Psychological determinants of physical activity: differences between adults with type 2 diabetes, prediabetes, and without diabetes

Cogent Psychology ISSN: 2331-1908 (Online) Journal homepage: www.tandfonline.com/journals/oaps20 Psychological determinants of physical activity: differences between adults with type 2 diabetes, prediabetes, and without diabetes Iris Pahmeier, Andrea Teti & Michael Braksiek To cite this article: Iris Pahmeier, Andrea Teti & Michael Braksiek (2024) Psychological determinants of physical activity: differences between adults with type 2 diabetes, prediabetes, and without diabetes, Cogent Psychology, 11:1, 2407694, DOI: 10.1080/23311908.2024.2407694 To link to this article: https://doi.org/10.1080/23311908.2024.2407694 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 01 Oct 2024. Submit your article to this journal Article views: 743 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oaps20 HealtH PsycHology | ReseaRcH aRticle Cogent PsyChology 2024, Vol. 11, no. 1, 2407694 Psychological determinants of physical activity: differences between adults with type 2 diabetes, prediabetes, and without diabetes iris Pahmeiera, andrea tetib and Michael Braksieka aDepartment of sport science, University of Vechta, Vechta, germany; bInstitute of gerontology, University of Vechta, Vechta, germany ABSTRACT this study aimed to compare psychological constructs related to physical activity (Pa) among adults with type 2 diabetes (t2D), prediabetes, and without diabetes. the focus was on understanding differences in social-cognitive determinants and individual motives regarding Pa in order to tailor Pa interventions more effectively. the study utilized a sample from the randomized controlled trial activate_Prevention, including participants with t2D, prediabetes and without diabetes at baseline (n = 214). socio-demographic variables, diabetes parameters, psychological constructs, and the level of sport activity were gathered. the data were analyzed using regression models. adults with t2D exhibited significantly lower self-efficacy, recovery self-efficacy, and affective attitude toward physical activity compared to those without diabetes. they also perceived stronger barriers. Participants with prediabetes and t2D exhibited stronger extrinsic motives but lower intrinsic motives than those without diabetes. the odds of not engaging in sport activity were higher in participants with prediabetes and t2D. the study highlights differences in socio-cognitive variables, barrier perception, and motivational profiles among adults with varying diabetes statuses. these findings underline the need for tailored intervention strategies that consider these differences, emphasizing intrinsic motives and addressing the physical and mental health benefits of exercise. Introduction type 2 diabetes (t2D) represents a prevalent chronic-degenerative disease associated with substantial physical and mental health implications. While lifestyle factors like physical activity (Pa) and reducing sedentary behavior significantly impact t2D management, individuals with t2D often struggle to adopt necessary behavioral changes to achieve optimal health effects (cleven et al., 2020; Kennerly & Kirk, 2018). Recent (sport) psychology research has underlined the influence of social-psychological variables within behavior change theories in driving health-related sports activities, particularly in populations with prediabetes and t2D (de Man et al., 2020; Heiss & Petosa, 2016; Rhodes et al., 2019). However, these variables appear notably distinct between individuals with different diabetes statuses, necessitating a deeper understanding of these variances (Plotnikoff et al., 2009). currently, limited knowledge exists about these specific differences regarding individuals with prediabetes, despite the potential of Pa-based interventions to reduce the risk of developing t2D in this cohort (Huttunen-lenz et al., 2018). Understanding these variations is crucial for developing tailored Pa interventions addressing the unique needs of patients with (pre-)diabetes. yet, theoretical frameworks beyond social-cognitive theories, such as humanistic and process-related perspectives, have received insufficient attention in this domain (Rhodes et al., 2019). exploring variables from these alternative frameworks could fill gaps left by social-cognitive theories and provide a more comprehensive understanding of personal determinants influencing Pa behavior (Rhodes et al., 2019). this study aimed to address this gap by comparing constructs from social-cognitive models, humanistic (goal content theory), and process-related models (health action process approach - HaPa) © 2024 the Author(s). Published by Informa UK limited, trading as taylor & Francis group CONTACT Iris Pahmeier [email protected] Department of sport science, University of Vechta, Driverstraße 22, Vechta 49377, germany https://doi.org/10.1080/23311908.2024.2407694 this is an open Access article distributed under the terms of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. the terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. ARTICLE HISTORY Received 10 January 2024 Revised 17 september 2024 accepted 18 september 2024 KEYWORDS type 2 diabetes; prediabetes; physical activity; humanistic approach; social-cognitive approach; motives SUBJECTS Physical activity and Health; sports Psychology; Diabetes 2 i. PaHMeieR etal. among adults with t2D, prediabetes, and without diabetes. Background t2D stands as one of the most prevalent chronic diseases among adults (scheidt-Nave & icks, 2019). characterized by elevated blood glucose levels (hyperglycemia), t2D is influenced by genetic, age-related, and lifestyle factors, including physical inactivity, poor nutrition, and smoking (german Medical association, 2023; scheidt-Nave & icks, 2019). Notably, low socioeconomic status is associated with an increased risk of developing t2D (agardh et al., 2011; Kautzky-Willer etal., 2019). elevated blood glucose levels but not at t2D levels, indicates prediabetes. Hyperglycemia associated with (pre-)diabetes leads to various complications affecting the eyes, kidneys, nerves, and cardiovascular system (german Medical association, 2023). Moreover, t2D correlates with higher occurrences of depressive symptoms and illnesses compared to unaffected individuals (Darwish et al., 2018). Regular Pa serves as a pivotal primary and tertiary preventive measure against associated diseases in t2D. studies emphasize its role in reducing the risk of t2D development, particularly when combined with maintaining a healthy weight (cloostermans etal., 2015). lifestyle modifications, including dietary changes and Pa promotion, are fundamental strategies for preventing and managing t2D (Merlotti etal., 2014; toplak etal., 2019). However, despite the recognized importance of Pa, individuals with t2D often struggle to initiate and maintain suitable Pa regimens, displaying lower activity levels and increased sedentary behavior compared to those without diabetes (Kennerly & Kirk, 2018; lees & Booth, 2004). they also showed low adherence in supervised programs accompanied by high dropout rates (Bullard et al., 2019). this presents a critical challenge in motivating and engaging individuals with t2D in regular sports and physical activities. Literature review Heiss and Petosa (2014) provided insights into demographic, behavioral, and environmental factors related Pa in adults with t2D. Factors such as higher BMi, older age, lower income, and perceived disease severity correlated with inactivity. Psychosocial factors like lower self-efficacy and perceived barriers, along with environmental aspects such as limited social support and weather constraints, were also associated with activity levels. When it comes to understanding, explaining and also promoting Pa, the use of theoretical frameworks has proven useful (Rhodes etal., 2019). social-cognitive perspectives, as exemplified by Plotnikoff et al. (2009), examined social-cognitive aspects regarding Pa in patients with type 1 diabetes (t1D), t2D, and those without diabetes. Notably, differences in Pa-related constructs from various social cognitive models were observed depending on these diabetes statuses. the models included the social cognitive theory, theory of planned behavior, transtheoretical model, and theory of protection motivation. While the intention-behavior gap remained a challenge in these theories, they explained only a modest portion of the variance in objective and self-reported Pa. the Health action Process approach (HaPa) addresses this gap by describing pre-intentional (motivational) and post-intentional (volitional) phases in behavior change. Volitional variables play a critical role in realizing intended behavior, especially concerning barriers commonly reported by patients with t2D (gómez et al., 2021; lidegaard et al., 2016). the HaPa’s inclusion of recovery self-efficacy acknowledges dealing with failure at a barrier as pivotal for sustained behavior change (schwarzer, 2008). Notably, lippke and Plotnikoff (2014) validated HaPa’s assumptions in the context of Pa among adults with t2D. Furthermore, the HaPa has demonstrated success in crafting interventions for individuals with t2D, facilitating specific health behavior changes like diet modification (MacPhail etal., 2014). However, its application in promoting Pa among this population remains relatively unexplored. Despite the HaPa’s advantages over classical social-cognitive theories due to its integration of volitional components, both theoretical frameworks primarily