Article Text
PDF
Abstract
Background Low physical activity (PA) levels which increase the risk of chronic disease are reported by two-thirds of the general UK population. Promotion of PA by primary healthcare professionals is advocated but more evidence is needed regarding effective ways of integrating this within everyday practice. This study aims to explore the feasibility of a randomised trial of a pedometer-based intervention, using step-count goals, recruiting patients from primary care.
Method Patients, aged 35–75, attending four practices in socioeconomically deprived areas, were invited to complete a General Practice PA Questionnaire during routine consultations. Health professionals invited ‘inactive’ individuals to a pedometer-based intervention and were randomly allocated to group 1 (prescribed a self-determined goal) or group 2 (prescribed a specific goal of 2500 steps/day above baseline). Both groups kept step-count diaries and received telephone follow-up at 1, 2, 6 and 11 weeks. Step counts were reassessed after 12 weeks.
Results Of the 2154 patients attending, 192 questionnaires were completed (8.9%). Of these, 83 were classified as ‘inactive’; 41(10 men; 31 women) completed baseline assessments, with the mean age of participants being 51 years. Mean baseline step counts were similar in group 1 (5685, SD 2945) and group 2 (6513, SD 3350). The mean increase in steps/day was greater in groups 1 than 2 ((2602, SD 1957) vs (748, SD 1997) p=0.005).
Conclusions A trial of a pedometer-based intervention using self-determined step counts appears feasible in primary care. Pedometers appear acceptable to women, particularly at a perimenopausal age, when it is important to engage in impact loading activities such as walking to maintain bone mineral density. An increase of 2500 steps/day is achievable for inactive patients, but the effectiveness of different approaches to realistic goal-setting warrants further study.
Statistics from Altmetric.com
Introduction
Non-communicable diseases place an increasingly large burden on our healthcare systems. The WHO estimated that in 2001 the major non-communicable diseases accounted for nearly 60% of all deaths annually and 47% of the global burden of disease.1 A major risk factor for non-communicable diseases is physical inactivity.1 Indeed, Lee et al2 report that physical inactivity accounts for 6–10% of all global deaths annually. Thus, the prevention of non-communicable diseases should be a main priority for healthcare practitioners. However, there has been minimal investment researching effective ways to encourage physical activity (PA) uptake within the UK.3
The UK PA guidelines have recently been updated4 and the WHO has advocated walking as a solution to solve the burden of physical inactivity.5 Tudor-Locke6 advises that walking can be easily performed at moderate intensity, fulfilling the exercise recommendations, which have been translated into pedometer targets. A cadence of approximately 100 and 130 steps/min is considered the floor value for moderate and vigorous-intensity activity, respectively.7
Pedometers have been shown to be accurate and reliable in measuring ambulatory activity 7–10 while also being relatively inexpensive. The Yamax pedometers have been shown to be the most accurate waist-borne instruments.8 A monitoring frame of 5–6 days has been shown to accurately describe the ambulatory activity pattern of young men.8
So why does everyone not undertake PA? There are significant health inequalities which exist in relation to PA participation. Both intrinsic and extrinsic factors including gender, age, ethnicity, disability, income,11 and the suitability and safety of the locality all can impact on PA participation. In particular, further research is required to determine effective ways of promoting PA among those living in areas with high levels of socioeconomic deprivation.
It is estimated that 78% of people consult their general practitioner (GP) at least once during each year11 ,12 and patients view GPs as a reliable source of advice.13 ,14 The UK government11 and National Institute of Health and Clinical Excellence (NICE)15 advise that routine health contacts should include practical advice on the benefits of increased levels of PA and how to initiate and maintain healthy behaviours. Thus, primary care encounters present a major opportunity to prevent disease and to promote healthy lifestyles.
However, the challenge remains to develop effective programmes within primary care to increase PA in the general population.10 ,16 ,17 Indeed, NICE has highlighted the need for further research into promoting PA in general practice through pedometers.15 This study aims to test the feasibility of conducting, within general practices in socioeconomically disadvantaged areas, a randomised controlled trial of interventions using pedometers and different approaches to set specific daily step-count goals.
