Study design:
This is an assessor-blinded randomized controlled trial that utilized a simple random sampling technique in sample selection. This study was registered with the Pan African Clinical Trials Registry - Trial ID: PACTR202001542824141, (https://pactr.samrc.ac.za/Search.aspx) on the 28/01/2020.
Participants
To be eligible to participate in this trial, women were: i. Diagnosed with breast cancer, ii. Aged 18 – 60 years, iii. Ambulating independently, iv. Not on sedatives at least 4 weeks prior to the study (since it decreases systemic vascular resistance due to inhibition of vasoconstrictor activity of the sympathetic nervous system, decreases cardiac contractility, and decreases preload46, v. Had no history of hypertension (because anti-hypertensive drugs influence the neurohumoral cardiovascular system and the concentration of hormones involved in blood pressure regulation47, vi. Staging of breast cancer was within stages I - III according to Surveillance, Epidemiology, and End Results (SEER) stage of disease classification,48 vii. Never-smokers (since smoking influences the serum level of Superoxide dismutase)49, viii. On primary treatment (chemotherapy, radiotherapy surgery) for ≤2 months, and ix. Without a visual loss. Participants were excluded if they had a history of other underlying pathological diseases of metabolic, orthopaedic or neurological nature.
Study setting
The study site was primarily at the Oncology centre, University of Nigeria Teaching Hospital, Ituku/Ozalla, Enugu, Enugu State, Nigeria. Enugu State, has an estimated population of 3.8 million, and is one of the 36 states of Nigeria.
Intervention
a. Intervention arm
The primary intervention comprised of two components delivered over a six‐week period.
Physical activity consultations, behavioural feedback and goal‐setting sessions were held with the participants to address the barriers to physical activity using the TTM [49], throughout the study duration. The TTM is a common theoretical framework for physical activity consultations [50], which has published guidelines for health professionals to conduct consultations [51, 52] and has been used successfully in intervention studies designed to increase physical activity [53, 54]. According to the guidelines, the consultations were semi-structured [51, 52], and adopted a guiding style by which the participants made decisions about how to modify and promote their walking behaviour [55] using various strategies. These strategies included: enhancing motivation, overcoming barriers and developing appropriate walking plans which were personalized to the individual as recommended by Ogilvie and colleagues [56]. The sessions also included a reflection on the three mediators of The TTM required for behaviour change [49], namely, i. self-efficacy (confidence in the ability to change), ii. decisional balance (pros and cons of change) and iii. processes of change (strategies and techniques used to change, e.g., social support). Although the sessions were flexible and personalized, The TTM was used to create a standard protocol and steps that were followed during the sessions (Table 1).
Table 1
Weekly goals of intervention group participants
Time-point
|
Goal
|
Week 1
|
To walk an extra 1,500 steps (from baseline value) on at least 3 days of the week
|
Week 2
|
To walk an extra 1,500 steps (from baseline value) on at least 3 days of the week
|
Week 3
|
To walk an extra 1,500 steps (from baseline value) on at least 5 days of the week
|
Week 4
|
To walk an extra 1,500 steps (from baseline value) on at least 5 days of the week
|
Week 5
|
To walk an extra 3,000 steps (from baseline value) on at least 3 days of the week
|
Weeks 6
|
To maintain walking levels using the week 5 goal
|
The walking programme lasted for six weeks, and varied as follows: the first 5 weeks consisted of graduated weekly goals with an aim for the increased walking behaviour to be maintained for the remaining one week (at the 6th week). The overall goal of the walking programme was for the participants to increase their mean daily step-count by 3,000 accumulated steps above their baseline value on five days of the week. Three thousand steps is based on the assumption that moderate brisk walking produces 100 steps a minute (1,000 steps per 10 minutes) [57]. Therefore, 3,000 steps would equate to approximately 30 minutes of moderate physical activity, in line with current physical activity recommendations [58]. This program has previously been used in sedentary adults, in the United Kingdom over a shorter time-frame [59]. Goals were retained for one week to enable participants to reinforce their increased levels of walking or to try other strategies to successfully accumulate the additional steps. Participants were advised on the nature of the intensity and duration of the desired increases in walking. Participants were familiarized with the Borg 6–20 scale [60] on which basis they were advised to ensure that the additional walking was a brisk nature that would leave them slightly breathless and hot, but still able to talk (indicated as between 12–14 on this scale). Bouts of at least 10 minutes in duration were recommended for additional walking, although the accumulation of walking during everyday tasks as much as possible was also advocated. The participants in the intervention group collected their pedometers at the end of baseline assessment visit and wore it unsealed until the end of week six of the study.
