Assessing Parental Discharge Readiness by Using the Ticket to Home Survey Tool Free

We conducted a quasi-experimental pilot study to test the effect of a standardized discharge process that used a predischarge survey tool (TTH). Postdischarge knowledge assessments were conducted in the intervention and usual-care control group to assess the effect of TTH on retention of discharge teaching. Parents were assigned to the intervention or control group on the basis of the inpatient medical team to which they were allocated on admission by using a 3:2 ratio (3 teams assigned to receive the intervention and 2 teams assigned to receive usual care). All pediatric hospital medicine (PHM) admissions are distributed evenly among the teams, regardless of diagnosis. The study took place at a tertiary care children’s hospital between April 2019 and January 2020. This is the region’s only freestanding children’s hospital, serving a catchment area of >900 000 children, with >18 000 annual admissions, 30% of which are admitted to the PHM service. PHM cares for patients with general or complex pediatric diagnoses, previously healthy or with complex medical conditions. Inpatient teams consist of medical students and residents, or a nurse practitioner, with an attending physician.

A convenience sample of parents were eligible to participate if they were English or Spanish speaking and had a child <18 years of age admitted to PHM. Exclusion criteria included primary language other than English or Spanish, parent unavailable at the time of enrollment, or children who were wards of the court. Within 36 hours of admission, we approached and invited eligible parents to participate. The institutional review board approved the study.

Our standard discharge process includes verbal teaching by the care team during rounds on the day of discharge plus patient- and disease-specific written education that is reviewed by nurses and/or respiratory therapists, depending on diagnosis. This written information includes the reason for admission and return precautions. Patients discharged with equipment may also receive device-specific education. Patients admitted on our clinical pathways often receive additional teaching through videos and standardized discharge instructions. For non–English-speaking families, discharge teaching is done by using in-person (Spanish) or video and/or phone interpretation (all languages).

The intervention group received all standard discharge practices. They also received the TTH on the day of discharge. TTH is a predischarge survey tool used to assess parental understanding around reason for admission, return precautions, follow-up appointments, and any necessary medications and/or equipment needed for home (Fig 1). With permission from the original creators of TTH, the study team made minor modifications to question formatting to streamline data collection. TTH was completed several hours before discharge to provide enough time for the team to assess parent’s responses and address gaps in knowledge.

For this pilot study, a provider survey was also created to serve as the gold standard to compare with parent TTH responses. A senior-level provider (pediatric nurse practitioner, senior resident, or PHM fellow or attending) caring for the patient completed this survey for each enrolled patient. We compared parent and provider responses and coded each as “concordant” or “discordant.” Discordant responses indicated that parents did not demonstrate accurate understanding and triggered a team member to address this discrepancy with the parent before discharge.

Parents in the intervention group were contacted 24 to 72 hours after discharge to complete TTH a second time to assess parental retention of discharge education. We also asked about perceived value of TTH and parental confidence in caring for their child’s medical needs related to their recent hospitalization (Supplemental Fig 1A). Finally, parents ranked with whom they would want to discuss postdischarge questions or concerns and review information regarding their hospitalization (Supplemental Fig 1B).

Parents from the control group received all standard discharge practices. They were contacted 24 to 72 hours postdischarge to complete the questions on the TTH tool to assess retention of discharge education. We also asked about confidence in caring for their child’s medical needs and communication preferences. Because they did not complete TTH before discharge, we did not ask them to rate TTH value.

The study team used Research Electronic Data Capture tools to administer all surveys, which were offered electronically in Spanish or English. While consenting, we asked parents’ age and how they preferred to receive the survey(s): E-mail or text message, telephone, or paper.

We also asked providers to rate the value of TTH from 3 perspectives: for families, the medical team, and the health care system. Value response items were assessed by using a 5-point Likert scale, from 1 = not at all beneficial to 5 = very beneficial (Supplemental Fig 1B).

