Skip to main content
Download PDF
Download ePub

Spillover effect of a dietary intervention on physical activity in a randomized controlled trial with colorectal cancer patients

International Journal of Behavioral Nutrition and Physical Activity volume 22, Article number: 54 (2025) Cite this article

Abstract

Background

Randomized controlled studies (RCTs) targeting dietary changes may also lead to other, untargeted changes in lifestyle habits, as spillover effects. In particular, the isolated impact of the dietary intervention may be difficult to separate due to spillover effects from changes in physical activity and physical function. Therefore, the aim of this study was to investigate the spillover effect of a one-year dietary intervention in post-surgery colorectal cancer patients by comparing the changes in physical activity and physical function between the diet intervention group and the control group in a randomized controlled trial, called the CRC-NORDIET study.

Methods

Men and women, aged 50–80 years were randomized into either the intervention group (n = 240) or the control group (n = 229). Both groups received similar incentives on physical activity. Activity sensors were used to collect data on physical activity at baseline, 6 months, and 12 months. Physical function was estimated by results from handgrip strength, 30 s sit-to-stand test and 6-min walking test. Anthropometric measurements and body composition were also measured.

Results

We found a significantly higher increase in moderate-to-vigorous intensity physical activity (MVPA) of 0.18 h per day from baseline to 6 months in the diet intervention group compared to the control group, respectively. However, the spillover effect of the dietary intervention on physical activity diminished to 0.10 h per day at 12 months follow-up which was not statistically significantly different (p = 0.24) from the control group. All measures of physical function increased in both groups from baseline to 6 months with no further increase at the 12-month follow-up.

Conclusions

The dietary intervention did not induce a significant spillover effect on physical activity after 12 months of baseline, which was the main timepoint of the intervention. Providing identical physical activity guidance to both study groups during the 12-month intensive dietary intervention period, ensured comparable levels of physical activity across both study groups. This approach facilitated the isolation and analysis of the dietary intervention's effects on primary endpoints, as well as effects of behaviour interventions in secondary preventions, such as the CRC-NORDIET study.

Trial registration

The study is registered on the National Institutes of Health Clinical Trials website (www.ClinicalTrials.gov; Identifier: NCT01570010).

Introduction

Lifestyle risk behaviors, such as poor diet, physical inactivity, alcohol consumption and smoking, are associated with risk of developing non-communicable diseases (NCDs), like cancer, cardiovascular disease, and diabetes [1,2,3,4]. These associations are primarily based on results from observational studies, which are subjected to confounders and bias and therefore have limited strength of causality [5, 6]. Randomized controlled trials (RCT) are considered the golden standard for providing the highest strength of causal inference due to the equal distribution of confounding factors through randomization and the inclusion of a control group. However, for RCTs with lifestyle interventions, it is challenging to implement traditional controls, as used in pharmacological RCTs. This is mainly because finding an appropriate control for health behaviors is not straightforward [5, 6]. Therefore, instead of a traditional control group an alternative is to establish a control group that do not receive any intervention but instead continue with their habitual lifestyle. While such alternative control group may lack the rigor of a control, it is still a valuable approach to assess the impact of lifestyle interventions on behavior change and subsequent health outcomes [6].

When implementing a lifestyle intervention trial aiming at changing one type of risk behavior e.g. diet, it is possible that other risk behaviors can change simultaneously and having synergistic or spillover impacts on the risk of NDCs accordingly [7,8,9,10]. The spillover effect, in the context of Multiple-Health-Behavior-Change (MHBC) interventions, refers to the phenomenon where changes made in one health behavior influence changes in other health behaviors, often leading to broader improvements in overall health outcomes [7, 8]. Essentially, it suggests that when individuals successfully modify one behavior, such as quitting smoking or increasing physical activity, they may become more motivated or confident to address other health-related behaviors, such as improving their diet or managing stress. MHBC interventions have been used to reduce recourses for intervention implementation and health care costs and to increase motivation and self-efficacy to improve certain health risk behaviors. [7, 8, 11,12,13]. However, the effects of MHBC interventions are difficult to separate from spillover effects to untargeted behavior. For diet interventions in particular, changes in physical activity have been suggested to promote such spillover effects [7, 8, 13, 14].

While it is well-established that a healthy diet may reduce risk of several cancers, few studies have assessed the effect of a healthy diet on long term health outcomes after primary treatment. Thus, there is a need for high quality and well-designed intervention studies evaluating the effectiveness of diet behavior in cancer survivors [6, 10, 15, 16].

In the present study and based on the CRC-NORDIET study we investigated the spillover effect of a dietary intervention on changes in moderate-to-vigorous intensity physical activity (MVPA) (hours per day and steps per day) in colorectal cancer (CRC) patients after 6 and 12 months follow up. Moreover, we also investigated the spillover effect of the dietary intervention on physical function in CRC patients after 6 and 12 months follow up.

Methods

Subjects and study design

The Norwegian Dietary Guidelines and Colorectal Cancer Survival (CRC-NORDIET) study was an RCT aimed to investigate the effect of a healthy diet according to the national recommendations on disease free- and overall survival among colorectal cancer patients 5, 10 and 15 years after baseline [17]. The design was a multi-centre two-arms RCT, i.e. an intensive dietary intervention group and a control group receiving usual care. The intensive dietary intervention lasted for 12 months and with long-term following up of 14 additional years from baseline. Due to the fact of the high interrelationship between diet and physical activity, both the intervention group and the control group received similar general advice to promote physical activity.

