Abstract
Keywords
Introduction
Acting volitionally according to one’s own desires (i.e., self-determination) contributes to subjective well-being and quality of life (Deci and Ryan, 2000; Ryan and Deci, 2000). Self-determination is thought to go along with satisfaction of three universal and innate basic psychological needs (BPNs), namely autonomy, competence, and relatedness (Deci and Ryan, 2000) including in persons with mild intellectual disabilities (Frielink et al., 2018). Life transitions may sometimes create new opportunities for supporting and achieving self-determination, but may also make it harder (Audet et al., 2021; Kins et al., 2009; Oliveira et al., 2014). Normative life transitions, for example, foster self-determination and quality of life in neurotypical children (e.g., Gillison et al., 2008). In social and health sciences, transition is a psychological process of adapting to change that involves an inner reorientation with distinct phases (Bridges, 2004). Transitions denote both shifts from one developmental stage to another and events that mark changes or disruptions in physical or mental health, education, work, daytime activities, finances, housing, relationships, or social circumstances (Kralik et al., 2006). The current longitudinal study aimed to explore how transitions may go along with changes in self-determination in persons with severe or profound intellectual and multiple disabilities and what role caregivers play during such transitions.
Individuals diagnosed with severe or profound intellectual and multiple disabilities experience comprehensive cognitive impairments that are accompanied by serious motor, sensory, communicative, and other health problems (Van der Putten et al., 2017; Van Timmeren et al., 2017). Consequently, they rely on extensive assistance from other people for daily functioning (Nakken and Vlaskamp, 2007). Given the focus of this study on the friction between this dependency and support for autonomy, we refer to this population as having complex support needs. People with intellectual and physical disabilities generally require significant accommodation to experience autonomy (Wehmeyer & Shogren, 2016, 2017). From a self-determination theory (SDT; Deci and Ryan, 2000) perspective, autonomy support includes acknowledging and respecting one’s perspectives, feelings, and choices. Additionally, it entails creating a climate rich in encouragement, empathy, and provision of practical information, while remaining free of coercion (Reeve, 2002; Ryan et al., 2006; Soenens et al., 2007). Transitions may offer persons with complex support needs new possibilities to reorganize life according to one’s wishes and desires. For example, transitioning from living with parents to moving into a group home may stimulate the development of new competencies, such as engaging with peers. However, Bigby et al. (2011) and Taylor et al. (2019) emphasized that successful outcomes of transitions largely depend on the adequacy of support. For instance, autonomy-supportive interventions for adolescents with intellectual or other disabilities in transition planning improved self-determination and quality of life compared to control groups (Nadig et al., 2018; Seong et al., 2015; Wehmeyer et al., 2011). It is therefore important to also explore how autonomy support during transitions may foster self-determination for persons with complex support needs.
Persons with complex support needs communicate their needs in idiosyncratic, unconventional, and pre-symbolic ways. Their methods may vary across physical contexts or interaction partners and are often delayed or reactive (Dhondt et al., 2021, 2023; Van Tuyll Van Serooskerken et al., 2022). Self-determination and autonomy support may result from socially mediated meaning-making, in which every effort is made to elicit, pick up, understand, and fulfill the person’s need expressions (Skarsaune et al., 2021; Skarsaune, 2023; Van Tuyll Van Serooskerken et al., 2022). Parents play a key role because of their unique and experiential knowledge through the lifelong and intensive connection with their child with complex support needs (De Geeter et al., 2002; Kruithof et al., 2020). Beyond parents, an autonomy-supportive context is fostered by other social partners involved in the care and upbringing, including other family members and healthcare professionals (Hostyn and Maes, 2009).
The outcomes of transitions may depend not only on support but also on the nature of the transition itself. Meleis et al. (2000) identified, among others, engagement in the transition process, temporal aspects, and expectations and attributions regarding the transition. Regarding engagement in children from the general population, for instance, a positive association was found between student involvement in transition planning and self-determination (Williams-Diehm et al., 2008). The impact of a transition can further vary depending on when it was completed. A recent transition may have different consequences than one completed some time ago, which may have allowed the effects on daily life time to develop for longer. Additionally, teachers’ expectations for students without disabilities have been linked to self-determination-related outcomes, such as student motivation and engagement (Hornstra et al., 2018). Outcomes of transitions may differ according to the possibilities that people involved see for achieving self-determination, which does not necessarily align with why a transition occurs (i.e., a blessing in disguise). Together, these perceived transition characteristics may shed light on changes in self-determination in persons with complex support needs.
Following SDT, this study focused on environmental autonomy support, BPN expressions, and subjective well-being as core constructs experienced by people with complex support needs according to their parents (Lachapelle et al., 2005; Ryan and Deci, 2000; Wehmeyer, 2020). The following research questions were formulated: I. To what extent did changes occur in a) the autonomy support for persons with complex support needs received from all key partners involved, b) the persons’ BPN expressions, and c) the persons’ subjective well-being, within the context of a relevant life transition (see Figure 1)? II. To what extent were the changes over time in these three self-determination-related concepts interrelated (see Figure 2)? III. To what extent were caregivers’ prior expectations of the impact of the transition on a) possibilities for BPN support and b) the persons’ BPN Expressions related to actual changes in BPN expressions (see Figure 3)? And IV. To what extent were a) the degree of “Caregiver engagement” in working towards the transition and b) the “Period of impact” after transition completion, associated with actual changes in the three self-determination-related concepts (see Figure 4)? Research question 1. To what extent did changes occur in self-determination-related concepts, within the context of a relevant life transition (visual representation)? Research question 2. To what extent were the changes in self-determination-related concepts interrelated (visual representation)? Research question 3. To what extent were caregiver expectations prior to the transition (i.e., transition characteristics) related to actual changes in BPN expressions (visual representation)? Research question 4. To what extent were caregiver engagement and period of impact (i.e., transition characteristics) associated with changes in self-determination-related concepts (visual representation)?



Methods
Sample
Information About Expected Transition (N = 40).
aBy work we mean special day service centers that resemble school or work tasks to accommodate and stimulate the possibilities of persons with disabilities.
Demographic Information of Participants and Persons with Complex Support Needs (N = 40).
Procedures
Recruitment was done through a variety of methods, using social media, newsletters, website posts, and word-of-mouth by support staff of several Dutch care and client advocacy organizations. Participants gave written consent before participation. Two measurement waves were planned for assessing independent, dependent, and background variables. T0 took place right after inclusion and thus prior to the expected transition in the life of the person with complex support needs. The second measurement (i.e., T1) took place around 6 months after T0 (M = 7.04, min = 4.91, max = 13.39 months, of which all measurements with more than 8 months apart were caused by the onset of COVID-19 in between). This period was chosen to create variability in characteristics of the transition in terms of caregiver engagement and period of impact. The same participant completed T0 and T1. An online questionnaire included questionnaires on perceived environmental autonomy support and subjective well-being of persons with complex support needs. A phone interview covered their BPN expressions and transition characteristics. When approximately half of the participants had completed the full study (n = 25) and the other half had only completed T0 (n = 15), data collection was put on hold for approximately 4.5 months due to the start of COVID-19. Upon restarting, the questionnaires were adapted to reflect and assess the impact of the pandemic on the care situation as well (e.g., Embregts et al., 2021). For 14 out of 15 persons with complex support needs, the living situation did not change due to the onset of COVID-19. One person had moved from sheltered care to living with the caregiver during the lockdown period. Five out of 15 participants saw their child as often during the pandemic as before, two participants saw their child more during, and eight participants were temporarily unable to see their child due to the lockdown measures, this period ranged between 4 weeks to 3.5 months. At the restart of T1, these eight latter participants had seen their child again for at least 1 month. Ethical approval was obtained from the Scientific and Ethical Review Committee of the Faculty of Behavioral and Movement Sciences, Vrije Universiteit Amsterdam, The Netherlands (registration number: VCWE-2019-047). For the COVID-19-related adjustments, an amendment was granted.
Materials
Environmental autonomy support
Autonomy support was measured with a questionnaire developed and preliminary validated for this target group (Van Tuyll Van Serooskerken et al., 2024). The basis for this questionnaire was the Perceptions of Parents Scale – College-Student Scale (POPS; Robbins, 1994). The translated and adapted version contains 15 items designed to assess environmental autonomy support for persons with complex support needs, from all key people currently involved in their lives (e.g., parents, relatives, and support staff). Each item starts with “To what extent do you see that important people in your child’s environment …” and is followed by, for example, “… understand his/her feelings?” or “… have enough time for him/her?” Participants rated each item on a Likert scale that ranged from 1 (absolutely disagree) to 5 (absolutely agree). Participants were instructed to answer the question from their child’s perspective as much as possible. After reversing the negatively phrased statements, a higher total score reflected higher autonomy support from the environment. Van Tuyll Van Serooskerken et al. (2024) demonstrated preliminary evidence for construct validity and excellent internal consistency (i.e., α = .92). Internal consistency in the current study was .91 at T0 and .91 at T1.
Basic psychological need expressions
The extent to which persons with complex support needs exhibited expressions of self-determination as reported by participants, was assessed using another specially developed and preliminarily validated questionnaire for this target group (Van Tuyll Van Serooskerken et al., 2024). The instrument has 16 items divided over two subscales (i.e., Noticing signals of Autonomy and Noticing signals of Competence/Relatedness), which were rated on a Likert scale that ranged from 1 (never) to 5 (always). Example items are “My child expresses him/herself when he/she wants something” and “My child enjoys the things he/she can do.” To reduce redundancy and the number of statistical analyses, the total score was used instead of the two separate subscale scores, with a higher score indicating more noticeable behaviors to communicate BPNs. Van Tuyll Van Serooskerken et al. (2024) demonstrated preliminary evidence for construct validity and excellent internal consistencies for both subscales (i.e., α = .88 and .84, respectively). Internal consistencies of the total scores in the current study were .69 at T0 and .77 at T1.
Transition characteristics
Participants were asked about four different transition-related features (i.e., expected impact on possibilities for BPN support, expected impact on BPN expressions, caregiver engagement, and period of impact). Parents’ expectations regarding the potential of transitions to enhance BPNs were assessed using two variables. First, the expected impact on possibilities for BPN support was measured with three questions, based on the work of Verhage et al. (2013). For instance, participants were asked “Working towards the transition, how well do you expect to be able to respond to what your child likes and dislikes?” and moved a sliding bar between 0 (not at all) and 100 (completely). The expected impact on BPN expressions was assessed by adapting three items from the BPN Expressions questionnaire (i.e., one for autonomy, one for competence, and one for relatedness). Participants were asked whether they expected the transition to have a lot of positive = 5, a little positive = 4, none = 3, a little negative = 2, or a lot of negative = 1 influence on this. A total score was calculated for both expected impact variables, by summing the three answers. For caregiver engagement, participants were asked at T1 to estimate the amount of time spent on activities such as searching for information, planning ahead, and making preparations during the period leading up to the transition. This resulted in a subjective indication of caregiver engagement starting from T0 (in months). In addition, participants were asked whether and, if so, how long ago the transition had taken place to indicate the period of impact. This was also converted to time in months.
Subjective well-being
Well-being was assessed with the Dutch translation (Maes et al., 2016) of the Mood, Interest and Pleasure Questionnaire (MIPQ; Ross and Oliver, 2003). This questionnaire contains 23 items on a Likert scale that ranged from 1 (never) to 5 (always) over the past two weeks. Each item had “not applicable” as an additional response category. The total score reflects an overall indicator of positive mood. Petry et al. (2010) demonstrated evidence for construct validity, good internal consistency (i.e., Cronbach’s alphas between .84 and .94), and good test-retest and inter-rater reliabilities (i.e., above .86 and .69 respectively) for all subscales and the total scale when used in this target population. Internal consistencies of the total scores in the current study were .88 at T0 and .92 at T1.
Analysis strategy
Data were analyzed using IBM SPSS Statistics (Version 27). Missing items scores at T0 and T1 within the Environmental Autonomy Support and BPN Expressions questionnaires were replaced when the total percentage of missing items within these questionnaires was well below 5% through single imputation with the participant’s scale average (Eekhout et al., 2015). Missing item scores within the Well-Being questionnaire were imputed as per instrument manual (Maes et al., 2016). Multiple imputation was used to estimate the missing data of the four participants who dropped out at T1 (Sterne et al., 2009). Ten iterated datasets were created (White et al., 2011). Analyses were performed on these datasets and outcomes were pooled.
To assess change during the transition (research question 1), three paired-samples t tests were performed, one for each construct. In addition, the absolute change score was computed for each construct following the methodology outlined by Evans et al. (1998), to evaluate whether a statistically significant number of participants exhibited individual-level alterations, irrespective of its direction. It was checked whether the start of COVID-19 between T0 and T1 was related to the magnitude of change from T0 to T1. Three repeated measures analyses of variance were performed with time as within-subject factor and group (i.e., pre versus peri COVID-19) as between-subject factor. Because the group variable could not be imputed for the 4 participants with a missing T1, these four participants were excluded from this analysis.
To examine associations between changes over time (research question 2), bivariate correlation coefficients were computed. Difference scores were calculated for all three self-determination-related concepts (i.e., T1–T0), with positive scores indicating an increase.
The associations between the two a priori estimations of the impact of the transition and BPN expressions (research question 3) were examined with linear regression. The difference score for BPN expressions calculated by the subtraction of T0 from T1, represented the dependent variable. The two perceived impact scores were added as independent variables in two separate analyses.
Associations between caregiver engagement or period of impact and changes in the self-determination-related concepts (research question 4) were also explored with three linear regression analyses, separately for both predictive variables. The T1 measure of each self-determination-related concept was entered as a dependent variable. Its T0 measure plus caregiver engagement or period of impact were entered in the same step as independent (i.e., control) variables. All statistical tests used an alpha level of .05.
Results
Missing data and data pre-treatment
Descriptive Statistics of Original and Pooled Imputations Dataset for Study Variables.
Note. SE = standard error; SEM = standard error of the mean; BPN = Basic psychological needs.
Changes in self-determination-related concepts
On average, there was no statistically significant change between both measurement points for Environmental Autonomy Support (t(318) = 0.66, p = .51, Cohen’s d = 0.12 (CI –0.18 – 0.45)), BPN Expressions (t(713) = 0.15, p = .89, Cohen’s d = 0.04 (CI –0.27 – 0.35)), and Well-Being (t(284) = −0.06, p = .95, Cohen’s d = 0.04 (CI –0.33 – 0.40)). However, the percentage of participants exceeding the absolute change score was 22.2% for Environmental Autonomy Support, 5.6% for BPN Expressions, and 24% for Well-Being. This indicated that there was a large variation in the difference scores, with increases and decreases in all self-determination-related constructs balancing out. Furthermore, there was no statistical between x within effect of COVID-19 on the change in all three self-determination-related constructs, indicating that the change from T0 to T1 was not statistically different for the pre and peri COVID-19 participants: F(1, 34) = 1.11, p = .30, ηp2 = .03 for Environmental Autonomy Support, F(1, 34) = 0.66, p = .42, ηp2 = .02 for BPN Expressions, and F(1, 23) = 1.16, p = .29, ηp2 = .05 for Well-Being.
Interrelated changes in self-determination-related concepts
There were no statistically significant correlations between the change in Environmental Autonomy Support and change in BPN Expressions, r = .31, p (two-tailed) = .06, between the change in Environmental Autonomy Support and change in Well-Being, r = .03, p (two-tailed) = .88, and between the change in BPN Expressions and change in Well-Being, r = .07, p (two-tailed) = .74.
Caregiver expectations and actual changes in BPN expressions
Linear Regression Results for Change in Basic Psychological Need Expressions.
Note. BPN = Basic psychological needs. Because SPSS does not provide a pooled version of the explained variance of steps 1 and 2, the range of the ten imputations was: R2 = .00 for step 1 Exp. Impact possibilities BPN support T0, R2 = .00 – .02 for step 1 Exp. Impact BPN expressions T0.
Caregiver engagement and period of impact and actual changes in self-determination-related concepts
Multiple Regression Results for Dependent Variable Environmental Autonomy Support T1.
Note. Because SPSS does not provide a pooled version of the explained variance of steps 1 and 2, the range of the ten imputations was: R2 = .35 – .47 for step 1, ΔR2 = .00 – .07 for step 2 caregiver engagement, ΔR2 = .01 – .05 for step 2 period of impact.
Multiple Regression Results for Dependent Variable Basic Psychological Need Expressions T1.
Note. BPN = Basic psychological needs. Because SPSS does not provide a pooled version of the explained variance of steps 1 and 2, the range of the ten imputations was: R2 = .43 – .58 for step 1, ΔR2 = .00 – .03 for step 2 caregiver engagement, ΔR2 = .00 – .01 for step 2 period of impact.
Multiple Regression Results for Dependent Variable Well-Being T1.
Note. Because SPSS does not provide a pooled version of the explained variance of steps 1 and 2, the range of the ten imputations was: R2 = .32 – .43 for step 1, ΔR2 = .01 – .09 for step 2 caregiver engagement, ΔR2 = .00 – .01 for step 2 period of impact.
Discussion and conclusion
Discussion
Self-determination-related concepts remained stable on average during transition but sizeable subgroups went to reliable increases or decreases. Whether participants expected possibilities for BPN support or the person’s BPN expressions to increase or decrease was not associated with the actual direction of change. Caregiver engagement in preparing for the transition, as well as the time elapsed since the transition occurred, showed no association with observed changes either.
Extant studies suggest that self-determination and the support thereof are interrelated over time in persons with and without intellectual or other disabilities (Audet et al., 2021; Gillison et al., 2008; Kins et al., 2009; Nadig et al., 2018; Oliveira et al., 2014; Seong et al., 2015; Wehmeyer et al., 2011). However, these studies did not focus on people with complex support needs. Given that intellectual disabilities hamper opportunities to express and fulfill preferences and desires (Wehmeyer and Abery, 2013; Wehmeyer and Shogren, 2017), the outcomes of transitions may be more difficult to control, despite everyone’s best efforts.
The importance of further conceptual work to understand self-determination in the face of transitions in this population is highlighted by the fact the percentage of participants exceeding the absolute change score surpassed the 5% benchmark in all three constructs. Importantly, changes could not be predicted by caregivers, which in itself is noteworthy and in need of replication as it may have implications for supporting families. Research designs like case studies or within-person designs might track the often idiosyncratic and person-specific characteristics associated with these changes more accurately. Complementing this study with qualitative research methods to further explore what could make it so difficult for caregivers to predict the course of transitions might also be interesting. Another suggestion for further research would be to extend the duration between measurement points, which might reveal more pronounced changes, although this proposition remains speculative. Comprehensive investigations into the long-term impact of caregiver or environmental factors on changes in various domains remain underexplored within this population (Van keer & Maes, 2018).
Limitations
In general, recruiting participants from this population for scientific research is challenging due to the low prevalence of individuals with complex support needs (Maes et al., 2021). Although recruitment was conducted through multiple channels, each of these channels had a limited reach. Reaching individuals without social media or a connection to parent organizations proved more challenging, potentially limiting the generalizability of the results. Both recruitment and data collection within this study were also severely limited by the sudden emergence of the pandemic. The resulting small sample size undeniably raises methodological issues (e.g., generalizability and statistical power), meaning that results should be interpreted with caution (Faber and Fonseca, 2014). In addition, participants with various different life transitions were included, rather than opting for inclusion based on one type of transition. Different transitions may present distinct opportunities for self-determination, which could have biased the results. Future studies with larger samples could explore subgroup differences to address this issue. Additionally, focusing on one specific type of transition such as moving from family home to sheltered care (e.g., Vereijken et al., 2024) could help reduce this bias, although it would limit the generalizability of the results to other transitions. Subgroup analyses based on living arrangements would also be valuable, considering that 35% of individuals with complex support needs did not reside in the family home. Parents of these individuals may be involved differently in transitions compared to parents of those living at home. Another factor to consider in follow-up research may be the age of persons with complex support needs. In typically developing children, autonomy support from parents differs across age groups (Grolnick, 2009; Vrolijk et al., 2022). Caregivers might see more opportunities for self-determination support during transitions for younger persons (e.g., adolescents) than for older persons with complex support needs (van Tuyll van Serooskerken et al., 2022). Finally, the BPN expressions questionnaire identifies observable signals in persons with complex support needs, but it may not fully reflect their actual expressions of BPNs or whether those needs are genuinely satisfied or unsatisfied. Also, changes in BPN expressions may be challenging to discern when caregivers, serving as informants in this study, are the only source of information. Including others, such as support staff, could enhance the precision of representing the subjective experiences of BPN satisfaction of persons with complex support needs. Nevertheless, proxy informants often struggle to accurately recognize and define internal states, such as psychological needs and subjective well-being, in individuals with intellectual disabilities, leading to variability in their assessments (De Geus-Neelen et al., 2014, 2019; Scott and Havercamp, 2018; Webb et al., 2024). Additionally, objectively measuring the construct of environmental autonomy support is challenging, as parents are inherently part of this environment. Therefore, individual change trajectories might be confounded with measurement error. Nonetheless, the degree to which the environment may be sensitive and empathic to signals could potentially influence the degree to which persons with complex support needs express them, and the reverse might also hold true (Hostyn and Maes, 2009; Skarsaune, 2024). In conclusion, it is imperative to exercise prudence when interpreting findings derived from proxy research, as they inherently represent an interpretation of reality, despite its origin from individuals intimately acquainted with the person under study (Nieuwenhuijse et al., 2023). It therefore remains crucial to develop innovative methods to capture the experiences of individuals within this target population, such as utilizing physiological data (Hammann et al., 2022) or modern Information and Communication Technologies (ICT; Kosiedowski et al., 2019).
Conclusion
This is a first longitudinal exploration of self-determination-related constructs in persons diagnosed with severe or profound intellectual and multiple disabilities. Previous research in the disability field on autonomy support, basic psychological needs, and quality of life has primarily focused on individuals with mild intellectual disability (e.g., Frielink et al., 2018; Shogren and Broussard, 2011). This study delved into the dynamics of self-determination during transitions, finding relative stability in self-determination-related constructs at the group level, even amidst disruptions like the COVID-19 lockdown. Neither pre-transition expectations, engagement levels during the transition, nor the time elapsed after the transition, significantly influenced these constructs. However, significant individual-level changes highlight the need for a more personalized approach to understanding these trajectories. Given that self-determination-related constructs changed in ways unexpected for caregivers and unrelated to their engagement in these transitions, there is a need for deeper insight into how context influences self-determination in this population. By incorporating a contextual perspective, this study reveals the potential for these variables to enhance research and address complex challenges in caring for individuals with complex support needs. The longitudinal approach adopted here offers valuable insights into the developmental aspects of self-determination, with the potential to inform future research and interventions, ultimately improving care for people with more severe intellectual disabilities.
Supplemental Material
Supplemental Material - Change in self-determination-related constructs in persons with severe or profound intellectual and multiple disabilities in the context of transitions
Supplemental Material for Change in self-determination-related constructs in persons with severe or profound intellectual and multiple disabilities in the context of transitions by Jacqueline M. van Tuyll van Serooskerken, Agnes M. Willemen, Petri J. C. M. Embregts, and Carlo Schuengel in Journal of Intellectual Disabilities
Footnotes
Acknowledgments
The authors would like to express their sincere gratitude to the parents, relatives, and family members with complex support needs who took the time and effort to participate in this study. Furthermore, the authors would like to thank the care and client advocacy organizations ’s Heeren Loo, EMB Nederland, and Per Saldo sincerely for their support with recruitment.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by The Netherlands Organisation for Health Research and Development (ZonMw) Nationaal Programma Gewoon Bijzonder, grant number 845004005; and Stichting Wetenschappelijk Onderzoek of ’s Heeren Loo, The Netherlands (i.e., partial funding).
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
