The effect of cardiorespiratory fitness assessment in preventive health checks: a randomised controlled trial

The effect of cardiorespiratory fitness assessment in preventive health checks: a randomised... Abstract Background Poor cardiorespiratory fitness (CRF) increases morbidity and mortality risks. Routine CRF assessment in clinical practice has thus been advocated, but little is known about the effect. In this study, we investigated the effect of CRF assessment on CRF in a preventive health check programme. Methods We used a randomised design, in which we invited 4153 middle-aged adults and included 2201 participants who received a preventive health check with CRF assessment (intervention) or without CRF assessment (control). After 1 year, participants were examined. The primary outcomes were adjusted absolute (l/min), relative (ml/kg/min), and poor (%) CRF assessed by the Astrand-Ryhming test. We adjusted for baseline physical activity and intra-cluster correlation within general practices. Results A total of 901 attended the 1-year follow-up. In the intervention group, absolute CRF, relative CRF, and poor CRF were 2.7 l/min (95% confidence interval [CI]: 2.6; 2.8), 34.5 ml/kg/min (95% CI: 33.5; 35.4), and 31.0% (95% CI: 26.8; 35.2). In the control group, the corresponding figures were 2.8 l/min (95% CI: 2.7; 2.9), 35.2 ml/kg/min (95% CI: 34.2; 36.1), and 25.9% (95% CI: 21.8; 30.0). Adjusted absolute CRF was lower in the intervention group (−0.1 l/min [95% CI: −0.2; −0.01]). Adjusted relative CRF (−0.7 ml/kg/min [95% CI: −2.0; 0.6]) and poor CRF (5.0% [95% CI: −0.002; 10.1]) did not differ between groups. No differences were found when adjusting for potential confounding factors. Conclusion Preventive health checks with CRF assessment did not provide higher CRF levels at 1-year follow-up than preventive health checks without CRF assessment. Introduction Poor cardiorespiratory fitness (CRF) increases the risk of morbidity and mortality.1–3 Even small CRF improvements are associated with reduced mortality risk.3,4 Moreover, high CRF levels are associated with reduced risk of chronic diseases as well as improved physical and mental health.1,3 Consequently, it has been advocated, most recently by the American Heart Association,3 that routine CRF assessment should be implemented in clinical practice.5,6 CRF can be improved by increasing the amount or intensity of physical activity.1,3 Other CRF determinants include age, sex, genetics, obesity, smoking and morbidity.3 Studies indicate that measurement of physical activity per se may influence the physical activity behaviour by increasing people’s awareness and motivation.7,8 Accordingly, CRF assessment may motivate people to increase their physical activity level. Maintaining good health is another important factor that motivate people to engage in physical activity.9 As the realisation of poor CRF and unfavourable health status may synergistically encourage people to increase their physical activity level, an optimal setting for CRF assessment could be preventive health check programmes, which are already being offered in some countries.10–12 Only few previous studies have included CRF assessment in preventive health checks offered to the general population.13,14 A Swedish study, which was conducted in 1969, showed no effect of extensive health checks on all-cause or cause-specific mortality.13 In contrast, the Danish Ebeltoft Health Promotion Project, which was conducted in 1991, showed a positive effect of preventive health checks on cardiovascular risk and estimated life expectancy at 5-year follow-up.14 Neither of these studies reported on the effect on CRF or physical activity levels. As little is known about the effect of CRF assessment in population-based preventive health checks, we investigated the effect on CRF of CRF assessment in a preventive health check programme. Methods The trial protocol (NCT02224248) has previously been published.15 The trial was approved by the Danish Data Protection Agency (record number 2013-41-2527) and complies with the Declaration of Helsinki. Setting and participants We conducted a two-armed randomised trial, which was nested in the Danish population-based Check your Health Preventive Programme (CHPP). Eligible CHPP participants were allocated to receive either a health check with CRF assessment (intervention group) or a health check without CRF assessment (control group). The ongoing CHPP offers a single preventive health check to citizens aged 30–49 years in Randers Municipality,16 which has 100 000 residents and is characterised by a less favourable health and socioeconomic profile compared with the entire Danish population.17,18 The health checks were performed by trained health professionals at Randers Health Centre. Before initiation of the CHPP, 26 216 citizens in the target age group were identified using the Danish Civil Registration System.19 The identified individuals were randomised into five equal-sized groups (figure 1) to distribute the workload for the health professional and enable the conduction of several trials during the first years of the CHPP. Group three (n = 5249) was allocated to the present trial. Before trial initiation on 1 November 2014, 342 eligible citizens had emigrated; this left 4907 for further randomisation. The inclusion criteria were: (i) participation in the CHPP and (ii) written consent for the collected data to be used in research. The only exclusion criterion was terminal illness reported by the general practitioners (GPs). Participants were enrolled consecutively until 30 November 2015. Figure 1 View largeDownload slide Flow of study participants. (a) The study period lasted from 1 November 2014 to 30 November 2015. CHPP, Check your Health Preventive Programme; HC, health check; FU, follow-up Figure 1 View largeDownload slide Flow of study participants. (a) The study period lasted from 1 November 2014 to 30 November 2015. CHPP, Check your Health Preventive Programme; HC, health check; FU, follow-up Procedures The CHPP included a health check with behavioural and clinical measurements followed by recommendations for subsequent actions according to a risk profile generated at the end of the health check. The CHPP questionnaire included questions concerning physical activity, smoking, alcohol use and self-rated health.15 In the present study, the Stages of Change Questionnaire20 was added as baseline measurement. The clinical measures included height, weight, body mass index (BMI), waist circumference, blood pressure, biochemical measures (lipid profile, HbA1c), lung function and the European Systematic Coronary Risk Evaluation (SCORE).15,16 In case of poor self-rated health, alcohol risk behaviour, high SCORE risk, or abnormal values of blood pressure, biochemical measures, or lung function, participants were advised to book a health-promoting consultation with their GP. In case of high BMI, smoking, or self-reported physical inactivity, participants were offered to attend health behavioural courses at Randers Health Centre (e.g. healthy diet and physical activity course).16 The stratification algorithm and the health behavioural courses are described elsewhere.16 Intervention: CRF assessment In the intervention group, the health check included a CRF assessment. We employed a modified version of the Astrand-Ryhming cycle ergometer single-stage test to estimate the age- and sex-standardised CRF on the basis of heart rate (HR) and workload.15 We used an initial workload of 75 watts (women) and 100 watts (men) and a pedalling frequency of 60–70 rounds/min. The HR was monitored continuously and recorded at 5.5 min if a steady-state HR was achieved within a target interval of 120–170 beats/min. The test was terminated after 6 min. If the target interval was not attained within the first 2 min of pedalling, the workload was increased by 25 (women) or 50 (men) watts. The test continued until steady-state HR was reached and recorded (average test time: 8–10 min). For safety reasons and to avoid unreliable HR responses, contraindications for performing the ergometer test were blood pressure above 180 (systolic) or 110 (diastolic) mmHg, pacemaker and beta-blocker use.1 Follow-up Two-week follow-up Two weeks after the health check, participants answered the Stages of Change Questionnaire again. The Stages of Change describes an individual’s intention to change behaviour as a progression through five stages: pre-contemplation (no intention to engage in regular moderate physical activity within the next 6 months), contemplation (intention to engage in regular moderate physical activity within the next 6 months), preparation (intention to engage in regular moderate physical activity within the next 30 days), action (regular moderate physical activity has been performed for less than 6 months) and maintenance (regular moderate physical activity has been performed for 6 months or more).20,21 The pre-action stages (i.e. pre-contemplation, contemplation and preparation) represent the motivational phase, in which motivation (or intention) is theorised to increase.20 Accordingly, stage progression from pre-action stages reflects increased motivation and the intermediate outcome at 2-week follow-up was the percentage of baseline pre-action stage participants making stage progression. Stage progression to the maintenance stage, however, was considered invalid in view of our short follow-up period. We defined regular physical activity as at least 30 min (or 3 × 10 min) on at least 5 days/week and moderate intensity as producing increased HR, while still being able to talk.15 One-year follow-up One year after the trial initiation on 1 November 2015, all participants were re-invited to attend an examination and complete a questionnaire. The 1-year follow-up proceeded until 30 June 2016. The primary outcomes were absolute CRF (L O2/min), relative CRF (mL O2/kg/min) and the prevalence of poor CRF (lowest of the five benchmark Astrand categories).22 Secondary outcomes included self-reported physical inactivity prevalence at 1-year follow-up and self-rated physical and mental health changes from baseline to 1-year follow-up. Self-reported physical inactivity was defined as the lowest of four leisure-time physical activity categories assessed using the Saltin and Grimby Questionnaire.23,Self-rated physical and mental health was assessed using the second version of the 12-item Short-Form Health Survey (SF-12) from which a physical component summary (PCS) score and a mental component summary (MCS) score can be derived.24 The scores were calculated using the standard SF-12 scoring algorithm.24 Randomisation and blinding Randomisation was based on postal addresses derived from the Danish Civil Registration System.15,19 The randomisation and implementation was performed by a data manager. Participants were enrolled automatically if they met the inclusion criteria. Intervention, outcomes, group assignment and follow-up assessments were not disclosed to the participants during the study period. The