Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Objectively measured physical activity levels and adherence to physical activity guidelines in people with multimorbidity—A systematic review and meta-analysis

  • Lars Bo Jørgensen ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark, Department of Physiotherapy and Occupational Therapy, Zealand University Hospital, Roskilde, Region Zealand, Denmark, Department of Sports Science and Clinical Biomechanics, Research Unit for Musculoskeletal Function and Physiotherapy, University of Southern Denmark, Odense, Denmark

  • Alessio Bricca,

    Roles Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark, Department of Sports Science and Clinical Biomechanics, Research Unit for Musculoskeletal Function and Physiotherapy, University of Southern Denmark, Odense, Denmark

  • Anna Bernhardt,

    Roles Conceptualization, Data curation, Investigation, Validation, Writing – original draft, Writing – review & editing

    Affiliation Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark

  • Carsten B. Juhl,

    Roles Conceptualization, Data curation, Formal analysis, Methodology, Software, Writing – original draft, Writing – review & editing

    Affiliation Department of Sports Science and Clinical Biomechanics, Research Unit for Musculoskeletal Function and Physiotherapy, University of Southern Denmark, Odense, Denmark

  • Lars Hermann Tang,

    Roles Conceptualization, Methodology, Writing – original draft, Writing – review & editing

    Affiliations Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark, The Department of Regional Health Research, University of Southern Denmark, Odense, Denmark

  • Sofie Rath Mortensen,

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliations Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark, Department of Sports Science and Clinical Biomechanics, Research Unit for Exercise Epidemiology, Centre of Research in Childhood Health, University of Southern Denmark, Odense, Denmark

  • Jonas Ahler Eriksen,

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliation Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark

  • Sisse Walløe,

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliations Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark, Department of Clinical Research, Research Unit OPEN, University of Southern Denmark, Odense, Denmark

  • Søren T. Skou

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Physiotherapy and Occupational Therapy, The Research Unit PROgrez, Næstved-Slagelse-Ringsted Hospitals, Slagelse, Region Zealand, Denmark, Department of Sports Science and Clinical Biomechanics, Research Unit for Musculoskeletal Function and Physiotherapy, University of Southern Denmark, Odense, Denmark



To determine levels of objectively measured physical activity (PA) and the proportion of adults with multimorbidity that adheres to PA guidelines.


All studies, where PA was measured at baseline using an activity monitor in an adult (≥18 years) multimorbid (≥80% of the population had ≥2 chronic conditions) population. A systematic literature search was performed in Medline, EMBASE, CINAHL, CENTRAL,, and from inception up until 18th of January 2022. Risk of bias was assessed with a modified version of the Quality Assessment Tool for Quantitative Studies. A random-effects meta-analyses was performed to estimate daily minutes of sedentary behavior (SB), light PA (LPA), moderate PA (MPA), moderate to vigorous PA (MVPA) and steps. Proportions adhering to PA guidelines was narratively synthesized. Certainty of evidence was determined using The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.


Fifteen studies (2,172 participants) were included. The most frequent combination of conditions were type 2 diabetes and hypertension (six studies). Participants spent a daily average of 500.5 (95% CI: 407.1 to 593.9) minutes in SB, 325.6 (95% CI: 246.4 to 404.7 minutes in LPA and 32.7 (95% CI: 20.2 to 45.3) minutes in MVPA. The mean daily number of steps was 5,145 (95% CI: 4264 to 6026) for people in free-living conditions. The proportion adhering to PA guidelines ranged widely (7.4% to 43%). All studies were rated as at high risk of bias and the certainty of evidence was very low.


PA levels and adherence varied from low to above guideline recommended levels for adults with chronic conditions, depending on PA intensity. The very low certainty of evidence calls for high quality studies focusing on detailed descriptions of PA behavior in people with multimorbidity.

PROSPERO registration number


1. Introduction

The prevalence of people having two or more chronic conditions—commonly referred to as ‘multimorbidity’—is estimated to be 33% in the general population [1, 2] and more than half of all adults with a chronic condition are multimorbid [3]. Multimorbidity is not only a burden for the individual, but also for society with an almost exponential association between the number of chronic conditions and associated health care costs [4]. Multimorbidity is considered to be the next global health priority [5], due to an expected increase in prevalence of multimorbidity in the near future [6], impacting the lives of millions of people worldwide [7].

Physical inactivity represents an important target in the growing burden of multimorbidity as it increases the risk of poorer health, development of further chronic conditions [8] and death [9]. It is well known that physical activity (PA) and exercise are effective in preventing and treating chronic conditions such as type 2 diabetes, heart and pulmonary diseases and osteoarthritis [8, 10]. Furthermore, PA and exercise appear both safe and beneficial in improving physical and psychosocial health in people with multimorbidity [11, 12].

The World Health Organization (WHO) recommends that adults with chronic conditions should perform at least 150–300 minutes of moderate-intensity aerobic PA, 75 to 150 minutes of vigorous-intensity PA (or an equivalent combination) per week [13]. However, despite the well-documented benefits of PA, many people do not adhere to PA guidelines [14]. This proportion increases with the number of chronic conditions i.e. the level of PA is lower among people with multimorbidity [1517]. A recent study found that nearly 68% of older adults with multimorbidity fail to meet the level of PA recommended by the WHO [18].

Unfortunately, and in spite of emerging evidence there is still limited knowledge on PA level in people with multimorbidity. Furthermore, no overview of the level of objectively measured PA and adherence to PA guidelines is available, although objectively measured PA is considered more accurate and less susceptible to bias than self-reported PA [1922]. Therefore, this systematic review and meta-analysis aimed to determine 1) the level of objectively measured PA in people with multimorbidity in free-living conditions and 2) the proportion that adheres to the WHO PA guidelines for people with chronic conditions.

