|Ahead of print
Assessment of health-related quality of life in cancer patients undergoing treatment using Health Utilities Index (HUI-3®) in east Delhi, India
Utsav Gupta, Madhu K Upadhyay, Rahul Sharma
Department of Community Medicine, University College of Medical Sciences and Guru Teg Bahadur Hospital, New Delhi, India
|Date of Submission||29-Dec-2019|
|Date of Decision||29-May-2020|
|Date of Acceptance||05-Jun-2020|
|Date of Web Publication||11-May-2021|
Department of Community Medicine, University College of Medical Sciences and Guru Teg Bahadur Hospital, New Delhi
Source of Support: None, Conflict of Interest: None
Background: Health-related quality of life (HRQOL) is a construct that focuses on the capacity for living afforded by the health status of a patient. Measurement of HRQOL allows a composite estimation of the capacity for living of a patient and can help capture the suffering experienced by the patient due to adverse effects of therapeutic interventions. This study was conducted to understand the health-related quality of life of cancer patients undergoing various modalities of treatment to generate evidence source for need-based intervention, to assess patients diagnosed with cancers, using Health Utilities Index - 3 (HUI-3) and assign them single-score values to gauge HRQOL and to measure the various domains of HRQOL and change in HRQOL after a period of three months of treatment.
Methods: A descriptive, longitudinal study was conducted amongst patients aged more than 18 years, who were diagnosed with cancer at a tertiary care multispecialty hospital in New Delhi. They were administered a standardized HUI-3 Questionnaire and their responses were recorded, simultaneously. Statistical significance for change in HRQOL score was assessed with paired t-test. Multivariate linear regression was used to identify the various correlates of HRQOL.
Results: The mean (± standard deviation) overall HRQOL score for all participants was 0.71 (± 0.262) [range=-0.09 to 1.00]. A follow-up assessment was carried out after three months and changes in health scores were subsequently recorded. A significant decrease in mean overall HRQOL score was seen for the entire group after three months of having been administered treatment. Single-attributes of emotion and ambulation were maximally affected amongst cancer patients after three months of treatment. On linear regression analyses, baseline HRQOL was a significant correlate of HRQOL at follow-up after three months.
Conclusion: Addressing the HRQOL of a cancer patient before starting treatment would address morbidity that might be present even after three months.
Keywords: Cancer, health-related quality of life, quality of life, utility score
In pursuit for improvements in clinical outcomes and long term survival of cancer patients, clinicians must also strive for enhancement in the health-related quality of life of the patients.
|How to cite this URL:|
Gupta U, Upadhyay MK, Sharma R. Assessment of health-related quality of life in cancer patients undergoing treatment using Health Utilities Index (HUI-3®) in east Delhi, India. Indian J Cancer [Epub ahead of print] [cited 2022 Oct 1]. Available from: https://www.indianjcancer.com/preprintarticle.asp?id=315799
| » Introduction|| |
The concept of “cure” in oncology has evolved from simple eradication of cancer to a more complex vision of “restoration of health,” in which “health” is not limited to physical aspects but also includes other features, grouped into the concept of “health-related quality of life” (HRQOL).
Most patients with advanced malignant disease suffer from a number of disturbing symptoms, including pain, asthenia, anorexia, nausea, and constipation. Despite the existence of published guidelines for pain management, many cancer patients are still inadequately treated for pain. According to a study, 42% cancer patients suffering from pain, reportedly received inadequate analgesic therapy. When one is measuring subjective outcomes such as symptoms, psychological well-being, and social functioning, it should be expected that the way in which individuals report their illness and suffering will be highly influenced by such factors as gender, age, culture, and socio-economic status. This has led to the development of the concept of Health-related Quality of Life, which pertains to the quality of life of a patient that is afforded by their health status.
