Moving on! Can a Tailored Treatment in a Primary Care Setting Reduce Symptoms, Healthcare Consumption, Sick Leave and Prevent Pain and Hypertension in Individuals With Anxiety Disorders?
Moving on! Can a Tailored Treatment in a Primary Care Setting Reduce Symptoms, Healthcare Consumption, Sick Leave and Prevent Pain and Hypertension in Individuals With Anxiety Disorders?
The goal of this clinical study is to investigate the effects of two different interventions in patients with anxiety disorder and who seeks help at primary care centers. The participants will choose between two interventions:
The participant will fill in self-assessment scales and questionnaires and also have BMI and blood pressure measured at the start of the study, at 12-weeks after the study start and after 1 year.
The main question the investigators aim to answer is if the interventions are better than treatment as usual for this group of patients regarding:
Every third patient in the waiting room at a Swedish primary healthcare center is seeking help for common mental disorders including anxiety (1). Anxiety disorders substantially reduce quality of life and daily functioning and are, among mental disorders, the second leading cause of years lived with disability and the sixth-leading cause of disability globally. Anxiety disorders are also associated with elevated risks of pain, cardiovascular disease (CVD) and premature mortality (2), emphasizing the importance of finding effective treatment strategies that can be addressed in primary care.
Specific research questions:
Significance and scientific novelty: Despite the high burden of anxiety, there are major knowledge gaps when it comes to secondary preventive measures. This intervention aims to reduce anxiety, improve the quality of life, work ability and health in primary health care patients with anxiety disorders. Little research has focused on lifestyle improvements or exercise specifically in the treatment of anxiety. In Sweden there are guidelines from The National Board of Health and welfare for treatment of depression with exercise, but guidelines for treatment of anxiety disorders are currently lacking. Also results from this study may extend the current understanding and practice in the field of treatment and secondary prevention of anxiety disorders within primary care.
This project aims to combine evidence-based interventions to tailor a lifestyle intervention with different steps focusing on patient control and participation since the individual chooses the intervention arm. Two evidence-based lifestyle interventions will combined and evaluated: the modified eHealth Lift/Care manager according to regional clinical guidelines and PHYSBI, a promising new strategy with an intensive training program, a new approach for treating anxiety disorders. If a simple intervention program in primary care can alleviate anxiety, improve quality of life, cognitive performance, and work ability, prevent pain and hypertension, this may have a great positive impact both at the individual level and from a societal perspective. The investigators hope to increase the availability of person-centered, evidence-based and cost-effective measures to promote both physical and mental health in patients with anxiety disorders. Moreover, the interventions are designed to easily be implemented in regular primary care activities.
Design of study: A clinical intervention with a matched design with two parallel arms and three assessment points (baseline, post-intervention and 1 year of follow-up).
Setting: Primary care centers in Region Västra Götaland (VGR), Sweden Participants: Patients aged 18-65 diagnosed with anxiety disorders attending 20 primary care centers (PCCs) within the VGR. Participants will be recruited by personnel at the PCCs (doctors, psychologists, nurses), through information material at the PCCs (posters, flyers and handouts) and by using the primary care system Medrave to identify and contact suitable patients. Eligibility criteria can be found elsewhere in the ClinicalTrials.gov protocol. Participants will sign a written informed consent and informed about the possibility to withdraw from the study at any time without disadvantages.
Matching: PCCs connected to a rehabilitation clinic will be assigned as intervention PCCs. Thsese PCCs will be matched for the following variables 1) size of the PCC, 2) socioeconomy as Care Need Index (CNI) and 3). Matching will be executed in order to identify suitable PCCs with care as usual (controls).
The intervention: The 10 PCCs allocated to intervention will offer all patients with a registered anxiety diagnosis in the PCCs electronic patient register as well as patients newly diagnosed with any of the above listed disorders to participate in an intervention program. The patient chooses the intervention arm, and the focus is on person-centredness, control, empowerment and participation.
The two intervention-options are:
CAU (PCCs allocated to non-intervention): At Swedish PCCs, care as usual (CAU) for patients with anxiety disorders can consist of visits to various healthcare professionals (such as physicians, nurses, psychologists, therapists, physiotherapists). The treatment should follow the regional guidelines, which are based on the national evidence-based clinical guidelines for anxiety disorders (National guidelines for care of depression and anxiety disorders, The Swedish National Board of Health and Welfare, Stockholm 2020). These guidelines include psychopedagogic support, psychotherapy (cognitive behavioural therapy (CBT) face-to-face or internet-mediated), psychopharmaceutical treatment and/or sick listing (preferably partial).
