Early psychological intervention model based on sensory integration and parental empowerment for dietary abnormalities in children with autism

Abstract

BACKGROUND

Autistic children commonly have dietary abnormalities, lacking efficient standardized early intervention.

AIM

To explore the effect of an early psychological intervention on abnormal eating behaviors in children with autism spectrum disorder (ASD).

METHODS

Children (n = 150) with abnormal eating behaviors linked to ASD who were admitted to the Second Affiliated Hospital of Soochow University between January 2022 and December 2024 were randomly divided into two groups (control and intervention; n = 75/group). A total of 142 patients (intervention, 72; control, 70) were followed up. The control group received a conventional rehabilitation intervention, while the intervention group received an early psychological intervention combining sensory integration and parental empowerment. The Child Eating Behavior Scale, Autism Behavior Scale, Child Anxiety and Depression Scale, Sensory Integration Development Scale, and Parental Care Ability Scale were used for assessment before the intervention (T0), at 6 months of intervention (T1), and at 12 months of follow-up (T2). The rates of reaching the standard of nutritional intake and regular diet, ability to eat autonomously, compliance with intervention, effective rate of intervention, and recurrence rate in the groups were compared. The bootstrap method was used to analyze the chain mediating effect of improvement in parental care ability and sensory integration function.

RESULTS

At T1 and T2, the Child Eating Behavior Scale, Autism Behavior Scale, and Child Anxiety and Depression Scale scores of the children in the intervention group were significantly lower than those in the control group. The Sensory Integration Development Scale and Parental Care Ability Scale scores were significantly higher than in the control group. The rates of reaching the nutritional intake standard and regular diet, ability to eat autonomously, total effective rate of intervention, and compliance rate of children in the intervention group were significantly higher. The recurrence rate was also significantly lower in the intervention group. Mediating effect analysis showed that the improvement of parental care ability and sensory integration function played a partial chain mediating role in the joint intervention promoting the improvement of children’s dietary behavior.

CONCLUSION

Early psychological interventions based on sensory integration and parental empowerment can effectively improve abnormal eating behaviors in children with ASD, with good clinical value.

Key Words: Autism; Abnormal dietary behavior; Sensory integration; Parental empowerment; Early psychological intervention; Chain mediating effect

Core Tip: Early psychological interventions that combine sensory integration and parental empowerment effectively improve dietary abnormalities in children with autism spectrum disorder. This mechanism is mediated by enhanced parental care abilities and improved sensory integration in children. This study aims to explore the effect of an early psychological intervention on abnormal eating behaviors in children with autism spectrum disorder.

INTRODUCTION

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized mainly by impaired social communication, repetitive and stereotypical behaviors, and narrow interests[1]. It usually begins in early childhood, and its diagnostic rate has been on the rise in recent years. Abnormal eating behavior is a common concomitant problem in children with ASD[2]. At least half of children with ASD have difficulties in feeding or eating, which mainly manifests as food refusal, selective eating, slow eating, and abnormal eating behaviors[3]. Such problems can lead to an imbalance in nutrient intake, increase nutritional risks, and affect growth and development, while significantly increasing the burden of daily meal management on caregivers[4]. Currently, interventions for dietary issues in ASD include various methods, such as dietary therapy, behavioral intervention, and adjustment of the dining environment[5]. However, the overall evidence suggests that the effects of a single model still exhibit heterogeneity. Individualized, multidimensional, and systematic interventions are more clinically significant. Research has shown that abnormal eating behaviors in children with ASD are closely related to sensory processing difficulties, especially oral sensory processing abnormalities, which are more prominently associated with food rejection and extensive feeding problems[6]. Sensory integration-related interventions are expected to alleviate abnormal eating manifestations by improving sensory processing and adaptive responses[7]. As the main intervention implementers, caregivers’ knowledge, skills, and participation in quality of care also directly affected the intervention effect and maintainability[8]. Based on this, this study combined the theory of sensory integration with the concept of parental empowerment to construct an early psychological intervention model for abnormal eating behaviors in children with ASD and explored its intervention effects and possible mechanisms of action, with the aim of providing a basis for standardized clinical intervention.