operate under the assumption that Pa is a consequence of deliberate contemplation about values and expectations (Rhodes et al., 2019). these frameworks predominantly focus on parameters rooted in conscious cognitive control. From a humanistic perspective, the importance of the influence of unconscious parts of behavior is emphasized. an established theory in this context is Deci and Ryan (1985, 2000) self-determination theory (sDt). studies highlighting the positive association between autonomous motivation and self-reported Pa among t2D patients support this theory (Miquelon & castonguay, 2016). gourlan et al. (2016) identified different motivational profiles among adults with t2D, emphasizing the need for person-centered approaches in understanding the motivation for Pa. cogeNt PsycHology 3 Within sDt, goal content theory differentiates between extrinsic (e.g., appearance and social recognition) and intrinsic goals (e.g., enjoyment and sense of belonging) suggesting that intrinsic goals promote greater motivation, need satisfaction, and well-being (Rheinberg & engeser, 2018). to date, there are several studies differentiating explicit goals and motives for recreational and health sports, such as health, figure and appearance, social contact and enjoyment (lehnert etal., 2011). thus, it is asked in more detail what motivates people to participate in sports and what their goals are. Understanding these motivations and goals facilitates a more tailored alignment between individual motives and the content and structure of Pa-based intervention programs. this tailored approach aims to optimize the effectiveness of such programs, promoting both the initiation and maintenance of Pa (schorno et al., 2022). However, current research lacks specific insights into content-related motives for Pa among individuals with (pre-)diabetes, which hinders the customization of interventions to match individual goals and motives effectively. in light of the empirical and theoretical frameworks elucidated, this study endeavored to advance our comprehension of theory-based Pa-related variables among adults diagnosed with prediabetes and t2D. our primary objective was to gain insights essential for customizing tailored Pa-based interventions for this clientele. to achieve this, we investigated differences in (1) variables derived from established social cognitive theories, further augmented by insights from process models, (2) perceptions surrounding various barriers that impede the initiation of Pa, and (3) motives driving engagement in Pa among this specific population. Methods Design and sample the sample consisted of participants of the randomized controlled trial activate_Prevention at baseline (Mähs et al., 2022). the primary aim of the trial was to investigate the effects of an activity tracker-based intervention and an additional digital motivational intervention on the physical activity level and health outcomes in adults with prediabetes, t2D and without diabetes. Baseline data was collected successively from the end of June 2021 to the beginning of May 2022. the intervention lasted for 6 months with a 3-month follow-up period. currently, the data collection is completed, and the analysis of the longitudinal data is ongoing. Participants with prediabetes and t2D were recruited from a partnered diabetes center. Healthy individuals were recruited from a local cooperating company, that was interested in the health promotion of their workers. additionally, participation in the study was promoted via local print media and the study website to recruit people from the general population, regardless of their diabetes status. Baseline assessments included a questionnaire capturing socio-demographic, health-related information, Pa, and psychosocial constructs. Diabetes parameters were clinically assessed by trained personnel. the analysis included participants with prediabetes (n = 33), t2D (n = 76), and those without metabolic disorders (n = 105), resulting in a total analyzed sample of 214 participants. Due to the predetermined project duration and a delayed start to recruitment caused by the coViD-19 pandemic, it was not possible to generate balanced subsamples. Recruitment had to be completed to ensure that all participants had sufficient time in the study. the discrepancy in participant numbers was primarily due to the varying prevalence of these conditions in the population and the challenges in recruiting individuals with prediabetes and t2D. the diabetes center involved in the study treats a limited number of patients, most of whom have t2D rather than prediabetes. additionally, no incentives were provided for participation in the study. ethical approval was obtained from the ethical Review Board of the medical association of lower saxony, germany (05/2021), and participants provided voluntary written consent. Measures Demographic characteristics of the participants were measured using standardized questions including age, gender, and socio-economic variables (lampert etal., 2013). Following lampert etal. (2013) the educational and professional qualifications and the net equivalent income of the participants were transformed into indices ranging from 1 to 7. Psychosocial constructs were measured using validated scales. the values for the internal consistencies of the scales are shown in table 1. Self-efficacy to exercise was assessed with a 5-point likert scale consisting of five items (e.g., ‘i am confident i can participate in regular exercise when i am tired’) (Fuchs, 1997; Marcus et al., 1992). the scale ranges from 0 (not safe at all) to 4 (extremely safe). Recovery self-efficacy was also assessed with a 5-point likert scale consisting of five items (e.g., ‘i am 4 i. PaHMeieR etal. sure that i can be physically active again on a regular basis, even if i have not been active regularly for more than a month’) (Brehm et al., 2010; schwarzer, 2008). the scale ranges from 0 (not safe at all) to 4 (extremely safe). to assess the participants’ volitional inhibition regarding exercise, the dimension of postponing training from the Volition in exercise Questionnaire was used (e.g., ‘i often experience that i need to pull myself together to participate in my exercise activity’) (Pfeffer et al., 2020). this scale consisted of four items that could be answered on a scale from 0 (strongly disagree) to 3 (strongly agree). to assess the participants’ individual motives and goals in exercise and sport the Bernese Motive and target inventory (lehnert et al., 2011) was used. the inventory consists of 24 items covering seven motives that can be reasons for individuals to do exercise and sport activities: contact (5 items), competition/ performance (4 items), activation/enjoyment (3 items), distraction/catharsis (4 items), figure/appearance (3 items), fitness/health (3 items), and aesthetics (2 items). each item was presented with a 5-point rating scale ranging from 0 (strongly disagree) to 4 (strongly agree). the affective attitude toward PA was measured using a semantic differential scale (Brand, 2006). Participants were presented with the statement ‘When i think about being physically active, i feel…’. Following this statement, they were shown pairs of opposite adjectives (e.g., ‘not relaxed’ versus ‘very relaxed’). For each pair, participants rate their feelings on a scale from 1 to 7, where 1 indicates one extreme of the feeling (e.g., ‘not relaxed’) and 7 indicates the opposite extreme (e.g., ‘very relaxed’) allowing them to indicate the degree to which each adjective describes their feelings about being physically active. Perceived barriers to PA were measured using four scales developed by Brehm et al. (2010) assessing the individual relevance of barriers using 7-point rating scales: lack of time/stress (4 items), lack of motivation (6 items), insecurity regarding physical competence (3 items), and social context (2 items). the scales range from 0 (strongly disagree) to 6 (strongly agree). a single item was used to measure the social support for PA (self-constructed: ‘i feel supported in my physical activities by my social environment (e.g., family, friends, acquaintances)’). the scale of the item ranges from 0 (strongly disagree) to 4 (strongly agree). Intention to PA was also measured with a single item (‘How strong is your intention to be regularly active in the coming weeks and months?’) (gerber et al., 2010). the scale of the item ranges from 0 (I have no intention at all) to 5 (I have a very strong intention). For all scales, mean values were used for the analyses. additionally, the mean values of the motives were processed according to the principle of intraindividual standardization to compensate for general Table 1. overview of variables, summary statistics and reliability. Variables non-diabetes Prediabetes type 2 diabetes Full sample Internal consistency (Mc Donald’s ω)MsD MsD MsD MsD self-efficacy to exercisea 2.3 0.9 2.1 0.8 1.8 1.0 2.1 1.0 0.83 Recovery self-efficacya 3.0 0.9 2.7 0.9 2.4 1.1 2.7 1.0 * Volitional inhibitionb0.7 0.7 1.1 0.8 1.0 0.9 0.9 0.8 0.90 Affective attitude toward PAc5.2 1.3 4.9 1.6 4.6 1.4 5.0 1.4 0.95 social supportb2.8 1.2 2.4 1.1 2.1 1.4 2.5 1.3 * Intentiond4.2 0.8 4.0 0.7 3.6 1.2 4.0 1.0 * Motives (standardized)e Fitness/health 111.2 4.1 108.1 7.0 110.5 4.7 110.5 5.0 0.81 Figure/ Appearance 104.1 9.1 111.3 6.4 110.2 6.9 107.2 8.6 0.91 Contact 98.4 7.0 99.2 8.2 99.6 7.6 99.2 7.5 0.94 Distraction/ Catharsis 103.2 7.3 102.9 6.1 100.5 6.2 102.1 6.9 0.90 Activation/Joy 107.9 5.4 102.5 5.5 105.4 6.8 106.2 6.3 0.83 Competition/ Performance 92.9 6.8 92.0 5.7 