Open Access

International Physical Activity Questionnaire (IPAQ) and New Zealand Physical Activity Questionnaire (NZPAQ): A doubly labelled water validation

  • Ralph Maddison1Email author,
  • Cliona Ni Mhurchu1,
  • Yannan Jiang1,
  • Stephen Vander Hoorn1,
  • Anthony Rodgers1,
  • Carlene MM Lawes1 and
  • Elaine Rush2
Contributed equally
International Journal of Behavioral Nutrition and Physical Activity20074:62

DOI: 10.1186/1479-5868-4-62

Received: 30 March 2007

Accepted: 03 December 2007

Published: 03 December 2007



Accurate measurement of physical activity is a pre-requisite for monitoring population health and for evaluating effective interventions. The International Physical Activity Questionnaire (IPAQ) is used as a comparable and standardised self-report measure of habitual physical activity of populations from different countries and socio-cultural contexts. The IPAQ has been modified to produce a New Zealand physical activity questionnaire (NZPAQ). The aim of this study was to validate the IPAQ and NZPAQ against doubly labelled water (DLW). Method: Total energy expenditure (TEE) was measured over a 15-day period using DLW. Activity-related energy expenditure (AEE) was estimated by subtracting the energy expenditure from resting metabolic rate and thermic effect of feeding from TEE. The IPAQ (long form) and NZPAQ (short form) were completed at the end of each 7-day period. Activity-related energy expenditure (IPAQAEE and NZPAQAEE) was calculated from each questionnaire and compared to DLWAEE.


Thirty six adults aged 18 to 56 years (56% female) completed all measurements. Compared to DLWAEE, IPAQAEE and NZPAQAEE on average underestimated energy expenditure by 27% and 59%, respectively. There was good agreement between DLWAEE and both IPAQAEE and NZPAQAEE at lower levels of physical activity. However there was marked underestimation of questionnaire-derived energy expenditure at higher levels of activity.


Both the IPAQ and NZPAQ instruments have a demonstrated systematic bias toward underestimation of physical activity-related energy expenditure at higher levels of physical activity compared to DLW. Appropriate calibration factors could be used to correct for measurement error in physical activity questionnaires and hence improve estimation of AEE.


The global estimate for physical inactivity (doing very little or no physical activity at work, at home, for transport or in discretionary time) is 17% [1], whereas the estimate for insufficient levels of physical activity (< 150 minutes moderate or < 60 minutes of vigorous activity per week) is 40% [1]. Lack of physical activity is associated with an increased risk of ischemic heart disease, type 2 diabetes, colon cancer, depression, and breast cancer [1]. Accurate measurement of physical activity is a pre-requisite for monitoring population health and evaluating effective interventions, with the preferred and most practical method in epidemiological studies being self-report questionnaires.

The International Physical Activity Questionnaire (IPAQ) is used as a standardised measure to estimate habitual practice of physical activities of populations from different countries and socio-cultural contexts [2, 3]. Two-forms of the IPAQ have been developed: a short and a long version, both of which involve 7-day recall of physical activity. The short-form (SF) was designed for use in surveillance studies, in which time is limited, and consists of 8 items to estimate the time spent performing physical activities (moderate to vigorous) and inactivity (time spent sitting). The long form (LF) was designed to provide a comprehensive evaluation of daily physical activities, and assesses the time spent walking, doing moderate-intensity and vigorous-intensity activity within the domains of work, transportation, domestic and gardening (yard) activities, and leisure-related activities.

In 2003, Craig et al [3] provided reliability and validity data from 12 countries for both the long-and short-forms, showing that the IPAQ questionnaires produced reliable data, as well as acceptable criterion validity against the CSA accelerometer (rs approximately 0.30). Although a number of reliability and validity studies have been conducted [37], the IPAQ has not yet been validated against the gold (criterion) standard of doubly labelled water (DLW) [8]. The DLW technique is a biochemical procedure used to assess total energy expenditure (TEE) through biological markers that reflect the rate of metabolism in the body [9]. Combined with the measurement of resting metabolic rate (RMR) this technique is considered the criterion measure of physical activity-related energy expenditure.