The CRC-NORDIET study enrolled 503 participants aged 50 to 80 years, newly diagnosed with non-metastatic colorectal cancer staged (tumor-node-metastasis (TNM) I-III (ICD-10 codes C18-C20) between 2012–2020 at Oslo University Hospital and Akershus University Hospital. Eligibility required attendance at the first study visit within nine months of surgery, Norwegian literacy, and provision of written informed consent to participate in the study. Exclusion criteria included dementia, mental impairments affecting comprehension of the study’s intervention, or participating in conflicting studies [17].

Physical activity in the CRC-NORDIET intervention

Enrolled participants were randomly assigned to either an intervention or a control group. The intervention group received a 12-month intensive, personalized dietary intervention, following guidelines as outlined by Henriksen et al. [17] (Supplementary Material 1), in addition to recommendations for general physical activity aligned with national standards [18]. The control group received the same advice on physical activity, and general information about the Norwegian food-based dietary guidelines (FBDG) [17, 18].

All participants were recommended to aim for at least 150 min (i.e. 2.5 h) MVPA per week. They received a booklet containing practical guidance for integrating physical activity into their daily lives. In addition, they were encouraged to take advantage of local resources, such as health training centers and swimming pools, to enhance their physical activity regimen [17].

In collaboration with"Active against cancer,"a nonprofit organization founded in 2007, the CRC-NORDIET study provided participants 6 months of complimentary access to the exercise facility “Pusterommet”. Located in several Norwegian hospitals, “Pusterommet” offers group training, yoga-classes, strength- and cardiovascular training. Additionally, physical therapists provide personalized exercise guidance, customizing activities to support each patient's rehabilitation process during and after cancer treatment. Participants were invited to attend an inspiration day within the first 12 months of the intervention. This event included a lecture on daily physical activity and interactions with physical therapists from “Pusterommet” [17].

Assessment of physical activity

Data on physical activity were collected at baseline and after 6 and 12 months of follow-up. The Sensewear Mini Armband (BodyMedia, Pittsburgh, Pennsylvania, USA) recorded daily physical activity levels, sedentary time, and energy expenditure [19,20,21]. The activity monitor uses a combination of sensors to measure various physiological parameters, including heat flux, galvanic skin response, tri-axial acceleration, and skin temperature. Algorithms combine various activity intensities to estimate energy expenditure in metabolic equivalents (METs), defined to 3.5 ml O2 per kg body weight x minutes [22]. Physical activity was categorized as light (1.5–3 METs), moderate (3–6 METs), vigorous (> 6 METs), and moderate-to-vigorous intensity (> 3 METs). Sedentary time included all activities < 1.5 METs, including nighttime (midnight to 6:00 a.m.). Intensities are recorded in 1-min intervals, with the total duration measured in hours per day [23].

Participants were instructed to wear the activity monitor on their non-dominant arm continuously for seven days after each study visit. Each monitor was individually programmed with demographic and biometric information, including age, weight, height, sex and smoking status. They were advised to adhere to their normal daily routines, except removing the monitor during water-based activities or if any contraindications occurred. Adequate data capture was defined as a minimum of 80% wear time over a 24-h period and minimum 4 consecutive days.

The recorded data was processed using the Sensewear Professional Software Version 7.0 BodyMedia Inc. (Pittsburg, Pennsylvania, USA). This analysis was conducted on a secure laptop, which was not connected to the internet to ensure data protection. All data in the study was safely stored in a secure server called TSD (Service for Sensitive Data), designed for storing and post-processing sensitive data in compliance with the Norwegian “Personal Data Act” and “Health Research Act”. TSD is developed and maintained by IT-Department (USIT) at the University of Oslo, Norway. Participants with pacemakers were excluded from this measurement [17].

Physical strength and function

Handgrip strength, a common indicator of overall body strength, was measured using MAP 80 K1 Handgrip dynamometer (KERN & SOHN GmbH, Balingen, Germany) according to the manufacturer’s protocol [17]. Women performed the tests using 40 kg springs, whereas men used 80 kg springs. Seated with elbows at a 90-degree angle, participants performed three maximum-effort grips with each hand, and the highest value was documented.

The 30-s sit-to-stand test (30STS) assessed lower body muscle endurance. A 44 cm tall chair, with no armrest, was used for all measurements. The participant started from a seated position, arms crossed at the chest and the legs parallel to the ground. The total number of full stands within 30 s was registered [17].

The 6-min’ walk test (6MWT) measured aerobic capacity and stamina. The test's purpose is to assess the distance an individual can walk within 6 min. The test was conducted in a straight, 30-m-long corridor with level surface. Pulse rate was monitored before and after the test, and subjective effort was quantified using the Borg Rating of Perceived Exertion scale [17].

Anthropometric measurements

Height and weight were measured at each study visit using a digital measuring station known as Seca 285 (Seca Birmingham, United Kingdom) [24]. The measurements were carried out dressed in light clothing with no shoes, as elaborated in detail in Henriksen 2017 [17].

Waist circumference was measured at the midpoint between the top of the iliac crest and the lower margin of the last palpable rib [17]. Hip circumference was measured around the widest portion of the hips, with the subjects'feet positioned 12–15 cm apart [17].