2. Methods

This review followed the recommendations for performing systematic reviews as described by Cochrane [23]. Reporting of the review was done in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses 2020 (PRISMA 2020) guidelines [24] (S1 File).

2.1 Eligibility criteria

2.1.1 Study design and participants.

Studies where PA was measured objectively (e.g., activity monitor such as a pedometer or accelerometer) in an adult multimorbid population (mean age ≥18 years) were included. Only baseline PA data was extracted. The population was considered multimorbid if ≥80% had two or more of the following chronic conditions; osteoarthritis (knee or hip), chronic heart disease (heart failure or ischemic heart disease), hypertension (systolic blood pressure ≥140 and diastolic blood pressure ≥90 and/or intake of anti-hypertensive medications), type 2 diabetes mellitus (T2DM), chronic obstructive pulmonary disease (COPD), depression, anxiety, or chronic low back pain (LBP). If multimorbidity was reported as a cumulative numerical score or an index score, they were excluded unless the number of conditions exceeded 80% of the conditions that we used to define multimorbidity in this review. These conditions were chosen as they are among the leading causes of global disability, affect hundreds of millions of people around the world, and often co-exist [25]. Studies were furthermore excluded if PA was measured for a period of less than 7 days or had reported less than 2 valid days of PA measurement.

2.2 Information sources

Literature was searched in the scientific databases Medline and EMBASE (via Ovid), CINAHL (via EBSCOhost) and CENTRAL, from inception up until 18th of January 2022. Grey literature was searched using a grey literature checklist developed by Godin and colleagues [26] on and via the webpages and Furthermore, reference lists of included studies were hand searched for eligible studies and citation tracking performed on included studies in Web of Science (WoS).

2.3 Search strategy

The search strategy was developed for Medline and then customized for the remaining databases (S2S5 Files). Search strategies were developed individually for, and (S6 File). No limits were set on language.

2.4 Selection process

Records were transferred to Covidence software [27] and duplicates removed using the Covidence software. Two reviewers (LBJ and AnB) independently screened records for eligibility on title and abstract and resolved any conflicts by discussion. For full text screening, the review team was expanded to include four additional members (AB, SM, SW and JE), forming three teams consisting of two reviewers. The remaining studies were full text screened for eligibility individually as follows (SW/JE: 60 studies, AB/SM: 60 studies and LBJ/AnB: 240 studies. Conflicts were resolved by discussion within the review team or by involvement of a more experienced senior review team-member (STS or CJ) if needed.

2.5 Data collection process

Data from included studies were extracted independently by two reviewers (LBJ and AB) using a data extraction form developed in Excel by LBJ. The extraction procedure was pilot tested on three studies and data compared among the reviewers in order to implement any adjustments to the data extraction form. No adjustments were made. Data were then extracted from the remaining studies and compared. Consensus on extracted data was reached through discussion. If multiple papers were published based on data from the same study, the paper with the largest sample size was used and other papers excluded.

2.6 Data items

Data was extracted regarding study characteristics, participant and outcome data.

Study characteristics included author, year, country of origin of the study, and study design (e.g., observational). Participant data included population number (n =), mean age, gender (percentage of female), body mass index (BMI), ethnicity and number of conditions, severity, and diagnosis.

Outcome data included type (e.g., pedometer or accelerometer), brand and placement of activity monitor, duration of the PA measurement (e.g., 7 days), number of valid days of measurement and definition of intensity level (e.g., cut point for moderate-vigorous PA (MVPA)). For continuous outcome (physical activity) data, mean and standard deviation, standard error, or 95% Confidence Interval was extracted of the following variables: minutes spent as sedentary behavior (SB), light PA (LPA), moderate PA (MPA), moderate to vigorous (MVPA), vigorous PA (VPA), daily steps or ‘other’ (i.e., activity counts, calories).

When PA was reported as weekly levels it was converted into daily levels by dividing it with the number of valid days of measurement reported in the individual study. In PA data presented with medians, these were considered equivalent to the mean, and interquartile ranges used to calculate standard deviations (SD) as recommended by Cochrane [23]. SDs of daily PA levels were converted to standard errors (SE) to perform the meta-analyses. If the exact number of participants that contributed with PA data was not reported in a study, the total number of the population was used in data conversions. In studies reporting two measures of PA (e.g., bouted and unbouted PA), the number of participants were split in two groups of smaller sample sizes, and results reported as two separate estimates as recommended by Cochrane [23].

The following items, although reported in the PROSPERO registration, were not presented, as they were not consistently reported in the retrieved studies: socioeconomic status, physical function, mental health, quality of life, wear time (hours per day), epoch length and cut point for non-wear time.

