Caregiving and Confidence to Avoid Hospitalization for Children with Medical Complexity.

OBJECTIVE: To test associations between parent-reported confidence to avoid hospitalization and caregiving strain, activation, and health-related quality of life (HRQOL). STUDY DESIGN: In this prospective cohort study, enrolled parents of children with medical complexity (n = 75) from 3 complex care programs received text messages (at random times every 2 weeks for 3 months) asking them to rate their confidence to avoid hospitalization in the next month. Low confidence, as measured on a 10-point Likert scale (1 = not confident; 10 = fully confident), was defined as a mean rating <5. Caregiving measures included the Caregiver Strain Questionnaire, Family Caregiver Activation in Transition (FCAT), and caregiver HRQOL (Medical Outcomes Study Short Form 12 [SF12]). Relationships between caregiving and confidence were assessed with a hierarchical logistic regression and classification and regression trees (CART) model. RESULTS: The parents were mostly mothers (77%) and were linguistically diverse (20% spoke Spanish as their primary language), and 18% had low confidence on average. Demographic and clinical variables had weaker associations with confidence. In regression models, low confidence was associated with higher caregiver strain (aOR, 3.52; 95% CI, 1.45-8.54). Better mental HRQOL was associated with lower likelihood of low confidence (aOR, 0.89; 95% CI, 0.80-0.97). In the CART model, higher strain similarly identified parents with lower confidence. In all models, low confidence was not associated with caregiver activation (FCAT) or physical HRQOL (SF12) scores. CONCLUSIONS: Parents of children with medical complexity with high strain and low mental HRQOL had low confidence in the range in which intervention to avoid hospitalization would be warranted. Future work could determine how adaptive interventions to improve confidence and prevent hospitalizations should account for strain and low mental HRQOL.

Linking Parent Confidence and Hospitalization through Mobile Health: A Multisite Pilot Study.

OBJECTIVE: To evaluate the associations between parent confidence in avoiding hospitalization and subsequent hospitalization in children with medical complexity (CMC); and feasibility/acceptability of a texting platform, Assessing Confidence at Times of Increased Vulnerability (ACTIV), to collect repeated measures of parent confidence. STUDY DESIGN: This prospective cohort study purposively sampled parent-child dyads (n = 75) in 1 of 3 complex care programs for demographic diversity to pilot test ACTIV for 3 months. At random days/times every 2 weeks, parents received text messages asking them to rate confidence in their child avoiding hospitalization in the next month, from 1 (not confident) to 10 (fully confident). Unadjusted and adjusted generalized estimating equations with repeated measures evaluated associations between confidence and hospitalization in the next 14 days. Post-study questionnaires and focus groups assessed ACTIV's feasibility/acceptability. RESULTS: Parents were 77.3% mothers and 20% Spanish-speaking. Texting response rate was 95.6%. Eighteen hospitalizations occurred within 14 days after texting, median (IQR) 8 (2-10) days. When confidence was <5 vs ≥5, adjusted odds (95% CI) of hospitalization within 2 weeks were 4.02 (1.20-13.51) times greater. Almost all (96.8%) reported no burden texting, one-third desired more frequent texts, and 93.7% were very likely to continue texting. Focus groups explored the meaning of responses and suggested ACTIV improvements. CONCLUSIONS: In this demographically diverse multicenter pilot, low parent confidence predicted impending CMC hospitalization. Text messaging was feasible and acceptable. Future work will test efficacy of real-time interventions triggered by parent-reported low confidence.

Variation in Hospitalization Rates Following Emergency Department Visits in Children with Medical Complexity.