The New Zealand (NZ) Ministry of Health and Sport and Recreation New Zealand (SPARC) recently developed two self-report questionnaires, The NZ Physical Activity Questionnaire (NZPAQ): both a more in-depth long form (LF) and a briefer short-form (SF). The NZPAQ-SF was developed as a surveillance tool and is a modified version of the IPAQ-SF. The modifications were intended to increase the relevance and cultural appropriateness of the instrument for New Zealand. The NZPAQ-SF questions assess the three dimensions of physical activity, 'frequency, duration, and intensity'. The format for the first 7 questions is for participants to recall the frequency they performed brisk walking, moderate, vigorous and a combination of both over the past 7 days. If indicated, participants then record the duration (time) they performed the activity. Question 8 is optional and indicates stage of exercise behaviour change. Visual show-cards are provided to the participant to given an indication of moderate and vigorous activities. The most important differences between the instruments are (1) the IPAQ-SF does not measure the frequency of activity, by contrast, the NZPAQ-SF includes an extra item to tap this dimension of physical activity. (2) The order of items in the NZPAQ-SF have been reversed compared to the IPAQ-SF. (3) The NZPAQ-SF does not include an item on sedentary behaviour and focuses on physical activity only.

In a validation study, total physical activity (including walking and moderate, and vigorous intensity activities) measured using the NZPAQ-SF was correlated with that measured using the IPAQ-LF (r = 0.33 p < 0.0001). However physical activity levels were overestimated at low activity levels and underestimated at high activity levels when compared to heart-rate derived measurements [10]. The indication for use of the IPAQ-LF is for research whereas the IPAQ-SF is for surveillance purposes. Because of the similarity between the IPAQ-SF and NZPAQ-SF we chose the latter for this study. The aim of this study was to validate both the IPAQ-LF and NZPAQ-SF against DLW.


Study participants and recruitment

Adults living in Auckland, New Zealand were eligible if they were aged between 18 and 65 years, able to give written informed consent, able to communicate in English, and weight stable (not dieting or had not experienced recent dramatic changes in weight). Potential participants were recruited via community advertisements, direct contact through the university, and word of mouth. Adults who used a walking aid (crutches, cane, or wheelchair) or who were unable to perform physical activity for medical reasons were excluded. The study protocol and related documents were approved by the Northern Regional Ethics Committee (NTY/06/02/011).

Overview of procedures

Each participant was assessed over a 15-day period. At baseline (Day 0), measurements were taken of height, weight, and resting metabolic rate (RMR). Baseline urine samples were collected prior to administration of DLW. Participants returned at days 1, 8, and 15 for spot urine sample collection and questionnaire completion.


Doubly labelled water (DLW)

DLW is considered the reference or gold standard for the assessment of total energy expenditure (TEE). When combined with a measure of resting metabolic rate (RMR), activity-related energy expenditure (AEE) can be calculated [11]. In this study TEE was determined using the DLW method [12, 13]. After a baseline urine sample was collected for determination of background isotope enrichment (Day 0), each participant ingested a single dose of DLW (2H and 18O; Cambridge Isotope Laboratories, Inc., Andover, MA), corresponding to 0.12 g of 99.9% of 2H2O and 2 g of 10% 18O per kg body weight. The dose was followed by the ingestion of two 50-mL washes of the container with tap water. Post enrichment urine samples were collected the following day (Day 1) and at Days 8 and 15. A trained research assistant collected all specimens and recorded the exact time of voiding. Samples were frozen at -21°C until assayed for 2H and 18O. A subsample of the dose was reserved and after appropriate dilution analyzed at the same time as the urine samples to serve as a reference point. Samples were analysed in triplicate using a TC/EA (Thermo Chemical Elemental Analyser) interfaced (ConFlo II; Thermo Finnigan, Bremen, Germany) to an isotope ratio mass spectrometer (Delta Plus XP; Thermo Finnigan, Germany). TEE was calculated using the two point approach with an assumed respiratory quotient (RQ) of 0.85 over the 14 d period [14]. Activity-related energy expenditure (AEE) was calculated as the difference between TEE and RMR. For the purpose of this study, diet-induced thermogenesis (DIT) was considered to be 10% of TEE [11].

Body composition

Anthropometric measurements were taken in the morning of Day 1. Height was measured to the nearest 0.1 cm with a portable stadiometer (Harpenden Stadiometer, Chasmors Ltd, London) and weight was measured to the nearest 0.1 kg on an electronic scale (Salter Scales). To assess body composition, resistance, reactance, impedance and phase were determined using single-frequency (50 kHz) bioimpedance analysis (Model BIM4, Impedimed, Capalaba, Australia) using a standardized protocol. The average of three repeated measurements of resistance agreeing to within 1 ohm of each other was used in subsequent analyses. Lukaski [15] equations were used to determine fat free mass (FFM) for European participants and Swinburn [16] equations to determine fat mass for Maori and Pacific Island participants. Fat mass (FM) was derived as the difference between body weight and FFM. Percentage of body fat was calculated as FM/body weight × 100.