GPs and the health professionals performing the health checks and the health behavioural courses were unblinded. At 1-year follow-up, an independent unblinded health professional performed the outcome assessment. Sample size The required sample size was 1500 participants based on the following assumptions: 1:1 randomisation, a false positive error rate of 5%, a power of 80%, an intra-cluster correlation coefficient of 0.05, 30% loss to follow-up and categorical analysis with a power to detect a difference of at least 10% in poor CRF prevalence between the study groups.15 Statistical analysis Statistical analyses were performed using the Stata 14.0 software package (StataCorp, College Station, TX, USA). Data are presented as mean ± standard deviation, median (interquartile range) or as absolute numbers and percentages. Linear regression was used for between-group comparison of continuous outcomes. Binomial linear regression with identity link was used to compare proportions of dichotomous outcomes. Analyses were stratified by sex or age group because motives to be physically active vary according to sex and age.9,25 CRF assessment may, therefore, appeal differently to men and women and across different age groups. We adjusted for baseline self-reported physical activity and accounted for intra-cluster correlation within general practices (n = 40) as risk factor management differs between GPs.26 We refrained from employing multiple imputation to handle missing outcome data due to the extensive loss to follow-up. Instead, dropout analyses using baseline data were performed with the unpaired t-test (continuous data) or χ2 test (categorical data) to reveal potential confounding factors, for which we adjusted in a post-hoc sensitivity analysis.27 We also performed an explorative sub-analysis of within-group CRF changes in the intervention group using the paired t-test (absolute and relative CRF) or McNemar’s test (poor CRF). The randomisation groups were preserved in all analyses, and the statistical significance level was set at 5%. Results Study population Of the invited 2234 intervention participants and 1919 control participants, 1194 (53%) and 1007 (52%) participants were included (figure 1). Baseline characteristics between study groups did not differ (table 1). The median age (interquartile range) was 45.2 years (39.9–49.6) in the intervention group and 44.9 years (39.9–49.1) in the control group. In both groups, 49% of participants were men. Table 1 Baseline characteristics of the study population by study group   N  Intervention  N  Control  N (%)    1194 (54.2)    1007 (45.8)  Age, years  1194  45.2 (39.9; 49.6)  1007  44.9 (39.9; 49.1)  Sex (male) (%)  1194  585 (49.0)  1007  493 (49.0)  Sociodemographic characteristics, N (%)   Immigrant/descendant  1193  69 (5.8)  1,001  79 (7.9)   Single  1194  296 (24.8)  1,007  236 (23.4)   Educational level  1176    982     ≤10 years    172 (14.6)    147 (15.0)   >10 to ≤ 15 years    618 (52.6)    495 (50.4)   >15 years    386 (32.8)    340 (34.6)  Behavioural measures   Smoker, N (%)  1123  266 (23.7)  958  249 (26.0)   Physical activity category, N (%)  1160    970     Physical inactivity    202 (17.4)    198 (20.4)   Low physical activity    646 (55.7)    484 (50.0)   Moderate physical activity    292 (25.2)    271 (27.9)   High physical activity    20 (1.7)    17 (1.8)   Stage of Change, N (%)  1135    935     Maintenance    639 (56.3)    547 (58.5)   Action    147 (13.0)    113 (12.1)   Preparation    141 (12.4)    105 (11.2)   Contemplation    96 (8.5)    75 (8.0)   Pre-contemplation    112 (9.9)    95 (10.2)   SF-12           MCS score  1110  53.0 (46.7; 57.2)  908  52.4 (45.3; 57.2)   PCS score  1110  52.9 (47.0; 56.1)  908  52.9 (47.0; 56.1)  Clinical measures   BMI, kg/m2  1191  26.5 (23.5; 29.8)  1005  26.4 (23.7; 29.4)   WC, cm  1190  91.5 (82.0; 102.0)  1003  91.5 (83.0; 101.0)   SBP, mmHg  1192  124.0 (114.0; 135.0)  1005  124.0 (114.0; 135.0)   DBP, mmHg  1192  82.0 (76.0; 89.0)  1005  82.0 (76.0; 89.0)   TC, mM  1194  4.9 (4.3; 5.5)  1006  5.0 (4.4; 5.6)   CRF, l/min, mean (SD)  1078  2.5 (0.7)    –   ml/kg/min, mean (SD)  1078  31.5 (9.0)    –   poor, % (N)  1078  39.4 (436)        N  Intervention  N  Control  N (%)    1194 (54.2)    1007 (45.8)  Age, years  1194  45.2 (39.9; 49.6)  1007  44.9 (39.9; 49.1)  Sex (male) (%)  1194  585 (49.0)  1007  493 (49.0)  Sociodemographic characteristics, N (%)   Immigrant/descendant  1193  69 (5.8)  1,001  79 (7.9)   Single  1194  296 (24.8)  1,007  236 (23.4)   Educational level  1176    982     ≤10 years    172 (14.6)    147 (15.0)   >10 to ≤ 15 years    618 (52.6)    495 (50.4)   >15 years    386 (32.8)    340 (34.6)  Behavioural measures   Smoker, N (%)  1123  266 (23.7)  958  249 (26.0)   Physical activity category, N (%)  1160    970     Physical inactivity    202 (17.4)    198 (20.4)   Low physical activity    646 (55.7)    484 (50.0)   Moderate physical activity    292 (25.2)    271 (27.9)   High physical activity    20 (1.7)    17 (1.8)   Stage of Change, N (%)  1135    935     Maintenance    639 (56.3)    547 (58.5)   Action    147 (13.0)    113 (12.1)   Preparation    141 (12.4)    105 (11.2)   Contemplation    96 (8.5)    75 (8.0)   Pre-contemplation    112 (9.9)    95 (10.2)   SF-12           MCS score  1110  53.0 (46.7; 57.2)  908  52.4 (45.3; 57.2)   PCS score  1110  52.9 (47.0; 56.1)  908  52.9 (47.0; 56.1)  Clinical measures   BMI, kg/m2  1191  26.5 (23.5; 29.8)  1005  26.4 (23.7; 29.4)   WC, cm  1190  91.5 (82.0; 102.0)  1003  91.5 (83.0; 101.0)   SBP, mmHg  1192  124.0 (114.0; 135.0)  1005  124.0 (114.0; 135.0)   DBP, mmHg  1192  82.0 (76.0; 89.0)  1005  82.0 (76.0; 89.0)   TC, mM  1194  4.9 (4.3; 5.5)  1006  5.0 (4.4; 5.6)   CRF, l/min, mean (SD)  1078  2.5 (0.7)    –   ml/kg/min, mean (SD)  1078  31.5 (9.0)    –   poor, % (N)  1078  39.4 (436)      Notes: Values are median (interquartile range) unless otherwise specified. Data are derived from the Check your Health Preventive Programme. Sociodemographic information was acquired from Statistics Denmark. SF-12, Short-Form 12, Health Survey; PCS, physical component summary; MCS, mental component summary; BMI, body mass index; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; CRF, cardiorespiratory fitness; SD, standard deviation. Motivation The sub-analysis of stage progression at 2-week follow-up included 183 intervention participants and 129 control participants (figure 1). Overall, we found no significant difference in stage progression (table 2). Table 2 Stage progression at 2-week follow-up, self-rated health changes at one-year follow-up and adjusted estimates of between-group differences stratified by sex and age groups   Intervention Crude  Control Crude  Adjusted differencea    N  2-week follow-up  N  2-week follow-up  2-week follow-up  Stage progression % (95% CI)  Total  183  45.9 (38.7;53.1)  129  37.2 (28.8;45.6)  8.3 (−2.6;19.2)  Men  97  44.3 (34.4;54.2)  53  35.8 (22.9;48.8)  6.4 (−10.4;23.2)  Women  86  47.7 (37.1;58.2)  76  38.2 (27.;49.1)  11.5 (−2.9;26.0)  30–30 years  81  42.0 (31.2;52.7)  57  40.4 (27.6;53.1)  2.1 (−11.7;16.0)  40–49 years  102  49.0 (39.3;58.7)  72  34.7 (23.7;45.7)  13.5 (−4.4;31.5)    N  1-year follow-up  N  1-year follow-up  1-year follow-up  SF-12 mean (95% CI)  MCS score change   Total  465  0.1 (−0.7;0.9)  430  −0.4 (−1.3;0.4)  0.5 (−0.9;1.8)   Men  234  −0.3 (−1.3;0.8)  219  −0.4 (−1.5;0.7)  0.1 (−1.5;1.6)   Women  231  0.5 (−0.7;1.7)  211  −0.5 (−1.7;0.8)  0.8 (−0.8;2.5)   30–30 years  190  0.6 (−0.7;1.9)  173  −0.3 (−1.6;1.1)  0.7 (−1.3;2.8)   40–49 years  275  −0.2 (−1.2;0.8)  257  −0.5 (−1.6;0.5)  0.3 (−1.0;1.6)  PCS score change             Total  512  −0.05 (−0.7;0.6)  430  −0.1 (−0.8;0.6)  0.1 (−0.8;1.0)   Men  234  −0.1 (−0.9;0.8)  219  −0.2 (−1.1;0.7)  0.2 (−1.0;1.4)   Women  231  −0.02 (−1.0;0.9)  211  0.02 (−1.0;1.0)  0.03 (−1.3;1.4)   30–30 years  190  −0.9 (−1.9;0.1)  173  −0.2 (−1.3;0.8)  −0.6 (−2.2;1.0)   40–49 years  275  0.6 (−0.2;1.4)  257  −0.002 (−0.8;0.8)  0.6 (−0.6;1.8)    Intervention Crude  Control Crude  Adjusted differencea    N  2-week follow-up  N  2-week follow-up  2-week follow-up  Stage progression % (95% CI)  Total  183  45.9 (38.7;53.1)  129  37.2 (28.8;45.6)  8.3 (−2.6;19.2)  Men  97  44.3 (34.4;54.2)  53  35.8 (22.9;48.8)  6.4 (−10.4;23.2)  Women  86  47.7 (37.1;58.2)  76  38.2 (27.;49.1)  11.5 (−2.9;26.0)  30–30 years  81  42.0 (31.2;52.7)  57  40.4 (27.6;53.1)  2.1 (−11.7;16.0)  40–49 years  102  49.0 (39.3;58.7)  72  34.7 (23.7;45.7)  13.5 (−4.4;31.5)    N  1-year follow-up  N  1-year follow-up  1-year follow-up  SF-12 mean (95% CI)  MCS score change   Total  465  0.1 (−0.7;0.9)  430  −0.4 (−1.3;0.4)  0.5 (−0.9;1.8)   Men  234  −0.3 (−1.3;0.8)  219  −0.4 (−1.5;0.7)  0.1 (−1.5;1.6)   Women  231  0.5 (−0.7;1.7)  211  −0.5 (−1.7;0.8)  0.8 (−0.8;2.5)   30–30 years  190  0.6 (−0.7;1.9)  173  −0.3 (−1.6;1.1)  0.7 (−1.3;2.8)   40–49 years  275  −0.2 (−1.2;0.8)  257  −0.5 (−1.6;0.5)  0.3 (−1.0;1.6)  PCS score change             Total  512  −0.05 (−0.7;0.6)  430  −0.1 (−0.8;0.6)  0.1 (−0.8;1.0)   Men  234  −0.1 (−0.9;0.8)  219  −0.2 (−1.1;0.7)  0.2 (−1.0;1.4)   Women  231  −0.02 (−1.0;0.9)  211  0.02 (−1.0;1.0)  0.03 (−1.3;1.4)   30–30 years  190  −0.9 (−1.9;0.1)  173  −0.2 (−1.3;0.8)  −0.6 (−2.2;1.0)   40–49 years  275  0.6 (−0.2;1.4)  257  −0.002 (−0.8;0.8)  0.6 (−0.6;1.8)  Note: Data are derived from the Check your Health Preventive Programme. a Estimates of mean difference (linear regression) or absolute percentage difference (binomial linear regression with identity link) between the intervention and control group at 2-week follow-up and 1-year follow-up adjusted for baseline physical activity and intra-cluster correlation within general practices. CI, confidence interval. Cardiorespiratory fitness At 1-year follow-up, 505 intervention participants and 484 control participants attended the examination (figure 1). Of these, 44 participants in each group were excluded due to missing CRF or baseline self-reported physical activity level. When comparing the study groups at follow-up, the adjusted absolute CRF was lower in the intervention group (−0.1 l/min [95% confidence interval (CI): −0.2; −0.01]) (table 3). No differences were found in the adjusted relative CRF or poor CRF prevalence. The stratified analyses showed that the adjusted absolute CRF was lower in the intervention group among men (−0.2 l/min [95% CI: −0.3; −0.02]) and 40–49 year-old participants (−0.2 l/min [95% CI: −0.3; −0.03]), whereas the adjusted poor CRF prevalence was higher in the intervention group among women (7.2% [95% CI: 1.9; 12.5]). In the intervention group, both absolute and relative CRF increased from baseline to follow-up regardless of sex and age group (Supplementary material table S4). The poor CRF prevalence tended to decrease, but only significantly among 40–49 year-old participants. Table 3 CRF levels and physical inactivity prevalence at 1-year follow-up and adjusted estimates of between-group differences stratified by sex or age group   Intervention Crude  