Method
This feasibility study was conducted in four general practices situated in socioeconomically disadvantaged areas of Belfast, defined by the measure of deprivation (Multiple Deprivation Measure (MDM) (NISRA, 2010)), from April 2011 to August 2011. Ethics approval was granted through the Office for Research Ethics Committees Northern Ireland, reference number 11/NIR03/2, 15/03/2011.
During a 2-week period in each practice, individuals, aged 35–75 years, attending for consultation, were invited to complete a General Practice PA Questionnaire (GPPAQ) by either the GP or practice nurse. GPPAQ has been proposed by NICE as a validated tool for use in general practice to screen patients for physical inactivity between the ages of 16 and 74 years, 15 and its routine use has also been encouraged by the UK government.18 GPPAQ provides a 4-level Physical Activity Index output categorising patients as active, moderately active, moderately inactive and inactive and can be completed by the patient or health professional in approximately 1 min.19 GPPAQ was developed from the basis of a short questionnaire by Wareham et al,20 and its use has been encouraged within the National Health Service through the ‘Let's Get Moving’ campaign.11
Those categorised as ‘inactive’ from the GPPAQ were asked if they would agree to being approached by the researcher: those who agreed were invited to reattend their own GP premises where the researcher obtained their informed consent and ensured that they had no contraindications to PA using PAR-Q.7 ,21
Data regarding age, gender, marital status, ethnic group, smoking habits, alcohol intake, social class and occupation were recorded by the researcher for the participants. Quality of life was assessed using EQ-5D-5L. Height was measured without shoes to the nearest 0.5 cm using a calibrated stadiometer (SECA Leicester Height Measure). Body weight was measured in light clothing and without shoes to the nearest 0.1 kg using a standard physician's balance scale (Tanita, Japan). A Prochaska stages of change questionnaire was also completed at the initial and final assessments.
Waist girth was calculated standing as described in Chan et al22 Resting heart rate and blood pressure were measured simultaneously using an electronic blood pressure monitor (Omron M5-I) after having the participant sit quietly for 5 min.
All participants were then given a pedometer (Yamax CW-700/701) with a 7-day memory, instructed how to use it and invited to return 1 week later. Participants were blinded to their baseline daily step counts by sealing the pedometer and were encouraged not to alter their usual PA habits over this initial 1-week assessment to allow an accurate measurement of basal activity.21
Participants were then asked to return the pedometer to their general practice surgery and to meet the researcher again, at the end of the assessment week, to receive their group allocation for the study. Individuals were randomised to one of two groups by an individual independent of the research team, using a computer-based programme, and allocations were not revealed until the initial assessment was complete.
Group 1 received information on their baseline step count, a brief health promotion leaflet on PA, obtained from the regional Public Health Agency, and, with the researcher, negotiated a self-determined step-count goal. The target was set by the participant after reviewing their baseline step count and considering their own perceived ability to increase this in the weeks following. This group then followed a 12-week pedometer-based PA programme, with follow-up telephone calls at 1, 2, 6 and 11 weeks, and were advised to record daily step counts within a step-diary. The follow-up telephone calls included a review of the recent pedometer step counts, if the participant was experiencing any problems with wearing the pedometer, and then a review of the step-count goal using motivational interview techniques as appropriate. Telephone calls were repeated until participant contact was made and no voicemail messages were left. If the pedometer was not worn on a day for any reason, participants were asked to leave the activity log blank on that day to indicate non-compliance.
Group 2 followed the same pedometer programme and received the same telephone review as group 1 but were given a specific, uniform daily step-count goal of 2500 steps/day above baseline. The only difference, therefore, between groups 1 and 2 was the nature of the step-count goal.
