Overall, we demonstrated a high completion rate for a complex feedback behavioral intervention that resulted in the majority of subjects successfully attending a PCP follow-up visit and discussing guideline-based preventive asthma care. The intervention was based on the CCM (29
) to enhance preventive care following an asthma ED visits by communicating individualized clinical information to the child’s PCP with guideline-based recommendations, provide caregivers self-management support, and facilitate proactive PCP and caregiver communication during a follow-up clinic visit. The individualized clinical data and specific recommendations provided to the PCP were designed to prompt initiation or adjustment of anti-inflammatory medication based on similar successful feedback interventions (27
). The PCP was asked to review and sign an asthma action plan at the visit, likely accounting for the high rate of caregiver report of discussion about asthma symptoms and medication and agreement with the asthma action plan instructions.
Our 71% PCP visit completion rate exceeds the 20–64% rate found in prior ED follow-up intervention studies with comparable populations (17
), or the 57% rate of high-engagement noted in a web-based asthma management intervention (28
). In the multi-city National Cooperative Inner-City Asthma Study (NCICAS), only half (52%) of enrolled children completed a PCP follow-up visit despite an intensive social worker and asthma education intervention targeting a comparable population as in our study (46
). Our 71% follow-up rate is more consistent with rates reported in ED interventions that scheduled a primary care appointment during the ED visit (64%) (17
), or provided transportation vouchers and reminder calls for follow-up visits (65%) (18
), or a onetime ED follow-up visit for asthma education/management (71%) (26
). Despite this high completion rate, it is concerning that one out of four children did not attend a scheduled PCP appointment. High engagement in a pediatric web-based asthma intervention was associated with increased controller medication use and medication adherence (28
) in contrast to our data, indicating no difference in controller medication use or medication adherence between completers and noncompleters. Failure to complete the PCP visit resulted in a missed opportunity for the nurse to advocate for the PCP to start or step up controller medication and to counsel the caregivers of children residing with a smoker to institute a total home smoking ban.
Several challenges in implementing the intervention warrant comment. First, many inner-city families experience barriers to accomplishing preventive care visits. Younger children were significantly more likely to attend the PCP follow-up visits than older children, possibly due to competing obligations of older children in school that impeded attending a PCP appointment for preventive care. Alternately, caregivers may be more concerned about recurrent symptoms and exacerbations in younger children who are less able to communicate their asthma symptoms and need for medication compared with older children or allowing older children to self-manage their asthma. We standardized the intervention protocol for caregivers across age groups without using a distinct protocol for older children. Second, lack of child care, transportation costs, or caregiver fear of missing work or school may have influenced the caregiver’s ability to complete the PCP visit. Although we subsidized transportation costs with bus tokens, this may have been insufficient to overcome barriers to attendance at the PCP visit for some families. Most children (91%) had medical assistance health insurance coverage; thus, health-care insurance should not have been a barrier to access to care. Third, caregiver asthma stress was lower in the group of children who completed PCP visits, suggesting non-completers may be more psychologically distressed. Alarmingly, the overall rate of reported general life stress in this sample was high and consistent with prior studies in comparable populations (48
) The presence of psychological stress may compromise the caregiver’s ability to recognize their child’s symptoms, weaken their problem-solving skills, and leave the caregiver feeling overwhelmed to perform essential asthma self-management behaviors (50
) or divert attention and resources away from the child (51
). Caregiver depression did not independently predict completion status, in contrast to a study of adults enrolled in a behavioral medical symptom reduction program that reported lower depression scores were associated with higher completion rates (52
), and pre-existing depression associated with failure to complete treatment for hepatitis C disease (53
). Stresses associated with residing in the inner city (e.g., housing and utilities instability, exposure to community violence, and poverty) increase caregiver stress and may outweigh their perception of need for preventive asthma care.
Understanding factors associated with study completion may help identify modifiable factors to improve completion rates and clinical outcomes. In our study, having an asthma action plan in the home at baseline was associated with completing the intervention and attending the PCP visit. The presence of an action plan indicates a previous connection to preventive care that may make follow-up with PCP more likely. Further, having an asthma action plan may serve as a reminder of the need for asthma follow-up. This is consistent with a recent study, finding substantially higher medical follow-up after an asthma ED visit in children who received an asthma action plan and a prescription for controller medication in the ED (54
). Interestingly, frequency of day and night symptoms was not independent predictors of intervention completion, suggesting that intervention completion is not substantially different by asthma severity or asthma control.
Prior to the PCP visit, the intervention involved home visits and almost all children completed this component of the intervention. The home may be the ideal location to teach asthma self-management, that is, appropriate controller medication use or inhaler device technique in the context of the child’s own environment and particularly for older children. While home visits can be more time-intensive, they may be more favorable to families with high stress because they reduce structural barriers to care such as transportation or family disorganization. For the guideline-based care to be achieved, the family must have a strong link to the PCP for medication and symptom management (27
). Alternate methods to assure access to PCP follow-up could include nurse-staffed call centers for nonacute asthma decisions (55
), community or school-based asthma care (56
), or the use of technology such as telemedicine to make appropriate preventive health care more accessible to high-risk populations.
Several lessons were learned from implementing this complex behavioral intervention. First, caregivers of older children may need further attention and require more accessible and flexible appointments after school hours to accomplish preventive care visits and not compete with school demands. Second, caregivers with high stress represent a particularly high-risk group for inadequate follow-up care. These caregivers should be identified and would benefit from referral to appropriate mental health and asthma care resources pre-intervention Third, the lack of an asthma action plan in the home may be a marker of poor access to preventive care and indicates the need for increased monitoring of follow-up. Last, new models of care for frequent ED utilizers should be considered, such as a short ED follow-up visit within 1 week after an ED asthma visit to transition high-risk children from an acute to primary care setting.
There are several limitations to this study that should be considered. First, we did not specifically collect data regarding the reasons for non-attendance of PCP visits. Second, we did not collect or examine PCP characteristics, that is, provider age, years in practice, type of practice, and communication style, that may influence caregiver decision to engage in preventive care and complete the intervention versus only seeking emergency care. Finally, findings from this study can only be generalized to similar high-risk populations. We purposely recruited high-risk children with frequent ED visits to maximize our chances to determine effectiveness of our intervention when all outcome data are available.