Kåre Birger Hagen

Objectives: To describe adherence to a 12-week web-based aerobic exercise programme, to compare characteristics between those who adhere or not, and to identify barriers for exercising in patients with hip or knee osteoarthritis. Design: Single-arm feasibility study. Subjects: Patients with hip or knee osteoarthritis in specialist healthcare, age 40–80 years, and not candidates for joint surgery. Methods: Adherence to a 12-week exercise programme was defined as having completed ≥ 2 exercise sessions a week for at least 8 weeks. Baseline differences between adherent and non-adherent groups in demographics, symptoms, disability, physical activity and fitness were assessed using Mann–Whitney U or χ2 tests. Reasons for not completing exercise sessions were reported in weekly diaries. Results: A total of 29 patients (median age 64 years, 72% female) were included. Median baseline pain (numerical rating scale 0–10) was 5. Fifteen patients adhered to the exercise programme, 14 did not. Non-adherent patients were less active (p = 0.032) and had lower cardiorespiratory fitness (p = 0.031). The most frequently reported barrier to exercising was sickness. Less than 10% reported pain as a barrier. Conclusion: Half of the patients with hip or knee osteoarthritis adhered to the digitally delivered exercise programme and the most frequently reported barrier for adherence was sickness, while less than 10% reported pain as a reason for not exercising. Trial registration: ClinicalTrials.gov, NCT04084834. The Regional Committee for Medical and Health Research Ethics South-East, 2018/2198.

The objective of this study was to develop prediction models and explore the external validity of the models in a large sample of patients with chronic widespread pain (CWP) and fibromyalgia (FM). Patients with CWP and FM referred to rehabilitation services in Norway (n = 986) self‐reported data on potential predictors prior to entering rehabilitation, and self‐reported outcomes at one‐year follow‐up. Logistic regression models of improvement, worsening and work status, and a linear regression model of health‐related quality of life (HRQoL), were developed using lasso regression. Externally validated estimates of model performance were obtained from the validation set. The number of participants in the development and the validation sets was 771 and 215 respectively; only participants with outcome data (n = 519–532 and 185, respectively) were included in the analyses. On average, HRQoL and work status changed little over one year. The prediction models included 10–11 predictors. Discrimination (AUC statistic) for prediction of outcome at follow‐up was 0.71 for improvement, 0.67 for worsening, and 0.87 for working. The median absolute error of predictions of HRQoL was 0.36 (0.22–0.51). Reasonably good predictions of working at follow‐up and HRQoL could be obtained using only the baseline scores as predictors. Moderately complex prediction models (10–11 predictors) generated poor to excellent predictions of patient‐relevant outcomes. Simple prediction models of working and HRQoL at follow‐up may be nearly as accurate and more practical. Prediction modelling of outcome in rehabilitation has been sparsely explored. Such models may guide clinical decision‐making. This study developed and externally validated prediction models for outcomes of people with chronic widespread pain and fibromyalgia in a rehabilitation setting. Multivariable prediction models generated poor to excellent predictions of patient‐relevant outcomes, but the complexity of these models may reduce their clinical utility. Simple univariable prediction models were nearly as accurate and may have more potential for use in clinical practice.

The clinical picture of fibromyalgia (FM) symptoms fluctuates, and the symptom severity varies within and between patients. The current study aimed to identify groups of PDS trajectories and to explore differences in baseline characteristics between the potential groups of trajectories. We included patients from a completed randomised controlled trial, in total 170 patients diagnosed with FM according to the ACR 2010 criteria. The mean age was 40 years, and 94% were women. Symptom severity was assessed by the Polysymptomatic distress scale (PDS) [range 0 (no symptoms) to 31] at four timepoints over 13–18 months. Latent class growth analysis was used to identify patient trajectories based on their response pattern on the PDS. Potential differences in baseline characteristics between the trajectories were compared using appropriate statistical tests. Two distinct PDS trajectories were identified with 110 patients (65%) classified as the “no improvement” group and 60 (35%) as the “some improvement” group. Mean PDS scores at pre-baseline were ≥ 20 in both groups. At 12 months, the groups diverged, mean (SD) PDS score was 14 (3.82) in the “some improvement” group and 21 (4.12) in the "no improvement" group. There were no significant differences in baseline characteristics between the groups of PDS trajectories. We identified one group of FM patients that improved slightly during the study period and one group that not improved. There were no differences in baseline characteristics between the two groups.

The objective of the study was to examine the hand-, knee- and hip osteoarthritis (OA) distribution, risk factor profiles and health-related quality of life (HRQoL) in a population-based OA cohort. Persons with self-reported OA responded to questionnaires and attended a clinical examination (N = 606). We analyzed cross-sectional associations to risk factor profiles and HRQoL dimensions (Short Form 36) in four mutually exclusive groups based on fulfillment of The American College of Rheumatology criteria: no OA (NOA), monoarticular upper extremity (hand) OA (MOAupper-ex.), monoarticular lower extremity (hip or knee) OA (MOAlower-ex.) and polyarticular OA (POA). Multivariate regression analyses and correspondence analysis were performed. The distribution of NOA, MOAupper-ex. MOAlower-ex. and POA was 21.1, 25.4, 22.4 and 31.0%, respectively. Compared to NOA, minor differences were found in risk factor profile in MOAupper-ex., whereas POA was significantly associated with sociodemographic, metabolic and mechanical features. The correspondence analysis identified different risk factor profiles between the four OA phenotypes, but the differences were not statistically significant (p = 0.13). Regarding HRQoL, neither OA groups were associated with poorer mental functioning. MOAlower-ex. and POA were associated with, among other things, poorer physical functioning (β = −6.2, 95% CI −11.2 to −1.2 and β = −12.5, 95% CI −17.4 to −7.9, respectively) and more pain (β = −5.4, 95% CI −10.4 to −0.3 and β = −10.5, 95% CI −15.3 to −5.8, respectively). In this cohort of persons with self-reported OA, POA was the most prevalent phenotype and was associated with an unfortunate risk factor profile and several diminished HRQoL dimensions. POA needs further attention in research and clinical care.

