Despite the importance of HRQL for long-term survivors, there are currently only two HRQL measures designed specifically for long-term survivors, one of which is a modification of the other. Researchers at the City of Hope National Medical Centre were originators in developing a HRQL measure for long-term survivors. They developed the Quality of Life-Cancer Survivors scale (QOL-CS) based on their conceptualization of HRQL as having four dimensions: physical, psychological, social, and spiritual [27
]. Although this scale reflects an attempt to recognize HRQL issues relevant to long-term cancer survivors, it has a number of limitations. Items were based on a small number of cancer survivors and validation of the scale was based on survivors ranging from 4 months to 28 years after diagnosis (thus including newly diagnosed patients). Some items have problematic wording, in that they ask about change but fail to indicate its direction (e.g., "has your illness or treatment caused changes in your self-concept?"). The domains often measure multiple constructs at once (e.g., social interaction includes appearance, sexual functioning, and family distress). Further, several items ask about distress at the time of diagnosis and treatment.
Wyatt and colleagues [15
] developed the Long-term Quality of Life (LTQL) questionnaire based on the same conceptual model. Based on data from a sample of female cancer survivors, they conducted a factor analysis, an internal consistency analysis, and determined content validity. These results yielded 34 items loading on 4 factors that are slightly different from the QOL-CS: somatic concerns, spiritual/philosophical view of life, fitness, and social support. While the psychometric approach of the LTQL is an advantage,, some of the items themselves are still problematic and the domains often encompass more than one important aspect of HRQL. For example, the somatic concerns domain includes both body-image problems and pain. Type of cancer and its treatment may well vary in terms of their impact on these two concerns. The broad domains do not allow investigators to look at more specific HRQL areas.
The QLACS was developed in response to these limitations. The analyses reported here suggest that the QLACS has good internal consistency reliability, and adequate concurrent and retrospective validity. Responsiveness to change is encouraging, but needs more testing in other samples.
Important findings of this study include the concurrent and retrospective validity of both the Generic Summary score and the Cancer-Specific Summary score in assessing HRQL of long term cancer survivors. We found that the Generic Summary score tracks well with some existing generic measures, while the Cancer-Specific Summary Score showed a much lower correlation with other generic QOL measures. Compared to other instruments, the Cancer-Specific Summary score exhibits relatively high predictive validity as judged by its associations with number of missed days of work/usual activity following diagnosis, the respondent's feelings of preparedness for dealing with cancer, and coping strategies. For example, the number of days missing work has a strong significant correlation with the QLACS CSS, suggesting that women who tend to miss a high number of workdays or usual activity immediately following diagnosis have a lower QOL several years later. Women who reported being well prepared (responded with 4 or 5 on the preparedness scale) showed substantially better quality of life when projected to several years later in the QLACS CSS measure than the FACT-B. These retrospective validity results suggest that the QLACS had better predictive validity than the FACT-B.
Our qualitative data suggests that some of the FACT-B items may not be relevant for long-term survivors. Despite answering all of the items, 52 women indicated that at least one FACT question was no longer relevant to them. This occurred most frequently for questions about pain (N = 30), side effects of treatment (N = 29), nausea (N = 14), and family communication about illness (N = 14). We should point out, however, that we did not ask respondents to indicate when QLACS questions were not relevant.
It is also worth comparing the QLACS scores in this sample with those of the breast cancer sample in the original QLACS paper. The only domain score that was significantly different was negative feelings, with the younger sample reporting greater negative feelings. This finding is consistent with other research showing greater psychological morbidity among younger women with breast cancer [42
The QLACS was generally responsive to self-reported changes. One finding about this population of long-term survivors is the absence of detectable changes in HRQL in several domains – Cognitive Problems, Sexual Problems, Family Distress, and Recurrence Distress (all p > 0.20), even when a change in these areas had been reported. There may be several explanations for this. While it is possible that the QLACS is not sufficiently responsive in these domains, the absence of detectable changes could also be due to this particular sample or the nature of the life change questions. More clinical measures might show greater responsiveness. We also observe that the instrument tends to be more responsive to negative changes. For example, for respondents reporting "Somewhat worse" or "Quite a bit worse" in wanting to socialize (Social Avoidance domain), the effect size was 2.98, compared to an effect size of 1.85 for the group that reported "Somewhat better" or "Quite a bit better" for wanting to socialize. It is important to continue evaluation of the responsiveness of the QLACS in different samples. To our knowledge, however, we are unaware of responsiveness data by domain for other scales such as the FACT.
There are several limitations of this study. The analyses reported here are based on a follow-up of well-educated, young, white female breast cancer survivors. Studies need to be conducted among more diverse samples of survivors of different ethnicities, ages, and cancers to gather more data on the usefulness of the QLACS for other cancer survivors. Another limitation of this work is the number of women lost to follow-up. Finally, our responsiveness data were limited to self-reported life. More research using clinical measures or clinical trial data to assess responsiveness is desirable. Nevertheless, the QLACS may be a a useful measure for assessing HRQL among long term cancer survivors and warrants further evaluation.