The headline number is stark: the average parenting stress score in the sample was 44.40, with a standard deviation of 12.22. Because the PSI-SF yields scores that can be benchmarked against clinical cutoffs, a mean in the mid-forties signals that a substantial share of these caregivers are carrying stress loads that would raise concern in a clinical setting. This is not merely an abstract statistic. Decades of research, including meta-analytic work comparing parents of children with and without ASD, show that elevated parenting stress is linked to poorer caregiver mental and physical health, harsher or less effective parenting practices, and worse developmental trajectories for the child. Chronic stress in caregivers has even been associated in recent neuroimaging studies with measurable changes in brain structure and function, underscoring that parenting stress is a genuine public health variable rather than a soft psychosocial endpoint.

Among the child-related predictors, age mattered. Caregivers of children older than six years reported significantly higher stress than those with younger children, with a regression coefficient of 2.376 points (95% confidence interval 0.424 to 4.329). This finding may seem counterintuitive at first glance, since early diagnosis and the toddler years are often portrayed as the most turbulent period. But the authors’ data align with a growing body of longitudinal work suggesting that stress does not necessarily attenuate as autistic children grow. School-age children face mounting demands around education, independence, and social integration, and in the Chinese context the pressure of academic expectations and the scarcity of inclusive educational placements can intensify rather than ease the caregiving burden as time passes.

Two specific child behavior patterns emerged as particularly potent stress amplifiers. Children who exhibited stereotyped behaviors — the repetitive, restricted patterns such as hand-flapping, rocking, lining up objects, or rigid insistence on sameness, measured here with the Repetitive Behavior Scale–Revised — added an average of 4.207 points to their caregivers’ stress scores (95% CI 1.504 to 6.909). Stereotypies are among the most visible and socially conspicuous features of autism, and they frequently attract unwanted attention in public spaces, compounding the emotional labor of managing the behavior itself. Meanwhile, mealtime behavior problems, assessed with the Brief Autism Mealtime Behavior Inventory, contributed 0.236 points per unit increase on that scale (95% CI 0.139 to 0.333). Feeding difficulties are remarkably common in autistic children, encompassing extreme food selectivity, ritualized eating patterns, and mealtime tantrums, and because meals occur three times a day, every day, they transform a routine family activity into a recurring battleground that erodes caregiver resilience incrementally but relentlessly.

The caregiver’s own circumstances proved equally decisive. Unemployed caregivers scored 3.182 points higher on average than those who were employed (95% CI 0.586 to 5.779). The direction of this association is complex: unemployment may reflect the practical reality that round-the-clock caregiving, therapy appointments, and school advocacy make holding a job untenable, while the resulting loss of income and social identity feeds back into stress. In China, where families of children with ASD often shoulder substantial out-of-pocket costs for rehabilitation services, the economic dimension of this loop is especially consequential. Previous economic analyses have estimated that the lifetime cost of supporting a child with autism can reach into the millions of dollars in high-income settings, and in regions with thinner public safety nets the financial strain falls almost entirely on households.

Perhaps the most striking social finding concerns public stigma. Caregivers who perceived high levels of public stigma, measured with the Perceived Public Stigma Scale, showed markedly elevated stress, with each unit increase on the stigma scale adding 0.534 points (95% CI 0.389 to 0.678). Stigma operates as a chronic, ambient stressor: it shapes how strangers react to a child’s meltdown in a supermarket, whether relatives offer help or judgment, and whether parents feel safe disclosing their child’s diagnosis. Research on affiliate stigma — the stigma experienced by association — has shown that in many East Asian communities, misconceptions about autism still cast blame on parenting or ancestry, isolating families precisely when they most need community support. The Lianyungang data quantify how strongly this social climate translates into individual psychological burden.

On the protective side, three factors were associated with significantly lower stress. Female caregivers scored 3.580 points lower than male caregivers in the sample (95% CI −6.195 to −0.964), a result the authors report with appropriate caution, since gender differences in stress reporting and coping styles complicate interpretation. More actionable were the two psychosocial buffers. Caregivers reporting good family function, assessed with the Family APGAR index — a five-item screen covering adaptation, partnership, growth, affection, and resolve — showed reduced stress, with each unit of better functioning subtracting 0.383 points (95% CI −0.760 to −0.006). And social support emerged as the strongest protective variable in the model: each unit increase in perceived support was associated with a 1.228-point reduction in stress (95% CI −1.734 to −0.723). These coefficients are consistent with a large international literature showing that cohesive families and robust support networks dampen the physiological and psychological impact of chronic caregiving demands.

