https://orcid.org/0000-0001-7434-658X
Abstract
Physical strength can be an important parameter to monitor for patients with diabetes mellitus type 2 (T2DM) to promote healthy lifestyle choices. Functional measurements can contribute to healthcare advice and possibly motivate more active lifestyles. The aim of the study was to investigate whether adding measurement and feedback concerning handgrip strength (HGS) to the parameters measured for patients with T2DM at annual check-ups leads to change in physical activity (PA) level, HGS, HbA1c or waist circumference.
Methods
Measurement of HGS with Jamar dynamometers was added to annual check-ups for patients with T2DM by diabetes nurses in primary care with feedback about normal values for age and sex in the intervention group. The control group had standard check-ups. Change in self-reported PA level was measured with questionnaires.
Results
Seven clinics and 334 patients participated. The intervention led to similar effects on PA in both groups. Patients with T2DM had comparable HGS to the general public. Regression analyses showed statistically significantly higher HGS in the intervention group than in the control group at follow-up and no improvement in PA, HbA1c, or waist circumference. Increased HGS was found for older people, men, and people with normal-to-high inclusion HGS, while patients with low inclusion HGS reduced their strength levels.
Conclusions
Measuring HGS and giving feedback to patients with T2DM can lead to increased HGS but does not seem to affect general PA level, HbA1c, or waist circumference. People over 65 years, men, and people with normal-to-high HGS were influenced positively by the intervention. Patients with low HGS may need personalised support to increase physical activity and improve function.
ClinicalTrials registration: NCT03693521
Introduction
The prevalence of diabetes mellitus type 2 (T2DM) has increased exponentially in both the developed and developing parts of the world, with millions of new cases being diagnosed worldwide over the last decade [Citation1]. There is good evidence that this trend is partially due to changing lifestyles with clear correlation to obesity and physical inactivity [Citation1]. People with T2DM have an increased risk for cardiovascular incidents and diseases, and for early mortality [Citation1].
In Sweden, management of most patients with T2DM is the responsibility of primary care. Standard management includes an annual check-up by a physician and annual screening by a diabetes nurse, who monitor progress of the disease and help patients manage their healthcare needs. Advice is given by both physicians and nurses regarding lifestyle, diet, and physical activity (PA) level [Citation2].
Many parameters are used to monitor progress of T2DM, to determine medication level, and to inform personal management and lifestyle choices. While blood glucose, cholesterol, blood pressure levels and weight are measured objectively and followed closely, PA level, which is recognised as an important parameter for progress of the disease, is most often self-reported. However, questionnaires are an inexpensive, quick, and easy measurement method and give results which can be followed over time.
Studies have indicated a negative correlation between general muscular strength and frequency and impact of severe health incidents [Citation3]. Physically weak people are more prone to cardiovascular incidents and have greater risk of early mortality [Citation3]. Physical frailty is a problem for many older people with T2DM and international recommendations include regular examinations of physical function for this patient group [Citation4].
Handgrip strength (HGS) is positively correlated to general strength [Citation5]. It is negatively correlated to early mortality, to cardiovascular incidents, and positively correlated to survival from cardiovascular incidents [Citation3]. It is associated with reduced risk for incident T2DM among middle-aged women [Citation6]. It is inexpensive, quick, and easy to measure and has been recommended as a suitable risk-stratifying method for several health disorders [Citation3]. Measurement of HGS was hypothesised to give healthcare personnel a new, relevant, objective measure of physical capacity for patients with T2DM and that feedback about personal function in relation to normal values might inspire lifestyle changes.
It is unknown whether measurement and feedback concerning HGS may lead to behavioural changes in the target population. Previous research highlights that people appreciate healthcare monitoring of functional capacity [Citation7,Citation8]. An as yet unpublished study by this research group has shown that structured motivational feedback concerning functional capacity increases motivation for health-promoting behaviours among healthy middle-aged people [Citation9]. Patients with T2DM receive advice and feedback regularly from their caregivers within primary care but may seldom discuss objectively measured strength level.
The aim of this study was, therefore, to investigate whether the addition of measurement of HGS and feedback about personal strength in relation to normal values for age and sex to the annual check-ups of patients with T2DM by diabetes nurses in primary care would affect self-reported PA levels or lead to change in HGS, HbA1c, or waist circumference.
