Chapter 3.Material and Methods
4.6 Differential item functioning analysis
Figure 4 to Figure 8 showed the results of differential item functioning of Rasch analysis. Our study found there was no differential item functioning of the EORTC QLQ-PR25 for prostate cancer patients, which indicated the measurement property were equivalent between two modes for each item.
(See the Supplementary Appendix for additional details in Table A3~ Table A5.)
Chapter 5.Discussion
This Study demonstrates that the touch-screen version of questionnaires was shown to be feasible in prostate cancer patients, and it appeared to be preferable to use than the paper version of the same questionnaires. In addition, the equivalence of the paper version and the touch-screen version of the EORTC QLQ-C30 and the EORTC QLQ-PR25 is shown, in terms of no mode effect in domain level by using cross-over regression analysis, high exact and global agreement in both item and domain level, and no DIF by using the modern test measurement analysis.
5.1 Feasibility assessment of two modes
Many studies examined the measurement equivalence of paper-based version and touch-screen computer-based version, and showed the touch-screen version was well accepted for most subjects1, 19, 27, 142
. In this study, most people preferred touch-screen version questionnaire than paper version, the result was consistent with previous studies (range 39%~57%).
About 92% of patients indicated that they liked using the touch-screen to complete the questionnaire; about 97% of patients thought the touch-screen interface was user-friendly; and about 67% patients reported that they prefer the touch-screen version to paper version. Moreover, most patients (92%) in our study reported that the touch-screen was easy to use. Similarly, Pouwer (1998) noted that a touch-screen questionnaire was easy (easier) for patients to complete even if they have rarely or never used a computer6.
As to suggestions to the touch-screen version questionnaire, 10%~24%
patients suggested improving some drawbacks on touch-screen version questionnaire, for example: font size and layout of the touch-screen.
5.1.1
Feasibility for the elderly
The average age of prostate cancer patients (eighty years old in this study)
is higher than the other diseases 3, 7, 9, 19, 21, 137
such as gastroesophageal reflux disease, asthma, rheumatoid arthritis, cancer, head and neck cancer 3, 7, 9, 19, 21, 137, in those diseases the average age ranged from 52.1~59.3 years. Even though, our results showed the feasibility assessment findings were fine.
There was just one item question on a screen in the touch-screen version questionnaires, for the elderly patients, can be clear about what they saw.The younger patients(less than 70 years old) were more feasible in touch-screen questionnaire than the older patients (greater than 70 years old). Furthermore, younger patients(less than 70 years old) spent less time than older patients (greater than 70 years old) to complete the questionnaires in both versions.
5.1.2
Feasibility for inexperienced computer user
Greenwood (2006) investigated the feasibility of collecting rheumatoid arthritis (RA) patient self-administered outcome data using touch-screen computers in a routine outpatient clinic. Forty patients with RA completed the touch-screen and paper Rheumatoid Arthritis Quality of Life Questionnaire (RAQol) in the clinic and rated ease of use and preference. The touch-screen RAQol took less time to complete, was preferred by 64% (33% had no preference) and also was significantly higher for ease of use (p=0.003, n=40) even by inexperience computer-using patients (p=0.031, n=24). In our study, inexperienced computer-using patients (n=80) showed comparable feasibility of touch-screen version in terms of acceptance, preference, suggestions with experienced computer-using patients (n=19).
5.2 Time management
On the average, touch-screen took more time than paper version (18.1 min vs. 16.3 min, p value=0.0018). This resulted from several reasons: first, the respondents have to acquaint with computer using; second, the respondents in both groups spent fewer time on the followed questionnaires than first one. According to our results, touch-screen would take more time.
However, it would be expected that it will take less time in the future, because when patients answer routine assessment of HRQL, they will be more familiar with questionnaires. Besides, the touch-screen version allows data to be automatically entered into the database of a computer server and to immediately calculate the scores, thereby saving the time of manual entry, scoring, and analyzing the data31, 158.
Comparing the time to completion between paper/touch group and touch/paper group, it took almost the same length of time to complete the paper and the touch-screen versions for paper version first followed by touch-screen version group. Interestingly, for touch-screen version first followed by paper version group, it took longer time to complete the touch-screen version. In both groups the first questionnaire took longer to complete, which is consistent with the pattern reported by others1, 19, 27, 142
. All patients were given a two-hour break between the two questionnaires to avoid washout effect. However, we can recognize that time on first questionnaire indeed takes more time than followed questionnaire and this result is same as previous studies. This finding suggests the time management will become more and more efficient in the follow-up assessment.
In addition, the touch-screen version of the questionnaire was able to guide patients to skip some non-relevant items which is not needed to response based on their previous response, in such way, the patients can save the response time by eliminating the need to “click” through all non-relevant items and the response error can also be reduced.
5.3 Data management
The use of the touch-screen questionnaire may reduce the missing data, because in this way, respondents were guided through the screen driving and were unable to skip any item which is relevant to answer. The touch-screen version can also eliminate the invalid data by permitting patients to select
only one of the on-screen response options. However, in the paper version, the respondent could further proceed even some items were not completed;
and some out of range or ambiguous data could be answered.
5.4 Equivalence assessment of two modes
To access the health-related quality of life by using touch screen mode has been shown to be feasible; the crossover randomized design for this comparison of both modes of paper-and-pencil and computerized version was commonly used in various diseases3, 7, 9, 19, 21, 137
, including gastroesophageal reflux disease, asthma, rheumatoid arthritis, cancer, head and neck cancer.
Many studies examined the measurement equivalence of paper-based version and touch-screen computer-based version, and showed the touch-screen version was well accepted for most subjects1, 19, 27, 142
. Our finding showed that all domains in the EORTC QLQ-C30 and the EORTC QLQ-PR25 were equivalence in prostate cancer patients. This finding can be an empirical evidence to understand the touch-screen mode can be another valuable option to assess the patient’s report quality of life.
Using crossover regression model analysis, overall, the mode effect was all no statistically significant, which supported the equivalence of measure properties. Global agreement in all domains reaches greater than 96% in the EORTC QLQ-C30 and the EORTC QLQ-PR25. Differential item functioning (DIF) analysis based on the modern test theory also supported the equivalent properties between two modes.
5.5 Advantages of the crossover design
There were advantages to crossover design. The reason to consider a crossover design when planning a clinical trial (or methodology) is that it could yield a more efficient comparison of treatments than a parallel design, i.e., fewer patients might be required in the crossover design in order to attain the same level of statistical power, precision, etc. Intuitively, this seems
reasonable because each patient serves as his/her own matched control. Every patient receives both treatments (methods) A and B. Crossover designs are popular in medicine, agriculture, manufacturing, education, and many other disciplines and a comparison is made of their response on A vs. B. Our results from crossover regression analysis showed there was no mode-order interaction effect for most domains, which implicated the carry-over effect did not exist; and when we refit the main effect removing the interaction term, the results showed the order effect did not exist for most domain. The above results supported the crossover randomized design in our study is rigorous.
5.6 Confirmation from modern measurement theory
Rasch model analysis is based on the modern measurement theory, originally developed in the fields of education and psychology, has been proven to be a powerful tool for patients reported outcome assessment159-160. This model comprises a set of statistical models suitable for analyzing a scale or survey instrument with multiple items that measure the same construct (e.g., physical functioning). Rasch model specifies how both person–trait level and item characteristics are related to a person’s item responses. This is different from the classical test theory (CTT) approach in which items and the person latent trait being measured are considered separately and, therefore, cannot be meaningfully and systematically compared 161-162. Many limitations of CTT approach can be solved rationally using modern measurement theory approach. Many useful statistics, such as differential item functioning (DIF) can be examined for measurement invariance161-162. Our analyses of the DIF revealed that four domains in the EORTC QLQ-PR25 to assess for prostate cancer patients exhibited no DIF across the two method groups (touch-screen vs. paper) displayed.
5.7 Improvement of quality of care
First, as mentioned in the literature review, paper questionnaire would
require more manpower to collect questionnaires and key-in data6, 136, so that touch-screen mode could save more time, manpower. For example, in our proceeding, the patients’ responses to the EORTC QLQ-C30 and the EORTC QLQ-PR25 were automatically entered into a desktop computer, scored, and printed as a graphic summary profile (see Figure 9). Although our results showed the touch-screen mode took more time to finish (see Table2), however, it can be expected that when the routine assessment of HRQL is required, the assessment time will become more shortly afterward.Before the start of the consulting in the visiting room, each patient completes a touch-screen version HRQL questionnaire in the waiting room, and then physicians receives the patients reported outcomes later immediately; in such way, the quality of care will be upgraded.
Figure 9. Example of graphic summary profile of quality of life questionnaire EORTC QLQ-C30
Second, the proceeding of data collection through touch-screen mode can help the integration of patient’s reported outcome and clinical information to promote the quality of health care. Paper-and-pencil would raise the manpower required to administer, collect, enter data and score an HRQL questionnaire6, 136. We postulate (believe) that the establishment of the touch-screen version would be useful to the integration of clinical informatics.
Third, several large studies in chronic diseases also suggested that incorporating standardized HRQL assessments in routine clinical oncology practice facilitates the discussion of the progression of HRQL issues and can heighten physicians’ awareness of their patients’ HRQL3, 10-13, 139-141
. Copies of the summary were given to the patient and physician immediately before
consultation. A copy was also placed in the medical records. At the each subsequent outpatient visits, a summarized report from patient’s report questionnaires included both the patients’ current scores and those elicited at the previous visit(s) can be displayed in the physician screen in real time, which can be an useful information to facilitate the communication between physicians and patients.
Finally, oncology settings system assessed the manner in which clinicians use this touch-screen questionnaire and identify the benefits and challenges that oncology clinics may face when adopting9, 137. For example, one research reported challenges included patient burden from the frequent need to answer the questionnaires, the development of short version of questionnaire could be one solution to solve the challenge. In addition, the setting of the overall computerized environment such as the integrated system of clinical informatics and the setting of computerized hardware plays an important key role in the performance and contribution of the data collection though the touch screen mode. In summary, touch-screen questionnaire assessments can be linked to the integration of routine assessment of patients’ symptoms and health-related quality of life into the daily flow of an oncology clinic, it offers advantages in terms of promote health care quality.
Chapter 6.Conclusion 6.1 Conclusion
The touch-screen mode had good feasibility, and was accepted for most prostate cancer patients, 92% patients showed the touch-screen version was easy to use. High percentages of patients thought they preferred touch-screen version to the paper-and-pencil version, which were 74% for the patients below 70 years old and 59.2% for the patients aged greater than 70 years. The younger patients spent less time than older patients to complete the questionnaires in both versions. As to suggestions to the touch-screen version questionnaire, only 10%~24% patients suggested improving some drawbacks on touch-screen version questionnaire, for example: font size and layout of the touch-screen.
The measurement properties of the EORTC QLQ-C30 and the EORTC QLQ-PR25 data by using the touch-screen version were shown to be equivalent to the paper-and-pencil version. The measurement effect between the touch-screen mode and the paper-and-pencil mode were no significant difference from the crossover regression model analysis. The percentages of global agreement in all domains reached greater than 96% in both the EORTC QLQ-C30 and the EORTC QLQ-PR25. Most ICC indices greater than 0.7 in both questionnaires indicated good equivalence. Differential item functioning (DIF) analysis based on the modern test theory also supported the equivalent properties between two modes.
Our study result provided an empirical evidence to support the touch-screen mode of the QLQ-C30 and the EORTC QLQ-PR25 for patients with prostate cancer can be an alternative choice of measurement mode in addition to paper-and-pencil mode to assess the patient’s report quality of life.
The e-data from the touch screen questionnaire can be easily integrated with other clinical data to provide real time diagnostic information in clinic. It may
not only improve medical care quality, but also promote the relationship between physician and patient.
6.2 Limitation
There are some limitations in the present study. First, since we excluded patients who could not read, speak and write Chinese, and who could not complete these questionnaires by themselves for the whole procedure, the results cannot be generalized to these patients. Second, the study was conducted in a single disease and a single hospital so the representative of all patients with prostate cancer in Taiwan may not be enough. In addition, the sample subjects were from the outpatient clinic, thus the results may not suit to the inpatients.
References
1. Matthew AG, Currie KL, Irvine J, Ritvo P, Santa Mina D, Jamnicky L, Nam R, Trachtenberg J. Serial personal digital assistant data capture of health-related quality of life: a randomized controlled trial in a prostate cancer clinic. Health Qual Life Outcomes, 2007; 5:38.
2. Sanda MG, Dunn RL, Michalski J, Sandler HM, Northouse L, Hembroff L, Lin X, Greenfield TK, Litwin MS, Saigal CS. Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med, 2008; 358(12):1250-1261.
3. Detmar SB, Muller MJ, Schornagel JH, Wever LD, Aaronson NK.
Health-related quality-of-life assessments and patient-physician communication: a randomized controlled trial. JAMA, 2002;
288(23):3027-3034.
4. Hahn EA, Bode RK, Du H, Cella D. Evaluating linguistic equivalence of patient-reported outcomes in a cancer clinical trial. Clin Trials, 2006;
3(3):280-290.
5. Wright WG, Jones JA, Spiro A, 3rd, Rich SE, Kressin NR. Use of patient self-report oral health outcome measures in assessment of dental treatment outcomes. J Public Health Dent, 2009; 69(2):95-103.
6. Pouwer F, Snoek FJ, van der Ploeg HM, Heine RJ, Brand AN. A comparison of the standard and the computerized versions of the Well-being Questionnaire (WBQ) and the Diabetes Treatment Satisfaction Questionnaire (DTSQ). Qual Life Res, 1998; 7(1):33-38.
7. Bushnell DM, Martin ML, Parasuraman B. Electronic versus paper questionnaires: a further comparison in persons with asthma. J Asthma, 2003; 40(7):751-762.
8. Mullen KH, Berry DL, Zierler BK. Computerized symptom and quality-of-life assessment for patients with cancer part II: acceptability and usability. Oncol Nurs Forum, 2004; 31(5):E84-89.
9. Young NL, Varni JW, Snider L, McCormick A, Sawatzky B, Scott M, King G, Hetherington R, Sears E, Nicholas D. The Internet is valid and reliable for child-report: an example using the Activities Scale for Kids (ASK) and the Pediatric Quality of Life Inventory (PedsQL). J Clin Epidemiol, 2009; 62(3):314-320.
10. Kazis LE, Callahan LF, Meenan RF, Pincus T. Health status reports in the care of patients with rheumatoid arthritis. J Clin Epidemiol, 1990;
43(11):1243-1253.
11. Wagner AK, Ehrenberg BL, Tran TA, Bungay KM, Cynn DJ, Rogers
WH. Patient-based health status measurement in clinical practice: a study of its impact on epilepsy patients' care. Qual Life Res, 1997;
6(4):329-341.
12. Velikova G, Brown JM, Smith AB, Selby PJ. Computer-based quality of life questionnaires may contribute to doctor-patient interactions in oncology. Br J Cancer, 2002; 86(1):51-59.
13. Lubeck DP. Health-related quality of life measurements and studies in rheumatoid arthritis. Am J Manag Care, 2002; 8(9):811-820.
14. Henderson A, Andreyev HJ, Stephens R, Dearnaley D. Patient and physician reporting of symptoms and health-related quality of life in trials of treatment for early prostate cancer: considerations for future studies. Clin Oncol 2006; 18(10):735-743.
15. Mills ME, Murray LJ, Johnston BT, Donnelly M. Feasibility of a standardised quality of life questionnaire in a weekly diary format for inoperable lung cancer patients. Eur J Oncol Nurs, 2008;
12(5):457-463.
16. Abernethy AP, Herndon JE, 2nd, Wheeler JL, Day JM, Hood L, Patwardhan M, Shaw H, Lyerly HK. Feasibility and acceptability to patients of a longitudinal system for evaluating cancer-related symptoms and quality of life: pilot study of an e/Tablet data-collection system in academic oncology. J Pain Symptom Manage, 2009;
37(6):1027-1038.
17. Albertsen PC, Nease RF, Jr., Potosky AL. Assessment of patient preferences among men with prostate cancer. J Urol, 1998;
159(1):158-163.
18. Anie KA, Jones PW, Hilton SR, Anderson HR. A computer-assisted telephone interview technique for assessment of asthma morbidity and drug use in adult asthma. J Clin Epidemiol, 1996; 49(6):653-656.
19. de Bree R, Verdonck-de Leeuw IM, Keizer AL, Houffelaar A, Leemans CR. Touch screen computer-assisted health-related quality of life and distress data collection in head and neck cancer patients. Clin Otolaryngol, 2008; 33(2):138-142.
20. Goodhart IM, Ibbotson V, Doane A, Roberts B, Campbell MJ, Ross RJ.
Hypopituitary patients prefer a touch-screen to paper quality of life questionnaire. Growth Horm IGF Res, 2005; 15(6):384-387.
21. Kleinman L, Leidy NK, Crawley J, Bonomi A, Schoenfeld P. A comparative trial of paper-and-pencil versus computer administration of the Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire. Med Care, 2001; 39(2):181-189.
22. Millsopp L, Frackleton S, Lowe D, Rogers SN. A feasibility study of computer-assisted health-related quality of life data collection in patients with oral and oropharyngeal cancer. Int J Oral Maxillofac Surg, 2006; 35(8):761-764.
23. Wolpin S, Berry D, Austin-Seymour M, Bush N, Fann JR, Halpenny B, Lober WB, McCorkle R. Acceptability of an Electronic Self-Report Assessment Program for patients with cancer. Comput Inform Nurs, 2008; 26(6):332-338.
24. Wright EP, Selby PJ, Crawford M, Gillibrand A, Johnston C, Perren TJ, Rush R, Smith A, Velikova G, Watson K. Feasibility and compliance of automated measurement of quality of life in oncology practice. J Clin Oncol, 2003; 21(2):374-382.
25. Gordon MM, Capell HA, Madhok R. The use of the Internet as a resource for health information among patients attending a rheumatology clinic. Rheumatology (Oxford), 2002; 41(12):1402-1405.
26. Greenwood MC, Hakim AJ, Carson E, Doyle DV. Touch-screen computer systems in the rheumatology clinic offer a reliable and user-friendly means of collecting quality-of-life and outcome data from patients with rheumatoid arthritis. Rheumatology 2006; 45(1):66-71.
27. Salaffi F, Gasparini S, Grassi W. The use of computer touch-screen technology for the collection of patient-reported outcome data in rheumatoid arthritis: comparison with standardized paper questionnaires. Clin Exp Rheumatol, 2009; 27(3):459-468.
28. Verhoef J, Oosterveld FG, Hoekman R, Munneke M, Boonman DC, Bakker M, Otten W, Rasker JJ, de Vries-Vander Zwan HM, Vliet Vlieland TP. A system of networks and continuing education for physical therapists in rheumatology: a feasibility study. Int J Integr
28. Verhoef J, Oosterveld FG, Hoekman R, Munneke M, Boonman DC, Bakker M, Otten W, Rasker JJ, de Vries-Vander Zwan HM, Vliet Vlieland TP. A system of networks and continuing education for physical therapists in rheumatology: a feasibility study. Int J Integr