2.7 Study risk of bias assessment

Risk of bias was assessed using a quality assessment tool designed specifically for this systematic review (S7 File). The tool was inspired by the Quality Assessment Tool for Quantitative Studies, developed for use in the Effective Public Healthcare Panacea Project (EPHPP) ( Questions not considered relevant for the aims of this review were deleted. The deleted questions were: Rating of study design, Confounders, Withdrawals and drop-outs, Intervention integrity and Analyses. Furthermore, options to answer, and wording was altered to fit the aims of the review e.g., ‘Are the individuals selected to participate in the study likely to be representative of the target population? was replaced with ‘How representative was the study participants of the multimorbid target population? This is a common procedure and has been done in several previous studies [28]. Risk of bias was assessed through six questions divided into three sections; selection bias, blinding (performance bias) and data collection methods (measurement bias). Each section was rated as strong, moderate, or weak (blinding could only be rated strong or weak) based on ratings of the questions in the section. Each study was given a global rating based on the ratings of the three sections. A low-quality rating (one or more weak ratings) was considered high risk of bias, moderate-quality rating (only strong or moderate ratings) as moderate risk of bias and high-quality rating (only strong ratings) as low risk of bias. Risk of bias assessment was performed by two review teams consisting of four reviewers LBJ/SM and SW/JE. Each team member assessed studies independently and compared ratings within their review team. To assist the rating, a dictionary was incorporated after each section of the quality assessment tool explaining the rationale behind rating. This was inspired by the original assessment tool. In case of disagreements consensus was reached through discussion.

2.8 Effect measures

2.8.1 Estimate measures.

Average minutes of daily activity spent as SB, LPA, MPA, MVPA, VPA and steps was assessed in separate meta-analyses.

2.9 Synthesis of results

2.9.1 Physical activity.

Meta-analyses were performed for minutes of activity spent as SB, LPA, MPA, MVPA, VPA and daily steps in free-living conditions. A random-effects model was used given the heterogeneous population included. Results were presented in forest plots with overall estimates of the PA level including subgroup-analyses of bouted and unbouted PA. Subgroup-analyses were performed in order to investigate the possible differences in PA arising from analyzing data as bouted versus unbouted, which have been suggested in earlier studies [2931]. Statistical heterogeneity was assessed using I2 statistics and interpreted according to the Cochrane recommendations as a continuous measure where an I-squared value of 0% indicates no inconsistency, and an I-squared value of 100% indicates maximal inconsistency. All statistical analyses were performed using Stata 17 (StataCorp. 2021. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC) using the ‘meta’ command.

If studies did not report enough data to be included in meta-analysis and authors did not provide the necessary data, a narrative synthesis of the results was performed in accordance with the Cochrane recommendations [23].

2.9.2 Adherence to physical activity guidelines.

The proportion of people adhering to PA guidelines was summarized in a narrative synthesis as there were only three studies with heterogeneous populations reporting proportions, or insufficient data available to calculate proportions. Therefore, meta-analysis was deemed not appropriate in line with the Cochrane recommendations for performing meta-analysis [23].

2.9.3 Reporting bias assessment.

Authors of potentially eligible studies were contacted about outcome and population characteristics. For example, asking for measure of variance (e.g. standard deviation) regarding PA measures and prevalence of chronic conditions in the reported population when it was only reported that the population had multimorbidity. All contact to author was done by e-mail to ensure the greatest response rate possible [32]. Initially an e-mail was sent to the corresponding author. If an answer was not received within fifteen workdays, the last author was contacted, and the deadline extended with fifteen more days. If no reply was received, the study was excluded.

2.9.4 Sensitivity analysis.

Sensitivity analysis was performed by removing studies presenting median values to check the robustness of the findings, give that mean and median values were pooled in the same meta-analyses as recommended by Cochrane [23].

2.9.5 Certainty assessment.

The certainty (overall quality) of evidence was determined for each meta-analyses using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach [33] for prognostic studies, given the nature of the included studies and in line with Iorio et al. [34]. Five domains; risk of bias, inconsistency, indirectness, imprecision, and publication bias were assessed individually by two reviewers (LBJ and AB). In case of disagreements, consensus was reached through discussion.

3. Results

3.1 Study selection and characteristics

A total of 17,537 records were identified through databases and registers. Another 19 records were identified by searching the webpages and Eight additional records were identified by screening references of published systematic reviews focusing on multimorbidity.

Of 360 full text screened reports, 345 were excluded (S8 File) with the main exclusion criteria being non-multimorbid populations or non-peer reviewed reports. No further studies were identified through hand search of reference lists of included studies or WoS citation tracking.

Nine studies [3543] appeared eligible after full text screening but had not reported PA data in a format making it possible to extract for meta-analyses (i.e. missing SD or SE), or lacked data on the proportion of population that was multimorbid. Authors of these studies were contacted. Four authors replied providing additional information, leading to inclusion of three additional studies [36, 42, 43] and exclusion of one study due to the population not being multimorbid [35]. The flow of studies is presented in Fig 1.

Fifteen studies were included [36, 4255] with a total of 2,172 participants. Publication year ranged from 2011 to 2021. The most common study designs were randomized controlled trials [42, 45, 47, 48, 50, 51, 53] and observational studies [43, 44, 46, 49, 52, 53, 55]. The most frequent combination of conditions was type 2 diabetes (T2DM) and hypertension (six studies) [36, 43, 49, 50, 52, 55]. No studies had populations where more than 80% of the participants had more than two chronic conditions.

Accelerometers were the most commonly used measurement method (11 of 15 studies) [36, 4244, 4648, 5154], and steps the most frequent way to report the PA level (8 of 15 studies) [42, 45, 49, 50, 5255].

Three studies reported adherence (or data making it possible to calculate such) to physical activity guidelines [36, 52, 54]. Study characteristics of included studies are presented in Tables 1 and 2.

Table 1. Conditions and selected demographic variables in the included studies.

Table 2. Physical activity measurements in the included studies.

3.2 Risk of bias in studies

One third of the included studies were rated as ‘weak’ in all three parts of the risk of bias assessment, mainly due to studies failing to describe the validity and reliability of the measurement method used. Overall, all included studies were of low quality (high risk of bias) (see Table 3).