The predicted increase in cancer incidence in the developing world, coupled with an unsatisfactory cure rate that is not likely to improve substantially, calls for redoubling of efforts in the area of palliative oncology. The primary goal of palliative treatment is to reduce symptoms, to improve patients' level of functioning, and to improve patients' overall quality of life (QOL). In studies of palliative treatment, HRQOL is rarely used as an outcome measure. Several factors may account for the paucity of HRQOL measurements in cancer clinical trials. First, in striving for improvements in cure rates and long-term survival, most clinical researchers, tend to minimize the importance of improvements in HRQOL. Another reason is that there is limited knowledge on how to measure HRQOL.
Nevertheless, a number of HRQOL measuring instruments have been developed for use in clinical trials. These scales allow a composite estimation of the adverse effects of therapeutic interventions during as well as after a treatment has been completed, and can be sufficiently sensitive to discriminate among the outcomes of different treatment intensities. These measures are useful in promoting greater accuracy in communication among patients, families, and health-care providers; and in helping to shape clinical policy and resource allocation. Quantitative estimates of HRQOL can capture disease burden from both patient and societal perspectives and are useful for planning health services and developing health policy.
Health Utilities Index Mark-3 (HUI-3) is a generic, preference-based, comprehensive system for measuring health status and health-related quality of life and for producing utility scores. HUI is reliable, valid, responsive, and easy to use tool, which evolved in response to the need for a standardized system to measure health status and HRQOL.
The HUI scoring systems provide utility scores on a generic scale where dead = 0.00 and perfect health = 1.00. Utilities are measured by the multi-attribute approach, in which a subject completes a questionnaire, highlighting areas of personal health like vision, mobility, pain, etc., The responses provide information about an individual's health status, which are then scored using a multi-attribute scoring function, derived from community preference measures for health states.
Limited studies from developing world, even lesser from India, have been reported where various domains of HRQOL have been assessed. A few studies have studied the effect of treatment on overall HRQOL in adult cancer patients. Co-morbidities like hypertension or diabetes might affect HRQOL of cancer patients, as well as demographic variables like socio-economic status, age, sex need to be mapped, to gather a more insightful representation of the actual disability. Better understanding of HRQOL might help to identify ways to improve health services delivery to cancer patients and make cancer treatment more affordable, at a time when World Health Organization (WHO) predicts the burden of cancer on healthcare services to increase, specifically in the South-East Asian region.
The current study was carried out with an aim to understand the HRQOL of cancer patients undergoing various modalities of treatment to generate evidence source for need-based intervention. The objectives of our study were to assess patients diagnosed with cancers using HUI-3 and assign them single-score values to gauge HRQOL. We also measured the various domains of HRQOL, change in HRQOL after a period of three months of treatment, and the correlates of HRQOL in patients undergoing treatment.
| » Methods|| |
Our study was a descriptive, longitudinal study, conducted amongst patients aged more than 18 years, who were diagnosed with cancer at a tertiary care multispecialty hospital in New Delhi. The newly diagnosed adult patients, who had not yet been put on any form of treatment, were recruited according to their willingness to participate, irrespective of their preliminary diagnosis, treatment regimen, or treating physician. Patients who were critically ill, not able to comprehend the questionnaire, or those who were not expected to complete at least three months of their treatment from the study setting were excluded from the study. Because our study setting was located in Delhi it receives patients from outside Delhi and also from outside the country; hence, it was possible that all of them would not have completed three months of treatment from this facility. Many patients leave for their native states after the diagnosis gets established. Moreover, the study setting was a private institution and considering that cancer treatment involves long-term and entails heavy expenses, some patients would be compelled to leave the treatment because of financial implications.
We used the standard deviation of HUI-3 score of 0.25, as reported from previous studies that had used HUI-3 amongst cancer patients, for the calculation of sample size. At a desired precision of 0.10 in the outcome measure and a 95% confidence interval; the sample size was calculated to be 96. We expected an attrition of 25%, which meant a minimum of 120 participants, would have to be enrolled in our study. We enrolled 135 patients, of which 122 were successfully followed up for three months.
