JMIR Publications

ABSTRACT

Background:

Since 2004, we have collected patient-reported outcome (PRO) data from several Danish patient populations for use at group and patient levels.

Objective:

The aim of this paper is to highlight trends during the last 15 years with respect to patient inclusion, the methods for collection of PRO data, the processing of the data, and the actual applications and use of the PRO measurements.

Methods:

All PRO data have been collected using the WestChronic/AmbuFlex PRO system, which was developed by the author in 2004 and has been continuously updated since. The analysis of trends was based on a generic model applicable for any kind of clinical health data, according to which any application of clinical data may be divided into four processes: patient identification, data collection, data aggregation, and the actual data use. Data were generated by a specific application in the system and transferred to analysis in the R package.

Results:

During the 15-year period, 77,828 patients answered 253,998 questionnaires containing a total of 13,204,223 responses. A number of marked changes have taken place. 1) The creation of cohorts for clinical epidemiological research purposes has shifted towards cohorts defined by clinical use of PRO data at the patient level 2) The development of AmbuFlex, where PRO data are used as the entire basis for outpatient follow-up instead of fixed appointments, has undergone exponential growth, and the system is currently in use in 47 ICD groups, covering 16,000 patients and 94 departments throughout Denmark. 3) In the first part of the period, a combination of paper- and web-based questionnaires secured response rates above 90% and low attrition rates; while in 2005, 66.5% of questionnaires were paper-based, this is the case for only 4.3% in 2019. 4) The approach methods for questionnaires and reminders have changed dramatically from letter, e-mails, and text-messaging to a national, secure electronic mail system, through which 93.2% of the communication to patients took place in 2019. The combination of secure email and web-based answering has resulted in a low turnaround time in which half of the responses are now received within 5 days.

Conclusions:

Substantial changes have occurred in the PRO scene during the 15-year period. The demand for clinical use of PRO measurements has increased, driven by a wish among patients as well as clinicians to use PRO as a mean to promote better symptom assessment, more patient-centred care, and more efficient use of resources. Important technological changes have occurred, creating new opportunities and making PRO collection and use cheaper and more feasible. Also, a number of legal changes have occurred and may constitute a barrier for further development as well as a barrier for better utilisation of patients’ questionnaire data. In the future, clinical settings will be the primary source of PRO data, not only for clinical application but also for other purposes like research and quality surveillance, and new ways to conduct complementary data collection will be necessary. However, the current legal restrictions on the joint use of health data imposed by the EU’s General Data Protection Regulation makes no distinction between use and misuse, and steps should be taken to alleviate these restrictions on the joint use of PRO data.