical activity for adults59 Sedentary behaviour was calculated through the following formula: [(sitting time in a week day x 5 + sitting time in a weekend day x 2)/7]60 and considered elevated from the 75th percentile60, corresponding to 2.2857 median value.

The ADC is a widely used instrument to identify DCD traits in adults2. It has demonstrated strong internal reliability, with alpha coefficients ranging from 0.87 to 0.95, and moderate, significant concurrent validity, with correlations ranging from 0.68 to 0.75 with the Handwriting Proficiency Screening Questionnaire (HPSQ)55. The checklist is composed of 40 items, divided into two sections. The first section comprises 10 questions about the difficulties during childhood, while the second section comprises 30 questions about the current difficulties in adult life. Each question is scored into a four-point Likert scale, considering the frequency of symptoms as follows: (0) Never; (1) Sometimes; (2) Frequently; and (3) Always3. In this study, the scores were summarized for each section as well as for the total score and participants were classified as “Probable DCD”, “at risk for DCD”, or “typically developing”61. During childhood (Sect.  1) scores ≥ 10 indicated “At risk for DCD” and scores ≥ 17 “Probable DCD”. Current adult life (Sect.  2) scores ≥ 39 indicated “At risk for DCD” and scores ≥ 48 “Probable DCD”. Total (Sect.  1 + Sect.  2) scores ≥ 56 indicated “At risk for DCD” and scores ≥ 65 “Probable DCD”. Scores under the thresholds classified typically developing participants61.

The sociodemographic questionnaire was a 14-item standard instrument developed by the researchers within the laboratory coordinated by the first author. The sociodemographic factors considered the personal, occupational, and demographic information, as follows: Sex (Male/Female); Age (years); Field of the course (Heath, Human or Exact Sciences); Marital Status (Married/Not married); If they received any financial assistance from the University (Yes/No); If they had any type of employment status (Yes/No); If they lived in the same city of the campus (Yes/No); Ongoing semester (Corresponding semester period in numbers); Self-reported race/colour (White/Black/Brown/Indigenous/Yellow/Not reported). The health-related factors considered health behaviours and clinical information, as follows: If they had previous neurodevelopmental disorders diagnosis (Yes/No); If they had previous medical condition diagnosis (Yes/No); If they consumed tobacco (Yes/No); If they consumed alcohol (Yes/No); and Controlled Medication use (Yes/No).

Statistical analyses

The participants’ characteristics and prevalence of those “At risk for DCD” and “Probable DCD” were reported using descriptive statistics by absolute and relative frequencies. Cases were merged by “At risk for DCD” and “Probable DCD” as “DCD traits” for analysis. The levels of physical activity and sedentary behaviour were the outcomes, DCD traits by each period of time were independent variables, while the sociodemographic and health-related factors were assumed as co-variables. Bivariate analyses tested the associations a priori using the chi-square test. Logistic regression analysis was used to assess levels of physical activity and sedentary behaviour in adults with DCD traits over time (childhood, adult life and total score). The analysis was adjusted for those sociodemographic and health-related factors with p ≤ 0.20 in the bivariate analyses. Each outcome was tested in a separate analysis, using three models. Considering the known role of sex and age in physical activity and sedentary behaviour30,62, these two variables were included in all models. In Model 1 the relationship between each outcome and DCD traits over time was controlled by sex and age, only. In Model 2 the relationship between each outcome and DCD traits over time was controlled by sex ang age, including a range of other sociodemographic variables. Finally, in the Model 3 the relationship between each outcome and DCD traits over time was controlled by sex and age, including a range of health-related variables.

An additional sensitivity analysis using these three models was performed, considering only the cutoff for probable DCD, which allowed tested the importance of the categorization made for DCD traits in this study. Odds-ratio (OR) with 95% of confidence interval tested the associations and the level of significance was set at p < 0.05. The null hypothesis, that DCD traits do not explain levels of physical activity and sedentary behavior, was rejected when the significance levels was below 0.05. All analyses were performed using the Statistical Package for Social Sciences (SPSS Inc. version 28).

Results

The age of the participants ranged from 17 to 55 years (25.48 ± 7.20) and the majority were female (65.5%). Insufficient levels of physical activity and elevated sedentary behaviour were observed in 50% and 27.4% of the sample, respectively. The occurrence of DCD Traits (ADC Total scores) ranged from 10.5% to 12.9%, respectively for those screened at risk for DCD and probable DCD. The main characteristics of the participants are described in the Table 1.

Table 1 Characteristics of the participants.
Full size table

Insufficient levels of physical activity were significantly more frequent in individuals with DCD traits, considering childhood (63.0%), adult life (68.4%), and total score (67.6%). Also, in those from exact sciences courses (71.4%), and those who used medication (75.0%) (Table 2).

