The Quebec Pain Registry (QPR) is a large research database of patients suffering from various chronic pain (CP) syndromes who were referred to one of five tertiary care centres in the province of Quebec (Canada). Patients were monitored using common demographics, identical clinical descriptors, and uniform validated outcomes. This paper describes the development, implementation, and research potential of the QPR. Between 2008 and 2013, 6902 patients were enrolled in the QPR, and data were collected prior to their first visit at the pain clinic and six months later. More than 90% of them (mean age ± SD: 52.76 ± 4.60, females: 59.1%) consented that their QPR data be used for research purposes. The results suggest that, compared to patients with serious chronic medical disorders, CP patients referred to tertiary care clinics are more severely impaired in multiple domains including emotional and physical functioning. The QPR is also a powerful and comprehensive tool for conducting research in a “real-world” context with 27 observational studies and satellite research projects which have been completed or are underway. It contains data on the clinical evolution of thousands of patients and provides the opportunity of answering important research questions on various aspects of CP (or specific pain syndromes) and its management.
In the field of pain research, like in other medical fields, randomized controlled trials (RCTs) are the gold standard for establishing the efficacy of interventions. However, RCTs have several limitations [
A patient registry is defined as “an organized system that uses observational study methods to collect uniform data (clinical and other) to evaluate specified outcomes for a population defined by a particular disease, condition or exposure, and that serves a predetermined scientific, clinical or policy purpose(s)” [
In 2008, the Health Ministry of the province of Quebec (Canada) designated four Pain Centres of Expertise within the Montreal, McGill, Sherbrooke, and Laval University Health Networks which altogether cover the entire province. The Ministry wished to monitor the clinical outcomes of patients treated in these newly designated centres (and especially in tertiary care pain clinics) and obtain relevant administrative statistics. In parallel, one of the strategic plans of the Quebec Pain Research Network for 2007–2011 was to develop a province-wide clinical pain research infrastructure to facilitate the conduct of large observational and clinical studies. To meet the objectives of both of these organizations, there was a need to develop a uniform multisite registry that documents the clinical condition and evolution of patients treated in tertiary care pain clinics. This gave the impetus to implement in the Quebec Pain Registry (QPR) project designed to serve both clinical/administrative and research purposes. To our knowledge, only two other registries of patients with various types of chronic pain disorders treated in multidisciplinary clinics have been developed so far, one in the UK (Pain Audit Collection System) [
The present paper describes how the QPR was developed and implemented detailing its strength and shortcomings with the aim of facilitating the creation of other pain patient registries. The QPR structure and content are also presented along with the characteristics of the enrolled patients. The policy and procedures for accessing QPR data sets for research purposes are described as well as the type of access requests made.
The aims of the QPR project were to (1) put in place a prospective web-based registry of ambulatory patients suffering from various types of pain syndromes who were referred for multidisciplinary treatment in large university-affiliated pain clinics in the province of Quebec, (2) assess and monitor their condition over time using common demographics, identical clinical descriptors, and uniform outcomes measured with standardized/validated measurement tools in each participating site, (3) document pain treatments patients received and/or used over time, (4) provide clinicians with a summary of the individual condition of their patients along with useful administrative statistics for their pain treatment facility, and (5) provide reliable “real-world” data to researchers wishing to answer important research questions or test hypotheses regarding various aspects of chronic pain (or specific pain syndromes) and its management, to assess study feasibility, and to facilitate and speed up patient recruitment in research projects or clinical trials.
Using the guidelines proposed by Solomon et al. (1991) [
The choice of items to be included in the QPR was made with the objective of creating a uniform
All the above information except for patient sociodemographics was incorporated into a single questionnaire named the Initial Nurse-administered Questionnaire. A second questionnaire was also developed using the approach described above in order to collect follow-up data after the patients’ first visit at the pain clinic (6-Month Nurse-administered Questionnaire).