The study team gathered additional information via chart review, including demographic information (patient sex, age, ethnicity, race, primary language, insurance type, and length of hospitalization), medical complexity (defined as technology dependent versus nontechnology dependent), principal diagnosis, and number of diagnoses being addressed during hospitalization. We categorized whether the principal diagnosis was an acute illness in a previously healthy child, presentation of a chronic diagnosis with an unrelated acute illness, or an acute exacerbation of an underlying chronic diagnosis. We used industry-standard categories in our electronic medical record for race (White, African American, Asian American, and other) and ethnicity (Hispanic or non-Hispanic). We defined technology dependence as having ≥1 of the following: ventriculoperitoneal shunt, gastrostomy tube, tracheostomy, and/or mechanical ventilation. The study team determined the reason for admission on the basis of the principal diagnosis during chart review and categorized on the basis of the system: respiratory, infectious disease, gastrointestinal, neurologic, rheumatologic, renal/urologic, and dermatologic. Discharge medications were categorized as “new” or “old/none.” Because standardized education occurs when a patient is admitted by using a clinical pathway, we obtained information regarding pathway use. The study team reviewed charts to gather rates of subsequent 7- and 30-day ED visits and readmissions. We used our hospital use committee’s assignment of “all cause” versus “related” and “preventable” versus “nonpreventable” to report ED return and readmission data.

We used descriptive statistics to examine the sample. We used between-group comparisons on the postdischarge TTH survey tool and questionnaires and used within-group comparisons between baseline and postintervention tools only for the intervention group, using t tests and χ² tests. Baseline surveys were not conducted in the control group because these questions could impact parent understanding of postdischarge care, and it would be unethical to identify knowledge gaps and not address them before discharge. Therefore, within-group comparisons were not performed for the control group. Logistic regression analysis, including all significant covariates, was conducted to determine which factors were associated with increased retention of discharge precautions. We used P value <0.05 to determine significance and SAS version 9.4 (SAS Institute, Inc, Cary, NC) for all analyses.

A total of 495 parents consented to participate in this pilot study. On the basis of team assignment, 319 received the intervention and 176 received usual care. All parents assigned to the intervention group were sent the initial TTH survey; 251 parents did not complete this survey (78.7%). Sixty-eight parents completed the initial survey, all of whom were sent the follow-up TTH survey 24 to72 hours after discharge; 27 of these parents did not complete the follow-up survey for a total loss to follow-up of 278 parents (87.1%) in the intervention group. All parents randomly assigned to the control group were sent the TTH tool 24 to 72 hours after discharge; 117 did not complete this survey (66.5%) (Fig 2). Only 100 completed the necessary surveys and were included in the analysis (41 in the intervention group and 59 in the control group). Among families who did complete the study, we found no differences in sex, race and ethnicity, primary language, number of diagnoses, or medical complexity. However, those who did not complete the survey were more likely to be on public insurance (Medi-Cal) (55.7% vs 39%; P < .01).

Among those that completed the surveys, patients in the intervention group were younger and less likely to be technology dependent or have new medications and/or equipment on discharge (Table 1). There was no difference between groups in length of stay, mean number of diagnoses, new versus old diagnoses, or number of diagnoses managed by clinical pathways. Most of the parents (92%) spoke English as their primary language. Thirty-three parents chose to receive surveys via E-mail, 55 via text message, and 11 via telephone; none chose paper.

Both groups showed high retention of understanding of the reason for admission, with a parent-provider concordance rate of 92.7% in the intervention group and 86.4% in the control group (P = .33) (Table 2). However, the intervention group had significantly higher parent-provider concordance for accuracy of return precautions than the control group (90.2% vs 58.2%; P < .001) (Table 2). In the multivariable analysis, controlling for significant covariates, parents in the intervention group still had increased odds of reporting accurate return precautions (odds ratio 6.24, 95% confidence interval [CI] 1.78–21.93) (Table 3).

In the intervention group at baseline, all parents rated themselves as “confident” or “very confident in caring for their child postdischarge,” with 35 of 41 (85.4%) rating themselves as very confident. There was a statistically significant decrease in confidence in the intervention group from baseline to follow-up (P = .002), with 33 of 41 (80.5%) rating their confidence as very confident at the time of follow-up (Table 4).

Almost all parents in the intervention group felt that sharing their TTH responses with their medical team was beneficial or very beneficial (95.0%). More than half of providers (n = 40) reported similar ratings for TTH benefit to families, the medical team, and the health care system (56.0%, 54.0%, and 56.0%, respectively). When parents were asked to choose a provider to contact to discuss questions or concerns, they most frequently chose their primary care physician (82.0% of respondents). However, when asked to choose a provider to review information regarding their hospitalization, they chose the hospital physician (61.0% of respondents).