Control Crude  Adjusted differencea    N  1-year follow-up  N  1-year follow-up  1-year follow-up  Absolute CRF, l/min mean (95% CI)   Total  461  2.7 (2.6; 2.8)  440  2.8 (2.7; 2.9)  −0.1 (−0.2; −0.01)*   Men  233  3.0 (2.9;3.1)  227  3.2 (3.1;3.4)  −0.2 (−0.3; −0.02)*   Women  228  2.4 (2.3;2.5)  213  2.4 (2.3;2.5)  −0.005 (−0.1;0.1)   30–39 years  192  2.9 (2.8;3.0)  178  2.9 (2.8;3.0)  −0.04 (−0.2;0.1)   40–49 years  269  2.6 (2.5;2.7)  262  2.8 (2.7;2.9)  −0.2 (−0.3; −0.03)*  Relative CRF, ml/kg/min mean (95% CI)   Total  461  34.5 (33.5;35.4)  440  35.2 (34.2;36.1)  −0.7 (−2.0;0.6)   Men  233  34.7 (33.5;36.0)  227  36.8 (35.4;38.2)  −1.9 (−3.9;0.08)   Women  228  34.3 (33.0;35.6)  213  33.4 (32.1;34.6)  0.6 (−1.1;2.3)   30–39 years  192  36.8 (35.2;38.4)  178  36.0 (34.6;37.4)  0.4 (−1.5;2.4)   40–49 years  269  32.8 (31.6;33.9)  262  34.6 (33.3;35.9)  −1.5 (−3.0;0.04)  Poor CRF % (95% CI)   Total  461  31.0 (26.8;35.2)  440  25.9 (21.8;30.0)  5.0 (−0.002;10.1)   Men  233  39.5 (33.2;45.8)  227  34.4 (28.2;40.5)  2.6 (−5.1;10.4)   Women  228  22.4 (17.0;27.8)  213  16.9 (11.9;21.9)  7.2 (1.9;12.5)*   30–39 years  192  32.8 (26.2;39.5)  178  29.2 (22.5;35.9)  4.5 (−4.8.;13.8)   40–49 years  269  29.7 (24.3;35.2)  262  23.7 (18.5;28.8)  5.5 (−0.005;11.5)  Physical inactivity % (95% CI)   Total  588  14.8 (11.9;17.7)  541  17.2 (14.0;20.4)  −0.7 (−3.3;1.9)   Men  286  18.9 (14.3;23.4)  265  15.1 (10.8;19.4)  −1.9 (−5.5;1.8)   Women  302  10.9 (7.4;14.4)  276  19.2 (14.6;23.8)  −2.6 (−6.2;1.0)   30–39 years  248  18.1 (13.3;22.9)  224  18.8 (13.6;23.9)  −0.6 (−4.7;3.4)   40–49 years  340  12.4 (8.9;15.9)  317  16.1 (12.0;20.1)  −0.8 (−4.3;2.6)    Intervention Crude  Control Crude  Adjusted differencea    N  1-year follow-up  N  1-year follow-up  1-year follow-up  Absolute CRF, l/min mean (95% CI)   Total  461  2.7 (2.6; 2.8)  440  2.8 (2.7; 2.9)  −0.1 (−0.2; −0.01)*   Men  233  3.0 (2.9;3.1)  227  3.2 (3.1;3.4)  −0.2 (−0.3; −0.02)*   Women  228  2.4 (2.3;2.5)  213  2.4 (2.3;2.5)  −0.005 (−0.1;0.1)   30–39 years  192  2.9 (2.8;3.0)  178  2.9 (2.8;3.0)  −0.04 (−0.2;0.1)   40–49 years  269  2.6 (2.5;2.7)  262  2.8 (2.7;2.9)  −0.2 (−0.3; −0.03)*  Relative CRF, ml/kg/min mean (95% CI)   Total  461  34.5 (33.5;35.4)  440  35.2 (34.2;36.1)  −0.7 (−2.0;0.6)   Men  233  34.7 (33.5;36.0)  227  36.8 (35.4;38.2)  −1.9 (−3.9;0.08)   Women  228  34.3 (33.0;35.6)  213  33.4 (32.1;34.6)  0.6 (−1.1;2.3)   30–39 years  192  36.8 (35.2;38.4)  178  36.0 (34.6;37.4)  0.4 (−1.5;2.4)   40–49 years  269  32.8 (31.6;33.9)  262  34.6 (33.3;35.9)  −1.5 (−3.0;0.04)  Poor CRF % (95% CI)   Total  461  31.0 (26.8;35.2)  440  25.9 (21.8;30.0)  5.0 (−0.002;10.1)   Men  233  39.5 (33.2;45.8)  227  34.4 (28.2;40.5)  2.6 (−5.1;10.4)   Women  228  22.4 (17.0;27.8)  213  16.9 (11.9;21.9)  7.2 (1.9;12.5)*   30–39 years  192  32.8 (26.2;39.5)  178  29.2 (22.5;35.9)  4.5 (−4.8.;13.8)   40–49 years  269  29.7 (24.3;35.2)  262  23.7 (18.5;28.8)  5.5 (−0.005;11.5)  Physical inactivity % (95% CI)   Total  588  14.8 (11.9;17.7)  541  17.2 (14.0;20.4)  −0.7 (−3.3;1.9)   Men  286  18.9 (14.3;23.4)  265  15.1 (10.8;19.4)  −1.9 (−5.5;1.8)   Women  302  10.9 (7.4;14.4)  276  19.2 (14.6;23.8)  −2.6 (−6.2;1.0)   30–39 years  248  18.1 (13.3;22.9)  224  18.8 (13.6;23.9)  −0.6 (−4.7;3.4)   40–49 years  340  12.4 (8.9;15.9)  317  16.1 (12.0;20.1)  −0.8 (−4.3;2.6)  Note: Data are derived from the Check your Health Preventive Programme. a Estimates of mean difference (linear regression) or absolute percentage difference (binomial linear regression with identity link) between the intervention and control group at 1-year follow-up adjusted for baseline physical activity and intra-cluster correlation within general practices. *Indicates P < 0.05 (intervention vs. control group). CI, confidence interval; CRF, cardiorespiratory fitness. Self-reported physical inactivity and self-rated health No differences were found for the 1-year adjusted self-reported physical inactivity prevalence (table 3) or the adjusted changes in PCS or MCS scores (table 2). Dropout and sensitivity analyses A post-hoc analysis comparing the baseline characteristics of intervention participants (n = 461) and control participants (n = 440) included in the primary analysis showed a lower percentage of singles and individuals with low educational level in the control group (figure 1). A post-hoc sensitivity analysis adjusting for these potential confounding factors showed no between-group differences in absolute CRF (−0.09 l/min [95% CI: −0.19;0.01]), relative CRF (−0.2 ml/kg/min [95% CI: −1.4;1.0]), or poor CRF prevalence (3,5% [95% CI: −2.6;9.7]). We also compared the baseline characteristics of analysed participants with those who were not analysed (i.e. lost to follow-up or excluded due to missing data). In both groups, the not-analysed participants were more likely to be physically inactive, smokers and to have a low PCS score. In the control group, the not-analysed participants were also more likely to have a low education level and higher levels of BMI, waist circumference, and diastolic blood pressure. In the intervention group, the not-analysed participants were less likely to be single. Discussion Preventive health checks with CRF assessment did not provide higher CRF levels at 1-year follow-up than preventive health checks without CRF assessment. Furthermore, no differences were found in self-reported physical inactivity prevalence or in self-rated physical or mental health changes. Additionally, the proportion of pre-action stage participants with increased motivation to perform regular moderate physical activity at 2-week follow-up did not differ between the groups. Our study provides novel insights into the effect of including CRF assessment in preventive health checks on CRF levels. The effect of measuring CRF and physical activity levels has received very little attention.7 A recent systematic review and meta-analysis, including five randomised trials focusing on physical activity behaviour, showed a small positive effect of questionnaire-based physical activity measurements on self-reported physical activity.8 The follow-up periods of the included studies ranged from 1 week to 3 months, with the exception of one study with 6 months of follow-up.8 Evidence suggests that a majority of people who succeed in increasing their physical activity level may relapse to a less active or inactive status within 6 months.28 This may partly explain the lack of effect on CRF at 1-year follow-up in our study. Our findings are consistent with a study by Sluijs et al., which showed no effect of accelerometer-based measurements in an intervention on objectively measured physical activity at 6-month follow-up.29 We could only explore CRF changes in the intervention group as CRF was not assessed in the control group at baseline; a post-hoc analysis showed small increases in CRF levels from baseline to 1-year follow-up. Generally, physical activity interventions in primary care and community settings have shown limited long-term effects on CRF and physical activity levels.30–32 Surprisingly, both positive and negative effects on self-rated health have been observed during physical activity promoting interventions.33,34 We found no differences in self-rated physical or mental health status. Importantly, these findings suggest that CRF assessment did not lead to unintended physical or mental harm. Strengths and limitations The strengths of our study are the randomised design, the objective CRF assessment and the relatively long follow-up period compared with previous studies.8 The sustainability of an intervention effect is key, because temporary CRF improvements are unlikely to provide sustained health benefits. We used a real-life setting and recruited participants without any screening prior to inclusion, and more than 50% of the invited individuals attended the CHPP in both study arms. This attendance rate is high compared with other real-life health check programmes.35 Despite the emigration of 342 eligible participants before randomisation and a CHPP non-attendance rate around 45%, the baseline characteristics of the study groups did not differ significantly. The risk of selection bias is, therefore, considered to be low. Our study also has some important limitations. The dropout rate above 50% at follow-up increased the risk of attrition bias. Our dropout analysis indicated that control participants who were included in the primary analysis had a more favourable social profile than intervention participants. As lower social status is associated with lower CRF,36,37 attrition bias is likely to explain the higher absolute CRF level in the control group. This was also supported by the post-hoc sensitivity analysis. The study power was reduced by loss to follow-up and exclusion of certain participants, which increases the risk of a type II error. In particular, the sub-analysis of stage progression had limited power due to the unexpected small percentage of participants categorising themselves in the pre-action stages at baseline. In addition, the Stages of Change Questionnaire has not been validated to measure changes over time. Our estimates of the effect on motivation should, therefore, be interpreted with caution. The poor CRF prevalence tended to be higher in the intervention group. However, the difference between the study groups was attenuated when further adjusting for social factors. Loss of power is, therefore, unlikely to have caused a type II error with respect to CRF. Another possible contributing factor to the little difference between the study groups could be contamination. While only four women in the control group received a CRF assessment, intervention participants may have influenced control participants through social groups and workplaces. Furthermore, health professionals and GPs may have been more attentive to control participants than would be expected in usual care without CRF assessment. This could have attenuated a potential intervention effect. Another limitation was the relatively wide error margin of the Astrand-Ryhming test.38 This imprecision may have further reduced any differences between the study groups. Overall, as some limitations may have biased our results towards the null, we cannot preclude that a true intervention effect was overlooked. Generalisability and clinical impact CHPP attenders generally had a higher social status than non-attenders.39 As these results agree with previous findings, our study population likely represents the attenders that would be expected in a real-life population-based setting.40 Still, our findings may not generalise to more disadvantageous populations given that the participants who were not included in the analyses overall had a less favourable sociodemographic and clinical profile. CRF assessment is highly topical in view of the recent scientific statement from the American Heart Association.3 Routine CRF assessment as part of preventive health checks was feasible in our study. Yet, our results imply that adding