2026, Vol. 6, Issue 10, Part A

Keywords: Sensory processing, Toilet training, Autism spectrum disorder, Interoception, Tactile responsivity, Proprioception, Adaptive behavior, Sensory subtypes, Toileting difficulty, Sensory-adapted intervention

Introduction

While most typically developing children achieve daytime continence between ages 2 and 4, autistic children often reach this milestone years later, and a sizeable proportion remain partially or fully incontinent well into middle childhood (Leader et al., 2018; Kroeger et al., 2009) ^{[14, 19]}. The gap between expected and actual toileting age creates ongoing stress for families, limits participation in school and community activities, and restricts access to inclusive educational placements where toilet training is a prerequisite (Bal et al., 2015) ^[3]. Despite the scale of the problem, the reasons why toilet training is so much harder for autistic children have received surprisingly little focused investigation.

The dominant approach to toilet training in autism has drawn on applied behavior analysis, using reinforcement schedules, systematic prompting, and structured sit times to shape toileting behavior (Kroeger et al., 2009) ^[19]. These methods have produced documented gains, but they work better for some children than others, and the reasons for uneven outcomes remain unclear (Leader et al., 2018) ^[14]. One explanation that has gained attention in the occupational therapy literature is that sensory processing differences, which affect the majority of autistic children, may interfere with specific components of the toileting sequence. Toilet training requires a child to detect internal bladder and bowel signals, tolerate the sensory features of the bathroom environment, sit on a surface that may feel unfamiliar or unstable, and manage clothing adjustments that involve fine motor and tactile processing (Beaudry-Bellefeuille et al., 2023) ^[1]. Each of these steps engages sensory systems that are often atypical in autism (Tomchek et al., 2007; Ben-Sasson et al., 2009) ^{[17, 18]}.

Sensory processing differences in autism are not a single phenomenon. Research using the Sensory Profile and related instruments has identified distinct sensory subtypes, including patterns characterized by seeking, avoiding, under-responsivity, and over-responsivity, each with different behavioral signatures and different implications for daily functioning (Ausderau et al., 2016; Lane et al., 2010) ^{[10, 20]}. Tomchek and colleagues ^[7] reported that sensory subtypes in preschool-aged autistic children were connected to different adaptive behavior profiles, and Ausderau and colleagues ^[10] showed that sensory subtypes were linked to communication, socialization, and daily living outcomes. However, the specific connection between sensory subtypes and toileting outcomes has not been tested directly.

A parallel line of research has examined interoception, the sense of internal bodily signals, as a possible contributor to toileting delay. DuBois and colleagues ^[9] reviewed evidence that interoceptive processing is altered in autism, and Price and colleagues ^[5] found through meta-analysis that autistic individuals show reduced interoceptive accuracy across multiple physiological domains. (Beaudry-Bellefeuille and Lane) ^[1] proposed a clinical model linking interoceptive differences to toileting difficulties, arguing that children who struggle to perceive bladder fullness signals will not develop the anticipatory awareness that successful toilet training requires. This theoretical framework has not been tested with a sample that also measures external sensory processing domains alongside interoception.

The present research addressed three questions: (1) which sensory processing domains are the strongest predictors of toilet-training difficulty in autistic children, (2) whether sensory subtypes differ in toileting outcomes, and (3) whether a sensory-adapted toileting intervention produces measurable improvements in a subset of participants. This research was conducted in Medellín, Colombia, where access to sensory-informed developmental services varies widely and where no previous investigation of the sensory-toileting connection has been published.

Material and methods

Material

Participants were 112 autistic children (78 male, 34 female) aged 3 to 8 years (M = 5.2, SD = 1.4) recruited from three developmental clinics affiliated with the Medellín Foundation for Child Development Research between January 2022 and December 2023. All children had a confirmed diagnosis of autism spectrum disorder based on DSM-5 criteria and received by a licensed clinical psychologist or developmental pediatrician. Children were excluded if they had a co-occurring medical condition known to affect continence (e.g., spina bifida, chronic urinary tract infections) or if they were receiving medication with known effects on bladder function. The sample was classified by cognitive level using the Bayley Scales of Infant Development-III or the Wechsler Preschool and Primary Scale of Intelligence-IV, depending on age: 42 children (37.5%) had intellectual disability ($IQ<70$), 38 (33.9%) had borderline cognitive functioning (IQ 70-84), and 32 (28.6%) had average or above-average cognitive ability ($IQ\ge 85$). Autism severity was measured using the Autism Diagnostic Observation Schedule-2 (ADOS-2) calibrated severity score.