3.3 Results of syntheses

3.3.1 Physical activity level.

No studies reported minutes of PA spent at all intensity levels and daily steps. The pooled mean daily minutes spent in SB and LPA was 500.5 (95% CI: 407.1 to 593.9) and 325.6 (95% CI: 246.4 to 404.7) (Figs 2 and 3). The pooled mean daily minutes of MVPA was 32.7 (95% CI: 20.2 to 45.3) (Fig 4). MPA and VPA were only reported in one study [54] and meta-analysis of these activity levels therefore not performed.

Fig 2. Forest plot showing daily minutes of Sedentary Behavior (SB) in people with multimorbidity.

Fig 3. Forest plot showing daily minutes of Light Physical Activity (LPA) in people with multimorbidity.

Fig 4. Forest plot showing daily minutes of Moderate to Vigorous Physical Activity (MVPA) in people with multimorbidity.

The pooled mean of daily steps was 5145 (95% CI: 4264 to 6026) (Fig 5). One study, not included in meta-analysis, included hospitalized participants reporting a median daily number of steps of 1170 (IQR: 523 to 2580).

Fig 5. Forest plot showing daily number of steps in people with multimorbidity.

Two studies [46, 51] reported PA as other than minutes or steps (activity counts and active hours) (Table 2). All meta-analyses showed high heterogeneity and subgroups analyses on PA stratified as bouted or unbouted did not explain the inconsistency of the results for all the meta-analyses (Figs 24). The sensitivity analysis removing studies reporting median values displayed similar results as the main analyses (S1S3 Figs).

3.3.2 Adherence to physical activity guidelines.

The proportion that adhered to physical activity guidelines was reported in three studies and ranged from 7.4% to 43%. The populations had combinations of T2DM and hypertension (two studies) [36, 52] and depression and anxiety (one study) [54].

3.4 Certainty of evidence

Certainty of evidence was assessed as being very low for all meta-analyses (Table 4).

4. Discussion

Results from this systematic review and meta-analysis showed that the daily level of MVPA exceeded what is recommended by the WHO for people with chronic conditions, and that the few studies specifically reporting proportions that adhere to PA guidelines found wide variation in adherence (7,4–43%). Results also showed that participants spent on average more than eight hours per day in SB and had a number of daily steps equivalent to a low active lifestyle among adults [13, 56]. The results should be interpreted with caution since few studies evaluating objectively measured PA levels in multimorbid populations were available, and due to the high risk of bias of the included studies and very low overall certainty of evidence.

Although challenged by the lack of available evidence, higher levels of daily MVPA and LPA and lower levels of SB were found than previously reported in people with multimorbidity. In a Canadian cohort study, Hains-Monfette et al. [57] reported that people with two chronic conditions performed a daily average of 10 minutes of MVPA, 171 minutes of LPA and 580 minutes of SB. These results were based on an adult population with somewhat similar chronic conditions (heart disease, diabetes, and cancer and/or COPD) to the ones included in this review. Direct comparison of PA levels is, however, challenged by the general lack of data on factors that could possibly affect PA levels, such as the prevalence and severity of the included conditions which has previously demonstrated to be associated with PA [58].

The mean number of daily steps found in the meta-analysis in this study were similar to studies including healthy adult general populations from Western countries measuring steps in free-living conditions [59, 60]. Considering existing evidence suggesting that people with chronic conditions are less physically active than the general population [15], the number of daily steps were expected to be lower in populations with multimorbidity compared to the general population. Importantly, results from the present study showed that people with multimorbidity only attained a daily step level equivalent to the level of ‘low PA’ using earlier published cut off points for healthy adults of 5000–7499 daily steps [56]. It is likely that the contradictive findings between an adequate daily MVPA level and ‘low active’ daily step level could be explained by the difference in participants entering the meta-analyses. Only one study [52] reported data on both MVPA and steps and interpretation of the PA levels was therefore made on different sets of participants. The difference in PA levels could therefore possibly be explained by factors such as age, the specific chronic conditions, and the severity of their conditions [61]. Due to an insufficient number of available studies, it was not possible to perform meta-regression analysis to investigate whether these and other factors affected the results.

A previous large observational study from the UK Biobank found that almost half of people with multimorbidity can be categorized as having a low physical activity level [62]. A large variation in reported proportions of adherence to PA guidelines was observed in the narrative synthesis in this study (7,4–43%). This appears conflicting to the relatively high daily level of MVPA found in the meta-analysis. Such variation is similar to findings from previous studies of self-reported PA in people with multimorbidity [61] and has also been observed in the general healthy population [63]. This appears conflicting to the relatively high daily level of MVPA found in the meta-analysis. Wide variation in PA levels have however also been demonstrated in previous studies of self-reported PA in people with multimorbidity [61] and in the general healthy population [63]. Possible explanations could be the use of different measurement methods (self-reported vs. objectively measured PA) or that selection bias was introduced in the present study and could have affected the estimates. This is likely since the included populations displayed differences in characteristics such as age or specific chronic conditions. Last, functional status of the different populations were unknown since this data was not available or inconsistently reported. If the some of the populations included in the MVPA meta-analyses had higher functional status level, this could have resulted in higher MVPA estimates as supported by previous research [64].

The GRANADA consensus statement on analytical approaches for accelerometer-determined physical behaviors [65] gives an optional recommendation to express estimates of time spent in PA as bouted and unbouted. Although not statistically significant, the results from this meta-analysis showed that the number of activity minutes varied greatly between studies analyzing data on bouted versus unbouted PA. Highest levels were found in meta-analysis of unbouted data, except for SB, suggesting that unbouted data could potentially display higher activity levels. In the recent update of PA guidelines from WHO (2020) [13], recommendations that PA should be performed in bouts of ≥10 minutes were removed in recognition that all PA, regardless of length, promotes health. This is not yet fully reflected in analyses of PA data in studies being published.