Body composition

Body composition assessment was conducted using the Lunar Dual-energy X-ray Absorptiometry (iDXA) machine (GE Healthcare Lunar, Buckinghamshire, United Kingdom) and the enCORE software version 18. High precision and valid measurements of body composition has been shown for this machine in a sub-population of the CRC-NORDIET study as well as in healthy subjects [25, 26]. Participants underwent the scanning process in a fasted state and dressed in light clothing. Certified and trained operators adhering to the protocols established by the International Society for Clinical Densitometry (ISCD) [27] performed all DXA scans. The scans provided measures for whole body fat mass, fat free mass and bone mineral density, which were included in the analysis.

Descriptive and clinical data

Descriptive and clinical data, including age, sex, weight, height, TNM-status, tumor location, time and type of surgery and additional treatment, were obtained from the CRC-NORDIET database. Demographic and comorbidity questionnaires were used to describe marital status, education level, work ability and concurrent diseases [17].

Statistical analysis

Linear mixed effects models with random intercepts were used to analyze the data. The linear mixed effects models were used to account for the nested data structure (i.e., time nested in individuals) and examine differences between the intervention and control group at the 6- and 12-months follow-up. Separate models were estimated for each of the primary (MVPA, steps) and secondary outcome variables (hand-grip strength, 6MWT, 30STS). An interaction term between treatment group and measurement point was included to examine differences between the intervention and control group (i.e., the treatment effect) from baseline to 6 months and from baseline to 12 months. The baseline values were assumed to be equal between the groups and are thus reflected in the intercept of the model [28]. Both unadjusted and adjusted models were estimated. In the adjusted models we adjusted for sex, education, TNM-status, age, time since surgery, and comorbidity as control variables. We used Stata version 18.0, maximum likelihood estimation, and an unstructured covariance structure when estimating the linear mixed effects models (Stata code for the models are presented in the Supplementary Material 2, Table S1. Under the assumption of missing at random, maximum likelihood estimation accommodates incomplete outcomes by including all available data in the estimation [29]. Hence, no imputation of missing data was conducted prior to the analyses. No sample size estimation was conducted specifically for the current exploratory study because this is secondary data analyses of the CRC-NORDIET study [17]. However, based on the primary outcome variables (MVPA and step count) in the current study, we included participants with physical activity data (MVPA and step counts) on at least one measurement point (i.e. at either baseline, 6- and/or 12-months following up. The graphical display of results presented in Fig. 2 are based on estimated marginal means from the adjusted linear mixed effects models.

We also conducted two sensitivity analyses to examine the impact of missing data on the results. The first sensitivity analysis only included cases with complete physical activity data across all three time points (n = 317). In the second sensitivity analysis we used the factored regressions approach with a partially factored specification that accounts for missing data in both outcome variables and predictor variables [29, 30]. Hence, these analyses include all 469 participants regardless of amount of missing data on predictor or outcome variables. The sensitivity analyses using the factored regressions approach were conducted in the software Blimp version 3.2.1 [31], which relies on Bayesian estimation and does not rely on a joint distribution of the analysis variables. Instead, in the factored regression approach the multivariate distribution is factored into the product of multiple univariate distributions. The models were estimated using 60,000 Markov Chain Monte Carlo iterations and 30,000 burn-in iterations, and a low potential scale reduction factor [32] was used as an indication of convergence (i.e., < 1.05). The results from the sensitivity analyses are briefly described in the results and presented in full in the supplemental material (see Supplementary Material 2 Tables S5-S14).

Results

Recruitment and baseline characteristics

The current study included 469 patients out of the original 503 enrolled in the CRC-NORDIET study and were used for the primary endpoint (Fig. 1). For two of the secondary outcome variables, i.e. the 6MWT and 30STS, the sample size was slightly lower due to additional missing data on these variables (394 and 455 participants, respectively).

Fig. 1
figure 1

Flow of participants

Full size image

Baseline characteristics are detailed in Table 1. In brief, of the included participants 250 (53.3%) and 219 (46.7%) were males and females, respectively. A mean age of 65 years and an average BMI of 27 kg/m2 were measured in both study groups. More than 60% of the subjects were classified as overweight or obese. The most frequent diagnosis was colon cancer (59%), with TNM stage II being predominant (~ 37%). More than 60% had at least one comorbidity, with musculoskeletal diseases being more prevalent in the intervention group (29%) than the control group (22%). Median MVPA was recorded as 509.8 and 468.0 min/week (or 8.5 and 7.8 h per week) in the diet intervention and control group, respectively. Average hand-grip strength, and 30STS were 33 kg and 15.6 stands in each study groups, respectively. The 6MWT did not differ between the study groups, with 584.4 and 576.1 m in the diet intervention av control group, respectively (Table 1).

Table 1 Baseline characteristics of the intervention and control group (n = 469)*
Full size table

Primary outcomes – physical activity

At the 6 months follow-up, the diet intervention group had increased their MVPA compared to the control group as indicated by the statistically significant intervention group*measurement point (6 months) interaction effect (γ = 0.18, SE = 0.08, p = 0.03, 95% CI [0.02, 0.35]). However, at the 12-months follow-up this difference between the diet intervention and control groups in MVPA was reduced and not statistically significant (γ = 0.10, SE = 0.09, p = 0.24, 95% CI [−0.07, 0.27]) as indicated by the non-significant intervention group*measurement point (12 months) interaction effect. A graphical depiction of the differences in MVPA across the three time points is shown in Fig. 2 (top left) and the results are presented in Table 2.

