Abstract
Little is known about services provided to children and youth (C/Y) discharged from an acute care facility. Recent research has provided a foundation for efforts to supplement or complement that early work. This research investigates post-acute care (PAC) in Texas. It focuses on what differentiates those discharges that receive PAC from those that do not and on what differentiates those C/Y who receive PAC in a health care facility from those who receive home health services. The results show that only 6.4% of discharges involving C/Y receive PAC and that many factors affected the 2 issues under investigation quite differently. These results clearly demonstrate the low prevalence of PAC use for C/Y and the clear preference of using PAC home health in this population.
Keywords
Background
Berry and colleagues recently noted that considerable information exists on post-acute care (PAC) for older persons, but little research has focused on PAC for children and youth (C/Y). Their research on hospital discharges for C/Y in 44 states provided new insights into PAC for C/Y and an important foundation for further research. 1
Those results indicated how individual characteristics (race, age), number of chronic diseases, 2 diagnoses of complex chronic conditions, 3 payment source (Medicaid), type of hospital (children’s), and region of the nation (northwest) affected PAC. 1
This research builds on that earlier effort. The research questions investigated here are (1) how did discharges with any type of formal PAC (PAC provided by a health professional) differ from discharges home with no services, and (2) how did discharges that received services from professional staff in an institutional setting, or health care facility (HCF), differ from discharges that received care via home health (HH)?
Methods
Hospital discharges for C/Y in Texas (October 2011 to March 2014) comprised the research database. 4 It included 555 746 discharges from over 400 hospitals involving patients aged under 18 years. To enhance comparability with earlier research, discharges of patients with a principal diagnosis of mental health problems, discharges involving perinatal care and care for pregnancy, discharges that had missing data on race, or discharges for normal births where no PAC was provided were excluded from the database. 1 Discharges involving a death or a patient leaving against medical advice were also excluded from analyses.
The first dependent variable differentiated between those receiving any formal PAC and those receiving none. The second dependent variable only dealt with those discharges involving PAC, differentiating between discharges receiving HH services and discharges receiving PAC in an HCF (any institutional setting where care was provided by health professionals). Both variables represent distinctions of considerable importance for patients, families, and providers.5,6
Independent variables included patient characteristics and contextual factors. The number of chronic conditions was determined using definitions derived from the Agency for Healthcare Research and Quality’s Health Care Utilization Project; classification of illness severity was determined based on All Patient Refined Diagnosis Related Groups; identification of complex chronic conditions was based on Feudtner and colleagues’ work (see references for detail on measures).2,3,7
Using STATA 14, 2 logistic regression models were estimated with clustered robust standard errors (hospitals were clusters). 8 Possible errors in data processing for these population data made using inferential statistics necessary. Given the number of parameters estimated, discussion focuses on results where P < .01.
Results
Table 1 indicates that 6.4% of discharges received formal post discharge services. Table 1 also indicates that the number of chronic disease diagnoses, number of complex chronic conditions, illness severity, length of stay, and age played major roles in PAC decision making. For example, of those with no chronic illnesses, 95% were discharged home with no services, whereas 5% received formal post-acute services. In contrast, for those with 4 or more diagnoses of a chronic illness, almost 14% received formal services. Column 1 of Table 2 indicates the percent of the study population presenting with a specific type of problem. Columns 2 and 3 indicate what percent of those with the condition went home with no services or received PAC. The primary presenting problems resulting in a higher likelihood (>6.9%) of receiving PAC were respiratory problems, injuries, neoplasms, and circulatory or musculoskeletal problems.
Descriptive statistics for pediatric discharges from hospital in Texas with or without post discharge services, 2011-2014.
Principal presenting problem for pediatric discharges from hospital in Texas with or without post discharge services, 2011-2014.
However, the multivariate model results displayed in Tables 3 and 4 provide clearer guidance concerning PAC decisions. Table Three shows that the number of chronic problems, number of complex chronic problems, and severity of the primary problem all significantly increased the likelihood of receiving formal PAC. If PAC was delivered, the number of complex conditions made it more likely that it would be delivered via HH, and only the severity of the illness made PAC in an HCF more likely.
Logistic regression results concerning whether and where formal post discharge services were received.
CI: confidence interval.
To adjust for any regional differences in outcomes, binary variables were included in the model to represent the public health regions in Texas. These results are not included in the table.
P < .01; ***P < .001.
Logistic regression results concerning effects of principal presenting problem on whether and where formal post discharge services were received.
The results presented in Table 4 were part of the models estimated for Table 3.
P < .01; ***P < .001.
Discharges involving women or Medicaid recipients were more likely to result in PAC, whereas only discharges involving those with no external payment source increased the likelihood that PAC would be provided in an HCF. All age groups were less likely than newborns less than a month old to receive PAC and were also less likely to receive formal services in an HCF. The type of hospital (specialty) and its location (urban) made the provision of PAC less likely and the provision of any PAC more likely to be provided in the home.
As Table 4 indicates, among the presenting problems, only injuries and musculoskeletal problems increased the likelihood of PAC provision; circulatory problems and injuries increased the likelihood of PAC in HCFs.
Discussion
The results emphasized the relatively low likelihood of PAC for children with a wide range of seemingly serious health problems. They also clarified the importance of chronic conditions, complex chronic conditions, and illness severity in increasing the likelihood of C/Y receiving PAC. These results also make clear the strong preference that PAC be delivered via HH, a preference shared among health professionals, families, and patients. 3 Only those with more severe presenting problems, no external payment source, or an injury were more likely to receive PAC in an HCF than via HH. In contrast to earlier research, 1 the number of chronic conditions had no effect on where PAC was delivered.
This research has limitations: data include only a single state; data came from a payment information system rather than a clinical database; emphasis here was on formal care, with no consideration of family care. In addition, these data did not allow investigation of important issues such as the appropriateness of the PAC decisions, PAC’s cost-effectiveness, or the dynamics of the PAC decision-making process.
Nonetheless, this research presented a useful way of classifying hospital discharges for C/Y. It documented the low prevalence of PAC use in this population. It also provided results concerning the effects of illness severity and presenting problems on PAC, issues not investigated in earlier research, and helped clarify the role of chronic conditions and complex chronic conditions in PAC decisions.
Footnotes
Peer review:
Five peer reviewers contributed to the peer review report. Reviewers’ reports totaled 1081 words, excluding any confidential comments to the academic editor.
Funding:
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests:
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Author Contributions
CDP suggested the research questions. HCK and CT developed the database. CDP and CT performed the analyses. HCK, ON, and RO contributed to the development of the analysis strategy, to the interpretation of the results, and to decisions concerning the conclusions. All authors reviewed and provided comments on all drafts.