4.1 Clinical implications and future research

The level of daily SB and steps found in this review revealed that there is still a potential to gain further health benefits by increasing PA in people with multimorbidity. These benefits could potentially be even more important in older populations, since studies have shown that older people spent the vast majority of their time in SB [66]. Furthermore, current evidence also supports that the level of PA decreases with increasing age in both healthy [67] and multimorbid populations [18]. Despite an extensive knowledge gap in the area, clinicians should therefore promote PA for people with chronic conditions and multimorbidity given the documented health benefits [10, 11]. However, they should also acknowledge the importance of primary prevention, since physical inactivity is a risk factor for development of chronic conditions and multimorbidity [8, 68]. In populations failing to adhere to PA guidelines, special attention should be given to the importance of replacing SB with PA of any intensity level as recommended by the WHO [13]. Future studies should aim to incorporate valid measurement methods that are able to measure the full continuum of PA from SB to VPA and steps so that detailed knowledge on PA can be incorporated more when developing tailored patient-centered strategies to increase PA in people with multimorbidity.

4.2 Strength and limitations

In the present review, multimorbidity was defined as encompassing common, but selected, chronic conditions. This definition was inspired by earlier published literature [11] and available evidence of the benefit of PA on the included conditions [10]. A recent (2021) review found that measurements and definitions of multimorbidity is poorly reported and varies greatly (from 2 to 285 conditions) [69]. It is possible that the use of a more exhaustive list of chronic conditions would have led to more studies being included. Despite this, the authors do believe that this review managed to capture the majority of studies published by selecting conditions with a high global prevalence [25].

All meta-analyses demonstrated high statistical heterogeneity among the included studies. Furthermore, only few relevant studies were identified, all associated with a high risk of bias, leading to an overall certainty in the evidence that was very low. Additional analyses on the impact of gender, ethnicity or socioeconomic status could not be performed since too few studies reported these variables consistently. The results should therefore be interpreted with caution. It is highly likely that future studies will change the PA estimates identified in this review.

Studies using subjective measurement methods of PA were not included in this review, since these have demonstrated less accuracy than objective measurements [21, 22]. Subjective methods have, however, been widely used to assess PA across different populations due to low costs and accessibility. Inclusion of such studies would most likely have increased the number of included studies in the review but challenged the validity. No perfect tool exists for measuring PA, and also objective measurement methods have displayed limitations [70]. It is however suggested that researchers should incorporate appropriate objective measures specific to the PA behavior of interest when examining PA in adults in free-living conditions [19]. This recommendation is reflected in the proportion of published studies using objective measures of PA that has increased from 4% to 71% from 2006 to 2016 [71] also justifying the focus on objectively measured PA in this review.

4.3 Conclusion

The results of this systematic review show that the daily level of MVPA appears to exceed the level recommended in the WHO PA guidelines, while the level of daily steps is equivalent to living a low active lifestyle. Adherence to PA guidelines are currently rarely reported and varies greatly from 7,4–43%. The average time spent in SB was more than eight hours, highlighting a potential for further improvement in PA levels among people with multimorbidity, in particular less active subgroups. In general, studies investigating objectively measured PA in populations with multimorbidity were few in numbers, associated with a high risk of bias and a very low overall certainty in the evidence calling for a cautious interpretation of the results. The review highlights an urgent need for further high-quality studies providing detailed descriptions of PA behavior among people with multimorbidity.

Supporting information

S2 File. Search strategy for Medline via OVID.


S3 File. Search strategy for EMBASE via OVID.


S4 File. Search strategy CIHNAL via EBSCOhost.


S6 File. Search strategy for grey literature.


S7 File. Quality assessment tool for quantitative studies.


S8 File. Studies excluded in full text screening.


S1 Fig. Sensitivity analysis with studies presenting sedentary behavior data with medians omitted.


S2 Fig. Sensitivity analysis with studies presenting light physical activity data with medians omitted.


S3 Fig. Sensitivity analysis with studies presenting moderate to vigorous physical activity data with medians omitted.


S1 Data. Data extraction form including data.