The duration of our study was one year, from January 2017 to December 2017. The participants were interviewed using a standardized HUI-3 Questionnaire, interviewer format. Responses were recorded simultaneously using pen-and-paper for each participant. A follow-up assessment was carried out after three months, during which the participants had received or were undergoing treatment. Changes in health scores were subsequently computed. Self-prepared, semi-open-ended questionnaire to elicit the socio-demographic and clinical details of the participants was also used. No local language translation was available for HUI-3. The questionnaire was translated to Hindi by an expert, and subsequently back-translated to English by an independent expert for validation. Also, the questionnaire was pilot-tested in a small sample population.
Subject responses were decoded to generate HUI-3 attribute levels for each individual attribute, as well as the overall HRQOL score was computed, ranging from 0.00 (death) to 1.00 (perfect health) for every study subject, according to the coding procedure outlined in the Health Utilities Incorporated (HUInc) Procedure Manual. As mentioned by the tool authors, scores may assume a negative value up to –0.37, signifying conditions worse than death. A change in the health score of 0.03 is considered clinically significant. Quantitative data was presented as means (± standard deviation (SD)). Statistical significance for qualitative data was assessed with Chi-square while that for quantitative data was done with paired/student t-test. For comparison between multiple groups, analysis of variance (ANOVA) was used. P value of <0.05 was considered statistically significant.
Multivariate linear regression analysis was used to identify the various correlates of HRQOL. Age and sex were put into the model for adjustment. Independent variables with P value <0.25 on univariate analysis were entered into the model.
Permission was obtained from the Institutional Ethics Committees to conduct the study.
| » Results|| |
A total of 122 participants were successfully followed-up for three months in our study. The mean age of all the participants was 53.5 (±12.36) (range = 20 - 79) years. There were a slightly higher number of men (66, 54%) in our study. A few participants reported having other chronic conditions like hypertension (25, 20%), diabetes mellitus (22, 18%), hypothyroidism (6, 5%), tuberculosis (3, 2%). Sixty-nine (56.5%) participants were diagnosed with cancers of the head and the neck region including oral cancer while 33 (27%) patients had been diagnosed with breast cancer. Staging according to the tumour nodes metastasis (TNM) staging criteria was done for 118 participants by their treating physicians when they were recruited, out of which twelve had a metastasised lesion. Surgery, either alone or in combination with other therapies, was the most common modality of treatment, planned for 86 (70.5%) participants.
On assessment of HRQOL, participants reported the lowest mean single-attribute score for emotion, followed by pain, cognition and speech at baseline. At baseline assessment, the mean overall HRQOL score for all the participants was 0.72 (±0.26) [range = -0.09 - 1.00]. After three months, a reassessment was carried out, and changes in HRQOL scores were recorded. A statistically significant decrease was seen in the scores for attribute of emotion and ambulation, signifying that more participants reported a worsening of these attributes over the three-month period. The mean overall HRQOL score after three months [(0.67 ± 0.26), range = 0.05 - 1.00] also showed a decrease of 0.04, which was clinically and statistically significant [Table 1]. Forty-one participants experienced fatigue and generalized body pain following three months of therapy, while 37 patients experienced loss of appetite.
|Table 1: Mean single-attribute and overall Health Utility Index-3 (HUI-3) scores before administration of treatment and after three months|
Click here to view
The effect of socio-clinical factors on the change in the mean overall HRQOL score was also analyzed. Individuals who were either above the age of 50 years, residing in Delhi, overweight (body mass index (BMI) >24.9 kg/m2), were not currently married, diagnosed with cancers of oral cavity or the breast, having metastasised lesions, having a co-morbidity, having underwent surgery, showed a statistically significant decrease in their HRQOL score over a period of three months [Table 2] and [Table 3].
|Table 2: Mean overall Health Utility Index-3 (HUI-3) scores according to the sociodemographic attributes of the participants|
Click here to view
|Table 3: Mean overall Health Utility Index-3 (HUI-3) scores according to the clinical attributes of the participants|
Click here to view
Linear regression analyses was conducted to examine the relationship between HRQOL scores at baseline and follow-up, and the various potential correlates, which provided the best fitting model for the covariates.
