www.elsevier.com/locate/ypmed

Preventive Medicine 4

Recommendations for and warnings against physical activity given to

older people by health care professionals

M. Hirvensalo Ph.D.a,b,*, E. Heikkinen M.D., Ph.D., M.Sc.b,c

T. Lintunen Ph.D.a, T. Rantanen Ph.D.b,c

aDepartment of Physical Education, University of Jyvaskyla, FinlandbThe Finnish Centre for Interdisciplinary Gerontology, University of Jyvaskyla, Finland

cDepartment of Health Sciences, University of Jyvaskyla, Finland

Available online 1 February 2005

Abstract

Background. Little is known about how health care professionals advice older people with chronic conditions about physical exercise.

This study investigated exercise counseling in the context of health care as perceived by older people, and factors associated with perceived

advice.

Design and methods. Participants were 580 non-institutional 73- to 92-year-old people who reported at least one contact with health care

during the previous 12 months.

Results. Of all the participants, 23% recalled solely recommendations to exercise, and 9% solely warnings against exercise. Additionally,

34% recalled receiving both recommendations for and warnings against physical activity, and 34% did not recall exercise-related advice at

all. Recalling solely recommendations to exercise was associated with having musculoskeletal diseases and impaired mobility. Reporting

solely warnings against physical activity was more common among those having heart conditions. Recalling both recommendations for and

warnings against exercise was associated with being physically active despite of having heart conditions, musculoskeletal diseases, and

impaired mobility. Recalling no exercise-related advice was most common among people who were sedentary and older, had fewer chronic

conditions and reported no mobility limitation.

Conclusions. A substantial proportion of older people recalled negative, no, or contradicting advice about exercise. As warnings against

physical activity may outweigh recommendations to exercise, special attention should be paid to the content of advice in order to avoid

discouraging older people from being active.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Aging; Counseling; Physical exercise; Health care

Introduction

Growth of the older population has led the health care

system and society in general to call for a search for new,

economically sustainable methods for increasing years of

healthy life and postponing loss of independence. In older

people, practically all of whom already have chronic

conditions, an important aim for health promotion is to

prevent or slow down the worsening of disability. Physical

0091-7435/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.ypmed.2004.11.020

* Corresponding author. Fax: +358 17 260 2101.

E-mail address: mirja.hirvensalo@sport.jyu.fi (M. Hirvensalo).

activity may be a potential means for preventing the

progression of disabilities [1–4].

Health care professionals have an opportunity to

influence older people’s physical activity [5]. For exam-

ple, the proportion of those who report advice or referrals

from health care professionals as a motive for participa-

tion in exercise classes, increases with age [6]. However,

according to patient reports, health care professionals

counsel only part of their customers on activity issues

[7–9].

Previous studies have mainly treated health care profes-

sionals’ exercise-related advice as if they were unidimen-

sional, i.e., as if they consisted only of positive messages to

1 (2005) 342–347

M. Hirvensalo et al. / Preventive Medicine 41 (2005) 342–347 343

increase activity while negative influences, such as

warnings against physical activity, have been ignored

[10]. The possibility of the multidimensional nature of

exercise counseling has not been addressed.

This article describes the content of advice on physical

activity given by health care professionals as perceived by

people aged 73 and over. The advice were categorized into

recommendations for and warnings against physical

activity. The second aim was to examine if symptoms,

chronic conditions, mobility status, and level of physical

activity were associated with perceived advice. The

potential confounders that were adjusted for in the logistic

regression models included age, gender, marital status,

education, and the number of contacts with health care

professionals.

Methods

This cross-sectional study uses data from a follow-up

interview of the Evergreen project, a population-based study

on health and functional capacity of residents in the City of

Jyv7skyl7 in Central Finland [11]. The follow-up interview

took place in 1996 when questions about health care

professionals’ advice on physical activity were added into

the study protocol. The 1224 baseline participants were first

studied in 1988. During the 8-year follow-up, 441 people

died and 23 moved away from the area. The follow-up

interview was conducted for 662 people, who at that time

were 73 to 92 years of age, and represented 87% of the

initial sample still alive and living in Jyv7skyl7. Respond-ents who reported having had a contact with health care

professionals within the previous 12 months (n = 580) and

who answered the questions on exercise advice were

eligible for this study.