MATERIALS AND METHODS

General information

Between January 2022 and December 2024, the Second Affiliated Hospital of Soochow University and Wujiang District Mental Rehabilitation selected 150 children diagnosed with ASD with abnormal eating and behavioral patterns in the hospital to participate in the study. Inclusion criteria: (1) Meeting the diagnostic criteria for ASD in the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition), and confirmed by a attending psychiatrist; (2) Aged 2 to 6 years old, in the critical period of early intervention; (3) A clear abnormal eating behavior, with a score of ≥ 60 points on the Children’s Eating Behavior Scale (CEBQ); (4) Score of < 80 for the Sensory Integration Development Scale (SITS), indicating sensory integration dysfunction; (5) Parents were the main caregivers, possessed basic communication skills, were willing to participate in the study, and completed the entire follow-up; and (6) Patient had no serious physical or neurological, diseases, other comorbidities, or mental disorders. Exclusion criteria were: (1) Withdrawal halfway through the study, loss to follow-up, or poor compliance (intervention compliance rate < 70%); (2) Severe malnutrition, congenital digestive tract diseases, and other physical diseases that affect diet; (3) Parents with serious psychological barriers unable to cooperate in completing empowerment training and intervention guidance; and (4) Patient had received similar sensory integration or parental training interventions.

A total of 150 children were divided into the control and intervention groups using the random number table method, with 75 children per group. This study employed a computer-generated random number table for randomization. The randomization sequence was independently generated by a researcher who was not involved in the clinical assessments or the delivery of the intervention, and allocation was concealed using numbered, sealed envelopes. After the pediatric patients had completed their baseline assessments, another researcher assigned them to groups according to the randomization sequence to minimize selection bias. Owing to the marked differences in the nature of the interventions, it was not possible to implement blinding for the study participants or the intervention staff; however, the outcome assessors and statisticians were not involved in the intervention process.

During the intervention process, 3 patients dropped out of the intervention group (2 due to family relocation and 1 due to parents’ inability to cooperate with the training) and 5 patients dropped out of the control group (3 due to the aggravation of the children’s conditions and transfer to another hospital, and 2 due to poor intervention effect and self-withdrawal). Eventually, 142 patients were followed up, including 72 in the intervention group and 70 in the control group. There were no statistically significant differences in the general data between the two groups (P > 0.05), and they were comparable (Table 1). The Medical Ethics Committee of our hospital approved this study, and all parents signed the informed consent form.

Table 1 Comparison of baseline data between the two groups, mean ± SD/n (%).

	Intervention group (n = 72)	Control group (n = 70)	χ2/t	P value
Gender (male/female)	48/24	46/24	0.014	0.905
Age (years)	4.25 ± 1.08	4.32 ± 1.15	0.374	0.709
Course of the disease (months)	18.62 ± 5.37	19.15 ± 5.62	0.575	0.566
Severity of autism			0.119	0.942
Mild	28 (38.89)	26 (37.14)
Moderate	36 (50.00)	35 (50.00)
Severe	8 (11.11)	9 (12.86)
Abnormal types of eating behavior
Food refusal	42 (58.33)	40 (57.14)	0.021	0.886
Selective eating	51 (70.83)	49 (70.00)	0.012	0.913
Bradyphagia	38 (52.78)	36 (51.43)	0.026	0.872
Emotional eating	32 (44.44)	30 (42.86)	0.036	0.849
Educational level of parents			0.019	0.990
Junior high school and below	18 (25.00)	17 (24.29)
Senior high school/secondary specialized school	27 (37.50)	26 (37.14)
College degree or above	27 (37.50)	27 (38.57)

Research methods

The control group received a conventional rehabilitation intervention with an intervention period of 6 months and a follow-up period of 12 months. Specific measures included: (1) Regular rehabilitation training, rehabilitation training for core symptoms of ASD three times a week, including social communication training and stereotyped behavior correction training, with each session lasting 45 minutes; (2) Basic dietary guidance, during which the nutritionist developed a personalized dietary plan based on the child’s age and nutritional needs, inform the parents of dietary combinations, eating patterns, and other precautions, and conduct a phone follow-up once a month to answer the parents’ questions related to diet; and (3) Regular health education, at the beginning of the intervention, a manual on dietary care for ASD was distributed to parents, explaining the common manifestations of abnormal dietary behaviors in ASD and basic care methods. No systematic sensory integration training, parent empowerment training, or emotional interventions were conducted.

