Larry Yin MD, MSPH

Despite documented healthcare disparities among Latino autistic children, little is known about how their families experience the autism “diagnostic odyssey.” Pediatricians have a critical role in the “diagnostic odyssey,” but when it becomes arduous, parents may also use complementary health approaches, particularly when conventional care does not adequately address their concerns. Shared decision-making is important in healthcare for autistic children, especially when parents also choose complementary health approaches; but little is known about shared decision-making among Latino parents of autistic children. We conducted a 12-month ethnography with 12 bicultural/bilingual Latino families of autistic children to understand their healthcare experiences (conventional and complementary health approaches) for their children, focusing on shared decision-making. Thematic analysis revealed: (1) most pediatricians were involved early in the “diagnostic odyssey” but were largely uninvolved thereafter; (2) conventional healthcare was satisfactory to the parents for physical health, but not developmental issues; and (3) parents who used complementary health approaches were more frustrated about a lack of autism information from pediatricians than those who did not. Finally, (4) we describe two exemplars of successful shared decision-making between Latino parents and pediatricians. Increasing pediatricians’ autism knowledge and ability to discuss complementary health approaches may facilitate shared decision-making and reduce healthcare disparities for Latino autistic children.

Background. Autistic children and youth have high rates of co-occurring conditions,but little is known about how autism subgroups (girls, non-White children) access healthcare to treat them. The purpose of this longitudinal retrospective cohort study was to investigate differences by sex, race, and ethnicity in non-acute (primary and subspecialty) healthcare use by autistic children and youth.

Method. We used the OneFlorida Data Trust to measure healthcare use for 82,566 autistic children ages 1–21 (2012–2018). We investigated (1) the odds of using any healthcare and (2) annual healthcare usage rates. We adopted a logistic regression and multiple linear regression for each of the dependent variables (total non-acute, primary care, gastroenterology (GI), developmental-behavioral pediatrics (DBP), psychiatry/psychology, neurology, and total subspecialty visits), adjusting for potential confounders (sex, race, ethnicity, age at visit, insurance type, urbanicity, and co-occurring conditions).

Results. Autistic boys had significantly higher odds of any neurology and psychiatry/psychology visits, but lower annual rates of primary care, GI, and neurology visits. Black/African American autistic children had significantly higher odds of any primary care, DBP, and neurology visits, but lower odds of any GI visits, and lower annual rates of primary care, GI, DBP, and neurology visits. Hispanic/Latinx autistic children had significantly higher odds of any primary care, DBP, and neurology visits, but lower odds of psychiatry/psychology visits; and higher annual rates of neurology visits, but lower annual rates of GI, DBP, and psychiatry/psychology visits.

Conclusions. We found significant differences by sex, race, and ethnicity in non-acute healthcare use by autistic children.

Keywords. Autism spectrum disorders; Sex/gender; Healthcare utilization

Background. Childhood obesity continues to be a substantial problem despite major public health efforts, and disproportionately impacts children from low-income families. Digital health tools and consumer technology offer promising opportunities for interventions, but few studies have evaluated how they might be incorporated into existing interventions or used to create new types of interventions. It remains unclear which approaches would be most beneficial for underserved pediatric populations.

Purpose. To describe the design and rationale of a single-center randomized, controlled trial evaluating the effects of personal activity tracker (PAT) use by parents on weight-status improvement in both parents and overweight children enrolled in BodyWorks (BW), a comprehensive behavioral family-lifestyle intervention program (CBFLI), in a primary-care clinic serving a predominantly low-income Latino population.

Methods. This study is being conducted in the AltaMed general pediatrics clinic at Children's Hospital Los Angeles. Eligible participants are families (child and adult caregiver) in which the child is between 7 and 18 years of age, has a BMI ≥85th percentile for age and sex, and has been referred to BW by their AltaMed pediatrician. BW consists of one weekly, two-hour session for 7 weeks. In a given cycle, the program is offered on two separate nights: Monday (Spanish) and Wednesday (English). Families self sort into one of two groups based on language preference. To ensure balanced allocation of language preference groups and prevent in-group cross contamination, block randomization is used to assign whole groups to either the intervention or control arms of the study. The control arm consists of usual care, while the intervention arm adds assigning a Fitbit PAT to the parents and training them in its proper use. Study personnel are blinded to group assignment during the analysis phase. Study outcomes include attendance rate, program completion rate, and changes in weight-status improvement, defined as change in weight and BMI for adults and change in BMI z-score for children. We hypothesize that the intervention arm will have better weight-related outcomes than the control arm. Study completion is anticipated in 2017, after the enrollment of approximately 150 families.

Conclusions. The study aim is to evaluate the effects of PATs on weight-related outcomes in overweight children and parents participating in a CBFLI. The results will be important for determining whether wearable devices are an effective addition to weight loss interventions for overweight and obese children.

Conversation and discourse analyses were used to examine medical problem presentation in pediatric care.Healthcare visits involving children with ASD and typically developing children were analyzed. We examined how children's communicative and epistemic capabilities, and their opportunities to be socialized into a competent patient role are interactionally achieved. We found that medical problem presentation is designed to contain a 'pre-visit' account of the interactional and epistemic work that children and caregivers carry out at home to identify the child's health problems; and that the intersubjective accessibility of children's experiences that becomes disrupted by ASD presents a dilemma to all participants in the visit. The article examines interactional roots of unmet healthcare needs and foregone medical care of people with ASD.

Substantial scholarship has been generated in medical anthropology and other social science fields on typically developing child-parent-doctor interactions during health care visits. This article contributes an ethnographic, longitudinal, discourse analytic account of a child with autism spectrum disorder (ASD)-parent-doctor interactions that occur during pediatric and neurology visits. The analysis shows that when a child with ASD walks into the doctor's office, the tacit expectations about the visit may have to be renegotiated to facilitate the child's, the parent's, and the doctor's participation in the interaction. A successful visit then becomes a hard-won achievement that requires the interactional and relational work of all three participants. We demonstrate that communicative and sensory limitations imposed by ASD present unique challenges to all the participants and consider how health care disparities may invade the pediatric encounter, making visible the structural and interactional processes that engender them.

Efforts to improve early diagnosis of autism spectrum disorder (ASD) in children are beginning to use machine learning (ML) approaches applied to real-world clinical datasets, such as electronic health records (EHRs). However, sex-based disparities in ASD diagnosis highlight the need for fair prediction models that ensure equitable performance across demographic groups for ASD identification. This retrospective case-control study aimed to develop ML-based prediction models for ASD diagnosis using risk factors found in EHRs and assess their algorithmic fairness. The study cohorts included 70,803 children diagnosed with ASD and 212,409 matched controls without ASD. We built logistic regression and Xgboost models and evaluated their performance using standard metrics, including accuracy, recall, precision, F1-score, and area under the curve (AUC). To assess fairness, we examined model performance by sex and calculated fairness-specific metrics, such as equal opportunity (recall parity) and equalized odds, to identify potential biases in model predictions between boys and girls. Our results revealed significant fairness issues in ML models for ASD prediction using EHRs.