Outcome measures: Individual data on symptoms of anxiety, depression, general health, pain, sleep, fatigue, work ability, physical activity and perceived quality of life will be obtained through established self-assessment scales and collected for each intervention arm. Patients will self-report smoking, concurrent illnesses, pain, usage of prescribed drugs and physical activity using a questionnaire designed by the research team. Data will also be obtained from the healthcare database VEGA (Database for health-care consumption in VGR), the national Patient register, the regional prescribed drug register Digitalis, the national Prescribed drug register and the MiDAS database at Försäkringskassan (includes data on sick leave >14 days). Specification of outcomes can be found elsewhere in the ClinicalTrials.gov protocol.
Timeline and Workplan: During 2024 a Clinical Trial registration will be submitted. The investigators already have established contacts with gyms, the PHCRs and the Research and Development primary healthcare centres. During 2024 the collaborations with the primary healthcare organization, the PHCRs and health centres will be further established. A steering committee will be formed and the study procedures tested. During autumn 2024 the PCCs will be invited to participate. At the intervention PCCs, the recruitment of patients will be facilitated through a research nurse. At every PCC, a unique care manager manager which has been implemented at all PCCs in VGR since 2016 (4) handles the modified eHealth Lift, and the physiotherapists implement PHYSBI. The PHCRs will perform the PHYSBI arm and the involved PCCs will implement the modified eHealth Lift arm continuously during the intervention year of 2025. Study start date is anticipated to be in Jan 2025 and primary completion date in Dec 2026 (1 year of follow-up). Outcome data will be collected throughout the period 2025-2026. In 2026-2027 data will be processed, analyzed and scientific papers produced.
Data analysis and statistics: The majority of outcome variables are scale variables that may be treated as continuous variables and therefore primarily modelled using linear regression. These are scales for symptoms of anxiety, depression, pain, sleep, general health, perceived quality of life, sick leave days, number of health care contacts and number and types of psychopharmaceuticals prescribed. Data will be presented using descriptive statistics, including means and standard deviations for continuous variables, and frequencies and percentages for categorical variables Group differences by intervention status (modified eHealth Lift, PHYSBI, or CAU) will be assessed by Kruskal-Wallis test and chi-square tests, respectively. The average change in continuous outcomes will be analyzed using linear regression, with intervention as categorical predictor and baseline levels as covariates.
In the case of count data the investigators will test whether Poisson or negative binomial models are more appropriate than linear regression. The correlation of outcomes between patients from the same PCC will be modelled using a hierarchical linear model including PCC as random effect. A more detailed analysis of the longitudinal change will be based on repeated-measure models that also account for the correlations within individuals. Interaction terms for PCC status and time will test whether time trends differ by intervention status. Binary endpoints such as sick leave will be analyzed using logistic regression with adjustment for correlations as described for the linear models. All regression models will be adjusted for age, sex, as well as for variables on PCC-level such as socioeconomy, number of patients listed per PCC (size), and proportion of patients with anxiety diagnoses.
Power considerations: A power calculations has been performed based on 20 health centres distributed 50:50 i.e. 10 controls and 10 with two interventions. However, as patients are allowed to choose between the two intervention types (modified eHealth Lift or PHYSBI) the investigators will consider an unequal distribution of intervention types, e.g. 70:30 % for the modified eHealth Lift and PHYSBI. As the main outcome measures are continuous, Cohen's d = mean difference/standard deviation is considered as outcome measure, assuming small to medium effect size, i.e. d = 0.2 - 0.5 [51]. Power analysis was performed using a STATA procedure for one-way ANOVA. For a change (baseline to 3 months) effects sizes with d ≥ 0.2 and a power of ≥0.9 (significance level = 0.05) can be observed with a total of 320 individuals, distributed as 160 patients without intervention, 112 with the modified eHealth Lift, and 48 with PHYSBI. The same sample size will yield even larger statistical power if the group size for the modified eHealth Lift and PHYSBI were more alike.
Health-economic analyses: Cost-effectiveness analyses will be performed separately, comparing the two intervention arms to no CAU, including both direct (healthcare and pharmaceutical utilization) and indirect (long-term sickness and absenteeism from work) costs. Cost-effectiveness analyses will be based on the primary data collected within the project and based on a combination of primary data and secondary data from other sources for lifetime time perspectives. The long-term analyses will be performed using a health-economic simulation model projecting the outcomes and cost-effectiveness at the 1-year follow-up to a longer time period (5 years and lifelong).
Inclusion Criteria:
Exclusion Criteria:
maria.aberg@gu.se+46 (0)709668274
jenny.nyberg@neuro.gu.se+46 (0)702180801
Gothenburg, 41320, Sweden
maria.aberg@gu.se+46 (0)709668274
jenny.nyberg@neuro.gu.se+46 (0)702180801