Trained rehabilitation therapists conducted the sensory integration training. Vestibular training included swing activities, balance exercises, rotation stimulation, and movement coordination tasks. Proprioceptive training included weight-bearing activities, pushing and pulling exercises, jumping activities, and deep pressure stimulation. Tactile training included exposure to different textures and food-related sensory experiences. Food exposure followed a gradual sequence of touching, smelling, tasting, and accepting food to reduce sensory-related food rejection.

The parent empowerment program consisted of 12 offline sessions and weekly online guidance over six months. The curriculum included ASD knowledge, identification of abnormal eating behaviors, dietary management skills, family sensory integration training, positive reinforcement strategies, emotional regulation, data recording, and risk identification. Mindfulness training was delivered weekly online and included breathing awareness, body scanning, emotion recognition, and stress regulation exercises to improve caregivers’ psychological adjustment.

The intervention group received early psychological intervention combining sensory integration and parental empowerment. The intervention and follow-up periods were 6 and 12 months, respectively. Before the intervention, a team of professionals from the fields of psychiatry, rehabilitation, psychology, and nutrition conduct a comprehensive assessment of the child’s dietary behavior, sensory integration function, core symptoms of ASD, emotional state, nutrition, and growth and development. Parents’ care abilities were simultaneously evaluated. Based on family characteristics, the causes of abnormal dietary behavior and difficulties in care were clarified, and a personalized intervention plan will be formulated. The core of the intervention included four parts: Sensory integration training, parent empowerment, dietary behavior and emotional guidance, and closed-loop feedback optimization. Sensory integration training was conducted four times a week, with each one-on-one session lasting 60 minutes. It aimed to improve sensory processing disorders and enhance sensory tolerance through vestibular sensing, proprioception, and touch training. Food-related sensory experiences such as touching, smelling, and trying food were integrated into training to reduce food rejection and avoid sensory disturbances in meal preparation and the dining environment. Parent empowerment adopted a combined online and offline model, with two centralized offline training sessions per month and one online question and answer session per week lasting for 6 months. The content covered knowledge on ASD psychiatry, dietary care skills, simple family sensory integration training and emotional intervention methods, data recording, and risk identification. It also alleviated parental care pressure through cognitive behavioral therapy and mindfulness training, building a parent communication community and community support network, and enhanced care capabilities and compliance. On this basis, dietary behavior and emotional interventions were conducted. New foods were introduced following the principle of gradual progress. Abnormal eating behaviors were corrected through gamified eating and positive reinforcement. Food properties were adjusted according to the sensory tolerance to create a relaxed dining atmosphere. Simultaneously, behavioral intervention and cognitive guidance were used to improve anxiety and depression in children and break the vicious cycle of emotional and dietary abnormalities. During the intervention period, monthly reevaluations were conducted to dynamically collect feedback and optimize the plan, forming a closed-loop management of assessment, intervention, feedback, and optimization to ensure the pertinence and sustainability of the intervention.

