Effect of exercise therapy and self-management support on multimorbidity: Secondary analysis of the MOBILIZE trial

Abstract

Background

Multimorbidity is linked to systemic low-grade inflammation, poor glycaemic control, dyslipidaemia, and hypertension, yet evidence on effective interventions is limited. We evaluated the impact of a 12-week personalised exercise therapy and self-management support programme, in addition to usual care, on these outcomes in individuals with multimorbidity.

Methods

This was a pre-planned secondary analysis of the MOBILIZE multicentre randomised controlled trial (NCT04645732). Participants (n = 228) had at least two of the following conditions: knee/hip osteoarthritis, chronic obstructive pulmonary disease, heart disease, hypertension, type 2 diabetes, or depression. The intervention included 24 supervised 60-minute group-based exercise sessions and 24 self-management sessions over 12 weeks. Outcomes were assessed at baseline and 4 months, including interleukin-1 receptor antagonist (IL-1ra), high-sensitivity C-reactive protein (hs-CRP), tumour necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), glycated Hemoglobin (HbA1c), fasting glucose, insulin, High-Density Lipoprotein (HDL), Low-Density Lipoprotein (LDL), triglycerides, and blood pressure.

Results

Compared to usual care, the intervention group shows a statistically significant reduction in systolic blood pressure (mean difference: −4.7 mmHg, 95% CI: −8.8 to −0.6). No significant between-group differences are observed for other biomarkers, although favouring the intervention. Sensitivity analyses—excluding participants with low adherence, those receiving supervised exercise in the control group, or undergoing surgery—support the primary findings.

Conclusions

A 12-week personalised exercise and self-management programme reduces systolic blood pressure in people with multimorbidity. These findings support incorporating exercise therapy into multimorbidity care guidelines as a non-pharmacological adjunct.

Plain language summary

We studied whether adding exercise and self-management support to usual care helps people with multiple long-term conditions. In this trial, participants followed a 12-week programme with supervised exercise and self-management group sessions. We found that the programme lowered blood pressure compared to usual care alone. However, it did not significantly change other health markers like blood sugar, cholesterol, or inflammation. These results suggest that exercise therapy may be a helpful addition to care for people with multiple long-term conditions, especially for managing blood pressure. It also highlights the need for more research on exercise therapy for this complex group.

Bricca et al. present secondary analysis of the MOBILIZE trial examining whether adding a 12-week exercise and self-management program to usual care benefits people with multiple long-term conditions. The program lowers systolic blood pressure compared to usual care but does not significantly change other biomarkers.

Introduction

Multimorbidity, defined as the co-occurrence of two or more chronic conditions, poses a significant health burden globally, affecting a third of the adult population¹. Individuals with multimorbidity are characterised by elevated levels of systemic low-grade inflammation, impaired glycaemic control, dyslipidaemia, and elevated blood pressure, which are linked to accelerated progression of chronic conditions, cardiovascular events, disability, reduced quality of life, and increased healthcare costs^2–11. Yet, despite the strong link between these biomarkers and multimorbidity^12,13, little is known about the effect of interventions on these outcomes^1,14.

Exercise therapy is a first-line treatment for individuals with single chronic conditions such as type 2-diabetes, osteoarthritis, hypertension, cardiovascular disease, chronic obstructive pulmonary disease, depression, and obesity^15–19. Systematic reviews suggested that exercise therapy may improve health outcomes also in people with multimorbidity with low- certainty of evidence^18,19, and our recent trial (MOBILIZE) improved the credibility of the systematic review findings by showing that exercise therapy is safe and has long-term benefits on health-related quality of life in people with multimorbidity²⁰. However, the effect of exercise therapy on biomarkers of multimorbidity has not been investigated in randomised controlled trials¹, with only two observational studies focusing on this topic^21,22. They both showed that higher levels of exercise is associated with lower levels of systemic low-grade inflammation and glycaemic control, lipids, and blood pressure in people with multimorbidity.

