Parent and Physician Perceptions Regarding Preventability of Pediatric Readmissions

Abstract

Objective

To assess the causes and preventability of pediatric readmissions from the perspectives of parents and their physicians to guide future interventions.

Patients and Methods

Parent interview, physician survey and medical record review were completed for children who were readmitted to a pediatric hospitalist service within 30 days of an index admission. Questions were asked regarding Health Belief Model constructs (perceived severity, susceptibility or preventability of admission and perceived barriers), discharge readiness and follow-up plans. Parent and physician perceptions about reasons for readmissions were examined and responses to open-ended questions were coded.

Results

60 parent-physician pairs completed the study. The mean age of the patients was 6.43 (SD 6.42) years; 45% (n=27) had a chronic disease, and 47% (n=28) of patients were readmitted with the same or similar condition as in the previous hospitalization. At readmission, parents were more likely than physicians to feel that the condition was serious (parent 98%; physician 76%; p<.001) and that the readmission could have been prevented (parent 59%; physician 36%; P=.04). Most parents (63%) and physicians (65%) felt it was likely that the child may have future hospitalizations. Opportunities to prevent readmission included need for parent education, improving medication access and adherence, and need for coordination of follow-up care.

Conclusions

Many parents and physicians felt the readmission was preventable and the majority of both felt that the patient was susceptible to another hospitalization. Parents and physicians suggest opportunities to improve care processes during hospitalization and in services provided after discharge to reduce readmissions.

Background

Hospital readmission rates are considered a marker of quality of care.^1,2 Hospital readmission occurs in 19.6% of medical-surgical adult Medicare inpatients within 30 days.³ The Centers for Medicare and Medicaid Services (CMS) have reported that up to three quarters of Medicare readmissions may be preventable and has reduced hospital reimbursement if the hospital’s observed readmission rate within 30 days is higher than the expected rate for patients with the principal diagnoses of acute myocardial infarction, congestive heart failure, pneumonia, stroke, hip/knee replacement and chronic obstructive pulmonary disease.⁴ The Children’s Health Insurance Program Re-authorization Act (CHIPRA) has identified pediatric readmissions as a focus for measure development.⁵

While there exists a body of literature on risk factors that lead to re-admissions among adults^6,7,8 less is known about readmissions in children. In a recent pediatric hospital database study, the 30-day adjusted readmission rate for all hospitalized children was 6.5%.⁹ Another study from 38 children’s hospitals found that 16.7% of patients aged 2 to 18 years were readmitted within 365 days of the initial hospitalization.¹⁰ Studies have found readmissions were more common among patients with complex chronic conditions, technology assistance, public insurance, non-Hispanic black race, longer length of stay during the initial admission and frequent previous admissions.^10,11 Berry et al found significant variability in pediatric readmission rates across hospitals, both for all-condition and most condition-specific admissions.⁹ They noted that a small cohort of patients were a major contributor to expenses related to readmissions and postulate that there is potential to reduce readmission rates through interventions during the hospitalization as well as in the care of patients post discharge.

Despite great attention to the issue of readmissions, the majority of existing research stems from secondary data analysis assessing patient risk factors.^6,7,8 There is a dearth of information regarding reasons and preventability of readmissions using primary data collection with direct perspectives from patients, parents and physicians. These perspectives are important and clearly needed to inform strategies to reduce readmission rates. The primary research questions for the current study were based on the Health Belief Model.^12,13 The Health Belief Model is an important framework that assesses perceived severity, susceptibility and preventability, as well as perceived barriers and benefits related to health behavior change. We have applied these concepts to examine hospital readmissions and potential targets for prevention. We sought to understand and compare parent and physician perspectives on the cause and preventability of pediatric hospitalization and readmission.

METHODS

Participants and setting

From March 2012 to September 2012, all patients age 0 to 18 years who were readmitted (with a non-elective readmission) within 30 days of a prior admission to the hospitalist service of a quaternary children’s hospital in the Southeastern United States (All Children’s Hospital, Johns Hopkins Medicine) were identified for recruitment at the time of the readmission. Our hospitalist service does not provide care for patients with cystic fibrosis, sickle cell or malignancies.

Exclusion criteria included patients with non-English speaking parents, patients previously discharged against medical advice, and previous study participants. The parents of eligible patients were recruited within 7 days of the readmission date. Parent written consent was obtained in person as required by the Institutional Review Board (IRB) by one of the study investigators. Child assent was obtained from all children 7-17 years of age.