With regard to patient pain diagnoses to be established by the pain physicians at the participating sites, it was felt that the International Classification of Diseases (ICD-9 or 10 systems) [
Paper Case Report Forms (CRFs) were then prepared and used in a pilot prospective study to test the feasibility of implementing/running the QPR. Additional pieces of information were collected during the pilot study such as the time taken to complete each questionnaires/interview, patients’ perceived degree of difficulty for filling out the questionnaires, and clinical usefulness of the collected material. Physicians were also asked to check the items they would like to be included in a clinical summary form. After having obtained institutional ethic approval of the research protocol, the pilot study was conducted in 2007-2008 with 90 consecutive patients recruited in three multidisciplinary pain treatment facilities which were candidates for becoming the designated tertiary care clinics of the Quebec Pain Centres of Expertise. These clinics were, respectively, located at the
Variables, outcomes, and measurement tools of the Quebec Pain Registry at each time point.
Variables/outcomes collected with the Patient self-administered Questionnaire (QP) and the Nurse-administered Questionnaire (NQ) | Initial visit | 6-month follow-up |
---|---|---|
Pain history | ||
(i) NQ: pain duration | X | |
(ii) NQ: circumstances surrounding the onset | X | |
(iii) NQ: 1st degree family history of chronic pain | X | |
(iv) NQ: date and reason of referral, speciality of the referring doctor | X | |
(v) NQ: number of pain-related visits to emergency (past 6 months) | X | X |
(vi) NQ: number of pain-related hospitalizations (past 6 months) | X | X |
(vii) NQ: time elapsed between consultation request and 1st visit at the Pain Clinic | X | |
Pain characteristics | ||
(i) NQ: frequency in the past 7 days (always, occasionally, no pain) | X | X |
(ii) PQ: intensity (pain now, average, and worst pain in the past 7 days) | X | X |
(iii) NQ: quality (neuropathic pain component) | X | X |
(iv) PQ: pain interference on daily activities | X | X |
(v) NQ: impact of pain on sleep | X | X |
(vi) NQ: mobility support required inside and/or outside the home | X | X |
(vii) NQ: pain diagnosi(e)s established at the pain clinic: location, type, suspected etiology | X | X |
Psychological well-being and quality of life | ||
(i) PQ: depression | X | X |
(ii) PQ: anger | X | X |
(iii) PQ: tendency to catastrophize in the face of pain | X | X |
(iv) PQ: health-related quality of life | X | X |
Pain treatments at the pain clinic | ||
(i) NQ: current pharmacological pain treatment (prescribed and not prescribed): medication name and posology | X | X |
(ii) NQ: side effects of current pharmacological pain treatment: type and severity | X | X |
(iii) NQ: past pharmacological pain treatment (prescribed and not prescribed): medication name and reason(s) for stopping | X | X |
(iv) NQ: type of current and past nonpharmacological pain treatments including interventions (e.g., injection therapy, surgery), psychological techniques (e.g., self-management program, individual psychotherapy), self-management strategies (e.g., relaxation/breathing exercises, self-support group), physical therapies (e.g., physiotherapy, electrostimulation, acupuncture), and complementary alternative therapies | X | X |
(v) NQ: type of health care professionals consulted since pain onset and in the months preceding follow-up | X | X |
(vi) NQ: continuation of treatment at the pain clinic (yes, no) | X | X |
(vii) NQ: patient’s disposition after treatment at the pain clinic | X | |
Patient expectations regarding treatment at the pain clinic | ||
(i) PQ: expected pain relief | X | |
(ii) PQ: patient expected global change regarding functioning level and quality of life | X | |
Patients’ perceived improvement and satisfaction with treatment at the pain clinic | ||
(i) PQ: patient perception of pain relief | X | |
(ii) PQ: patient expected global impression of change regarding functioning level and quality of life | X | |
(iii) PQ: patient satisfaction with treatment | X | |
Medical history | ||
(i) NQ: current and past medical history (type of disorders other than chronic pain) | X | X |
(ii) NQ: type of current medication for medical condition | X | X |
(iii) PQ: consumption habits (cigarettes, alcohol, illicit drugs) | X | X |
(iv) PQ: risk of alcohol and drug abuse/misuse | X | |
(v) NQ: risk of opioid abuse/misuse | X | |
Demographics | ||
(i) PQ: date of birth | X | |
(ii) PQ: sex | X | |
(iii) PQ: ethnic group | X | |
(iv) PQ: first language | X | |
(v) PQ: education level | X | |
(vi) PQ: current living conditions | X | X |
(vii) PQ: civil status | X | X |
(viii) PQ: current work status | X | X |
(ix) PQ: family income | X | X |
(x) PQ: main source of income | X | X |
(xi) PQ: disability benefits | X | X |
(xii) PQ: litigation regarding disability benefits | X | X |
NQ, Nurse-administered Questionnaire; PQ, Patient self-administered Questionnaire (PQ).