Exercise therapy, both aerobic and strengthening, may positively influence biomarkers of multimorbidity through several mechanisms. Notably, its anti-inflammatory effects help reduce oxidative stress and modulate immune responses contributing to improved cardiovascular and metabolic health^23,24. Exercise also targets key pathophysiological processes specific to each condition. For instance, in people with type 2-diabetes, it enhances insulin sensitivity and promotes glucose uptake by skeletal muscles, thereby lowering blood glucose levels²⁵. In people with hypertension, regular aerobic exercise can reduce sympathetic nervous system activity and improve endothelial function, leading to decreased vascular resistance and lower blood pressure²⁶.

The dose of exercise that may improve biomarkers of multimorbidity appears high. For instance, in people with type-2 diabetes, the DOSE-EX study showed a signal consistent with beneficial effects on markers of low-grade inflammation only with exercise performed six times per week, but not three times per week²⁷. Similarly, the U-TURN trial showed that five to six aerobic sessions per week, over 12 months, at a moderate to high intensity decreased low-grade systemic inflammation but not other biomarkers^28,29. In contrast, other biomarkers reflecting, for example, glycaemic control and anthropometric outcomes, appear to respond to lower exercise doses, as demonstrated in both the U-TURN and DOSE-EX studies^30,31. Additionally, a systematic review of RCTs including older adults with at least one chronic condition showed that at least 3 times a week of strength training at ≥70% of 1 repetition maximum could decrease biomarkers of inflammation³². Together, these effects highlight the therapeutic potential of exercise as a core component in the management of complex chronic conditions¹⁸. Additionally, from a feasibility and implementation perspective, it is crucial to determine whether smaller, more achievable doses than 5–6 sessions per week can still provide meaningful benefits, as this would greatly influence real-world adoption and sustainability.

Therefore, in this pre-planned secondary analysis of the MOBILIZE trial, we aimed to assess whether a 12-week personalised exercise therapy and self-management programme in addition to usual care, compared to usual care alone, can reduce systemic low-grade inflammation, glycaemic control, lipids, and blood pressure in people with multimorbidity. Based on the available evidence, we hypothesised that the exercise therapy and self-management programme would improve biomarkers related to inflammation, glycaemic control, lipids, and blood pressure to a greater extent than usual care alone.

Methods

The statistical analysis plan for this pre-planned secondary analysis of the pragmatic, assessor-blinded, multicenter, parallel-group RCT (1:1 treatment allocation) was made available in OSF (See Supplementary Statistical Analysis Plan) before any analyses commenced, and the reporting of this study follows the CONSORT Statement for Randomized Trials of Nonpharmacologic Treatments³³. The protocol of the MOBILIZE RCT and the statistical analysis plan have been published elsewhere^34,35, ClinicalTrials.gov registration: NCT04645732 (27/11/2020—First Posted). Data in the MOBILIZE trial were assessed at baseline, 4 months (questionnaire, physical test, and biomarkers), 6 months (only questionnaire), and 12 months (questionnaire and physical test) after randomisation. This pre-planned secondary analysis includes data from the baseline and 4-month assessment, where the biomarkers were collected.

The study was approved by the Regional Committees on Health Research Ethics for Region Zealand (SJ513 857), the Danish Data Protection Agency (Region Zealand, Denmark, REG-015-2020) and pre-registered at ClinicalTrials.gov (NCT04645732).

Equity, diversity, and inclusion statement

The author group is gender-balanced, consisting of four women and three men. Five members are researchers: two are early-career, two are mid-career, and one is a senior researcher. The remaining two members are clinicians. The group’s professional backgrounds include one exercise physiologist, one biostatistician, one medical doctor, and four physiotherapists. All members are affiliated with three different research groups based in Denmark. In the trial, all patients adhering to the eligibility criteria, regardless of sex and gender, were included, but we did not plan or conduct a formal analyses related to sex or gender (see Recruitment and Retention section).

Patient and public involvement

Patient and public involvement (PPI) has been central in all phases of the MOBILIZE project. Throughout, a group of up to eight patients with multimorbidity and carers were involved in key meetings and decisions. They shared their experiences, needs, and preferences, and helped shape the intervention, recruit participants and co-develop, feature in and ensure the clarity of the information communicated from the MOBILIZE project. Our approach followed the “Collaborate” level on the IAP2 Spectrum of Public Participation, emphasizing active partnership. The PPI is described in full in a separate publication²⁰.