b. Control arm
Participants assigned to the control group were asked to maintain their normal physical activity level/walking behaviour between baseline and week six. They were advised to continue their current management program and scheduled for follow-up at the end of week five. The participants in the control group collected their pedometers at week 0, and wore it sealed in order to gain a record of their daily pedometer step counts at baseline. They returned it at the end of week 0, but were given the pedometers again at the end of week five, and were required to wear it sealed all through week six to gain a record of their step-counts. At the end of week 0 and week 6, other health outcomes apart from pedometer step counts/day, were also measured among all the participants. Prior to the assessment session, each participant was rested for 30 minutes.
Outcome variables
The outcome variables of interest that were measured in this study include:
-
A. Independent variables
-
1. Physical Activity
The overall daily physical activity was measured using two methods, namely pedometer step counts/day and physical activity expressed as MET-min per week
-
Pedometer step counts/day: The primary outcome measure is pedometer step counts/day measured by Fitbit Charge HR (Fitbit International Ireland). This was measured as step counts/day. The cut off points according to Jordan et al [70] are:
-
low step count = ≤7,999 step counts/day
-
high step count = ≥8000 step counts/day
-
high physical activity level=3000 MET-min/week
-
moderate physical activity level=600-3000MET-min/week
-
low physical activity level = 600MET-min/week
-
B. Dependent variables
-
2. Body composition. The cut-off points according to World Health Organisation (WHO) [1] are: -
-
Waist-to-hip ratio according to previous studies [1-3] is categorised into:
-
normal adiposity for females = <0.85;
-
abdominal adiposity/obesity = >0.85
-
Body mass index according to previous studies [1-3] is categorised into:
-
normal weight category =≤20Kg/m2 - ≥ 25Kg/m2
-
overweight = 25Kg/m2 - 29Kg/m2
-
Obesity = ≥ 30Kg/m2
-
Percentage body fat mass: Cut-off for percentage body fat mass based on previous studies [71, 72] are:
-
Overweight = >30% - <32%
-
Obesity = ≥32%
Sample size
The sample size of this study is based on our primary aim: to determine the efficacy of the pedometer-based physical activity behaviour promotion in influencing physical activity level of breast cancer survivors, of which the primary outcome variable is minutes per week of moderate to vigorous activity. Using the statistical superiority sample size design formula [44], assuming a between group change of 75 minutes per week in moderate to vigorous physical activity (MVPA), standard deviation of change of 120 minutes [45], power of 80% and 95% confidence interval, the estimated sample size accounting for 10% dropout is 44 participants per group (N=88).
Randomisation, sequence generation and allocation
Randomisation, sequence generation and allocation of participants were done after their baseline assessment to determine the step count/day and physical activity level on which basis they were stratified, randomised and allocated into groups.