Fig. 2
figure 2

Differences in MVPA (h/d) (top left), steps (stands per day) (top right), hand-grip strength (kg) (mid left), 30STS (stands per 30 s) (mid right) and 6MWT (meters) (bottom left) across the three time points (baseline, 6 months, 12 months) between the diet intervention group (black dots) and the control group (grey dots)

Full size image
Table 2 Unadjusted and Adjusted Linear Mixed Effects Model with MVPA as Dependent Variable
Full size table

There were no statistically significant differences in steps between the diet intervention group and control group at the 6- (γ = −165.38, SE = 233.62, p = 0.48, 95% CI [−623.27, 292.50]) or 12-months (γ = 165.69, SE = 247.02, p = 0.50, 95% CI [−318.46, 649.84]) follow-up as indicated by the non-significant intervention group*measurement point interaction effects. A graphical depiction of the differences in steps across the three measurement points is shown in Fig. 2 (top right) and the results are presented in Table 3. Both groups increased the number of steps slightly from baseline to 6 months (average increase across both groups ≈ 354 steps), but at the 12 months follow-up the average number of steps across both groups were close to baseline values.

Table 3 Unadjusted and Adjusted Linear Mixed Effects Model with Steps as Dependent Variable
Full size table

Secondary outcomes – physical function

There were no statistically significant differences in hand-grip strength between the diet intervention group and control group at the 6- (γ = 0.06, SE = 0.30, p = 0.85, 95% CI [−0.53, 0.64]) or 12-month (γ = 0.27, SE = 0.34, p = 0.43, 95% CI [−0.40, 0.93]) follow-up as indicated by the non-significant intervention group*measurement point interaction effects. There was, however, a general increase in both groups from baseline to 12 months (average increase across both groups ≈ 1.3 kg) in hand-grip strength (see Table S2 in Supplementary Material 2 and Fig. 2 (mid left)).

There were no statistically significant differences in the 30STS between the diet intervention group and control group at the 6- (γ = 0.03, SE = 0.31, p = 0.93, 95% CI [−0.58, 0.63]) or 12-month (γ = 0.09, SE = 0.37, p = 0.80, 95% CI [−0.62, 0.81]) follow-up as indicated by the non-significant intervention group*measurement point interaction effects. There was, however, a general increase in both groups from baseline to 12 months (average increase across both groups ≈ 2.3 stands) in the 30STS (see Table S3 in Supplementary Material 2 and Fig. 2 (mid right)).

There were no statistically significant differences in the 6MWT between the diet intervention group and control group at the 6- (γ = −5.24, SE = 8.66, p = 0.55, 95% CI [−22.21, 11.73]) or 12-month (γ = −1.52, SE = 8.38, p = 0.86, 95% CI [−17.94, 14.90]) follow-up as indicated by the non-significant intervention group*measurement point interaction effects. There was, however, a general increase in both groups from baseline to 12 months (average increase across both groups ≈ 30 m) in 6MWT (see Table S4 in Supplementary Material 2 and Fig. 2 (bottom left).

Sensitivity analyses

Results from the two sensitivity analyses are presented in Additional files Tables S5-S14. The two sensitivity analyses generally supported the findings from the main analyses and the effects on MVPA, hand-grip strength, 30STS, and 6MWT were almost identical (with minor differences in magnitude). One difference between the main findings and sensitivity analyses were observed in the complete case analyses for the main effect on steps at 6 months, which was weaker and not statistically significant in the complete case analyses (γ = 253.94, SE = 194.69, p = 0.19). However, although the main effect of time on steps was not statistically significant, it was in the same direction as the main analysis but slightly weaker in magnitude.

Discussion

In the current study, we investigated the possible spillover effect of a dietary intervention on physical activity in colorectal cancer patients 6- and 12 months after baseline. Moreover, we also investigated effect of the dietary intervention on physical function at the same time points. We found a higher significant increase in MVPA of 0.18 h per day from baseline to 6 months in the diet intervention group when compared to the control group. However, the effect of the dietary intervention on MVPA diminished after 12 months of baseline and was no longer significant different between the two study groups. All tests estimating physical function increased in both groups from baseline to 6 months. No further increase was observed at the 12-month follow-up. These results persisted after adjusting for sex, age, education, TNM-stages, time since surgery and comorbidities.

Controls for lifestyle interventions in RCTs have been found challenging, because a controlled lifestyle intervention may always have some effect on the participants behavior, like attention, changes in awareness of own health due to biological measurements and other monitoring of lifestyle habits [6]. It is therefore recommended to offer the controls either an alternative lifestyle intervention or usual care as in real-world practice, such as some lifestyle advice [6, 33]. Based on these findings, the CRC-NORDIET study provided comparable advice, which were presented in both written and oral formats, without implementing a specific exercise intervention. The guidance advised participants in both study groups to engage in physical activity, including strength training, at least twice a week, in accordance with the Norwegian national guidelines on physical activity. This study design enhanced our capacity to evaluate the effects of the dietary intervention on health outcomes while allowing for the monitoring of physical activity levels in both groups to control for potential confounding variables.

The CRC-NORDIET study was designed to explore the effect of a 12-months intensive dietary intervention on survival outcomes [17]. The result from the current study indicates that the design of the CRC-NORDIET study worked as intended, showed by no significant difference in physical activity between the study groups after 12 months of intervention. This is of great value for future studies estimating effects of the 12-months dietary intervention on both primary and secondary endpoints of dietary interventions like the CRC-NORDIET study. Moreover, these results may also be useful when implementing secondary prevention interventions among colorectal cancer survivors in clinical health care practices.

