Arthritis & Rheumatism (Arthritis Care & Research) Vol. 57, No. 3, April 15, 2007, pp 454 – 460 DOI 10.1002/art.22613 © 2007, American College of Rheumatology

ORIGINAL ARTICLE

Daily Health Status Registration (Patient Diary) in Patients With Rheumatoid Arthritis: A Comparison Between Personal Digital Assistant and Paper-Pencil Format TURID HEIBERG,1 TORE K. KVIEN,2 ØYSTEIN DALE,3 PETTER MOWINCKEL,3 GERD J. AANERUD,4 ANN B. SONGE-MØLLER,5 TILL UHLIG,3 AND KÅRE B. HAGEN3 Objective. The patient perspective workshops at the Outcome Measures in Rheumatology Clinical Trials have included daily measures of health status (patient diary) and use of electronic tools for data collection in the research agenda. The objective of this study was to compare daily and weekly registrations of self-reported health status measures between personal digital assistant (PDA) and paper-pencil (PP) format regarding scores, variation, and feasibility. Methods. Thirty-eight patients with stable rheumatoid arthritis recorded their health status during 84 days in a repeated crossover design, using PDA or PP format during four 21-day periods. Visual analog scales (VAS) for pain, fatigue, and global disease and the Rheumatoid Arthritis Disease Activity Index were scored daily; the Short Form 36 and Modified Health Assessment Questionnaire were scored weekly. Results. The average scores and measures of variation of the 4 daily health status measures over 21 days did not differ significantly between PDA and PP formats in either of the 2 crossover periods. The values for the average range between the maximum and minimum values for daily measures were similar between the 2 formats, but showed considerable variation (e.g., range for pain VAS was 19 –28 mm over each 21-day period). The time to complete the instruments was similar between the 2 formats. Missing daily data entries were generally low for both periods and somewhat higher for PDA. The majority of patients (82.9%) preferred using PDA. Conclusion. Daily assessments with PDA may be efficiently used for frequent data collection because this format performs similarly to the traditional PP format. KEY WORDS. Patient perspective; Electronic diary; Rheumatoid arthritis; Health status measures; Outcome; Personal digital assistant.

INTRODUCTION “I have been keeping my diary for years, to learn about my disease relative to my life. The diary could have been Ms Heiberg’s work was supported by research grants from the Eastern Health Region Research Foundation. 1 Turid Heiberg, RN, MSN: Centre for Education and Skill Training, Ulleval University Hospital, Oslo, Norway; 2Tore K. Kvien, MD, PhD: Diakonhjemmet Hospital, Oslo, Norway; 3 Øystein Dale, MHS, Petter Mowinckel, MSc, Till Uhlig, MD, PhD, Kåre B. Hagen, PT, PhD: National Resource Center for Rehabilitation in Rheumatology, Diakonhjemmet Hospital, Oslo, Norway; 4Gerd J. Aanerud: Elgtråkket 77, Oslo, Norway; 5Ann B. Songe-Møller: Månev. 14, Sandvika, Norway. Address correspondence to Turid Heiberg, RN, MSN, Centre for Education and Skill Training, Division of Administration, Ulleval University Hospital, N-0407 Oslo, Norway. E-mail: [email protected]. Submitted for publication June 5, 2006; accepted in revised form August 15, 2006.

454

useful at consultations, but so far, it has not been of interest. Your data collection as researchers and clinicians is not sufficiently frequent to capture the daily and weekly variations in my symptoms.” This was a message from a patient to researchers at the 2002 meeting of Outcome Measures in Rheumatology Clinical Trials (OMERACT) (1). This sixth OMERACT meeting was the first to invite patients as independent delegates (2). Because rheumatoid arthritis (RA) is a disease with a fluctuating and progressive disease course (3,4), the patient delegates at the 2002 OMERACT and the subsequent OMERACT meeting in 2004 referred to their own practice of keeping diaries to document the variations in their condition and especially to look for possible associations with changes in lifestyle or environmental factors (5). Diaries that include frequent disease assessment by patient-reported health status measures may, however, be challenging, both for the patient reporting the data and for the data manager. Touch screens in clinics (6), the Internet