Materials and methods
Procedure
A randomised controlled trial was initiated at seven primary care centres in western Sweden. Diabetes nurses at these centres were instructed to invite all patients with T2DM who came for their annual check-ups to participate in the study. There were no exclusion criteria. Patients received verbal and written information by the nurse who randomised those who agreed to participate to either intervention or control by means of prepared, sealed envelopes. The randomisation sequence was determined by digital algorithm with block size of 20. In the intervention group, HGS was measured with Jamar hand dynamometers (Performance Health International LTD, Nottinghamshire, UK) at inclusion and again at the next annual check-up by the diabetes nurse. Participants were informed of their results in relation to normal values for age and sex. Any ensuing discussion and advice from the diabetes nurse was individually determined based on the nurse's assessment and the patient's situation. The study protocol did not attempt to dictate the content of this discussion. The control group had their usual annual check-up at inclusion. At the following annual check-up, HGS was measured for the control group as well. Self-reported PA was measured at inclusion and follow-up in both groups with questionnaires.
Outcomes
The primary outcome was change in self-reported PA level determined by two questions according to the American College of Sports Medicine (ASCM) and the American Heart Association (AHA) recommendations [Citation10]: How many days during the last week did you participate in moderate intensity physical activity for a minimum of 30 min (at least 10 min at a time) that made you breathe somewhat harder and increased your heart rate moderately – for example a brisk walk? and How many days during the last week did you participate in intense physical activity/exercise for at least 20 min that made you very short of breath and increased your heart rate markedly? The second question was weighted 1.7 times higher than the first. Levels below five points are deemed insufficient according to ASCM-AHA recommendations.
Measurements of HGS with Jamar hand dynamometers were made in the seated position without backrest. The measured arm was held in a 90ᵒ angle at the elbow with a small gap between elbow and trunk. The participant was instructed to squeeze the handle of the dynamometer as hard as possible for a few seconds. The instrument marks the highest force achieved. Measurements were made in kilograms. Three measurements were made per arm, alternating left and right, to ensure a short rest of approximately 20 s between repeated measurements on the same arm. Rather than the mean, the maximum value of the three measurements is reported as recommended in relevant literature [Citation3,Citation11,Citation12]. Change in maximum value between inclusion and follow-up was examined as a secondary outcome. The difference between measured HGS and published normal values for age and sex (Additional file) was calculated and compared between groups at follow-up.
HbA1c and waist circumference, which are regularly measured parameters for this population, were assessed according to established routines at each clinic at inclusion and follow-up. No instructions were given for these measurements in the study protocol. Changes in diabetes medication were recorded as decreased, no change, increased, unclear, or unknown. The number of patients given Physical Activity on Prescription (PAP) [Citation13] at the inclusion visit were also reported. PAP is an accepted healthcare practice in western Sweden for promoting physical activity.
Data analysis
Descriptive statistics were used to examine demographics and primary and secondary outcomes. Differences between measured HGS and published normal values for age and sex for HGS (normalised HGS (NHGS)) [Citation11,Citation14,Citation15] () were calculated for each person and timepoint, with positive values indicating higher strength than normal and negative values indicating lower strength than normal. Correlations between PA at inclusion and change in PA were examined for the two groups. Multiple linear regression was used to analyse whether group allocation predicted change in self-assessed physical activity level or in secondary outcomes when adjusted for age, sex, clinic, inclusion values of all outcomes except HGS, medication changes and whether or not physical activity was prescribed according to Swedish routines for PAP [Citation13]. As HGS was only measured in the intervention group at inclusion, it could not be included as a predictor in the regression. In the regression analyses, participating clinics were coded to approximately follow the socioeconomic index for the area in which the clinic was situated [Citation16]. Sub-group analyses were made for patients with low (values < 0 for either right or left hand) and normal-to-high NHGS at inclusion, for a younger and an older age-group (cut-off 65 years) and for men and women.
Table 1. Descriptive statistics at inclusion.
The trial was registered prospectively at ClinicalTrial.gov (NCT03693521) on 24 August 2018.
Results
Recruitment to the study was initiated in September 2019 but was hindered by the 2020–2021 Covid-19 pandemic. Some clinics were able to continue having regular annual check-ups for patients with diabetes as usual and others were not. In total, 334 patients were recruited from seven different clinics and 300 were followed up (). The follow-up times were in many cases prolonged because of the pandemic. Follow-ups were completed in June 2023. Descriptive statistics are presented in . There were no significant differences between intervention and control groups at inclusion regarding demographics or outcomes.