Outcomes

Outcomes of the study were recalled advice on physical

activity. The recommendations for exercise were studied

by the following questions: bHave you been recommended

by health care professionals to practice calisthenics at

home, gym training, walking for fitness, other exercise, or

to participate in supervised exercise classes?Q The warn-

ings against physical activity was studied by querying:

bHave you been told by health care professionals to avoid

heavy exertion, walking long distances, exercise in cold

weather, exercise in hot weather, walking on slippery

roads, climbing stairs, climbing on a chair or ladder, or

other exercise?Q The formulation of the questions was

based on clinical experience of the research group

comprising of medical doctors, physiotherapists, nurses,

and physical exercise professionals. The participants were

categorized to separate groups based on whether they (1)

recalled solely recommendations for, (2) solely warnings

against, (3) both recommendations for and warnings

against physical activity, or (4) did not recall any

exercise-related advice.

Factors hypothesized to influence the advice given and

recalled included symptoms and chronic conditions, mobi-

lity status (impaired vs. intact), and physical activity level.

The potential confounders included age, gender, marital

status, education, and the number of contacts with health

care professionals during the previous 12 months. The sum

of all reported contacts with health care professionals

(physicians, physiotherapists, nurses, and others) in public

and private sector institutions was computed based on

responses to a given comprehensive list of health providers.

Chronic conditions and symptoms

Chronic conditions were queried by the following

question: bHave you been told by a doctor that you have

some chronic condition or disability?Q The respondents

listed freely the chronic conditions that they were experi-

encing. Chronic conditions included asthma, bronchitis or

emphysema, diabetes, cardiovascular diseases, such as

ischemic heart disease, coronary artery disease, hyper-

tension, heart insufficiency, cardiac infarction, arrhythmia,

intermittent claudication, musculoskeletal diseases such as

hip fracture and arthritis, neurological diseases such as

epilepsy and parkinsonism, stroke, and mental diseases.

Respondents were also asked about chest pain or pressure

and musculoskeletal pain they had experienced in a 2-week

period prior to the interview.

Mobility status

Mobility status was classified as intact for those who

reported being able to walk for 2 km and climb one flight of

stairs without difficulty. Those who reported difficulties in

one or both of these activities were classified as having

impaired mobility.

Physical activity

The level of physical activity and its intensity was

assessed using a six-point scale modified by the scale of

Grimby [12]: (1) moving only in necessary chores, (2)

walking or other outdoor activities 1–2 times/week, (3)

walking or other outdoor activities several times/week, (4)

1–2 times/week to the point of perspiring and heavy

breathing, (5) several times/week to the point of perspiring

and heavy breathing, (6) keep-fit exercise or competitive

sport several times a week. In addition, a list of various

sport activities was provided. Participants pointed to the

activities they did and indicated the frequency of doing the

sport. To be rated as active, subjects had to rate their overall

level of activity in categories 3 to 6, or report sport

activities such as swimming, cycling, dancing, skiing,

participating in supervised physical training, and other

activities, such as ball games or rowing, at least once a

M. Hirvensalo et al. / Preventive Medicine 41 (2005) 342–347344

week. If these criteria were not met, the participant was

rated as sedentary.

Statistical analysis

To consider the factors related to perceived advice,

differences in chronic conditions and symptoms, mobility

status, level of physical activity, age, gender, education,

marital status, and the number of contacts with health care

professionals were compared between the groups formed on

the basis of whether they recalled solely recommendations

for, solely warnings against, both recommendations for and

warnings against physical activity, or did not recall any

exercise-related advice at all. Cross-tabulation with Chi-

square test and one-way analysis of variance were used.

After that, four multiple logistic regression models were

computed to examine factors associated with perceived

advice. The reference groups were formed as follows to

avoid any overlap in the perceived advice between the

outcome group and the reference group: (1) recalling solely

recommendations vs. recalling solely warnings against

physical activity or no exercise-related advice, (2) recalling

solely warnings against physical activity vs. recalling solely

recommendations for physical activity or no exercise-related

advice, (3) recalling both recommendations for and warn-

ings against physical activity vs. not recalling exercise-

related advice, and (4) recalling no exercise-related advice

vs. others. All independent variables and confounders were

included in each model. The SPSS 9.0 software package

was used to carry out the analyses [13].