Patients

We enrolled adult patients aged 18 or older with multimorbidity, defined as having at least two of the following conditions: knee or hip osteoarthritis (OA), chronic obstructive pulmonary disease (COPD), heart disease (heart failure or coronary heart disease), hypertension, type 2 diabetes mellitus (T2DM), and depression. Patients with additional comorbidities were not excluded. Participants had to meet several eligibility criteria.

Inclusion criteria:

• Ability to walk three metres without assistance.

• A score of ≥ 3 on the Bayliss Disease Burden: Morbidity Assessment by Self-Report scale for at least one of the listed conditions and a score of ≥ 2 for at least one other listed condition.

• Willingness and ability to participate in a 12-week supervised exercise therapy and self-management program twice a week.

Exclusion criteria:

• Participation in supervised systematic exercise for any of their conditions within the last 3 months.

• Having an unstable health condition or being at risk of serious adverse events as assessed by a medical specialist.

• Being terminally ill or having a life expectancy of less than 12 months.

• Classified as Class IV on the New York Heart Association (NYHA) Functional Classification scale, as the benefits and harms of exercise in this population are uncertain.

• Having psychosis disorders, post-traumatic stress disorder, obsessive-compulsive disorder, attention deficit hyperactivity disorder, autism, anorexia nervosa/bulimia nervosa, and/or dependency disorders.

• Other reasons for exclusion included inability to understand Danish or being mentally unable to participate.

These exclusion criteria were based on a participant’s inability to provide informed consent or safely participate in the intervention.

Recruitment and retention

The recruitment and retention strategy is described in full elsewhere³⁴. Briefly, Participants were recruited from various healthcare facilities in the Region of Zealand, Denmark, including general practitioners, psychiatric facilities, and hospital departments, as well as through self-referral. Recruitment methods included direct consultations, social media ads, local newspaper articles, posters, and handouts. Eligible individuals visiting recruitment sites were invited to participate, and patient records were reviewed to identify potential participants who were then contacted by phone. Self-referrals were assessed for eligibility by a project team member and a medical specialist. Once participants agreed to join, written informed consent was obtained before enrollment.

Blinding

The biomarkers assessors and the research assistant handling the data were blinded to the randomization.

Interventions

Briefly, the MOBILIZE RCT investigated the effects of a personalised exercise therapy and self-management support programme, in addition to usual care in patients with multimorbidity. 228 participants were randomised (1:1) to a 12-week intervention, in addition to usual care or usual care alone. The programme included 24 supervised exercise sessions (60 min/session) and 24 self-management sessions (30 min/session), and primarily focused on empowering patients to enhance their ability to engage in their own care more actively. By doing so, the program seeks to reduce symptoms associated with individual conditions, improve quality of life and physical function, and help prevent the onset of additional health problems. The intervention development is described in detail elsewhere^34,36. Briefly, the self-management support sessions aimed to improve self-management skills and motivation to maintain an active lifestyle and better quality of life after the program. The sessions covered topics such as sleep, pain management, and physical activity. These modules were designed to equip participants with practical tools to manage life with multiple chronic conditions better.

Each exercise session comprised a warm-up (8 min), balance exercises (5 min), strengthening exercises (20 min), a participant’s choice segment, which could include additional strengthening, aerobic, or functional exercises (20 min), and a cool-down (7 min). Physiotherapists guided participants to reach optimal exercise intensity for health benefits as recommended by the WHO³⁷, targeting a 12-14 level on the BORG scale for aerobic activities and a 5–7 level on the OMNI scale for strengthening or functional exercises^38–40. The strengthening regimen began with 2 sets of 10 repetitions in the first week, progressing to 3 sets of 12 repetitions by weeks 11 and 12, following ACSM guidelines⁴¹. The intervention was co-developed with a range of stakeholders, including a strong patient and carer involvement^34,36, and the exercise therapy programme can be found alongside the main paper of the RCT²⁰. Adherence and compliance were monitored throughout the intervention using diaries, kept at the intervention sites, for exercise therapy (including the “participant’s choice segment”) and self-management sessions. Adherence was assessed as the number of exercise and self-management support sessions attended out of the total number of sessions available. Compliance was assessed by the patients in the diaries by reporting the rate of perceived exertion in any set of the prescribed program using the OMNI scale for strengthening exercises and the BORG scale for aerobic exercises. The diaries were systematically collected and reviewed by the physiotherapists delivering the intervention to ensure accurate reporting of participation^36,42.