Use of Electronic Patient Diaries in Rheumatoid Arthritis (7), and hand-held computers known as personal digital assistants (PDAs) (8) represent available technological tools (9) that can be used for frequent disease assessments. Electronic diaries may be relevant for research (8,10) and for the clinical setting (11–13) and, last but not least, for the patients themselves (5). Potential advantages of electronic compared with paper-pencil data entry include the opportunity for capturing real-time data, reduction in the number of skipped responses, and avoidance of incorrect manual data entry (14,15). However, new tools for data collection need to undergo scientific evaluation of the applicability in their intended setting, meeting standard criteria of validity, reproducibility, and feasibility (16). The research agenda of the OMERACT patient perspective workshops in 2002 and 2004 included the development of standardized patient diaries and the use of information technology for repeated measurement (1,5). We wanted to address this research agenda and include patients as research partners (GJA, ABS-M) (1,17,18). We have previously shown that the clinimetric performances of self-reported health status measures in a PDA version and a paper-pencil (PP) version were similar with regard to test–retest reliability, agreement between single-entry scores, and feasibility (19). The goal of the present study was to examine whether PDA and PP formats are comparable in daily and weekly registrations of commonly used health status measures with regard to scores, variation, and feasibility.

PATIENTS AND METHODS Patients were recruited from the Oslo RA register (ORAR). This register was established in 1994 and includes patients with RA (20) who have a residential address in Oslo. ORAR has been shown to have a completeness of 85% (21,22). Inclusion criteria for this study were age 25–79 years, stable disease (i.e., no changes in medication or surgical procedures during the last 4 weeks), and ability to communicate in Norwegian. The predefined number of patients for this study was 40. The number of potential candidates for the study in ORAR was 1,348 patients. Of these, 300 were randomly selected and approached by mail to provide information about recent changes in surgery and drug therapy. Fifty-one of the 300 expressed interest in participation, 84 declined to take part, and 165 did not respond. Finally, 40 patients with stable medication consented to take part, but 2 of these later declined participation before the baseline visit. The 38 participating patients signed an informed consent form, and the study was evaluated by the regional ethics committee. We arranged 8 introductory meetings for small groups of patients. The project was presented, and the PDA devices were demonstrated and tested under supervision of the project coordinator and 2 research assistants. The participants also received a detailed user’s guide, and a telephone number for technical support was provided. Every third week, the patients received personal followup telephone calls. The patients performed daily assessments at home in a

455

Figure 1. The repeated crossover design. In period A (days 1– 42), patients completed the questions on either personal digital assistant (PDA) or paper-pencil (PP) during the first 21 days and then switched to the other format for 21 days. The procedures were repeated over period B (days 43– 84).

repeated crossover design (Figure 1). Each period of reporting was 21 days/3 weeks for each of the 2 formats. Half of the patients (group 1) started with the PDA (Dell Axim, model x5, Dell Norge, Lysaker, Norway) and switched to the PP format after 3 weeks (period A, 6 weeks). The other half (group 2) participated in the opposite order. The procedure was then repeated (period B, 6 weeks) (Figure 1). Patients were asked to perform the assessments in the morning. A variety of self-reported health status measures were completed daily or weekly. Electronic versions of these questionnaires were developed, and the respondents completed these versions by tapping on a touch sensitive screen with a plastic pen. The data were recorded on the PDA and were directly transferred electronically in an encrypted format via the mobile telephone network and over the Internet to a central server in the hospital. Incoming PDA data were checked weekly to trace possible failures. Data from the PP format were entered manually into a statistical software package (SPSS, version 11, Chicago, IL). Pain, fatigue, and global disease activity were scored daily on visual analog scales (VAS) during a total of 84 days. The anchoring points were no and extreme pain, fatigue, and global disease activity. Additionally, all items of the Rheumatoid Arthritis Disease Activity Index (RADAI) (23) were completed daily. The RADAI is a selfreported questionnaire including questions on disease activity, joint tenderness, pain, morning stiffness, and perceived joint pain in 16 joint areas. The scores from the 5 items are summarized into a disease activity index with a range from 0 to 10. Weekly assessments were performed with the Modified Health Assessment Questionnaire (MHAQ) (24) and Short Form 36 (SF-36) (25,26). The MHAQ is an 8-item questionnaire that measures ability to perform daily activities (scale range 1– 4, where 4 ⫽ worst health). The SF-36 is a 36-item generic measure of 8 health dimensions (physical functioning, physical role, bodily pain, general health, vitality, social functioning, emotional role, mental health) with scales from 0 to 100, where 0 ⫽ worst health. The time frames of the SF-36 were modified from last month to last week. Feasibility was assessed as the mean time to complete the daily health status measures, patients’ preference for format, and the mean number of missing daily entries during 21 days. Mean time to complete was the mean of the difference between the time recorded by the patients