Observation indicators

The two groups of children and their parents were evaluated at three time points: Before the intervention (T0), 6 months after the intervention (T1), and at 12 months of follow-up (T2). (1) Eating behavior of the child: The CEBQ consists of 26 items divided into four dimensions: Food refusal, emotional eating, slow eating, and selective eating. Each item is scored on a scale of 1 to 5 points. The higher the total score, the more severe the child’s abnormal eating behavior; (2) Psychological and behavioral outcomes of the children: The Autism Behavior Scale (ABC) and Child Anxiety and Depression Scale (CDI) were used. The ABC scale consists of 57 items. The higher the total score, the more severe the core symptoms of ASD in children. The CDI comprises 27 items. The higher the total score, the more evident the child’s anxiety and depression. In this study, the total scores of the two scales were used as the main observation indicators to comprehensively evaluate changes in the core symptoms and emotional states of the children; (3) Sensory integration function of child patients was evaluated using SITS. This scale consists of 120 items with a total score of 120. Higher scores indicate better sensory integration function of the child. A score of < 80 points indicated the presence of sensory integration dysfunction. In the present study, the total SITS score was used as the observation index; (4) Parental care ability was evaluated using the Parental Care Ability Scale (PCCS), which includes four dimensions, care knowledge, care skills, psychological adjustment, and social support, and a total score. The scale consists of 20 items, each scored on a scale of 1 to 5. The higher the total score, the stronger the parents’ care ability; (5) Dietary and nutritional status of the child: The nutritionist recorded the daily intake of protein, vitamins, and minerals of the child patient. The rate of meeting nutritional intake standards was calculated in combination with the dietary reference intake for Chinese children. Simultaneously, the rate of regular eating and the ability to eat independently were evaluated to reflect changes in the child’s nutritional status after the improvement in dietary behavior; (6) Intervention-related indicators. These included three indicators: Intervention compliance, recurrence rate of abnormal dietary behavior, and intervention effectiveness rate. Intervention compliance = (actual number of completed interventions/planned number of interventions) × 100%, which was divided into complete compliance (≥ 90%), partial compliance (70%-89%), and non-compliance (< 70%). Total compliance rate = (number of complete compliance cases + number of partial compliance cases)/total number of cases × 100. Recurrence is defined as a situation where the CEBQ score is again ≥ 60 points after 6 months of intervention and persists for more than 2 weeks. Recurrence rate = number of recurrence cases/total number of cases × 100%. The intervention effectiveness rate was determined based on the improvement in the CEBQ score in combination with the improvement in the ABC scale score: Effective (CEBQ score drops by ≥ 30% and ABC score drops by ≥ 20%), effective (CEBQ score drops by 10% to 29% and ABC score drops by 10% to 19%), and ineffective (not meeting the effective standard or score increases). The effective rate = (number of effective cases + number of effective cases)/total number of cases × 100.

Statistical analysis

Data processing was conducted using SPSS26.0 statistical software. After normality test of measurement data, those conforming to normal distribution were expressed as (mean ± SD), and independent sample t-test was used for comparison between the two groups. Count data were expressed as n (%), and the χ² test was used for comparison between groups. Repeated-measures analysis of variance was used for repeated-measurement data at time points T0, T1, and T2 to compare the effects between groups, time effects, and interaction effects. Further, taking the group as the independent variable, ΔCEBQ as the dependent variable, and ΔPCCS and ΔSITS as the mediating variables, a chain mediating effect analysis was conducted using the biases-corrected bootstrap method. Sampling was repeated 5000 times. If the 95% confidence interval did not include zero, the mediating effect was considered significant. The test level α = 0.05, and a P value < 0.05 was considered statistically significant.

RESULTS

Comparison of dietary behavior scores of the two groups of children before and after intervention

At baseline (T0), no statistically significant differences were observed in CEBQ scores between the two groups (P > 0.05). After the intervention, CEBQ scores decreased significantly in both groups, with greater reductions observed in the intervention group at T1 and T2 (group-by-time interaction, P < 0.001). At T2, the total CEBQ score was 41.25 ± 5.96 in the intervention group compared with 59.23 ± 6.81 in the control group (P < 0.001), indicating a significant improvement in abnormal eating behaviors following the combined intervention, as shown in Table 2.

Table 2 Comparison of dietary behavior scores of the two groups of children before and after intervention, mean ± SD.