The lead researcher was a doctor in the final year of his family practice training, and he adopted motivational interviewing techniques in both groups at baseline and during the telephone follow-up calls using behaviour change counselling as per the PRE-EMPT study14 and the theory of planned behaviour.23 This involved following the ‘5 As’ guideline.24 ,25 No formal training in motivational interviewing techniques was undertaken beyond that experienced within undergraduate medical training.
All participants were invited to return to their general practice at the end of the 12-week programme for a repeat of the baseline measurements and a further 7-day assessment of their PA using a sealed pedometer.
Statistical analysis
No sample size calculation was undertaken as this was a pilot study. Intention-to-treat analysis was used to estimate the effects of the intervention. Six individuals who agreed to take part in the pedometer programme did not complete their baseline assessment. No analysis of their data was undertaken.
One participant (group 2) dropped out after 4 weeks due to a local allergy to the pedometer metal backing but agreed to be included in the study analysis. She was treated as if the intervention had no impact on her measurements and her baseline parameters were substituted for follow-up measurements.
Another participant (group 1) dropped out after 8 weeks as she relocated to another country. She completed follow-up measurements at 8 weeks and was treated like she had completed the full study (figure 1).
Study flow diagram.
The baseline and follow-up scores were analysed using SPSS (Statistical Package for Social Sciences, V.19). The main outcome measure was the mean number of steps/day. Normally distributed continuous data were analysed using student's t tests. Non-parametric data were analysed using the Mann-Whitney U test and Wilcoxon Signed Rank test. χ2 Tests were used to test for differences in categorical data.
Results
Baseline characteristics
Overall, there were 2154 eligible consultations in the four practices. 192(8.9%) completed questionnaires and 152 of these were completed by the GP. Eighty-three (43%) were categorised as inactive: 41(49%) of these participated in the trial (10 men, 31 women) and 21 were assigned to group 1 with 20 in group 2. Recruitment rates in different practices varied from 23/29 (79%) in Practice 1 to 10/36 (28%) in Practice 2 (figure 1). All participants received the same number of telephone contacts and all were contacted within three attempts.
An MDM was derived for each study participant from their postcode.26 There are 890 MDM rank scores for postcodes in Northern Ireland, 1 being the most deprived and 890 the least deprived. The majority of participants lived in some of the most deprived areas of Northern Ireland: 50% of group 1 and 52.4% of group 2 lived in areas within the lowest quintile of scores.
No significant differences were found between the groups in baseline parametric data (table 1). Of note, group 2 was approximately 7 kg heavier and appeared more active by just over 800 steps/day. The intraclass correlation coefficient for baseline step counts in the participating practices was zero.
Parametric baseline characteristics for the two groups
Non-parametric baseline data (tables 2 and 3) showed no statistically significant differences between the groups.
Non-parametric baseline characteristics for the two groups
Non-parametric baseline characteristics for the two groups
Changes in outcome measurements
Group 1, with the self-determined daily step-count goal, achieved an average increase of over 2600 steps/day from baseline, which was statistically significant compared with the increase in group 2 (mean=748, SD 1997; table 4). Group 1 also reduced on average their body mass index (BMI) and weight by 0.29 kg/m2 and 0.71 kg, respectively, although these differences did not prove statistically significant when compared with group 2. There was also a reduction in waist girth of approximately 2 cm in each group, although no statistically significant within or between group difference was detected.
Parametric post-intervention results for the two groups
No significant differences between the groups in the quality of life, change in ‘stages of change’ or categorical level of PA were found (table 5). All telephone contacts with participants were delivered as planned and their step-diary returns for both groups showed evidence of similar levels of compliance in recording daily step counts.
Non-parametric post-intervention results for the two groups
Two participants noted the effects of local irritation from the pedometer metal back at their waist. No other injuries were reported from the participants.
Discussion
This randomised controlled feasibility study has shown that pedometers and interventions using self-determined step-count targets can be evaluated by recruiting patients within routine GP contacts. The study participants lived in some of the most socioeconomically deprived areas of Northern Ireland—this is a group of patients who need focused PA promotion to help bridge known health inequalities.11 ,27 The findings suggest that pedometer-based interventions in this context can support inactive patients in achieving increases in their levels of PA.