  1. 1. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet Lond Engl. 2012;380: 37–43.
  2. 2. Nguyen H, Manolova G, Daskalopoulou C, Vitoratou S, Prince M, Prina AM. Prevalence of multimorbidity in community settings: A systematic review and meta-analysis of observational studies. J Comorbidity. 2019;9: 1–15. pmid:31489279
  3. 3. Fortin M, Stewart M, Poitras M-E, Almirall J, Maddocks H. A systematic review of prevalence studies on multimorbidity: toward a more uniform methodology. Ann Fam Med. 2012;10: 142–151. pmid:22412006
  4. 4. Lehnert T, Heider D, Leicht H, Heinrich S, Corrieri S, Luppa M, et al. Review: health care utilization and costs of elderly persons with multiple chronic conditions. Med Care Res Rev MCRR. 2011;68: 387–420. pmid:21813576
  5. 5. Lancet T. Making more of multimorbidity: an emerging priority. The Lancet. 2018;391: 1637.
  6. 6. Kingston A, Robinson L, Booth H, Knapp M, Jagger C, for the MODEM project. Projections of multi-morbidity in the older population in England to 2035: estimates from the Population Ageing and Care Simulation (PACSim) model. Age Ageing. 2018;47: 374–380. pmid:29370339
  7. 7. Multimorbidity: a priority for global health research | The Academy of Medical Sciences. [cited 23 Jul 2019].
  8. 8. Booth FW, Roberts CK, Laye MJ. Lack of exercise is a major cause of chronic diseases. Compr Physiol. 2012;2: 1143–1211. pmid:23798298
  9. 9. Physical activity. [cited 14 Jan 2022].
  10. 10. Pedersen BK, Saltin B. Exercise as medicine—evidence for prescribing exercise as therapy in 26 different chronic diseases. Scand J Med Sci Sports. 2015;25 Suppl 3: 1–72. pmid:26606383
  11. 11. Bricca A, Harris LK, Jäger M, Smith SM, Juhl CB, Skou ST. Benefits and harms of exercise therapy in people with multimorbidity: A systematic review and meta-analysis of randomised controlled trials. Ageing Res Rev. 2020;63: 101166. pmid:32896665
  12. 12. A B, M J, M J, G Z, Lk H, J M, et al. Effect of In-Person Delivered Behavioural Interventions in People with Multimorbidity: Systematic Review and Meta-analysis. Int J Behav Med. 2022 [cited 7 Jun 2022]. pmid:35484462
  13. 13. Bull FC, Al-Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54: 1451–1462. pmid:33239350
  14. 14. Guthold R, Stevens GA, Riley LM, Bull FC. Worldwide trends in insufficient physical activity from 2001 to 2016: a pooled analysis of 358 population-based surveys with 1·9 million participants. Lancet Glob Health. 2018;6: e1077–e1086. pmid:30193830
  15. 15. Brawner CA, Churilla JR, Keteyian SJ. Prevalence of Physical Activity Is Lower among Individuals with Chronic Disease. Med Sci Sports Exerc. 2016;48: 1062–1067. pmid:26741117
  16. 16. da S Gomes R, Barbosa AR, Meneghini V, Confortin SC, d’Orsi E, Rech CR. Association between chronic diseases, multimorbidity and insufficient physical activity among older adults in southern Brazil: a cross-sectional study. Sao Paulo Med J. 2020;138: 545–553. pmid:33331604
  17. 17. Cimarras-Otal C, Calderón-Larrañaga A, Poblador-Plou B, González-Rubio F, Gimeno-Feliu LA, Arjol-Serrano JL, et al. Association between physical activity, multimorbidity, self-rated health and functional limitation in the Spanish population. BMC Public Health. 2014;14: 1170. pmid:25404039
  18. 18. Salman A, Sellami M. Do Older Adults with Multimorbidity Meet the Recommended Levels of Physical Activity? An Analysis of Scottish Health Survey. Int J Environ Res Public Health. 2019;16: 3748. pmid:31590293
  19. 19. Dowd KP, Szeklicki R, Minetto MA, Murphy MH, Polito A, Ghigo E, et al. A systematic literature review of reviews on techniques for physical activity measurement in adults: a DEDIPAC study. Int J Behav Nutr Phys Act. 2018;15: 15. pmid:29422051
  20. 20. Rj S. Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med. 2003;37. pmid:12782543
  21. 21. Prince SA, Adamo KB, Hamel ME, Hardt J, Connor Gorber S, Tremblay M. A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008;5: 56. pmid:18990237
  22. 22. Adams SA. The Effect of Social Desirability and Social Approval on Self-Reports of Physical Activity. Am J Epidemiol. 2005;161: 389–398. pmid:15692083
  23. 23. Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al. Cochrane handbook for systematic reviews of interventions. John Wiley & Sons; 2019.
  24. 24. Page MJ, Moher D, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ. 2021;372: n160. pmid:33781993
  25. 25. GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Lond Engl. 2020;396: 1204–1222. pmid:33069326
  26. 26. Godin K, Stapleton J, Kirkpatrick SI, Hanning RM, Leatherdale ST. Applying systematic review search methods to the grey literature: a case study examining guidelines for school-based breakfast programs in Canada. Syst Rev. 2015;4: 138. pmid:26494010
  27. 27. Covidence—Better systematic review management. [cited 5 Jan 2022].
  28. 28. Farrah K, Young K, Tunis MC, Zhao L. Risk of bias tools in systematic reviews of health interventions: an analysis of PROSPERO-registered protocols. Syst Rev. 2019;8: 280. pmid:31730014
  29. 29. Loprinzi PD, Davis RE. Bouted and non-bouted moderate-to-vigorous physical activity with health-related quality of life. Prev Med Rep. 2015;3: 46–48. pmid:26844186