The RA contacted the Registry Nurse (RN) who conducted a structured telephone or face-to-face interview with the patients in the days/hours preceding their first appointment at the pain clinic using the Initial Nurse Questionnaire. Depending on the patient’s clinical condition, the interview lasted between 30 and 90 minutes, and the information was entered by the RN or RA into the online QPR database. A summary of the patient’s clinical condition (e.g., pain duration and intensity, and analgesic intake) was then generated from the database and transmitted to the treating physician of the pain clinic. Follow-up data were collected using the 6-Month Patient and Nurse Questionnaires using a similar methodology as the one described above. Additional follow-up data were gathered at 12 and 24 months after the initial visit but only in those patients who had not been discharged from the pain clinic in the meantime. These data were collected with questionnaires containing the same measures as those administered at the 6-month follow-up. Due to financial considerations, follow-up data were not collected at 12 and 24 months in newly enrolled patients after March 2012. Due to budget cuts in the QPR project for the year 2014-2015, collection of 6-month follow-up data had to be interrupted in newly registered patients after June 2014, and enrolment of new patients ended in November 2014.
In order to have a more complete picture of patient pain management than the one provided in medical records, all data about pharmacotherapy (prescribed and over-the-counter medication) and nonpharmacological treatments used inside and outside of the pain clinic (including complementary alternative therapy) were collected by the Registry Nurses who did not have clinical duties in the QPR participating sites. The purpose of having nurses rather than RA for conducting the interviews with the patients was twofold: (1) to ensure accuracy of the patient clinical summary transmitted to the treating physician at the pain clinic and (2) to optimize the quality of the medical/clinical data contained in the QPR database.
Quality monitoring of the QPR database was under the responsibility of the Registry Nurse Coordinator and her assistant. Each participating site was requested to provide a monthly report of the number of patients enrolled in the registry, reasons for exclusion, number of questionnaires not completed and reasons why, losses to follow-up, and so on. With regard to data quality monitoring, a series of quality controls were programmed in the QPR database to allow instant automated data validation checks (e.g., out-of-range values, logical inconsistencies). To facilitate medication data entry and ensure consistency (e.g., generic versus brand name), a medication dictionary was built in the database. Manual data cleaning was also carried out on a regular basis to identify discrepancies and missing data on variables targeted as important (e.g., patient diagnosis, pain duration, medication, and medical history) and to generate “queries” to be sent to the participating sites for resolution. Statistical programs to identify errors or inconsistencies on specific measures were also part of quality control activities. Errors identified when data were analysed for administrative or research purposes were also corrected in the database.
Datasets of patients who gave informed consent can be accessed for conducting observational studies. Assessment of feasibility of research projects or clinical trials is also possible (e.g., number of QPR female patients aged between 30 and 50 years with a diagnosis of complex regional pain syndrome). The QPR can also be used to conduct “satellite” research projects, that is, studies in which data contained in the QPR (e.g., age, sex, and types of pain medication) are linked to other sets of data (e.g., governmental administrative databases, data collected in the context of a new study on variables other than the ones contained in the QPR). However, the research protocol and the accompanying patient consent form of the satellite projects have to receive prior approval by the clinical team of the participating site(s), the central REB of the CHUM, and the local REBs. Once the project is approved by these authorities, the Medical Director of the participating clinic(s) sends a letter to inform the eligible QPR patients about the research project and invites them if they are interested in participating to contact the person in charge of the study or her/his representative. Finally, the QPR data can be accessed to facilitate and speed up patients’ recruitment in research projects or clinical trials; the procedure is the same as the one used for satellite research projects.
