Health-related quality of life measurement in pediatric clinical practice: An appraisal and precept for future research and application

One potential benefit of incorporating HRQOL assessment into clinical practice is the impact on patient-physician communication [7]. The adult literature supports the idea that both patients and physicians believe HRQOL feedback to physicians assists communication [3, 8‐11]. For example, patients are often willing to discuss HRQOL with their physicians even when they are not experiencing significant impairments [8, 12]. Unfortunately, although patients desire to share HRQOL information, physicians are not likely to ask [8, 12‐14]. In fact, research evidence suggests that physician behaviors (e.g., neglecting to ask about HRQOL issues) can impede a patient's willingness to introduce HRQOL concerns, that physicians vary greatly in their ability to elicit important information from patients, and that patients vary in their ability or desire to articulate it [15, 16]. Patients are, however, generally willing to complete a HRQOL questionnaire at outpatient visits [8, 12]. In this way, HRQOL measures can be a catalyst for communication in that they can objectively identify areas of concern that can then be discussed at the point of healthcare service.

Unfortunately, there are no randomized controlled studies in the published literature that have examined the impact of pediatric patient self-report or parent proxy-report HRQOL measurement on patient-physician communication. In pediatric practice, communication can be a key issue, as young children may not have the language skills necessary to accurately communicate their symptoms or feelings verbally. For these children, child self-report HRQOL measures that are sensitive to children's language and cognitive development may indicate problem areas, even those unrelated to the patient's chief complaint, that need to be further explored. In this way, one benefit of standardized HRQOL measurement in pediatric practice is the communication of healthcare needs by the child who may not be able to otherwise accurately report symptoms or other problems with daily functioning.

The adult literature suggests that patient satisfaction with medical care has been related to physician discussion of psychosocial topics and active physician elicitation of patient opinion [17, 18]. Street and colleagues [13] reported that physicians who asked about pain, health perceptions, vitality, and role limitations due to physical problems were evaluated more favorably than physicians who did not. Another study by Brody et al.[19] reported that patients of physicians who received mental health feedback reported more satisfaction with their physician than patients in a control group. In the pediatric literature, no research has examined patient satisfaction and HRQOL assessment in pediatric practice. While it is unclear whether children would respond similarly, parents of pediatric patients are likely to report satisfaction or dissatisfaction with medical care based on their perceptions of their child's HRQOL outcomes as a result of treatment.

The adult literature has indicated a disparity in physician and patient ratings of health status [20‐23]. Physicians frequently underestimate or fail to detect psychosocial and functional disabilities in patients, and often even high levels of psychological distress are undetected and untreated [22, 24‐28]. Studies with adult patients, however, indicate that physicians who receive HRQOL data have significantly higher detection rates for psychological problems [29‐32], and for functional and HRQOL impairments [10, 33]. With this information a physician may be better able to make appropriate referrals and tailor interventions to the specific needs of his/her patients.

Research in pediatric healthcare has also demonstrated the continuing under-identification of psychosocial problems, termed the "new hidden morbidity" [34], in routine pediatric practice [35, 36]. Given the marked under-detection of these problems by physicians, HRQOL measures can serve as standardized screening instruments for identifying physical and psychosocial health concerns from the perspectives of both the child and parent at the point of service that pediatricians might otherwise overlook [37]. However, there are no existing randomized controlled pediatric trials that address this issue. Nevertheless, it seems reasonable to assume that pediatricians are likely to be assisted by standardized HRQOL measurement in the identification of hidden morbidities that might not otherwise be detected.

In addition to traditional morbidity and mortality outcomes, evidence-based clinical decision-making requires validated patient-reported outcomes, particularly for chronic health conditions [38]. The impact of a HRQOL measurement instrument on clinical decision-making can be tested under the working hypothesis that HRQOL measurement and feedback to physicians must occur at the point of service in order to improve health outcomes [37]. Clinical decision-making may also benefit from a serial screening approach, which can assist with tracking changes in HRQOL over time. This longitudinal monitoring of HRQOL can help physicians evaluate the effectiveness of treatment or monitor disease status and associated symptoms such as pain, fatigue, coughing, wheezing, and the like.

In the adult literature, it has been estimated that up to 52% of physicians make management decisions (e.g., medication changes, ordering lab tests) on the basis of patient-reported HRQOL [6, 8, 39]. One study found that functional status feedback provided new information for approximately 25% of patients, resulting in specific clinical management actions by physicians in approximately 40% of those patients [3]. HRQOL measurement may also assist physicians with the referral process, in that they may be more likely to refer patients to mental health specialists or other health professionals as a result of this information [8, 31].