Age, sex, occupation, site of cancer, presence of metastasis, and symptoms at presentation (viz. oral ulcers, local lump) were included in the regression model for HRQOL at baseline. Occupational status and presence of metastasis was significantly and negatively associated with HRQOL. Sex was also significantly associated with HRQOL. The proportion of variation explained (R2) was modest although overall regression power for the model was signiﬁcant (p value = 0.007) [Table 4].
|Table 4: Multivariate linear regression of baseline Health Utility Index-3 (HUI-3) with the socio-clinical variables|
Click here to view
Sex, occupation, marital status, bleeding manifestations, administration of surgery, side effects of treatment, and baseline HRQOL were included in the regression model for change in HRQOL. With the exception of occupation, all variables were negatively associated with change in HRQOL. The baseline HRQOL and sex were significant correlates of change in HRQOL. The proportion of variation explained by the model was R2 = 0.268. Overall power for the model was signiﬁcant (p value <0.001) [Table 5].
|Table 5: Multivariate linear regression of change in Health Utility Index-3 (HUI-3) with the socio-clinical variables|
Click here to view
| » Discussion|| |
In single-attribute mean scores, significant decrease was seen in the attributes of Emotion and Ambulation. Lovrics et al. has reported significant decrease in attribute of Pain, Emotion, and Ambulation following surgery. In our study, we found virtually no change in Pain scores after three months of follow-up. In contrast, Falicov et al. has reported a significant increase in Pain scores at three months using HUI. Batra et al. made similar observations for pediatric patients diagnosed with lymphoma. In our study, the mean overall HUI-3 (HRQOL) score decreased significantly over the three-month period, which was in contrast to the findings made by Rodrigues, Bank, and Furlong. They reported an increase in mean overall score after three months using the HUI-3 questionnaire.,,
Previous studies have found that HRQOL is most compromised during the three to six months after diagnosis, regardless of treatment protocol. Lovrics et al. and Falicov et al. found the overall HRQOL score fall over the same time, although that change was not statistically significant., Studies where patients were followed-up for a longer duration reported that HRQOL scores returned to baseline or increased after one-year. Tempier found a clinically and statistically significant increase in overall HRQOL scores at one-year follow-up.
Sociodemographic factors (age, occupation, sex), as significant predictors of HRQOL score, were reported in numerous studies.,,, Women were far more likely to experience a decrease in their HRQOL score compared to men. Individuals who were above the age of 50 years were also more likely to show a decrease in overall HRQOL scores when compared to their counterparts. In contrast, Ramsey et al. and Krahn et al. did not find age, sex, or socio-economic status to be a predictor of overall HUI-3 (HRQOL) scores.
At baseline, sociodemographic factors like sex and occupation status were found to be significant correlates of HUI-3 score on multivariate regression analysis. Education status was not used in the model since it has high correlation with occupation. Lovrics found only occupation to be a significant predictor of HRQOL.
Clinical factors found to be predictors of baseline overall HUI-3 score (HRQOL) were lung cancer, metastasised lesions, and primary symptom of oral ulcers and/or local lump. Co-morbidities were found to be significant predictors in models computed in four previous studies.,,, Although our results indicated that, the presence of a co-morbidity predisposed the study participant to a significant decrease in the overall HUI-3 (HRQOL) score after three months, presence of a co-morbidity was not a significant predictor of HUI-3 score according to the regression model.
Baseline overall HUI-3 score was found to be a significant predictor of overall HUI-3 scores at follow-up. Penn et al. made similar findings, where scores at baseline strongly predicted the score at one-year follow-up. The significant relationship between HRQOL measured early after diagnosis and that measured three months later offers an opportunity to identify those most at risk of poor HRQOL later on. Men, married individuals, those reporting bleeding manifestation at baseline, undergoing surgery and those who experienced local reaction as a side-effect of treatment, best predicted the change in HRQOL over a period of three months in our regression model.