Each identified patient readmission in this study had: 1) a parent interview; 2) physician survey and 3) medical record review. Parents were compensated with a $25 gift card. All 12 hospitalist physicians in the hospitalist group provided written consent for participation in the study. IRB approval was obtained for the study from the All Children’s Hospital IRB.

Study instruments

Parent interviews took an average of 10 minutes to complete. Questions were framed around Health Belief Model constructs and identified factors such as discharge preparedness, prescriptions, and follow up appointments that have been associated with decreased readmissions in the literature.^14,15 The Health Belief Model constructs include perceived susceptibility as an element that frames behavior (reduce the threat through personal action). The physician survey was developed to mirror the same questions asked in the parent interview. The answers were in Yes/No or 5-point Likert scale (definitely agree, somewhat agree, somewhat disagree, definitely disagree and do not know). Parents and physicians who definitely or somewhat agreed that the readmission was preventable were asked to complete open-ended replies about why they thought so. Physician surveys and medical record review were completed for all patients for whom a parent interview was completed. Seven physicians who were the attendings at the time of the readmission completed the surveys. In only 2 patients reviewed, the index admission discharging attending and the readmission attending were the same physician

The medical record review was completed by two investigators (who were not the attendings for these patients) and included abstraction of demographic information, presence of clinical discharge threshold criteria (no critical values for heart rate, respiratory rate, blood pressure and laboratories, no weight decrease or new oxygen requirement in the last 2 days before discharge), the duration of the index hospitalization, previous hospitalizations at this hospital, the existence of chronic illness and technology dependence. Chronic disease condition was defined as a primary or secondary diagnosis that had these elements as described by Feudtner et al¹⁶ “condition lasting at least 12 months and either affecting multiple systems or one system that required management at a tertiary care center.”

Double data entry was completed for the parent interview, physician survey and medical record review and demonstrated less than 1% differences in quantitative data.

Data Analysis

The Chi-square test or Wilcoxon rank-sum test was used to evaluate significant differences across readmission groups for demographic characteristics and readmission status (Table 1). Agreements between parent and physician perceptions were examined using McNemar’s test for paired binary responses. The exact McNemar probability was adopted to report the significance.

Table 1.

Patient Characteristics (N=60)

	2 admissions (N = 33)	> 2 admissions (N = 27)	p-value
Age of patients	Mean 4.79 years, SD 6.28, Range .06-17	Mean 8.38 years, SD 6.15, Range .23-17	0.03
Race	84% White 16 % African American	89 % White 11 % African American	0.58
Ethnicity	91 % Non-Hispanic 9 % Hispanic	89 % Non-Hispanic 11 % Hispanic	0.83
Insurance	55 % Medicaid 46 % Commercial	67 % Medicaid 22 % Commercial 7% Both 4% None	0.11
Days between current admission and previous discharge	Mean 9.1 days, SD 8.68, Range 0-29	Mean 11.3 days, SD 7.21, Range 1-25	0.30
Length of previous hospitalization	Mean 4.2 days, SD 5.10, Range 1-21	Mean 4.8 days SD 4.90, Range 1-23	0.63
Chronic Disease (% yes)	25 %	76 %	<0.001
Readmitted for same condition (% yes)	53 %	44 %	0.49
Technology Dependent (IV therapy, home ventilator, continuous feeds) (% yes)	9 %	30 %	0.05
Criteria met for initial discharge (% met) *	91 %	96 %	0.41

^*No clinical instability during last 2 days of previous admission [no critical HR, RR, BP or new oxygen requirement; no critical last laboratory values; no decrease in weight]

All analyses were conducted using Stata Version 12¹⁷ with the threshold for statistical significance set at p < .05. Responses to open-ended questions were reviewed and themes coded independently using grounded theory methodology¹⁸ by two investigators (DA and TC). Disagreements were resolved with consensus scoring during regular investigator meetings.

RESULTS

During the study interval, a total of 1562 admissions occurred on the hospitalist service with an all cause readmission rate of 5.4% (85 readmissions). Parent interviews, physician surveys and medical record reviews were completed on 60 patients as shown in Figure 1. Parent interviews were completed within 7 days of the readmission by a research assistant. Seventeen (28%) were completed by phone, 43 (72%) were completed in person. The physician surveys were completed within a mean of 4 days after readmission. Socio-demographic characteristics of the patients are summarized in Table 1 stratified by number of previous hospitalizations (at this hospital) to illustrate the differences in these populations. For all our patients, the mean age was 6.43 years (SD 6.42), 45% (n=27) had a chronic disease, 18% (n=11) were technology dependent and 47% (n=28) of patients were readmitted with the same or similar condition as in the previous hospitalization. The majority of patients were White, Non-Hispanic and had Medicaid insurance. For all patients in the study, the mean and median number of days from previous discharge to readmission was 10.1 and 7.5 days respectively. Length of stay of the previous hospitalization was a mean of 4.5 days and median of 2 days. Results from sensitivity analyses examining potential outliers in length of stay demonstrated no change to substantive findings.