Objective To evaluate the state of quality of care for OA by summarizing studies that have assessed the care provided to patients. Methods A systematic review of community-based observational studies of actual clinical practice treating people with OA compared with quality indicators (QIs) was performed. Four databases were searched from January 2000 to November 2015. Two reviewers independently determined study eligibility, assessed risk of bias and extracted study data. The outcome was adherence to the QIs (pass rate). The overall pass rate (the total number of indicators passed divided by the total number of indicators for which the patients was eligible) was extracted from each study. When at least 50% of the studies had comparable individual QIs, the data were pooled with proportion meta-analyses. Results Fifteen studies comprising 16103 patients were included, and the median overall pass rate across studies was 41% (ranging from 22% to 65%). The pooled pass rates for individual QIs were: “Referral to orthopedic surgeon if no response to other therapy” 59.4% (95% Confidence Interval 47.5 - 70.3), “Paracetamol or acetaminophen first drug used” 46.0% (26.6 - 66.7), “Assessed for pain and/or function” 45.5% (33.9 – 57.6), “Referral or recommendation to exercise” 38.7% (28.9 – 49.5) “Offered education and self-management” 35.4% (27.8 - 44.0), and “Informed about potential risks if NSAIDs prescribed” 34.1% (24.7-44.9). Conclusion There is room for improvement in community-based OA care.

People with osteoarthritis desire high quality care, support and information. However, the quality of care for people with OA in general practice is not routinely collected. Quality Indicators can be used to benefit patients by measuring whether minimum standards of quality care are being met from a patient perspective. The aim of this study was to describe how a Research User Group (RUG) worked alongside researchers to co-produce a set of self-reported quality indicators for people with osteoarthritis when visiting their general practitioner or practice nurse (primary care). These were required in the MOSAICS study, which developed and evaluated a new model of supported self-management of OA to implement the NICE quality standards for OA. This article describes the public involvement in the MOSAICS study. This was 1) the co-development by RUG members and researchers of an Osteoarthritis Quality Indicators United Kingdom (OA QI (UK)) questionnaire for use in primary care, and 2) the comparison of the OA QI (UK) with a similar questionnaire developed in Norway. This study shows how important and effective a research user group can be in working with researchers in developing quality care indicators for osteoarthritis for use in a research study and, potentially, routine use in primary care. The questionnaire is intended to benefit patients by enabling the assessment of the quality of primary care for osteoarthritis from a patient’s perspective. The OA QI (UK) has been used to examine differences in the quality of osteoarthritis care in four European countries. Background People with osteoarthritis (OA) desire high quality care, support and information about OA. However, the quality of care for people with OA in general practice is not routinely collected. Quality Indicators (QI) can be used to benefit patients by measuring whether minimum standards of quality care (e.g. NICE quality standards) are being met from a patient perspective. A Research User Group (RUG) worked with researchers to co-produce a set of self-report, patient-generated QIs for OA. The QIs were intended for use in the MOSAICS study, which developed and evaluated a new model of supported self-management of OA to implement the NICE guidelines. We report on 1) the co-development of the OA QI (UK) questionnaire for primary care; and 2) the comparison of the content of the OA QI (UK) questionnaire with a parallel questionnaire developed in Norway for the Musculoskeletal Pain in Ullensaker (MUST) study. Methods Researchers were invited to OA RUG meetings. Firstly, RUG members were asked to consider factors important to patients consulting their general practitioner (GP) for OA and then each person rated their five most important. RUG members then discussed these in relation to a systematic review of OA QIs in order to form a list of OA QIs from a patient perspective. RUG members suggested wording and response options for a draft OA QI (UK) questionnaire to assess the QIs. Finally RUG members commented on draft and final versions of the questionnaire and how it compared with a translated Norwegian OA-QI questionnaire. Results RUG members (5 males, 5 females; aged 52–80 years) attended up to four meetings. RUG members ranked 20 factors considered most important to patients consulting their GP for joint pain. Following discussion, a list of eleven patient-reported QIs for OA consultations were formed. RUG members then suggested the wording and response options of 16 draft items – four QIs were split into two or more questionnaire items to avoid multiple dimensions of care quality within a single item. On comparison of this to the Norwegian OA-QI questionnaire, RUG members commented that both questionnaires contained seven similar QIs. The RUG members and researchers agreed to adopt the Norwegian OA-QI wording for four of these items. RUG members also recommended adopting an additional seven items from the Norwegian OA-QI with some minor word changes to improve their suitability for patients in the UK. One other item from the draft OA QI (UK) questionnaire was retained and eight items were excluded, resulting in a 15-item final version. Conclusions This study describes the development of patient-reported quality indicators for OA primary care derived by members of a RUG group, working in partnership with the research team throughout the study. The OA QI (UK) supports the NICE quality standards for OA and they have been successfully used to assess the quality of OA consultations in primary care in the MOSAICS study. The OA QI (UK) has the potential for routine use in primary care to assess the quality of OA care provided to patients. Ongoing research using both the UK and Norwegian OA-QI questionnaires is assessing the self-reported quality of OA care in different European populations.