Outcomes

The participants were instructed to be fasting and not eat or drink anything except water for 10 h before they had their blood test. The concentration of the following molecular biomarkers were included: Interleukin-1 receptor antagonist (IL-1ra pg/ml), high-sensitivity C-reactive protein (hs-CRP mg/l), Tumour necrosis factor (TNF-α pg/ml), Interleukin 6 (IL-6 pg/ml), High-Density Lipoprotein (HDL) cholesterol (mmol/l) and Low-Density Lipoprotein (LDL) cholesterol (mmol/l), triglycerides (mmol/l), HbA1c (pg/ml), fasting glucose (mmol/l) and fasting insulin (mU/L). These biomarkers were included due to their relevance in evaluating the systemic effects of exercise on inflammation and cardiometabolic health. IL-1ra, TNF-α, IL-6, and hs-CRP are established mediators and markers of systemic low-grade inflammation, which can be modulated by exercise-induced immune and stress responses^23,24,43,44. HDL cholesterol, LDL cholesterol, and triglycerides also provide critical insights into lipid metabolism and cardiovascular health⁴⁵, while HbA1c, fasting glucose, and fasting insulin are central to assessing glycaemic control and insulin sensitivity⁴⁶.

The samples were stored at the local Departments of Clinical Biochemistry in Holbæk, Roskilde, Næstved, Slagelse, Nykøbing Falster, and Rigshospitalet in Copenhagen. Fasting plasma samples were centrifuged and stored at −80 ◦C until analyses. Markers of low-grade systemic inflammation were measured using immunoassay kits from Meso Scale Discovery, USA, as follows: IL-1Ra (V-PLEX Cytokine Panel 2 (human) kit), CRP (V-PLEX vascular injury Panel 2 (human) kit), TNF-α and IL-6 (V-PLEX Proinflammatory Panel1(human)).

Systolic and diastolic blood pressures were measured after 5 min of rest and reported as the mean of three consecutive measurements using the Rossmax X5 blood pressure monitor (Rossmax, Heerbrugg, Switzerland) at baseline and 4-month follow-up. Participants were asked to avoid smoking, consuming alcohol, or drinking coffee for half an hour before the blood pressure measurement.

Randomisation procedure

Participants who met the eligibility criteria and signed the informed consent form were randomised in a 1:1 allocation ratio following baseline assessment. The statistician had previously prepared a computer-generated randomisation schedule using permuted blocks of four or six individuals, stratified by the number of chronic conditions (2 or 3 + ) and recruitment centres. Allocation numbers were concealed in opaque sealed envelopes, only accessible to a study coordinator after informed consent and baseline assessment were completed.

Sample size

Sample size for the MOBILIZE trial was determined to detect a difference of 0.074 points between the two groups in the primary outcome (EQ-5D) from baseline to the 12-month follow-up³⁴. This study presents an exploratory, pre-planned analysis and therefore did not include a separate sample size calculation.

Statistical analysis

Between-group comparisons of change from baseline to the 4 months follow-up in continuous outcomes were analysed using repeated measures mixed-effects linear model, in line with the analysis of the primary outcome, reported in a separate publication²⁰. Visits (baseline and 4 months), treatment arm (Exercise therapy and self-management support programme versus usual care), and interaction between the 4-month visit and treatment arm were included as fixed effects. The interaction term is the main test of effect. The model was adjusted for the randomisation stratification factors (number of chronic conditions (2 or 3 + ) and recruitment centre (hospitals, general practitioners, and self-referrals) by including them as fixed effects. Throughout the report, 95% confidence intervals, standard deviation, and/or interquartile ranges are provided, as appropriate to demonstrate the precision of the estimates. All analyses were performed in STATA BE-Basic Edition version 18.5.