Figure 1.

Recruitment Flow Diagram

We assessed agreement between parent and physician perceptions on the initial hospitalization and readmission regarding seriousness of illness, preventability of readmission and the child’s susceptibility for readmission (Table 2) in order to identify common themes that could present opportunities for improvement. Although many parents and a third of physicians felt the readmission was preventable, parents and physicians differed significantly in the perception of the seriousness of the illness for the previous admission and the readmission, and the need for the readmission. Parents were more likely than physicians to feel that the medical team or the hospital could do more to prevent another hospitalization. Although 56 % of parents versus 40% physicians felt that the parent could do more to prevent another hospitalization, this result was not statistically significant due to a limited number of paired responses. After the readmission, most parents (67%) and physicians (75%) expressed great likelihood for the child to have another hospitalization.

Table 2.

Parent and Physician Perceptions of Initial Admission and Re-admission Health Belief Model Constructs of Perceived Seriousness & Need for Admission, Susceptibility, and Preventability (n=60)

QUESTION	PARENT	PHYSICIAN	P-value
Previous admission: Perceived Seriousness How serious do you think your child’s illness was in his/her last hospitalization? How serious do you think this patient’s illness was in his/her last hospitalization? (% very or somewhat serious)	97%	81%	0.02
This readmission: Perceived Seriousness How serious do you think your child’s illness was during this hospitalization? How serious do you think this patient’s illness was during hospitalization? (% very or somewhat serious)	98%	76 %	<0.001
Perceived Susceptibility for Another Hospitalization How likely is it that your child may have another hospitalization like this? How likely is it that this patient may have another hospitalization like this? (% very or somewhat likely)	67 %	75 %	0.39
Perceived Need for (this) Readmission How likely is it that your child may have another hospitalization like this? How likely is it that this patient may have another hospitalization like this? (% definitely or somewhat true)	98 %	88 %	0.07
Perceived Preventability of Previous Admission Do you feel the situation that resulted in your child’s last admission could have been prevented? Do you feel the situation that resulted in your patient’s last admission could have been prevented?(% definitely or somewhat true)	26 %	13 %	0.14
Perceived Preventability of (this) Readmission Do you feel the situation that resulted in your child’s admission this time could have been prevented? Do you feel the situation that resulted in your patient’s admission this time could have been prevented? (% definitely or somewhat true)	59 %	36 %	0.04
Parent could do more to prevent another hospitalization Do you feel that you can do something to avoid another hospitalization for this patient? Do you feel that the parent can do something to avoid another hospitalization for this patient?(% definitely agree or somewhat agree)	56%	40 %	0.21
Medical Team could do more to avoid another hospitalization Do you feel that you can do something to avoid another hospitalization for this patient? Do you feel that the parent can do something to avoid another hospitalization for this patient? (% definitely agree or somewhat agree)	51%	23%	0.01
Hospital could do more to avoid another hospitalization Do you feel that the hospital or health care system could do something to avoid another hospitalization for your child? Do you feel that the hospital or health care system could do something to avoid another hospitalization for this patient? (% definitely agree or somewhat agree)	37%	16%	0.04

Themes identified during review of the parent and physician open-ended responses regarding preventability included 1) parent concern about rushed discharge; 2) need for parent education; 3) outpatient medication access and adherence; and 4) access to follow-up appointments. Some parents felt their previous hospital discharge was premature and felt “rushed out the door”. One parent stated that “they [hospital staff] could have observed [him] longer to make sure he was improving.” Another stated “[the patient] should have stayed longer to make sure the medication will take effect.” Other parents said it was important to “find out the cause” of the medical disorder before discharge.