Comparison with the previous literature
Patients at baseline were on average doing approximately 5500–6500 steps/day, which is similar to the daily step count previously shown for healthy adults of between 6000 and 7000 steps/day.6 ,16 There was no statistically significant difference between the groups in their sex composition; 76% women uptake is in keeping with previous studies.16 This is important, as national statistics often highlight the sex difference in PA levels, with men being more active than women.28 ,18 Moreover, the mean age of participants was just over 50 years, which is approximately the mean UK menopausal age. Pedometers, therefore, appear to be one suitable mode of PA promotion for women, particularly at the perimenopausal age, when it is important to engage in impact loading activities such as walking to maintain bone mineral density.29 ,30
Group 1 achieved over 2500 steps/day increase after having set their own goals, following a consideration of their baseline step counts and their own perceived ability to change their level of activity. The group given the specific 2500 steps/day goal increase only achieved an increase of approximately 800 steps/day. Previous studies have shown an increase in step counts of between 1800 and 4500 steps/day.16 This study confirms that an increase of 2500 steps/day is achievable even in an inactive subgroup. However, if inactive patients, who typically do between 6000 and 7000 steps/day, are initially presented with this data, they may think that this goal is unachievable and it may reduce their enthusiasm for the intervention. The findings tend to support patients setting their own step-count targets as opposed to receiving a predefined goal.
It is important to encourage patients to set realistic self-determined step-count targets and then review their level of achievement. Thus, our study expands the literature on appropriate goal-setting for using pedometers in inactive participants recruited from ‘routine’ general practice. Recent studies31 have advocated the role of walking in increasing PA at a population level and pedometers appear to be one acceptable form of facilitating this. Indeed, De Cocker et al32 reviewed the long-term step-count compliance of people undertaking the ‘10 000steps Ghent project’ and found that the step-count increase was maintained at 1-year follow-up for those who were given pedometers compared to reduced step counts in the non-intervention group.
Secondary outcomes
The study was not powered to identify statistically significant changes in biophysical outcome measures and none were observed. In their review of the effects of pedometer interventions, Bravata et al33 found that a reduction in BMI and systolic blood pressure was associated with increased step counts. Our group 1 participants, with their increased steps/day, showed a mean weight reduction of 0.71 kg. A similar weight loss has been observed in previous studies.34 Group 1, compared with group 2, also showed a reduction in BMI of 0.29 kg/m2, and waist girth fell in both groups. Waist girth is associated with the metabolic syndrome and shows a correlation with future cardiovascular events,35 which needs to be emphasised to the public and may incentivise some to consider increasing their PA.
Within both groups, the mean BMI was greater than 28. In the context of the study setting, there was an available free 12-week PA scheme in a local fitness centre (‘Healthwise’), open for referral by health professionals, for patients with, for example, high blood pressure, BMI>25 kgm2 and diabetes. Despite this having no cost implications for either the patient or the practice, only one referral was made from the participating practices over the 8-week recruitment period, suggesting that this was not a preferred management option. A systematic review on exercise referral schemes36 concluded that gym-referral schemes are ‘probably not an efficient use of resources.’ Pedometers appear to offer a realistic management option for use in primary care to help support patients in increasing their levels of PA.
Strengths and weaknesses
The strengths of the study included that baseline assessments were completed before randomisation of the study participants, thus minimising the risk of bias. The groups were well matched in baseline parameters and there was a high level of participant retention. The dropout rates for the study (2/41; 4.8%) are lower than those for previous studies in the primary care setting. For example, Bull and Milton 37 reported a 30% completion rate for their 3-month PA programme. We did, however, note differences in the numbers of questionnaires completed and recruitment rates in different practices: these differences are partly explained by variations in practice size (numbers of patients registered) and systems of questionnaire administration but require further study using qualitative methods prior to undertaking a definitive trial. The qualitative exploration of participants’ and staff views about the study processes and the challenges faced in recruitment and in seeking to achieve behavioural goals will also inform the design of future work, to ensure its relevance in practice.
Owing to funding constraints, it was not possible to blind the researcher to the allocation of patients to the study groups, which may have introduced an element of bias.
A further limitation of the study is the small sample of general practices included in the study, but these were drawn from different areas of Belfast and included participants who resided in socioeconomically deprived areas, in which there is a need to expand knowledge regarding how to tackle the known health inequalities which exist with physical inactivity. However, this is a feasibility study and its findings support the development of a definitive trial of a primary care-based pedometer intervention.
The duration of follow-up was 12 weeks, which is typical of PA interventions, but there is relevance and importance in performing longer term assessments of outcomes within a definitive trial.
Conclusions
This study has provided information on the use of pedometers within primary care in promoting PA specifically within a socioeconomically deprived section of the population, using the theory of planned behaviour in PA promotion and using step-count targets. The study has shown that pedometers are acceptable tools for patients to promote PA, and the goal of increasing the number of steps taken per day by 2500 steps/day is achievable for inactive patients in general practice aged 35–75 years. However, it is important to help patients achieve this goal through self-determined step-count targets.
Patients attending GPs can be informed about their current PA level using the GPPAQ. Using the pedometer for 1 week provides a baseline daily step-count goal from which practitioners can advise an increase of 2500 steps/day, through the setting of achievable goals, reviewed regularly. If participants achieve this goal, they may expect reductions in BMI, weight and waist girth. Using GPPAQ to assess PA and offering a pedometer programme would allow the UK GPs to achieve the proposed 2012/13 Quality and Outcomes Framework indicators for PA.38 The provision of more financial assistance to healthcare providers to allow them to place greater emphasis on health promoting behaviour, including PA, should be a priority for the UK government as they attempt to realise the promised legacy of the 2012 London Olympics of increased PA levels among the general population. The current findings suggest that further research using pedometers is warranted, to help identify the most effective methods of promoting PA, especially in socioeconomically disadvantaged communities.
New findings
General Practice Physical Activity Questionnaire can be used within routine general practice to educate patients on their current level of physical activity (PA).
An increase of 2500 steps/day from baseline is a realistic goal for inactive individuals from general practice to use.
However, it is important to achieve this goal through self-determined targets negotiated with the individual.
The metal backing of the pedometer can promote a local reaction and discourage some individuals from using the pedometer, and thus strategies to reduce this must be considered when using pedometers.
How might it impact on clinical practice in the near future?
Pedometers appear to offer a realistic management option for use in primary care to help support patients in increasing their levels of physical activity.
A target of 2500 steps/day increase from baseline is achievable for inactive individuals from primary care through self-determined targets. If this target is achieved over a 12-week programme, the individual can expect improvements in their weight, body mass index and waist girth.
It is important to warn individuals of the risk of a local reaction to the pedometer metal backing when using them in clinical practice and for the individual to use preventative measures (eg, barrier cream) if they have a history of local reactions.
Acknowledgments
I would like to acknowledge the help of my coauthors, the four general practices which participated in the trial, all study participants and Queen's University in helping me carry out this piece of research.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online references
Footnotes
-
Correction notice This article has been corrected since it was published Online First. Figure 1 has been replaced with a correct version, and the redundant supplementary file (containing the references) has been removed.
-
Contributors The article has been planned, conducted and reported by NH with supervision, guidance and review at all stages by MT, MMK and MC. MC will be the guarantor of the overall content along with myself.
-
Competing interests None.
-
Patient consent Obtained.
-
Ethics approval Office for Research Ethics Committees Northern Ireland (ORECNI), reference number 11/NIR03/2, 15/03/2011.
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement There remains qualitative work from this piece of research which is currently unpublished. The hope is that the remaining data from the study will be published in due course.