The methodological design deserves attention for what it does and does not establish. As a cross-sectional study, it captures a single moment in time, so the causal arrows remain ambiguous: do mealtime problems cause stress, or do stressed caregivers perceive mealtime behavior as more problematic? Longitudinal and intervention studies, including a recent randomized controlled trial of hospital–family collaborative intervention in China, suggest the relationship is bidirectional and that improving child symptoms and family functioning can indeed reduce stress. The authors also took standard precautions against statistical artifacts, using variance inflation factors to check for multicollinearity among predictors and reporting confidence intervals throughout, which strengthens confidence in the precision of the estimates even if causality cannot be inferred.

The practical implications, however, are clear and potentially far-reaching. The study was funded by the Lianyungang Science and Technology Bureau and the city’s maternal and child health research program, signaling local government interest in translating findings into services. The authors conclude that interventions to improve family function and social support, and to reduce public stigma, may help alleviate parenting stress in caregivers of children with ASD. Concretely, that points toward parent-mediated intervention programs that strengthen family cohesion, peer support networks and community education campaigns that normalize autism, respite care and employment protections that keep caregivers connected to the workforce, and feeding-focused behavioral therapy that defuses one of the most frequent daily flashpoints. As autism prevalence continues to rise globally — recent surveillance from the United States estimated that roughly one in thirty-one eight-year-olds is identified with ASD — the caregivers behind those statistics are asking a simple question: what would actually lighten the load? This study offers a data-driven answer, and it is less about changing the child than about changing the ecosystem around the family.

Subject of Research: Factors associated with parenting stress among caregivers of children with autism spectrum disorder in eastern China

Article Title: The Associated Factors of Parenting Stress Among Caregivers of Children With Autism Spectrum Disorders in Lianyungang, Eastern China: A Cross-Sectional Study

Article References: Chen, X., Tong, J., Ma, S., Zhu, L., Wang, X., Tao, Y., Zhao, H., Liu, C., Wang, Y., Wang, T., & Yan, D. (2026). The Associated Factors of Parenting Stress Among Caregivers of Children With Autism Spectrum Disorders in Lianyungang, Eastern China: A Cross-Sectional Study. Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-026-07559-y

Image Credits: AI Generated

DOI: 10.1007/s10803-026-07559-y

Keywords: parenting stress, autism spectrum disorder, caregivers, Lianyungang, public stigma, social support, family function, mealtime behavior, stereotyped behaviors, cross-sectional study, PSI-SF, China

The research team, led by Shiye Deng and corresponding author Lei Qiu of Hainan Medical University, recruited 1,294 women seeking treatment at two tertiary reproductive care centers in Hainan Province, a tropical island region where cultural expectations around childbearing remain strong. Every participant completed a battery of validated psychometric instruments: the Infertility Stigma Scale, which captures both externally imposed public stigma and internally directed self-devaluation; the Fertility Problem Inventory, which measures five distinct domains of fertility-related stress including social concern, sexual concern, relationship concern, the need for parenthood, and rejection of a childfree lifestyle; the Patient Health Questionnaire-9 for depressive symptoms; and the Family APGAR Questionnaire, a brief measure of perceived family function across adaptation, partnership, growth, affection, and resolve.

Methodologically, the study stands out for pairing two complementary analytical frameworks. The first, latent profile analysis, is a person-centered technique that searches for hidden subgroups of individuals who share similar patterns across multiple measured variables. Rather than asking which symptoms correlate with which, it asks whether distinct types of people exist. The researchers fed four stigma subscales and five stress subscales into the model and tested solutions with increasing numbers of profiles, using statistical fit indices including the Akaike and Bayesian information criteria, adjusted Bayesian information criterion, the Lo-Mendell-Rubin test, and a bootstrap likelihood ratio test. As a robustness check, they repeated the classification with k-means clustering, a fundamentally different algorithm.

The answer that emerged was strikingly orderly. Three profiles were retained, and they differed from one another in degree rather than in kind. The largest group, 563 women or 43.5 percent of the sample, showed low stigma paired with low stress. A nearly equal second group of 540 women, 41.7 percent, occupied the moderate range on both dimensions. The smallest and most concerning group comprised 191 women, 14.8 percent, who carried high levels of both stigma and stress simultaneously. Notably, the researchers found no evidence of profiles in which stigma was high but stress was low, or vice versa, which is precisely the pattern one would expect if the two constructs formed genuinely distinct psychosocial subtypes. Instead, the data describe a severity continuum.