All randomised participants with at least one valid assessment of the outcomes of interest were included in this study. In the sensitivity analyses, the following participants were excluded: (i) participants in the exercise and self-management group participating in less than 18 out of the 24 self-management and exercise sessions (this was a pre-defined threshold chosen to deem the attendance to the intervention satisfactory)³⁴; (ii) participants in the usual care group participating in 12 or more supervised exercise therapy sessions for one of their diseases during follow-up or participants in both groups who underwent major surgeries and/or were hospitalised for more than seven consecutive days during the 4-month follow-up. Finally, we performed a sensitivity analysis imputing missing data with conditional mean imputation.

Results

Patient disposition

Between January 18, 2022, and May 30, 2023, 663 patients with multimorbidity were assessed. Of these, 228 were randomised, achieving a recruitment rate of 36%. Participants were allocated to either the exercise therapy and self-management support intervention group (n = 115) or the usual care group (n = 113). One individual in the intervention group withdrew the written consent, including the permission to use data. Therefore, the intervention group ultimately included 114 individuals. The flow for the inclusion of participants has been published in the paper reporting the findings on the primary outcome (i.e., health-related quality of life)²⁰. However, in Fig. 1 and Tables 1 and  2, we report the number of individuals with a valid assessment of each biomarker at baseline and the 4-month follow-up. Reasons for not having a valid assessment included not showing up at the blood sample collection, or the biomarkers not being stored at the right temperature, and biomarkers with a coefficient of variation, which measures the variability of the data, greater than 20% which is the threshold used by the U.S. Food and Drug Administration⁴⁷. The median session attendance for the exercise therapy and self-management support group was 76%. Non-attendance was mainly due to illness, vacations, or appointments with healthcare providers (e.g., planned hospital visits)²⁰. Participant characteristics were comparable between the intervention and usual care groups (Table 1). In the total group, the average age was 69.8 years (SD 8.4), with a mean BMI of 30.9 (SD 5.7), 43% were female, and patients had an average of 7 chronic conditions (SD 3, range 2–19).

Fig. 1. Flow diagram.

Flow of study patients.

Table 1.

Baseline characteristics. Data are reported as median and interquartile range, unless otherwise stated

		Exercise and self-management	Usual care
	N		N
Age (mean and SD)	114	70.0 (8.7)	113	69.6 (8.1)
Body mass index (mean and SD)	114	31.6 (6.0)	113	30.3 (5.5)
Sex (%) Female Male	45	39%	53	47%
	69	61%	60	53%
Number of chronic conditions per individual (mean and SD)	114	7 (2.9)	113	7 (2.8)
Interleukin-1 receptor antagonist (pg/ml)	102	206.9 (145.0–394.9)	102	195.1 (148.4–337.7)
High-sensitivity C-reactive protein (mg/l)	103	2.40 (1.12–4.97)	103	1.91 (0.99–4.64)
Tumour necrosis factor-α (pg/ml)	103	1.26 (0.99–1.73)	102	1.19 (0.93–1.62)
Interleukin 6 (pg/ml)	101	1.38 (0.92–2.19)	101	1.41 (0.87–2.29)
High-Density Lipoprotein cholesterol (mmol/l)	105	1.3 (1.1–1.7)	104	1.4 (1.1–1.7)
Low-Density Lipoprotein cholesterol (mmol/l)	103	2.1 (1.4–2.9)	104	1.9 (1.4–2.6)
Cholesterol P (total from plasma) (mmol/l)	105	4 (3.5–5.1)	104	4 (3.5–4.8)
Triglycerides (mmol/l)	105	1.5 (1.0–2.1)	104	1.4 (0.9–2.4)
HbA1c (pg/ml)	105	42 (37–53)	104	43 (38–49)
Fasting glucose (mmol/l)	105	6.9 (6.3–8.5)	104	7.1 (6.4–7.9)
Fasting insulin (mU/L)	101	6.6 (4.1–9.4)	100	5.9 (3.3–10.7)
Systolic blood pressure (mmHg)	114	138 (126–150)	113	137 (124–150)
Diastolic blood pressure (mmHg)	114	80 (71–88)	113	78 (71–88)

pg/ml picograms per millilitre, mmol/l millimoles per liter, mU/L milliunits per liter, mmHg millimetre of mercury.

Table 2.