91.3 4.6 92.2 6.0 0.92 Aesthetics 93.6 7.8 92.1 5.3 92.9 5.7 93.0 6.7 * Barriersf lack of time/stress 1.9 1.6 2.4 1.5 2.0 1.5 1.9 1.5 0.88 lack of motivation 1.3 1.1 2.2 1.4 2.0 1.4 1.6 1.3 0.89 Insecurity regarding physical competence 0.6 0.8 1.3 1.2 1.4 1.1 1.0 1.0 0.72 social context 0.9 1.0 1.4 1.3 1.4 1.4 1.1 1.2 * *less than three items. aRange: 0–4. bRange: 0–3. cRange: 1–7. dRange: 0–5. eRange: 80–120. fRange: 0–6. cogeNt PsycHology 5 response tendencies and to better account for relational expressions of the motives (sudeck etal., 2011). Values around 100 correspond to the average importance of a motive in relation to the other individual motives (conzelmann et al., 2012). Pa was assessed using a scale from the Physical activity, exercise, and sport Questionnaire (Fuchs et al., 2015). Participants reported up to three different sport and exercise activities engaged in over the last four weeks, detailing frequency, and duration. the sport activity index (sai) was derived from these data, representing minutes per week engaged in sports activities. Statistical analyses Data were analyzed using sPss 27 (iBM corp, 2020) and R studio (Rstudio team, 2020). to investigate differences in the psychological variables between the three groups (prediabetes, t2D and non-diabetic) controlling for gender, age, educational and professional qualifications (ePQ), and the net equivalent income (NeQ), linear regression models were conducted (tabachnick & Fidell, 2018). Beforehand, missing values were analyzed, and the assumptions of the regression analyses were checked. there were only missing values in the indices for ePQ, NeQ and sai. these were missing completely at random (little’s test: χ2 = 7.309, p = 1.00). subsequently, they were imputed by the R package mice using multivariate imputation by chained equations (van Buuren & groothuis-oudshoorn, 2011). Regression assumptions were checked using standard diagnostic plots (for linearity, normality, homogeneity, and outliners) and variance inflation factor (ViF) (for multicollinearity) (Bruce et al., 2020; Pituch & stevens, 2016). the plot models indicated that most of the assumptions were met. However, some plots and White’s test indicated heteroscedasticity in the models. therefore, the R packages sandwich and lmtest were used to generate heteroscedasticity-consistent covariance matrices and heteroscedasticity-robust standard errors for the regression models to handle heteroscedasticity. the ViFs of all predictors were below 10 (1.1–1.5) indicating that multicollinearity did not inflate the estimations. a power analysis was conducted using g*Power version 3.1.9.6 (Faul et al., 2009) for sample size estimation, based on results from comparable studies (Plotnikoff et al., 2009). With an error probability of α = 0.05, power = 0.80 and six predictors in the regression models the minimum sample size for small effects (f2= 0.05) in single regression coefficients is N = 159. thus, the obtained sample size of N = 214 is adequate to test the study hypothesis. the sport activity index (sai) analysis revealed a substantial number of zeros and a right-skewed distribution in the dataset. to address this, zero-inflated negative binomial regression models were employed utilizing the pscl package in R (Zeileis et al., 2008). Zero-inflated negative binomial regressions assume data to stem from a negative binomial distribution, accommodating the high number of zero values. this approach involves a two-part model: first, a logistic regression model determines the probability of zero counts versus non-zero counts, followed by a negative binomial regression model that examines count data among non-zero counts (slymen et al., 2006). Results table 2 displays descriptive statistics. the sample consisted of 61% females and respondents had an average age of 51 years. Participants without diabetes represented the largest group (49.1%) following by the participants with t2D (35.5%) and prediabetes (15.4%). on average, individuals with t2D were older than the other groups. socio-demographically, participants without diabetes had slightly higher educational qualifications and net equivalent income than the other groups. Weekly Pa showed that participants without diabetes engaged for over two hours, prediabetes participants for slightly over one and a half hours, and those with t2D for less than an hour. table 1 presents summary statistics for sociocognitive variables. internal consistency of the scales was good, enabling their utilization for further analysis. Descriptive values aligned closely with theoretical means of scales. Motive values indicated low, medium, and high importance, without extreme values. constructs theoretically associated positively with Pa (e.g., self-efficacy) tended to surpass scale Table 2. sociodemographic characteristics of participants at baseline. Baseline characteristic non-diabetes Prediabetes type 2 diabetes Full sample n%n%n%n% gender 105 33 76 214 Female 62 59.0 27 81.8 41 53.9 130 60.7 Male 43 41.0 6 18.2 35 46.1 84 39.3 M SD M SD M SD M SD Age 48.8 13.9 41.7 12.6 58.1 12.5 51.3 14.4 sAI 137.6 184.8 93.3 185.8 50.9 93.9 100.2 163.4 ePQ 5.5 1.5 4.8 1.4 4.3 1.3 5.0 1.5 neI 4.5 1.7 3.7 1.8 3.3 1.8 4.0 1.8 sAI: sport activity index; ePQ: educational and professional qualifications; neI: net equivalent income. 6 i. PaHMeieR etal. means, while those negatively linked tended to fall below (e.g., barriers). the results of the regression models are shown in the tables 3–6. all models were used to analyze the difference in the Pa-related variables or sport activity among participants with prediabetes, t2D and without diabetes, controlling for age, gender, and socio-economic factors. table 3 indicates that individuals with t2D displayed significantly lower self-efficacy to exercise, recovery self-efficacy, and affective attitude toward exercise compared to those without diabetes. they also perceived lower social support for Pa, exhibited lower intention for Pa, and demonstrated higher volitional inhibition than participants without diabetes. No differences were observed between participants without diabetes and those with prediabetes. among control variables, only age showed a positive association with perceived social support. the overall significance of the models indicated a significant portion of variance explained by predictors (6% ≤ adj. R2 ≤ 12%). table 4 shows that both participants with prediabetes and t2D perceived lack of motivation and insecurity regarding physical competence as stronger barriers to Pa compared to those without diabetes. additionally, participants with t2D felt time constraints and stress more acutely. there were no differences regarding the social context as potential barrier. Regarding the control variables, females and younger individuals perceived fewer time constraints and less stress. additionally, age was negatively associated with a lack of motivation. the socioeconomic variables were not associated with any perception of barriers. all models explained a significant part of the variance in the barriers ranging from 8% to 13%. Differences in motives for exercise and sport are shown in table 5. Participants with prediabetes and t2D expressed a stronger figure/appearance motive but a lower activation/joy and distraction/catharsis motive than those without diabetes. there were no differences regarding the orientation to fitness and health, contact, competition and performance as well aesthetics between the analyzed groups, but there was a difference between genders: women had a significantly lower orientation to competition and performance but higher expressions of the distraction/catharsis and activation/joy motives. age and socioeconomic factors showed varied associations with different motives. the motive contact was not affected by any of the predictors at all resulting in an insignificant overall model. all other models explained a significant part of the variance in the motives but with differences in the effect size. With 22% (adjusted R2), a comparatively higher part of the variance of the motive activation/joy could be explained, whereas the variance of motive aesthetics could be explained less well (6%). to ensure that there are also differences between the groups in terms of sport activity, an additional model was calculated to test for differences. table 6 shows that the odds of not engaging in any sport activity are significantly higher in participants with prediabetes with a factor of 2.6 and with a factor of 3.3 in participants with t2D. When being active, the minutes of the sport activity are not associated with the diabetes status. No control variable was associated with the sport activity. Table 3. Regression analyses of socio-cognitive variables. self-efficacy to exercise Recovery self-efficacy Volitional inhibition Affective attitude social support Intention Prediabetesa−0.14 −0.24 0.30 −0.20 −0.23 −0.17 (0.19) (0.20) (0.17) (0.34) (0.34) (0.24) type-2 diabetesa−0.61*** −0.62*** 0.31* −0.61* −0.69** −0.65** (0.15) (0.17) (0.14) (0.26) (0.26) (0.22) Age 0.005 0.004 −0.01 0.01 0.02* 0.01 (0.005) (0.01) (0.004) (0.01) (0.01) (0.01) sexb−0.10 −0.01 0.02 0.05 0.13 0.20 (0.14) (0.15) (0.11) (0.22) (0.22) (0.18) neI 0.03 −0.02 −0.002 0.09 0.03 0.03 (0.04) (0.05) (0.03) (0.06) (0.06) (0.05) ePQ 0.02 0.03 −0.05 −0.02 0.09 0.02 (0.05) (0.05) (0.04) (0.07) (0.07) (0.06) Constant 2.04*** 2.67*** 1.35*** 4.40*** 1.14 3.27*** (0.45) (0.48) (0.37) (0.69) (0.69) (0.64) observations 214 214 214 214 214 214 R20.10 0.08 0.07 0.06 0.11 0.12 Adjusted R20.07 0.05 0.05 0.03 0.08 0.09 Residual std. error (df = 207) 0.92 1.00 0.77 1.40 1.20 0.92 F statistic (df = 6; 207) 3.88** 2.87* 2.67* 2.26* 4.29*** 4.57*** ePQ: educational and professional qualifications; neI: net equivalent income. *p < .05; **p < .01; ***p < .001. aReference category are participants without diabetes. bReference category are male participants. cogeNt PsycHology 7 Discussion the primary aim of this study was to deepen our understanding of the disparities in psychological variables related to Pa among adults based on their diabetes status – non-diabetes, prediabetes, and t2D. this exploration serves as a foundation for tailoring interventions in both primary and secondary prevention contexts, aiming to bolster the effectiveness of these initiatives and mitigate early drop-out rates. the approach involved a comprehensive comparison of variables derived from established social cognitive theories, extended by facets from process models. additionally, it delved into understanding the perception of barriers hindering Pa initiation and the motives driving Pa practice across different diabetes statuses among adults. the findings generally indicated less favorable values for adults with t2D across most analyzed variables, with some similarly disadvantageous outcomes for prediabetic individuals in specific variables. these results reinforce and expand upon earlier research, particularly in echoing socio-cognitive perspective findings (Plotnikoff et al., 2009). the majority of previously analyzed variables align with the pre-intentional phase in the Health action Process approach (HaPa), such as self-efficacy to exercise. However, this study reveals differences extending into the post-intentional phase (e.g., recovery self-efficacy and volitional inhibition), highlighting the necessity for intervention strategies that bridge the intention-behavior gap specifically concerning sports and exercise activities. strategies focused on action and coping planning, as advocated in HaPa (schwarzer etal., 2011), have shown potential in promoting healthy eating among adults with t2D (MacPhail et al., 2014) and in enhancing medical parameters relevant to t2D prevention (laatikainen et al., 2007), although not directly targeted at individuals with t2D. Table 4. Regression analyses of barriers. lack of time/stress lack of motivation Insecurity regarding physical competence social context Prediabetesa0.36 0.77** 0.70** 0.34 (0.31) (0.29) (0.24) (1.13) type-2 diabetesa 0.53* 0.77*** 0.73*** 0.53 (0.24) (0.21) (0.17) (0.74) Age −0.04*** −0.02* −0.001 −0.02 (0.01) (0.01) (0.01) (0.03) sexb−0.42* −0.07 −0.09 −0.24 (0.21) (0.19) (0.15) (0.69) neI 0.02 −0.06 −0.04 −0.08 (0.06) (0.06) (0.04) (0.18) ePQ −0.04 −0.01 −0.004 −0.04 (0.08) (0.07) (0.05) (0.27) Constant 4.67*** 2.56*** 1.01* 2.68 (0.73) (0.66) (0.49) (2.71) observations 214 214 214 214 R20.16 0.13 0.14 0.11 Adjusted R20.13 0.11 0.12 0.08 Residual std. error (df = 207) 1.44 1.25 0.97 1.18 F statistic (df = 6; 207) 6.51*** 5.21*** 5.78*** 4.16*** ePQ: educational and professional qualifications; neI: net equivalent income. **p < .05; **p < .01; ***p < .001. aReference category are participants without diabetes. bReference category are male participants. Table 5. Regression analyses of motives (standardized). Fitness/ health Figure/ Appearance Contact Distraction/ Catharsis Activation/Joy Competition/ Performance Aesthetics Prediabetesa−2.11 6.47*** 1.10 −1.35*** −4.55*** −0.78 −0.89 (1.52) (1.76) (1.25) (0.20) (0.17) (1.21) (1.37) type-2 diabetesa−1.14 5.35*** 1.60 −1.02*** −4.16*** −0.53 −1.24 (1.35) (1.28) (1.12) (0.17) (0.14) (1.17) (1.14) Age 0.10* −0.04 0.04 −0.16*** 0.15*** −0.09** 0.10** (0.05) (0.04) (0.03) (0.01) (0.004) (0.03) (0.03) sexb−0.06 −0.26 0.37 0.58*** 0.77*** −3.19** 1.73 (1.24) (1.14) (0.98) (0.15) (0.11) (0.97) (0.95) neI −0.01 −0.38 0.31 −0.14** 0.09* 0.16 −0.04 (0.34) (0.31) (0.26) (0.05) (0.03) (0.30) (0.28) ePQ 0.64 −0.39 −0.21 0.42*** 0.17*** −0.23 0.24 (0.42) (0.43) (0.33) (0.05) (0.04) (0.34) (0.34) Constant 103.26*** 110.21*** 95.65*** 108.19*** 98.35*** 102.90*** 84.70*** (4.24) (3.74) (3.20) (0.48) (0.37) (3.43) (3.19) observations 214 214 214 214 214 214 214 R20.15 0.16 0.03 0.14 0.22 0.11 0.06 Adjusted R20.12 0.13 −0.003 0.12 0.20 0.08 0.03 Residual std. error (df = 207) 4.66 7.98 7.47 6.47 5.62 5.78 6.58 F statistic (df = 6; 207) 5.92*** 6.45*** 0.89 5.65*** 9.64*** 4.08*** 2.06 ePQ: educational and professional qualifications; neI: net equivalent income. *p < .05; **p < .01; ***p < .001. aReference category are participants without diabetes. bReference category are male participants. 