Although these studies with adult patients suggest that improving detection through the use of HRQOL measurement does affect clinical care in various ways, the data regarding changes to clinical practice, after receiving HRQOL feedback, is inconclusive. For example, some adult studies have reported a lack of statistically significant differences between groups of physicians who received HRQOL feedback versus those who did not with regard to changes in prescription of medications, ordering of tests, referrals, or resource utilization [2, 6, 10, 39‐41].

The available evidence suggests that simply providing primary care physicians with HRQOL screening information without specific resource and management suggestions may not be sufficient in changing healthcare provider behavior [2, 6]. Some physicians even express concern that HRQOL instruments measure some problems they are not prepared to treat; that is, they question the extent to which they could intervene, practically and financially, when certain problems are detected beyond their time and resources to address adequately [42, 43]. The failure to produce changes in treatment may also be related to physician's limited experience with HRQOL measures or difficulty in translating HRQOL data into specific interventions to improve functioning.

The clinical utility of HRQOL measurement in clinical practice is likely to be enhanced by the availability of measures that are easy to use and interpret, physician education regarding the purpose and utility of HRQOL measurement, and specific management guidelines for deficits in HRQOL [6, 44, 45]. In fact, some investigators have concluded that the provision of targeted recommendations in conjunction with HRQOL scores are needed to yield beneficial effects of HRQOL measurement information on outcomes of care [8, 40, 46, 47]. Providing practitioners with easy to use linkage tools (e.g., to appropriate referrals, resources, and evidence-based interventions) may greatly improve the integration of HRQOL data into clinical practice. However, linking HRQOL data to appropriate treatment recommendations, with subsequent improvements in HRQOL outcomes, has not yet been demonstrated empirically in pediatrics.

Thus, while the above studies provide valuable information about the impact of HRQOL on clinical decision making in adult patients, virtually no pediatric studies exist. One study with the Pediatric Quality of Life Inventory™ (PedsQL™) Generic Core Scales and Rheumatology Module in pediatric rheumatology [48], however, found that for those patients for whom the pediatrician reported that HRQOL measurement at the point of service influenced her clinical decision-making, subsequent HRQOL outcomes demonstrated significant improvement in comparison to patients in which HRQOL assessment was not utilized [37]. Nevertheless, this single empirical study only highlights the need for further research.

A major hypothesized benefit to incorporating HRQOL measurement into routine clinical practice is the potential for identifying symptoms and problems that may result in improved patient outcomes over time. The findings are nevertheless mixed, with some physician-feedback intervention studies in the adult literature demonstrating no significant differences in patient outcomes such as functional status or health status between intervention and control groups [2, 40, 44, 48, 49], while other studies has yielded some support for improvement in mental health (e.g., depression) [10, 48] and role functioning [10] following feedback interventions.

Overall, the results from the adult literature suggest that routine implementation of standardized HRQOL screening may be a necessary but not sufficient condition for enhancing patients' HRQOL. For example, Rubenstein and colleagues [47] reported that providing physicians with patient functional status information in conjunction with management and community resource suggestions resulted in better mental health and social activity scores. Thus, incorporating specific resource management suggestions, such as appropriate referrals and tailored treatments, may enhance the potential potency of HRQOL measurement by providing physicians with variable options to identified problems.

In sum, the adult research on the integration of HRQOL measurement into clinical practice has demonstrated that the provision of HRQOL data to physicians is more likely to impact detection and diagnosis, but is generally not sufficient in changing physicians' management behaviors unless the HRQOL findings are linked to the available resources necessary to improve HRQOL outcomes. The general failure to produce changes in treatment may be related to physician's limited experience with such measures or the difficulty of healthcare professionals in translating HRQOL data into specific interventions to improve functioning. Finally, there may be significant barriers to the implementation of HRQOL measures into clinical practice, as discussed next.

Based on their semi-structured interviews of physicians concerning practice-based functional health assessment, McHorney and Bricker [42] found that physicians desired a point-of-service data collection process that did not interrupt the flow of care; perceived barriers included insufficient staff and computing resources, as well as the need for summary output to be able to be immediately implemented. A study in an outpatient oncology clinic addressed several of these issues, reporting that completion, scoring, and printing of patient-reported HRQOL data could be conducted during waiting room time [9]. Additionally, summaries provided to physicians did not lengthen average consultation time, and anecdotally, some physicians expressed the belief that the summary increased their efficiency [9]. Thus, intervention studies comparing duration of physician consultation have found that having access to HRQOL data does not necessarily increase the amount of time spent with patients [9, 10].

