Pilot Project: Discovering What Health Outcomes Matter Most to Children and Parents

PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.

Background

Healthcare researchers need to know what outcomes are most important to patients so that they can study the outcomes that matter to them. In many cases, researchers do not know patients’ priorities when they are doing research on treatments for different conditions. This is especially true of studies that examine conditions that affect children because the common view is that children cannot tell us what matters to them. Without information on what health outcomes are important to patients, researchers may identify “best” treatments that are not what children and their parents would truly prefer.

Project Purpose

The goal of this project was to create and test a way to allow children and their parents to prioritize health outcomes that matter most to them.

Methods

The researchers reviewed two well-known lists of health outcomes and looked at outcome measures used to study six childhood conditions.

The researchers then interviewed 63 youth ages 12 to 17 and 147 parents of children ages 5 to 17. The children and youth were attending clinics at Children’s Hospital of Philadelphia. The researchers asked the children, youth, and parents to describe health outcomes that mattered most to them.

Using the outcome measures from the lists of health outcomes, measures used in research, and information gathered from the interviews, the researchers created a list of health outcome names and definitions that children and their parents could choose from when indicating what outcomes are most important to them.

The researchers tested the list of outcome names and definitions with 21 high school students ages 14 to 17 and 22 parents of children ages 8 to 17. The researchers tested whether the names and definitions were easy for youth and parents to understand. The research team revised and retested outcome names and definitions until most of the youth and parents understood them.

The researchers then created an online tool using the tested names and definitions of healthcare outcomes. The tool lets children and parents rank the outcomes in the order of those that matter most to them.  

The researchers tested how well the online tool could work for young people with autism spectrum disorder (ASD). The team asked 32 youth and young adults with ASD ages 12 to 30 and 109 parents of youth and young adults with ASD to use the tool to choose the outcomes that mattered most to them. Then the researchers compared what young people with ASD and their parents chose with what researchers use when they study ASD.

Findings

The researchers created 68 health outcome names and definitions that youth and parents understood. Almost half of the outcomes that youth and parents said were most important to them were not commonly used in research studies on ASD.

Limitations

Because the researchers used studies of just six childhood conditions to create their list of outcomes, the list may be missing some health outcomes that are important for other conditions.

Importance of Findings

This project showed that researchers may not be studying many health outcomes that children, youth, and their parents think are the most important. The study identified a method that researchers can use to find out what outcomes are most important to children, youth, and their parents.

Sharing the Results

The researchers presented their work at several national and international conferences. They are preparing articles to send to journals. They plan to make the online tool publicly available.

PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.

Project Purpose

Researchers often select outcome measures without sufficient patient input, a deficiency that precludes assessment of the full spectrum of patient concerns and, therefore, compromises the patient- centeredness of research and clinical care. This deficiency is particularly true for children whose viewpoints are often devalued based on the erroneous assumption that they are unable to provide insight into their own health experiences. Failure to incorporate patient and family views may undermine the meaningfulness of pediatric research by leading to the identification of “best practice” interventions that have little or no effect on outcomes that matter most to children and parents. In response, the researchers sought to develop and evaluate methods that enable children and their families to identify high-priority patient-centered health outcomes.

The researchers had three specific objectives: (1) to develop a child health outcome typology that is relevant to the experiences and preferences of children, (2) to ensure the comprehensiveness and comprehensibility of the health outcome typology for children and their parents, and (3) to develop and test the feasibility and practicality of methods that use the typology to set priorities for health outcomes from the perspectives of children and their families, using Autism Spectrum Disorder (ASD) as a proof-of-principle condition.

Study Design

This study attempted to harmonize two major health outcome typologies: the International Classification of Functioning, Disability and Health (ICF) and the Patient-Reported Outcome Measurement Information System (PROMIS®) domain framework. Researchers conducted a systematic review of health status measures used in research on six childhood conditions; carried out youth and parent semi-structured interviews to identify patient-centered health outcomes; and completed youth and parent cognitive interviews to evaluate the comprehensibility of outcome labels and definitions. Researchers also performed a formative evaluation of an outcome prioritization methodology.

Participants, Interventions, Settings, and Outcomes

Researchers conducted semi-structured interviews of 63 youth ages 12–17 and 147 parents of children ages 5–17, all recruited from ambulatory care settings at the Children’s Hospital of Philadelphia. Cognitive interviewees came from a general community population of public high school students in Philadelphia and included 21 youth ages 14–17 and 22 parents of children ages 8–17. For the pilot test of outcome prioritization, 32 youth/young adults with ASD ages 12–30 and 109 parents of youth/young adults with ASD were included.

Data Analysis

Item concepts derived from patient-reported outcome measures were mapped to the combined ICF/PROMIS framework following standardized procedures. The research team analyzed semi-structured interview transcripts using a meaning condensation procedure in which text was divided into “meaning units” that reflect outcomes of concern. Three raters identified outcome themes expressed in each meaning unit and mapped them to the typology. Two reviewers coded the cognitive interview responses as 1 = not understood, 2 = somewhat understood, and 3 = completely understood. For each outcome, labels and definitions were iteratively refined and retested until the average comprehension rating across five youths exceeded 2.5. The online prioritization task applied a best-worst procedure to generate individual preference scores for each outcome, which were averaged across youth and parent subgroups.

Researchers generated an initial version of the typology by harmonizing the ICF and PROMIS frameworks. They identified 1,602 articles that described the development or application of patient-reported outcomes in the target populations. Researchers gathered 137 instruments from which 5,072 item concepts were derived. The researchers then mapped the item concepts to the combined ICF/PROMIS framework. The typology was revised to ensure that child and family perspectives, identified in the semi-structured interviews, were represented and defined in patient-centered terms. Average Flesch-Kincaid reading level for the “patient-centered” definitions was 5.0 (compared with 12.0 for the ICF/PROMIS technical definitions). Based on cognitive interviews, 44 percent (30 of 68) of the outcome domain labels and definitions were revised. The researchers developed a customizable web-based platform through which youth and parent outcome priorities can be assessed using a best-worst scaling approach.

Findings

Nearly 50 percent of youth- and parent-identified outcomes were underrepresented in research involving youth with ASD.

Limitations

Because the typology was informed by the literature and child and parent input pertaining to six of the most common childhood conditions, it contains a wide variety of outcomes of importance to a large number of children and families. However, it is not (and may never be) exhaustive. Rather, the typology is a working product that investigators should revise to study other conditions and patient experiences. Of particular importance is expanding the typology to include outcome priorities for children with moderate to severe physical disabilities and their families.

Conclusions

Researchers generated a pediatric outcome typology that includes 68 domains of health and functioning (most with two to seven subdomains) that children with chronic conditions and their families value. In the evaluation of six common childhood conditions among youth with ASD, the researchers found that nearly 50 percent of parent- and youth-identified outcome priorities were not included in research focused on these conditions. Thus, research evidence of intervention effects on outcomes that matter most to children and families may be lacking. The typology is a feasible approach to setting priorities for outcomes that may be useful for identifying patient-centered study endpoints.