After completing cancer treatment, including surgery, chemotherapy, and radiation therapy, individuals often strive to regain their health and well-being as quickly as possible. Fast recovery after cancer treatment is of utmost importance. The significant higher increase in MVPA in the diet intervention group after 6 months, together with the observation that the control group has higher body weight, BMI, fat mass and visceral adipose tissue than compared to the intervention group [34] may suggest that they recover faster than the control group halfway through the diet intervention.

Spillover effects of lifestyle interventions have been shown reduced after long term follow up in other studies as well [33, 35, 36] and spillover effects to untargeted behaviors in Multiple behavior change interventions may occur and therefore important to investigate [7, 8]. In the Chapman`s review of lifestyle interventions, dietary changes can be achieved when the intervention is most intense, but will diminish afterwards in the long run [35].

Spillover effects on untargeted behaviors (e.g. diet) was also found in the CanChange study [36] investigating effects on a telephone-delivered health coaching 12-months intervention on physical activity and monitoring of dietary intakes. Gerstel et al. investigated the impact of a lifestyle intervention on body weight, metabolic syndrome parameters, nutrition and physical activity in home-care providers [33]. They found an intermediate effect of weight loss also among the controls after 6 months which was not maintained at 12 months of intervention. The I CAN study investigated the feasibility of a 12-months lifestyle intervention among cancer patients undergoing curative and palliative chemotherapy [37]. They found a significant increase in dietary habits and physical activity after 4 months of intervention, which reduced to baseline levels after 12 months follow-up [37]. This may be due to the “ceiling” effect also observed in other lifestyle interventions [38, 39] referring to a phenomenon whereby the potential for sustainable improvement in measured outcomes has reached its threshold for further improvements among the participants.

The CRC-NORDIET study included general advice for physical activity for both the diet intervention group and the control group, but no specific behavior intervention aiming at improvements in physical function. However, since physical function is an important health outcome among cancer patients, the CRC-NORDIET study monitored physical function by different tests. Therefore, the current study investigated possible spillover effects of the diet intervention on physical function, estimated by hand grip strength, 30STS and 6MWT, and found no differences between the study groups at neither time points. However, improvements in both groups were documented from baseline to 12-months follow-up.

Strengths and limitations

The RCT design of the CRC-NORDIET study is unique within lifestyle interventions, as observational and mechanistic study design are dominated among cancer patients. The strength is in the randomization to an individualized dietary intervention or an active control group with similar advice on being physical active in both study arms. Moreover, since it takes time to change behavior and to develop health effects on disease and survival, the intensive dietary intervention had a long-term focus with a duration of 12 months, followed by a maintenance period of additional 14 years. This was based on experiences from other lifestyle interventions among cancer patients [35, 40,41,42]. By this design, effect of the dietary intervention on the survival outcomes are possible to be estimated. Objective measurement tools, like accelerometers, are increasingly used in lifestyle interventions and gives more accurate measures of activity than questionnaires [43,44,45]. The present study used accelerometers to measure physical activity, which gives more precise and accurate measurements on possible differences in physical activity between the study groups than by using self-reported data from questionnaires.

It is important to note that the activity sensors (SWA) used to measure physical activity across various intensity levels may lead to an overestimation of physical activity [21]. Participants wore the sensors continuously, both day and night, over a one-week period, facilitating the registration of all activities in one-minute intervals during this time frame. This methodology accounts for the higher duration of moderate- to- vigorous physical activity recorded compared to self-reported physical activity data obtained through questionnaires. National guidelines of physical activity are usually based on epidemiological studies using questionnaires as assessment tool for physical activity and to investigate the effect on health outcomes in the population.

A limitation in the current study is the dropouts on the visits after 6 and 12 months follow up, which were higher in the control group compared to the diet intervention group. By using the linear mixed model analysis, we included participants with physical activity data on at least one measurement point and the assumption of missing at random and accommodates incomplete outcomes by including all available data in the estimation. The participants in the study were higher educated compared to the general Norwegian population [46]. However, the distribution of colon and rectal cancer cases, as well as the TNM stages (I-III), within the CRC-NORDIET study population appears to reflect similar proportions observed in the general population of colorectal cancer patients in Norway in 2012 [47]. This is also the year when recruitment for the study commenced. Thus, the CRC-NORDIET study population may be representative (i.e. external valid) for all CRC patients with primary invasive CRC, TNM stages I-III and ages between 50–80 years old.

Based on the finding in this study, identifying the synergistic interplay of health risk behaviors, such as the spillover effect of a target behavior to an untargeted behavior, has great impact on the design of the study and the effect of intervention on health outcomes.

Conclusions

In the current study, no spillover effect of the intensive dietary intervention on physical activity was found at 12 months follow-up, despite the significant effect at the 6 months follow-up (i.e. half-way of the dietary intervention). The number of steps increased in both study groups from baseline to 6 months, however, neither of these effects were maintained at the 12-month follow-up. Moreover, no statistically significant differences between the diet intervention and control group were observed in any of the physical function outcomes. Both groups showed an increase in all three physical function tests from baseline to 6 months and from baseline to 12 months, which indicates a general improvement in physical function across the intervention.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

MVPA:

Moderate-to-vigorous physical activity

30STS:

30 Seconds sit-to-stand test

6MWT:

6 Minutes walking test

PG-SGA:

Patient-Generated Subjective Global Assessment

TNM:

Tumor Node Metastasis

BMI:

Body Index Mass

RCT:

Randomized controlled trials

NCD:

Non-communicable diseases

CRC-NORDIET study:

The Norwegian Dietary Guidelines and Colorectal Cancer Survival study

FBDG:

Food-based dietary guidelines

METs:

Energy expenditure in metabolic equivalents

DXA:

Lunar Dual-energy X-ray Absorptiometry

ISCD:

International Society for Clinical Densitometry

References

  1. Continuous Update Project Report. Food, Nutrition, Physical Activity, and the Prevention of Colorectal Cancer. Washington D.C.: World Cancer Research Fund / American Institute for Cancer Research; 2011.

