NZMA Home

Table of contents
Current issue
Search journal
Archived issues
NZMJ Obituaries
Classifieds
Hotline (free ads)
How to subscribe
How to contribute
How to advertise
Contact Us
Copyright
Other journals
The New Zealand Medical Journal

 Journal of the New Zealand Medical Association, 25-January-2008, Vol 121 No 1268

A single question reliably identifies physically inactive women in primary care
Sally B Rose, C Raina Elley, Beverley A Lawton, Anthony C Dowell
Abstract
Aim To validate a single-item screening question for systematic use in primary health care to identify physically inactive adults, who may benefit from physical activity intervention.
Methods The single-item physical activity screening question was administered to 1171 women aged 51–74 years recruited from 10 general practices, followed by a longer validated physical activity questionnaire (the NZPAQ-LF). Sensitivity, specificity, likelihood ratios, positive and negative predictive values, and a Kappa statistic were calculated to assess validity of the screening question.
Results The sensitivity of the single-item question was 76.7% (95% confidence interval [CI] 73.5–79.7). It had high specificity (81.1%, 95%CI 77.2–84.4), and a high positive predictive value (86.7%, 95%CI 83.8–89.1). The positive likelihood ratio was 4.05 (3.33–4.93), and negative likelihood ratio was 0.29 (0.25– 0.33). The Kappa statistic calculated for the single-item screening question when validated against the NZPAQ-LF was 0.56 (p<0.001).
Conclusions The single-item screening question has good sensitivity, specificity, and concordance with a validated physical activity questionnaire. The question is easy to administer and elicits a simple yes/no response from patients. This validated tool can now be used in practice to identify women who would benefit from physical activity interventions in primary care.

Physical inactivity is a predictor of high existing morbidity and is an independent risk factor for type 2 diabetes and cardiovascular disease.1 In 1996, the US Surgeon General recommended a minimum of 30 minutes of moderate intensity physical activity on most days of the week for health benefits.1 However, much of the evidence for health benefit relates to the accumulation of 150 minutes per week of moderate activity rather than 30 minutes on 5 days of the week.2
General practice-based interventions such as exercise on prescription programs that involve physical activity counselling have been shown to be effective at increasing physical activity levels for individual health benefit.3–5 The health benefits of increasing physical activity are greatest amongst individuals with high cardiovascular risk,6,7 and those who are least active and least physically fit.8 Ninety-two percent of physically inactive adults identified in general practice also have at least one pre-existing cardiovascular risk factor.9 Therefore, screening for physical inactivity in everyday general practice would identify individuals who would benefit most from physical activity interventions.9
Using validated physical activity questionnaires for the assessment of physical activity levels may be the most appropriate method for large-scale screening or epidemiological studies, but administering such questionnaires in a busy general practice is not practical or economical. Several validation studies of shorter physical activity screening tools have been reported, but no single-item question that is sensitive and specific for physical inactivity has been validated for use with adults within a general practice setting.
Previous studies of physical inactivity screening measures have been conducted in settings other than general practice, including household surveys,10 community and workplace settings,11–13 via telephone or mailouts,14,15 and most included more than one question.10–12,16,17 Those that described single-item questions have limitations in that they did not assess the single-item question against another measure of physical activity,13 did so against a physical activity measure that had not been validated,11 did not report sensitivity and specificity,14 or reported a low level of agreement with the validated measure of physical activity.15
A two-question tool validated against objective measures of physical activity in 20–60 year old adults attending general practice had good validity,16 but required the administrator to assign scores to each question, then combine scores to determine whether individuals fell into ‘sufficiently active’ or ‘sufficiently inactive’ categories. We argue that this tool would be too complex for use in a busy clinical practice setting.
This study uses the single-item screening question “As a rule, do you do at least half an hour of moderate or vigorous exercise (such as walking or a sport) on five or more days of the week?” which has been used in two studies to screen for physical inactivity in New Zealand general practice settings but has not yet been validated.3,18 This paper describes the validation of this single-item physical activity screening question against a validated long physical activity questionnaire.19

