Summary
This is the first observational study regarding the effects of the SCS on outcomes of care, that is, the degree of pain, physical function, self-efficacy, and active pain coping, in patients with hip or knee OA over a period of 2 years after implementation of the SCS in primary care. Although more than half of the patients received health care that was considered inconsistent with the SCS recommendations, statistically significant differences were not identified in any of the outcomes of care between patients receiving care inconsistent with the SCS recommendations and patients receiving care consistent with the SCS recommendations.
The results raise several important issues that need to be considered in further research regarding the additional value of a complex intervention such as the SCS.
Strengths and limitations
First, patients who received SCS-consistent care were younger, had fewer comorbidities and painful joints, and were more likely to have additional healthcare insurance. The SCS takes contextual factors into account only to a minor extent. Therefore, SCS-inconsistent care may sometimes be unavoidable or even preferable. For example, a regular exercise programme for OA is not feasible for patients with severe cardiovascular disease. Also, not having additional health insurance may be an insurmountable barrier to use SCS-recommended modalities. Therefore, further exploration of GPs’ reasons for providing SCS-inconsistent care is needed to gain insight and redefine SCS-consistent care in a more sophisticated way. As patients were only included if they were able to communicate in Dutch, any cultural factors were not addressed. When exploring GPs’ reasons, cultural factors should also be considered as they may influence the choice for treatment as well.
Second, it appeared that relevant differences existed in healthcare use between the two groups. For example, patients in the SCS-consistent group were found to be more likely to receive education and lifestyle advice, while patients in the SCS-inconsistent group were more likely to receive intra-articular injections. This may have influenced outcome differences, because the effect of education and lifestyle on pain may be relatively small, whereas the effect of intra-articular injections may be relatively large.14 SCS-inconsistent care (for example, receiving injections before lifestyle advice) may thus show better outcomes. An important aim of the SCS is to reduce the inappropriate use of advanced treatment modalities, such as arthroplasties, by making optimal use of the non-surgical treatment modalities for as long as possible.1 Therefore, side effects and costs were important drivers for the SCS recommendations. SCS-consistent care may in this sense lead to high-quality care at the cost of pain and dysfunction. The added value of the SCS can only be appraised by considering a range of outcome measures simultaneously over a prolonged period of time in which cost–benefits and the number of side effects should be included.15
Third, the time-frame of this study may have been too short to detect differences in these outcomes. Evidence shows that changes in pain and physical function within the first years of follow-up in patients with hip or knee OA are relatively small and, thus, the potential benefit of the SCS may only be seen after a prolonged period of time.16–18 Also, the results suggest that a relevant difference in change of pain scores cannot be excluded in favour of the patients who received SCS-inconsistent care, as the minimal important change of the WOMAC pain score19 lies within the 95% CI. Further, self-efficacy and active pain coping hardly changed during the follow-up period, whereas the self-management booklet was intended to strengthen patient empowerment on these issues. These findings were in line with a recent randomised controlled trial regarding the effect of a self-management intervention for long-term chronic conditions in primary care, which concluded that the intervention had no noticeable value to existing care.20 It may require more advanced interventions such as cognitive-behavioural group interventions,21 pain-coping skill training,22,23 or self-management training24 to effectively improve self-efficacy and an active coping style.
Finally, an observational study may not be the ideal design to find a potential association between SCS-consistent care and better health outcomes, as the most important limitation of such design is the threat of potential confounding by indication. Although baseline differences were adjusted for between the two groups, unknown and not measured differences cannot be adjusted for. Many will prefer the use of controlled designs over observational designs to evaluate complex interventions such as the SCS, as those can potentially counterbalance this bias. However, these designs have their own limitations, such as the risk of contamination, recruitment bias, the need for larger study samples, or costs.25 The external validity of controlled trial results can often be questioned in studies with complex interventions. Therefore, the trade-off between the costs to overcome the practical consequences of controlled designs and the value of the evidence that can be gathered given these constraints should be considered in further research.
To the authors’ knowledge, this is the first study examining the association between care that is provided according to implemented guidelines and health outcomes.
