Background/Purpose: Registries have become a common tool for collecting patient-centered outcome measures. Clinical effectiveness research may be improved if data from multiple registries could be combined allowing for richer data sets in that patients in one registry could supplement data which has been collected in a separate registry. However, the question remains as to whether data that is collected in two separate registries are equivalent. While the registries may ask the same clinical question, if they collect it in unique formats, such as online at home versus in the clinical setting, it is not clear if the metrics are equal. We attempted to address this question of data equivalence through evaluating a patient entered registry versus a registry collected at the point of clinical care.

Methods: Patients were participants in a rheumatology clinic research database that collected patient data via paper questionnaires at clinic visits. They were also participants in a longitudinal, observational study that collected patient data at 6-month intervals from their home via paper, web, or telephone-interview questionnaires at patient’s preference from 2007-2012. Four patient measures were congruent between the studies: HAQ-II and pain, patient global assessment, and fatigue visual analog scales. General estimating equations (GEE) assessed longitudinal effects between clinical and at-home data collection methods adjusting for sociodemographic status and number of clinic visits.

Results: A total of 1439 patients enrolled in both studies (40% rheumatoid arthritis, 22% osteoarthritis, and 38% other rheumatic diseases). Mean (SD) age at enrollment was 56.5 (14.0) years, 20.5% were male, 92.0% were Caucasian, and mean (SD) education was 14.0 (2.2) years. Primary rheumatic disease duration was 8.2 (9.3) years and rheumatic disease comorbidity index was 2.0 (1.6). Baseline measures were: HAQ-II 0.87 (0.65), pain 4.0 (2.8), global 3.8 (2.5), and fatigue 4.6 (3.1). Results of the GEE are shown in Table 1; clinic paper questionnaire as well as non-clinic formats of web and telephone were compared to non-clinic paper questionnaires.

Conclusion: On average patients report a higher level of disease severity during a clinic visit than when they report from home while those who respond from home via the web do not differ from paper respondents. As expected, patients that choose telephone interviews reported worse outcomes except for global assessment. Patients respond at home when they are willing and available which may delay responses during spans of poorer health. Patients seen in the clinic, notably those seen more often, could have more severe health care needs from increased disease severity. Further study is needed to determine if the accuracy of such assessments differ.

Table 1. Relative effect of location and questionnaire media on outcomes through general estimating equations

Outcome Measure	Clinic	Non-clinic (e.g., at home)
	Paper	Paper (ref)	Web	Telephone
HAQ-II (0-3)	0.07 (0.06, 0.08)	0.0	0.02 (0.00, 0.04)	0.18 (0.15, 0.21)
Pain (0-10)	0.98 (0.90, 1.06)	0.0	-0.12 (-0.24, 0.01)	0.71 (0.51, 0.91)
Global assessment (0-10)	0.55 (0.47, 0.62)	0.0	-0.02 (-0.14, 0.10)	0.00 (-0.19, 0.18)
Fatigue (0-10)	0.75 (0.67, 0.83)	0.0	-0.02 (-0.15, 0.10)	1.12 (1.01, 1.39)

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/is-there-a-difference-in-rheumatology-patient-reported-outcomes-when-measured-at-home-versus-the-clinic-setting/