Parent education was a frequent theme. One parent said “if I knew more, I could have prevented the readmission.” Another parent stated “give the proper, clear, specific instructions. Come talk to me instead of giving me a bunch of papers to read. Physicians noted the need to “educate families about disease process and treatment strategies” and to educate about the need for adherence to medications and follow-up appointments. Physicians also noted that on occasion there were family anxiety and mental health issues that needed to be addressed to prevent readmission

Discharge readiness and arrangements for follow-up were also assessed in the parent interview. Table 3 summarizes these results. 81% of parents reported that it was easy to obtain medications after discharge. Of those patients who had new prescriptions to obtain after discharge, 67% of patients obtained them on the day of discharge. Parents reported problems such as, “the liquid medication was not covered by Medicaid and was too expensive,” “I couldn’t find a compounding pharmacy” and “I couldn’t get special authorization for the new medication.” When parents were asked if they were able to give their child medications as directed, 88% agreed. When physicians were asked if the patient took their medications as directed after the last discharge, 52% agreed.

Table 3.

Parent Report of Discharge Preparation and Follow up (N=60)

QUESTION	Parent Response
Discharge Readiness (% Yes) • “I feel the child was discharged at the right time.” • “I felt ready to take child home as planned.” • “My child was ready to go home.”	80% 75% 79%
Prescription Needs • “Did you have prescriptions to fill after discharge?”  (% Yes) • If had prescriptions, “was it easy to fill the  prescriptions?” • If had prescriptions, “when were you able to get all  the prescriptions filled?”	72% 81% easy, 19% difficult 67% day of discharge 14% one day later 15% ≥2 days later 4% don’t know
Primary Care Follow Up • “Were you advised to have a follow up appointment  with your child’s primary care provider?” (% Yes) • “Did you feel that making the follow-up appointment  was…”	97% 94% easy, 2% difficult
Specialist Follow Up • “Were you advised to have a follow up appointment  with your child’s specialist doctor?” (% Yes) • “Did you feel that making the follow-up appointment  was…”	73% 79% easy, 21% difficult

In the majority of cases (97%) parents reported that they were advised to have a follow-up appointment with their primary care physician (PCP) and 55% were seen by the PCP prior to the readmission. Parent comments regarding PCP follow-up included “her doctor did not know what was going on” and “I didn’t think it was necessary to keep that appointment.” When parents were asked what they did when their child “got worse” before the readmission (multiple choices were allowed) 30% of parents reported that they called the PCP, 32% called the specialist, 10% called both and 33% did not call either the PCP or the specialist prior to their emergency department visit. Twenty one percent of parents reported that they had difficulty arranging a follow-up appointment with a specialist and most children who had scheduled specialist follow-up were readmitted before those scheduled appointments. Parents also stated “we didn’t have insurance authorization for the specialist” or “we didn’t have transportation.”

DISCUSSION

To our knowledge our study is the first to include parents and physicians perspectives on contributing factors to readmission. Parent and physician perspectives are critical missing components in the discussion on strategies to prevent readmissions. The finding that a substantial proportion of both parents and physicians felt that the patient’s readmission was preventable suggests room to improve. Parents were more likely than physicians to feel that the child’s condition was serious, that the readmission was preventable and that the medical team could do more to avoid another hospitalization.

In our study, physicians felt that for 36% of patients it was definitely or somewhat true that the readmission could have been prevented. This is similar to the findings of Hain et al.¹⁹ who assessed physician perspectives on preventability of pediatric readmissions through retrospective chart review of patients readmitted within 15 days and concluded that 27% of non-elective readmission were preventable. These findings suggest opportunity for improvement and are consistent with existing research.

Family-centered care requires an understanding of the parents’ perspectives on readmissions. Although our chart review indicated that a majority of patients were discharged under optimal conditions and most parents agreed that the discharge was at the right time, many parents felt that their child’s readmission could have been prevented. They indicated that education, obtaining medication, follow-up care and contingency plans could be improved. Some felt that the discharge was rushed and occurred prior to establishing a firm diagnosis or demonstrating effectiveness of therapy. Berry et al²⁰ surveyed 348 parents following their child’s hospital discharge and found a lower readmission rate when parents strongly agreed with the statement “I felt that my child was healthy enough to leave the hospital.” Together with our finding (that some parents reported a rushed discharge) two potential strategies to reduce readmissions include: 1) assessing parent agreement that the child was healthy enough for discharge; and/or 2) ensuring smooth transition to outpatient care.