OBJECTIVES: To evaluate factors associated with admission from emergency department (ED) encounters for children with medical complexity (CMC) and to quantify the hospital admission rate as well as variation in adjusted hospital admission rates across EDs. STUDY DESIGN: Retrospective study of 271 806 visits to 37 EDs in freestanding children's hospitals from January 1, 2014, to June 30, 2017, for patients of all ages with a complex chronic condition. Associations between patient demographic, clinical, and health services characteristics and the likelihood of hospital admission were identified using generalized linear models, which were then used to calculate adjusted hospital admission rates. RESULTS: Hospital admission occurred with 25.7% of ED visits. Characteristics with the greatest aOR of hospitalization were ≥3 compared with 0 prior hospitalizations in 365 days (4.7; 95% CI, 4.5-4.9), ED arrival overnight compared with during workday 3.2 (95% CI, 3.1-3.3)], and ≥6 vs 0-1 chronic conditions (1.6; 95% CI, 1.5-1.6). Adjusted hospital admission rates varied significantly (P < .001) across EDs (21.1% [10th percentile]) and 30.0% [90th percentile]). Significant variation remained when excluding low-intensity ED visits, excluding hospitalizations requiring surgery and/or intensive care, or restricting the cohort to overnight ED arrival and to children with ≥3 prior hospitalizations. CONCLUSIONS: CMC are frequently admitted from the ED. Substantial variation in CMC hospital admission rates across EDs exists after case-mix adjustment.

Association of Home Respiratory Equipment and Supply Use with Health Care Resource Utilization in Children.

OBJECTIVE: To compare health care use and spending in children using vs not using respiratory medical equipment and supplies (RMES). STUDY DESIGN: Cohort study of 20 352 children age 1-18 years continuously enrolled in Medicaid in 2013 from 12 states in the Truven Medicaid MarketScan Database; 7060 children using RMES were propensity score matched with 13 292 without RMES. Home RMES use was identified with Healthcare Common Procedure Coding System and International Classification of Diseases codes. RMES use was regressed on annual per-member-per-year Medicaid payments, adjusting for demographic and clinical characteristics, including underlying respiratory and other complex chronic conditions. RESULTS: Of children requiring RMES, 47% used oxygen, 28% suction, 22% noninvasive positive-pressure ventilation, 17% tracheostomy, 8% ventilator, 5% mechanical in-exsufflator, and 4% high-frequency chest wall oscillator. Most children (93%) using RMES had a chronic condition; 26% had ≥6. The median per-member-per-year payments in matched children with vs without RMES were $24 359 vs $13 949 (P < .001). In adjusted analyses, payment increased significantly (P < .001 for all) with mechanical in-exsufflator (+$2657), tracheostomy (+$6447), suction (+$7341), chest wall oscillator (+$8925), and ventilator (+$20 530). Those increased payments were greater than the increase associated with a coded respiratory chronic condition (+$2709). Hospital and home health care were responsible for the greatest differences in payment (+$3799 and +$3320, respectively) between children with and without RMES. CONCLUSION: The use of RMES is associated with high health care spending, especially with hospital and home health care. Population health initiatives in children may benefit from consideration of RMES in comprehensive risk assessment for health care spending.

Trends in Hospital Treatment of Empyema in Children in the United States.

OBJECTIVE: To evaluate trends in procedures used to treat children hospitalized in the US with empyema during a period that included the release of guidelines endorsing chest tube placement as an acceptable first-line alternative to video-assisted thoracoscopic surgery. STUDY DESIGN: We used National Inpatient Samples to describe empyema-related discharges of children ages 0-17 years during 2008-2014. We evaluated trends using inverse variance weighted linear regression and characterized treatment failure using multivariable logistic regression to identify factors associated with having more than 1 procedure. RESULTS: Empyema-related discharges declined from 3 in 100 000 children to 2 in 100 000 during 2008-2014 (P = .04, linear trend). There was no significant change in the proportion of discharges having 1 procedure (66.1% to 64.1%) or in the proportion having 2 or more procedures (22.1% to 21.6%). The proportion coded for video-assisted thoracoscopic surgery as the only procedure declined (41.4% to 36.2%; P = .03), and the proportions coded for 1 chest tube (14.6% to 20.9%; P = .04) and 2 chest tube procedures (0.9% to 3.5%; P < .01) both increased. The median length of stay for empyema-related discharges remained unchanged (9.3 days to 9.8 days; P = .053). Having more than 1 procedure was associated with continuous mechanical ventilation (adjusted OR, 2.7; 95% CI, 1.8-4.1) but not with age, sex, payer, chronic conditions, transfer admission, hospital size, or census region. CONCLUSIONS: The use of video-assisted thoracoscopic surgery to treat children in the US hospitalized with empyema seems to be decreasing without associated increases in length of stay or need for additional drainage procedures.