-
Baseline assessment of the stratification variable
At onset, the participants collected an individually-calibrated pedometer from the research centre and wore it sealed during the week zero to gain a record of their baseline step-counts/day. All participants completed a baseline week wearing a pedometer for seven days at all times (except when showering, sleeping or taking part in structured sport or exercise (e.g., athletics, gym workouts, football, dance aerobics, etc.) with instructions not to alter their daily routine. Irregular bouts of structured sport or exercise that could significantly affect an individual's mean step-count was recorded. Each pedometer was individually calibrated consistently with manufacturers guidelines to within 5% of actual steps walked in a 100-step test. Participants were required to provide at least five days of step counts including at least one weekend day to gain an accurate reflection of physical activity levels. At the end of the baseline week, the participants returned the pedometers to the study centre. During the visit, the baseline The International Physical Activity Questionnaires (IPAQ) scores of each participant was recorded. The participants were subsequently stratified into low (<600MET-min/week), moderate (600-3000MET-min/week) and high levels (3000 MET-min/week) of physical activity on the basis of their total physical activity (MET-min/week) and the frequency of activities.70 The participants within each IPAQ score category above were further stratified into two groups according to their step counts/day using baseline low step count of ≤7,999, and high step count of ≥8000. The value of 8,000 step count/day was used as a stratification variable to account for individuals with a high baseline step-count. This value was previously used as a baseline descriptor for sedentarism. 73
From a frame of the breast cancer survivors who met the eligibility criteria, participants were randomized into groups using computer-generated numbers and concealed allocation. Thus, using block allocation, participants from each of the stratum mentioned previously (i.e., low step count of ≤7,999, and high step count of ≥8000) were also randomized into the two study groups, namely: -
-
Intervention group – which comprised of participants who were given a pedometer to monitor their step counts/day, and,
-
Control group - which comprised of patients without a pedometer to monitor their step counts/day.
To achieve this, random numbers were generated to allocate the patients into intervention and control groups in blocks (10 numbers equally distributed into intervention and control groups). The blocks were divided into two, namely – low step count of ≤7,999, and high step count of ≥8000, and concealed in an envelope. The number allocation was done by independent research support and was done in such a manner that until the 10 numbers in an envelope were exhausted before a new block of 10 numbers were allocated. This process was continued until the sample was exhausted.
Blinding
The principal investigator, who was the assessor, was blinded to allocation. A qualified registered physiotherapist was appointed as the research assistant 1 (RA 1) for the study. The research assistant 1 kept the names of the randomised participants in a centrally locked steel-drawer and arranged for the participants to undergo assessment by the assessor without informing the assessor which groups a participant belonged to. The research assistant 1 implemented the intervention programme and did not allow the assessor access to the groups during intervention period.
Data Collection
Recruitment
Recruitment strategy involved 6-week publicity campaign about the study involving public health talks for Breast cancer survivors who attended the Oncology Clinic., Flyers were placed in the Physicians consulting rooms in private and public health facilities in Enugu Metropolis, Enugu State, Nigeria. Announcements were made in local religious congregations, interested participants contacted the researcher through email, phone number, and social media handle provided in newsletters. Interested participants were screened for eligibility and recruitment started on 27 July, 2016. The study lasted from 27 July 2016 and 26 July, 2017. The participants were followed-up from 27 July 2017 to 26 January, 2018. Each participant had a documented diagnosis, treatment and history of being followed-up at the Oncology Clinic. The time of recruitment and assessment was the usual routine clinic hours (8 – 4 pm) at the oncology clinic.
Procedure.
The study process involved three stages: obtaining informed consent, assessment of body anthropometry and physical activity profile, and laboratory study. At the onset of the study, written informed consent was obtained from all participants prior to their involvement in the pedometer-based walking intervention trials. All participants were provided with a diary to record their daily physical activity, a pedometer with instructions on how to use it and IPAQ-SF questionnaire for baseline and post-intervention data collection. The clinical trials commenced at week zero (0) at the hospital. The participants who were eligible, and willing to participate in the study were reviewed by the research assistant 1 (RA 1) for pre-identified parameters of breast cancer at stages I and II as well as evaluate their physical activity profile after obtaining their biodata. Participants were required to complete the Physical Activity Readiness Questionnaire which have been used in previous studies [46, 47] to assess their suitability for an exercise programme. A “yes” answer to any of the questions requires that approval be obtained from the participant's physician in order to take part. They were educated on the purpose/procedure for the study. For participants in the intervention (pedometer) group, a physiotherapist (RA2) administered the six-week pedometer-based walking programme according to the Trans-Theoretical Model (TTM) framework to ensure quality [48]. A follow-up home programme guideline was provided. The physiotherapist also gave them a follow-up guide on how to continue the programme at home using the standardized guideline. The Physiotherapist was not blinded to participant allocation but the assessor was.