The total number of patients enrolled in the QPR between November 1, 2008, and December 21, 2014, is 9363 (
The number and percentages of the Patient and Nurse Questionnaires which were completed prior to the initial visit at the pain clinic and at 6-month follow-up were computed. In order to assess if missing questionnaires qualified as “missing at random” [
Descriptive statistics including measures of central tendency (mean or median) and dispersion (standard deviation or range) along with frequency tables were used to document the characteristics of the patients enrolled in the QPR. Due to space limitation, only a subset of the variables in Table
Descriptive statistics were computed on the kind and number of requests made for accessing QPR data and the type of users.
Of the 8,233 patients who were referred to the participating pain clinics between November 2008 and December 2013 (inclusively), 7021 (85.3%) qualified for enrolment in the QPR and only 1.7% refused to do so (Figure
Flow of participants through the QPR during the study period.
Patients enrolled in the QPR during the study period were aged between 18 and 88 years (mean = 52.76, SD = 14.6), 59.1% were female, and the vast majority (92.0%) were of Caucasian origin (Table
Demographic characteristics of the 6,337 patients enrolled in the Quebec Pain Registry.
Mean | SD | |
---|---|---|
Age | 52.76 | 14.6 |
| ||
| % | |
| ||
Sex | ||
Female | 3742 | 59.1 |
Male | 2595 | 40.9 |
Education | ||
None | 23 | 0.4 |
Primary | 488 | 7.8 |
Secondary | 2361 | 37.9 |
College | 1749 | 28.1 |
University | 1613 | 25.9 |
Civil status | ||
Married/common law | 3562 | 57.1 |
Single | 1458 | 23.4 |
Separated/divorced | 910 | 14.6 |
Widowed | 309 | 4.9 |
Ethnicity | ||
Caucasian | 5755 | 92.0 |
Black descent | 162 | 2.6 |
Asian | 91 | 1.5 |
Hispanic | 83 | 1.3 |
Native | 83 | 1.3 |
Mixed race | 81 | 1.3 |
Work status | ||
Full-time work | 1236 | 19.8 |
Part-time work | 521 | 8.3 |
Temporary disability income | 1212 | 19.4 |
Permanent disability income | 1113 | 17.8 |
Retired | 1299 | 20.8 |
Unemployed/laid-off | 368 | 5.9 |
Homemaker | 366 | 5.9 |
Student | 102 | 1.6 |
Volunteer | 18 | 0.3 |
Other | 21 | 0.3 |
The median amount of time elapsed between patients’ referral and first visit at the pain clinic was 3.5 months; 35.0% of the patients waited more than 6 months for their appointment, some of them (5.2%) having waited between 2 and 4 years (Table
Pain-related characteristics of patients enrolled in the Quebec Pain Registry up to December 31, 2013.
Mean | SD | |
---|---|---|
Average pain intensity in the past 7 days | 6.71 | 2.0 |
Worst pain intensity in the past 7 days | 8.16 | 1.8 |
Physical Health-Related QOL (SF-12v2) | 29.07 | 8.9 |
Mental Health-Related QOL (SF-12v2) | 40.48 | 11.7 |
| ||
| % | |
| ||
Evidence of neuropathic pain | ||
(i) No | 1336 | 23.9 |
(ii) Yes | 1732 | 31.0 |
Mixed evidence | 2511 | 45.0 |
Pain duration | ||
(i) <1 year | 702 | 13.1 |
(ii) 1 year to <3 years | 1201 | 22.4 |
(iii) 3 years to <5 years | 964 | 18.0 |
(iv) 5 years to <10 years | 1098 | 20.4 |
(v) ≥10 years | 1405 | 26.2 |
Time elapsed between referral and 1st visit | ||
(i) <0.5 year | 3603 | 65.0 |
(ii) 0.5 year to <2 years | 1620 | 29.2 |
(iii) 2 years to <4 years | 297 | 5.4 |
(iv) ≥4 years | 23 | 0.4 |
Pain interference over the past 7 days (BPI score ≥ 7/10) | ||
(i) General activity | 3684 | 59.1 |
(ii) Mood | 2931 | 47.0 |
(iii) Walking ability | 2789 | 44.7 |
(iv) Normal work | 4016 | 64.4 |
(v) Relations with other people | 2225 | 35.7 |
(vi) Sleep | 3410 | 54.7 |
(vii) Enjoyment of life | 2337 | 37.5 |
(viii) Self-care | 1653 | 26.5 |
(ix) Recreational activities | 4069 | 65.2 |
(x) Social activities | 3275 | 52.5 |
Depressive symptoms (BDI-I) | ||
(i) None or minimal (0–9) | 1335 | 21.4 |
(ii) Mild (10–18) | 2167 | 34.8 |
(iii) Moderate (19–29) | 1786 | 28.7 |
(iv) Severe (30–63) | 946 | 15.2 |
An accident or a trauma was at the origin of the pain in more than half of the cases (52.5%); 31.4% of the patients reported that their pain occurred during or following an illness or and 14.2% after a surgery while one patient out of five (22.5%) was unable to associate the onset of her/his their pain to any precise event. Figure
Top 10 pain diagnoses made by the physicians of the pain clinics.