Outcomes at 4-month follow-up

Outcome	Change from baseline to 4 months in the intervention group (N, mean change and 95% CI)	Change from baseline to 4 months in the usual care group (N, mean change and 95% CI)	Between-group difference in mean improvement (crude) (95% CI)	Between-group difference in mean improvement (adjusted)a (95% CI)
Interleukin-1 receptor antagonist (pg/ml)	N = 90 –19.8 (−54.8 to15.2)	N = 79 25.4 (−32.6 to 83.4)	−40.7 (−105.7 to 24.3)	−41.1 (−106.1 to 23.9)
High-sensitivity C-reactive protein (mg/l)	N = 91 –0.90 (−3.59 to 1.78)	N = 81 0.60 (−0.31 to 1.51)	−1.56 (−4.30 to 1.18)	−1.55 (−4.29 to 1.19)
Tumour necrosis factor-α (pg/ml)	N = 91 –0.43 (−0.12 to 0.03)	N = 81 –0.0005 (−0.14 to 0.14)	−0.03 (−0.18 to 0.11)	−0.03 (−0.18 to 0.12)
Interleukin 6 (pg/ml)	N = 89 –0.38 (−1.09 to 0.33)	N = 79 –0.03 (−0.30 to 0.23)	−0.30 (−1.06 to 0.45)	−0.31 (−1.06 to 0.44)
High-Density Lipoprotein cholesterol (mmol/l)	N = 93 –0.02 (−0.06 to 0.03)	N = 84 –0.03 (−0.07 to 0.01)	−0.05 (−0.18 to 0.08)	0.01 (−0.05 to 0.08)
Low-Density Lipoprotein cholesterol (mmol/l)	N = 91 –0.02 (−0.18 to 0.13)	N = 82 0.06 (−0.05 to 0.16)	−0.11 (−0.30 to 0.08)	−0.11 (−0.29 to 0.08)
Cholesterol (total from plasma) (mmol/l)	N = 93 –0.09 (−0.26 to 0.07)	N = 82 0.06 (−0.05 to 0.16)	−0.11 (−0.30 to 0.09)	−0.11 (−0.30 to 0.09)
Triglycerides (mmol/l)	N = 93 –0.14 (−0.28 to 0.01)	N = 84 –0.03 (−0.13 to 0.07)	−0.09 (−0.27 to 0.09)	–0.09 (−0.27 to 0.09)
HbA1c (pg/ml)	N = 93 –0.81 (−0.164 to 0.02)	N = 84 –0.06 (–1.11 to 0.99)	–0.70 (–2.01 to 0.60)	−0.70 (–2.01 to 0.60)
Fasting glucose (mmol/l)	N = 93 –0.11 (−0.24 to 0.01)	N = 84 –0.02 (−0.18 to 0.12)	–0.08 (−0.27 to 0.10)	–0.08 (−0.27 to 0.10)
Fasting insulin (mU/L)	N = 89 –0.03 (−1.19 to 1.13)	N = 77 –2.19 (−7.16 to 2.78)	1.71 (–2.57 to 6.01)	1.71 (–2.58 to 6.01)
Systolic blood pressure (mmHg)	N = 98 –5.8 (−8.9 to −2.8)	N = 91 –1.19 (−4.0 to 1.7)	–4.6 (−8.7 to −0.6)	−4.7 (−8.8 to −0.6)
Diastolic blood pressure (mmHg)	N = 98 –3.3 (−6.0 to −0.6)	N = 91 –1.2 (−2.8 to 0.3)	−2.2 (−5.3 to 0.8)	−2.4 (−5.5 to 0. 7)

SD standard deviation, mmHg millimetres of mercury.

^aAdjusted for the randomisation stratification factors (number of chronic conditions (2 or 3 + ) and recruitment centre (hospitals, general practitioners, and self-referrals) by including them as fixed effects.

Effect of exercise therapy and self-management support on systemic low-grade inflammation, glycaemic control, lipids, and blood pressure

The between-group analysis identified a statistically significant change in systolic blood pressure between groups from baseline to 4 months (Table 2, Fig. 2). The mean (95% CI) differences in change were -4.7 (mmHg) (-8.8 to -0.6) in favour of the exercise therapy and self-management group in both the crude and adjusted analyses (Table 2). The difference in change between groups for the other outcomes did not reach statistical significance, although most favoured the exercise therapy and self-management group (Fig. 2).