Methods

Design—The criterion-related validity of the single-item screening question to identify physically inactive women in general practice—As a rule, do you do at least half an hour of moderate or vigorous exercise (such as walking or a sport) on five or more days of the week?—was assessed using the validated New Zealand physical activity questionnaire long form (NZPAQ-LF)19 as a reference standard.
Participants—Participants included women aged 51–74 years drawn from a previous observational study cohort that had been recruited between 1999 and 2002 from 10 general practices in the greater Wellington area.20 The response rate for the original cohort was 55%, and exclusion criteria included severe, debilitating or terminal clinical conditions such as breast cancer, schizophrenia, or impaired mobility. Data presented here were collected during the recruitment phase of a randomised controlled trial (the ‘Women's Lifestyle Study).21 Recruitment for this study was carried out between February and July 2005 by letter of invitation to all women in the Wellington cohort, followed by a phone call from a research nurse. The Central Regional Ethics Committee approved the study in September 2004 (Ref WGT/04/08/061).
Data collection—Data collection took place at face-to-face interviews conducted by six primary care research nurses in community healthcare settings. Research nurses took part in a comprehensive training session on the administration of the questionnaire prior to the interviews. All participants answered the single-item screening question first, followed by the NZPAQ-LF later in the interview. The research nurses were aware of the results of the screening question when administering the NZPAQ-LF.
Outcome measures—The single-item physical activity question was coded as a categorical yes/no variable. The NZPAQ-LF questionnaire has previously been validated against both an objective measure of activity (heart rate monitoring (HRM) and sub-maximal VO2), self-reported physical activity logs, and the validated international physical activity questionnaire (IPAQ-Long).22 The validation study showed that the NZPAQ-LF tended to overestimate physical activity levels compared to HRM, but more accurately reflected activity levels than the IPAQ-Long.22
The NZPAQ-LF asks participants about physical activity carried out in the last 7 days in relation to activity type, context, intensity and duration. Show cards with grouped activities performed in different contexts (sport and physical recreation, transport, occupation, cultural/other/incidental) are presented to the participant to assist with recall of activities performed in the previous week. Definitions of moderate and vigorous intensity are provided for the participant to categorise their own activity intensity.
Activities performed in the context of sport and physical recreation and transport recorded on the NZPAQ-LF were collated to determine the total number of minutes spent being physically active in the past seven days, as well as the number of days on which at least 30 minutes of moderate or vigorous intensity activity was achieved. The duration of activities recorded as vigorous-intensity was multiplied by two to achieve moderate-intensity equivalents in line with the NZPAQ-LF administration guidelines.19
Statistical analyses—To assess validity, the sensitivity, specificity, positive likelihood ratio and positive predictive value of the screening question were determined using the NZPAQ-LF as the reference standard test. An online statistical calculator on the Evidence Based Medicine University of Toronto Website (http://www.cebm.utoronto.ca/) was used. Kappa statistics were calculated using the SPSS (v11.5) statistical program.

Results

Letters of invitation were sent to 2021 women, of whom 230 women were no longer resident at the recorded address and could not be contacted. Replies were received from 94.6% of those contacted (1695/1791). Of those, 69.6% were willing to participate and attended an interview with a general practice research nurse (1179/1695). Of those interviewed, 1171 were eligible for participation in the validation study; 8 were not eligible as they responded to the single-item question over the telephone rather than in a face-to-face setting. Characteristics of the participants are presented in Table 1.

Table 1. Characteristics of the female participants aged 51–74 years (N=1171)

Characteristics
n
%
Mean age in years
62.5 (SD 5.35)
Tertiary education
747
63.8
New Zealand European ethnicity
1031
88
Māori ethnicity
33
2.8
Married/permanent partner
829
70.8
Current smoker
89
7.6
Diagnosed diabetes
67
5.7
Previous cardiovascular disease*
83
7.1
Obese (BMI≥30)
340
29
*Includes diagnosis of myocardial infarction, angina, coronary artery bypass graft, angioplasty, or stroke; BMI=Body mass index (weight in kilograms divided by height in metres squared).