Methods

Sensory processing was assessed using the Sensory Profile-2 (SP-2), a caregiver-report questionnaire that yields scores across six sensory domains (tactile, auditory, visual, vestibular, proprioceptive, and oral) and four sensory processing quadrants (seeking, avoiding, sensitivity, and registration) (Little et al., 2022; Tomchek et al., 2007) ^{[2, 17]}. The SP-2 was administered in Spanish using the validated Latin American adaptation. Toileting status was recorded using a structured interview that classified each child as toilet trained (consistently dry during daytime for at least 3 months), in progress (actively engaged in toilet training with partial success), or not yet trained (no systematic toilet-training program initiated or no meaningful progress). Toilet-training difficulty was quantified using the Toilet Training Difficulty Index (TTDI), a 15-item composite score developed for this research that captures resistance to sitting on the toilet, avoidance of bathroom environments, lack of awareness of bladder fullness, distress during clothing changes, and frequency of toileting accidents (Hedgecock et al., 2023) ^[4]. The TTDI showed acceptable internal consistency (Cronbach alpha = 0.84) and test-retest reliability (ICC = 0.79) in a pilot sample of 30 children prior to the main data collection.

Sensory subtypes were derived using k-means cluster analysis on the six SP-2 domain scores, with the number of clusters determined by silhouette analysis. Hierarchical multiple regression was used to identify sensory predictors of TTDI scores, with age, sex, cognitive level, and ADOS-2 severity entered in the first block and SP-2 domain scores entered in the second block. Between-group comparisons across sensory subtypes used one-way ANOVA with Games-Howell post hoc tests. A subsample of 24 children with the highest TTDI scores participated in a 12-week sensory-adapted toileting intervention that included graded tactile exposure to bathroom surfaces, proprioceptive preparation activities before toilet sits, visual schedules with sensory cues, and interoceptive awareness training using body-scan exercises (Beaudry-Bellefeuille et al., 2023; Schaaf et al., 2018) ^{[1, 11]}. Pre-post comparisons used paired-samples t-tests and Cohen d effect sizes.

Ethical considerations

This research received approval from the Ethics Committee of the Medellín Foundation for Child Development Research (Protocol MFCDR-2021-038). All procedures followed the principles of the Declaration of Helsinki. Written informed consent was obtained from each child’s main caregiver, and verbal assent was obtained from children who were able to provide it. Caregivers were informed of their right to withdraw from the research at any point without consequences for their child’s clinical care. All data were de-identified and stored on encrypted institutional servers accessible only to the research team.

Patient characteristics

The 112 children in the sample ranged from 3.0 to 8.4 years of age. The majority were male (69.6%), which is consistent with the higher prevalence of autism diagnoses in boys. By toileting status, 36 children (32.1%) were classified as toilet trained, 38 (33.9%) as in progress, and 38 (33.9%) as not yet trained. Mean ADOS-2 calibrated severity scores were 7.1 (SD = 1.8) for the full sample, with no significant differences across toileting status groups (F_(2,109) = 1.42, p = 0.25). Mean SP-2 total scores differed across groups: toilet-trained children had a mean of 58.3 (SD = 14.2), in-progress children had a mean of 72.6 (SD = 16.8), and not-yet-trained children had a mean of 84.1 (SD = 15.3), with the between-group difference reaching significance (F_(2,109) = 22.47, p < 0.001). Children in the not-yet-trained group were younger on average (M = 4.6, SD = 1.2) than those in the trained group (M = 6.1, SD = 1.3), and cognitive level was lower in the not-yet-trained group ($\text{mean}IQ=68.4,SD=16.2$) compared with the trained group ($\text{mean}IQ=82.7,SD=14.8$).

Results

The hierarchical regression analysis revealed that sensory processing domains added meaningful predictive power beyond demographic and diagnostic variables. After controlling for age, sex, cognitive level, and ADOS-2 severity, the six SP-2 domain scores explained an additional 16% of the variance in TTDI scores (ΔR² = 0.16, p < 0.001). Tactile processing emerged as the single strongest predictor (β = 0.38, p < 0.001), followed by proprioceptive processing (β = 0.31, p < 0.001) and vestibular processing (β = 0.22, p = 0.003). Auditory processing reached significance (β = 0.18, p = 0.024), while oral and visual domains did not independently predict TTDI scores after the other domains were accounted for.

Cluster analysis of the six SP-2 domain scores yielded five sensory subtypes, consistent with previous classification work using similar instruments (Ausderau et al., 2016; Lane et al., 2010) ^{[10, 20]}. The over-responsive subtype (n = 16) had the highest mean TTDI score (M = 68.4, SD = 12.1), while the sensory-seeking subtype (n = 34) had the lowest (M = 44.2, SD = 15.3), with the between-group difference reaching significance (F_(4,107) = 14.83, p < 0.001). Post hoc comparisons showed that the over-responsive group scored higher on the TTDI than all other subtypes (all p < 0.01), and the seeking group scored lower than the avoiding and mixed groups (p < 0.05).

The intervention subsample (n = 24) showed meaningful improvements across all measured toileting outcomes after 12 weeks of the sensory-adapted protocol. The largest effect was observed for the overall TTDI score (d = 1.45), with the reduction from a mean of 62.4 to 41.7 representing a shift from severe to moderate difficulty. Bladder awareness, which had been one of the lowest-scoring components at baseline, improved by 4.5 points (d = 1.02), suggesting that the interoceptive awareness training component may have helped children attend to internal signals they had previously missed. Toileting resistance and sitting duration also improved with large effect sizes, consistent with the expectation that graded tactile exposure would reduce the aversive quality of the toilet and bathroom environment. Accidents per week decreased from 8.6 to 4.1, though considerable variability remained within the group.

When the relationships between sensory domains and specific toileting components were examined, tactile processing showed the strongest links with toileting resistance (r = 0.54, p < 0.001) and bathroom tolerance (r = 0.49, p < 0.001), while proprioceptive processing was most strongly connected to sitting duration (r = -0.47, p < 0.001) and clothing management (r = 0.41, p < 0.001). Vestibular processing was connected to sitting duration and balance-related toileting behaviors (r = 0.38, p < 0.001). These domain-specific patterns support the argument that different sensory systems contribute to different components of the toileting sequence, and that intervention planning should target the domains most relevant to each child’s specific barriers (Beaudry-Bellefeuille et al., 2023; Schaaf et al., 2018) ^{[1, 11]}.

Discussion

The finding that tactile and proprioceptive processing were the strongest predictors of toilet-training difficulty aligns with clinical observations that many autistic children resist sitting on unfamiliar surfaces and struggle with the clothing adjustments required for independent toileting (Beaudry-Bellefeuille et al., 2023; Hedgecock et al., 2023) ^{[1, 4]}. (Beaudry-Bellefeuille and Lane) ^[1] proposed that tactile defensiveness may make the physical contact with the toilet seat aversive, while proprioceptive processing differences may reduce the child’s ability to maintain a stable seated posture, both of which would directly interfere with successful toilet use. The present findings provide empirical support for that model. The effect sizes for these two domains were substantially larger than for auditory and visual processing, pointing to the importance of body-contact senses over environmental senses in the toileting context.

The identification of five sensory subtypes mirrors the work of Ausderau and colleagues ^[10] and Lane and colleagues ^[20], who found similar clustering patterns using the Sensory Profile in North American samples. The present research extends their findings by showing that these subtypes are connected to different toileting trajectories. The over-responsive subtype, which represented the smallest group (14.3%), had the worst toileting outcomes, suggesting that these children may need the most intensive sensory preparation before toilet-training programs begin. In contrast, children in the sensory-seeking group, while still delayed relative to typically developing peers, showed the most promising training outcomes, possibly because their sensory-seeking tendencies can be directed toward the proprioceptive and tactile features of toileting activities rather than away from them.

The interoceptive dimension of toileting has been less thoroughly researched than external sensory processing. DuBois and colleagues ^[9] reviewed evidence that autistic individuals process internal bodily signals differently, and Price and colleagues ^[5] confirmed through meta-analysis that interoceptive accuracy is reduced in autism. The improvement in bladder awareness scores following the sensory-adapted intervention in this research backs up the theoretical link between interoception and toileting proposed by (Beaudry-Bellefeuille and Lane) ^[1], though the absence of a control group limits causal conclusions. The finding that bladder awareness improved with a large effect size (d = 1.02) over 12 weeks suggests that body-scan and interoceptive training exercises may be a productive addition to existing toilet-training protocols, an approach that Green and colleagues ^[12] have advocated for based on their work with sensory behaviours in autistic children.

