8 The evidence

This section lists the evidence statements from the review provided by external contractors (see What evidence is the guidance based on?) and links them to the relevant recommendations. (See Summary of the methods used to develop this guidance for the key to quality assessments.)

This section also sets out a brief summary of findings from the economic analysis.

The evidence statements are short summaries of evidence in a review. Each statement has a short code indicating which document the evidence has come from. The letter(s) in the code refer to the type of document the statement is from, and the numbers refer to the document number, and the number of the evidence statement in the document.

Evidence statement number PA8 indicates that the linked statement is numbered 8 in the review 'Physical activity: brief advice for adults in primary care'.

The review and economic analysis are available.

Where a recommendation is not directly taken from the evidence statements, but is inferred from the evidence, this is indicated by IDE (inference derived from the evidence).

Recommendation 1: evidence statements PA8, PA9, PA12, PA16, PA20; IDE

Recommendation 2: evidence statements PA1, PA8, PA9, PA16, PA18, PA19, PA20; IDE

Recommendation 3: evidence statements PA12, PA15, PA16, PA23, PA25, PA30; IDE

Recommendation 4: evidence statements PA11, PA16, PA23, PA27, PA30; IDE

Recommendation 5: evidence statements PA8, PA9, PA10, PA12, PA13, PA15, PA23, PA26, PA28, PA29, PA30

Evidence statements

Please note that the wording of some evidence statements has been altered slightly from those in the evidence review(s) to make them more consistent with each other and NICE's standard house style.

Evidence statement PA1

Moderate evidence from 15 studies; 4 nRCTs (4 [−]2,3,14,15), 4 cluster RCTs (2[++]4,5, 1 [+]6 and 1 [-]7) and 7 RCTs (1 [++]8 4 [+]1,10,11,12 , 2 [−]9,13) suggests that there is an increase in the self-reported physical activity levels in those participants who received brief advice, or who were seen by primary care professionals trained to deliver brief advice.

1 Bull et al. 1998 ([+] Australia)

2 Calfas et al. 1996 ([−] USA)

3 Marcus et al. 1997 ([−USA)

4 Elley et al. 2003 ([++] New Zealand)

5 Grandes et al. 2009 ([++] Spain)

6 Goldstein et al. 1999 ([+] USA)

7 Marshall et al. 2005 ([−] Australia)

8 Petrella et al. 2003 ([++] Canada)

9 Hillsdon et al. 2002 ([−] UK)

10 Harland et al. 1999 ([+] UK)

11 Halbert et al. 2000 ([+] Australia)

12 Little et al. 2004 ([+] UK)

13 Lewis et al. 1993 ([−] USA)

14 Smith et al. 2000 ([−] Australia)

15 Naylor et al. 1990 ([−] UK)

Evidence statement PA8

Moderate evidence from 5 studies; 2 qualitative (1 [++]1 and 1 [+]2) and 3 quantitative studies (all [+]3,4,5), suggests that perceived patient characteristics affect a practitioner's decision to discuss and/or prescribe physical activity.

1 Ampt et al. 2009 ([++] Australia)

2 Melillio et al. 2000 ([+] USA)

3 Booth et al. 2006 ([+] Australia)

4 Kreutzer et al. 1997 ([+] USA)

5 Royals et al. 1996 ([+] USA)

Evidence statement PA9

Moderate evidence from 18 studies; 8 qualitative (3 [++]1,7,8, 4 [+]11,12,17,18 and 1 [−]8) and 10 quantitative studies (all [+]2,3,4,5,6,10,13,14,15,16) suggests that perceived likely uptake of advice, motivation to change, and receptiveness affects a practitioner's decision to discuss and/or prescribe physical activity. Practitioners are more likely to provide brief physical activity advice to patients who they perceive are most likely to act on the advice given.