When asked the single-item physical activity screening question, 45% (532/1171) of participants said that they do at least 30 minutes of moderate or vigorous activity such as walking or sport on 5 or more days of the week.
Data collected from the NZPAQ-LF showed that 38.3% (449/1171) had achieved at least 150 minutes of moderate intensity activity in the past 7 days, and 21.3% (249/1171) had achieved at least 30 minutes on 5 or more of the past 7 days. The average number of minutes of physical activity reported was 150.6 (SD 187.2).
Table 2 presents results of those classified as ‘inactive’ according to the NZPAQ-LF, presented as those achieving less than 150 minutes per week, as well as those achieving less than 30 minutes five times per week as measured by the longer questionnaire, to reflect the two definitions of physical inactivity commonly used in the literature. The prevalence of physical inactivity assessed in these two ways was 61.7% and 78.7%. The sensitivity, specificity, positive predictive value, likelihood ratios, and Kappa statistic are also presented for the single-item screening question in Table 2.

Table 2. Sensitivity, specificity, likelihood ratios, positive predictive value, and Kappa statistic of single-item screening question when compared with the NZPAQ-LF

Variable
‘Physically inactive’ as assessed by the NZPAQ-LF
Did less than 150 minutes in the last 7 days
Did not do 30 minutes on 5 or more of the last 7 days
Physically inactive (assessed using the NZPAQ-LF)
722
61.7%
922
78.7%
True positive responses (inactive)
True negative responses (active)
% Sensitivity (95%CI)
% Specificity (95%CI)
Positive likelihood ratio (95%CI)
Negative likelihood ratio (95%CI)
Positive predictive value (95%CI)
Kappa statistic*
554
364
76.7
81.1
4.05
0.29
86.7
0.56
47.3%
31.1%
73.5–79.7
77.2–84.4
3.33–4.93
0.25–0.33
83.8–89.1
p<0.001
630
240
68.3
96.4
18.9
0.33
98.6%
0.46
53.8%
20.5%
65.3–71.3
93.3–98.1
9.94–36.0
0.30–0.36
97.3–99.3
p<0.001
*Proposed cut-points for Kappa (a measure of agreement) values: 0.2–0.4 considered to represent fair agreement; 0.4–0.6 represent moderate agreement, 0.6–0.8 represent substantial agreement.23

Sensitivity refers to the proportion of those with the condition or disease (i.e. are inactive as assessed by the NZPAQ-LF) who have a positive test result (i.e. are inactive as assessed by single-item question), whereas the specificity refers to the proportion of those who do not have the condition, who have a negative test result. Kappa statistics were calculated using two criteria for inactivity derived from responses to the NZPAQ-LF: those who did not achieve 30 minutes on 5 or more of the past 7 days (Kappa=0.46) and those who achieved less than 150 minutes in total over the past 7 days (Kappa=0.56). These values represent ‘moderate’ agreement between the short and long assessment tools.