Dimensionality	Group	T0	T1	T2	F among groups/P value	F time/P value	F interactive/P value
Food refusal	Intervention group (n = 72)	3.25 ± 0.48	2.01 ± 0.36	1.82 ± 0.35	137.642/< 0.001	378.910/< 0.001	130.359/< 0.001
	Control group (n = 70)	3.28 ± 0.46	2.89 ± 0.41	2.95 ± 0.42
Emotional eating	Intervention group (n = 72)	3.18 ± 0.45	1.92 ± 0.34	1.76 ± 0.32	153.272/< 0.001	452.895/< 0.001	182.498/< 0.001
	Control group (n = 70)	3.21 ± 0.43	2.85 ± 0.39	2.88 ± 0.40
Bradyphagia	Intervention group (n = 72)	3.22 ± 0.47	2.05 ± 0.37	1.85 ± 0.36	131.763/< 0.001	353.137/< 0.001	139.331/< 0.001
	Control group (n = 70)	3.25 ± 0.45	2.92 ± 0.42	2.91 ± 0.43
Selective eating	Intervention group (n = 72)	3.15 ± 0.44	1.89 ± 0.33	1.79 ± 0.33	156.814/< 0.001	526.763/< 0.001	152.472/< 0.001
	Control group (n = 70)	3.18 ± 0.42	2.82 ± 0.38	2.85 ± 0.41
CEBQ Total Score	Intervention group (n = 72)	68.52 ± 7.35	45.12 ± 6.28	41.25 ± 5.96	144.685/< 0.001	575.820/< 0.001	108.889/< 0.001
	Control group (n = 70)	68.87 ± 7.28	58.96 ± 6.73	59.23 ± 6.81

CEBQ: Children’s Eating Behavior Scale.

Comparison of psychological and behavioral outcomes of the two groups of children before and after intervention

Compared with baseline values, both groups showed reductions in ABC and CDI scores after intervention. However, the intervention group exhibited significantly greater improvements at T1 and T2 (all P < 0.001). At T2, the ABC score decreased to 25.49 ± 4.76 in the intervention group compared with 42.71 ± 6.98 in the control group, while the CDI score decreased to 7.87 ± 1.67 compared with 16.36 ± 2.86, respectively (both P < 0.001), as shown in Table 3.

Table 3 Comparison of core symptom scores of autism spectrum disorder in the two groups of children before and after intervention, mean ± SD.

Dimensionality	Group	T0	T1	T2	F among groups/P value	F time/P value	F interactive/P value
ABC Total Score	Intervention group (n = 72)	50.36 ± 8.74	29.58 ± 5.44	25.49 ± 4.76	82.977/< 0.001	396.255/< 0.001	127.132/< 0.001
	Control group (n = 70)	50.57 ± 8.91	42.30 ± 6.85	42.71 ± 6.98
CDI Total Score	Intervention group (n = 72)	20.11 ± 3.65	9.97 ± 2.00	7.87 ± 1.67	170.825/< 0.001	624.828/< 0.001	158.186/< 0.001
	Control group (n = 70)	20.23 ± 3.71	16.19 ± 2.80	16.36 ± 2.86

ABC: Autism Behavior Scale; CDI: Child Anxiety and Depression Scale.

Comparison of sensory integration function scores of the two groups of children before and after intervention

The intervention group demonstrated significantly greater improvements in sensory integration function and parental care ability compared with the control group. At T2, the SITS score increased to 98.42 ± 11.35 in the intervention group vs 76.15 ± 9.58 in the control group (P < 0.001). Similarly, the PCCS total score was significantly higher in the intervention group than in the control group (94.14 ± 13.91 vs 62.88 ± 10.70, P < 0.001), as shown in Table 4.

Table 4 Comparison of sensory integration function scores of the two groups of children before and after intervention, mean ± SD.

Group	T0	T1	T2	F among groups/P value	F time/P value	F interactive/P value
Intervention group (n = 72)	65.32 ± 8.16	92.65 ± 10.23	98.42 ± 11.35	122.684/< 0.001	438.517/< 0.001	96.273/< 0.001
Control group (n = 70)	65.18 ± 8.21	75.36 ± 9.42	76.15 ± 9.58

Comparison of care ability scores of parents in the two groups before and after intervention

At T0, there was no significant difference in each dimension or the total PCCS score between the two groups (P > 0.05). At T1 and T2, the intervention group scores were significantly higher than those of the control group. The intergroup, time, and interaction effects were all significant (P < 0.05), as shown in Table 5.

Table 5 Comparison of care ability scores of parents in the two groups before and after intervention, mean ± SD.