1 Ampt et al. 2009 ([++] Australia)

2 Bize et al. 2007 ([+]Switzerland)

3 Bull et al. 1995 ([+] Australia)

4 Bull et al. 1997([+] Australia)

5 Buchholz et al. 2007 ([+] USA)

6 Burns et al. 2000 ([+] USA)

7 Douglas et al. 2006a ([++] UK)

8 Douglas et al. 2006b (([++] UK)

9 Gould et al. 1995 ([−] UK)

10 Heintze et al. 2010 ([+] Germany)

11 Horsley Tompkins et al. 2009 ([+] USA)

12 Huang et al. 2004 ([+] USA)

13 Kennedy et al. 2003 ([+] Canada)

14 Kreuter et al. 1997 ([+] USA)

15 Lawlor et al. 1999 ([+]UK)

16 Long et al. 1996 ([+] USA)

17 Walsh et al. 1999 ([+], USA)

18 Winzenberg et al. 2009 ([+] Australia)

Evidence statement PA10

Moderate evidence from 8 studies; 5 qualitative (1 [++]2, 3 [+]4,7,8 and 1 [−]6) and 3 quantitative studies (all [+]1,3,5) suggests that practitioner behaviour is influenced by perceived evidence for effectiveness of physical activity advice as well as the perceived effectiveness of physical activity to improve health. Practitioners who believe that physical activity improves health are more likely to deliver brief physical activity advice.

1 Bull et al. 1995 ([+] Australia)

2 Douglas et al. 2006a ([++] UK)

3 Horsley Tompkins et al. 2009 ([+] USA)

4 Huang et al. 2004 ([+] USA)

5 Kennedy et al. 2003 ([+] Canada)

6 Swinburn et al. 1997 ([−] New Zealand)

7 Ribera et al. 2005 ([+] Spain)

8 Winzenberg et al. 2009 ([+] Australia)

Evidence statement PA11

Moderate evidence from 12 studies: 7 qualitative (3 [++]1,6,7, and 4 [+]2,8,11,12) and 5 quantitative studies (all [+]3,4,5,9,10) suggests that practitioners consider a lack of provision of print materials, incentives, or other support resources to be a barrier to discussing and/or prescribing physical activity. It may be that better provision of print materials to hand out to patients, financial reward for providing brief physical activity advice or addition provision of other support recourses would increase the delivery of brief physical activity advice.

1 Ampt et al. 2009 ([++] Australia)

2 Bize et al. 2007 ([+] Switzerland)

3 Bull et al. 1995 ([+] Australia)

4 Bull et al. 1997([+] Australia)

5 Burns et al. 2000 )[+] USA)

6 Douglas et al. 2006a ([++] UK)

7 Douglas et al. 2006b ([++] UK)

8 Huang et al. 2004 ([+] USA)

9 Long et al. 1996 ([+] USA)

10 McDowell et al. 1997 ([+] UK)

11 Pinto et al. 1998 ([+] UK)

12 Ribera et al. 2005 ([+] Spain)

Evidence statement PA12

Moderate evidence from 19 papers; 9 qualitative (2 [++]7,8 [+]2,9,14,15,16,19 and 1 [−]17), 9 quantitative studies (all [+]1,3,4,5,6,10,11,12,13), and 1 mixed methods evaluation [+]18 suggests that practitioners considered that time resources and conflicting priorities affected their ability to discuss and/or prescribe physical activity. Time acts as a 'proxy' for related factors such as increased workload, resulting in conflicting priorities and a need to choose between physical activity promotion and other factors which may be seen as more central to the practitioner role.

1 Albright et al. 2000 ([+] USA)

2 Bize et al. 2007 ([+]Switzerland)

3 Bull et al. 1995 ([+] Australia)

4 Bull et al. 2010 ([+] UK)

5 Buchholz et al. 2007 ([+] USA)

6 Burns et al. 2000 ([+] USA)

7 Douglas et al. 2006a ([++] UK)

8 Douglas et al. 2006b ([++] UK)

9 Huang et al. 2004 ([+] USA)

10 Kennedy et al. 2003 ([+] Canada)

11 Lawlor et al. 1999 ([+] UK)

12 Long et al. 1996 ([+] USA)

13 McKenna et al. 1998 ([+] UK)

14 Melillo et al. 2000 ([+] USA)

15 Patel et al. 2011 ([+] UK)

16 Ribera et al. 2005 ([+] Spain)

17 Swinburn et al. 1997 ([−] New Zealand)

18 Van Sluijs et al. 2004 ([+] Netherlands)

19 Winzenberg et al. 2009 ([+] Australia)

Evidence statement PA13

Moderate evidence from 18 studies; 9 qualitative (one [++]1, 7 [+]7,8,9,12,14,15,16 and 1 [−]3) and 9 quantitative studies (all [+]2,3,4,5,6,10,12,17,18) suggests that practitioner confidence and knowledge (including the need for further training/support) affected their ability to discuss and/or prescribe physical activity. Greater practitioner confidence/knowledge (created through better training) increases the likelihood of delivery brief advice.