The Latin American context of this research is worth noting. Leader and colleagues ^[14] documented that toileting problems in autistic children are widespread across cultural settings, but the available evidence base for sensory-informed toileting interventions draws almost exclusively from high-income English-speaking countries. Schaaf and colleagues ^[11] showed the effectiveness of Ayres Sensory Integration approaches in controlled trials, yet these methods require specialized training and equipment that may not be available in many Colombian clinical settings. The sensory-adapted protocol used here was designed to be implementable by caregivers with minimal professional supervision, using commonly available materials for tactile and proprioceptive activities. This design feature may increase the relevance of the approach for resource-limited settings, though replication with a larger sample and a comparison group would be needed to confirm its effectiveness (Kojovic et al., 2022; McCormick et al., 2016) ^{[6, 13]}.

McCormick and colleagues ^[13] found that sensory symptoms in autism are relatively stable over time, which would mean that sensory processing differences identified during the preschool years are likely to persist into later childhood unless targeted by intervention. This temporal stability supports the argument that sensory assessment should be incorporated into toilet-training readiness evaluations, rather than waiting for training difficulties to emerge. The regression model in this research accounted for 37% of the variance in toilet-training difficulty, leaving 63% unexplained. Variables not measured here, including specific interoceptive accuracy for bladder signals, caregiver training practices, previous toileting experiences, and behavioral rigidity, likely account for portions of the remaining variance (Williams et al., 2023; Crasta et al., 2020; Neil et al., 2016) ^{[8, 15, 16]}.

Limitations

Several limitations should be considered when interpreting these findings. The cross-sectional design of the main analysis prevents causal conclusions about the relationship between sensory processing and toileting difficulty; it remains possible that unsuccessful toileting experiences shape sensory responses rather than the reverse. The sensory-adapted intervention component lacked a randomized control group, so improvements cannot be attributed exclusively to the intervention rather than to maturation, practice, or caregiver attention effects. The SP-2 is a caregiver-report instrument, and caregiver perceptions of sensory behavior may not align perfectly with objective measures of sensory processing. The sample was drawn from developmental clinics in a single Colombian city, and families who attend these clinics may not be representative of the broader population of autistic children in the region. The TTDI was developed for this research and has not yet been validated in independent samples. Finally, interoceptive processing was assessed indirectly through the bladder awareness subscale of the TTDI rather than through a direct measure of interoceptive accuracy, limiting conclusions about the specific interoceptive mechanisms involved (Price et al., 2022; DuBois et al., 2016) ^{[5, 9]}.

Conclusion

This research set out to determine whether sensory processing differences predict toilet-training difficulty in autistic children and whether sensory subtypes are connected to different toileting outcomes. The results confirm that sensory processing, particularly in the tactile and proprioceptive domains, is a meaningful predictor of how much difficulty autistic children experience during toilet training, even after accounting for age, cognitive ability, and autism severity. The five sensory subtypes identified in this sample showed distinct toileting profiles, with the over-responsive subtype experiencing the most severe toileting difficulties and the sensory-seeking subtype showing the best outcomes.

The practical message emerging from these findings is straightforward. Before beginning a toilet-training program for an autistic child, clinicians and caregivers should assess the child’s sensory processing profile across multiple domains. Children who show high levels of tactile over-responsivity may benefit from gradual exposure to toilet-related surfaces and textures before any expectation of toilet use is introduced. Children with proprioceptive processing differences may need seating supports, weighted lap pads, or preparatory heavy-work activities to help them maintain a comfortable and stable position during toilet sits. Children who show reduced bladder awareness may need explicit interoceptive training that teaches them to attend to and interpret internal signals. These sensory strategies are not replacements for established behavioral approaches to toilet training; rather, they address the sensory barriers that may prevent behavioral approaches from working as intended.

The pilot intervention results, while preliminary, are encouraging. The large effect sizes observed for toileting resistance, bladder awareness, and sitting duration suggest that a relatively brief period of sensory-adapted support can meaningfully improve toileting readiness, even in children who have been resistant to previous training attempts. The finding that the intervention was implementable by caregivers with minimal professional support adds to its potential utility in settings where access to specialized occupational therapy or sensory integration services is limited.

Future work should test sensory-adapted toileting interventions through randomized controlled trials with active comparison conditions, develop direct interoceptive measures suitable for young autistic children, and examine whether early sensory intervention during the preschool period can prevent the toileting difficulties that many autistic children encounter. Longitudinal designs that follow children from initial sensory assessment through toilet-training outcomes would clarify the direction of the sensory-toileting relationship and identify the developmental windows during which sensory-adapted approaches are most effective. Cross-cultural replication across Latin American and other under-represented regions would strengthen the evidence base and inform the development of culturally responsive clinical guidelines.