8 i. PaHMeieR etal. additionally, the results unveiled a lower affective attitude toward Pa among adults with t2D. Prior studies often merged affective and cognitive attitudes following theories like the theory of planned behavior (Plotnikoff et al., 2010). given recent research emphasizing the pivotal role of affective factors in Pa maintenance (Brand & ekkekakis, 2018), the discrepancies found in the affective attitude dimension signal the importance of focusing on affective mechanisms when designing sports programs for adults with t2D or exploring reasons for drop-out from these programs. the absence of differences between prediabetic individuals and those without diabetes in social cognitive variables suggests that distinctions tend to manifest as the disease progresses. Future longitudinal studies should investigate the causal relationship underlying this development. the socio-cognitive determinants of Pa analyzed in this study are instrumental in surmounting barriers that impede Pa initiation or continuity. adults with t2D reporting higher perceptions of time constraints and motivation deficiencies align with prior research (gómez et al., 2021; Plotnikoff et al., 2009). this study extended these findings, revealing that a lack of motivation similarly impedes adults with prediabetes from engaging in Pa to the same extent as adults with t2D. additionally, both groups expressed uncertainty about their physical competence, hindering their Pa. Heiss and Petosa (2016) previously linked lower confidence in performing physical exercises properly with reduced moderate Pa in adults with t2D. this study expands upon this understanding by highlighting that this sentiment is more pronounced in adults with prediabetes and t2D than in those without diabetes. these individuals are not only aware that this feeling hinders their Pa but also perceive this barrier as challenging as the lack of motivation. this necessitates considerations beyond psychological strategies, urging the inclusion of exercise and training content adjustments in preventive and promotional measures for patients with prediabetes and t2D. encouragingly, prediabetic and diabetic adults do not perceive their social environment as an obstacle to being physically active. the analyzed individual motives for engaging in sports can be classified in extrinsic, purpose-related (e.g., fitness/health), and intrinsic, activity-related motives (e.g., activation/joy) (Rheinberg & engeser, 2018). this study unveiled that the extrinsic motive of shaping figure and outer appearance is much more important for adults with prediabetes and t2D. in contrast, intrinsic motives like distraction/catharsis and activation/joy appear to be less influential reasons for their participation in sports. Regarding results from previous studies, showing that intrinsic motivation is generally more important for being physically active, adults with prediabetes and t2D seem to have less favorable motive profiles concerning Pa and sports. Furthermore, the results provide more detailed insights into which motives are more or less relevant for adults with prediabetes and t2D compared to those without diabetes. Despite the less favorable extrinsic motives for Pa, it remains crucial to offer sports activities that align with individual motives, particularly for sustaining Pa (schorno et al., 2022; sudeck & conzelmann, 2011). consequently, the motive of body image and appearance should be more strongly considered in health promotion measures for this clientele. given indications that individuals with t2D may have a negative body image (Preiss et al., 2013; Weinberger et al., 2016) addressing body image concerns psychologically alongside physical transformations should be contemplated. the fact that the intrinsic motives distraction/ catharsis and activation/joy are minor reasons for adults with prediabetes and t2D to engage in exercise and sport activities provides a novel insight. this could stem from the lower engagement or absence of these individuals in physical activities, leading to a lack of the experiences associated with these motives. alternatively, this could arise from their inability to regulate or control their exercise routines to attain such experiences. it is possible that individuals in this group may engage in activities aimed at quicker physiological effects (e.g., Hiit, (Nardi et al., 2018)) Table 6. Regression analyses of activity. sport activity Count model Zero-inflated model estimates (SE) odds estimates (SE) odds Prediabetesa0.06 1.06 0.94* 2.57 (0.24) (0.44) type-2 diabetesa−0.28 0.75 1.18* 3.27 (0.20) (0.44) Age −0.01 0.99 −0.01 0.99 (0.01) (0.01) sexb0.02 1.02 −0.35 0.70 (0.17) (0.31) ePQ 0.02 1.02 0.07 1.08 (0.06) (0.11) neI −0.01 0.99 −0.06 0.94 (0.05) (0.09) Constant 5.74*** 311.97 0.35 1.42 (0.46) (1.04) ePQ: educational and professional qualifications; neI: net equivalent income. *p < .05; **p < .01; ***p < .001. aReference category are participants without diabetes. bReference category are male participants.