456

Heiberg et al

Table 1. Comparison of participants versus nonparticipants* Participants Nonparticipants (n ⴝ 38) (n ⴝ 262) Age Disease duration Female sex, % Full-time employment, %‡

58.4 ⫾ 12.9 14.9 ⫾ 11.7 65.8 40.0

61.2 ⫾ 14.2 16.0 ⫾ 10.7 79.8 20.7

P† 0.26 0.57 0.05 0.06

* Values are the mean ⫾ SD unless otherwise indicated. † Two-sample t-test for continuous variables, chi-square test for categorical variables. ‡ Data available from 156 patients.

when starting (“We wish to know how long the registrations take you: what time is it now?”) and after completing the data entry (“We wish to know how long the registrations took you: what time is it now?”). The question addressing the preferred format had 3 response options (PDA, PP, or do not know). Systematic differences between the 2 formats were examined by computing mean scores and studying the variation over time. The average scores of the 21 daily and the 3 weekly assessments in period A and period B (Figure 1) were computed separately. To address the variations in reported health status, we computed the pooled SDs of the 21 daily and 3 weekly assessments and the ranges between the maximum and minimum reported values for each of the daily measures in the crossover periods for the 2 formats (Figure 1). Examinations for normality with Q–Q plots, distribution parameters, and histogram with displayed normal curve revealed that a high proportion of the computed scores had a skewed distribution. Comparisons

between the 2 different formats (PDA and PP) were therefore performed using Wilcoxon’s signed rank test, with presentation of median values of average scores, pooled SDs, and ranges. Wilcoxon’s signed rank test was also used to compare the number of missing entries over 21 days for the daily assessments and time to complete the questionnaires over 21 days. Reported preference for a particular format was examined with a binomial test and was compared across categorical variables by chi-square test. The level of statistical significance was P less than 0.05, and we chose not to correct for the number of statistical tests because we wanted an explorative and conservative approach in the examination of differences between the tools.

RESULTS Of a total of 300 patients who were asked to participate, 38 patients were included (mean ⫾ SD age 58.4 ⫾ 12.9, 65.8% women). Participants and nonparticipants (n ⫽ 262) were similar regarding disease duration and age, but the frequency of male sex and full-time employment was slightly higher among participants (Table 1). The average scores of the 4 daily health status measures did not differ between the PDA and PP formats in both groups combined, or in the first (period A) or second (period B) crossover period (Table 2). However, 3 comparisons were borderline significant, and all 3 tended to have higher scores with the PDA format than the PP format. A statistically significant difference was observed between the PDA and PP formats for 2 of the SF-36 scales (role physical in period A and social functioning in period B), but the differences were in opposite directions (Table 2). The average scores from the other weekly assessed

Table 2. Comparison of average scores from 21 daily and 3 weekly health status assessments between PDA and PP formats, in both groups combined, in 2 consecutive crossover periods (A and B), and differences between the PDA scores minus PP scores for each of the 2 periods* Health status measures Daily Pain VAS Fatigue VAS Global VAS RADAI Weekly MHAQ SF-36 Physical Role physical Pain General health Vitality Social functioning Role emotional Mental

Period A: day 1–42, week 1–6 PDA

PP

30.2 35.5 35.8 3.15

28.9 30.9 29.6 2.99

1.54

1.51

56.7 29.2 44.3 50.7 45.0 79.2 66.7 78.0

47.5 41.7 46.0 46.3 42.5 77.1 88.9 76.7

PDA ⴚ PP

Period B: day 43–84, week 7–12 P†

PDA

PP

PDA ⴚ PP

P†

3.4 3.5 4.5 0.17

0.11 0.15 0.06 0.06

29.4 31.4 32.4 3.00

30.4 32.6 29.4 2.75

0.0 2.3 2.6 0.03

0.47 0.11 0.06 0.55

0.00

0.90

1.63

1.50

0.00

0.38

0.0 0.0 0.0 0.0 0.0 4.2 0.0 0.0

0.84 0.66 0.64 0.77 0.79 0.01 0.91 0.82

⫺0.8 0.0 0.0 1.7 ⫺0.8 0.0 0.0 ⫺0.7

0.84 0.02 0.83 0.86 0.43 0.28 0.37 0.48

53.3 25.0 44.0 52.0 41.7 79.2 55.6 77.3

51.7 33.3 44.3 50.3 43.3 75.0 66.7 76.0

* Values are the median unless otherwise indicated. PDA ⫽ personal digital assistant; PP ⫽ pencil-paper; VAS ⫽ visual analog scale; RADAI ⫽ Rheumatoid Arthritis Disease Activity Index; MHAQ ⫽ Modified Health Assessment Questionnaire; SF-36 ⫽ Short Form 36. † Wilcoxon’s signed rank test.

Use of Electronic Patient Diaries in Rheumatoid Arthritis

457

Table 3. Comparison of the pooled standard variation from 21 daily and 3 weekly health status assessments between PDA and PP formats, in both groups combined, in 2 consecutive crossover periods, and differences between the pooled variation with PDA minus PP for each of the 2 periods* Period A: day 1–42, week 1–6

Health status measures Daily Pain VAS Fatigue VAS Global VAS RADAI Weekly MHAQ SF-36 Physical Role physical Pain General health Vitality Social functioning Role emotional Mental

Period B: day 43–84, week 7–12

PDA

PP

PDA ⴚ PP

P†

PDA

PP

PDA ⴚ PP

P†

7.7 8.4 7.5 0.47

7.48 7.34 7.08 0.60

⫺0.01 1.4 ⫺0.1 ⫺0.05

0.70 0.16 0.82 0.42

4.9 7.7 5.5 0.45

5.5 7.2 6.1 0.46

⫺0.5 0.3 ⫺0.3 0.00

0.34 0.66 0.41 0.83

0.10

0.13

0.00

0.70

0.07

0.07

0.00

0.98

3.21 0.00 5.20 5.13 5.77 7.22 0.00 4.00

0.0 0.0 0.3 0.0 2.4 0.0 0.0 0.0

0.98 0.37 0.14 0.64 0.14 0.15 0.003 0.67

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.75 0.51 0.70 0.77 0.68 0.63 0.21 0.54

5.0 7.2 5.8 7.6 7.6 7.2 19.3 4.6

5.0 14.4 5.8 5.0 5.0 7.2 19.3 4.0

3.5 14.4 5.5 5.1 5.0 7.2 0.0 4.6

* Values are the median unless otherwise indicated. See Table 2 for definitions. † Wilcoxon’s signed rank test.

SF-36 scales as well as for the MHAQ were similar (Table 2). The median values of the pooled individual variations were similar between the 2 formats, except for role mental of the SF-36 in the first crossover period (Table 3). Large daily variations were seen for all daily measures (pain, fatigue, global disease activity, and RADAI) but without any differences between PDA and PP (Table 4). For example, the median (25th, 75th percentile) values of the average range between the maximum and minimum values for reported pain during period A were 28 mm (19.5, 38.3) for PDA and 26.5 mm (17.5, 48.3) for PP. The time to complete the instruments was similar in 37 of the 42 days. For 4 of the days the difference was in the direction of spending more time to complete with the PP format (data not shown). The majority of the patients (82.9%) preferred using the PDA (P ⬍ 0.001). This preference was reported by 76.5% of those starting with the PDA and by 88.9% of those finishing with the PDA (P ⫽ 0.53). Preferences were similar across sex (P ⫽ 0.83). The median values for missing daily data entries were

generally low, 1 for both periods with PDA and 0 with PP for all 4 measures, both in periods A and B. However, missing daily entries were significantly higher for PDA than PP, and this difference was especially pronounced in period B. For weekly data entries, missing data were few and were similar between PDA and PP (data not shown).

DISCUSSION This report addresses 2 important issues from the OMERACT research agenda (1,5) on the patients’ perspective on outcome measures: patient diaries to capture daily and weekly variations in symptoms and, in particular, the use of electronic tools for data recording. However, new tools for capturing data should undergo rigid validation (16) and this study demonstrates that PDA and PP formats are comparable in daily and weekly registrations of commonly used health status measures. We applied a variety of well-known and frequently used health status measures ranging from simple VAS to more

Table 4. Comparison of the average ranges between the maximum and minimum reported scores over 21 daily health status assessments between PDA and PP formats, in both groups combined, in the 2 crossover periods, and differences between the ranges with PDA minus PP for each of the 2 periods* Health status measures

PDA

PP

PDA ⴚ PP

P†

PDA

PP

PDA ⴚ PP

P†

Pain VAS Fatigue VAS Global VAS RADAI

28.0 32.0 28.5 1.72

26.5 28.5 27.0 1.98

0.0 0.5 0.0 ⫺0.06

0.65 0.32 0.61 0.76

19.0 26.0 21.0 1.63

20.0 26.0 24.0 1.53

⫺3.0 ⫺1.0 0.0 0.03

0.26 0.99 0.58 0.89

Days 1–42, period A

Days 43–84, period B

* Values are the median unless otherwise indicated. See Table 2 for definitions. † Wilcoxon’s signed rank test.

458 complex composite instruments. This variety of instruments increases the generalizability of our findings, because the results were consistent across the different types of health status measures. A possible exception was that most of the VAS scores were numerically higher for PDA than PP (Table 2). A similar trend was also seen in a previously reported reliability study (19). We recruited patients from ORAR to have a group of patients that could represent a larger background population within a geographic area. Random selection of patients was chosen to obtain a representative sample. The participating patients also had health status profiles similar to the 1,342 patients from which they were selected (data not shown). Concerning introducing a new tool, we foresaw some concerns of systematic bias effects. For the patients, we were concerned about how they would manage the electronic devices, and for the project group, we saw some challenges on technical issues. To address these challenges, we organized introductory group meetings for the patients and provided technical support during the study. A log of support calls showed 64 inquiries, all of them concerning technical issues. Introductory training and support has also been a common feature in previous studies of electronic diaries (10,27–29). We performed this study with a crossover design instead of a parallel design. The crossover design does not require as many patients and allows the use of paired statistical analyses. Crossover designs are appropriate when performing studies on chronic, stable disease, which may be a challenge for studies in RA, with its fluctuating disease course. The comparison between PDA and PP could have been influenced by changes in disease activity from weeks 1–3 to weeks 4 – 6. However, because the design was repeated over two 6 week periods (Figure 1), we had the opportunity to compare PDA scores during period A and period B, and in the same way compare scores obtained with the PP format during the 2 periods. No significant differences between scores from the 2 periods were seen (data not shown), indicating that the disease was stable on the group level. The repeated crossover design gave us the opportunity to compare PDA and PP during 2 periods. The consistency in the findings regarding average scores (Table 2) and variation over time (Tables 3 and 4) supports the robustness of the results. The number of missing values increased slightly in the second period. Not many studies on electronic diaries have had a duration similar to or exceeding the 84 days of this study. The patients showed an impressive and persistent motivation to accomplish the assessments and recordings of data. The median number of missing data entries in the current study never exceeded 1 during the 21-day study period and the average number of missing observations never exceeded 15% of the expected data entries. Two common weaknesses in crossover studies are carryover effects and dropouts. In the present study, carryover effect was considered to possibly affect the preference of format. In accordance with this assumption, we found a slight numeric, but not significant preference at the last visit in favor of the last format in use. There were

Heiberg et al no dropouts during this study in the sense of withdrawal from participation, but one of the patients discontinued data entries, which were recorded as missing values. The technical problems that we faced, such as quick out loading of batteries and some data transfer problems, probably accounted for most of the missing values with PDA, and probably also for the increase in missing values with PDA in the last periods of the study. However, the significantly increased number of missing values with PDA compared with PP has also been previously described (27,30) and may have multiple explanations. Patients who were at the prestudy meeting indicated that retrospective completion of data on the PDA was impossible. PDA captures real-time data (8,31), whereas PP may be vulnerable to retrospective completion (27,30). Some studies have detected faked compliance (entries that were not real time) in up to 70 – 80% of cases (10,32). Retrospective completion may also have occurred with PP in the present study and may explain some of the higher amounts of missing data on PDA compared with PP in both crossover periods. Recall biases may occur with retrospective completion (33,34), but there is also evidence that recalled pain may be as valid as momentary data for many patients (35). The weekly assessments in our study did not show a significant number of missing entries, and it is possible that retrospective completion was done simultaneously with the weekly assessments, as also seen in previous research projects (10,14). In summary, it is difficult to establish systems that prevent fake data entry with PP, but electronic devices provide the opportunity for accurate and documented realtime data entry. Missing values may be interpreted as a measure of feasibility. Other considerations of feasibility are time to complete, patients’ preferences, costs, and resources. We found that the preference for the electronic format was independent of sex (data not shown). Previous studies have also shown that preference for electronic format is not associated with age, sex, educational level, or even computer literacy (11,36). The time to complete was similar in the PP format, and this finding is in accordance with previous results (37). The clear preference for the electronic version by the patients also echoes previous findings (29,36), even if there was a slight carryover effect for the last format in use. The components contributing to the total costs of PP and PDA differ fundamentally. PP formats are much cheaper than electronic devices, but wireless transferral directly into a server obviously is more efficient, accurate, and cost saving than manual data entry into a computer (14,15,29). In this project ⬃7 hours were allotted to data entry and administrative handling of the PP format in each patient. We estimated that this project would have to involve between 20 and 25 hours (salary $30/hour) of data entry and administration to balance the costs of purchasing a PDA ($500) and to provide technical support ($150) to 1 patient. However, PDAs are initial investments that can benefit several studies, whereas the costs for data entry, administrative handling, and technical support represent continuous expenses. Therefore, electronic formats may be more cost effective in the long run (37), especially when individual patients are entering large amounts of data. Because PDA provides good data

Use of Electronic Patient Diaries in Rheumatoid Arthritis

459

Table 5. Strengths and weaknesses of PDA and PP data collection formats* PDA Strengths Excellent compliance control Electronic reminder to respondent possible No manipulated data entries Accurate and good data quality Improved data management

PP Weaknesses

Initial costly implementation, development, and infrastructure Validation of psychometric properties required Training and support required Open-ended questions difficult Technical problems may cause loss of data

Strengths

Weaknesses

Established method

No compliance control

Familiar to respondents

Retrospective completion possible Time-consuming manual data entry Errors in data entries

Cheap to conduct Usually validated in terms of psychometric properties Not dependent on technology that may malfunction

Long time to data lock

Preferred by respondents * See Table 2 for definitions.

quality and effective data management and is respondents’ preferred format (Table 5), PDA can also be used for data collection in clinical trials, especially in situations with frequent measurements to detect and document onset of action of drugs. After termination of the project, the patient research partners performed structured group interviews. The responses from the patients can be understood as issues of feasibility that may be of interest for further studies. The introductory meeting and the user’s guide were unanimously appreciated by the patients. Patients also appreciated receiving followup telephone calls focusing on the daily performance with the tools. Employed patients reported that completion of the questionnaires was somewhat stressful before going to work. Retrospective completion of items in the questionnaires with PP was confirmed, as some patients stated “that it was easier to cheat on assessment time with PP than with PDA.” This study also confirms patients’ concerns about symptom variation over time. Large daily variations were observed within individuals, e.g., the median value for the range between the maximum and minimum value for reported pain was between 19 mm and 28 mm within a 21-day period (Table 4). Recognition of this variation over time may be important when planning intervention trials. It is relevant to consider whether a certain number of repeated pre-intervention measures could improve the precision between subject variation, and accordingly influence the number of measures and participants required in randomized trials (38). Electronic data capturing is being used increasingly in research and clinical practice, introducing the opportunity for frequent measurement and the entry of huge data materials (39). Further studies that may raise awareness of the pitfalls and concealed information about tools for data collection are as important as exploring the more obvious advantages. The OMERACT meeting requested that standard health status measures should undergo reliability, validity, and feasibility testing when used within an electronic format, either PDA, Internet, computer with touch screen, or other devices (1,5,16). Our study, as well as a series of other

studies in rheumatic diseases, have not found any major differences in clinometric properties when using electronic devices compared with PP (7,14,19,40,41). Similar findings have been reported in other areas of medicine (42). In conclusion, the overall research until now supports that widely used health status measures in rheumatology can be used in various electronic formats, but we still need to pay cautious attention to challenges that may emerge during new applications. Electronic data entry represents a future opportunity to obtain a view on disease courses and patterns not only in research and in clinical settings, but also for the patients who are truly involved in the course of their own health.

ACKNOWLEDGMENTS The authors thank the patients for participation in the data collection and for valuable feedback after the project. We also express our gratitude to Kjersti L. Damman and Tone Omreng for their coordination of the patient participation and support. AUTHOR CONTRIBUTIONS Ms Heiberg had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Heiberg, Kvien, Dale, Aanerud, Songe-Møller, Uhlig. Acquisition of data. Heiberg, Kvien, Dale, Aanerud, SongeMøller. Analysis and interpretation of data. Heiberg, Kvien, Dale, Mowinckel, Aanerud, Songe-Møller, Uhlig, Hagen. Manuscript preparation. Heiberg, Kvien, Dale, Mowinckel, Aanerud, Songe-Møller, Uhlig, Hagen. Statistical analysis. Heiberg, Kvien, Mowinckel, Hagen. Patients’ evaluation of the project. Heiberg, Dale, Aanerud, Songe-Møller. PhD supervision. Hagen.

REFERENCES 1. Kirwan J, Heiberg T, Hewlett S, Hughes R, Kvien T, Ahlmen M, et al. Outcomes from the Patient Perspective Workshop at OMERACT 6 [review]. J Rheumatol 2003;30:868 –72.

460 2. Saag KG. OMERACT 6 brings new perspectives to rheumatology measurement research. J Rheumatol 2003;30:639 – 41. 3. Wolfe F. A reappraisal of HAQ disability in rheumatoid arthritis. Arthritis Rheum 2000;43:2751– 61. 4. Kirwan JR. Links between radiological change, disability, and pathology in rheumatoid arthritis. J Rheumatol 2001;28: 881– 6. 5. Kirwan JR, Ahlmen M, de Wit M, Heiberg T, Hehir M, Hewlett S, et al. Progress since OMERACT 6 on including patient perspective in rheumatoid arthritis outcome assessment. J Rheumatol 2005;32:2246 –9. 6. Buxton J, White M, Osoba D. Patients’ experiences using a computerized program with a touch-sensitive video monitor for the assessment of health-related quality of life. Qual Life Res 1998;7:513–9. 7. Athale N, Sturley A, Skoczen S, Kavanaugh A, Lenert L. A web-compatible instrument for measuring self-reported disease activity in arthritis. J Rheumatol 2004;31:223– 8. 8. Koop A, Mosges R. The use of handheld computers in clinical trials. Control Clin Trials 2002;23:469 – 80. 9. Wolters FL, van Zeijl G, Sijbrandij J, Wessels F, O’Morain C, Limonard C, et al. Internet-based data inclusion in a population-based European collaborative follow-up study of inflammatory bowel disease patients: description of methods used and analysis of factors influencing response rates. World J Gastroenterol 2005;11:7152– 8. 10. Stone AA, Shiffman S, Schwartz JE, Broderick JE, Hufford MR. Patient compliance with paper and electronic diaries. Control Clin Trials 2003;24:182–99. 11. Carlson LE, Speca M, Hagen N, Taenzer P. Computerized quality-of-life screening in a cancer pain clinic. J Palliat Care 2001;17:46 –52. 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;86:51–9. 13. Cull A, Gould A, House A, Smith A, Strong V, Velikova G, et al. Validating automated screening for psychological distress by means of computer touchscreens for use in routine oncology practice. Br J Cancer 2001;85:1842–9. 14. Wilson AS, Kitas GD, Carruthers DM, Reay C, Skan J, Harris S, et al. Computerized information-gathering in specialist rheumatology clinics: an initial evaluation of an electronic version of the Short Form 36. Rheumatology (Oxford) 2002;41:268 – 73. 15. Ryan JM, Corry JR, Attewell R, Smithson MJ. A comparison of an electronic version of the SF-36 General Health Questionnaire to the standard paper version. Qual Life Res 2002;11: 19 –26. 16. Boers M, Brooks P, Strand CV, Tugwell P. The OMERACT filter for Outcome Measures in Rheumatology [editorial]. J Rheumatol 1998;25:198 –9. 17. Shea B, Santesso N, Qualman A, Heiberg T, Leong A, Judd M, et al. Consumer-driven health care: building partnerships in research. Health Expect 2005;8:352–9. 18. Telford R, Boote JD, Cooper CL. What does it mean to involve consumers successfully in NHS research? A consensus study. Health Expect 2004;7:209 –20. 19. Kvien TK, Mowinckel P, Heiberg T, Dammann KL, Dale O, Aanerud GJ, et al. Performance of health status measures with a pen based personal digital assistant. Ann Rheum Dis 2005; 64:1480 – 4. 20. Arnett FC. Revised criteria for the classification of rheumatoid arthritis. Orthop Nurs 1990;9:58 – 64. 21. Kvien TK, Glennas A, Knudsrod OG, Smedstad LM, Mowinckel P, Forre O. The prevalence and severity of rheumatoid arthritis in Oslo: results from a county register and a population survey. Scand J Rheumatol 1997;26:412– 8. 22. Kvien TK, Uhlig T. The population based studies in rheumatoid arthritis: a method of longterm followup studies. J Rheumatol Suppl 2004;69:35– 40. 23. Fransen J, Langenegger T, Michel BA, Stucki G. Feasibility and validity of the RADAI, a self-administered rheumatoid

Heiberg et al

24.