References

1. 1.Beaudry-Bellefeuille I, Lane SJ. Interoception, body awareness, and toileting difficulties in children with neurodevelopmental conditions. Am J Occup Ther. 2023;77(1):7701205010.
2. 2.Little LM, Dean E, Tomchek SD, Dunn W. Classifying sensory profiles of children in the general population. Child Care Health Dev. 2022;48(3):361-368.
3. 3.Bal VH, Kim SH, Cheong D, Lord C. Daily living skills in individuals with autism spectrum disorder from 2 to 21 years of age. Autism. 2015;19(7):774-784.
4. 4.Hedgecock JB, Stanton M, Engel C. Toileting difficulties and sensory sensitivity in young children: A cross-sectional analysis. J Pediatr Nurs. 2023;68:e34-e41.
5. 5.Price R, Marsh AJ, Magistro D. Interoception in autism: A systematic review and meta-analysis. Neurosci Biobehav Rev. 2022;141:104803.
6. 6.Kojovic N, Natraj S, Franchini M, Schaer M. Sensory processing trajectories and adaptive functioning in autism spectrum disorder. Mol Autism. 2022;13(1):42.
7. 7.Tomchek SD, Little LM, Myers J, Dunn W. Sensory subtypes in preschool children with autism spectrum disorder. J Autism Dev Disord. 2022;52(5):2224-2232.
8. 8.Williams KL, Wierzbicki R, Henderson H, Lerner MD. Daily living skills and sensory features across adolescence in autism. Autism Res. 2023;16(4):856-868.
9. 9.DuBois D, Ameis SH, Lai MC, Casanova MF, Bhatt S. Interoception in autism spectrum disorder: A review. Int J Dev Neurosci. 2016;47(Pt B):74-80.
10. 10.Ausderau KK, Sideris J, Little LM, Furlong M, Bulluck JC, Baranek GT. Sensory subtypes and associated outcomes in children with autism spectrum disorders. Autism Res. 2016;9(12):1316-1327.
11. 11.Schaaf RC, Dumont RL, Arbesman M, May-Benson TA. Efficacy of occupational therapy using Ayres Sensory Integration: A systematic review. Am J Occup Ther. 2018;72(1):7201190010.
12. 12.Green D, Chandler S, Charman T, Simonoff E, Baird G. Brief report: DSM-5 sensory behaviours in children with and without an autism spectrum disorder. J Autism Dev Disord. 2016;46(11):3597-3606.
13. 13.McCormick C, Hepburn S, Young GS, Rogers SJ. Sensory symptoms in children with autism spectrum disorder, other developmental disorders and typical development: A longitudinal examination. Autism. 2016;20(5):572-579.
14. 14.Leader G, Francis K, Mannion A, Chen J. Toileting problems in children and adolescents with parent-reported diagnoses of autism spectrum disorder. J Dev Phys Disabil. 2018;30(3):307-327.
15. 15.Crasta JE, Salzinger E, Lin MH, Gavin WJ, Davies PL. Sensory processing and attention profiles among children with sensory processing disorders and autism spectrum disorders. Front Integr Neurosci. 2020;14:22.
16. 16.Neil L, Olsson NC, Pellicano E. The relationship between intolerance of uncertainty, sensory sensitivities, and anxiety in autistic and typically developing children. J Autism Dev Disord. 2016;46(6):1962-1973.
17. 17.Tomchek SD, Dunn W. Sensory processing in children with and without autism: A comparative study using the Short Sensory Profile. Am J Occup Ther. 2007;61(2):190-200.
18. 18.Ben-Sasson A, Hen L, Fluss R, Cermak SA, Engel-Yeger B, Gal E. A meta-analysis of sensory modulation symptoms in individuals with autism spectrum disorders. J Autism Dev Disord. 2009;39(1):1-11.
19. 19.Kroeger KA, Sorensen-Burnworth R. Toilet training individuals with autism and other developmental disabilities: A critical review. Res Autism Spectr Disord. 2009;3(3):607-618.
20. 20.Lane AE, Young RL, Baker AEZ, Angley MT. Sensory processing subtypes in autism: Association with adaptive behavior. J Autism Dev Disord. 2010;40(1):112-122.