Resting Metabolic Rate (RMR)

RMR was measured by indirect calorimetry (Day 0) using the MetaMax®3B (Cortex, Biophysik) portable gas exchange analysis system, a device with demonstrated precision [17, 18]. This methodology incorporates the fraction of inspired oxygen and expired O2 and CO2, pump rate, ambient conditions in association with the Weir equation [19] to calculate RMR. Participants were instructed to limit any vigorous physical activity the day before RMR measurement, and tests were conducted in the morning after an overnight fast. A face mask (Hans Rudolf, Kansas City, MO) held in place by a nylon harness covered the participant's nose and mouth. The mask was attached to a bidirectional digital turbine flow meter to measure the volume of inspired and expired air. A sample line between the turbine and analyzer unit determined O2 and CO2 content. A two point calibration procedure was conducted before each testing session according to the manufacturer's guidelines (Cortex Biophysik MetaMax®3B Manual). Participants rested for a period of 20 minutes in the supine position before RMR testing, during which breath by breath measurements were recorded for 30 minutes. A coefficient of variation of less than 10% of consecutive energy expenditure measurements was used as steady state [9].

Physical activity questionnaires

The IPAQ-LF was administered at day 0, day 8, and 15. The questionnaire took approximately 15 minutes to complete, and a trained research assistant was present to assist participants with the questionnaire.

The NZPAQ-SF was administered at day 8 and 15. Participants indicated the number of times in the past seven days they spent walking, and in moderate, and vigorous activity) and also the usual amount of time spent per day. Physical activity show cards provided examples of the intensity of various physical activities (e.g., Rugby Union/carrying heavy loads = vigorous, doubles tennis/lawn mowing = moderate). NZ Maori (indigenous New Zealanders)-specific activities are included in this questionnaire. The questionnaire took approximately 10 minutes to complete and a trained research assistant was present to assist participants.

Energy expenditure estimation

Physical activity data obtained from the IPAQ-LF and NZPAQ-SF were computed for metabolic equivalent (MET)-minutes per week, calculated as the MET intensity multiplied by the minutes for each activity over the seven day period. METs are multiples of the RMR and a MET-minute is computed by multiplying the MET score of an activity by the minutes. It was assumed that 1 MET was equivalent to 1 kcal/kg/hr for all participants [20]. Adopting this approach, the MET values (e.g. 3.3) were multiplied by the participant's body weight (kg) and then divided by 60 minutes to convert the unit into kcal/min (e.g. 3.3 * 50 kg/60 min). Energy expenditure values were then converted to kJ (1 kcal = 4.18 kJ) for direct comparison with DLW. Energy expenditure values from week 1 (Days 1–7) and week 2 (Days 8–15) for the IPAQ-LF and NZPAQ-SF, respectively were summed and divided by the number of days to provide average daily physical-activity related energy expenditure.


Statistical analysis was performed using SAS (Statistical Analysis Systems) version 9.1.3 and S-PLUS 6.1 for Windows. Means and standard deviation (SD) were calculated. The significance level was set at 5%. Average daily activity-related energy expenditure (AEE) measured using DLW, IPAQ-LF, and NZPAQ-SF was calculated for each participant. Bland-Altman plots were first used to evaluate the level of agreement in activity-related energy expenditure (kJ) between DLW and the different measures. Spearman correlation coefficients were then calculated to assess the degree of linear association between measurements. Regression analysis was finally conducted to explore the relationships between the dependent variable (DLW derived AEE) and independent variable (AEE derived from the IPAQ-LF and NZPAQ-SF, respectively). Potential covariates of interest (age, sex, weight, fat free mass) were considered and kept in the regression model if they were statistically significant.

To examine the level of agreement between the IPAQ-LF, NZPAQ-SF, and DLW we produced Bland-Altman plots [21, 22]. For each series, the Y axis represents the difference in AEE between each measure (IPAQ and NZPAQ) and DLW. The x axis of these plots represents the AEE from DLW. The limits of agreement (dotted lines) equalling two standard deviations of the mean difference above and below the mean (solid line) were plotted.


Participants were 36 adults (20 female and 16 males) aged 18 to 64 years (Table 1). The participants were, on average 39 years old, predominantly of European ethnicity (75%). According to international [23] BMI cut-off points, participants were on average overweight (Mean BMI = 25.9 kg/m2), which is typical of the New Zealand population (BMI = 26.1 kg/m2) [24]. No sex differences were found for BMI, percent body fat, weight, TEE, AEE. The males, however, had significantly more lean body mass (62 kg) than the females (48 kg) (p < 0.001).
Table 1

Participant's characteristics






Age (yr)





Height (cm)





Weight (kg)





BMI (kg/m2)





FM (kg)





FFM (kg)





Sex (n/%)



20 (56%)



16 (44%)


Ethnicity (n/%)


NZ European

27 (75%)


NZ Maori

4 (11%)



1 (3%)



2 (6%)



5 (15%)


Reliability of the measures (IPAQ and NZPAQ) The test-retest reliability coefficients of the IPAQ-LF questionnaires were acceptable with Spearman correlation coefficients of 0.79 (p < 0.0001) between 0 and 8 days and 0.74 (p < 0.0001) between days 8 and 15. A similar correlation coefficient was found for the NZPAQ-SF questionnaire between days 8 and 15 (0.69, p < 0.0001). Both parametric and non-parametric tests were conducted and revealed no significant differences in MET-minutes per week from baseline to Day 8 for the IPAQ, suggesting physical activity behaviour was not affected by participation in this study

Energy expenditure

AEE values for DLW, IPAQ-LF, and NZPAQ-SF are presented in Table 2. The correlations between IPAQAEE, NZPAQAEE, and DLWAEE were of a moderate magnitude (rs = 0.31–0.38, respectively). Because AEE values (kJs) were derived from MET/min estimates from the IPAQ-LF and NZPAQ-SF correlations were high (Table 3). At the group level, compared to DLWAEE, mean IPAQAEE and NZPAQAEE underestimated energy expenditure by 27% and 59%, respectively. At the individual, compared to DLWAEE there was overestimation of energy expenditure with the IPAQ-LF for 7 participants at lower levels of physical activity (equivalent to 376 daily MET/min per day). However for the remaining participants, as activity levels increased the IPAQ-LF underestimated AEE compared to DLWAEE (Figure 1) from approximately 1000 kJ (equivalent to 187 MET/min per day). Results were similar for NZPAQAEE with marked underestimation of energy expenditure as levels of activity increased from approximately 1000 kJ (Figure 2).
Table 2

Energy expenditure (Means ± SD)














Females (n= 20)

12174 ± 2695

3628 ± 1957

2225 ± 1892

1318 ± 1040

439 ± 336

269 ± 206

Males (n = 16)

13603 ± 3112

3224 ± 2377

2855 ± 2224

1497 ± 1372

503 ± 397

261 ± 232

Total (n = 36)

12809 ± 2935

3448 ± 2131

2505 ± 2041

1398 ± 1184

467 ± 361

265 ± 214


DLW = Doubly labelled water

LF = Long form

SF = Short form

TEE = Average total energy expenditure per day

AEE = Average activity-related energy expenditure per day

MET/Min = Daily Metabolic Equivalent Minutes

Table 3

Spearman Correlation Coefficients between Different EE Measures





IPAQ Met/Min



















IPAQ Met/Min







* p = 0.06,

** p < 0.05

*** p < 0.01

Figure 1

Difference between AEE determined with DLW and IPAQ-LF.

Figure 2

Difference between AEE determined with DLW and NZPAQ-SF.

Due to the demonstrated systematic bias toward underestimation of AEE at higher levels of physical activity, regression analyses were conducted where DLWAEE was the dependent variable and the respective self-report measures were each used as independent variables (Table 4). In addition, because the NZPAQ-SF was developed from the IPAQ we conducted a regression analysis with IPAQAEE as the dependant variable and NZPAQAEE as the independent variable (Table 4, Set 3). Potential covariates of interest (age, sex, weight, fat [kg], and FFM) were considered for an adjusted analysis. Except for sex and FFM (IPAQ vs. NZPAQ), all other covariates were highly non-significant and therefore dropped from the regression model. Approximately 38% of the variance in DLWAEE was explained by the IPAQ-LF and NZPAQ-SF, respectively. The NZPAQAEE explained approximately 53% of the variance of the IPAQAEE.
Table 4

Analysis of Covariance (Regression Equation)

Set 1


Dependent Variable (Y):

DLW average daily activity-related EE (kJ)

Independent Variable (X):

IPAQ average daily activity-related EE (kJ)




p-value (t-test)









R2 = 0.25


Regression Line: DLWAEE = 2137 + 0.52 * IPAQ

With covariates


















R2 = 0.38


Regression Line: DLWAEE = -2082 + 0.45 * IPAQAEE + 2589 * sex + 97 * FFM

Set 2


Dependent Variable (Y):