Table 1

Characteristics of participants grouped according to perceived exercise advice (sole

and warnings against physical activity, no exercise-related advice)

Solely recommendations Solely warnings

n = 134 n = 50

M SD M SD

Age 78.5 5.2 79.2 4.6

No. of contactsa 9.9 11.2 8.5 7.8

% %

Married 29 44

Male gender 36 30

Middle school or more 13 20

Heart conditions 52 72

Musculoskeletal diseases 41 40

Neurological conditions 3 2

Mental disease 2 2

Stroke 9 6

Asthma 7 10

Diabetes 8 2

Chest pain or pressure 4 12

Musculoskeletal pain 15 12

Mobility difficulties 57 46

Sedentary 23 20

Means and standard deviations (SD) followed by one-way ANOVA, and percenta The number of contacts with a health care professional (physicians, physiothera

Results

Of the 662 older subjects, 580 reported having had at

least one contact with a health care professional during the

last 12 months. The mean sum of contacts with physicians,

physiotherapists, nurses, and other health providers was 8.5

(SD 11.2). Of the participants, 23% recalled solely

recommendations for (n = 134) and 9% recalled solely

warnings against physical activity (n = 50), 34% recalled

both recommendations for and warnings against physical

activities (n = 198), while 34% did not recall being advised

on physical activity issues (n = 198). Those who recalled no

advice were older, more sedentary, had less heart conditions,

musculoskeletal diseases, and physical symptoms and had

had less contacts with health care professionals than those

who recalled advice (Table 1).

Table 2 presents all variables included in the multiple

logistic regression analyses in four separate models.

Among those reporting solely recommendations for phys-

ical activity, the odds for having musculoskeletal diseases

(OR 1.87, 95% CI 1.13–3.11) and mobility difficulties

(OR 1.98, 95% CI 1.09–3.58) was double compared to

those who reported solely warnings against physical

activity or did not recall any exercise-related advice.

Younger age was also associated with reporting solely

recommendations. Perceiving solely warnings against

physical activity was associated with having heart con-

ditions (OR 2.66, 95% CI 1.32–5.39) and being married

(OR 2.95, 95% CI 1.32–6.60). Perceiving both recom-

mendations for and warnings against physical activity was

ly recommendations for, solely warnings against, both recommendations for

Both recommendations and warnings No advice F

n = 198 n = 198

M SD M SD P value

79.0 4.9 80.2 5.0 b0.001

10.7 10.8 7.6 12.4 0.001

% % v2

P value

32 24 0.035

33 29 0.563

23 15 0.085

69 46 b0.001

44 22 b0.001

2 1 0.364

2 3 0.984

9 6 0.583

14 8 0.086

7 6 0.500

8 5 0.107

27 10 0.010

64 47 0.003

15 31 0.003

ages followed by Chi-square test (v2).

pists, nurses, and other health care professionals) during the last 12 months.

Table 2

Multivariate logistic regression models predicting perceived advice: solely recommendations for, solely warnings against, both recommendations for and

warnings against physical activity, no exercise-related advice

Solely

recommendationsaSolely warningsb

OR

Both recommendations

and warningsc OR

No adviced

OR

OR, 95% CI OR, 95% CI OR, 95% CI OR, 95% CI

Age 0.93, 0.88–0.97 1.01, 0.94–1.08 0.95, 0.88–1.03 1.07, 1.03–1.11

Married 0.85, 0.46–1.57 2.95, 1.32–6.60 2.20, 0.82–5.91 0.70, 0.42–1.18

Male gender 1.47, 0.82–2.63 0.51, 0.22–1.17 1.40, 0.53–3.69 0.85, 0.52–1.40

Middle school or more 0.98, 0.50–1.90 1.49, 0.65–3.38 3.50, 1.33–9.26 0.64, 0.38–1.10

Heart condition 1.12, 0.70–1.81 2.66, 1.32–5.39 3.43, 1.64–7.15 0.48, 0.32–0.71

Musculoskeletal disease 1.87, 1.13–3.11 1.77, 0.88–3.56 3.91, 1.76–8.69 0.47, 0.30–0.74

Neurological conditions 4.18, 0.58–30.3 1.36, 0.12–14.9 – 0.23, 0.03–2.11

Mental disease 0.74, 0.15–3.68 1.39, 0.15–13.1 0.59, 0.47–7.48 1.36, 0.36–5.13

Stroke 1.94, 0.75–5.00 1.08, 0.27–4.37 2.24, 0.25–20.4 0.50, 0.21–1.19

Asthma 0.55, 0.21–1.40 2.07, 0.68–6.44 2.47, 0.69–8.81 0.75, 0.38–1.49

Diabetes 1.54, 0.56–4.24 0.29, 0.03–2.38 2.18, 0.44–10.9 0.82, 0.35–1.94

Pressure/pain on the chest 0.41, 0.13–1.26 2.59, 0.81–8.33 0.31, 0.07–1.48 1.15, 0.49–2.73