Instruments
Participants in the Intervention (pedometer) and control groups were provided with pedometers which they wore on their wrist. The pedometer step counts/day was measured with the Fitbit Charge HR (Fitbit International Ireland) (Plate 1). The choice of this technology was made because the inter-device reliability of the Fitbit is good in measuring steps at all levels of aggregation (minutes, hours, days), but especially when steps were measured per day, and therefore individuals can reliably relate their daily physical activity scores with peers [61]. Furthermore, a systematic review on walking- and running-based pedometer trials indicated consistently high inter-device reliability for steps (Pearson and intraclass correlation coefficient - CC 0.76–1.00), distance (Interclass CC 0.90–0.99), and energy expenditure (Pearson and intraclass CC 0.71–0.97) with Fitbit [62]. To the researcher's knowledge, this model of pedometer has been utilized and validated in many previous intervention studies. In fact, good inter-device reliability was reported in seven studies using the Fitbit [62]. The Fitbit charge HR has several features beneficial to an intervention of this nature, including a cover to prevent accidental resetting, continuous heart rate measurement for 24 hours, tracks all physical activities in a day by providing number of steps, distance covered, floors climbed, calories burned, active minutes, time spent, hours slept, time awakened, sleep quality, display the statistics in wirelessly synchronized smartphones/computer monitors, and a 7-day memory which negates the need for participants to record their own step-counts.
The International Physical Activity Questionnaire-short form (IPAQ-SF) is a subjective measure of physical activity was used to examine changes in modes of physical activity that would not be measured by the pedometer such as swimming or other forms of structured sport or exercise. Thus, daily physical activity was also measured using the IPAQ-SF [63]. The IPAQ-SF is a valid and reliable seven-item tool applied in measuring self-reported physical activity in the last seven days [63, 64].
It collects information about moderate and vigorous physical activity across four domains: work-related, transportation, housework/gardening, and leisure-time physical activity. The detail on specific physical activity domains, therefore, enabled the identification of specific domains of physical activity where changes may have occurred. Walking time is also included for the work, transport and leisure domains. Two additional questions measured time spent sitting which can be used as an indication of sedentary time. This questionnaire asks about three specific types of activity, namely walking, moderate-intensity activities, and vigorous-intensity activities [63]. The minutes spent every week on each type of activity were computed separately by multiplying the duration and frequency of activity [65]. A continuous activity score was calculated by multiplying the selected metabolic equivalent (MET) value and weekly minutes of activity, therefore expressing physical activity as MET-min per week [66]. The MET values and the levels of physical activity were calculated according to the guidelines for data processing and analysis of the IPAQ-SF [66] [67]. Criterion validity of IPAQ-SF had a median of about 0.30, which was comparable to most other self-report validation studies [67]. Test-retest reliability data for the IPAQ-SF using Spearman correlation coefficients across 12 countries ranged from 0.96 (USA2) to 0.46 (SA Ru), but most were around 0.8 [63].
Body anthropometry was measured while the participants were standing, wearing light clothing and no shoes.
-
Height The height was measured as the distance from the height scale platform to the vertex of the head using a stadiometer (Sartorius, AG Gottingen). The height was then, read-off, and recorded to the nearest 1.0 the centimetres [68].
-
Body mass index: The Fat Loss Monitor with Scale (OMRON® HBF-400) was used to measure the body weight in Kg. The body mass index was also calculated as the square of height (m)/weight (kg). [69].
-
Waist-hip-ratio: The Waist and hip circumference were measured using a SECA 200 (SECA, Birmingham, UK) measuring tape. The waist-hip-ratio was calculated from these measurements.
