clinical and methodological issues in a cost-utility analysis alongside a
randomized clinical trial in patients with rectal cancer undergoing
total mesorectal excision
Brink, Mandy van den
Citation
Brink, M. van den. (2005, June 28). Economic evaluation of preoperative radiotherapy in rectal cancer : clinical and methodological issues in a cost-utility analysis alongside a randomized clinical trial in patients with rectal cancer undergoing total mesorectal excision. Retrieved from https://hdl.handle.net/1887/4273
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in theInstitutional Repository of the University of Leiden Downloaded from: https://hdl.handle.net/1887/4273
ASSESSING COSTS,
QUALITY OF LIFE AND
PREFERENCES IN PATIENTS DIAGNOSED
W ITH RECTAL CANCER:
TELEPHONE OR
FACE-
TO-
FACE FOLLOW -
UP INTERVIEW S?
Mandy van den Brink, Wilbert B. van den Hout, Sylvia J.T. Jansen, Job Kievit, Cornelis J.H. van de Velde, Anne M. Stiggelbout
Abstract
Purpose To investigate the feasibility and convergent validity of telephone and face-to-face follow-up interviews for the assessment of costs, quality of life, utility values and probability trade-offs in recurrence-free patients previously diagnosed with rectal cancer.
Design We randomly assigned 87 patients with rectal cancer to receive telephone (n=42) or face-to-face interviews (n=45) at 3 and 12 months after surgery for their primary tumor. During a baseline face-to-face interview before surgery all patients received a binder containing all measurement instruments. Questions and interview structure were identical for all patients.
Introduction
The assessment of all relevant outcomes in each patient is the best guarantee of internal validity in cost-utility analyses performed alongside clinial trials (1;2). Data may be obtained from medical records for hospitalization costs and medical outcomes, but patients' self-report is often needed to assess non-hospital care (e.g. informal care, out-of-pocket costs), quality of life, utility values, and treatment preferences. Questionnaires and diaries, administered by mail, telephone or face-to-face interview, or via the world wide web, may all be used to obtain patient specific data, each with its own (dis)advantages in terms of feasibility and validity (3).
Self-administered questionnaires and diaries may be used for the assessment of costs and quality of life (4), but are generally considered unsuitable for the assessment of preferences (time trade-off values or probability trade-offs). The elicitation of preferences involves sensitive measures in which the patient is asked to think about trade-offs between quality and length of life, or side-effects and benefits of treatments (5). Further, these are complex cognitive tasks, requiring personal interviews with well-trained interviewers to overcome problems related to the understanding and interpretation of those questions (6;7).
Personal interviews may be carried out by telephone or in face-to-face contacts. Face-to-face interviews may provide better opportunities for instruction and support, thereby resulting in more valid answers. In addition, face-to-face interviews may promote patients’ commitment to the study and thereby enhance their motivation to participate in following interviews or to perform other tasks. However, face-to-face interviews may also be logistically more difficult to organize, take more time, and thus are more costly than telephone interviews (8).
The comparability of face-to-face and telephone interviews for the assessment of health status has been investigated for several patient groups (9-14). Most studies concluded that there were no or only very small differences in response rates and obtained estimates of health status and quality of life. Evidence on the comparability of face-to-face and telephone interviews for the assessment of costs and preferences is scarce. Only one study compared preferences elicited by telephone or face-to-face interviews, and found no significant differences in estimated utilities in patients with arterial occlusive disease (15). To our knowledge, no study has yet addressed this issue in oncological patients, or used telephone interviews in oncological patients, possibly owing to the fact that preference questions may be more burdensome and sensitive for patients diagnosed with cancer, and that telephone interviews may provide less opportunities for emotional support.
Methods
Design and subjects
Data were obtained as part of a cost-utility analysis alongside a clinical trial (16,17). The main objective of this trial was to assess the additional value of preoperative radiotherapy (PRT) to Total Mesorectal Excision (TME) in patients with primary rectal cancer. Between January 1996 and January 2000, 1530 Dutch patients with resectable rectal cancer from 84 hospitals were randomized for TME-surgery with or without PRT. At the time of our analyses all patients had been followed up for at least 2 years.
From February 1999, 169 patients were asked to participate in face-to-face (at home baseline interviews just before treatment and in face-to-face (at home) or telephone follow-up interviews at 3 (T3) and 12 (T12) months after surgery (to assess short and long term effects of treatment). One hundred and twelve patients (66%) consented and were randomly assigned to receive face-to-face or telephone follow-up interviews. Most refusals were due to the patient's belief that the interviews would be too burdensome (67%). One patient explicitly mentioned the telephone interviews as the reason not to participate. All baseline interviews took place in a face-to-face situation, because we did not want to elicitate preferences by telephone in patients very recently diagnosed with rectal cancer. In case of a local or distant recurrence, telephone interviews were replaced by face-to-face interviews, to be better able to support those patients emotionally (18). Thus, 23 patients (21%) with a recurrence within 12 months after TME-surgery were excluded from analysis. Two patients (2%) withdrew from the study after the baseline interview, because they felt the interview was too burdensome and were excluded as well, leaving 87 patients for analysis.
Interviews
All interviews were conducted by the same trained interviewer (MvdB). The format and content of the face-to-face interview and a follow-up telephone interview were pre-tested in a pilot study among 10 recurrence-free patients who had been diagnosed with rectal cancer during the past 3 years, and were adapted accordingly. The pilot study showed that telephone follow-up interviews were feasible.
were reminded by telephone at 2 and 4 weeks after the interview. In the face-to-face group, all instruments were collected by the interviewer.
Instruments
The cost diary included structured no/yes questions on hospitalizations, outpatient visits, home help, district nursing, informal care, medications and care products, and has been described elsewhere (19).
Quality of life questionnaires consisted of the Rotterdam Symptom CheckList (RSCL), the EuroQoL (EQ-5D), a visual analogue scale (VAS), and the Health and Labour Questionnaire (HLQ),. The RSCL is a 38-item cancer specific questionnaire which includes scales for physical symptom distress (23 items), psychological symptom distress (7 items), and limitations in activities of daily living (8 items). All items are measured on 4-point Likert scales, ranging from 1 (not at all) to 4 (very much) (20). The summed scores of the items of the subscales were standardized to scale scores ranging from 0 (no problems at all) to 100 (most problems). The EQ-5D consists of 5 descriptive health status items (21), from which TTO-values assigned by the general public can be derived, which may range from –0.6 (worst possible health state) to 1 (perfect health) (22). The VAS is a 100 mm. line, anchored by 0 (death) to 100 (perfect health), measuring overall perceived health in the past week. The HLQ is a 20-item questionnaire measuring paid and unpaid labour (23).
Preference measures consisted of a time trade-off (TTO) and a probability trade-off method (PTO) (5). Flowcharts were used to administer the TTO and PTO (figures 1 and 2) (24). An introduction, which included definitions of current and perfect health for the TTO, and description of the treatments and their potential side-effects for the PTO, preceded each flowchart and were read by the interviewer. For the TTO, patients were asked to make an imaginary choice between a situation in current health for their remaining life time or a situation in perfect health for a shorter life time. For the PTO, patients were asked to indicate whether they would choose for a treatment with PRT, or not, at the provided hypothetical percentages of local recurrence risk. Both methods started with the two extremes and narrowed down in three (TTO) and two (PTO) steps in a pingpong fashion (figure 1). Thereafter, patients were asked to indicate their indifference point (if they had not reached their indifference point before).
The interview evaluation questionnaire contained 4 items on the perceived burden of the interview, measured on 4-point Likert scales ranging from 1 (not at all burdensome) to 4 (very burdensome), and, for patients in the telephone group, a question on the patient's preference for face-to-face or telephone interviews
Analyses
To assess feasibility, the response rates, missing values, interview duration, and interview evaluations were compared between randomization groups (face-to-face or telephone interviews). Measurement instruments not returned within 6 weeks (T3) and 3 months (T12) of the interview were considered missing. Chi-square tests were used to compare proportions, T-tests and Mann-Whitney tests to compare continuous variables.
6. You thought 8 years and 9 months in perfect health was better than 10 years in your current health.
7. You thought 7 years and 6 months in perfect health was worse than 10 years in your current health.
8. Could you indicate at what point you cannot make a choice between a shorter time in perfect health or 10 years in your current health?
9. I feel 8 years and 6 months in perfect health are equal to 10 years in my current health.
Figure 1. Example of a filled out TTO flow-chart
Figure 2. Example of a filled out PTO flow-chart
To assess convergent validity, estimated total costs from the diary, activities of paid and unpaid labour, quality of life scores, EQ-5D utilities, VAS scores, TTO values, and PTO scores were compared by repeated measures analyses with time as a within-subjects factor and randomization group (face-to-face or telephone interviews) as between-subjects factor. To study the sensitivity of our results to the assumptions of an intention-to-treat analysis, all analyses were also performed on a per-protocol basis.
Operation 85% no recurrence 15% recurrence Irradiation + operation 85% no recurrence 15% recurrence ❑ 1 ❑ ? Operation 85% no recurrence 15% recurrence ❑ Irradiation + operation 100% no recurrence 0% recurrence 2 ❑ ?
Could you please indicate below how large the chance of no recurrence should minimally be before you would choose for irradiation? 96% no recurrence, 4% recurrence 97% no recurrence, 3% recurrence 98% no recurrence, 2% recurrence 99% no recurrence, 1% recurrence 5b
Could you please indicate below how large the chance of no recurrence should minimally be before you would choose for irradiation? 91% no recurrence, 9% recurrence 92% no recurrence, 8% recurrence 93% no recurrence, 7% recurrence 94% no recurrence, 6% recurrence 5a
Results
Patients
The characteristics of the 87 patients are shown by randomization group in table 1. No significant differences were found concerning gender, age, preoperative radiotherapy, and hospital stay between randomization groups.
Table 1. Patient characteristics by randomization group. Telephone follow-up (n=42) Face-to-face follow-up (n=45) P-value Characteristic Males (%) 55 71 0.11
Mean age in years (sd) 63 (12) 64 (9) 0.81
Preoperative radiotherapy (%) 45 53 0.45
Median hospital stay for surgery (days) 15 16 0.42
Feasibility
At T3 and T12, 87 (100%) and 86 (99%) patients were interviewed. At T12, one patient in the telephone group was not interviewed, because she received adjuvant chemotherapy and felt too ill to participate in an interview. In the telephone group, the scheduled follow-up interviews were replaced by face-to-face interviews in 4 patients (9.5%), because of hearing problems (n=2), problems with understanding the questions (n=1), and a patient's strong preference for a face-to-face interview (n=1).
Table 2. Percentages of missing values by randomization group.
Time and randomization 3 months 12 months
Telephone follow-up (n=42) Face-to-face follow-up (n=45) p-value Telephone follow-up (n=42) Face-to-face follow-up (n=45) p-value Cost diary 17 16 0.89 17 4 0.06 Paid labour 7 11 0.52 12 0 0.02 Unpaid labour 10 7 0.62 10 0 0.03 Quality of life (RSCL) Physical scale Psychological scale Activity level scale
2 2 5 0 0 0 0.30 0.30 0.14 5 5 7 2 2 2 0.52 0.52 0.27 EuroQoL 2 4 0.60 2 2 0.30
Overall health (VAS) 12 0 0.02 5 0 0.14
Time trade-off 7 7 0.93 2 2 0.96
Probability trade-off 5 2 0.52 2 0 0.30
The mean duration of the telephone interviews was 98 (range 60-150) and 88 (range 45-150) minutes at T3 and T12 respectively, which was significantly shorter (both p<0.01) than the mean duration of the face-to-face interviews (T3: 110, range 70-150 and T12: 110, range 45-150).
The perceived burden of the interviews was not different between randomization groups (table 3). At T3 and T12, 43% and 42% of the patients in the telephone group indicated that they preferred face-to-face interviews, and respectively 2% and 16% preferred telephone interviews. The remaining patients had no marked preference. About 75% of the patients with a preference for face-to-face interviews did so because of the more personal contact with the interviewer, 15% said they were better able to understand the questions in a face-to-face situation, and about 10% said they thought the interview was too long to conduct by telephone. The reasons for preferring telephone interviews were more diverse, e.g. 'less intensive, so more easy', 'you can be yourself more fully', and 'saves the interviewer travel time and money'.
Convergent validity
assessed at T3 or T12. For several variables there was a significant overall effect of time, which shows, for example, that costs decrease and quality of life increases over time. Re-analysis using per-protocol analyses did not substantially change any of the results (data not shown).
Table 3. Average estimated costs, quality of life scores, utility values, and probability trade-offs (sd) by randomization group
Time and randomization 3 months 12 months P-values* Telephone
follow-up Face-to-face follow-up
Telephone
follow-up Face-to-face follow-up
Randomi-zation Randomi-zation by time
Time
Costs from the diary(€)† 1980 (1230) 1730 (970) 1020(1290) 1080 (840) 0.64 0.31 <0.001
Paid labour (%) 26 23 19 18 0.99 0.95 0.32
Hours of unpaid labour per week 19 (12) 18 (14) 20 (14) 21 (17) 0.94 0.43 0.10 Quality of life (RSCL)
Physical scale Psychological scale Activity level scale
14 (11) 11 (11) 10 (21) 12 (8) 12 (13) 7 (18) 13 (10) 13 (16) 7 (14) 11 (9) 12 (15) 2 (8) 0.37 0.99 0.18 0.51 0.53 0.66 0.33 0.62 0.02 EuroQoL 0.85 (0.16) 0.82 (0.21) 0.88 (0.17) 0.90 (0.14) 0.87 0.22 0.01 Overall perceived health (VAS) 75 (13) 76 (15) 83 (11) 80 (18) 0.71 0.41 0.003 Time trade-off 0.90 (0.13) 0.86 (0.21) 0.92 (0.15) 0.88 (0.19) 0.15 0.87 0.31 Probability trade-off‡ % never PRT % required benefit of PRT§ 8 3.1 (4) 9 2.0 (3) 7 2.5 (3) 13 2.6 (3) 0.99 0.35 0.84
Perceived interview burden 1.4 (0.5) 1.4 (0.5) 1.3 (0.5) 1.2 (0.4) 0.69 0.45 0.32 * Repeated measures analyses with time as within-subjects and randomization as between-subjects factor.
† All costs were standardized to a time period of 3 months to facilitate comparison over time. ‡ With rank ordered probability trade-off scores.
§ Based on the patients that did not refuse PRT at any benefit.
Discussion
The 66% response rate is comparable to the response rate found by Jansen et al. (25), who also performed utility interviews in patients that had recently been diagnosed with cancer. The reasons for non-participation were mostly related to the believed burden of the interviews in general. Only one patient mentioned the possibility of telephone interviews as the reason for non-participation. Therefore, we do not believe that this has biased our results. In the telephone group, somewhat more missing values occurred with respect to the cost diary and questionnaires that were to be self-completed. Patients in the telephone group may have been less motivated to fill out and return those instruments. In the face-to-face group the interviewer read through the diary and the questionnaires during the interview, which may have caused patients to fill out those questionnaires more meticulously. The difference in missing values did not translate into significantly different estimates of costs and of activities of paid and unpaid labour between randomization groups.
We compared telephone and face-to-face follow-up interviews for a wide variety of measures and found no significant differences between interview methods for any of the obtained estimates. The interviews were relatively long, both for the telephone and the face-to-face group. In the telephone group, most patients had a preference for face-face-to-face interviews. Still, all but one patient completed both follow-up interviews and the interviews were not evaluated as burdensome by the patients. This indicates that patients were very motivated to participate in our study. Our general impression was that, although the questions were sometimes perceived as difficult and confronting, patients also appreciated the conversation about their situation with an outside person. The personal contact with the interviewer during the baseline interview may also have contributed to the high compliance. In other settings and for other patient groups, results may differ. In patients without a life-threatening disease, the assessment of costs, quality of life, and preferences by telephone may be less burdensome, which would strengthen our conclusion that telephone interviews are a feasible alternative to face-to-face interviews. The findings of several other authors, who found no differences between telephone and face-to-face interviews in other patient groups support this hypothesis (9-14). On the other hand, in a less controlled setting patients may be less motivated to answer all questions, which could result in more missing values and might lead to less valid estimates.
In conclusion, telephone follow-up interviews are feasible in recurrence-free patients recently diagnosed with rectal cancer and yield similar estimates of costs, quality of life, utility values, and probability trade-offs. In economic evaluations of oncological treatments, telephone follow-up interviews may replace face-to-face follow-up interviews, thereby saving interviewer and travel time.
References
(1) Oostenbrink JB, Koopmanschap MA, Rutten FF. Standardisation of costs: the Dutch Manual for Costing in economic evaluations. Pharmacoeconomics 2002; 20(7):443-454.
(2) Schulman KA, Boyko WL, Jr. Evaluating cancer costs in NCI trials. Cancer Treat Res 1998; 97:37-52. (3) Biemer PP, Groves RM, Lyberg LE, Mathiowetz NA, Sudman S. Measurement errors in surveys. New
(4) Goossens ME, Rutten-van Molken MP, Vlaeyen JW, van der Linden SM. The cost diary: a method to measure direct and indirect costs in cost- effectiveness research. J Clin Epidemiol 2000; 53(7):688-695.
(5) Torrance GW. Utility approach to measuring health-related quality of life. J Chronic Dis 1987; 40(6):593-603.
(6) Jansen SJ, Stiggelbout AM, Nooij MA, Kievit J. The effect of individually assessed preference weights on the relationship between holistic utilities and nonpreference-based assessment. Qual Life Res 2000; 9(5):541-557.
(7) Revicki DA. Relationship between health utility and psychometric health status measures. Med Care 1992; 30(5 Suppl):MS274-MS282.
(8) van Campen C, Sixma H, Kerssens JJ, Peters L. Comparisons of the costs and quality of patient data collection by mail versus telephone versus in-person interviews. Eur J Publ Health 1998; 8(1):66-70. (9) Fenig S, Levav I, Kohn R, Yelin N. Telephone vs face-to-face interviewing in a community psychiatric
survey. Am J Public Health 1993; 83(6):896-898.
(10) Korner-Bitensky N, Wood-Dauphinee S, Siemiatycki J, Shapiro S, Becker R. Health-related information postdischarge: telephone versus face-to-face interviewing. Arch Phys Med Rehabil 1994; 75(12):1287-1296.
(11) McCormick MC, Workman-Daniels K, Brooks-Gunn J, Peckham GJ. When you're only a phone call away: a comparison of the information in telephone and face-to-face interviews [see comments]. J Dev Behav Pediatr 1993; 14(4):250-255.
(12) Rintala DH, Willems EP. Telephone versus face-to-face mode for collecting self-reports of sequences of behavior. Arch Phys Med Rehabil 1991; 72(7):477-481.
(13) Rohde P, Lewinsohn PM, Seeley JR. Comparability of telephone and face-to-face interviews in assessing axis I and II disorders. Am J Psychiatry 1997; 154(11):1593-1598.
(14) Wu AW, Jacobson DL, Berzon RA, Revicki DA, van der Horst C, Fichtenbaum CJ et al. The effect of mode of administration on medical outcomes study health ratings and EuroQol scores in AIDS. Qual Life Res 1997; 6(1):3-10.
(15) van Wijck EE, Bosch JL, Hunink MGM. Time-tradeoff values and standard-gamble utilities assessed during telephone interviews versus face-to-face interviews. Med Dec Making 1998; 18(4):400-405. (16) Kapiteijn E, Marijnen CA, Nagtegaal ID, Putter H, Steup WH, Wiggers T et al. Preoperative radiotherapy
combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001; 345(9):638-646.
(17) van den Brink M, van den Hout WB, Stiggelbout AM, Klein Kranenbarg E, Marijnen CAM, van de Velde CJH, Kievit J. Cost-utility analysis of preoperative radiotherapy in patients with rectal cancer
undergoing total mesorectal excision: A study of the Dutch Colorectal Cancer Group. Journal of Clinical Oncology 2004; 22(2):244-253.
(18) Stiggelbout AM, Kiebert GM, Kievit J, Leer JW, Habbema JD, de Haes JC. The "utility" of the Time Trade-Off method in cancer patients: feasibility and proportional Trade-Off. J Clin Epidemiol 1995; 48(10):1207-1214.
(19) van den Brink M, van den Hout WB, Stiggelbout AM, van de Velde CJH, Kievit J. Cost measurement in economic evaluations of health care: Whom to ask? Medical Care 2004; 42(8):740-746.
(20) de Haes JC, van Knippenberg FC, Neijt JP. Measuring psychological and physical distress in cancer patients: structure and application of the Rotterdam Symptom Checklist. Br J Cancer 1990; 62(6):1034-1038.
(21) EuroQol--a new facility for the measurement of health-related quality of life. The EuroQol Group. Health policy 1990; 16(3):199-208.
(22) Dolan P. Modeling valuations for EuroQol health states. Med Care 1997; 35(11):1095-1108. (23) van Roijen L, Essink Bot ML, Koopmanschap MA, Bonsel G, Rutten FF. Labor and health status in
economic evaluation of health care. The Health and Labor Questionnaire. Int J Technol Assess Health Care 1996; 12(3):405-415.
(24) Bosch JL, Hunink MG. The relationship between descriptive and valuational quality-of- life measures in patients with intermittent claudication. Med Decis Making 1996; 16(3):217-225.