Musculoskeletal pain 0.98, 0.47–2.06 0.74, 0.25–2.14 1.18, 0.38–3.25 0.73, 0.38–1.33

Mobility difficulties 1.98, 1.09–3.58 0.70, 0.31–1.58 2.32, 1.00–5.40 0.50, 0.31–0.82

Sedentary 0.67, 0.38–1.27 0.77, 0.32–1.86 0.25, 0.09–0.71 2.18, 1.33–3.57

No. of contacts with health

care professionals

1.01, 0.99–1.03 1.00, 0.98–1.03 1.06, 1.00–1.12 0.98, 0.96–1.00

Odds ratios, OR; 95% confidence interval, CI.a Recalled solely recommendations for physical activity (n = 134), vs. those who recalled solely warnings against physical activity or no exercise-related

advice (n = 248).b Recalled solely warnings against physical activity (n = 50), vs. those who recalled solely recommendations for physical activity or no exercise-related advice

(n = 332).c Recalled both recommendations for and warnings against physical activity (n = 198), vs. those who did not recall recommendations for or warnings against

physical activity (n = 198).d Did not recall advice on physical activity (n = 198), vs. those who recalled recommendations, warnings or both (n = 382).

M. Hirvensalo et al. / Preventive Medicine 41 (2005) 342–347 345

associated with having heart conditions (OR 3.43, 95% CI

1.64–7.15), musculoskeletal diseases (OR 3.91, 95% CI

1.76–8.69), and mobility difficulties (OR 2.32, 95% CI

1.00–5.40). Being sedentary decreased the probability of

recalling either recommendations for or warnings against

physical activity (OR 0.25, 95% CI 0.09–0.71). The

analyses also indicated that more educated participants

were more likely (OR 3.50, 95% CI 1.33–9.26) to report

both recommendations for and warnings against physical

activity. Recalling no advice was more common among

those who had fewer chronic conditions, intact mobility,

who were sedentary (OR 2.18, 95% CI 1.33–3.57) and

among older participants.

Discussion

Our study showed that majority (77%) of older people

recalled negative, no, or contradicting advice about exercise.

Our study provides new evidence about the perceived

exercise-related advice given to older people by health care

professionals. Previously, studies have mainly concentrated

on perceptions of positive, encouraging advice about

physical activity ignoring the potential restrictive aspects

of exercise-related advice. To the best of our knowledge,

this is the first study to capture the potentially conflicting

contents of exercise-related advice given as part of the

operation of the health care system. The simultaneous

conflicting advice including both recommendations to

exercise and warnings about the risks of exercise may

actually decrease physical activity rather than increase it

[10]. Warnings may cause older people to become over-

cautious even about doing moderate physical activities of

everyday life and eventually completely give up physical

activities. Warning against specific activities may be

interpreted as an instruction to be inactive or rest.

Our results indicate that also the social context of a

person may have an effect on the advice they recall. In the

current study, participants who were married were three

times more likely to report solely warnings against physical

activity than widows or unmarried people. It is possible that

worried spouses and younger family members reinforce the

negative advice received from the health care professionals,

for example, by taking over tasks in the house and then

justifying the situation by citing health care professionals’

advice. Constantly reminding older people about the

potential risks involved in the physical activities and the

unsuitability of the activity is detrimental to their exercise-

related self-efficacy and will likely result in a decrease in the

activity level [14–16].

A potential limitation of this study is that the results may

be subject to recall bias. First of all, some respondents may

have forgotten discussions with health care professionals on

exercise, even though they occurred. In the current study,

M. Hirvensalo et al. / Preventive Medicine 41 (2005) 342–347346

sedentary participants were more likely to not recall any

exercise-related advice than physically active participants.

Physically active people may be better able to associate

exercise information with their own previous experiences,

which results in a more accurate recall. Similarly, in the year

2000 supplemental survey of USA adult population, those

who reported the lowest activity level were three times less

likely to recall being counseled on exercise compared to

those reporting high level of activity [7]. Secondly, people

with chronic conditions and mobility difficulties had

probably had more contacts with health care professionals,

and therefore had more opportunities to receive advice.

However, the number of contacts with health professionals

within the last 12 months was taken into account in our

analyses. A third potential limitation is that information on

perceived advice was collected through interviewing the

subjects, not through direct observation of the situation.

Consequently, we do not know the actual content of the

advice. However, Stage et al. [17] compared direct observa-

tions of patients’ visits to information collected using

questionnaires and found that there was moderate to high

sensitivity for health habit counseling (kappa 0.42). Accord-

ing to Wee et al. [9], the patients’ recall accurately reflected

the quality of discussions and, in particular, the influence of

the counseling on the patient. Consequently, research on self-

reports of advice provides information highly relevant for

planning of experimental health promotion studies and health

promotion interventions.

Furthermore, the perceived advice in our study was in

line with existing recommendations indirectly supporting

the validity of our data. American College of Sports

Medicine [18], a gold standard of exercise recommendations

for health care providers, emphasizes the benefits of

endurance exercise in cardiac diseases while simultaneously

points out potential health risks related to exertion. This is in

line with our study, as those having heart conditions

typically reported both warnings against and recommenda-

tions for physical activity or solely warnings against

physical activity. For musculoskeletal diseases, ACSM

[18] recommends diversity of exercise including strength,

balance, and stretching exercises with less emphasis on risks

of exercise. Also in our study, recalling solely recommen-

dations to exercise was associated with musculoskeletal

diseases and mobility difficulties.

Currently, widely accepted recommendations about if

frail older people with comorbidity should be encouraged to

exercise and how it should be done do not exist, even

though the benefits from physical activities in terms of

preventing further disability are particularly evident among

them [2,19–24]. Recently, however, recommendations about

how to design randomized controlled trials on prevention of

progressive disability among frail older people have been

given [25]. The main idea is that exclusion criteria should be

minimized.

The current findings provide new perspectives on the

existing practices of exercise counseling within the context

of health care organizations as perceived by older people.

Previously, it has been shown that counseling by health care

providers has great potential to increase the level of physical

activity of sedentary patients in follow-ups of 6 weeks to 6

months duration [26–31]. Our results indicate that a large

proportion of older people may receive no, negative, or

conflicting advice about exercise in the context of health

care. This finding warrants further study using methods of

direct observation and also controlled trials. However, the

current finding also suggests that health care providers

should start paying attention to the content of their exercise

advice, in order to avoid discouraging older people from

being active.

References

[1] Evans WJ. Exercise training guidelines for the elderly. Med Sci Sports

Exerc 1999;31:12–7.

[2] Hirvensalo M, Rantanen T, Heikkinen E. Mobility difficulties and

physical activity as predictors of mortality and loss of independence in

the community living older population. JAGS 2000;48:493–8.

[3] Rantanen T, Guralnik JM, Sakari-Rantala R, Simonsick EM, Ling S,

Fried LP. Disability, physical activity and muscle strength in older

women: the Women’s Health and Aging Study. Arch Phys Med

Rehabil 1999;80:130–5.

[4] Leveille SG. Aging successfully until death in old age: opportunities for

increasing active life expectancy. Am J Epidemiol 1999;149:654–64.

[5] Booth ML, Bauman A, Owen N, Core CJ. Physical activity

preferences, preferred sources of assistance, and perceived barriers

to increased activity among physically inactive Australians. Prev Med

1997;26:131–7.

[6] Hirvensalo M, Lampinen P, Rantanen T. Physical exercise in old age:

An eight-year follow-up study on involvement, motives and obstacles

among persons age 65–84. J Aging Phys Act 1998;6:157–68.

[7] Damush TM, Stewart AL, Mills KM, King AB, Ritter PL. Prevalence

and correlates of physician recommendations to exercise among older

adults. J Gerontol Med Sci 1999;4A:423–7.

[8] Pender NJ, Sallis JF, Long BJ, Calfas KJ. Health-Care provider

counseling to promote physical activity. In: Dishman R.K., editor.

Advances in exercise adherence. Champaign, IL7 Human Kinetics;

1994. p. 213–35.

[9] Wee CC, McCarthy EP, Davis RB, Phillips RS. Physician counseling

about exercise. JAMA 1999;282:1583–8.

[10] Chogahara MA. Multidimensional scale for assessing positive and

negative social influences on physical activity in older adults.

J Gerontol Soc Sci 1999;54B:S356–67.

[11] Heikkinen E. Background, design and methods of the project. In:

Heikkinen E, Heikkinen R-L, Ruoppila I, editors. Functional capacity

and health of elderly people—The Evergreen project. Scand J Soc

Med Suppl, vol. 53, 1997. p. 1–18.

[12] Grimby G. Physical activity and muscle training in the elderly. Acta

Med Scand, Suppl 1986;711:233–7.

[13] Norusis MJ. SPSSR advanced statistics, vol. 6.1. Chicago, IL7 SPSS

Inc; 1994.

[14] Bandura A. Self-efficacy. The exercise of control. New York7 WH

Freeman and Co; 1997.

[15] Calfas KJ, Sallis JF, Oldenburg B, Ffrench M. Mediators of change in

physical activity following an intervention in primary care: PACE.

Prev Med 1997;26:297–304.

[16] King AC, Sallis JF, Dunn AL, Simons-Morton DG, Albright CA,

Cohen S, et al. Overview of the Activity Counseling Trial (ACT)

intervention for promoting physical activity in primary health care

settings. Med Sci Sports Exerc 1998;30:1086–96.

M. Hirvensalo et al. / Preventive Medicine 41 (2005) 342–347 347

[17] Stage KC, Zyzanski SJ, Smith TF, Kelly R, Langa DM, Flocke

SA, Jaen CR. How valid are medical records and patient question-

naires for physician profiling and health services research? A com-

parison with direct observations of patients visits. Med Care 1998;

36:851–67.

[18] American College of Sports Medicine Position Stand. Exercise and

physical activity for older adults. Med Sci Sports Exerc 1998;30:

992–1008.

[19] Fried LP, Guralnik JM. Disability in older adults: evidence regarding

significance, etiology and, risk. J Am Geriatr Soc 1997;45:92–100.

[20] Ettinger WH, Burns R, Messier SP, Applegate W, Rejeski J, Morgan

T, et al. A randomized trial comparing aerobic exercise and resistance

exercise with a health education program in older adults with knee

osteoarthritis. JAMA 1997;277:25–31.

[21] Fiatarone MA, O’Neill EO, Ryan ND, Clements KM, Solares GD,

Nelson ME, et al. Exercise training and nutritional supplementation

for physical frailty in very elderly people. N Eng J Med

1994;330:1769–75.

[22] LaCroix AZ, Guralnik JM, Berkman LF, Wallace RB, Satterfield S.

Maintaining mobility in late life. II. Smoking, alcohol consumption,

physical activity and body mass index. Am J Epidemiol 1993;137:

858–69.

[23] Jette A, Lachman M, Giorgetti MM, Assmann SF, Harris BA,

Levenson C, et al. Exercise—It’s never too late the strong-for-life

program. Am J Public Health 1999;89:66–72.

[24] Tinetti ME, Baker DI, McAvay G, Claus EB, Garrett P, Gottschalk

M, et al. A multifactorial intervention to reduce the risk of falling

among elderly people living in the community. N Engl J Med

1994;331:821–7.

[25] Ferrucci L, Guralnic JM, Studenski S, Fried LP, Cutler GB, Walston

JD. Designing randomised, controlled trials aimed at preventing or

delaying functional decline and disability in frail, older persons: a

consensus report. J Am Geriatr Soc 2004;52:625–34.

[26] Bull FC, Jamrozik K. Advise from a family physician can help

sedentary patients to become active. Am J Prev Med 1998;15:85–94.

[27] Calfas KJ, Long BJ, Sallis JF, Wooten WJ, Pratt M, Patrick K. A

controlled trial of physician counseling to promote the adoption of

physical activity. Prev Med 1996;25:225–33.

[28] Elley R, Kerse N, Arroll B, Robinson E. Effectiveness of counselling

patients on physical activity in general practice. Cluster randomised

controlled trial. BMJ 2003;326:793.

[29] Lewis BS, Lynch WD. The effect of physician advice on exercise

behavior. Prev Med 1993;22:110–21.

[30] Marcus BH, Goldstein MG, Jette A, Simkin-Silverman L, Pinto BM,

Milan F. Training physicians to conduct physical activity counselling.

Prev Med 1997;26:382–8.

[31] Swinburn BA, Walter LG, Arrol B, Tilyard MW, Russell DG. The

green prescription study: a randomized controlled trial of written

exercise advise provided by general practitioners. Am J Pub Health

1998;88:288–91.