1 Ampt et al. 2009 ([++] Australia)

2 Buchholz et al. 2007 ([+] USA)

3 Buffart et al. 2012 ( [+] Australia)

4 Bull et al. 1995 ([+] Australia)

5 Bull et al.1997([+] Australia)

6 Burns et al. 2000 ([+]USA)

7 Douglas et al. 2006a ([++] UK)

8 Douglas et al. 2006b ([++] UK)

9 Eadie et al. 1996 ([+], Qualitative, UK)

10 Gould et al. 1995 ([−] UK)

11 Gribben et al. 2000 ([+] New Zealand)

12 Huang et al. 2004 ([+] USA)

13 Kennedy et al. 2003 ([+] Canada)

14 Pinto et al. 1998 ([+] UK)

15 Ribera et al. 2005 ([+] Spain)

16 Sims et al. 2004 [+] (Australia)

17 Van der Ploeg et al. 2007([+] Australia)

18 Walsh et al. 1999 ([+] USA)

Evidence statement PA15

Moderate evidence from 6 studies; 2 qualitative (all [++]4,5) and 4 quantitative studies (all [+]1,2,3,6), suggests that practitioner willingness to discuss and/or prescribe physical activity was influenced by whether they perceived this activity to be within their remit/role. Those who saw physical activity promotion as within their role were more likely to provide brief physical activity advice.

1 Booth et al. 2006 ([+] Australia)

2 Buffart et al. 2012 ([+] Australia)

3 Bull et al. 1995 ([+] Australia)

4 Douglas et al. 2006a ([++] UK)

5 Douglas et al. 2006b ([++] UK)

6 Van der Ploeg et al. 2007 ([+] Australia)

Evidence statement PA16

Moderate evidence from 18 studies; 11 qualitative (3 [++]1,4,5 6 [+]2,11,13,14,15,18 and 2 [−]6,17) and 7 quantitative studies (all [+]3,7,8,9,10,12,16), suggests that practitioners were more willing to discuss and/or prescribed physical activity where this was linked to the presenting condition (rather than as a preventative measure), that is to provide curative rather than preventative advice.

1 Ampt et al. 2009 ([++] Australia)

2 Bize et al. 2007 ([+] Qualitative, Switzerland)

3 Bull et al. 1995 ([+] Australia)

4 Douglas et al. 2006a ([++] UK)

5 Douglas et al. 2006b ([++] UK)

6 Gould et al. 1995 ([−] UK)

7 Gribben et al. 2000 ([+] New Zealand)

8 Horsley Tompkins et al. 2009 ([+] USA)

9 Kreuter et al. 1997 ([+] USA)

10 Lawlor et al. 1999 ([+]UK)

11 Leijon et al. 2010 ([+] Sweden)

12 McDowell et al. 1997 ([+] UK)

13 Melillo et al. 2000 ([+] USA)

14 Patel et al. 2011 ([+] UK)

15 Ribera et al. 2005 ([+] Spain)

16 Schmid et al. 2009 ([+] Switzerland)

17 Swinburn et al. 1997 ([−] New Zealand)

18 Winzenberg et al. 2009 ([+] Australia)

Evidence statement PA18

Moderate evidence from 4 qualitative studies (all [+]1,2,3,4) suggests that patient willingness to comply with brief physical activity advice is affected by their recall and understanding of advice. Patients who understand the advice are more likely to comply with it.

1 Huang et al. 2004 ([+] USA)

2 Ribera et al. 2006 ([+] Spain)

3 Pinto et al. 1998 ([+] UK)

4 Sims et al. 2004 ([+] Australia)

Evidence statement PA19

Moderate evidence from 1 qualitative study (all [+]1), suggests that patients felt they needed to receive more preventative advice (that is, advice not linked to a presenting condition).