Dimensionality	Group	T0	T1	T2	F among groups/P value	F time/P value	F interactive/P value
Care knowledge	Intervention group (n = 72)	12.35 ± 2.14	21.68 ± 3.25	23.52 ± 3.46	155.005/< 0.001	274.393/< 0.001	128.397/< 0.001
	Control group (n = 70)	12.40 ± 2.18	15.36 ± 2.62	15.48 ± 2.67
Care skills	Intervention group (n = 72)	13.12 ± 2.26	22.53 ± 3.38	24.61 ± 3.59	155.391/< 0.001	233.605/< 0.001	211.956/< 0.001
	Control group (n = 70)	13.18 ± 2.30	16.25 ± 2.75	16.39 ± 2.80
Psychological adjustment	Intervention group (n = 72)	12.86 ± 2.21	21.95 ± 3.31	23.86 ± 3.52	165.859/< 0.001	231.409/< 0.001	219.775/< 0.001
	Control group (n = 70)	12.91 ± 2.25	15.82 ± 2.69	15.95 ± 2.73
Social support	Intervention group (n = 72)	11.95 ± 2.08	20.36 ± 3.12	22.15 ± 3.34	108.572/< 0.001	246.829/< 0.001	108.394/< 0.001
	Control group (n = 70)	12.00 ± 2.12	14.93 ± 2.56	15.06 ± 2.60
PCCS Total Score	Intervention group (n = 72)	50.28 ± 8.69	86.52 ± 13.06	94.14 ± 13.91	128.910/< 0.001	239.601/< 0.001	174.553/< 0.001
	Control group (n = 70)	50.49 ± 8.85	62.36 ± 10.62	62.88 ± 10.70

PCCS: Parental Care Ability Scale.

Comparison of dietary and nutritional status between two groups of children

The post-intervention nutritional intake compliance rate, dietary regularity rate, and eating autonomy of the intervention group were significantly higher than those of the control group (P < 0.05; Table 6).

Table 6 Comparison of dietary and nutritional status between the two groups of children, n (%).

Indicator	Intervention group (n = 72)	Control group (n = 70)	χ2 value	P value
Nutrition intake attainment rate	65 (90.28)	42 (60.00)	17.519	< 0.001
Dietary regularity rate	63 (87.50)	39 (55.71)	17.722	< 0.001
Autonomous eating ability	61 (84.72)	36 (51.43)	18.174	< 0.001

Comparison of intervention-related indicators between the two groups

The total compliance rate and effective rate of the intervention in the intervention group were significantly higher than those in the control group, whereas the recurrence rate was significantly lower than that in the control group (P < 0.05), as shown in Table 7.

Table 7 Comparison of intervention-related indicators between the two groups, n (%).

	Complete adherence	Partial adherence	Nonadherence	Overall adherence rate	Markedly effective	Effective	Ineffective	Overall effective rate	Recurrence rate
Intervention group (n = 72)	56 (77.78)	14 (19.44)	2 (2.78)	70 (97.22)	38 (52.78)	29 (40.28)	5 (6.94)	67 (93.04)	3 (4.17)
Control group (n = 70)	32 (45.71)	25 (35.71)	13 (18.57)	57 (81.43)	15 (21.43)	32 (45.71)	23 (32.86)	47 (67.14)	18 (25.71)
χ2				9.371				15.055	13.077
P value				0.002				< 0.001	< 0.001

Analysis of mediating effects

The mediating effect test was conducted using the bias-corrected bootstrap method with 5000 repeated samples. Taking the group (intervention group = 1, control group = 0) as the independent variable, the improvement value of children’s dietary behavior (ΔCEBQ, defined as T0-T2) as the dependent variable, and the improvement value of parents’ care ability (ΔPCCS, defined as T2-T0) and the improvement value of sensory integration function (ΔSITS, defined as T2-T0) as the mediating variables. A chain mediation model of “group → ΔPCCS → ΔSITS → ΔCEBQ” was constructed and the results showed that the total effect of the group on the CEBQ was 17.624, the direct effect was 5.983, and the total indirect effect was 11.641. The bootstrap test results indicated that the independent mediating effect of ΔPCCS, the independent mediating effect of ΔSITS, and the chain mediating effect of ΔPCCS → ΔSITS were all significant, and their 95% confidence interval did not cross 0. It is suggested that the improvement of parents’ care ability and sensory integration function play a partial chain mediating role in the process of promoting the improvement of children’s dietary behavior through combined psychological intervention, as shown in Table 8.

