Collaborative nurse-led self-management support for primary care patients with anxiety, depressive or somatic symptoms: Cluster-randomised controlled trial (findings of the SMADS study)

Abstract

BACKGROUND: Collaborative, nurse-led care is a well-established model of ambulatory care in many healthcare systems. Nurses play a key role in managing patients' conditions as well as in enhancing symptom- and self-management skills. OBJECTIVE: The SMADS trial evaluated the effectiveness of a primary care-based, nurse-led, complex intervention to promote self-management in patients with anxiety, depressive or somatic symptoms. Change in self-efficacy 12 months post baseline was used as the primary outcome. DESIGN: The SMADS trial set up a two-arm, cluster randomised controlled trial in the city of Hamburg, a large metropolitan area in the North West of Germany. SETTING: We randomly allocated participating primary care practices to either the intervention group (IG), implementing a nurse-led collaborative care model, or to the control group (CG), where patients with the above psychosomatic symptoms received routine treatment. PARTICIPANTS: Patients from 18 to 65 years of age, regularly consulting a participating primary care practice, scoring≥5 on the anxiety, depressive or somatic symptom scales of the Patient Health Questionnaire (PHQ-D), German version. METHODS: A mixed model regression approach was used to analyse the outcome data. Analyses were based on the intention-to-treat population: All enrolled patients were analysed at their follow-up. Additionally, we reported results as effect sizes. The robustness of the results was investigated by performing an observed cases analysis. RESULTS: 325 participants (IG N=134, CG N=191) from ten practices in each study arm consented to take part and completed a baseline assessment. The mean group difference (ITT-LOCF, IG vs. CG) in self-efficacy at the post baseline follow-up (median 406days) was 1.65 points (95% CI 0.50-2.8) in favour of IG (p=0.004). This amounts to a small Cohen's d effect size of 0.33. An observed cases analysis (168 participants, IG=56, CG=105) resulted in a mean difference of 3.13 (95% CI 1.07-5.18, p=0.003) between the groups, amounting to a moderate effect size of d=0.51. CONCLUSION: A complex, nurse-led intervention, implemented as a collaborative care model, increased perceived self-efficacy in patients with symptoms of anxiety, depression or somatisation compare to control patients. For the first time in the German healthcare system, the SMADS trial validated the belief that a nurse can successfully complement the work of a general practitioner - particularly in supporting self-management of patients with psychosomatic symptoms and their psychosocial needs.