1 Horne et al. 2010 ([+] UK)

Evidence statement PA20

Moderate evidence from 2 qualitative studies (all [+]1,2) suggests that patients were less receptive to brief physical activity advice if they were unaware of physical activity recommendations. Making patients aware of physical activity recommendations would increase their willingness to comply with brief physical activity advice.

1 Horne et al. 2010 ([+] UK)

2 Sims et al. 2004 ([+] Australia)

Evidence statement PA23

Moderate evidence from 10 studies; 5 qualitative (4 [+]3,5,7,8, and 1 [−]9), 3 quantitative (all [+]1,2,4), and 2 mixed methods studies (all [+]6,10), suggests that interventions to encourage practitioners to administer brief physical activity advice can be effective in improving practitioners' views of brief physical activity advice, which may lead to positive effects on patients' physical activity behaviours.

1 Albright et al. 2000 ([+] USA)

2 Booth et al. 2006 ([+] Australia)

3 Bull et al. 2010 ([+] UK)

4 Gribben et al. 2000 ([+] New Zealand)

5 Leijon et al. 2010 ([+] Sweden)

6 Long et al. 1996 ([+] USA)

7 Patel et al. 2011 ([+] UK)

8 Pinto et al. 1998 ([+] UK)

9 Swinburn et al. 1997 ([−] New Zealand)

10 Van Sluijs et al. 2004 ([+] Netherlands

Evidence statement PA25

Moderate evidence from 14 studies; 7 effectiveness studies (2 [++]1,3 3 [+]4,9,13 and 2 [−])10,12, and 7 barriers and facilitators studies (1 [++]7, 5 [+]2,5,6,11,14 and 1 [‑]8), suggests that the provision of incentives to encourage practitioners to administer brief physical activity advice or provision of incentives to patients to encourage them to act on brief physical activity advice may overcome barriers to delivery/uptake but this cannot be validated through the effectiveness evidence.

1 ACT 2001 ([++] Australia)

2 Bize et al. 2007 ([+] Switzerland)

3 Bolognesi et al. 2006 ([++] Italy

4 Bull et al. 1998 ([+] Australia)

5 Bull et al. 1995 ([+] Australia)

6 Burns et al. 2000 ([+]USA)

7 Douglas et al. 2006a ([++] UK)

8 Gould et al. 1995 ([−] UK)

9 Harland et al. 1999 ([+] UK)

10 Lewis 1993 ([−] USA)

11 McDowell et al. 1997 ([+] UK)

12 Naylor et al. 1999 ([− UK)

13 Pinto et al. 2005 ([+] USA)

14 Ribera et al. 2005 ([+] Spain)

Evidence statement PA26

Moderate evidence from 23 studies; 9 effectiveness studies (5 [++]2,9,12,19,20 2 [+]4,10, and 2 [−]16,17), and 14 barriers and facilitators studies (1[++]1, and 13 [+]3,5,6,7,8,11,13,14,15,16,21,22,23) suggests that the provision of training may encourage practitioners to administer brief physical activity advice and that the education of patients may encourage them to act on brief physical advice. In particular this may be effective in improving intervention outcomes in populations where this knowledge is found to be lacking.