Figure 1. Consort 2021 flow diagram [Citation17].
![Figure 1. Consort 2021 flow diagram [Citation17].](/cms/asset/802dd473-e722-4777-9dae-efd7a651687c/ipri_a_2373298_f0001_c.jpg)
Missing follow-up results were due to death (n = 7), illness (n = 4), moved from area (n = 9), own reasons (n = 9) and deviations from study protocol (n = 5). There is also limited internal missing results due to deviations from study protocol.
Non-adjusted changes in PA and secondary outcomes are presented in . Mean HGS at first measurement occasion in each group was slightly higher than normal values for the general population, indicating that muscular weakness is not a widespread problem among patients with T2DM. At follow-up, NHGS was significantly higher in the intervention group than in the control group for both right (p = 0.030) and left (p = 0.011) hands. There was a moderate negative correlation between inclusion PA and change in PA at follow-up for the whole group (R = −0.436, 95%CI −0.525, −0.337), the intervention group (R=-0.466, 95%CI −0.585, −0.327) and the control group (R = −0.389, 95%CI −0.521, −0.239).
Table 2. Change in primary and secondary outcomes (non-adjusted).
The results of the regression analyses for PA regarding group allocation and adjusted for age, sex, clinic, inclusion values for PA, HbA1c and waist circumference, as well as medication changes and whether PAP was given at inclusion are presented in . Inclusion PA was the only predictor which significantly affected change in PA with patients with low inclusion self-reported PA increasing self-reported PA at follow-up to a greater extent.
Table 3. Multivariable linear regression analyses for change in physical activity level.
Secondary outcomes included between-groups comparisons for HGS and NHGS at follow-up and for change in HbA1c, and waist circumference between inclusion and follow-up. Results of regression analyses are presented in . Allocation to the intervention group significantly and positively affected HGS and NHGS but not HbA1c. There was a very poor model fit for change in waist circumference and with very low R2 (0.026), which limits interpretation of the results.
Table 4. Multivariable linear regression analyses for change in secondary outcomes.
Results of the sub-group analyses can be found in . All regression models were statistically significant except for those regarding change in waist circumference for all sub-groups and for NHGSRight for women. The intervention had no significant effect on the primary outcome – change in self-reported PA level – in any of the sub-groups. As in the whole group analysis, inclusion PA had a considerable negative association with change in PA in all sub-groups and in both intervention and control groups. Nor did the intervention significantly affect change in HbA1c or change in waist circumference in most cases. There was a significant group effect only in the sub-group with low NHGS, indicating that the intervention group had higher HbA1c at follow-up (). Regarding HGS, it was found that intervention group allocation led to significantly higher HGS and NHGS at follow-up for older people, men, and people with normal-to-high initial NHGS. It did not significantly affect HGS and NHGS for women or for younger people. Also, in the low NHGS group, being allocated to the intervention group had a negative effect on strength.
Discussion
The intervention of adding measurements and feedback concerning HGS to annual check-ups for patients with T2DM by diabetes nurses in a pragmatic trial in primary care settings did not lead to increased self-reported PA but did seem to lead to increased HGS and NHGS in the intervention group. Those with lower inclusion PA values in both groups reported increased PA at follow-up to a greater extent than those who were already more physically active. The increase in NHGS was significant for patients over 65 years, men, and patients with normal-to-high inclusion HGS, whereas women and younger people were not significantly affected and, somewhat unexpectedly, HGS for people with low inclusion HGS was negatively affected by the intervention. This latter group was the only sub-group which showed a negative group effect on strength and whose HbA1c level increased significantly. Participating clinics were coded to approximately follow the socioeconomic index for their catchment area and there was a clear clinic-related association with increased HGS in areas with higher socioeconomic index in the whole group and in all sub-groups except the group over 65 years of age. Also, there was a clinic-related association with reduction of HbA1c in the sub-groups consisting of stronger patients, older patients, and women. At inclusion, HGS for patients with T2DM was comparable with or slightly higher than population norms for age and sex.
Muscular weakness is a risk factor for many types of illness, including cardiovascular disease [Citation3]. Strength is lost through lack of muscular exertion. Increased strength comes through physically exerting activities, which can be achieved through specific training or through more subtly exerting daily activities. Our hypothesis that receiving feedback about personal HGS might increase motivation for general PA was not supported, at least not as we chose to measure PA. There was a moderate negative correlation between inclusion PA and change in PA at follow-up in both groups, and in all sub-groups, which can be interpreted in two ways: either people who described low PA at inclusion felt a need to report higher PA at follow-up regardless of actual activity or that measuring PA in itself inspired to increased activity. Self-reported PA has poor validity according to some studies, but is often the most accessible measure available [Citation18]. Self-reported PA values can be highly variable and are susceptible to perceived desirability of the reported value [Citation19,Citation20]. As it is not likely that most inactive patients with T2DM generally increase their activity level from year to year spontaneously, it may be that measuring PA by questionnaire is also an intervention which affects how people behave or at least how they respond to questions about PA. The ACSM-AHA PA questions, which were used in this study, focus on aerobic activity and do not measure participation in resistance training. In this study, we found increased HGS in the intervention group and increased PA among inactive patients, but no relation between the two measures. Objectively measured PA with accelerometers, on the other hand, has been shown to correlate with objectively measured function [Citation21].
The sub-group with negative effect of the intervention was the group with below normal HGS. While this is the group which stands to gain the most by increasing both strength and PA, changes in the opposite direction were seen. It is likely that this group needs more specific individual support by relevant caregivers to make necessary lifestyle changes regarding physical exertion. This group also increased their HbA1c level during the study period. Deterioration of such traditional T2DM indicators seems to correspond to both lower PA and lower strength.
Increased HGS after the intervention was most clearly seen in the sub-groups composed of men and of people over 65 years of age. It is particularly important to note the significant improvement in strength in the older group as they are at highest risk for adverse effects of diabetes, cardiovascular incidents, and sarcopenia [Citation4,Citation22]. The stronger effect on men regarding increased HGS may reflect a more competitive spirit in the male group. Men were overrepresented in the study, but it is unknown whether this was caused by men being more interested in measuring their HGS or because of the male overrepresentation in Sweden among patients with T2DM [Citation23]. There was no significant effect on HGS for women or people under 65 years. These groups may need more support or more information about the effect of muscular strength on general health to make changes. People with normal-to-high HGS were also inspired to maintain or increase their strength by simply measuring it and receiving feedback. As it is easier to maintain function than to build it up once problems have been established, it is important to encourage normal function in many patient groups and to begin informing about the advantages of resistance training before functional deficits become measurable. These sub-group analyses give relevant information to clinically active diabetes nurses who need to prioritise among many different measures when they hold annual check-ups. Knowing who may be most easily influenced by functional measurements can be useful.
The clinics were distributed throughout areas with varying socioeconomic conditions. Attending clinics located in areas with higher socioeconomic index was associated with improvement in HbA1c and HGS. Whether this is due to patient characteristics, nurse routines, or other factors cannot be determined through the data, but it can be suspected that socioeconomic conditions may affect patients' ability or motivation to make lifestyle-related changes relevant to this intervention. This is consistent with many previous studies showing clear associations between socioeconomy and incident diabetes, cardiovascular risk factors, physical fitness, and PA [Citation24,Citation25]. In this study, the areas with low socioeconomic index had higher proportions of immigrants, including new immigrants. It is possible that detailed exercise advice was difficult to fit in when there were language difficulties or that people who originated in countries without the same exercise culture as in Sweden were more difficult to reach with exercise advice.
The nurses were asked to report whether they gave PAP at the inclusion visit. PAP has been shown to have good effect on increasing PA in patients with metabolic syndrome as well as in other populations [Citation26]. Since the nurses now had concrete measures of both PA and strength, discerning those in need of PAP should have been relatively easy. In the older group, PAP was significantly associated with better HGS and HbA1c values. Medication changes, on the other hand, were only significantly associated with reducing HbA1c in the younger group. Perhaps more effort should be made to motivate increased PA and resistance training among older people to regulate HbA1c, with PAP being a particularly useful method for older patients.
Study limitations
The effect of the Covid-19 pandemic on the study is unclear. Recruitment of participants was postponed at several clinics and follow-ups were delayed or replaced with telephone or digital contacts which were not conducive to following study protocol. Because of this, it is unknown how large proportion of possible participants were asked to participate. None of the nurses reported frequent active refusal. As both study groups were affected equally, it is unlikely that the main results would have been different if the recruitment procedure had been more optimal. It is possible that, if a higher number of participants could have been recruited, more detailed results could have been discerned, especially in the sub-group analyses where there were several predictors which approached significance. It is unlikely that a group effect on the primary outcome would have been established as there were no trends in this direction. There is also the probability that many people belonging to risk groups for Covid-19, such as people with T2DM, were more isolated, avoided gyms and similar establishments, and likely became less active than usual during a long interval within the study period. This was not seen in our follow-up self-reported PA measurements but may, nonetheless, have had an effect on some people.
The clinically active nurses who participated in the study had no problem learning to measure HGS or to follow study protocol. However, some found it difficult to remember the rationale behind the study and the role muscular strength plays in general health which may have affected the advice they gave or did not give. It is possible that not all diabetes nurses feel confident giving advice on musculoskeletal health and function or on exercise alternatives. Given the importance that strength, fitness, and PA play in the progression of T2DM, solutions for this situation should be examined.
We have not found any previous research concerning HGS for patients with T2DM. Our intervention included measurement of HGS, feedback and possibly advice from diabetes nurses. Instead of designing the study to simply measure PA in the control group at intervention and follow-up, we added measurement of HGS at follow-up in the control group to ensure the groups were comparable regarding HGS. Although we do not have HGS measurements at inclusion for the control group, only negligible changes in strength level would be expected due to the natural course of events; perhaps a minimal decrease due to ageing which would affect both groups. Except among the very old and the very young, one year is probably a too short period of time to see those age effects regarding strength that could be expected without behavioural change. In addition, by comparing each patient's results to population norms, we have tried to compensate for the lack of inclusion values for HGS in the control group.
Another limitation is the presence of multiple testing in the analysis. Several of the p-values are connected to confounder adjustment (age, sex, clinic, inclusion values for PA, HbA1c, waist circumference, medication changes and whether PAP was given at inclusion). Therefore, interpretation of results for confounders should be made cautiously.
Implications
This relatively simple intervention raises some interesting points concerning clinical management of patients with T2DM. Should strength measurements be included in annual check-ups? If so, is HGS the most suitable measurement to choose? Should strength be measured for all patients or only certain sub-groups? How should the management of weaker patients be structured? Also, should PA be measured and if so, how? And what about PAP – who should healthcare professionals prioritise when prescribing PA?
In this study, clinical nurses with no general previous experience of measuring strength incorporated measuring HGS into their normal routines without any known problems. They found the method easy to learn and managed to fit the extra measurement into their allotted consultation time. The dynamometer is relatively inexpensive. Therefore, since the costs are minimal, if negative health events can be prevented by following strength measurements in a risk population such as patients with T2DM, it is possible that the intervention could be cost-effective. Preventive effects are, however, difficult to measure and usually require following large groups of people over long periods of time. Within clinical trials, it can be useful to use a proxy such as PA or strength which predict many types of negative health events [Citation3,Citation27]. In this study, the lack of change in PA contradicts the implications indicated by the increase in HGS. So it remains unclear whether measuring HGS is a useful parameter for patients with T2DM.
Since muscular strength is an important factor for both morbidity and mortality, it can be a relevant parameter for healthcare to follow. While most sub-groups of patients had either positive or negligible effects of the intervention, the group with lower HGS had a negative effect and clearly needs more than just feedback about their status to make function and activity-based lifestyle changes. Support, encouragement, and strategies to increase strength should be offered. This can be done by the patient's regular caregivers, through PAP, or in other lifestyle support programs. If no progress is made, perhaps a physiotherapist should be consulted. Weaker people stand to gain health advantages by increasing strength, although they possibly need extra support. People with normal to high strength levels seem to make efforts to maintain or increase their strength when it is followed by healthcare. Therefore, functional testing should play a role in yearly check-ups.
Regularly measuring PA with a questionnaire seems to be a simple and positive addition to a clinical check-up. It increases the focus on PA and the importance that healthcare places on PA and encourages patients to consciously consider their own PA level. Inactive people seem to increase PA when they know they will need to account for it. More objective methods would be even better, but objective measurement of PA is more complicated and likely difficult to implement on a large scale [Citation28,Citation29].
