Nurses provide valuable proxy assessment of the health-related quality of life of children with Hodgkin disease


  • We thank the many nurses who helped out with this study, including Deanna Lawson, Maryanne Gibson and Catherine Bourne. We would like to acknowledge Dr. Isabelle Gaboury, Dr. Murray Krahn, Dr. Lawrence Jardine, and Dr. Dorothy Barnard for their contributions as coauthors.



We prospectively compared the proxy reporting of health-related quality of life (HRQL) by parents and nurses of children with Hodgkin disease to see how well they correlated with the children's report.


Children with all stages of Hodgkin disease, their parents, and the clinic nurse were all asked to complete 4 different HRQL measures at 4 time points: 2 weeks after the first course of chemotherapy, on the third day of the second course of chemotherapy, during the third week of radiation, and 1 year after diagnosis.


Fifty-one patients from 12 centers across Canada were enrolled in the study between May 1, 2002 and March 31, 2005. Two patients were excluded. The children's Pediatric Quality of Life Inventory (PedsQL) generic scores increased from 64 at base line to 81 at the end of the study. There was substantial agreement (intraclass correlation coefficient >0.6) among the participants' scores at most time points except when the child was actively receiving inpatient chemotherapy. At that time, there was only fair to moderate agreement between the child and parent, with the parents on average rating the PedsQL generic score 5 points lower than the child. The nurses still had substantial agreement at that time point with the PedsQL generic and cancer module as well as the EuroQol EQ-5D visual analogue scale visual analogue scale.


Over the course of treatment for Hodgkin disease, there was substantial agreement among the parent's, nurse's, and children's reported HRQL scores. Nurses contribute valuable additional information as proxy respondents. Cancer 2010. © 2010 American Cancer Society.

Once a child is diagnosed with cancer, they embark on a plan of treatment that is complicated by numerous side effects including nausea, vomiting, fatigue, and anorexia. These side effects, added on to the significant symptoms from their underlying malignancy, result in a quality of life that is significantly decreased compared with the general population.1, 2 The focus of cancer therapy is understandably to cure as many patients as possible, but this involves aggressive multiagent chemotherapy that contributes to the significant morbidity seen in these patients. An important consideration when comparing treatment arms should be to look at health-related quality of life (HRQL). If therapies have equivalent survival results, then the HRQL results can be used to select the arm with the least morbidity.

Fortunately, the number of HRQL measures available for children has increased in the last decade, with more rigorous methodology ensuring that these measures can be reliably used as outcomes.2-4 The one issue that remains is how best to measure HRQL. Ideally, the child should complete the measures themselves, but as many children are too young or too sick to complete the tools, proxy input is a good second choice. Many studies have looked at the agreement between parents and children,1, 5, 6 but few have looked at other potential proxy assessors, in particular, nurses.7-9

The purpose of this study was to determine the agreement in HRQL scores among children recently diagnosed with Hodgkin disease, their parents, and their clinic nurse followed for 1 year. The other objective was to provide an overview of the change in HRQL seen in patients with Hodgkin disease during chemotherapy, radiation, and off treatment.

A companion study looking at the ability of the 4 different HRQL measures used in this study to detect change in HRQL over time (responsiveness) has been submitted for publication.


Recruitment of Participants

Children with a new presentation of pathologically confirmed Hodgkin disease between 8 and 17.99 years of age were recruited from oncology clinics at 12 pediatric centers across Canada: B.C. Children's Hospital, Vancouver, British Columbia; Alberta Children's Hospital, Calgary, Alberta; Stollery Children's Hospital, Edmonton, Alberta; Saskatoon Cancer Centre, Saskatoon, Saskatchewan; Cancer Care Manitoba, Winnipeg, Manitoba; Children's Hospital of Western Ontario, London, Ontario; McMaster Hospital, Hamilton, Ontario; The Hospital for Sick Children, Toronto, Ontario; Children's Hospital of Eastern Ontario Ottawa, Ontario; Kingston General Hospital, Kingston, Ontario; Montreal Children's Hospital, Montreal, Quebec and IWK Health Centre, Halifax, Nova Scotia. A lower age limit of 8 was chosen so that participants could self-complete all of the questionnaires. Research ethics board approval was obtained from each institution and signed informed consent was obtained before enrolment.