The majority of the patients (85.0%) reported that their pain was present continuously in the 7 days preceding their first visit at the pain clinic. Mean pain intensity scores for the “average” pain and “worse” pain during this time period were 6.71 (SD = 2.0) and 8.16 (SD = 1.80), respectively. Patients’ ratings on the interference scales of the Brief Pain Inventory-10 [
Mean scores on the physical and mental summary scales of the SF-12v2 in QPR patients (
When questioned about the expected percentage of pain relief at six months after initiating treatment at the pain clinic, more than half of the patients (53.4%) anticipated pain relief ranging between 50 to 80% while one patient out of four (25.0%) expected pain relief superior to 80%. A large percentage of patients also anticipated that their functioning level (63.4% of patients) and quality of life (65.3% of patients) would be greatly or considerably improved over the next six months.
Table
Type and number of studies for which access to QPR data has been requested up to February 2016.
Observational studies | Satellite research project | Feasibility studies | Patient recruitment for external studies | Total | |
---|---|---|---|---|---|
Students | |||||
(i) Undergraduate | 1 | 1 | |||
(ii) M.S. | 3 | — | — | — | 3 |
(iii) Ph.D. | 4 | 2 | — | — | 6 |
(iv) Postdoctoral | 2 | 2 | — | 1 | 5 |
(v) Research/clinical fellowship | 3 | 1 | — | 1 | 5 |
Academic researchers | 4 | 1 | 6 | 1 | 11 |
Industry researchers | 4 | — | 1 | 1 | 6 |
Clinicians | — | — | 2 | — | 2 |
| |||||
Total | 21 | 6 | 9 | 4 | 40 |
As shown in the present paper, developing and implementing a multisite patient registry is a complex task. Although maintaining a registry such as the QPR is very costly, we have shown that it is feasible to collect uniform and reliable data in a large number of tertiary care patients suffering from a variety of pain syndromes and across different clinics. The collected information can help clinicians in making their diagnosis and management plan and can provide participating pain clinics with useful statistics on their practices. In addition to documenting the characteristics and management of patients referred to multidisciplinary pain clinics, our results showed that the QPR made possible the conduct of observational studies and satellite research projects using “real-world” data on various aspects of chronic pain.