Among patients over 65 years of age, PAP was significantly associated with increased HGS. Older people stand to reduce health risks substantially by becoming both more physically active and stronger. In parts of Sweden, PAP is already an established management strategy, but the results of this study suggest that it should perhaps be employed more systematically among the elderly.
More research is needed to evaluate different strength measures and their utility for this population and to investigate need for support in increasing functional level among risk groups. Also, longer term follow-ups are needed to determine whether regular HGS measurements lead to maintenance of better strength levels over time and to evaluate the relationship between HGS and other parameters for T2DM and general health over time.
Conclusions
Handgrip strength level for patients with T2DM is comparable to to that for the general population. Measuring HGS and giving feedback to patients with T2DM about their strength level in relation to normal values can lead to increased HGS but does not seem to affect general PA level, HbA1c, or waist circumference. People over 65 years, men, and people with normal-to-high HGS were influenced positively by the extra focus on strength measurements. The sub-group of weaker patients with T2DM likely needs personal support and instruction to become more physically active and to gain better function.
Acknowledgements
The authors would like to thank all the participants and all the diabetes nurses and primary care centres who performed the data collection. A special thanks to Dr Margareta Hellgren for her support in designing the study.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Additional information
Funding
References
- Glovaci D, Fan W, Wong ND. Epidemiology of diabetes mellitus and cardiovascular disease. curr cardiol rep. 2019;21(4):21.
- Trout KK, McCool WF, Homko CJ. Person-centered primary care and type 2 diabetes: beyond blood glucose control. J Midwifery Womens Health. 2019;64(3):312–323. doi: 10.1111/jmwh.12973.
- Leong DP, Teo KK, Rangarajan S, et al. Prognostic value of grip strength: findings from the prospective urban rural epidemiology (PURE) study. Lancet. 2015;386(9990):266–273. doi: 10.1016/S0140-6736(14)62000-6.
- Sinclair AJ, Abdelhafiz A, Dunning T, et al. An international position statement on the management of frailty in diabetes mellitus: summary of Recommendations 2017. JFrailty Aging. 2018;7(1):10–20.
- Samson MM, Meeuwsen IB, Crowe A, et al. Relationships between physical performance measures, age, height and body weight in healthy adults. Age Ageing. 2000;29(3):235–242. doi: 10.1093/ageing/29.3.235.
- Karvonen-Gutierrez CA, Peng Q, Peterson M, et al. Low grip strength predicts incident diabetes among mid-life women: the Michigan Study of Women's Health Across the Nation. Age Ageing. 2018;47(5):685–691. doi: 10.1093/ageing/afy067.
- Puthoff M. Participants' perceptions and the implementation of a physical fitness screen for aging adults. J Geriatr Phys Ther. 2021;44(1):E1-E8. doi: 10.1519/JPT.0000000000000241
- Bornhöft L, Arvidsson D, Bergenheim A, et al. Development and feasibility of a function-based preventive intervention for lifestyle-related disorders. BMC Public Health. 2024;24(1):681. doi: 10.1186/s12889-024-18017-8.
- Bornhöft L, Bergenheim A, Börjesson M, et al. Pilot evaluation of a function-based primary preventive intervention on physical activity level and risk for lifestyle-related disorders among inactive 40-year-old people. 2024;
- Haskell WL, Lee I-M, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation. 2007;116(9):1081–1093. doi: 10.1161/CIRCULATIONAHA.107.185649.
- Leong DP, Teo KK, Rangarajan S, et al. Reference ranges of handgrip strength from 125,462 healthy adults in 21 countries: a prospective urban rural epidemiologic (PURE) study. J Cachexia Sarcopenia Muscle. 2016;7(5):535–546. doi: 10.1002/jcsm.12112.
- Bohannon RW, Peolsson A, Massy-Westropp N, et al. Reference values for adult grip strength measured with a Jamar dynamometer: a descriptive meta-analysis. Physiotherapy. 2006;92(1):11–15. doi: 10.1016/j.physio.2005.05.003.
- Onerup A, Arvidsson D, Blomqvist Å, et al. Physical activity on prescription in accordance with the Swedish model increases physical activity: a systematic review. Br J Sports Med. 2019;53(6):383–388. doi: 10.1136/bjsports-2018-099598.
- Bohannon RW, Wang YC, Yen SC, et al. Handgrip strength: a comparison of values obtained from the NHANES and NIH toolbox studies. Am J Occup Ther. 2019;73(2):7302205080p1–7302205080p9. doi: 10.5014/ajot.2019.029538.