The application of computer-assisted assessment technology to the measurement of patient self-report and parent proxy-report in pediatric clinical care may reduce a number of the perceived barriers associated with the administration, completion, and scoring of standardized HRQOL instruments, and consequently represents one method for potentially overcoming several barriers to the use of these measures in pediatric practice. Computer-assisted HRQOL measurement as a screening methodology may facilitate the identification of areas of concern at the point of service, eliminate a lengthy interview process, and assist with targeting interventions [51‐54]. For instance, a one page electronically-generated report can provide a brief summary of symptoms and problem areas for the individual patient on a real time basis, which may influence clinical decision-making at the point of service. This summary report could also generate targeted referral suggestions to facilitate the clinical decision-making process and make suggestions for targeted evidence-based healthcare based on the HRQOL findings.

One attitudinal barrier to the incorporation of HRQOL assessment in clinical care is clinicians' views of the relevance and value of this information in patient care. Physicians vary greatly in their degree of willingness to discuss HRQOL issues with patients, and this willingness impacts patients' behavior (e.g., patients do not discuss their HRQOL concerns with their physician unless prompted to do so) [12, 16, 55].

Measuring the HRQOL of children in clinical practice may also meet with some resistance by clinicians who feel that the completion of these measures represents too great a burden for children and their parents, particularly children with serious illnesses. Oftentimes, a more qualitative approach is sought, with the belief that such methods are less burdensome and intrusive than standardized quantitative methods [56]. While an understandable concern, similar concerns have been seen in the past in regards to pain measurement in children [57] and HRQOL measurement in children with newly diagnosed cancer [58]. These concerns were met with brief instruments (to reduce respondent burden), developed with focus groups and cognitive interviews (to hear the voices of children and their parents), and by careful attention to the methodological details involved in establishing the reliability and validity of these instruments [59]. Subsequent to the development, field-testing, and documentation of the measurement properties of these standardized quantitative instruments, clinical trials have been conducted targeting pain management and enhancing HRQOL through symptom reduction using these quantitative instruments as outcome measures. Thus, standardized HRQOL instruments have the potential to improve the lives of children by providing the systematic documentation of the efficacy of treatment interventions designed for symptom relief [60]. In palliative care, for instance, standardized measures have the potential to increase the accountability and the quality of the care provided by comparing healthcare institutions or practitioners, and utilizing that information to inform consumers and aid quality improvement efforts [61].

A related attitudinal barrier to use in pediatric clinical practice is the perception by some physicians that HRQOL measures are neither sufficiently associated with nor predictive of subtle changes in physiological parameters [51, 62]. For example, small-to-medium correlation effect sizes were found between perceived HRQOL and Hb_a1c in children and adolescents with Type 1 diabetes [63], a finding consistent with the broader literature across diseases [64]. However, as succinctly summarized by McHorney [64]: "QOL scores correlate modestly at best with clinical outcomes. This finding suggests that clinical and human function are relatively independent. It does not imply that one or the other is inherently superior or correct. They simply measure different things, and using both will likely yield more information than any set alone" (p III-58). In fact, the use of physiological parameters to assess disease severity may be problematic, as many health conditions lack physiological markers, and such markers do not often apply across disease categories [65]. Furthermore, a patients' HRQOL may change in the absence of a measured biological change, or vice versa [62]. From this perspective, then, the incorporation of HRQOL measures into clinical practice should not be viewed as proxies for physiological parameters, but rather as a means of providing a more comprehensive evaluation of patient functioning across multiple life domains, particularly when meaningful physiological parameters are not available. This can be the hallmark for providing pediatric comprehensive care; multidimensional assessment leading to targeted interventions based on patient perceived needs.

As medicine shifts to a more patient-centered paradigm [38], willingness to incorporate HRQOL measurement is likely to increase. Physician education can be used to target the aforementioned attitudinal barriers, but Golden [45] notes the inadequacy of continuing medical education in changing physician behaviors and underscores the need for physician leaders who will advocate for and model the use of HRQOL measurement in clinical practice.

Finally, positive experiences with HRQOL measurement may reduce the aforementioned attitudinal barriers. Physicians in intervention studies who have used HRQOL measures report positive attitudes toward HRQOL assessment, expressing beliefs that the HRQOL data were accurate, were useful in clinical practice, could improve patient health status and satisfaction with care, provided new patient information, and facilitated physician-patient communication [2, 8‐10, 29, 39, 40, 44, 46]. These positive experiences are a necessary but not sufficient condition for change, as detailed later.