In terms of the feasibility of implementing the QPR, our results showed that only 1.7% of the potentially eligible patients refused to complete any questionnaires and 8.2% did not consent that their data be used for research purposes. Close to 100% of patients completed both the Initial Patient Questionnaire and Nurse Questionnaire. At 6-month follow-up, the percentage decreased but the overall retention rate remained high; that is, nearly 90% completed at least one of the two questionnaires. Some statistically significant differences were found in terms of patients’ age, sex, and study site between those who did and did not answer the questionnaires at each time point and between time points. However, all effect sizes were not clinically meaningful suggesting that missing questionnaires did not introduce bias in the QPR data [
Compared to the majority of existing pain patient registries [
With regard to the characteristics of the participants enrolled in the QPR prior to their first visit at the pain clinic, our results highlight the fact that patients attending tertiary care pain clinics are significantly impaired in multiple domains. The majority of patients experienced continuous pain that reached intensity levels severe enough to interfere substantially with various aspects of their daily living including emotional well-being. Consistent with earlier results obtained in a smaller sample [
Interestingly, the median time elapsed between patients’ referral and first visit at the pain clinic was found to be 3.5 months. Eleven years ago, Veillette et al. (2005) examined the waitlists of pain services across the province of Quebec and found that two-thirds of the patients were waiting for 9 months or more [
In terms of expectations toward treatment, our results revealed that close to 70% of the patients anticipated great or considerable improvement in their functioning and quality of life while one patient out of five was expecting 80–100% pain relief following treatment at the pain clinic. Whether such high expectations can be detrimental to patients’ outcomes was recently investigated using the initial visit and 6-month follow-up QPR data. The results of this study suggest that individuals who expected positive changes were more inclined to perceive improvements in their overall condition, leading to superior clinical outcomes [
Although our above study findings are informative and are based on a large sample size of patients followed prospectively in several sites and in a real-world context, they have limitations that should be acknowledged. First, they characterized only a small proportion of the chronic pain population, that is, those who are referred to tertiary care clinics, such that the results cannot be generalized to other populations of patients treated in primary or secondary care settings. Second, it is important to point out that access to tertiary care clinics in the province of Quebec requires a physician referral; access to these clinics is free but limited due to relatively long waiting lists as is the case in other Canadian provinces [
The present paper finally illustrates how “real-world” patient registries such as the QPR can be valuable and powerful research tools [
The QPR is a vast registry of patients referred to multidisciplinary pain treatment clinics that was designed for clinical/administrative and research purposes. This registry provides numerous opportunities to study various aspects of chronic pain (or specific pain syndromes) and its management using longitudinal “real-world” data on a large set of variables collected in tertiary care patients. The most important challenge posed by the QPR remains to be its maintenance costs which have compromised its long-term sustainability and its expansion in other pain clinics.
The Quebec Pain Registry (QPR) Project, led by Drs. Manon Choinière (M.C.) and Mark Ware (M.W.), was initially funded by the Quebec Pain Research Initiative which was itself financed by governmental grants from Valorisation Recherche Québec and Canada Foundation for Innovation. Then, the QPR was supported by the Quebec Pain Research Network (QPRN) which was itself funded by a governmental grant from the Fonds de Recherche du Québec, Santé (FRQS). The QPRN was also supported by the Quebec Health Ministry, Pfizer Canada Inc., and Astra Zeneca Inc. and to a lesser extent by Janssen Inc. whose contributions were all channeled through the FRQS via an official financial partnership. All funding sources had no involvement in data analysis/interpretation and manuscript preparation. When the QPR was developed and implemented, M. Choinière and M. A. Ware were research scholars of the FRQS. Dr. Gabrielle Pagé is currently a recipient of a postdoctoral research award from the Canadian Institutes of Health Research and Dr. Philippe Sarret holds a Canada Research Chair on neurophysiopharmacology of chronic pain. Between 2009 and 2012, M. Choinière and M. A. Ware were members of the Scientific Committee of the Pfizer Neuropathic Pain Research Award and received honoraria from Pfizer Canada Inc. for reviewing grant applications and assisting in the meetings of the committee; the same was true for Dr Aline Boulanger but only for 2009-2010 and 2010-2011.
All authors of the present paper certify that they have no conflict of interests with any financial organization regarding the material presented and discussed in this manuscript.
The authors thank all the nurses and assistants for their dedicated work during the development/implementation of the QPR and the data collection process at the multidisciplinary pain treatment clinics of the CHUM, MUHC, CHUS, CHUQ, and HDL. Thanks are also due to the clinicians working in each participating site and to the patients who gave consent for their QPR data to be used for research purposes. Mr. John Padoba and his team from Dacima Software Inc. also deserve thanks for their work developing the first version of the electronic web-based software for inputting QPR data. The authors thank Mr. Benoit Duchaine and his team from Typhon Solutions Inc. who developed the updated electronic CRFs and database of the QPR. Finally, thanks are due to Mr. Marc Dorais (StatSciences Inc.) who conducted the statistical analyses carried out in the early phases of the QPR project and to Mr Hichem Saïdi (Research Centre of the CHUM) who participated into the editing process of the manuscript.