objective measurement of CRF to a traditional risk factor assessment is unlikely to improve the effectiveness of population-based preventive health checks in terms of increasing CRF. Less resource-demanding non-exercise-based methods may provide valid alternatives.3 A pertinent question is how identification of poor CRF should be reflected in the subsequent care. Additionally, it remains to be investigated if CRF assessment should be confined to high-risk individuals who may derive particular benefit. Conclusion Preventive health checks with CRF assessment did not afford higher CRF levels, lower self-reported physical inactivity prevalence, or improved self-rated physical or mental health at 1-year follow-up compared with preventive health checks without CRF assessment. Key points CRF assessment in preventive health checks has received little attention in research. This study concludes that CRF assessment as part of a preventive health check did not result in higher CRF levels at 1-year follow-up. Our results imply that adding measurement of CRF to a traditional risk factor assessment is unlikely to improve the effectiveness of population-based preventive health checks in terms of increasing CRF. Several questions remain unanswered: Should CRF assessment be broadly utilised or confined to high-risk individuals? Which subsequent care should be offered for identified poor CRF? Public health policy and practice should consider these issues when planning future interventions aiming to decrease health risks resulting from poor CRF. Supplementary data Supplementary data are available at EURPUB online. Conflicts of interest: None declared. Acknowledgements The authors wish to thank the staff at Randers Health Centre and the Check Your Health group at Aarhus University. This work was supported by Aarhus University, the Central Denmark Region (file numbers 1-15-1-72-13-09 and 1-30-72-141-12), the municipality of Randers, the Danish foundation TrygFonden and the Aase and Ejnar Danielsen’s Foundation. References 1 Pescatello LS. ACSM's Guidelines for Exercise Testing and Prescription , 9th edn Philadelphia, PA: Lippincott Williams & Wilkins; 2014. 2 Kodama S, Saito K, Tanaka S, et al.   Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women: a meta-analysis. JAMA  2009; 301: 2024– 35. Google Scholar CrossRef Search ADS PubMed  3 Ross R, Blair SN, Arena R, et al.   Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign: a scientific statement from the American Heart Association. Circulation  2016; 134: e653– 99. Google Scholar CrossRef Search ADS PubMed  4 Lee DC, Sui X, Artero EG, et al.   Long-term effects of changes in cardiorespiratory fitness and body mass index on all-cause and cardiovascular disease mortality in men: the Aerobics Center Longitudinal Study. Circulation  2011; 124: 2483– 90. Google Scholar CrossRef Search ADS PubMed  5 Myers J, McAuley P, Lavie CJ, et al.   Physical activity and cardiorespiratory fitness as major markers of cardiovascular risk: their independent and interwoven importance to health status. Prog Cardiovasc Dis  2015; 57: 306– 14. Google Scholar CrossRef Search ADS PubMed  6 Franklin BA. Exercise capacity: a crystal ball in forecasting future health outcomes?. Phys Sportsmed  2009; 37: 154– 6. Google Scholar CrossRef Search ADS PubMed  7 French DP, Sutton S. Reactivity of measurement in health psychology: how much of a problem is it? What can be done about it?. Br J Health Psychol  2010; 15: 453– 68. Google Scholar CrossRef Search ADS PubMed  8 Rodrigues AM, O'Brien N, French DP, et al.   The question-behavior effect: genuine effect or spurious phenomenon? A systematic review of randomized controlled trials with meta-analyses. Health Psychol  2015; 34: 61– 78. Google Scholar CrossRef Search ADS PubMed  9 Zunft H-F, Friebe D, Seppelt B, et al.   Perceived benefits and barriers to physical activity in a nationally representative sample in the European Union. Public Health Nutr  1999; 2: 153– 60. Google Scholar PubMed  10 Dalton AR, Soljak M. The nationwide systematic prevention of cardiovascular disease: the UK's health check programme. J Ambul Care Manage  2012; 35: 206– 15. Google Scholar CrossRef Search ADS PubMed  11 Sox HC. The health checkup. Was it ever effective? Could it be effective?. JAMA  2013; 309: 2496– 7. Google Scholar CrossRef Search ADS PubMed  12 Geue C, Lewsey JD, MacKay DF, et al.   Scottish Keep Well health check programme: an interrupted time series analysis. J Epidemiol Community Health  2016; 70: 924– 9. Google Scholar CrossRef Search ADS PubMed  13 Theobald H, Bygren LO, Carstensen J, et al.   Effects of an assessment of needs for medical and social services on long-term mortality: a randomized controlled study. Int J Epidemiol  1998; 27: 194– 8. Google Scholar CrossRef Search ADS PubMed  14 Lauritzen T, Jensen MS, Thomsen JL, et al.   Health tests and health consultations reduced cardiovascular risk without psychological strain, increased healthcare utilization or increased costs. An overview of the results from a 5-year randomized trial in primary care. The Ebeltoft Health Promotion Project (EHPP). Scand J Public Health  2008; 36: 650– 61. Google Scholar CrossRef Search ADS PubMed  15 Hoj K, Skriver MV, Hansen AL, et al.   Effect of including fitness testing in preventive health checks on cardiorespiratory fitness and motivation: study protocol of a randomized controlled trial. BMC Public Health  2014; 14: 1057. Google Scholar CrossRef Search ADS PubMed  16 Maindal HT, Stovring H, Sandbaek A. Effectiveness of the population-based Check your health preventive programme conducted in primary care with 4 years follow-up [the CORE trial]: study protocol for a randomised controlled trial. Trials  2014; 15: 341. Google Scholar CrossRef Search ADS PubMed  17 Christensen AI, Davidsen M, Ekholm O, et al.   Danskernes sundhed. Tal fra den Nationale Sundhedsprofil 2010 og 2013. [The Health of the Danes. Figures from the National Health Profile 2010 and 2013]. Available at: http://www.danskernessundhed.dk/ (5 July 2016, date last accessed). 18 Danmarks Statistik. [Statistics Denmark]. Available at: http://www.statistikbanken.dk/statbank5a/default.asp?w=1920 (9 September 2016, date last accessed). 19 Schmidt M, Pedersen L, Sorensen HT. The Danish Civil Registration System as a tool in epidemiology. Eur J Epidemiol  2014; 29: 541– 9. Google Scholar CrossRef Search ADS PubMed  20 Nigg CR. There is more to stages of exercise than just exercise. Exerc Sport Sci Rev  2005; 33: 32– 5. Google Scholar PubMed  21 Nigg CR, Geller KS, Motl RW, et al.   A research agenda to examine the efficacy and relevance of the transtheoretical model for physical activity behavior. Psychol Sport Exerc  2011; 12: 7– 12. Google Scholar CrossRef Search ADS PubMed  22 Astrand I. Aerobic work capacity in men and women with special reference to age. Acta Physiol Scand Suppl  1960; 49: 1– 92. Google Scholar PubMed  23 Saltin B, Grimby G. Physiological analysis of middle-aged and old former athletes. comparison with still active athletes of the same ages. Circulation  1968; 38: 1104– 15. Google Scholar CrossRef Search ADS PubMed  24 Ware JE, Kosinski M, Turner-Bowker DM, Gandeck B, eds. Useŕs Manual for the SF-12 v2TM Health Survey (with a Supplement Documenting the SF-12 Health Survey) . Lincoln, RI: QualityMetric Incorporated, 2007. 25 Skov-Ettrup LS, Petersen CB, Curtis T, et al.   Why do people exercise? A cross-sectional study of motives to exercise among Danish adults. Public Health  2014; 128: 482– 4. Google Scholar CrossRef Search ADS PubMed  26 Dallongeville J, Banegas JR, Tubach F, et al.   Survey of physicians' practices in the control of cardiovascular risk factors: The EURIKA study. Eur J Prev Cardiol  2012; 19: 541– 50. Google Scholar CrossRef Search ADS PubMed  27 Groenwold RH, Moons KG, Vandenbroucke JP. Randomized trials with missing outcome data: how to analyze and what to report. CMAJ  2014; 186: 1153– 7. Google Scholar CrossRef Search ADS PubMed  28 Amireault S, Godin G, Vezina-Im L. Determinants of physical activity maintenance: a systematic review and meta-analyses. Health Psychol Rev  2013; 7: 55– 91. Google Scholar CrossRef Search ADS   29 van Sluijs EM, van Poppel MN, Twisk JW, van Mechelen W. Physical activity measurements affected participants' behavior in a randomized controlled trial. J Clin Epidemiol  2006; 59: 404– 11. Google Scholar CrossRef Search ADS PubMed  30 Orrow G, Kinmonth AL, Sanderson S, Sutton S. Effectiveness of physical activity promotion based in primary care: systematic review and meta-analysis of randomised controlled trials. BMJ  2012; 344: e1389. Google Scholar CrossRef Search ADS PubMed  31 Baker PR, Francis DP, Soares J, Weightman AL, Foster C. Community wide interventions for increasing physical activity. Cochrane Database Syst Rev  2015; 1: CD008366. Google Scholar PubMed  32 Baumann S, Toft U, Aadahl M, et al.   The long-term effect of screening and lifestyle counseling on changes in physical activity and diet: the Inter99 study - a randomized controlled trial. Int J Behav Nutr Phys Act  2015; 12: 33. Google Scholar CrossRef Search ADS PubMed  33 Kinmonth AL, Wareham NJ, Hardeman W, et al.   Efficacy of a theory-based behavioural intervention to increase physical activity in an at-risk group in primary care (ProActive UK): a randomised trial. Lancet  2008; 371: 41– 8. Google Scholar CrossRef Search ADS PubMed  34 Petersen CB, Severin M, Hansen AW, et al.   A population-based randomized controlled trial of the effect of combining a pedometer with an intervention toolkit on physical activity among individuals with low levels of physical activity or fitness. Prev Med  2012; 54: 125– 30. Google Scholar CrossRef Search ADS PubMed  35 Robson J, Dostal I, Sheikh A, et al.   The NHS health check in England: an evaluation of the first 4 years. BMJ Open  2016; 6: e008840. Google Scholar CrossRef Search ADS PubMed  36 Hoj K, Vinther Skriver M, Terkildsen Maindal H, et al.   High prevalence of poor fitness among Danish adults, especially among those with high cardiovascular mortality risk. Eur J Public Health  2016; 27: 569– 74. 37 Lindgren M, Borjesson M, Ekblom O, 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– 9. Google Scholar CrossRef Search ADS PubMed  38 Åstrand PO, Rodahl K. Textbook of Work Physiology: Physiological Bases of Exercise , 4th edn Champaign, IL: Human Kinetics; 2003. 39 Bjerregaard AL, Maindal HT, Bruun NH, Sandbaek A. Patterns of attendance to health checks in a municipality setting: the Danish ′Check your health preventive program′. Prev Med Rep  2016; 5: 175– 82. Google Scholar CrossRef Search ADS PubMed  40 Tolonen H, Ahonen S, Jentoft S, et al.   Differences in participation rates and lessons learned about recruitment of participants: the European Health Examination Survey Pilot Project. Scand J Public Health  2015; 43: 212– 9. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

The effect of cardiorespiratory fitness assessment in preventive health checks: a randomised controlled trial

Loading next page...