Fig. 2. Spider graph summarising the within-group changes in the outcomes between baseline and the 4-month follow-up. Values closer to −1.0 represent a greater improvement.

The red line represents no within-group change. The values on the axes range from −1.0 to 0.0 and are the normalised means (z-scores) of the biomarkers. The scale for HDL cholesterol has been inverted to ensure that an increase in HDL is considered a beneficial improvement. BP: Blood Pressure; HbA1c: Glycated Haemoglobin, Glucose: Blood Glucose Levels, CholeP: Total Cholesterol; HDL: High-Density Lipoprotein Cholesterol, LDL: Low-Density Lipoprotein Cholesterol, hs-CRP: High Sensitive C-Reactive Protein, TNF-α: Tumour Necrosis Factor, IL6: Interleukin 6, IL1-RA: Interleukin 1 Receptor Antagonist, Systolic: Systolic Blood Pressure, Diastolic: Diastolic Blood Pressure.

The sensitivity analyses, including only individuals attending 75% or more of the intervention sessions (Supplementary Table 1), without individuals in the usual care group that received supervised exercise therapy sessions or without individuals that in both groups undergo surgery in the 4-month follow up (Supplementary Table 2).and, the analysis with the imputed dataset, were in line with the main analyses (Supplementary Table 3).

Discussion

This is the first study to investigate the effects of personalised exercise therapy and self-management support, in addition to usual care, on biomarkers related to systemic low-grade inflammation, glycaemic control, blood lipids, and blood pressure in individuals with multimorbidity. Although the MOBILIZE intervention was not designed to target biomarkers of multimorbidity primarily, and as this secondary analysis was not powered for between-group differences on the biomarkers, we found a statistically significant reduction in systolic blood pressure in the intervention group in addition to usual care as compared to the usual care group alone. Additionally, IL-1ra, hs-CRP, IL-6, LDL, triglycerides, HbA1c, fasting glucose, and diastolic blood pressure also favoured the intervention group but without demonstrating a statistically significant difference from usual care.

Systolic blood pressure decreased 4.7 mm Hg (95% CI −8.8 to −0.6) more in response to exercise therapy and self-management support than usual care alone. A reduction of 5 mm Hg in systolic blood pressure is associated with approximately a 10% decrease in the risk of major cardiovascular events⁴⁸. This reduction also corresponds to decreases in the risk of stroke (13%), heart failure (13%), ischaemic heart disease (8%), and cardiovascular death (5%)⁴⁸. The effect of exercise therapy on blood pressure has been investigated in several studies. Our finding aligns with previous studies investigating the effect of exercise therapy on blood pressure. However, given that the 95%CI of the between-group difference covers a reduction smaller than the 5 mm Hg, the clinical relevance of our findings needs to be interpreted with caution. The two most extensive systematic reviews on exercise and blood pressure included 197 and 270 trials, respectively^17,49, showed that different forms of exercise therapy, such as aerobic and resistance exercise, reduced systolic blood pressure by about 5 mm Hg. However, they also highlighted that most trials (72%) excluded individuals with high blood pressure ( ≥ 140 mmHg), and none included people with multimorbidity, despite its prevalence and burden¹. Therefore, our results add significantly to the available evidence by providing the foundation for generalising these results to people with multimorbidity.