The child, 1 parent, and the clinic nurse were all asked to complete a package of HRQL measures (the Health Utilities Index Mark 2 and Mark 3 [HUI 2 and 3], the Pediatric Quality of Life Inventory [PedsQL] generic and cancer modules, the EuroQol EQ-5D visual analogue scale [EuroQol] and the Lansky Play-Performance Scale [Lansky]) at 4 time points: 2 weeks after the first course of chemotherapy (Time 1), on the third day of the second course of chemotherapy (Time 2), during the third week of radiation (Time 3), and 1 year after diagnosis (Time 4). The timing was selected so that all stages of disease could participate. We did not contact patients before the first cycle of chemotherapy as many of the patients were being approached to participate in biology and treatment studies at that time. Patients who did not receive radiation did not complete Time 3.

Parents were encouraged not to interact with their child when completing the questionnaires. If the patient had questions, then they were directed to the research assistant, not their parent or nurse. Questionnaires were given to the participants at a clinic visit and they were requested to complete the package that same day before leaving the clinic. Nurses completed their questionnaire packages separate from the child and parent. To provide consistent reporting, the same nurse and parent were asked to complete all of the questionnaires.


All of the measures were framed to encompass the HRQL of the preceding week.

PedsQL 4.0 Generic Core Scales and PedsQL 3.0 Cancer Module

The PedsQL takes a modular approach to measuring HRQL, comprising 1 generic core scale in combination with a disease-specific module. The measure was originally developed in the pediatric oncology population.10 There is no summary score for the cancer module but to facilitate comparison across measures the score was calculated using the sum of the items over the number of items answered, as is used for the generic core scale.

Health Utilities Index Mark 2 and 3

The HUI 2 and 3 are generic, multiattribute questionnaires that measure health status and HRQL. The HUI 2 and 3 are complimentary measures and were used in tandem for this study. The scores are defined on a scale from 0-1, but to facilitate comparison with the other measures, the results were multiplied by 100.

The EuroQol EQ-5D

The EuroQol EQ-5D was developed by an international, multidisciplinary group of researchers in adult patients to establish a standardized generic instrument capable of being expressed as a single index value. It is made up of 2 components: a description of the patient's health status and a self-rating of their current status, using a visual analogue scale (VAS). The VAS is a vertical thermometer anchored by the “best imaginable health state” at the top of the scale (100) and the “worst imaginable health state” at the bottom (0). For the purposes of this study, we used only the VAS, as there are no “value sets” for populations younger than 20 years of age to use with the descriptive component.

The Lansky Play-Performance Scale

The Lansky is a pediatric adaptation of the Karnofsky Performance Status Scale Index of global functioning. It comprises a single domain with 11 items scaled from 0 to 100, ranging from unresponsive (0) to full active (100). This measure has been widely used in pediatric oncology but is limited by looking solely at functioning and, therefore, is not truly a measure of HRQL. Because it is widely used by pediatric oncologists, we have included it despite its limitations.

A Priori Hypothesis

We expected that the parents would have moderate to substantial concordance with the children's response, whereas the nurses would have fair to moderate concordance. We expected that the HRQL would be lowest during chemotherapy, increase during radiation, and would be the most favorable at 1-year postdiagnosis.

Statistical Analysis

Interrater Agreement

We evaluated the concordance among instrument scores completed by patients, parents, and nurses, using an average measure intraclass correlation coefficient (ICC) with a random effects model. Agreement is classified as recommended by Landis et al, with an ICC of <0.0 indicating poor, 0.0-0.20 slight, 0.21-0.40 fair, 0.41-0.60 moderate, 0.61-0.80 substantial, and 0.81-1.0 almost perfect agreement.11


Fifty-one children were prospectively enrolled in the study, with 49 patients included in the final analysis. Two patients were excluded: 1 patient died shortly after the first time point and the other patient failed to complete any of the questionnaires. Four patients relapsed during the study follow-up period (8%). The participant flow chart is shown in Figure 1. 45% of the children were male, with a mean age of 14.7 years and a range of 8.9 to 18 years. Stage IIA was the most common stage at diagnosis (39%), with IVA (18%), IV B (12%), and IIB (10%) being the next most common stages. Twenty-seven percent of the patients did not receive radiation.

Figure 1.

Participant flow is shown in this diagram.

The child-reported PedsQL total score 1 year from diagnosis improved from 64 at baseline to 81. Figure 2 demonstrates the changes in child scores of the measures over the 4 time periods.

Figure 2.

Change in child-reported HRQL scores is shown from the time of enrollment to 1 year after diagnosis: (A) change in Health Utilities Index Mark 2 and Mark 3 scores, (B) change in the Pediatric Quality of Life Inventory generic and cancer module scores, and (C) change in the Lansky Play-Performance Scale and EuroQol EQ-5D visual analogue scale scores.

Concordance Between Raters

There was, at least, fair agreement between the parent and nursing proxy scores and the child scores, with the majority of responses having substantial agreement (Table 1). The agreement between parent and child was weakest at time 2, on the third day of the second course of chemotherapy, with the parents scoring, on average, the children's HRQL 2-5 points lower than the child rating. The nurses still had substantial agreement at that time point with both PedsQL measures and the EuroQol.

Table 1. Intraclass Correlation Statistics
  • PedsQL indicates Pediatric Quality of Life Inventory; HUI2, Health Utilities Index Mark 2; HUI3, Health Utilities Index Mark 3; Lansky, Lansky Play-Performance Scale; EuroQol, EuroQol EQ-5D visual analogue scale.

  • a

    Substantial or greater agreement (ICC > 0.6).

PedsQL Cancer0.75a0.440.69a0.580.67a0.69a0.77a0.76a


From the time a child is diagnosed with Hodgkin disease until they have completed therapy, there is a substantial change in their quality of life. We used several different types of measures, all of which demonstrated a significant improvement in HRQL over the 1-year period.

We asked parents and nurses to perform a fairly difficult cognitive exercise: they were asked to try to decide if a particular issue was a problem for their child (parent) or patient (nurse). This act of being a proxy involves significant mental gymnastics; therefore, it is not surprising that many studies indicate poor agreement between parents and children.12-14 Conversely, our study showed substantial agreement among children and their parents and clinic nurses for two thirds of the measures and time points. In fact, almost 90% of the summary scores had at least moderate (>0.4) concordance with the children's scores. This is very reassuring, indicating that proxy respondents, even clinic nurses who see the patients only intermittently, can do a reasonable job of estimating a child's overall quality of life. This is consistent with an analysis of 8406 parent-proxy reports involving the PedsQL generic measure that found moderate to substantial agreement across most of the scales and summary scores.15 This is essential information for when the child is either too young (in general younger than 7 years) or too sick to complete the questionnaire.

The one time the parents seemed to have trouble was during active chemotherapy treatment, which, for children with Hodgkin disease, is primarily done inpatient. The parents had only poor to moderate concordance with all of the measures, while the nurses had moderate to substantial agreement. The reason for this is not clear, but it may be because of emotional impact on parents of having their child admitted to hospital, resulting in scores that were 2-5 points lower. The nurses can provide a more objective assessment of the children's situation, providing better concordance with the children's report. A qualitative study looking at discordance between the parent and their child indicated that the differences were primarily due to differences in reasoning and response styles, which may have also played a role in our study.16

There are only limited studies looking at nurses as proxy respondents in pediatrics, with most of the work being done with the HUI3. Two of the studies found that there were significant missing data and recommended that nurse proxies not be used.8, 17 This was because of the explicit design constraints of those studies, which required the nurses to actually observe each item of the questionnaire.18 We did not put those restrictions on the nurses involved in our study and clearly they did an excellent job. It should be mentioned that our clinic nurses had an ongoing relationship with the patients, which may have contributed to their substantial concordance with the children's reports. Our study should put an end to the practice of requiring proxies to report only on items they observe, at least in the setting where there is an established relationship between the nurse and patient.

Limitations of our study include the fact that we did not collect data on nonrespondents, and so we cannot be certain of how representative our sample is. Also, our trial did not include children younger than 8 years, who have been shown in other studies to provide unique information about HRQL. A randomized placebo controlled study of children with cancer receiving weekly epoetin alpha found only a significant difference among children aged 5-7 years.19 This reinforces the importance of using self-report whenever possible and relying on proxy report only when pediatric patients are not able to provide self-report.


Over the course of treatment for Hodgkin disease, there is considerable agreement among the parent's, nurse's, and children's reported HRQL scores. Nurses contribute valuable additional information as proxy respondents. We would recommend that they be included, in addition to parents to provide a valuable “second opinion” of a children's HRQL, particularly in the setting when the child is unable to provide a self-report.


This research was supported by the Hospital for Sick Children Foundation.