 
/lp/ou_press/the-effect-of-cardiorespiratory-fitness-assessment-in-preventive-ZXoI8iEhL5
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/ckx108
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Poor cardiorespiratory fitness (CRF) increases morbidity and mortality risks. Routine CRF assessment in clinical practice has thus been advocated, but little is known about the effect. In this study, we investigated the effect of CRF assessment on CRF in a preventive health check programme. Methods We used a randomised design, in which we invited 4153 middle-aged adults and included 2201 participants who received a preventive health check with CRF assessment (intervention) or without CRF assessment (control). After 1 year, participants were examined. The primary outcomes were adjusted absolute (l/min), relative (ml/kg/min), and poor (%) CRF assessed by the Astrand-Ryhming test. We adjusted for baseline physical activity and intra-cluster correlation within general practices. Results A total of 901 attended the 1-year follow-up. In the intervention group, absolute CRF, relative CRF, and poor CRF were 2.7 l/min (95% confidence interval [CI]: 2.6; 2.8), 34.5 ml/kg/min (95% CI: 33.5; 35.4), and 31.0% (95% CI: 26.8; 35.2). In the control group, the corresponding figures were 2.8 l/min (95% CI: 2.7; 2.9), 35.2 ml/kg/min (95% CI: 34.2; 36.1), and 25.9% (95% CI: 21.8; 30.0). Adjusted absolute CRF was lower in the intervention group (−0.1 l/min [95% CI: −0.2; −0.01]). Adjusted relative CRF (−0.7 ml/kg/min [95% CI: −2.0; 0.6]) and poor CRF (5.0% [95% CI: −0.002; 10.1]) did not differ between groups. No differences were found when adjusting for potential confounding factors. Conclusion Preventive health checks with CRF assessment did not provide higher CRF levels at 1-year follow-up than preventive health checks without CRF assessment. Introduction Poor cardiorespiratory fitness (CRF) increases the risk of morbidity and mortality.1–3 Even small CRF improvements are associated with reduced mortality risk.3,4 Moreover, high CRF levels are associated with reduced risk of chronic diseases as well as improved physical and mental health.1,3 Consequently, it has been advocated, most recently by the American Heart Association,3 that routine CRF assessment should be implemented in clinical practice.5,6 CRF can be improved by increasing the amount or intensity of physical activity.1,3 Other CRF determinants include age, sex, genetics, obesity, smoking and morbidity.3 Studies indicate that measurement of physical activity per se may influence the physical activity behaviour by increasing people’s awareness and motivation.7,8 Accordingly, CRF assessment may motivate people to increase their physical activity level. Maintaining good health is another important factor that motivate people to engage in physical activity.9 As the realisation of poor CRF and unfavourable health status may synergistically encourage people to increase their physical activity level, an optimal setting for CRF assessment could be preventive health check programmes, which are already being offered in some countries.10–12 Only few previous studies have included CRF assessment in preventive health checks offered to the general population.13,14 A Swedish study, which was conducted in 1969, showed no effect of extensive health checks on all-cause or cause-specific mortality.13 In contrast, the Danish Ebeltoft Health Promotion Project, which was conducted in 1991, showed a positive effect of preventive health checks on cardiovascular risk and estimated life expectancy at 5-year follow-up.14 Neither of these studies reported on the effect on CRF or physical activity levels. As little is known about the effect of CRF assessment in population-based preventive health checks, we investigated the effect on CRF of CRF assessment in a preventive health check programme. Methods The trial protocol (NCT02224248) has previously been published.15 The trial was approved by the Danish Data Protection Agency (record number 2013-41-2527) and complies with the Declaration of Helsinki. Setting and participants We conducted a two-armed randomised trial, which was nested in the Danish population-based Check your Health Preventive Programme (CHPP). Eligible CHPP participants were allocated to receive either a health check with CRF assessment (intervention group) or a health check without CRF assessment (control group). The ongoing CHPP offers a single preventive health check to citizens aged 30–49 years in Randers Municipality,16 which has 100 000 residents and is characterised by a less favourable health and socioeconomic profile compared with the entire Danish population.17,18 The health checks were performed by trained health professionals at Randers Health Centre. Before initiation of the CHPP, 26 216 citizens in the target age group were identified using the Danish Civil Registration System.19 The identified individuals were randomised into five equal-sized groups (figure 1) to distribute the workload for the health professional and enable the conduction of several trials during the first years of the CHPP. Group three (n = 5249) was allocated to the present trial. Before trial initiation on 1 November 2014, 342 eligible citizens had emigrated; this left 4907 for further randomisation. The inclusion criteria were: (i) participation in the CHPP and (ii) written consent for the collected data to be used in research. The only exclusion criterion was terminal illness reported by the general practitioners (GPs). Participants were enrolled consecutively until 30 November 2015. Figure 1 View largeDownload slide Flow of study participants. (a) The study period lasted from 1 November 2014 to 30 November 2015. CHPP, Check your Health Preventive Programme; HC, health check; FU, follow-up Figure 1 View largeDownload slide Flow of study participants. (a) The study period lasted from 1 November 2014 to 30 November 2015. CHPP, Check your Health Preventive Programme; HC, health check; FU, follow-up Procedures The CHPP included a health check with behavioural and clinical measurements followed by recommendations for subsequent actions according to a risk profile generated at the end of the health check. The CHPP questionnaire included questions concerning physical activity, smoking, alcohol use and self-rated health.15 In the present study, the Stages of Change Questionnaire20 was added as baseline measurement. The clinical measures included height, weight, body mass index (BMI), waist circumference, blood pressure, biochemical measures (lipid profile, HbA1c), lung function and the European Systematic Coronary Risk Evaluation (SCORE).15,16 In case of poor self-rated health, alcohol risk behaviour, high SCORE risk, or abnormal values of blood pressure, biochemical measures, or lung function, participants were advised to book a health-promoting consultation with their GP. In case of high BMI, smoking, or self-reported physical inactivity, participants were offered to attend health behavioural courses at Randers Health Centre (e.g. healthy diet and physical activity course).16 The stratification algorithm and the health behavioural courses are described elsewhere.16 Intervention: CRF assessment In the intervention group, the health check included a CRF assessment. We employed a modified version of the Astrand-Ryhming cycle ergometer single-stage test to estimate the age- and sex-standardised CRF on the basis of heart rate (HR) and workload.15 We used an initial workload of 75 watts (women) and 100 watts (men) and a pedalling frequency of 60–70 rounds/min. The HR was monitored continuously and recorded at 5.5 min if a steady-state HR was achieved within a target interval of 120–170 beats/min. The test was terminated after 6 min. If the target interval was not attained within the first 2 min of pedalling, the workload was increased by 25 (women) or 50 (men) watts. The test continued until steady-state HR was reached and recorded (average test time: 8–10 min). For safety reasons and to avoid unreliable HR responses, contraindications for performing the ergometer test were blood pressure above 180 (systolic) or 110 (diastolic) mmHg, pacemaker and beta-blocker use.1 Follow-up Two-week follow-up Two weeks after the health check, participants answered the Stages of Change Questionnaire again. The Stages of Change describes an individual’s intention to change behaviour as a progression through five stages: pre-contemplation (no intention to engage in regular moderate physical activity within the next 6 months), contemplation (intention to engage in regular moderate physical activity within the next 6 months), preparation (intention to engage in regular moderate physical activity within the next 30 days), action (regular moderate physical activity has been performed for less than 6 months) and maintenance (regular moderate physical activity has been performed for 6 months or more).20,21 The pre-action stages (i.e. pre-contemplation, contemplation and preparation) represent the motivational phase, in which motivation (or intention) is theorised to increase.20 Accordingly, stage progression from pre-action stages reflects increased motivation and the intermediate outcome at 2-week follow-up was the percentage of baseline pre-action stage participants making stage progression. Stage progression to the maintenance stage, however, was considered invalid in view of our short follow-up period. We defined regular physical activity as at least 30 min (or 3 × 10 min) on at least 5 days/week and moderate intensity as producing increased HR, while still being able to talk.15 One-year follow-up One year after the trial initiation on 1 November 2015, all participants were re-invited to attend an examination and complete a questionnaire. The 1-year follow-up proceeded until 30 June 2016. The primary outcomes were absolute CRF (L O2/min), relative CRF (mL O2/kg/min) and the prevalence of poor CRF (lowest of the five benchmark Astrand categories).22 Secondary outcomes included self-reported physical inactivity prevalence at 1-year follow-up and self-rated physical and mental health changes from baseline to 1-year follow-up. Self-reported physical inactivity was defined as the lowest of four leisure-time physical activity categories assessed using the Saltin and Grimby Questionnaire.23,Self-rated physical and mental health was assessed using the second version of the 12-item Short-Form Health Survey (SF-12) from which a physical component summary (PCS) score and a mental component summary (MCS) score can be derived.24 The scores were calculated using the standard SF-12 scoring algorithm.24 Randomisation and blinding Randomisation was based on postal addresses derived from the Danish Civil Registration System.15,19 The randomisation and implementation was performed by a data manager. Participants were enrolled automatically if they met the inclusion criteria. Intervention, outcomes, group assignment and follow-up assessments were not disclosed to the participants during the study period. The GPs and the health professionals performing the