Ensuring smooth transition from inpatient to outpatient care should involve prompt and detailed communication with the PCP throughout the hospitalization and at discharge, creating and communicating goals for care and future contingency plans and involvement of the family in planning. Some parents in our study were unsure that their PCP was aware of their child’s situation or able to handle the condition. Communication and involvement of the PCP with the patient and family during hospitalization could improve care. Studies have found that discharge summaries may not have critical data and are frequently not sent to the PCP in a timely manner. ^21,22 van Walraven found that for patients discharged from a Canadian teaching hospital, the relative risk for readmission was decreased for patients who were seen in follow-up by a physician who had received a hospitalization summary.²² CMS guidelines on meaningful use of the electronic medical record urge detailed transmission of a care plan including goals and instructions to the PCP and the patient²³ which has the potential to improve communication.

In a study of Medicare patients, those receiving hospitalist care had a shorter hospital length of stay compared to those receiving care from their primary care physician but had higher medical utilization (emergency department visits and readmission) within 30 days of discharge.²⁴ Lower utilization among those with PCP care may have been due to longer length of stay or smoother transition to outpatient care because of PCP familiarity. With the push for shorter lengths of stay and parent concerns about being rushed out and needing education, more research and innovation is needed on family-centered models to improve transition to outpatient care.

A substantial number of parents in our study felt that they could have done more to prevent their child’s readmission. They voiced that they “could have done more if [they] knew more.” Physicians in this study also noted the need for enhanced education. Despite practices within the institution, including family-centered rounds and written educational materials for all discharge patients, many parents felt inadequately prepared. CMS has emphasized standard written discharge instructions as a strategy to improve transitions in care²⁵ but a recent study has shown that standardized discharge instructions were not associated with a reduction in readmissions within 30 days.²⁶ Utilizing known education strategies such as “teach back,” where patients are encouraged to verbalize instructions, could help ascertain patient/parent understanding. Asking families about their confidence in caring for their child and providing support and contingency plans are other important strategies. There is a need for development and evaluation of new tools and methods for parent education.

On discharge from their previous admission most patients had prescriptions to fill. A concerning 19% had difficulty filling prescriptions and many were not able to fill the prescription on the day of discharge. Parents described problems with insurance authorization for prescriptions, pharmacy availability and pharmacy compounding. Systems for easier insurance authorization and hospital dispensing of prescriptions prior to discharge are needed to alleviate these barriers. In a small observational study, medication counseling and delivery of medications by hospital based pharmacists improved patient satisfaction and understanding of medications.²⁷

Most patients were advised to seek primary care and specialty care follow-up. While the majority of parents described getting primary care appointments to be easy, this was less true for specialty care appointments. Of those advised to follow up with specialists, one in five parents stated it was difficult to make the appointment and many readmissions occurred prior to scheduled specialty appointments (which were often delayed to more than 2 weeks after discharge). Attention to subspecialty appointment access barriers is needed. Among other interventions, Jack et al found that arranging appointments and confirming transportation plans prior to discharge decreased readmissions.¹⁴ There is a body of evidence now suggesting that care coordination with social support services can lead to lower health care use including decreased emergency department visits and readmissions.²⁸ In addition, studies on follow-up phone calls from pharmacists and trained registered nurses after discharge have demonstrated effectiveness in reducing readmissions.^14,29

There are limitations and strengths of this study. A strength of this study is the mixed method approach utilizing both quantitative and qualitative data. Mixed methods provided a more comprehensive and nuanced understanding of perceptions surrounding pediatric readmissions. This study involved English- speaking parents of patients readmitted to a hospitalist service at a quaternary care children’s hospital and may not be generalizable to other institutions or populations. In addition, there was not a control group (those not readmitted) and only those who were readmitted provided information about their experience. There may have been recall bias regarding previous events though assessments were done soon after readmission. Social desirability bias in parent and physician responses was also possible. However, with the current focus on readmissions, few studies have explored patient, family and clinician perspectives. Perceptions influence behavior and this study utilizes a patient and family-centered approach grounded in a Health Belief Model framework.

Conclusions

In this study of readmitted patients, a substantial proportion of both parents and physicians felt the readmission was preventable. Many patients had chronic conditions (as defined in our study) and a majority of parents and physicians felt that the patient was susceptible to another hospitalization. There is great opportunity to improve care processes during hospitalization and in services provided after discharge to impact readmissions and enhance child health.

What’s new.

This study adds knowledge about parents and physicians perceptions on preventability of readmissions. A significant number of parents/families and more than a third of physicians felt that the patient’s readmission was preventable and identified contributing factors for the readmission that could be addressed for prevention

List of abbreviations

CMS

Centers for Medicare and Medicaid Services

CHIPRA

Children’s Health Insurance Program Re-authorization Act

IRB

Institutional Review Board

PCP

Primary Care Provider