25. 26.

27.

28.

29.

30.

31. 32. 33. 34. 35.

36. 37. 38.

39.

40.

41.

42.

arthritis disease activity index. Rheumatology (Oxford) 2000; 39:321–7. Pincus T, Summey JA, Soraci SA Jr, Wallston KA, Hummon NP. Assessment of patient satisfaction in activities of daily living using a modified Stanford Health Assessment Questionnaire. Arthritis Rheum 1983;26:1346 –53. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473– 83. Kvien TK, Kaasa S, Smedstad LM. Performance of the Norwegian SF-36 Health Survey in patients with rheumatoid arthritis. II. A comparison of the SF-36 with disease-specific measures. J Clin Epidemiol 1998;51:1077– 86. Gaertner J, Elsner F, Pollmann-Dahmen K, Radbruch L, Sabatowski R. Electronic pain diary: a randomized crossover study [published erratum appears in J Pain Symptom Manage 2004; 28:626]. J Pain Symptom Manage 2004;28:259 – 67. Nyholm D, Kowalski J, Aquilonius SM. Wireless real-time electronic data capture for self-assessment of motor function and quality of life in Parkinson’s disease. Mov Disord 2004; 19:446 –51. Palermo TM, Valenzuela D, Stork PP. A randomized trial of electronic versus paper pain diaries in children: impact on compliance, accuracy, and acceptability. Pain 2004;107: 213–9. Lauritsen K, Degl’ Innocenti A, Hendel L, Praest J, Lytje MF, Clemmensen-Rotne K, et al. Symptom recording in a randomised clinical trial: paper diaries vs. electronic or telephone data capture. Control Clin Trials 2004;25:585–97. Quinn P, Goka J, Richardson H. Assessment of an electronic daily diary in patients with overactive bladder. BJU Int 2003; 91:647–52. Broderick JE, Schwartz JE, Shiffman S, Hufford MR, Stone AA. Signaling does not adequately improve diary compliance. Ann Behav Med 2003;26:139 – 48. Eich E, Reeves JL, Jaeger B, Graff-Radford SB. Memory for pain: relation between past and present pain intensity. Pain 1985;23:375– 80. Eich E, Rachman S, Lopatka C. Affect, pain, and autobiographical memory. J Abnorm Psychol 1990;99:174 – 8. Jamison RN, Raymond SA, Slawsby EA, McHugo GJ, Baird JC. Pain assessment in patients with low back pain: comparison of weekly recall and momentary electronic data. J Pain 2006; 7:192–9. Bliven BD, Kaufman SE, Spertus JA. Electronic collection of health-related quality of life data: validity, time benefits, and patient preference. Qual Life Res 2001;10:15–22. Chan SS, Chu CP, Cheng BC, Chen PP. Data management using the personal digital assistant in an acute pain service. Anaesth Intensive Care 2004;32:81– 6. Mowinckel P, Hagen KB, Heiberg T, Kvien TK. Repeated measures in rheumatoid arthritis (RA) can effectively reduce the between subject variation and thereby the sample size in clinical trials [abstract]. Ann Rheum Dis 2006;65 Suppl 2:97. Greenwood MC, Hakim AJ, Carson E, Doyle D. Touch screen computers 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 (Oxford) 2006;45:66 –71. E-pub 2005 Nov 1. Williams CA, Templin T, Mosley-Williams AD. Usability of a computer-assisted interview system for the unaided self-entry of patient data in an urban rheumatology clinic. J Am Med Inform Assoc 2004;11:249 –59. Mosley-Williams A, Williams CA. Validation of a computer version of the American College of Rheumatology Patient Assessment questionnaire for the autonomous self-entry of self-report data in an urban rheumatology clinic. Arthritis Rheum 2004;50:332–3. Dale O, Hagen KB. Despite technical problems personal digital assistants outperform pen and paper when collecting patient diary data. J Clin Epidemiol 2007;60:8 –17. E-pub 2006 Aug 30.