-
Body fat mass; The percentage body fat mass was estimated using the Fat Loss Monitor with Scale (OMRON® HBF-400. Though the DEXA (Dual Energy X-Ray Absorptiometry) has been the established method for the accurate evaluation of body composition. However, OMRON has used research information from several hundred people from 10 to 80 years of age using the DEXA method to develop the method by which the Fat Loss MONITOR with Scale works. The body fat mass and body fat percentage are quantified by a method that includes five factors: electrical resistance, height, weight, age and gender. Body fat percentage was measured from 5 - 60% in 0.1% increments [69].
Laboratory study of all participants was done by a trained laboratory scientist. At baseline and end of week 6, each participant’s blood (5mls) was drawn from the subcutaneous vein of the arm following standard blood sample collection procedures. Subsequently, the collected blood samples were stored in Serum Separator Tube (SST) without an anticoagulant, thereby allowing the clotting of the blood sample. This facilitated the separation of the serum from blood cells. The blood sample was kept at 40οC prior to centrifugation (3000 rpm, 10 min, 0-40C). The serum level of Superoxide dismutase was measured by the procedure described by the manufacturer of the kit (BioAssay Systems, USA). A clear flat-bottom 96 well plate was used which contained serum samples as dilutes, 1:5 prior to assay. Thereafter, 20µL superoxide dismutae standards and samples were put into separate wells of the 96-well plate, with a clear flat-bottom, into which 160 µL working reagent (160µL Assay Buffer, 5µL Xanthine and 5µL WST-1) was added and mixed thoroughly. Subsequently, 20µL of diluted XO enzyme (1:20 in diluent) was rapidly put into the plate using a multichannel pipette, and mixed. The optical density (OD) of the mixture was read instantly at 440 nm (OD0). The clear flat-bottom 96-well plate was at that point incubated at 60 minutes in the dark, at room temperature. The mixture was again read after incubation at OD 440nm (OD60). For each standard and sample, we calculated ∆OD60 =OD60 - ODO. A standard curve was plotted and utilised to estimate the sample concentration. Superoxide dismutase was measured in (U/ml).
Covariates
Covariates assessed and documented at baseline included: marital status, (single, married, widowed); occupation (Housewife, Farmer, Civil servant, Trader, Student). Breast cancer stage at diagnosis, family history of cancer, and treatment received - surgery (had surgery - Lumpectomy and Partial mastectomy; or no surgery; chemotherapy; and radiotherapy) were reported at initial review to assess eligibility prior to study enrolment.
Follow-up:
As the primary outcome measure of this study is physical activity level, strict follow-up on physical activity behavioural pattern was indicated. After completion of the six-week pedometer-based walking programme, physiotherapists set the optimal exercise level that should be adhered to during the follow-up periods. Over the follow-up period, participants were evaluated by their physiotherapist and medical oncologist at six weeks, twelve weeks and six months. At each assessment session, participants’ history was obtained and likewise their physical activity level, antioxidant (superoxide dismutase) level and body composition. Any treatment administered to any breast cancer survivor for second, third or subsequent recurrence of breast cancer was appropriately documented. All the collected data were recorded on a follow-up Case Record Form (CRF).
Data Analysis
Data were analyzed using XLSTAT statistical and data analysis solution software (Boston, USA). Descriptive statistics were used to present the data in tables as frequency and percentages. Data collected were tested for normality using the Kurtosis (Fisher) Normality test. Subsequently, baseline differences in steps/day, physical activity level (MET-min/week), body composition (waist-to-hip ratio, percentage body fat mass and body mass index) and serum level of Superoxide dismutase between the intervention and control groups were tested for statistical significance using independent t-test. Statistical significance is defined as p<0.05 for all tests with data presented as mean and standard deviation.
Ethical consideration:
All methods were performed in accordance with the relevant guidelines and regulations of the University of Nigeria Health Research Ethics Committee, and in accordance with the Declaration of Helsinki.