health checks and the health behavioural courses were unblinded. At 1-year follow-up, an independent unblinded health professional performed the outcome assessment. Sample size The required sample size was 1500 participants based on the following assumptions: 1:1 randomisation, a false positive error rate of 5%, a power of 80%, an intra-cluster correlation coefficient of 0.05, 30% loss to follow-up and categorical analysis with a power to detect a difference of at least 10% in poor CRF prevalence between the study groups.15 Statistical analysis Statistical analyses were performed using the Stata 14.0 software package (StataCorp, College Station, TX, USA). Data are presented as mean ± standard deviation, median (interquartile range) or as absolute numbers and percentages. Linear regression was used for between-group comparison of continuous outcomes. Binomial linear regression with identity link was used to compare proportions of dichotomous outcomes. Analyses were stratified by sex or age group because motives to be physically active vary according to sex and age.9,25 CRF assessment may, therefore, appeal differently to men and women and across different age groups. We adjusted for baseline self-reported physical activity and accounted for intra-cluster correlation within general practices (n = 40) as risk factor management differs between GPs.26 We refrained from employing multiple imputation to handle missing outcome data due to the extensive loss to follow-up. Instead, dropout analyses using baseline data were performed with the unpaired t-test (continuous data) or χ2 test (categorical data) to reveal potential confounding factors, for which we adjusted in a post-hoc sensitivity analysis.27 We also performed an explorative sub-analysis of within-group CRF changes in the intervention group using the paired t-test (absolute and relative CRF) or McNemar’s test (poor CRF). The randomisation groups were preserved in all analyses, and the statistical significance level was set at 5%. Results Study population Of the invited 2234 intervention participants and 1919 control participants, 1194 (53%) and 1007 (52%) participants were included (figure 1). Baseline characteristics between study groups did not differ (table 1). The median age (interquartile range) was 45.2 years (39.9–49.6) in the intervention group and 44.9 years (39.9–49.1) in the control group. In both groups, 49% of participants were men. Table 1 Baseline characteristics of the study population by study group   N  Intervention  N  Control  N (%)    1194 (54.2)    1007 (45.8)  Age, years  1194  45.2 (39.9; 49.6)  1007  44.9 (39.9; 49.1)  Sex (male) (%)  1194  585 (49.0)  1007  493 (49.0)  Sociodemographic characteristics, N (%)   Immigrant/descendant  1193  69 (5.8)  1,001  79 (7.9)   Single  1194  296 (24.8)  1,007  236 (23.4)   Educational level  1176    982     ≤10 years    172 (14.6)    147 (15.0)   >10 to ≤ 15 years    618 (52.6)    495 (50.4)   >15 years    386 (32.8)    340 (34.6)  Behavioural measures   Smoker, N (%)  1123  266 (23.7)  958  249 (26.0)   Physical activity category, N (%)  1160    970     Physical inactivity    202 (17.4)    198 (20.4)   Low physical activity    646 (55.7)    484 (50.0)   Moderate physical activity    292 (25.2)    271 (27.9)   High physical activity    20 (1.7)    17 (1.8)   Stage of Change, N (%)  1135    935     Maintenance    639 (56.3)    547 (58.5)   Action    147 (13.0)    113 (12.1)   Preparation    141 (12.4)    105 (11.2)   Contemplation    96 (8.5)    75 (8.0)   Pre-contemplation    112 (9.9)    95 (10.2)   SF-12           MCS score  1110  53.0 (46.7; 57.2)  908  52.4 (45.3; 57.2)   PCS score  1110  52.9 (47.0; 56.1)  908  52.9 (47.0; 56.1)  Clinical measures   BMI, kg/m2  1191  26.5 (23.5; 29.8)  1005  26.4 (23.7; 29.4)   WC, cm  1190  91.5 (82.0; 102.0)  1003  91.5 (83.0; 101.0)   SBP, mmHg  1192  124.0 (114.0; 135.0)  1005  124.0 (114.0; 135.0)   DBP, mmHg  1192  82.0 (76.0; 89.0)  1005  82.0 (76.0; 89.0)   TC, mM  1194  4.9 (4.3; 5.5)  1006  5.0 (4.4; 5.6)   CRF, l/min, mean (SD)  1078  2.5 (0.7)    –   ml/kg/min, mean (SD)  1078  31.5 (9.0)    –   poor, % (N)  1078  39.4 (436)        N  Intervention  N  Control  N (%)    1194 (54.2)    1007 (45.8)  Age, years  1194  45.2 (39.9; 49.6)  1007  44.9 (39.9; 49.1)  Sex (male) (%)  1194  585 (49.0)  1007  493 (49.0)  Sociodemographic characteristics, N (%)   Immigrant/descendant  1193  69 (5.8)  1,001  79 (7.9)   Single  1194  296 (24.8)  1,007  236 (23.4)   Educational level  1176    982     ≤10 years    172 (14.6)    147 (15.0)   >10 to ≤ 15 years    618 (52.6)    495 (50.4)   >15 years    386 (32.8)    340 (34.6)  Behavioural measures   Smoker, N (%)  1123  266 (23.7)  958  249 (26.0)   Physical activity category, N (%)  1160    970     Physical inactivity    202 (17.4)    198 (20.4)   Low physical activity    646 (55.7)    484 (50.0)   Moderate physical activity    292 (25.2)    271 (27.9)   High physical activity    20 (1.7)    17 (1.8)   Stage of Change, N (%)  1135    935     Maintenance    639 (56.3)    547 (58.5)   Action    147 (13.0)    113 (12.1)   Preparation    141 (12.4)    105 (11.2)   Contemplation    96 (8.5)    75 (8.0)   Pre-contemplation    112 (9.9)    95 (10.2)   SF-12           MCS score  1110  53.0 (46.7; 57.2)  908  52.4 (45.3; 57.2)   PCS score  1110  52.9 (47.0; 56.1)  908  52.9 (47.0; 56.1)  Clinical measures   BMI, kg/m2  1191  26.5 (23.5; 29.8)  1005  26.4 (23.7; 29.4)   WC, cm  1190  91.5 (82.0; 102.0)  1003  91.5 (83.0; 101.0)   SBP, mmHg  1192  124.0 (114.0; 135.0)  1005  124.0 (114.0; 135.0)   DBP, mmHg  1192  82.0 (76.0; 89.0)  1005  82.0 (76.0; 89.0)   TC, mM  1194  4.9 (4.3; 5.5)  1006  5.0 (4.4; 5.6)   CRF, l/min, mean (SD)  1078  2.5 (0.7)    –   ml/kg/min, mean (SD)  1078  31.5 (9.0)    –   poor, % (N)  1078  39.4 (436)      Notes: Values are median (interquartile range) unless otherwise specified. Data are derived from the Check your Health Preventive Programme. Sociodemographic information was acquired from Statistics Denmark. SF-12, Short-Form 12, Health Survey; PCS, physical component summary; MCS, mental component summary; BMI, body mass index; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; CRF, cardiorespiratory fitness; SD, standard deviation. Motivation The sub-analysis of stage progression at 2-week follow-up included 183 intervention participants and 129 control participants (figure 1). Overall, we found no significant difference in stage progression (table 2). Table 2 Stage progression at 2-week follow-up, self-rated health changes at one-year follow-up and adjusted estimates of between-group differences stratified by sex and age groups   Intervention Crude  Control Crude  Adjusted differencea    N  2-week follow-up  N  2-week follow-up  2-week follow-up  Stage progression % (95% CI)  Total  183  45.9 (38.7;53.1)  129  37.2 (28.8;45.6)  8.3 (−2.6;19.2)  Men  97  44.3 (34.4;54.2)  53  35.8 (22.9;48.8)  6.4 (−10.4;23.2)  Women  86  47.7 (37.1;58.2)  76  38.2 (27.;49.1)  11.5 (−2.9;26.0)  30–30 years  81  42.0 (31.2;52.7)  57  40.4 (27.6;53.1)  2.1 (−11.7;16.0)  40–49 years  102  49.0 (39.3;58.7)  72  34.7 (23.7;45.7)  13.5 (−4.4;31.5)    N  1-year follow-up  N  1-year follow-up  1-year follow-up  SF-12 mean (95% CI)  MCS score change   Total  465  0.1 (−0.7;0.9)  430  −0.4 (−1.3;0.4)  0.5 (−0.9;1.8)   Men  234  −0.3 (−1.3;0.8)  219  −0.4 (−1.5;0.7)  0.1 (−1.5;1.6)   Women  231  0.5 (−0.7;1.7)  211  −0.5 (−1.7;0.8)  0.8 (−0.8;2.5)   30–30 years  190  0.6 (−0.7;1.9)  173  −0.3 (−1.6;1.1)  0.7 (−1.3;2.8)   40–49 years  275  −0.2 (−1.2;0.8)  257  −0.5 (−1.6;0.5)  0.3 (−1.0;1.6)  PCS score change             Total  512  −0.05 (−0.7;0.6)  430  −0.1 (−0.8;0.6)  0.1 (−0.8;1.0)   Men  234  −0.1 (−0.9;0.8)  219  −0.2 (−1.1;0.7)  0.2 (−1.0;1.4)   Women  231  −0.02 (−1.0;0.9)  211  0.02 (−1.0;1.0)  0.03 (−1.3;1.4)   30–30 years  190  −0.9 (−1.9;0.1)  173  −0.2 (−1.3;0.8)  −0.6 (−2.2;1.0)   40–49 years  275  0.6 (−0.2;1.4)  257  −0.002 (−0.8;0.8)  0.6 (−0.6;1.8)    Intervention Crude  Control Crude  Adjusted differencea    N  2-week follow-up  N  2-week follow-up  2-week follow-up  Stage progression % (95% CI)  Total  183  45.9 (38.7;53.1)  129  37.2 (28.8;45.6)  8.3 (−2.6;19.2)  Men  97  44.3 (34.4;54.2)  53  35.8 (22.9;48.8)  6.4 (−10.4;23.2)  Women  86  47.7 (37.1;58.2)  76  38.2 (27.;49.1)  11.5 (−2.9;26.0)  30–30 years  81  42.0 (31.2;52.7)  57  40.4 (27.6;53.1)  2.1 (−11.7;16.0)  40–49 years  102  49.0 (39.3;58.7)  72  34.7 (23.7;45.7)  13.5 (−4.4;31.5)    N  1-year follow-up  N  1-year follow-up  1-year follow-up  SF-12 mean (95% CI)  MCS score change   Total  465  0.1 (−0.7;0.9)  430  −0.4 (−1.3;0.4)  0.5 (−0.9;1.8)   Men  234  −0.3 (−1.3;0.8)  219  −0.4 (−1.5;0.7)  0.1 (−1.5;1.6)   Women  231  0.5 (−0.7;1.7)  211  −0.5 (−1.7;0.8)  0.8 (−0.8;2.5)   30–30 years  190  0.6 (−0.7;1.9)  173  −0.3 (−1.6;1.1)  0.7 (−1.3;2.8)   40–49 years  275  −0.2 (−1.2;0.8)  257  −0.5 (−1.6;0.5)  0.3 (−1.0;1.6)  PCS score change             Total  512  −0.05 (−0.7;0.6)  430  −0.1 (−0.8;0.6)  0.1 (−0.8;1.0)   Men  234  −0.1 (−0.9;0.8)  219  −0.2 (−1.1;0.7)  0.2 (−1.0;1.4)   Women  231  −0.02 (−1.0;0.9)  211  0.02 (−1.0;1.0)  0.03 (−1.3;1.4)   30–30 years  190  −0.9 (−1.9;0.1)  173  −0.2 (−1.3;0.8)  −0.6 (−2.2;1.0)   40–49 years  275  0.6 (−0.2;1.4)  257  −0.002 (−0.8;0.8)  0.6 (−0.6;1.8)  Note: Data are derived from the Check your Health Preventive Programme. a Estimates of mean difference (linear regression) or absolute percentage difference (binomial linear regression with identity link) between the intervention and control group at 2-week follow-up and 1-year follow-up adjusted for baseline physical activity and intra-cluster correlation within general practices. CI, confidence interval. Cardiorespiratory fitness At 1-year follow-up, 505 intervention participants and 484 control participants attended the examination (figure 1). Of these, 44 participants in each group were excluded due to missing CRF or baseline self-reported physical activity level. When comparing the study groups at follow-up, the adjusted absolute CRF was lower in the intervention group (−0.1 l/min [95% confidence interval (CI): −0.2; −0.01]) (table 3). No differences were found in the adjusted relative CRF or poor CRF prevalence. The stratified analyses showed that the adjusted absolute CRF was lower in the intervention group among men (−0.2 l/min [95% CI: −0.3; −0.02]) and 40–49 year-old participants (−0.2 l/min [95% CI: −0.3; −0.03]), whereas the adjusted poor CRF prevalence was higher in the intervention group among women (7.2% [95% CI: 1.9; 12.5]). In the intervention group, both absolute and relative CRF increased from baseline to follow-up regardless of sex and age group (Supplementary material table S4). The poor CRF prevalence tended to decrease, but only significantly among 40–49 year-old participants. Table 3 CRF levels and physical inactivity prevalence at 1-year follow-up and adjusted estimates of between-group differences stratified by sex or age group   Intervention Crude  Control Crude  Adjusted differencea    N  1-year follow-up  N  1-year follow-up  