Discussion

This is the first validated single-item screening question to identify physically inactive women in general practice. The single-item screening question had good sensitivity (77%) and high specificity (81%) for identifying those women undertaking less than 150 minutes of physical activity per week. The Kappa statistic (0.56) showed moderate agreement when compared with the longer validated questionnaire. The positive predictive values calculated show that of those who answer ‘no’ to the single-item question, 87% would be undertaking less than 150 minutes of physical activity a week and 97% of them less than 30 minutes on 5 days of the week when assessed using the longer questionnaire. These results can give confidence to health practitioners that with only one question, women who would most benefit from an intervention can be accurately identified in the general practice setting.
A strength of the present study is the large sample size (n=1171) and good response rate (70%) from a women who were recruited from 10 general practices. A limitation is that participants were not drawn from a random sample, and results can only be generalised to mid-life and older women.
The proportion of physically active women in the present study compares well to another large study of postmenopausal women in which 34% reported doing at least 150 minutes of moderate or strenuous physical activity per week.24 Rates of diabetes, obesity, and physical inactivity are comparable with New Zealand population data.25,26
Furthermore, the group sampled comprises a large portion of general practice attendees, who will be at increasing risk for the development of chronic diseases that might be prevented through lifestyle interventions. The single-item question has been used successfully to screen men for physical inactivity in general practice,3 but has not yet been validated for this group.
The NZPAQ-LF used as a reference standard for comparison showed good validity when compared with objective and subjective criteria,22 but is also a self-reported instrument so may have had self-report errors that correlate to those of the screening question.27 The possibility of experimenter bias cannot be ruled out, as research nurses were not blind to participants’ answers to the single-item screening question when administering the NZPAQ-LF.
Regarding the implications for clinical practice, the question validated here was previously found to be acceptable by participants and practitioners in urban and rural healthcare settings,3,18 and one study found that more than 99% of all adult patients in general practice would complete a lifestyle screening form containing this question.18 These studies demonstrate that the single-item screening question could be administered within usual clinical practice. While validated only when asked face-to-face, it is conceivable that the single question could also be administered in written form (for example at the reception desk of a general practice) as was shown to be practical in a previous study.3
Moreover, there is evidence from a well-designed, randomised controlled trial to demonstrate the effectiveness of physical activity interventions in primary care. General practice based ‘exercise on prescription’ interventions involving physical activity counselling have been shown to be both effective and cost-effective.3,28 Furthermore, three large randomised controlled trials have shown that lifestyle modification programs are effective in increasing physical activity, reducing weight and the incidence of diabetes in those at high risk.29–31
Therefore, screening for physical inactivity is justified, and this newly validated single-item screening tool is a practical way to identify those who would most benefit from physical activity interventions.
Competing interests: None known.
Author information: Sally B Rose, Research Fellow, Department of Primary Health Care and General Practice, Wellington School of Medicine and Health Sciences, University of Otago, Wellington; C Raina Elley, Senior Lecturer, Department of General Practice & Primary Health Care, School of Population Health, University of Auckland; Beverley A Lawton, Senior Research Fellow, Department of Primary Health Care and General Practice, Wellington School of Medicine and Health Sciences, University of Otago, Wellington; Anthony C Dowell, Professor, Department of Primary Health Care and General Practice, Wellington School of Medicine and Health Sciences, University of Otago, Wellington
Acknowledgements: The National Heart Foundation of New Zealand (Grant 1091, 1164), Lottery Health Grants Board (Grant AP102414) and the Hutt Valley District Health Board funded the recruitment phase of the Women's Lifestyle Study. We also thank Selina Brown and the team of research nurses for their role in patient recruitment and data collection; SPARC and Karen Moy for assistance with training on the use of the NZPAQ-LF; and importantly, the women who generously gave us their time to participate in this research.
Correspondence: Sally Rose, Women’s Health Research Centre, Wellington School of Medicine and Health Sciences, University of Otago, Wellington, New Zealand. Fax: +64 (0)4 3855473; email: Sally.Rose@otago.ac.nz
References:
  1. Centers for Disease Contol and Prevention. Physical activity and health: A report of the Surgeon General. Atlanta: US Dept of Health and Human Services. Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996.
  2. Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393–403.