    Google Scholar 

  2. WHO-Physical activiy: World Health Organization (WHO); 2020 [cited 2022 09/29]. Available from: https://www.who.int/news-room/fact-sheets/detail/physical-activity.

  3. Murray CJL, Aravkin AY, Zheng P, Abbafati C, Abbas KM, Abbasi-Kangevari M, et al. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet. 2020;396(10258):1223–49. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/S0140-6736(20)30752-2.

    Article  Google Scholar 

  4. Noncommunicable diseases: WHO; 2024 [cited 2024 December]. Available from: https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases.

  5. Margetts BM, Nelson M. Design concepts in nutritional epidemiology. New York: Oxford University Press; 1997.

    Book  Google Scholar 

  6. Younge JO, Kouwenhoven-Pasmooij TA, Freak-Poli R, Roos-Hesselink JW, Hunink MM. Randomized study designs for lifestyle interventions: a tutorial. Int J Epidemiol. 2015;44(6):2006–19. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ije/dyv183.

    Article  PubMed  Google Scholar 

  7. Prochaska JJV, W.F; Nigg,C.R.; Prochaska, J.O. Methods of quantifying change in multiple risk factor interventions. Prev Med. 2008;46:260–5.

  8. Prochaska JJS, B.; Nigg,C.R. Multiple health behavior change research: an introduction and overview. Prev Med. 2008;46:181–8.

  9. Fine LJ, Philogene GS, Gramling R, Coups EJ, Sinha S. Prevalence of multiple chronic disease risk factors. 2001 National Health Interview Survey :Washington, DC. 2004. p. 18–24.

  10. Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS, et al. Heart Disease and Stroke Statistics—2022 Update: A Report From the American Heart Association. Circulation. 2022;145(8):e153–639. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/CIR.0000000000001052.

    Article  PubMed  Google Scholar 

  11. Sarma EA, Moyer A, Messina CR, Laroche HH, Snetselaar L, Van Horn L, Lane DS. Is There a Spillover Effect of Targeted Dietary Change on Untargeted Health Behaviors? Evidence From a Dietary Modification Trial. Health Educ Behav. 2019;46(4):569–81. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/1090198119831756. (PubMed PMID: 30808245).

    Article  PubMed  Google Scholar 

  12. Emmons KM, Marcus BH, Linnan L, Rossi JS, Abrams DB. Mechanisms in multiple risk factor interventions: smoking, physical activity, and dietary fat intake among manufacturing workers. Working Well Research Group: San Diego, CA. 1994. p. 481–9.

  13. Wilcox S, King AC, Castro C, Bortz W. Do changes in physical activity lead to dietary changes in middle and old age? [Washington, DC] :2000. p. 276–83.

  14. Dutton GR, Napolitano MA, Whiteley JA, Marcus BH. Is physical activity a gateway behavior for diet? Findings from a physical activity trial. [San Diego, CA] :2008. p. 216–21.

  15. World Cancer Research Fund/American Institute for Cancer Research. Continuous Update Project Expert Report 2018. Diet, nutrition and physical activity. 2018.

  16. Kerr J, Anderson C, Lippman SM. Physical activity, sedentary behaviour, diet, and cancer: an update and emerging new evidence. Amsterdam ; New York :2017. p. e457-e71.

  17. Henriksen HB, Ræder H, Bøhn SK, Paur I, Kværner AS, Billington S, et al. The Norwegian dietary guidelines and colorectal cancer survival (CRC-NORDIET) study: a food-based multicentre randomized controlled trial. BMC Cancer. 2017;17(1):83. Epub 2017/02/01. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12885-017-3072-4. PubMed PMID: 28137255; PubMed Central PMCID: PMCPMC5282711.

  18. Kostråd for å fremme folkehelsen og forebygge kroniske sykdommer: metodologi og vitenskapelig kunnskapsgrunnlag. Oslo: Nasjonalt råd for ernæring, Helsedirektoratet; 2011. 353 s. : ill. p.

  19. Mackey DC, Manini TM, Schoeller DA, Koster A, Glynn NW, Goodpaster BH, et al. Validation of an armband to measure daily energy expenditure in older adults. Washington, DC :2011. p. 1108–13.

  20. Cereda E, Turrini M, Ciapanna D, Marbello L, Pietrobelli A, Corradi E. Assessing energy expenditure in cancer patients: a pilot validation of a new wearable device. J Parenter Enteral Nutr. :2007. p. 502–7.

  21. Berntsen S, Hageberg R, Aandstad A, Mowinckel P, Anderssen SA, Carlsen KH, Andersen LB. Validity of physical activity monitors in adults participating in free-living activities. Br J Sports Med 44(9):2010. p. 657–64.