Table 2 Distribution DCD traits, sociodemographic and health characteristics according to the levels of physical activity in university students.
Full size table

Elevated sedentary behaviour was significantly more frequent in individuals with DCD traits, considering childhood (50.0%), adult life (55.3%), and total score (52.9%). Also, in those from Social and Human sciences courses (71.4%), previously diagnosed with neurodevelopmental disorders (70.0%), and those who used medication (60.0%) (Table 3).

Table 3 Distribution DCD traits, sociodemographic and health characteristics according to sedentary behaviour in university students.
Full size table

Participants who reported DCD traits had significantly increased odds for both outcomes across all time periods. Although controlling for health-related variables (Model 3) eliminated the significant association between DCD traits during childhood and insufficient levels of physical activity, and between DCD traits total score and insufficient levels of physical activity, the association remained significant across all time periods for adult life, and after controlling for sociodemographic variables for childhood and total score. On the other hand, despite a slight reduction in the odds ratios, significant associations between DCD traits and elevated sedentary behaviour persisted across all time periods, even after adjustments in Models 2 and 3 (Table 4).

Table 4 Association between DCD traits over time periods and levels of physical activity and sedentary behaviour in university students.
Full size table

Considering only probable DCD (pDCD), the significant association observed in childhood and adult life with insufficient levels of physical activity (Model 1) remained significant after adjustments in Model 2 but was no longer significant after adjustments in Model 3. The presence of probable DCD remained significantly associated with elevated sedentary behaviour across all Models in adult life. However, in childhood and total scores, this association was observed only after controlling for sociodemographic variables (Model 2) (Table 5).

Table 5 Association between pDCD over time periods and levels of physical activity and sedentary behaviour in university students.
Full size table

Discussion

This study showed that DCD traits during childhood, adult life and combining both periods (total score) increased the risk of insufficient levels of physical activity by 2.19, 2.88 and 2.59 times, respectively. After adjustments by sociodemographic factors, the relationship with physical activity persisted for childhood, adult life, and total scores. However, after adjustments by health-related factors, the relationship with physical activity persisted only for adult life. It may highlight that the burden of such sociodemographic and health-related factors on physical activity appears to differ across developmental periods. This suggests that during childhood, these factors play a less prominent role in shaping physical activity behaviours compared to adulthood, even in the presence of DCD traits. In other words, these determinants seem to gain greater importance later in life, potentially acting as more significant barriers to physical activity as individuals age. This hypothesis aligns with the synthesis by Telama63, which pointed out that the stability of regular physical activity varies across age groups, being notably less stable during the transition from adolescence to adulthood.

The odds were greater for sedentary behaviour, increasing the risk for elevated sedentary behaviour by 5.81, 6.72, and 4.95 times, respectively for childhood, adult life and total score. The relationship with sedentary behaviour persisted significantly for all time periods, with odds ratios increasing in Model 2 for childhood, and slightly decreasing for adult life and total score. This aligns with findings that a physical activity deficit may begin in early childhood64, but its persistence can vary depending on sociodemographic or health-related factors.

Although physical activity has recently gained greater public health attention in Brazil65, there is still a significant gap in opportunities to be physically active across the country. For instance, a pooled analysis from 2013 to 201966 revealed substantial inequalities in leisure-time physical activity, particularly among individuals with lower levels of education, women and non-white populations. This context may explain our findings, especially considering that the university where the study was conducted is known for its inclusion and equity policies. Many students come from low-resource backgrounds and are often the first in their families to access higher education. A large proportion are from rural areas, spend long hours sitting during classes and commuting by bus, and often need to work to support themselves financially. These circumstances place our sample at a significant disadvantage when it comes to engaging in physical activity, and make them more likely to lead sedentary lifestyles. Given this background, it becomes clear that reducing sedentary behaviour requires considerable resilience, something that may be particularly challenging for individuals with DCD traits, who often have reduced physical literacy11.

However, for fair comparisons, methodological differences between the studies should be considered. Although both this and the previous study66 used a similar weekly minutes-based reference to assess physical activity levels, our study included only university students, while the other study66 recruited adults from the general population. Additionally, our participants completed the questionnaire online via Google Forms, whereas participants in the other study66 were interviewed by telephone.

The role of DCD traits on physical activity and sedentary behaviour is consistent with previous literature, indicating that deficits in motor skills may limit individuals’ participation in physical activities, contributing to increased sedentary behaviours67 Although, we found significant associations between both outcomes and DCD traits, the variations in DCD traits across developmental periods and in odds ratios across the three models suggest that DCD traits present a greater burden on sedentary behaviour than on physical activity, especially when considering sociodemographic and health-related factors.

This was not replicated in a study conducted with Korean adults found significantly lower levels of physical activity in adults with DCD compared to typically developing adults, but no significant differences in sedentary behaviour68. In contrast, a study conducted with Finnish adults69 found significantly higher levels of sedentary behaviour in individuals with DCD, as measured by accelerometry compared to their peers without DCD.