  3. Elley CR, Kerse N, Arroll B, Robinson E. Effectiveness of counselling patients on physical activity in general practice: cluster randomised controlled trial. BMJ. 2003;326:793.
  4. Petrella RJ, Koval JJ, Cunningham DA, Paterson DH. Can primary care doctors prescribe exercise to improve fitness? The step test exercise prescription (STEP) project. Am J Prev Medicine. 2003;24(4):316–22.
  5. The Writing Group for the Activity Counseling trial Research Group. Effects of Physical activity counseling in primary care: the activity counseling trial: a randomized controlled trial. JAMA. 2001;286(6):677–87.
  6. Fagard RH. Exercise characteristics and the blood pressure response to dynamic physical training. Medicine & Science in Sports & Exercise. 2001;33(6 Suppl):S484–92; discussion S493–4.
  7. Wilmore JH. Dose-response: variation with age, sex, and health status. Medicine & Science in Sports & Exercise 2001;33(6 Suppl):S622–34; discussion S640–1.
  8. Blair SN, Kohl HW, 3rd, Barlow CE, et al. Changes in physical fitness and all-cause mortality. A prospective study of healthy and unhealthy men.[see comment]. JAMA. 1995;273(14):1093–8.
  9. Elley CR, Kerse NM, Arroll B. Why target sedentary adults in primary health care? Baseline results from the Waikato Heart, Health, and Activity Study. Prev Med. 2003;37(4):342–8.
  10. Weiss TW, Slater CH, Green LW, et al. The validity of single-item, self-assessment questions as measures of adult physical activity. J Clin Epidemiol. 1990;43(11):1123–9.
  11. Li S, Carlson E, Holm K. Validation of a single-item measure of usual physical activity. Perceptual & Motor Skills. 2000;91(2):593–602.
  12. Iwai N, Hisamichi S, Hayakawa N, et al. Validity and reliability of single-item questions about physical activity. Journal of Epidemiology. 2001;11(5):211–8.
  13. Schechtman KB, Barzilai B, Rost K, Fisher EB, Jr. Measuring physical activity with a single question. American Journal of Public Health. 1991;81(6):771–3.
  14. Macera CA, Ham SA, Jones DA, et al. Limitations on the Use of a Single Screening Question to Measure Sedentary Behavior. American Journal of Public Health. 2001;91(12):2010–2.
  15. Snyder DC, Sloane R, Lobach D, et al. Agreement between a brief mailed screener and an in-depth telephone survey: Observations from the Fresh Start study. Journal of the American Dietetic Association. 2004;104(10):1593.
  16. Marshall AL, Smith BJ, Bauman AE, et al. Reliability and validity of a brief physical activity assessment for use by family doctors * Commentary. Br J Sports Med 2005;39(5):294-297.
  17. Prochaska JJ, Sallis JF, Long B. A physical activity screening measure for use with adolescents in primary care. Archives of Pediatrics & Adolescent Medicine. 2001;155(5):554–9.
  18. Goodyear-Smith F, Arroll B, Sullivan S, et al. Lifestyle screening: development of an acceptable multi-item general practice tool. N Z Med J. 2004;117(1205). http://www.nzma.org.nz/journal/117-1205/1146/
  19. McLean G, Tobias M. The New Zealand Physical Activity Questionnaires. Report on the validation and use of the NZPAQ-LF and NZPAQ-SF self-report physical activity survey instruments. 2004. http://www.sparc.org.nz/research-policy/research/nzspas-97-01/nzpaq
  20. Rose S, Lawton B, Dowell T, Fenton AJ. Risk factors for Type 2 diabetes in postmenopausal women: a cross-sectional study. N Z Med J. 2004;117(1207). http://www.nzma.org.nz/journal/117-1207/1206/content.pdf
  21. Rose SB, Lawton BA, Elley CA, et al. The 'Women's Lifestyle Study' 2-year randomised controlled trial of physical activity counselling in primary health care: rationale and study design. BMC Public Health. 2007;7:166.
  22. Moy K, McFarlane K, Scragg R, Robinson S. Validation of MOH-short and SPARC-long physical activity questionnaires. Final report: University of Auckland, Sport and Recreation New Zealand; 2003.
  23. McGinn T, Wyer PC, Newman TB, et al. Tips for learners of evidence-based medicine: 3. Measures of observer variability (kappa statistic).[see comment][erratum appears in CMAJ. 2005 Jul 5;173(1):18]. CMAJ Canadian Medical Association Journal. 2004;171(11):1369–73.
  24. Hsia J, Wu L, Allen C, et al. Physical activity and diabetes risk in postmenopausal women. American Journal of Preventive Medicine. 2005;28(1):19–25.
  25. Ministry of Health. NZ Food, NZ People: Key results of the 1997 National Nutrition Survey. Wellington, New Zealand: Ministry of Health; 1999.
  26. van Aalst I, Kazakov D, McLean G. SPARC facts. Results of the New Zealand sport and physical activity surveys (1997-2001). Wellington: Sport and Recreation New Zealand; 2003.
  27. Sallis JF, Saelens BE. Assessment of physical activity by self-report: Status, limitations, and future directions. [Erratum appears in Res Q Exerc Sport 2000 Dec: 71(4):409]. Res Q Exerc Sport 2000;71(2 Suppl):S1–4.
  28. Elley CR, Kerse N, Arroll B, et al. Cost-effectiveness of physical activity counselling in general practice. N Z Med J. 2004;117(1207). http://www.nzma.org.nz/journal/117-1207/1216/
  29. Diabetes Prevention Program Research Group. Reduction in the incidence of Type 2 Diabetes with lifestyle intervention or Metformin. New Eng J Med. 2002;346(6):393–403.
  30. Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. The Da Qing IGT and Diabetes Study. Diabetes Care. 1997;20(4):537–44.
  31. Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344(18):1343–50.
     
Current issue | Search journal | Archived issues | Classifieds | Hotline (free ads)
Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals