Our main findings were that a primary care-based 3-month group exercise intervention increased leg-muscle strength and function assessed as maximal step-up height (MSH). Furthermore, we found that MSH was declining to below baseline level at 14-30 months follow-up. By that time, our female patients had stopped with their regular group training 2-3 times per week, but reported brisk walking most days of the week as their main exercise. Furthermore, at the long-term follow-up, at a mean of 22 months, the patients still had remaining positive effects on most outcomes. Moreover, at the first follow-up directly after the 3-month period of group exercise as part of the intervention program, patients had a significant decline in weight and BMI. Weight stability compared to baseline remained at the long-term follow-up. Clinically acceptable inter-examiner variation at 4 cm found in this study is a minor concern compared to test-retest variation, which measures at 6 cm . This means that to achieve high validity of MSH results within a study, it is more important that examiners are robust in their testing methods than that several different examiners are used. Furthermore, the inter-examiner repeatability was tested at long-time follow-up to mimic future clinical situations with repeated testing, and with patients with different degrees of physical fitness.
Group exercise included different workouts, mostly mixed aerobic fitness and strength training with a moderate to high intensity level 13-15 on 6-20 Borg RPE scale, which the patients trained to use for intensity control. However, also sessions with merely strength, balance and coordination training were offered to the patients, see Additional file1.
Analyzing MSH subgroups according to level of maintenance of MSH at long-term follow-up, we found that even the subgroup with the lowest maintenance of MSH maintained their MSH at the follow-up directly after the 3-month group-training period. Thereafter their MSH declined from different MSH baseline levels (Figure 1a and 1b), and did so more steeply than for participants in the other subgroups. Most important for MSH decline—in a logistic regression multivariate analysis with maintenance of MSH as outcome—was the length of time before long-term follow-up, indicating a continuous decline of leg-muscle strength and function detectable within a mean period of 22 months. Secondly, an increase in BMI and thereafter higher age (increased sarcopenia) and higher MSH levels at baseline were important for MSH decline.
In the subgroup with the lowest maintenance of MSH, 44% of the female patients had been put on sick leave by the time of the long-term follow-up, compared to 27% of the patients with the highest maintenance of MSH. Changes in MSH and increased sedentary behavior (SB)—due to reduced working capacity and cessation of transportation to work—are further discussed in Additional file 3, section I. These could be reasons for steeper MSH decline in our study but SB was not assessed. Furthermore, evidence has been presented in a recent meta-analysis in which lifestyle interventions showed the potential to reduce SB in adults .
Thus, those who during holiday season—the three summer months—could find regular group exercise with mixed aerobic fitness and strength training, increased their ability to maintain leg-muscle strength and function assessed as MSH. Brisk walking most days of the week was not enough to maintain these individuals’ leg strength and function. Also, lower levels of support from primary health care professionals after the program to encourage patients to prioritize regular group exercise—with mixed aerobic fitness and strength training 2-3 times per week with the subjective intensity of 13-15 on the 6-20 Borg scale—could be a reason for the steeper MSH decline beyond the expected due to age.
MSH along a time axis visualizes changes in muscle strength
Patients with high baseline MSH appeared more frequently early on the time axis in the on-going intervention project (see Figure 1a). One important reason for this was that patients with many diseases and consequently lower MSH were deemed not to manage the training intervention by their GP, for both physical and mental reasons. The opinions of GPs, however, changed over time after positive reports from patients. According to a recent European multicentre study paper, the greatest difference in mortality risk was observed between the two lowest activity groups for a population with abdominal and general adiposity, as in our female patient population .
Mixed aerobic fitness and strength training with higher intensity important for MSH
In a meta-analysis, combined aerobic and resistance training—the most effective training modality to reduce anthropometric outcomes—have been recommended in the prevention and treatment of overweight and obesity whenever possible . Mixed aerobic fitness and strength training 2-3 times per week in our study (see Additional file 1) could significantly improve MSH and VO2-max at both follow-ups and also waist circumference at long-term follow-up. Brisk walking—as reported in our female patient population as the main exercise at long-term follow-up with the mean of 22 months—was at the level of the recommended 150 minutes per week with moderate activity , but this was not enough for maintenance of their MSH. In a systematic review, interventions increased walking among targeted participants on an average of up to 30-60 minutes per week, at least in the short term. Important for a successful increase was when intervention was tailored to participants’ needs and targeted at those with low physical activity . The group exercise intervention in this study was designed with the same intentions and could be recommended for women with common diseases, and with low working capacity, as well as to patients with cardio-metabolic risks.
Women in their 50s and cardio-metabolic risk
It is known that cardiorespiratory fitness in general declines at a nonlinear rate which accelerates after 45 years of age , and possibly the situation is the same for MSH. However, the women patients in their 50s (table 2; 3c) in our study had lower aerobic fitness compared to the oldest age group, see Additional file 3, section II.
Another notable lifestyle change was that eleven out of twenty smokers had stopped smoking at long-term follow-up.