Table 8 Analysis of mediating effects.

Effect pathway	Effect size	Bootstrap SE	95%CI	Proportion of effect (%)
Total effect	17.624	1.229	15.213-20.036	100.00
Direct effect	5.983	1.101	3.824-8.141	33.94
Total indirect effects	11.641	1.394	9.038-14.505	66.06
Group → ΔPCCS → ΔCEBQ	4.280	0.846	2.711-6.084	24.29
Group → ΔSITS → ΔCEBQ	2.978	0.751	1.624-4.619	16.90
Group → ΔPCCS → ΔSITS → ΔCEBQ	4.383	0.813	2.965-6.102	24.87

CEBQ: Children’s Eating Behavior Scale; PCCS: Parental Care Ability Scale; SITS: Sensory Integration Development Scale; CI: Confidence interval.

DISCUSSION

Abnormal eating and behavioral patterns in children with ASD have often been overlooked for a long time, but they can further lead to nutritional imbalance, emotional problems, and an increased burden on family care, which are important challenges in the field of child rehabilitation[9]. This study developed an early psychological intervention model based on sensory integration and parental empowerment. The results showed that this model can significantly improve abnormal eating behaviors, core symptoms of ASD, and emotional states in children; enhance sensory integration function and parental care ability; increase the rate of meeting nutritional intake standards and the effectiveness of intervention; and reduce the recurrence rate. Mediating analysis further confirmed that there is a chain mediating effect between parents’ care ability and sensory integration function in the intervention effect, suggesting that this model can achieve continuous benefits through dual paths and provide a replicable and scalable standardized intervention idea for clinical practice[10].

The prevalence of abnormal eating behaviors among children with ASD is significantly associated with atypical sensory processing and integration[11,12]. Many pediatric patients exhibit pronounced sensory sensitivities to food touch, texture, smell, and taste, which predispose them to issues such as food refusal, selective eating, and slow eating[12,13]. In this study, the intervention group exhibited significant improvements in SITS score and reductions in sensory sensitivity levels through targeted training in vestibular sense, proprioception, and touch, thereby establishing a foundation for enhanced dietary behavior. In contrast to simple behavioral interventions or nutritional guidance, sensory integration training emphasizes the correction of abnormal response patterns from the perspective of sensory processing and neurodevelopment, which effectively diminishes non-adaptive dietary rejection behaviors[14]. Furthermore, this study integrated sensory food experiences into a training regimen. By employing progressive exposure techniques such as touching, smelling, and tasting, while minimizing sensory disturbances in meal preparation and the dining environment, the relevance and sustainability of the intervention were further enhanced.

Parental empowerment is essential for maintaining the stability of the intervention effects. Prior interventions have primarily targeted the children, often overlooking the knowledge, skills, and psychological well-being of caregivers. This oversight has led to challenges in sustaining therapeutic effects and a high recurrence rate. Recent research indicates that enhancing caregivers’ capabilities is crucial for the enduring effects of family interventions for ASD. Family centered and parent-mediated interventions facilitate the integration of treatment strategies into everyday life scenarios[15,16]. The research developed an “online + offline” systematic empowerment model aimed at enhancing parents’ capabilities across four domains: Disease knowledge, care skills, psychological adjustment, and social support. The findings indicated that the dimensions and total scores of the PCCS in the intervention group were significantly higher than those in the control group. The results demonstrated that integrating remote support with a family centered model effectively improved parental participation, self-efficacy, and access to interventions[17]. Following the enhancement of parents’ knowledge and care skills, simple sensory integration training, emotional guidance, and management of eating behaviors can be consistently implemented within the family context. Furthermore, parent-mediated programs can enhance feeding responses and bolster implementation confidence, thereby facilitating the ongoing management of problematic behaviors[18]. Simultaneously, psychological adjustment and social support can alleviate caregiver stress, diminish the spread of negative emotions, and foster a positive dining environment[19]. Consequently, parental empowerment not only improves the quality of care but also mitigates the risk of recurrence of abnormal eating behaviors.

The most significant finding of this study is the mediating chain effect: The intervention operates through the sequence of “enhancing parental care ability > improving sensory integration function > promoting dietary behavior improvement”, with the indirect effect constituting 66.06%, surpassing the direct effect. Prior research has demonstrated that enhancing caregivers’ skills in parent-mediated interventions serves not only as an auxiliary condition, but also as a crucial mechanism influencing children’s outcomes[20,21]. Consequently, training children in isolation is unlikely to yield stable therapeutic outcomes. Concurrently elevating parental care levels is essential to further enhance sensory integration and rectify abnormal eating behaviors. Furthermore, existing evidence indicates that sensory integration and sensory processing interventions can improve sensory integration and related behaviors in children with ASD with dietary abnormalities being closely linked to sensory dysfunction, thereby supporting the theoretical validity of this pathway[22,23]. Additionally, the proportions of ΔPCCS, ΔSITS, and chain mediating effects were 24.29%, 16.90%, and 24.87%, respectively. These findings indicate that parent empowerment and sensory integration training do not operate independently; rather, they exhibit synergistic enhancement[23,24]. This mechanism elucidates the high efficiency and low recurrence rate observed in this study and serves as a foundation for developing a comprehensive multidimensional intervention model[20,22,24].

In terms of clinical outcomes, the intervention group had better rates of achieving nutritional intake standards, dietary regularity, and autonomous eating ability than the control group, indicating that improving dietary behavior can help improve nutritional status, growth, and development and reduce nutritional risks. Previous studies have shown that eating problems in children with ASD are often associated with limited food choices, decreased dietary quality, and abnormal nutritional status[25,26]. Simultaneously, the intervention group showed a significant decrease in the ABC and CDI scores, indicating that the synchronous relief of core symptoms may accompany improvements in dietary behavior, anxiety, and depression. Related studies have also indicated that dietary intake, gastrointestinal and core symptoms, repetitive stereotyped behavior, sensory characteristics, and eating behavior are closely related, and emotional problems often interact with these abnormalities, jointly forming an adverse cycle[27,28]. Additionally, the intervention group had high compliance and a low recurrence rate, indicating that this model has a good long-term execution. Recent studies have shown that carver-mediated feeding interventions can enhance parents’ feeding confidence, self-efficacy, and positive feeding responses and have good acceptability and promotional value[29,30]. Clinical significance analysis: In addition to statistical differences, the intervention group showed a greater reduction in CEBQ score over time, suggesting meaningful improvement in daily eating behavior. Improvements were also observed in the attainment of nutritional intake, dietary regularity, and autonomous eating ability, indicating potential benefits in real-world clinical practice.

This study has several limitations. First, this was a single-center prospective study with a relatively small sample size, and participants were recruited from specific clinical settings, which may limit the generalizability of the findings to broader populations of children with ASD. In future, multicenter studies with larger samples are needed to further validate the effectiveness and applicability of this intervention model. Second, the intervention was a comprehensive program integrating sensory integration training, parental empowerment, dietary behavioral guidance, and emotional intervention. Although the mediation analysis explored potential mechanisms, the individual contributions of each intervention component could not be fully uncoupled. Future studies using factorial designs or comparative intervention groups are warranted to clarify the independent effects and synergistic mechanisms of different components. Third, although this study included a 12-month follow-up period and demonstrated sustained improvements, longer-term follow-up is still needed to determine whether the intervention effects can be maintained throughout different developmental stages. Finally, some outcome measures, including assessments of dietary behavior and parental care ability, relied on caregiver-reported questionnaires, which may introduce subjective reporting bias. Future research should incorporate more objective indicators, such as direct behavioral observations, nutritional biomarkers, and standardized feeding assessments, to provide more comprehensive evidence.

CONCLUSION

In summary, the early psychological intervention model based on sensory integration and parental empowerment can improve the abnormal eating behaviors of children with ASD at multiple levels, such as neurodevelopment, psychological behavior, and family support. The mechanism of action is clear, the clinical effect is precise, and the long-term stability is good. It can provide a scientific basis and practical reference for early standardized interventions for abnormal eating behaviors in children with ASD, which is of great significance for improving the physical and mental health of children with ASD, reducing the burden of family care, and improving the quality of family life.