  22. Jetté M, Sidney K, Blümchen G. Metabolic equivalents (METS) in exercise testing, exercise prescription, and evaluation of functional capacity. Mahwah, NJ :1990. p. 555–65.

  23. Ainsworth BE, Haskell WL, Herrmann SD, Meckes N, Bassett DR, Tudor-Locke C, et al. 2011 Compendium of Physical Activities: a second update of codes and MET values. [Baltimore, MD] :2011. p. 1575–81. https://doiorg.publicaciones.saludcastillayleon.es/10.1249/MSS.0b013e31821ece12.

  24. Seca 285 [1 December 2015]. Available from: http://www.meddevicedepot.com/PDFs/seca285.pdf.

  25. Alavi DH, Henriksen HB, Lauritzen PM, Kværner AS, Sakinis T, Langleite TM, et al. Quantification of adipose tissues by Dual-Energy X-Ray Absorptiometry and Computed Tomography in colorectal cancer patients. Clinical Nutrition ESPEN. 2021. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.clnesp.2021.03.022.

    Article  PubMed  Google Scholar 

  26. Henriksen H, Alavi DH, Blomhoff R. Precision of Lunar Dual-energy X-ray Absorptiometry (iDXA) in measuring body composition among colorectal cancer patients and healthy subjects. Clinical Nutrition ESPEN. 2021;44:316–23. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.clnesp.2021.05.025.

    Article  PubMed  Google Scholar 

  27. International Society for Clinical Densitometry (ISCD) 2020 [cited 2020]. Available from: https://www.iscd.org/resources/calculators/precision-calculator/.

  28. Twisk J, Bosman L, Hoekstra T, Rijnhart J, Welten M, Heymans M. Different ways to estimate treatment effects in randomised controlled trials. Contemporary Clinical Trials Communications. 2018;10:80–5. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.conctc.2018.03.008.

    Article  Google Scholar 

  29. Enders CK. Missing data: An update on the state of the art. Psychol Methods. 2025;30(2):322-339. https://doiorg.publicaciones.saludcastillayleon.es/10.1037/met0000563.

  30. Enders CK, Du H, Keller BT. A model-based imputation procedure for multilevel regression models with random coefficients, interaction effects, and nonlinear terms. Psychol Methods. 2020;25(1):88-112. https://doiorg.publicaciones.saludcastillayleon.es/10.1037/met0000228.

  31. Enders CK. Applied missing data: Blimp3 2023 [cited 2024 April]. Available from: www.appliedmissingdata.com/blimp

  32. Gelman A, Rubin DB. Inference from Iterative Simulation Using Multiple Sequences. Statistical Science. 1992;7(4):457–72, 16.

    Article  Google Scholar 

  33. Gerstel E, Pataky Z, Busnel C, Rutschmann O, Guessous I, Zumwald C, Golay A. Impact of lifestyle intervention on body weight and the metabolic syndrome in home-care providers. Diabetes Metab. 2013;39(1):78-84. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.diabet.2012.07.00.

  34. Alavi DH, Henriksen H, Lauritzen PM, Kværner AS, Sakinis T, Langleite TM, et al. Quantification of adipose tissues by Dual-Energy X-Ray Absorptiometry and Computed Tomography in colorectal cancer patients. Clinical Nutrition ESPEN. 2021;43:360–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.clnesp.2021.03.022.

    Article  PubMed  Google Scholar 

  35. Chapman K. Can people make healthy changes to their diet and maintain them in the long term? A review of the evidence Appetite. 2010;54(3):433–41. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.appet.2010.01.017.

    Article  PubMed  Google Scholar 

  36. Hawkes AL, Chambers SK, Pakenham KI, Patrao TA, Baade PD, Lynch BM, et al. Effects of a telephone-delivered multiple health behavior change intervention (CanChange) on health and behavioral outcomes in survivors of colorectal cancer: a randomized controlled trial. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2013;31(18):2313–21. Epub 2013/05/22. https://doiorg.publicaciones.saludcastillayleon.es/10.1200/jco.2012.45.5873. PubMed PMID: 23690410.

  37. Vassbakk-Brovold K, Berntsen S, Fegran L, Lian H, Mjåland O, Mjåland S, et al. Lifestyle changes in cancer patients undergoing curative or palliative chemotherapy: is it feasible? Acta Oncologica. 2018;57(6):831–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/0284186X.2017.1413247.

  38. Kleinke F, Ulbricht S, Dörr M, Penndorf P, Hoffmann W, van den Berg N. A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study. PLoS One. 2021;16;16(9):e0257326. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0257326.

  39. Ramsey KA, Yeung SSY, Rojer AGM, Gensous N, Asamane EA, Aunger JA, et al. Knowledge of Nutrition and Physical Activity Guidelines is Not Associated with Physical Function in Dutch Older Adults Attending a Healthy Ageing Public Engagement Event. Clin Interv Aging. 2022;17:1769-1778. https://doiorg.publicaciones.saludcastillayleon.es/10.2147/CIA.S353573.

  40. Demark-Wahnefried W, Morey MC, Sloane R, Snyder DC, Miller PE, Hartman TJ, Cohen HJ. Reach out to enhance wellness home-based diet-exercise intervention promotes reproducible and sustainable long-term improvements in health behaviors, body weight, and physical functioning in older, overweight/obese cancer survivors. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2012;30(19):2354–61. Epub 2012/05/23. https://doiorg.publicaciones.saludcastillayleon.es/10.1200/jco.2011.40.0895. PubMed PMID: 22614994; PubMed Central PMCID: PMCPmc3675693.

  41. Pekmezi DW, Demark-Wahnefried W. Updated evidence in support of diet and exercise interventions in cancer survivors. Acta oncologica (Stockholm, Sweden). 2011;50(2):167–78. Epub 2010/11/26. https://doiorg.publicaciones.saludcastillayleon.es/10.3109/0284186x.2010.529822. PubMed PMID: 21091401; PubMed Central PMCID: PMCPmc3228995.

  42. Stull VB, Snyder DC, Demark-Wahnefried W. Lifestyle interventions in cancer survivors: designing programs that meet the needs of this vulnerable and growing population. J Nutr. 2007;137(1 Suppl):243s-s248 (Epub 2006/12/22 PubMed PMID: 17182834).

    Article  CAS  PubMed  Google Scholar 

  43. Ekelund U, Tarp J, Fagerland MW, Johannessen JS, Hansen BH, Jefferis BJ, et al. Joint associations of accelero-meter measured physical activity and sedentary time with all-cause mortality: a harmonised meta-analysis in more than 44 000 middle-aged and older individuals. Br J Sports Med. 2020;54:1499–1507. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/bjsports-2020-103270.

  44. Prince SA, Adamo KB, Hamel ME, Hardt J, Gorber SC, Tremblay M. A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008;5(1):56. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/1479-5868-5-56.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Matthews CE, Keadle SK, Troiano RP, Kahle L, Koster A, Brychta R, et al. Accelerometer-measured dose-response for physical activity, sedentary time, and mortality in US adults. The American journal of clinical nutrition. 2016;104(5):1424–32. Epub 20161005. https://doiorg.publicaciones.saludcastillayleon.es/10.3945/ajcn.116.135129. PubMed PMID: 27707702; PubMed Central PMCID: PMCPMC5081718.

  46. Educational attainment of the population: Statistics Norway; 2024 [10.04.2025]. Available from: https://www.ssb.no/en/utdanning/utdanningsniva/statistikk/befolkningens-utdanningsniva.

  47. Norway Cro. Cancer in Norway 2012 2012 [cited 2012 10.04.2025]. Available from: https://www.kreftregisteret.no/globalassets/cancer-in-norway/2012/cin_2012-web.pdf.

Download references

Acknowledgements

We would like to thank the patients for their invaluable contribution to this study.

Funding

Open access funding provided by University of Oslo (incl Oslo University Hospital) This project has received funding from Research Council of Norway, Throne Holst Foundation of Nutrition Research, Norwegian Cancer Society, and the Southeastern Norway Regional Health Authority. The funder of the study had no role in data collection, the decision to publish or the preparation of the manuscript.

Author information

Authors and Affiliations

  1. Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, Blindern, P.O. Box 1049, 0316, Oslo, Norway

    Hege Berg Henriksen, Åshild Kolle, Ingvild Paur & Rune Blomhoff

  2. Division of Health Services, Norwegian Institute of Public Health, Oslo, Norway

    Hege Berg Henriksen

  3. Department of Psychology, Umeå University, Umeå, Sweden

    Andreas Stenling

  4. Department of Sport Science and Physical Education, University of Agder, Kristiansand, Norway

    Andreas Stenling & Sveinung Berntsen

  5. Norwegian Advisory Unit On Disease-Related Undernutrition, Oslo University Hospital, Oslo, Norway

    Ingvild Paur

  6. Department of Clinical Service, Division of Cancer Medicine, Oslo University Hospital, Oslo, Norway

    Ingvild Paur & Rune Blomhoff

  7. Biotechnology and Food Science, Norwegian University of Life Sciences, Ås, Norway

    Siv Kjølsrud Bøhn

  8. Møre Og Romsdal Hospital Trust, Ålesund, Norway

    Pernille Brøto

  9. Department of Diet Care and Nutrition, Bærum Municipality, Bærum, Norway

    Tuva Syrdal Tronstad

Authors
  1. Hege Berg Henriksen
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Åshild Kolle
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Andreas Stenling
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Ingvild Paur
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Siv Kjølsrud Bøhn
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Pernille Brøto
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Tuva Syrdal Tronstad
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Rune Blomhoff
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Sveinung Berntsen
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

HBH had the main responsibility for writing the manuscript. HBH, ÅK, AS, IP, SKB, PB, TST, RB and SB, contributed to the conception and the design of the study and drafting of the manuscript. HBH, ÅK, PB, TST, IP, SKB contributed to acquisition of data. AS and ÅK contributed with analyses. All authors contributed with the interpretation of the data. All authors contributed to the writing and approval of the final manuscript.

Corresponding author

Correspondence to Hege Berg Henriksen.

Ethics declarations

Ethics approval and consent to participate

The CRC-NORDIET study is carried out in accordance with the Helsinki Declaration. The study, including all secondary outcomes, was approved by the Regional Committees for Medical and Health Research Ethics (REC Protocol Approval 2011/836) and by the data protection officials at Oslo University Hospital and Akershus University Hospital. Informed consent was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Henriksen, H.B., Kolle, Å., Stenling, A. et al. Spillover effect of a dietary intervention on physical activity in a randomized controlled trial with colorectal cancer patients. Int J Behav Nutr Phys Act 22, 54 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12966-025-01757-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12966-025-01757-0

Keywords

International Journal of Behavioral Nutrition and Physical Activity

ISSN: 1479-5868

Contact us