1 Ampt et al. 2009 ([++] Australia)

2 Bolognesi et al. 2006 ([++] Italy)

3 Buchholz et al. 2007 ([+] USA)

4 Bull et al.1998 ([+] Australia)

5 Burns et al.2000 ([+]USA)

6 Douglas et al. 2006a ([++] UK)

7 Douglas et al. 2006b ([++] UK)

8 Eadie et al.1996 ([+] UK)

9 Elley et al. 2003 ([++] New Zealand)

10 Goldstein et al. 1999 ([+] USA)

11 Goodman et al. 2011 ([+] UK)

12 Grandes et al. 2009 ([++] Spain)

13 Horne et al. 2010 ([+] UK)

14 Huang et al. 2004 ([+] USA)

15 Kennedy et al. 2003 ([+] Canada)

16 Lewis et al. 1993 ([−] USA)

17 Marcus et al. 1997 ([−] USA)

18 McDowell et al. 1997 ([+] UK)

19 Petrella et al. 2003 ([++] Canada)

20 Pinto et al. 2005 ([+] USA)

21 Ribera et al. 2006 ([+] Spain)

22 Sims 2004 ([+] Australia)

23 Walsh et al. 1999 ([+] USA)

Evidence statement PA27

Moderate evidence from 22 studies; 11 effectiveness studies (3 [++]1,9,10,4 [+]4,18,21,22 and 4 [−]12,14,15,17), and 11 barriers and facilitators studies (3 [++]2,7,8 and 8 [+]3,5,6,11,13,16,19,20), suggests no benefit from the addition of written support materials to a brief advice intervention. However it may be that the quality of currently available materials needs to improve to see an effect.

1 ACT 2001 ([++] Australia)

2 Ampt et al. 2009 ([++] Australia)

3 Bize et al. 2007 ([+]Switzerland)

4 Bull et al. 1998 ([+] Australia)

5 Bull et al. 1995 ([+] Australia)

6 Burns et al. 2000 ([+]USA)

7 Douglas et al. 2006a ([++] UK)

8 Douglas et al. 2006b ([++] UK)

9 Elley et al. 2003 ([++] New Zealand)

10 Grandes et al. 2009 ([++] Spain)

11 Huang et al. 2004 ([+] USA)

12 Little et al. 2004 ([-] UK)

13 Long et al. 1996 ([+] USA)

14 Marcus et al. 1997 ([−] USA)

15 Marshall et al. 2005 ([−] Australia)

16 McDowell et al. 1997 ([+] UK)

17 Naylor 1999 ([−] UK)

18 Pfeiffer et al. 2001([+] USA)

19 Pinto et al. 1998 ([+] UK)

20 Ribera et al. 2005 ([+] Spain)

21 Smith et al. 2000 ([+] Australia)

22 Swinburn et al. 1998 ([+] New Zealand)

Evidence statement PA28

Moderate evidence from 18 studies; 9 effectiveness studies (2 [++]6,9, 4 [+]3,8,10,11 and 3 [-]5,13,15), and 9 barriers and facilitators studies (8[+]1,2,4,7,12,14,16,17 and 1 [−]18), suggests that whilst the evidence of relative effectiveness for brief interventions of 5 minutes or longer versus interventions of very short duration (less than 5 minutes) is inconclusive, structured interventions can help to overcome practitioner barriers to prescribing brief advice.

1 Albright et al. 2000 ([+] (USA)

2 Booth et al. 2006 ([+] Australia)

3 Bull et al. 1998 ([+] Australia)

4 Bull et al. 2010 ([+] UK)

5 Calfas et al. 1996 ([−] USA)

6 Elley et al. 2003 ([++] New Zealand)

7 Gribben et al. 2000 ([+] New Zealand)

8 Goldstein et al. 1999 ([+] USA)

9 Grandes et al. 2009 ([++] Spain)

10 Halbert et al. 2000 ([+] Australia)

11 Hillsdon et al. 2002 ([+] UK)

12 Leijon et al. 2010 ([+] Sweden)

13 Lewis et al. 1993 ([−] USA)

14.Long et al. 1996 ([+] USA)

15 Marcus et al. 1997 ([−] USA)

16 Patel et al. 2011 ([+] UK)

17 Pinto et al. 1998 ([+] UK)

18 Swinburn et al. 1997 ([−] New Zealand)

Evidence statement PA29

Moderate evidence from 7 barriers and facilitators studies (2 [++]1,2, 4 [+]3,4,5,6, and 1 [−]7), suggests that time constraints resulted from conflicting priorities, and unfavourable working conditions. It seems likely that practitioners report lack of time as a proxy for a wide range of barriers to delivering brief physical activity advice and that overcoming problems such as lack of training, knowledge and confidence could act to remove the perceived barrier of lack of time.