Change in MSH and metabolic risk
The subgroup with highest maintenance of MSH at long-term follow-up had lower mean MSH (cm) (SD) at baseline, 26.7 (4.1) compared to 29.6 (6.0) in the group with the lowest maintenance of MSH, see Additional file 3, section III. No significant difference was seen in mean levels of sick leave between groups at baseline. When a person with overweight changes lifestyle in a short time—for example, being put on sick leave and becoming less physically active without simultaneous weight loss—the decline in leg-muscle strength gives reduced ability to move the body vertically as assessed with the MST. Change in MSH indicate metabolic risk. Also, with rapid reduction in muscle strength and function and no change in energy intake, an increased weight may start a negative spiral towards an even higher BMI, leading to further decline in MSH.
Change in MSH and obesity and type 2 diabetes
Our data presented in Additional file 2 shows that patients with obesity (BMI ≥ 30) had a higher MSH at baseline than the group with overweight and normal BMI and that the body height did not differ between these groups. Furthermore, the obese patients showed less improvement of MSH during intervention and a steeper decline at long-term follow-up. More information about baseline MSH and the MSH changes in different subgroups are presented in Additional files 2 and 3.
Change in SF-36 scores and SF-36 physical function items
In addition to patients’ increased metabolic and cardiovascular risks, mentioned above, lower self-reported quality of life in itself is associated with an increased risk of ill health. Comparing our patients’ mean SF-36 scores (broad standard deviations) in table 1b, the eight scales at baseline, with normative Swedish female population scores (narrow confidence interval) with women in the age groups in question of 45-54 years  the differences in scores are PF 19, RP 49, BP 31, GH 30, VT 32, SF 22, RE 33 and MH 16. Significantly improved values compared to baseline were in our study measured at both follow-ups (table 1b) and we suggest that an intervention of this kind should be further investigated in a larger context.
We analysed in more detail some questions in SF-36, PF which measures the ability of completing activities of moderate strain level where leg strength might have importance to the estimated function (table 1b). The ability to walk more than 2 km at a time without strain was the easiest task to achieve and keep until the long-term follow-up, but also those who improved their ability to escalate several stairs without large or small difficulty could maintain this ability long term. One third of the patients with reported large problems, mostly with knee bending, reported decreased limitations at the 3-month follow-up shortly after the period of group training. At long-term follow-up the positive effect on knee function had returned to baseline, indicating the need for regular mixed aerobic fitness and strength training. We conclude that the SF-36 items 3b, 3d, 3f and 3g are useful in clinical practice, both for screening of leg muscle strength and function, to enhance individually prescribed training, and be a help for setting goals.
Strengths and limitations of the study
The study population was representative of female patients in Swedish primary health care. The intervention program was received well by patients, despite substantial burdens of disease. There were no accidents requiring medical attention during or after the training sessions. All training sessions, except Nordic Walking, included music and in a recent study this was recommended to promote PA among sedentary individuals . The baseline and 3-month follow-up tests were distributed across three out of the four seasons because of the way in which patients were added to the program, and no seasonal influence was discovered. One limitation is that we have not measured SB or PA objectively, including measuring intensity with an accelerometer. Such measurements may have identified additional factors for maintaining MSH. Another limitation is that when presenting the long-term results from subgroups, each subgroup is comparatively small. Because this study is the first to describe long-term results of MSH change after an intervention has been completed, we believe that our results are valuable for planning future intervention studies, see Additional file 2. Another limitation is that the study did not investigate its findings in a population of men.
Advantages and possibilities with the maximal step-up test
For the diagnosis of sarcopenia, a consensus report recommends the use of the presence of both low muscle mass and low muscle function (strength or performance) . An algorithm recommended for sarcopenia in older individuals includes measurements of gait speed, grip strength and muscle mass. A recently published study showed that handgrip strength may not be an appropriate surrogate for lower-body strength, power or balance, and is proposed only to be used describing upper-body strength or functionality . MST is a weight-bearing test that in an integrated and functional way assesses leg-muscle strength, power, mobility, balance and coordination and is also indirectly a test of metabolic and cardiovascular risk as measuring the function of the body’s largest tissue with secretory capacity—the skeletal muscle.
Exercise has been recommended as a first-line treatment of degenerative joint disease of the hip and knee. The 30-s chair-stand test, 40 m fast-paced walk test and a stair-climb test were recently recommended as the minimal core set of performance-based tests for hip or knee osteoarthritis research and in clinical practice [29, 47]. New research recommends that exercise should more closely target the sensorimotor deficiencies and functional instability associated with the degenerative joint disease of the knee than in traditionally used training methods . MST has been proposed as a recommended performance-based test to assess leg strength and function and side difference in people diagnosed with knee osteoarthritis . Muscle strength, power and joint mobility are perishable, which is why repeated testing is needed, especially in patients with cardiovascular and metabolic diseases and symptomatic degenerative joint disease of the knee and hip, see also Additional file 3, section IV.