According to modality of therapy administered, participants who underwent radiotherapy reported a mean overall HUI-3 score lower than those who underwent surgery or chemotherapy. Most common side effects reported following radiotherapy were oral ulcers, difficulty swallowing, local mucositis, other reactions, etc. Most patients underwent radiotherapy as an adjuvant mode of treatment, after they had already undergone surgery. Because of this, patients reported side effects of radiotherapy at follow-up, which were conspicuously unreported by patients who underwent surgery only. Batra et al. found that patients who underwent chemotherapy alone or in combination with surgery, reported a better HRQOL score after four months of follow-up compared to patients who underwent radiation in combination. A few authors have reported the volume of irradiation to be a significant predictor of HRQOL in cancer patients.,, Krahn et al. and Hammerlid et al. also found radiotherapy to be adversely associated with HRQOL.
Our study is a comprehensive assessment of HRQOL of cancer patients using a validated tool. This is one of the few studies carried out in northern India, investigating the effect of cancer and its treatment on HRQOL amongst adult patients. This study gives an estimate of the unreported morbidity prevalent amongst patients when they are diagnosed with a prolonged illness and after they have been administered treatment. Study participants assessed their own health status without the use of proxy assessors. Proxy assessors in the form of primary caregivers, attendants, etc., tend to overestimate the morbidity of the patient compared to the patient themselves.
Socio-clinical factors, which predict the change in HRQOL, have been analyzed and reported in our study. Treating physicians can recognize these factors and the specific single-attributes measured in HUI-3, prior to the initiation of a treatment in cancer patients. They can, hence, formulate protocols that can alleviate emotional and cognitive morbidity, which might cause deterioration in overall HRQOL immediately after treatment has been administered.
Our study had certain limitations. The sample size recruited for our study was adequate to capture the change in HRQOL over three months, but was too small to capture differences in HRQOL scores across various socio-clinical groups (e.g., age, sex, site of cancer, delay in diagnosis, modality of treatment, duration of treatment, co-morbidities, etc.). There was a potential source of selection bias, due to the inclusion criteria of only recruiting patients who had not been previously treated for cancer. Many patients diagnosed with cancers of blood, lung, gastrointestinal tract (GIT), etc., were naturally excluded as they had been misdiagnosed and were put on treatments that had relieved them symptomatically and influenced their baseline HRQOL. Patients who were critically ill, or those who were not expected to complete at least three months of follow-up from the institute were also excluded from our study and were a source of potential bias. Our cohort comprised predominantly of patients with oral cancer from a single private tertiary-care institute, which might affect generalizability of our results to all types of cancer patients.
| » Conclusion|| |
In our study, we found a significant deterioration in HRQOL scores over a period of three months. This change was largely dependent on the baseline HRQOL score that was recorded. Addressing the HRQOL of a cancer patient before initiation of treatment would address morbidity, which might be prevalent even after three months.
The follow-up period of our study was only three months. Most authors have found a worsening of HRQOL after a period of three months, and a return to baseline HRQOL values by one year. Hence, studies having a longitudinal design with multiple follow-ups that capture health status up to one year after initiation of treatment should be conducted. This will accurately register quality of life amongst patients who have undergone long regimens of treatment.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Schwartz CL. Health status of childhood cancer survivors. JAMA 2003;290:1641-3.
Cleeland CS, Gonin R, Hatfield AK, Edmonson JH, Blum RH, Stewart JA, et al
. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 1994;330:592-6.
Park KU. Assessment of change of quality of life in terminally ill patients under cancer pain management using the EORTC Core Quality of Life Questionnaire (QLQ-C30) in a Korean sample. Oncology 2008;74:7-12.
Kaasa S. Using quality of life assessment methods in patients with advanced cancer: A clinical perspective. Eur J Cancer 1995;31A: S15-7.
Barr RD, Simpson T, Whitton A, Rush B, Furlong W, Feeny DH. Health-related quality of life in survivors of tumours of the central nervous system in childhood--A preference-based approach to measurement in a cross-sectional study. Eur J Cancer 1999;35:248-55.
Feeny D, Furlong W, Mulhern RK, Barr RD, Hudson M. A framework for assessing health-related quality of life among children with cancer. Int J Cancer Suppl 1999;12:2-9.
Battié MC, Jones CA, Schopflocher DP, Hu RW. Health-related quality of life and comorbidities associated with lumbar spinal stenosis. Spine J 2012;12:189-95.