1-year follow-up  Absolute CRF, l/min mean (95% CI)   Total  461  2.7 (2.6; 2.8)  440  2.8 (2.7; 2.9)  −0.1 (−0.2; −0.01)*   Men  233  3.0 (2.9;3.1)  227  3.2 (3.1;3.4)  −0.2 (−0.3; −0.02)*   Women  228  2.4 (2.3;2.5)  213  2.4 (2.3;2.5)  −0.005 (−0.1;0.1)   30–39 years  192  2.9 (2.8;3.0)  178  2.9 (2.8;3.0)  −0.04 (−0.2;0.1)   40–49 years  269  2.6 (2.5;2.7)  262  2.8 (2.7;2.9)  −0.2 (−0.3; −0.03)*  Relative CRF, ml/kg/min mean (95% CI)   Total  461  34.5 (33.5;35.4)  440  35.2 (34.2;36.1)  −0.7 (−2.0;0.6)   Men  233  34.7 (33.5;36.0)  227  36.8 (35.4;38.2)  −1.9 (−3.9;0.08)   Women  228  34.3 (33.0;35.6)  213  33.4 (32.1;34.6)  0.6 (−1.1;2.3)   30–39 years  192  36.8 (35.2;38.4)  178  36.0 (34.6;37.4)  0.4 (−1.5;2.4)   40–49 years  269  32.8 (31.6;33.9)  262  34.6 (33.3;35.9)  −1.5 (−3.0;0.04)  Poor CRF % (95% CI)   Total  461  31.0 (26.8;35.2)  440  25.9 (21.8;30.0)  5.0 (−0.002;10.1)   Men  233  39.5 (33.2;45.8)  227  34.4 (28.2;40.5)  2.6 (−5.1;10.4)   Women  228  22.4 (17.0;27.8)  213  16.9 (11.9;21.9)  7.2 (1.9;12.5)*   30–39 years  192  32.8 (26.2;39.5)  178  29.2 (22.5;35.9)  4.5 (−4.8.;13.8)   40–49 years  269  29.7 (24.3;35.2)  262  23.7 (18.5;28.8)  5.5 (−0.005;11.5)  Physical inactivity % (95% CI)   Total  588  14.8 (11.9;17.7)  541  17.2 (14.0;20.4)  −0.7 (−3.3;1.9)   Men  286  18.9 (14.3;23.4)  265  15.1 (10.8;19.4)  −1.9 (−5.5;1.8)   Women  302  10.9 (7.4;14.4)  276  19.2 (14.6;23.8)  −2.6 (−6.2;1.0)   30–39 years  248  18.1 (13.3;22.9)  224  18.8 (13.6;23.9)  −0.6 (−4.7;3.4)   40–49 years  340  12.4 (8.9;15.9)  317  16.1 (12.0;20.1)  −0.8 (−4.3;2.6)    Intervention Crude  Control Crude  Adjusted differencea    N  1-year follow-up  N  1-year follow-up  1-year follow-up  Absolute CRF, l/min mean (95% CI)   Total  461  2.7 (2.6; 2.8)  440  2.8 (2.7; 2.9)  −0.1 (−0.2; −0.01)*   Men  233  3.0 (2.9;3.1)  227  3.2 (3.1;3.4)  −0.2 (−0.3; −0.02)*   Women  228  2.4 (2.3;2.5)  213  2.4 (2.3;2.5)  −0.005 (−0.1;0.1)   30–39 years  192  2.9 (2.8;3.0)  178  2.9 (2.8;3.0)  −0.04 (−0.2;0.1)   40–49 years  269  2.6 (2.5;2.7)  262  2.8 (2.7;2.9)  −0.2 (−0.3; −0.03)*  Relative CRF, ml/kg/min mean (95% CI)   Total  461  34.5 (33.5;35.4)  440  35.2 (34.2;36.1)  −0.7 (−2.0;0.6)   Men  233  34.7 (33.5;36.0)  227  36.8 (35.4;38.2)  −1.9 (−3.9;0.08)   Women  228  34.3 (33.0;35.6)  213  33.4 (32.1;34.6)  0.6 (−1.1;2.3)   30–39 years  192  36.8 (35.2;38.4)  178  36.0 (34.6;37.4)  0.4 (−1.5;2.4)   40–49 years  269  32.8 (31.6;33.9)  262  34.6 (33.3;35.9)  −1.5 (−3.0;0.04)  Poor CRF % (95% CI)   Total  461  31.0 (26.8;35.2)  440  25.9 (21.8;30.0)  5.0 (−0.002;10.1)   Men  233  39.5 (33.2;45.8)  227  34.4 (28.2;40.5)  2.6 (−5.1;10.4)   Women  228  22.4 (17.0;27.8)  213  16.9 (11.9;21.9)  7.2 (1.9;12.5)*   30–39 years  192  32.8 (26.2;39.5)  178  29.2 (22.5;35.9)  4.5 (−4.8.;13.8)   40–49 years  269  29.7 (24.3;35.2)  262  23.7 (18.5;28.8)  5.5 (−0.005;11.5)  Physical inactivity % (95% CI)   Total  588  14.8 (11.9;17.7)  541  17.2 (14.0;20.4)  −0.7 (−3.3;1.9)   Men  286  18.9 (14.3;23.4)  265  15.1 (10.8;19.4)  −1.9 (−5.5;1.8)   Women  302  10.9 (7.4;14.4)  276  19.2 (14.6;23.8)  −2.6 (−6.2;1.0)   30–39 years  248  18.1 (13.3;22.9)  224  18.8 (13.6;23.9)  −0.6 (−4.7;3.4)   40–49 years  340  12.4 (8.9;15.9)  317  16.1 (12.0;20.1)  −0.8 (−4.3;2.6)  Note: Data are derived from the Check your Health Preventive Programme. a Estimates of mean difference (linear regression) or absolute percentage difference (binomial linear regression with identity link) between the intervention and control group at 1-year follow-up adjusted for baseline physical activity and intra-cluster correlation within general practices. *Indicates P < 0.05 (intervention vs. control group). CI, confidence interval; CRF, cardiorespiratory fitness. Self-reported physical inactivity and self-rated health No differences were found for the 1-year adjusted self-reported physical inactivity prevalence (table 3) or the adjusted changes in PCS or MCS scores (table 2). Dropout and sensitivity analyses A post-hoc analysis comparing the baseline characteristics of intervention participants (n = 461) and control participants (n = 440) included in the primary analysis showed a lower percentage of singles and individuals with low educational level in the control group (figure 1). A post-hoc sensitivity analysis adjusting for these potential confounding factors showed no between-group differences in absolute CRF (−0.09 l/min [95% CI: −0.19;0.01]), relative CRF (−0.2 ml/kg/min [95% CI: −1.4;1.0]), or poor CRF prevalence (3,5% [95% CI: −2.6;9.7]). We also compared the baseline characteristics of analysed participants with those who were not analysed (i.e. lost to follow-up or excluded due to missing data). In both groups, the not-analysed participants were more likely to be physically inactive, smokers and to have a low PCS score. In the control group, the not-analysed participants were also more likely to have a low education level and higher levels of BMI, waist circumference, and diastolic blood pressure. In the intervention group, the not-analysed participants were less likely to be single. Discussion Preventive health checks with CRF assessment did not provide higher CRF levels at 1-year follow-up than preventive health checks without CRF assessment. Furthermore, no differences were found in self-reported physical inactivity prevalence or in self-rated physical or mental health changes. Additionally, the proportion of pre-action stage participants with increased motivation to perform regular moderate physical activity at 2-week follow-up did not differ between the groups. Our study provides novel insights into the effect of including CRF assessment in preventive health checks on CRF levels. The effect of measuring CRF and physical activity levels has received very little attention.7 A recent systematic review and meta-analysis, including five randomised trials focusing on physical activity behaviour, showed a small positive effect of questionnaire-based physical activity measurements on self-reported physical activity.8 The follow-up periods of the included studies ranged from 1 week to 3 months, with the exception of one study with 6 months of follow-up.8 Evidence suggests that a majority of people who succeed in increasing their physical activity level may relapse to a less active or inactive status within 6 months.28 This may partly explain the lack of effect on CRF at 1-year follow-up in our study. Our findings are consistent with a study by Sluijs et al., which showed no effect of accelerometer-based measurements in an intervention on objectively measured physical activity at 6-month follow-up.29 We could only explore CRF changes in the intervention group as CRF was not assessed in the control group at baseline; a post-hoc analysis showed small increases in CRF levels from baseline to 1-year follow-up. Generally, physical activity interventions in primary care and community settings have shown limited long-term effects on CRF and physical activity levels.30–32 Surprisingly, both positive and negative effects on self-rated health have been observed during physical activity promoting interventions.33,34 We found no differences in self-rated physical or mental health status. Importantly, these findings suggest that CRF assessment did not lead to unintended physical or mental harm. Strengths and limitations The strengths of our study are the randomised design, the objective CRF assessment and the relatively long follow-up period compared with previous studies.8 The sustainability of an intervention effect is key, because temporary CRF improvements are unlikely to provide sustained health benefits. We used a real-life setting and recruited participants without any screening prior to inclusion, and more than 50% of the invited individuals attended the CHPP in both study arms. This attendance rate is high compared with other real-life health check programmes.35 Despite the emigration of 342 eligible participants before randomisation and a CHPP non-attendance rate around 45%, the baseline characteristics of the study groups did not differ significantly. The risk of selection bias is, therefore, considered to be low. Our study also has some important limitations. The dropout rate above 50% at follow-up increased the risk of attrition bias. Our dropout analysis indicated that control participants who were included in the primary analysis had a more favourable social profile than intervention participants. As lower social status is associated with lower CRF,36,37 attrition bias is likely to explain the higher absolute CRF level in the control group. This was also supported by the post-hoc sensitivity analysis. The study power was reduced by loss to follow-up and exclusion of certain participants, which increases the risk of a type II error. In particular, the sub-analysis of stage progression had limited power due to the unexpected small percentage of participants categorising themselves in the pre-action stages at baseline. In addition, the Stages of Change Questionnaire has not been validated to measure changes over time. Our estimates of the effect on motivation should, therefore, be interpreted with caution. The poor CRF prevalence tended to be higher in the intervention group. However, the difference between the study groups was attenuated when further adjusting for social factors. Loss of power is, therefore, unlikely to have caused a type II error with respect to CRF. Another possible contributing factor to the little difference between the study groups could be contamination. While only four women in the control group received a CRF assessment, intervention participants may have influenced control participants through social groups and workplaces. Furthermore, health professionals and GPs may have been more attentive to control participants than would be expected in usual care without CRF assessment. This could have attenuated a potential intervention effect. Another limitation was the relatively wide error margin of the Astrand-Ryhming test.38 This imprecision may have further reduced any differences between the study groups. Overall, as some limitations may have biased our results towards the null, we cannot preclude that a true intervention effect was overlooked. Generalisability and clinical impact CHPP attenders generally had a higher social status than non-attenders.39 As these results agree with previous findings, our study population likely represents the attenders that would be expected in a real-life population-based setting.40 Still, our findings may not generalise to more disadvantageous populations given that the participants who were not included in the analyses overall had a less favourable sociodemographic and clinical profile. CRF assessment is highly topical in view of the recent scientific statement from the American Heart Association.3 Routine CRF assessment as part of preventive health checks was feasible in our study. Yet, our results imply that adding objective measurement of CRF to a traditional risk factor assessment is unlikely to improve the effectiveness