  30. 30. Wolff-Hughes DL, Fitzhugh EC, Bassett DR, Churilla JR. Total Activity Counts and Bouted Minutes of Moderate-to-Vigorous Physical Activity: Relationships With Cardiometabolic Biomarkers Using 2003–2006 NHANES. J Phys Act Health. 2015;12: 694–700. pmid:25109602
  31. 31. Sagelv EH, Ekelund U, Pedersen S, Brage S, Hansen BH, Johansson J, et al. Physical activity levels in adults and elderly from triaxial and uniaxial accelerometry. The Tromsø Study. PloS One. 2019;14: e0225670. pmid:31794552
  32. 32. Young T, Hopewell S. Methods for obtaining unpublished data. Cochrane Database Syst Rev. 2011; MR000027. pmid:22071866
  33. 33. GRADE handbook. [cited 24 Mar 2022].
  34. 34. Iorio A, Spencer FA, Falavigna M, Alba C, Lang E, Burnand B, et al. Use of GRADE for assessment of evidence about prognosis: rating confidence in estimates of event rates in broad categories of patients. BMJ. 2015;350: h870. pmid:25775931
  35. 35. Kirk A., Barnett J., Leese G., Mutrie N. A randomized trial investigating the 12-month changes in physical activity and health outcomes following a physical activity consultation delivered by a person or in written form in Type 2 diabetes: Time2Act. Diabet Med. 2009;26: 293–301. pmid:19317825
  36. 36. Whipple MO, Schorr EN, Talley KMC, Wolfson J, Lindquist R, Bronas UG, et al. Influence of Changes in Sedentary Time on Outcomes of Supervised Exercise Therapy in Individuals with Comorbid Peripheral Artery Disease and Type 2 Diabetes. Ann Vasc Surg. 2020;68: 369–383. pmid:32278867
  37. 37. Li Y-N, Shapiro B, Kim JC, Zhang M, Porszasz J, Bross R, et al. Association between quality of life and anxiety, depression, physical activity and physical performance in maintenance hemodialysis patients. Chronic Dis Transl Med. 2016;2: 110–119. pmid:29063031
  38. 38. Murphy SL, Kratz AL, Williams DA, Geisser ME. The Association between Symptoms, Pain Coping Strategies, and Physical Activity Among People with Symptomatic Knee and Hip Osteoarthritis. Front Psychol. 2012;3: 326. pmid:22969747
  39. 39. Pozehl BJ, Mcguire R, Duncan K, Hertzog M, Deka P, Norman J, et al. Accelerometer-Measured Daily Activity Levels and Related Factors in Patients With Heart Failure. J Cardiovasc Nurs. 2018;33: 329–335. pmid:29538050
  40. 40. Miller G.D., Jakicic J.M., Rejeski W.J., Whit-Glover M.C., Lang W., Walkup M.P., et al. Effect of varying accelerometry criteria on physical activity: The look AHEAD study. Obesity. 2013;21: 32–44. pmid:23505166
  41. 41. Zhang Q, Schwade M, Schafer P, Weintraub N, Young L. Characterization of Sedentary Behavior in Heart Failure Patients With Arthritis. Cardiol Res. 2020;11: 97–105. pmid:32256916
  42. 42. Holber JP, Abebe KZ, Huang Y, Jakicic JM, Anderson AM, Belnap BH, et al. The Relationship Between Objectively Measured Step Count, Clinical Characteristics, and Quality of Life Among Depressed Patients Recently Hospitalized With Systolic Heart Failure. Psychosom Med. 2021. pmid:34724453
  43. 43. Correia MA, Silva GO, Longano P, Trombetta IC, Consolim-Colombo F, Ritti-Dias RM, et al. In peripheral artery disease, diabetes is associated with reduced physical activity level and physical function and impaired cardiac autonomic control: A cross-sectional study. Ann Phys Rehabil Med. 2021;64: 101365. pmid:32145411
  44. 44. Oliveira NL, Alves AJ, Ruescas-Nicolau M-A, Silva G, Teixeira M, Ribeiro F, et al. Arterial Stiffness is Associated With Moderate to Vigorous Physical Activity Levels in Post-Myocardial Infarction Patients. J Cardiopulm Rehabil Prev. 2019;39: 325–330. pmid:30913044
  45. 45. Piette JD, Richardson C, Himle J, Duffy S, Torres T, Vogel M, et al. A randomized trial of telephonic counseling plus walking for depressed diabetes patients. Med Care. 2011;49: 641–648. pmid:21478777
  46. 46. Reddy YNV, Rikhi A, Obokata M, Shah SJ, Lewis GD, AbouEzzedine OF, et al. Quality of life in heart failure with preserved ejection fraction: importance of obesity, functional capacity, and physical inactivity. Eur J Heart Fail. 2020;22: 1009–1018. pmid:32150314
  47. 47. Schlenk EA, Fitzgerald GK, Rogers JC, Kwoh CK, Sereika SM. Promoting Physical Activity in Older Adults With Knee Osteoarthritis and Hypertension: A Randomized Controlled Trial. J Aging Phys Act. 2020;29: 207–218. pmid:32887850
  48. 48. Schneider KL, Panza E, Handschin B, Ma Y, Busch AM, Waring ME, et al. Feasibility of Pairing Behavioral Activation With Exercise for Women With Type 2 Diabetes and Depression: The Get It Study Pilot Randomized Controlled Trial. Behav Ther. 2016;47: 198–212. pmid:26956652
  49. 49. Zucatti ATN, de Paula TP, Viana LV, DallAgnol R, Cureau FV, Azevedo MJ, et al. Low Levels of Usual Physical Activity Are Associated with Higher 24 h Blood Pressure in Type 2 Diabetes Mellitus in a Cross-Sectional Study. J Diabetes Res. 2017;2017: 6232674. pmid:29018827
  50. 50. Paula TP, Viana LV, Neto ATZ, Leitão CB, Gross JL, Azevedo MJ. Effects of the DASH Diet and Walking on Blood Pressure in Patients With Type 2 Diabetes and Uncontrolled Hypertension: A Randomized Controlled Trial. J Clin Hypertens Greenwich Conn. 2015;17: 895–901. pmid:26041459