The stronger association of DCD traits with sedentary behaviour was further supported by the sensitivity analysis using pDCD. The pDCD during childhood and adult life increasing the risk of insufficient physical activity by 3.96 and 3.14 times respectively in Model 1, remained significant after adjustments by sociodemographic factors (Model 2). On the other hand, pDCD markedly increased the risk of sedentary behaviour, ranging from 3.99 to 11.99 times when controlled for sex and age (Model 1), and from 3.47 to 11.35 times after adjustment for sociodemographic (Model 2) and health-related (Model 3) factors. These findings are in line with previous literature27,70, which suggests that motor difficulties in adults with DCD can explain avoidance behaviours70, including disengagement from physical activities.

It is important to note that sedentary behaviour and physical activity are distinct, though related. Each outcome can manifest differently, especially in emerging adulthood. For example, Aira et al.71 through a cohort study, found that adolescents who were already physically inactive were more likely to increase sedentary behaviour over time. Similarly, Smith et al.72 found that adolescents with low gross motor coordination at age 16 reported more screen time and sedentary behaviour into adulthood, a pattern that persisted at age 42. However, the relationship between motor coordination and physical activity was not significant at age 16 but became significant by age 42, suggesting a non-persistent effect of motor difficulties on physical activity over time or a delayed effect.

However, the literature presents mixed evidence regarding differences in physical activity and sedentary behaviour during childhood between individuals with and without DCD. A systematic review9 showed lower levels of physical activity in children with DCD, but study heterogeneity limited the strength of this evidence. Another study73 found that children with higher motor proficiency were 2.46 times more likely to meet the 60-minute daily physical activity recommendation. Yu et al.67 did not find significant differences in physical activity or sedentary behaviour between children with and without DCD, but did observe sex-based differences, with girls being less active and more sedentary67.

Our study suggests that DCD traits may explain insufficient levels of physical activity and sedentary behaviour in adults when childhood is considered alone, and only when controlling for sex and age. As this association became non-significant in Model 3, the findings suggest that there are other health-related factors may mediate physical activity in childhood more than DCD traits.

The persistent associations observed for adulthood, but not childhood, support the hypothesis that DCD traits may have been concealed or unrecognized during developmental stages, where movement challenges remained within their capacities. During adulthood, however, new challenges come up, individuals with DCD may fully manifest traits that were not apparent in childhood but become evident in adulthood as these individuals likely face demands beyond their capacities2. A series of case studies43 has suggested that the limited motor capacities associated with DCD are often overshadowed by non-motor issues in adulthood, highlighting the presence of emotional comorbidities, such as anxiety and depression. While these findings are novel, more research is needed to clarify the mediating role of persistent DCD traits on physical activity and sedentary behaviour deficits in adults, as evidence remains limited.

Strengths and limitations

Exploring DCD traits and pDCD to understand the risks for insufficient levels of physical activity and elevated sedentary behaviour in adults is a significant strength of this study, given that the literature on DCD in adults is still limited. Dividing the DCD traits into three periods was crucial for identifying its role on physical activity and sedentary behaviour persistently. However, the lack of objective measures for physical activity and sedentary behaviour represents a limitation of this study. Future research should consider employing of self-reported and objectively measured assessments of physical activity and sedentary behaviours. Further research is also necessary to determine whether sedentary behaviour continues to pose a greater burden than physical inactivity in this population. Additionally, the absence of reliability and validity studies of the ADC checklist for the Brazilian population should indeed be acknowledged as a limitation of this study. However, the ADC checklist is the most widely used instrument to screen for DCD traits in adults and is, to date, essentially the only option available in the literature for assessing DCD in this age range. The original English questions were translated into Brazilian Portuguese, and a panel of the research team was convened to at least ensure the instrument was understandable for Brazilian participants. We anticipate that a proper cultural adaptation of the instrument will be conducted soon to minimize potential bias in future studies. Finally, self-report measurements cand introduce bias in this study and affect the validity of the reported associations. Objective measurements should be included in future studies.

Conclusions

DCD traits were associated with reduced levels of physical activity and increased sedentary behaviour in adults. After adjustments for sociodemographic and health-related factors, the persistent effect of DCD traits remained significant only for sedentary behaviour, and was even stronger for pDCD. While the impact of DCD on insufficient physical activity was overshadowed by its impact in increasing sedentary behaviour, further studies are needed to clarify the relationships observed in this study. This study also highlights its unique contribution to the literature, as to the best of our knowledge, it is the first investigation of DCD traits among adults in Brazil. From a practical perspective, information about potential DCD traits should be included in the health history of adults when participating in physical activity programs. As a recommendation, interventions should prioritize reducing sedentary behaviour among adults with DCD traits, incorporating psychosocial and behavioural strategies, preferably managed by the University.