- Mathiowetz V, Kashman N, Volland G, et al. Grip and pinch strength: normative data for adults. Arch Phys Med Rehabil. 1985;66(2):69–74.
- Boverket. Segregationsbarometern: boverket.se; 2021; [cited 2023 2023-11-13]. Available from: https://segregationsbarometern.boverket.se/kommun/goteborg/.
- Schulz KF, Altman DG, Moher D, . CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. BMC Med. 2010;8(1):18. doi: 10.1186/1741-7015-8-18.
- Shephard RJ. Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med. 2003;37(3):197–206; discussion 206. doi: 10.1136/bjsm.37.3.197.
- Hicks J, Adams-Campbell LL, Lu J, et al. Validation study of physical activity and sedentary behavior in African-American Men and Women. J Racial Ethn Health Disparities. 2023; doi: 10.1007/s40615-023-01734-y.
- Arumugam A, Mohammad Zadeh SA, Zabin ZA, et al. Sedentary and physical activity time differs between self-reported ATLS-2 physical activity questionnaire and accelerometer measurements in adolescents and young adults in the United Arab Emirates. BMC Public Health. 2023;23(1):1045. doi: 10.1186/s12889-023-15881-8.
- Mickute M, Henson J, Rowlands AV, et al. Device-measured physical activity and its association with physical function in adults with type 2 diabetes mellitus. Diabet Med. 2021;38(6):e14393. doi: 10.1111/dme.14393.
- Evans WJ, Guralnik J, Cawthon P, et al. Sarcopenia: no consensus, no diagnostic criteria, and no approved indication-How did we get here? Geroscience. 2023;46(1):183–190. doi: 10.1007/s11357-023-01016-9.
- Wändell PE, Carlsson AC. Gender differences and time trends in incidence and prevalence of type 2 diabetes in Sweden–a model explaining the diabetes epidemic worldwide today? Diabetes Res Clin Pract. 2014;106(3):e90-2–e92. doi: 10.1016/j.diabres.2014.09.013.
- Agardh E, Allebeck P, Hallqvist J, et al. Type 2 diabetes incidence and socio-economic position: a systematic review and meta-analysis. Int J Epidemiol. 2011;40(3):804–818. doi: 10.1093/ije/dyr029.
- Lindgren M, Börjesson M, Ekblom Ö, et al. Physical activity pattern, cardiorespiratory fitness, and socioeconomic status in the SCAPIS pilot trial - A cross-sectional study. Prev Med Rep. 2016;4:44–49. doi: 10.1016/j.pmedr.2016.04.010.
- Lundqvist S, Börjesson M, Larsson MEH, et al. Physical activity on prescription (PAP), in patients with metabolic risk factors. A 6-month follow-up study in primary health care. PLOS One. 2017;12(4):e0175190. doi: 10.1371/journal.pone.0175190.
- Myers J, Kokkinos P, Nyelin E. Physical activity, cardiorespiratory fitness, and the metabolic syndrome. Nutrients. 2019;11(7):1652. doi: 10.3390/nu11071652.
- Atienza AA, Moser RP, Perna F, et al. Self-reported and objectively measured activity related to biomarkers using NHANES. Med Sci Sports Exerc. 2011;43(5):815–821. doi: 10.1249/MSS.0b013e3181fdfc32.
- Arvidsson D, Fridolfsson J, Börjesson M. Measurement of physical activity in clinical practice using accelerometers. J Intern Med. 2019;286(2):137–153. doi: 10.1111/joim.12908.
Appendix
Table A1. Normal values handgrip strength.
Table A2. Regression analyses for change in primary and secondary outcomes for the Sub-group with low handgrip strengthTable Footnote* for age and sex at inclusion
Table A3. Regression analyses for change in primary and secondary outcomes for the Sub-group with normal-to-high handgrip strengthTable Footnote* for age and sex at inclusion
Table A4. Regression analyses for change in primary and secon − ary outcomes for the Sub-group under 65 years of age at inclusion
Table A5. Regression analyses for change in primary and secondary outcomes for the Sub-group 65 years of age and over at inclusion
Table A6. Regression analyses for change in primary and secondary outcomes for the male Sub-group
Table A7. Regression analyses for change in primary and secondary outcomes for the female Sub-group