Hospitalizations for Ambulatory Care-Sensitive Conditions among Children with Chronic and Complex Diseases.

OBJECTIVE: To evaluate ambulatory-care sensitive (ACS) hospitalizations for children with noncomplex chronic diseases (NC-CD) and children with medical complexity (CMC), and identify associations with ambulatory care characteristics. Although ACS hospitalizations are potentially preventable in general populations, the specific ambulatory care predictors and influence of medical complexity on them is poorly understood. STUDY DESIGN: Retrospective cohort study of NC-CD and CMC hospitalizations at a children's hospital during 2007-2014, excluding labor/delivery and children over 21 years. Pediatric medical complexity algorithm identified NC-CD or CMC. ACS hospitalizations were identified using Agency for Healthcare Research and Quality indicator definitions. Demographic and ambulatory care characteristics were compared between ACS and non-ACS hospitalizations with logistic regression clustered by patient. Measures of ambulatory care during 2 years before admission were explored with 20% random sample of general pediatrics discharges. RESULTS: Among 4035 children with NC-CD, 14.6% of 4926 hospitalizations were ACS hospitalizations. Among 5084 CMC, 5.3% of 14 390 discharges were ACS hospitalizations. Among NC-CD discharges, ACS hospitalizations were more likely with no prior-year outpatient visits (OR 1.4, 95% CI 1.1-1.7) and less likely with timely well checks (OR 0.8, 95% CI 0.6-0.9) and phone encounters in the month before admission (OR 0.5, 95% CI 0.2-1.0). Among CMC discharges, the only association observed was with provider continuity (OR 0.3, 95% CI 0.1- 1.0). CONCLUSIONS: Provider continuity may be associated with fewer CMC ACS hospitalizations, however, measures of ambulatory care were more consistently associated with ACS hospitalizations for NC-CD. CMC may need more precise ACS hospitalization definitions.

Predictors of 30-day readmission and association with primary care follow-up plans.

OBJECTIVE: To test the hypothesis that missing primary care follow-up plans in the discharge summary is associated with higher 30-day readmissions. STUDY DESIGN: This retrospective cohort study included pediatric patients discharged from Mattel Children's Hospital, University of California, Los Angeles between July 2008 and July 2010. Exclusions included deaths, transfers, neonatal discharges, stays under 24 hours, and patients over 18 years of age. Bivariate and propensity weighted multivariate logistic regressions tested relationships between 30-day readmission and patient demographics, illness severity, and documentation of primary care provider (PCP) follow-up plans at discharge. RESULTS: There were 7794 index discharges (representing 5056 unique patients), with 1457 readmissions within 30 days (18.7%). Average length of stay was 6.3 days. Being 15-18 years old, (OR 1.42 [1.02-1.96]), having public insurance (OR 1.48 [1.20-1.83]), or having higher All-Patient Refined Diagnosis-Related Group severity scores (for severity = 4 vs 1, OR 6.88 [4.99-9.49]) was associated with increased odds of 30-day readmission. After adjusting for insurance status, Asian (OR 1.46 [1.01-2.12]) but not Black or Hispanic, race/ethnicity was associated with greater odds of readmission. Fifteen percent of 172 medical records from a randomly selected month in 2010 documented PCP follow-up plans. After adjusting for demographics, length of stay and severity, documenting PCP follow-up plans was associated with significantly increased odds of 30-day readmission (OR 4.52 [1.01-20.31]). CONCLUSION: Readmission rates are complex quality measures, and documenting primary care follow-up may be associated with higher rather than lower 30-day readmissions. Additional studies are needed to understand the inpatient-outpatient transition.