Health care reuse was low in both groups for ED visits and hospital readmissions. There were 10 30-day all-cause ED visits, 4 (9.8%) in the intervention group and 6 (10.2%) in the control group (P = .70). Only 1 of these ED visits (control group) was related to the initial hospitalization and determined to be nonpreventable. There were 2 30-day all-cause readmissions among the study cohort, both in the control group and unrelated to the original hospitalization.

Effective discharge processes can have a critical impact on the care of children when they return home. The goal of this pilot study was to assess the effect of a survey tool on the pediatric discharge process, specifically on parent understanding of the reason for admission and return precautions. We found that 24 to 72 hours postdischarge, parents in the intervention group had significantly higher concordance with health care providers regarding return precautions compared with the control group. Our results support previous studies in which researchers suggested that conducting processes to evaluate and strengthen parental understanding may help in the transition process.^6,17  However, TTH did not significantly improve retention of reason for admission, likely because parent-provider concordance was already high at baseline. This is also consistent with a previous study in which researchers reported parent understanding of their child’s diagnosis was concordant with the medical record ∼95% of the time when surveyed 24 hours postdischarge.¹⁷ 

Of note, there was a statistically significant decrease in confidence in postdischarge care needs among parents in the intervention group. Although 2 parents increased their confidence rating from 4 to 5 (confident to very confident) on the follow-up survey, 4 parents who initially ranked their confidence a 5 (very confident) subsequently reported a confidence rate of 4 (confident). This decrease could be due to parents realizing the difficulty of postdischarge care. Because these are small numbers with a high baseline confidence rate, this may also represent a regression to the mean and not be clinically significant.

A significant portion of patients in both groups were admitted by using a clinical pathway. These patients receive standardized education, including standard written discharge instructions, and with some pathways, additional videos and teaching. Thus, one could hypothesize that parents of these children would have a better understanding of discharge instructions. In addition, the control group was older and more medically complex, which may have led to bias in our results. However, even after controlling for these significant covariates in the multivariable analysis, receipt of TTH was still associated with greater concordance on return precautions, albeit with wide CIs. Larger studies should be conducted to assess the impact of predischarge education efforts and include traditional random assignment schemes to ensure baseline similarities in control and intervention groups.

When assessing the importance of this discharge readiness tool, parents in the intervention group placed high value on sharing this information with the medical team. However, given our low retention rate, these results are subject to bias; it is possible that parents who did not complete the surveys did so because they did not feel it would be beneficial. Previous studies have revealed that families desire engagement in the discharge process⁷  and that patient-engagement programs improve health outcomes.^18,19  Thus, the impact of such tools as TTH on patient engagement may be an area for further study.

We also gathered information on preferred point of contact for postdischarge questions. Many parents wanted to discuss questions or concerns after discharge with their child’s primary care physician. However, if parents wanted to review information about their child’s recent hospitalization, most reported preferring contact from one of their hospital physicians. This suggests that there may be a role for postdischarge phone calls from an inpatient medical team member, as noted by researchers in several previous studies.^{3,10,11,20–22} 

Despite revealing the value of TTH on improving retention of return precautions and parent engagement, our study had several limitations. First, our survey completion rate was low at 20.2%. However, this completion rate is similar to other studies with Internet-based surveys.^14,23,24  We had particular difficulty retaining adequate numbers for the intervention group because it required parents to complete 2 surveys. Consequently, noncompletion bias may have affected our results, with the possibility that current respondents did not accurately reflect the general population of hospitalized children. In addition, there was a differential loss to follow-up (greater in the intervention group), which may have resulted in a false-positive finding (type 1 error) in which the null hypothesis was erroneously rejected. Some of this loss to follow-up may have been due to the limited time frame in which to complete the initial TTH survey in the intervention group. Nevertheless, this pilot study served to provide preliminary information on the effectiveness of TTH. Most of our cohort also spoke English as a primary language and included few children who were technology dependent, thereby limiting generalizability to more medically complex patients and diverse populations. The control group was also older and more medically complex, likely a result of our quasi–random assignment approach, which was based on inpatient team assignment. This bias may have also contributed to a type 1 error. Further research should include larger studies in which researchers use a traditional random assignment approach and powered to detect differences in more complex patient subgroups. Lastly, this study was completed at a single center with patients admitted to only the PHM service; however, as PHM cares for a wide variety of pediatric patients, TTH reveals promise as a valuable tool for addressing concerns and engaging families.

In this pilot study, we found that use of TTH improved parental retention of return precautions, which is one of the critical features of safe transitions to home. Future efforts include testing in a larger study with more diverse and complex populations.