1 Douglas et al. 2006a ([++] UK)

2 Douglas et al. 2006b ([++] UK)

3 Huang et al. 2004 ([+] USA)

4 McKenna et al. 1998 ([+] UK)

5 Patel et al. 2011 ([+] UK)

6 Ribera et al. 2005 ([+] Spain)

7 Swinburn et al. 1997 ([−] New Zealand)

Evidence statement PA30

Moderate evidence from 1 effectiveness ([−]6), and 8 barriers and facilitators studies (1 [++]2, and 7 [+]1,3,4,5,6,7,8), suggests that the structure of the actual 'system' in which the intervention is delivered has the potential to affect both the effectiveness of the intervention and its acceptability to both patients and practitioners. It is important to note that all the structural factors outlined here need to be considered together rather than in isolation to facilitate positive changes in intervention delivery and physical activity uptake.

1 Bize et al. 2007 ([+] Switzerland)

2 Douglas et al. 2006b ([++] UK)

3 Gribben et al. 2000 ([+] New Zealand)

4 Leijon et al. 2010 ([+] Sweden)

5 Long et al. 1996 ([+] USA)

6 Marcus et al. 1997 ([−] USA)

7 McDowell et al. 1997 ([+] UK)

8 Pinto et al. 1998 ([+] UK)

9 Walsh et al. 1999 ([+] USA)

Cost effectiveness

There was a review of economic evaluations, a review of economic barriers and facilitators and an economic modeling exercise.

Review of economic evaluations

Three papers were reviewed, 2 of which were based on a UK and an Australian population. The only overlap with previous economic literature influencing public health guidance in this area was the cost-effectiveness model developed for the previous NICE guidance (Matrix 2006).

Moderate, but limited evidence from 3 studies suggested that brief advice on physical activity in primary care is more cost effective than usual care. The evidence should be interpreted with caution as the evidence based on effectiveness was weak and did not fully explore uncertainty. Therefore, a de novo modelling of the cost effectiveness of brief advice was needed to improve knowledge on its efficiency.

Review of economic barriers and facilitators

Six papers were reviewed: 5 quantitative studies from the USA and 1 qualitative study from New Zealand.

Poor quality evidence suggested that a perceived lack of adequate financial incentive for healthcare professionals is a barrier to the delivery of brief advice on physical activity in primary care. This was irrespective of whether the advice was provided by a nurse or GP.

Moderate evidence suggested a weakly positive correlation between the time spent on (or available for) counselling and the delivery of brief advice on physical activity in primary care, regardless of whether it was provided by a GP or nurse. There was no interpretable policy-relevant evidence on the role of remuneration in the delivery of brief advice on physical activity. There was no interpretable evidence on the role of other resources in the delivery of brief advice on physical activity.

Economic modelling

A number of assumptions were made which could underestimate or overestimate the cost effectiveness of the interventions (see economic modelling report for further details).

The analysis adopted a lifetime horizon, an NHS/Personal Social Service perspective and discounted quality-adjusted life years (QALY) as a key outcome. Uncertainty over the model results was estimated by deterministic sensitivity analysis, scenario analysis, and probabilistic sensitivity analysis. Deterministic analysis was used to estimate the impact of alternative model scenarios. Compared with usual care, the incremental cost-effectiveness ratio (ICER) of brief advice was £1730 and thus can be considered cost effective.

When brief advice was compared with usual care (the 'base case'), uncertainties were explored through a series of analyses. In most cases the base case results were robust, but they were sensitive to the duration of protective effects of physical activity, mental health gains from physical activity, changes in infrastructure and age of cohort.

The impact of changing the age at which physical activity started, post-brief advice, to 54 years and older resulted in brief advice being cheaper and more effective compared with usual care (this is termed as brief advice 'dominating' usual care). Thus, the strength of the cost-effectiveness results was even greater for people aged 54 years and older. Uncertainty over the model results was estimated by probabilistic sensitivity analysis. Probabilistic sensitivity analysis showed a 99.9% chance that brief advice will be cost effective if an additional QALY is valued at £20,000.

While the economic model was based on the previous economic model (Matrix 2006) used to support developing NICE public health guidance 2, this model offered a number of improvements including:

1) time-based modelling

2) mental health and wellbeing as well as infrastructure (considered where permitted by the evidence)

3) more extensive exploration of uncertainty around the ICERs

4) more conservative assumptions around changes in physical activity over time

5) use of meta-analysed effectiveness data.

Full modelling report and results are reported in the economic modelling report.

Overall, brief physical activity advice in primary was found to be cost effective.

  • National Institute for Health and Care Excellence (NICE)