  51. 51. Freedland KE, Carney RM, Rich MW, Steinmeyer BC, Rubin EH. Cognitive Behavior Therapy for Depression and Self-Care in Heart Failure Patients: A Randomized Clinical Trial. JAMA Intern Med. 2015;175: 1773–1782. pmid:26414759
  52. 52. Hult A, Johansson J, Nordström P, Nordström A. Objectively Measured Physical Activity in Older Adults With and Without Diabetes. Clin Diabetes Publ Am Diabetes Assoc. 2019;37: 142–149. pmid:31057220
  53. 53. Izawa KP, Watanabe S, Oka K, Hiraki K, Morio Y, Kasahara Y, et al. Association between mental health and physical activity in patients with chronic heart failure. Disabil Rehabil. 2014;36: 250–254. pmid:23614372
  54. 54. Lambert JD, Greaves CJ, Farrand P, Price L, Haase AM, Taylor AH. Web-Based Intervention Using Behavioral Activation and Physical Activity for Adults With Depression (The eMotion Study): Pilot Randomized Controlled Trial. J Med Internet Res. 2018;20: e10112. pmid:30012547
  55. 55. Moreira JSR, de Paula TP, Sperb LF, Miller MEP, Azevedo MJ, Viana LV. Association of plasma vitamin D status with lifestyle patterns and ambulatory blood pressure monitoring parameters in patients with type 2 diabetes and hypertension. Diabetes Res Clin Pract. 2018;139: 139–146. pmid:29518487
  56. 56. Tudor-Locke C, Hatano Y, Pangrazi RP, Kang M. Revisiting “how many steps are enough?” Med Sci Sports Exerc. 2008;40: S537–543. pmid:18562971
  57. 57. Hains-Monfette G, Atoui S, Needham Dancause K, Bernard P. Device-Assessed Physical Activity and Sedentary Behaviors in Canadians with Chronic Disease(s): Findings from the Canadian Health Measures Survey. Sports. 2019;7: 113. pmid:31100779
  58. 58. Gimeno-Santos E, Frei A, Steurer-Stey C, de Batlle J, Rabinovich RA, Raste Y, et al. Determinants and outcomes of physical activity in patients with COPD: a systematic review. Thorax. 2014;69: 731–739. pmid:24558112
  59. 59. Bassett DR, Wyatt HR, Thompson H, Peters JC, Hill JO. Pedometer-measured physical activity and health behaviors in U.S. adults. Med Sci Sports Exerc. 2010;42: 1819–1825. pmid:20305579
  60. 60. Althoff T, Sosič R, Hicks JL, King AC, Delp SL, Leskovec J. Large-scale physical activity data reveal worldwide activity inequality. Nature. 2017;547: 336–339. pmid:28693034
  61. 61. Vancampfort D, Koyanagi A, Ward PB, Rosenbaum S, Schuch FB, Mugisha J, et al. Chronic physical conditions, multimorbidity and physical activity across 46 low- and middle-income countries. Int J Behav Nutr Phys Act. 2017;14: 6. pmid:28100238
  62. 62. Chudasama YV, Khunti KK, Zaccardi F, Rowlands AV, Yates T, Gillies CL, et al. Physical activity, multimorbidity, and life expectancy: a UK Biobank longitudinal study. BMC Med. 2019;17: 108. pmid:31186007
  63. 63. Loyen A, Van Hecke L, Verloigne M, Hendriksen I, Lakerveld J, Steene-Johannessen J, et al. Variation in population levels of physical activity in European adults according to cross-European studies: a systematic literature review within DEDIPAC. Int J Behav Nutr Phys Act. 2016;13: 72. pmid:27350359
  64. 64. Steeves JA, Shiroma EJ, Conger SA, Van Domelen D, Harris TB. Physical activity patterns and multimorbidity burden of older adults with different levels of functional status: NHANES 2003–2006. Disabil Health J. 2019;12: 495–502. pmid:30871954
  65. 65. Migueles JH, Aadland E, Andersen LB, Brønd JC, Chastin SF, Hansen BH, et al. GRANADA consensus on analytical approaches to assess associations with accelerometer-determined physical behaviours (physical activity, sedentary behaviour and sleep) in epidemiological studies. Br J Sports Med. 2022;56: 376–384. pmid:33846158
  66. 66. Koolhaas CM, van Rooij FJA, Schoufour JD, Cepeda M, Tiemeier H, Brage S, et al. Objective Measures of Activity in the Elderly: Distribution and Associations With Demographic and Health Factors. J Am Med Dir Assoc. 2017;18: 838–847. pmid:28602617
  67. 67. Lounassalo I, Salin K, Kankaanpää A, Hirvensalo M, Palomäki S, Tolvanen A, et al. Distinct trajectories of physical activity and related factors during the life course in the general population: a systematic review. BMC Public Health. 2019;19: 271. pmid:30841921
  68. 68. Skou ST, Mair FS, Fortin M, Guthrie B, Nunes BP, Miranda JJ, et al. Multimorbidity. Nat Rev Dis Primer. 2022;8: 48. pmid:35835758
  69. 69. Ho IS-S, Azcoaga-Lorenzo A, Akbari A, Black C, Davies J, Hodgins P, et al. Examining variation in the measurement of multimorbidity in research: a systematic review of 566 studies. Lancet Public Health. 2021;6: e587–e597. pmid:34166630
  70. 70. Strath SJ, Kaminsky LA, Ainsworth BE, Ekelund U, Freedson PS, Gary RA, et al. Guide to the Assessment of Physical Activity: Clinical and Research Applications: A Scientific Statement From the American Heart Association. Circulation. 2013;128: 2259–2279. pmid:24126387
  71. 71. Silfee VJ, Haughton CF, Jake-Schoffman DE, Lopez-Cepero A, May CN, Sreedhara M, et al. Objective measurement of physical activity outcomes in lifestyle interventions among adults: A systematic review. Prev Med Rep. 2018;11: 74–80. pmid:29984142