Despite the reductions in the intervention group’s biomarkers of systemic low-grade inflammation, glucose, lipids, and diastolic blood pressure, these changes did not reach statistical significance. This includes biomarkers such as IL-1ra, hs-CRP, TNF-α, IL-6, HbA1c, fasting glucose, fasting insulin, HDL, LDL, triglycerides, and diastolic blood pressure. The absence of statistically significant changes could be due to the study’s sample size and the duration, intensity and dose of the intervention. While the intervention was intended to be performed at moderate intensity, as recommended by the WHO for maximising general health benefits from exercise, the dose may have been too low to elicit measurable changes in the biomarkers we included^23,28,29. Trials with adults with type 2 diabetes, with baseline values of systemic low-grade inflammation comparable to our study, have shown that only a high volume of exercise, including five to six aerobic exercise sessions per week for 12-months, decreased low-grade systemic inflammation measured with IL-1ra. Yet, no changes were observed in CRP, TNF-α, IL-6, HbA1c, triglycerides, total cholesterol, HDL, LDL, and diastolic blood^28,29. Additionally, the observed biomarker changes align with improvements in objectively measured physical function and activity in the MOBILIZE trial²⁰, which improved in the intervention group, without reaching statistical significance²⁰. Given the strong link between systemic inflammation, disability and physical activity, this may partly explain the lack of statistically significant reduction of low-grade systemic inflammation biomarkers¹². Finally, hypertension was highly prevalent in our study population²⁰, making systolic blood pressure a shared and clinically relevant risk factor across conditions. This may aso explain why significant reductions were observed in systolic blood pressure, whereas other biomarkers despite numerical improvements were more heterogeneous, better regulated at baseline, or less responsive within the intervention period.

The main limitations in terms of detecting effects on biomarkers of this pre-specified secondary analysis of the MOBILIZE trial²⁰, are: (i) the lack of statistical power to detect between group differences in the biomarkers of interest and, (ii) the lack of data on changes in medication during the trial, which could either mask a higher intervention effect (if reduced) or explain the observed reduction in blood pressure (if intensified). Although we collected medication information, reporting was highly inconsistent, preventing meaningful classification or subgroup analyses. However, as our participants were randomised and all received the usual care, the risk of this being a significant problem was kept at a minimum.

We recommend that future studies investigate the feasibility and effectiveness of exercise therapy interventions with higher doses and longer follow-up (e.g., greater intensity and duration for 12 months) on biomarkers related to systemic low-grade inflammation, glycaemic control, lipids, and blood pressure. Additionally, these studies should explore the mechanisms underlying the observed effects and identify which subgroups may benefit most. Given that individuals with multimorbidity have more complex needs than those with single chronic conditions¹, a co-development approach focusing on creating an exercise regimen with a higher dose that can be integrated into their daily lives is warranted⁵⁰. Furthermore, future research should examine dose-response relationships and incorporate device-based monitoring such as heart rate sensors, VO₂ measurement devices, accelerometers, and smart resistance equipment that allow to record load, repetitions, and velocity during strength training, to accurately capture exercise intensity, exertion, and adherence.

In conclusion, our results suggest that personalised exercise therapy and self-management support in addition to usual care, is more effective than usual care alone in decreasing systolic blood pressure at 4 months in adults with multimorbidity, but not diastolic blood pressure, or biomarkers related to systemic low-grade systemic inflammation, glycaemic control, or lipids. While the results on systolic blood pressure may be considered clinically relevant, future fully powered studies are needed to confirm these results.

Author contributions

Study conception and design: A.B., M.N., G.E.L., M.D., G.Z., L.C.T., S.T.S.,. and Data acquisition: M.N., M.D., Data analysis: A.B., G.E.L., L.C.T. Drafting of the article: A.B. and S.T.S. Critical revision of the article: A.B., M.N., G.E.L., M.D., G.Z., L.C.T., S.T.S. Final approval of the article: A.B., M.N., G.E.L., M.D., G.Z., L.C.T., S.T.S.

Peer review

Peer review information

Communications Medicine thanks Francesco Bettariga and Vincent R. Singh for their contribution to the peer review of this work. A peer review file is available.

Data availability

De-identified data and data dictionaries from the MOBILIZE study are available from the principal investigator (Prof. Søren T. Skou, stskou@health.sdu.dk) after publication of the primary publications and until 5 years after the publication of this manuscript. However, restrictions apply to the availability of the de-identified data due to GDPR and study-specific regulations, and access requires a data sharing agreement and a research proposal that will be evaluated by the study group. Requests to access data can expect to be answered within 3 months.

Competing interests

S.T.S. is the co-founder of Good Life with Osteoarthritis in Denmark (GLA:D®), a non-profit initiative hosted at the University of Southern Denmark. This initiative aims to implement clinical guidelines, including exercise and self-management support, for individuals with osteoarthritis in clinical practice. The authors affirm that they have no other competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s43856-026-01479-9.