Who ended up in the high-burden group was far from random. Using an automatic three-step multinomial logistic regression corrected for classification error, the team identified several sociodemographic and clinical correlates. Women living in rural areas, those with lower educational attainment, and those reporting greater perceived financial strain had significantly higher odds of belonging to the high stigma-high stress profile. The diagnostic picture also mattered: women whose infertility was attributed to male factors or remained of unknown origin had lower odds of high-burden membership than women diagnosed with female-factor infertility, a finding that likely reflects the differential blame and social judgment attached to female infertility in many communities. Meanwhile, a longer time since diagnosis was associated with membership in the moderate-burden profile rather than the low-burden one, suggesting that the psychological weight of infertility does not simply dissipate with time spent in the treatment system.

The second analytical framework, symptom-level network analysis, shifted the lens from whole persons to individual symptoms and their pairwise relationships. Network analysis treats psychological variables as nodes in a graph, with statistical associations drawn as edges connecting them. Using the EBICglasso estimator, which applies a least absolute shrinkage and selection operator penalty with an extended Bayesian information criterion to prune spurious connections, the researchers built a network spanning all stigma, stress, depressive symptom, and family function items. They then assessed the stability of the resulting structure through bootstrapping, verified that centrality rankings held across alternative estimation methods, estimated exploratory networks within each latent profile, and formally compared networks across profiles using statistical difference tests.

The full-sample network pinpointed the specific nodes that act as the psychological system’s most influential hubs. On the stigma and stress side, public stigma, the perception of devaluing attitudes from the wider community, and self-devaluation, the internalization of those attitudes, carried the highest expected influence, a centrality metric that captures a node’s total capacity to activate the rest of the network. Among depressive symptoms, sad mood, self-harm and suicidal ideation, and psychomotor symptoms ranked highest, alongside sexual concern from the stress inventory. When the researchers examined which nodes bridged distinct psychological communities, self-devaluation again led, followed by the need for parenthood and social withdrawal. These findings carry a practical implication: interventions that target self-devaluation, such as cognitive approaches that interrupt the internalization of stigmatizing messages, may propagate benefits across stigma, stress, and depressive symptom domains simultaneously.

Family function emerged as a protective counterweight within the network. Its nodes were negatively connected with distress-related symptoms, meaning that women who perceived their families as adaptable, communicative, and emotionally supportive reported fewer and less strongly interconnected psychological difficulties. The expected influence estimates showed good stability under bootstrapping, and centrality rankings were consistent across estimators, lending credibility to the hub identification. Perhaps the most sobering network result, however, was a null one: when the researchers compared networks across the three latent profiles, they found no statistically detectable differences in global network strength or overall structure after applying Benjamini-Hochberg correction for multiple comparisons. The psychological architecture of low-burden and high-burden women, in other words, appears fundamentally similar; what differs is how intensely the same machinery is running.

Taken together, the study’s two strands deliver a coherent message for clinicians and public health planners. Because stigma and stress co-occur along a severity gradient rather than forming discrete subtypes, screening in fertility clinics can reasonably use a single integrated assessment of psychosocial burden rather than separate pathways for stigma and stress. Because a substantial minority of roughly one in seven women occupies the high-burden tier, and because rural residence, lower education, and financial strain flag elevated risk, outreach resources can be targeted toward the women least likely to advocate for themselves. And because self-devaluation, public stigma, sad mood, and suicidal ideation sit at the network’s center, clinicians should treat self-blame not as a downstream byproduct of infertility but as a plausible intervention target with cascading effects. The authors caution that the cross-sectional design cannot establish causal direction, and the sample was limited to treatment-seeking women in a single Chinese province, so generalization requires care. Still, in a field where psychological screening is often an afterthought to hormonal protocols and embryo transfers, this analysis makes a data-driven case that the wound of infertility is as much social as it is medical, and that the two wounds share a single nervous system.

Subject of Research: Infertility stigma and fertility-related stress profiles and their symptom-level associations with depression and family function among infertile women in China

Article Title: Infertility stigma-fertility stress profiles and symptom-level associations among infertile women in Hainan Province: a latent profile and network analysis

Article References: Deng, S., Wang, Y., Liu, J., Wang, M., Lin, C., Liu, M., Petersen, J. D., Yang, H., & Qiu, L. (2026). Infertility stigma-fertility stress profiles and symptom-level associations among infertile women in Hainan Province: a latent profile and network analysis. BMC Public Health. https://doi.org/10.1186/s12889-026-29453-z

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29453-z

Keywords: infertility stigma, fertility-related stress, latent profile analysis, network analysis, depressive symptoms, family function, infertile women, BMC Public Health, Hainan Province, public stigma, self-devaluation, reproductive psychology