of population-based preventive health checks in terms of increasing CRF. Less resource-demanding non-exercise-based methods may provide valid alternatives.3 A pertinent question is how identification of poor CRF should be reflected in the subsequent care. Additionally, it remains to be investigated if CRF assessment should be confined to high-risk individuals who may derive particular benefit. Conclusion Preventive health checks with CRF assessment did not afford higher CRF levels, lower self-reported physical inactivity prevalence, or improved self-rated physical or mental health at 1-year follow-up compared with preventive health checks without CRF assessment. Key points CRF assessment in preventive health checks has received little attention in research. This study concludes that CRF assessment as part of a preventive health check did not result in higher CRF levels at 1-year follow-up. Our results imply that adding measurement of CRF to a traditional risk factor assessment is unlikely to improve the effectiveness of population-based preventive health checks in terms of increasing CRF. Several questions remain unanswered: Should CRF assessment be broadly utilised or confined to high-risk individuals? Which subsequent care should be offered for identified poor CRF? Public health policy and practice should consider these issues when planning future interventions aiming to decrease health risks resulting from poor CRF. Supplementary data Supplementary data are available at EURPUB online. Conflicts of interest: None declared. Acknowledgements The authors wish to thank the staff at Randers Health Centre and the Check Your Health group at Aarhus University. This work was supported by Aarhus University, the Central Denmark Region (file numbers 1-15-1-72-13-09 and 1-30-72-141-12), the municipality of Randers, the Danish foundation TrygFonden and the Aase and Ejnar Danielsen’s Foundation. References 1 Pescatello LS. ACSM's Guidelines for Exercise Testing and Prescription , 9th edn Philadelphia, PA: Lippincott Williams & Wilkins; 2014. 2 Kodama S, Saito K, Tanaka S, et al.   Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women: a meta-analysis. JAMA  2009; 301: 2024– 35. Google Scholar CrossRef Search ADS PubMed  3 Ross R, Blair SN, Arena R, et al.   Importance of assessing cardiorespiratory fitness in clinical practice: a case for fitness as a clinical vital sign: a scientific statement from the American Heart Association. Circulation  2016; 134: e653– 99. Google Scholar CrossRef Search ADS PubMed  4 Lee DC, Sui X, Artero EG, et al.   Long-term effects of changes in cardiorespiratory fitness and body mass index on all-cause and cardiovascular disease mortality in men: the Aerobics Center Longitudinal Study. Circulation  2011; 124: 2483– 90. Google Scholar CrossRef Search ADS PubMed  5 Myers J, McAuley P, Lavie CJ, et al.   Physical activity and cardiorespiratory fitness as major markers of cardiovascular risk: their independent and interwoven importance to health status. Prog Cardiovasc Dis  2015; 57: 306– 14. Google Scholar CrossRef Search ADS PubMed  6 Franklin BA. Exercise capacity: a crystal ball in forecasting future health outcomes?. Phys Sportsmed  2009; 37: 154– 6. Google Scholar CrossRef Search ADS PubMed  7 French DP, Sutton S. Reactivity of measurement in health psychology: how much of a problem is it? What can be done about it?. Br J Health Psychol  2010; 15: 453– 68. Google Scholar CrossRef Search ADS PubMed  8 Rodrigues AM, O'Brien N, French DP, et al.   The question-behavior effect: genuine effect or spurious phenomenon? A systematic review of randomized controlled trials with meta-analyses. Health Psychol  2015; 34: 61– 78. Google Scholar CrossRef Search ADS PubMed  9 Zunft H-F, Friebe D, Seppelt B, et al.   Perceived benefits and barriers to physical activity in a nationally representative sample in the European Union. Public Health Nutr  1999; 2: 153– 60. Google Scholar PubMed  10 Dalton AR, Soljak M. The nationwide systematic prevention of cardiovascular disease: the UK's health check programme. J Ambul Care Manage  2012; 35: 206– 15. Google Scholar CrossRef Search ADS PubMed  11 Sox HC. The health checkup. Was it ever effective? Could it be effective?. JAMA  2013; 309: 2496– 7. Google Scholar CrossRef Search ADS PubMed  12 Geue C, Lewsey JD, MacKay DF, et al.   Scottish Keep Well health check programme: an interrupted time series analysis. J Epidemiol Community Health  2016; 70: 924– 9. Google Scholar CrossRef Search ADS PubMed  13 Theobald H, Bygren LO, Carstensen J, et al.   Effects of an assessment of needs for medical and social services on long-term mortality: a randomized controlled study. Int J Epidemiol  1998; 27: 194– 8. Google Scholar CrossRef Search ADS PubMed  14 Lauritzen T, Jensen MS, Thomsen JL, et al.   Health tests and health consultations reduced cardiovascular risk without psychological strain, increased healthcare utilization or increased costs. An overview of the results from a 5-year randomized trial in primary care. The Ebeltoft Health Promotion Project (EHPP). Scand J Public Health  2008; 36: 650– 61. Google Scholar CrossRef Search ADS PubMed  15 Hoj K, Skriver MV, Hansen AL, et al.   Effect of including fitness testing in preventive health checks on cardiorespiratory fitness and motivation: study protocol of a randomized controlled trial. BMC Public Health  2014; 14: 1057. Google Scholar CrossRef Search ADS PubMed  16 Maindal HT, Stovring H, Sandbaek A. Effectiveness of the population-based Check your health preventive programme conducted in primary care with 4 years follow-up [the CORE trial]: study protocol for a randomised controlled trial. Trials  2014; 15: 341. Google Scholar CrossRef Search ADS PubMed  17 Christensen AI, Davidsen M, Ekholm O, et al.   Danskernes sundhed. Tal fra den Nationale Sundhedsprofil 2010 og 2013. [The Health of the Danes. Figures from the National Health Profile 2010 and 2013]. Available at: http://www.danskernessundhed.dk/ (5 July 2016, date last accessed). 18 Danmarks Statistik. [Statistics Denmark]. Available at: http://www.statistikbanken.dk/statbank5a/default.asp?w=1920 (9 September 2016, date last accessed). 19 Schmidt M, Pedersen L, Sorensen HT. The Danish Civil Registration System as a tool in epidemiology. Eur J Epidemiol  2014; 29: 541– 9. Google Scholar CrossRef Search ADS PubMed  20 Nigg CR. There is more to stages of exercise than just exercise. Exerc Sport Sci Rev  2005; 33: 32– 5. Google Scholar PubMed  21 Nigg CR, Geller KS, Motl RW, et al.   A research agenda to examine the efficacy and relevance of the transtheoretical model for physical activity behavior. Psychol Sport Exerc  2011; 12: 7– 12. Google Scholar CrossRef Search ADS PubMed  22 Astrand I. Aerobic work capacity in men and women with special reference to age. Acta Physiol Scand Suppl  1960; 49: 1– 92. Google Scholar PubMed  23 Saltin B, Grimby G. Physiological analysis of middle-aged and old former athletes. comparison with still active athletes of the same ages. Circulation  1968; 38: 1104– 15. Google Scholar CrossRef Search ADS PubMed  24 Ware JE, Kosinski M, Turner-Bowker DM, Gandeck B, eds. Useŕs Manual for the SF-12 v2TM Health Survey (with a Supplement Documenting the SF-12 Health Survey) . Lincoln, RI: QualityMetric Incorporated, 2007. 25 Skov-Ettrup LS, Petersen CB, Curtis T, et al.   Why do people exercise? A cross-sectional study of motives to exercise among Danish adults. Public Health  2014; 128: 482– 4. Google Scholar CrossRef Search ADS PubMed  26 Dallongeville J, Banegas JR, Tubach F, et al.   Survey of physicians' practices in the control of cardiovascular risk factors: The EURIKA study. Eur J Prev Cardiol  2012; 19: 541– 50. Google Scholar CrossRef Search ADS PubMed  27 Groenwold RH, Moons KG, Vandenbroucke JP. Randomized trials with missing outcome data: how to analyze and what to report. CMAJ  2014; 186: 1153– 7. Google Scholar CrossRef Search ADS PubMed  28 Amireault S, Godin G, Vezina-Im L. Determinants of physical activity maintenance: a systematic review and meta-analyses. Health Psychol Rev  2013; 7: 55– 91. Google Scholar CrossRef Search ADS   29 van Sluijs EM, van Poppel MN, Twisk JW, van Mechelen W. Physical activity measurements affected participants' behavior in a randomized controlled trial. J Clin Epidemiol  2006; 59: 404– 11. Google Scholar CrossRef Search ADS PubMed  30 Orrow G, Kinmonth AL, Sanderson S, Sutton S. Effectiveness of physical activity promotion based in primary care: systematic review and meta-analysis of randomised controlled trials. BMJ  2012; 344: e1389. Google Scholar CrossRef Search ADS PubMed  31 Baker PR, Francis DP, Soares J, Weightman AL, Foster C. Community wide interventions for increasing physical activity. Cochrane Database Syst Rev  2015; 1: CD008366. Google Scholar PubMed  32 Baumann S, Toft U, Aadahl M, et al.   The long-term effect of screening and lifestyle counseling on changes in physical activity and diet: the Inter99 study - a randomized controlled trial. Int J Behav Nutr Phys Act  2015; 12: 33. Google Scholar CrossRef Search ADS PubMed  33 Kinmonth AL, Wareham NJ, Hardeman W, et al.   Efficacy of a theory-based behavioural intervention to increase physical activity in an at-risk group in primary care (ProActive UK): a randomised trial. Lancet  2008; 371: 41– 8. Google Scholar CrossRef Search ADS PubMed  34 Petersen CB, Severin M, Hansen AW, et al.   A population-based randomized controlled trial of the effect of combining a pedometer with an intervention toolkit on physical activity among individuals with low levels of physical activity or fitness. Prev Med  2012; 54: 125– 30. Google Scholar CrossRef Search ADS PubMed  35 Robson J, Dostal I, Sheikh A, et al.   The NHS health check in England: an evaluation of the first 4 years. BMJ Open  2016; 6: e008840. Google Scholar CrossRef Search ADS PubMed  36 Hoj K, Vinther Skriver M, Terkildsen Maindal H, et al.   High prevalence of poor fitness among Danish adults, especially among those with high cardiovascular mortality risk. Eur J Public Health  2016; 27: 569– 74. 37 Lindgren M, Borjesson M, Ekblom O, 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– 9. Google Scholar CrossRef Search ADS PubMed  38 Åstrand PO, Rodahl K. Textbook of Work Physiology: Physiological Bases of Exercise , 4th edn Champaign, IL: Human Kinetics; 2003. 39 Bjerregaard AL, Maindal HT, Bruun NH, Sandbaek A. Patterns of attendance to health checks in a municipality setting: the Danish ′Check your health preventive program′. Prev Med Rep  2016; 5: 175– 82. Google Scholar CrossRef Search ADS PubMed  40 Tolonen H, Ahonen S, Jentoft S, et al.   Differences in participation rates and lessons learned about recruitment of participants: the European Health Examination Survey Pilot Project. Scand J Public Health  2015; 43: 212– 9. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.

Journal

The European Journal of Public HealthOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from
Google Scholar,
PubMed
Create lists to
organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off