Horsman J, Furlong W, Feeny D, Torrance G. The health utilities index (HUI
): Concepts, measurement properties and applications. Health Qual Life Outcomes 2003;13:1-13.
Lovrics PJ, Cornacchi SD, Barnabi F, Whelan T, Goldsmith CH. The feasibility and responsiveness of the health utilities index in patients with early-stage breast cancer: A prospective longitudinal study. Qual Life Res 2008;17:333-45.
Falicov A, Fisher CG, Sparkes J, Boyd MC, Wing PC, Dvorak MF. Impact of surgical intervention on quality of life in patients with spinal metastases. Spine (Phila Pa 1976) 2006;31:2849-56.
Batra P, Kumar B, Gomber S, Bhatia M. Assessment of quality of life during treatment of pediatric oncology patients. Indian J Public Health 2014;58:168-73.
] [Full text]
Rodrigues KEDS, Machado STC, Ferreira MA, Martins TF, Viana MB, Oliveira BM De. Health-related quality of life among teenagers during cancer treatment in a developing country: Patients' and proxies' reports. Pediatr Hematol Oncol 2013;30:307-16.
Banks BA, Barrowman NJ, Klaassen R. Health-related quality of life: Changes in children undergoing chemotherapy. J Pediatr Hematol Oncol 2008;30:292-7.
Furlong W, Rae C, Feeny DH, Gelber R, Laverdiere C, Michon B, et al
. Health-related quality of life among children with acute lymphoblastic leukemia. Pediatr Blood Cancer 2012;59:717-24.
Tempier R, Pawliuk N. Measuring patient quality of life: Is the health utility index useful? Psychiatr Q 2007;78:53-62.
Mols F, Coebergh JWW, van de Poll-Franse LV. Health-related quality of life and health care utilisation among older long-term cancer survivors: A population-based study. Eur J Cancer 2007;43:2211-21.
Heydarnejad MS, Hassanpour Da, Solati DK. Factors affecting quality of life in cancer patients undergoing chemotherapy. Afr Health Sci 2011;11:266-70.
Wexler DJ, Grant RW, Wittenberg E, Bosch JL, Cagliero E, Delahanty L, et al
. Correlates of health-related quality of life in type 2 diabetes. Diabetologia 2006;49:1489-97.
Alessi D, Dama E, Barr R, Mosso ML, Maule M, Magnani C, et al
. Health-related quality of life of long-term childhood cancer survivors: A population-based study from the Childhood Cancer Registry of Piedmont, Italy. Eur J Cancer 2007;43:2545-52.
Ramsey SD, Ph D, Berry K, Ph D, Moinpour C, Ph D, et al
. Quality of Life in Long Term Survivors of Colorectal Cancer. Am J Gastroenterol 2002;97:1228-34.
Krahn MD, Bremner KE, Alibhai SMH, Ni A, Tomlinson G, Laporte A, et al
. A reference set of health utilities for long-term survivors of prostate cancer: Population-based data from Ontario, Canada. Qual Life Res 2013;22:2951-62.
Ko CY, Maggard M, Livingston EH. Evaluating health utility in patients with melanoma, breast cancer, colon cancer, and lung cancer: A nationwide, population-based assessment. J Surg Res 2003;114:1-5.
Penn A, Lowis S, Stevens M, Hunt L, Shortman R, McCarter R, et al
. Family, demographic and illness-related determinants of HRQL in children with brain tumours in the first year after diagnosis. Pediatr Blood Cancer 2009;53:1092-9.
Barrera M, Atenafu E, Hancock K. Longitudinal health-related quality of life outcomes and related factors after pediatric SCT. Bone Marrow Transplant 2009;44:249-56.
Feeny D, Furlong W, Barr RD, Torrance GW, Rosenbaum P, Weitzman S. Health status of survivors of childhood cancer. J Clin Oncol 1992;10:923-8.
Hammerlid E, Silander E, Hörnestam L, Sullivan M. Health-related quality of life three years after diagnosis of head and neck cancer--A longitudinal study. Head Neck 2001;23:113-25.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]