DLW average daily activity-related EE (kJ)

Independent Variable (X):

NZPAQ-SF average daily activity-related EE (kJ)




p-value (t-test)









R2 = 0.38


Regression Line: DLW = 1885 + 1.18 * NZPAQAEE

Set 3


Dependent Variable (Y):

IPAQ-LF average daily activity-related EE (kJ)

Independent Variable (X):

NZPAQ-SF average daily activity-related EE (kJ)




p-value (t-test)









R2 = 0.47


Regression Line: IPAQ = 848 + 1.18 * NZPAQ

With covariates














R2 = 0.53


Regression Line: IPAQAEE = -1135 + 1.25 * NZPAQAEE + 49 * Age


To the best of our knowledge, this is the only study to date to provide DLW validation of the IPAQ-LF and NZPAQ-SF. Compared with DLW, our data showed that both self-report measures on average underestimated physical activity-related energy expenditure especially at higher levels of physical activity. Although accurate measurement of daily free-living physical activity is important to describe the behaviour, its impact on health, and to assess the effectiveness of interventions, only a handful of studies have tested the ability of different epidemiological methods to assess EE in free-living individuals [25, 26].

Compared to DLWAEE, on average, the IPAQ-LF underestimated activity-related energy expenditure. There was good approximation at the lowest levels of physical activity (1000–2000 kJ or equivalent to ~187–376 daily MET/mins), with greater underestimation as physical activity increased (greater than 3000 kJs or ~565 daily MET/mins). Similar findings were found for the NZPAQ-SF, except that there was a systematic underestimation starting at 1000 kJ, with greatest underestimation at over 5000 kJ (~942 daily MET/min). The best overall approximation of AEE with DLW was with the IPAQ-LF with a 27% deficit, compared to the 59% deficit of the NZPAQ-SF. These findings are comparable with other studies that have shown that 7-day recall questionnaires (e.g., Cross-Cultural Activity Participation Study Typical Week Activity Recall questionnaire; Minnesota Leisure Time PA questionnaire; College Alumni Score) underestimated energy expenditure by 10–30% compared to DLW measurement [2527]. However, other studies have shown that 7-day physical activity recall questionnaires (e.g., Stanford Seven Day Recall Questionnaire) overestimate energy expenditure compared to DLW in free-living adults [25, 2729].

The systematic bias in estimation of energy expenditure using the NZPAQ-SF is consistent with the NZPAQ development validation work, which showed self-reported physical activity levels were overestimated at low activity levels and underestimated at high activity levels in comparison to heart-rate derived measurements [10]. This bias may be due to the limited number and the nature of the questions in the short form, which asks participants to rate time spent in moderate to vigorous activities only. Light-intensity activities are not captured by items in the NZPAQ-SF but would be included in the DLW estimate. It is possible that various activities are not accurately recalled, either because people forget, or they cannot recall unstructured bouts of activity (e.g., such as climbing stairs, household work), thereby underestimating total physical activity.

Both physical activity questionnaires (IPAQ-LF and NZPAQ-SF) were not specifically designed to measure activity-related energy expenditure, rather MET values have been provided with the IPAQ scoring protocol to give an indication of levels of physical activity. In the present study we estimated AEE from MET minutes, which may have contributed to the underestimation of energy expenditure. For example, allocated MET values from the IPAQ scoring protocol may not reflect the intensity of all activities for all people. For example, an individual who indicates they walk 5 days a week for approximately 30 minutes/day equates to 150 minutes of activity. According to the IPAQ scoring protocol, MET-minutes/week = 3.3 × walking minutes × walking days in the week, which is equivalent to 495 MET-minutes/week. However one's walking speed and intensity vary according to many variables (e.g., weather, geography etc), and may in fact be greater or less than 3.3 METS [30], resulting in a misrepresentation of the energy cost of the activity. Future studies might choose to control for individual levels of physical fitness and exercise levels. Individuals with greater physical fitness, and who participate in large amounts of activity may perceive moderate and vigorous activity differently compared to those who are more sedentary, thereby confounding self-report-derived energy expenditure.

Although our results generally suggest the IPAQ-LF on average underestimated physical activity-related energy expenditure, simple adjustments to the energy expenditure estimations could be used to correct for this systematic bias. There was evidence supporting a linear relationship between AEE and IPAQ/NZPAQ, therefore, a crude approach to adjust energy expenditure would be to take the IPAQ estimate of AEE, divide by 2 and add approximately 2000 kJ. This approach does however suggest a constant difference in energy expenditure. Our regression analysis showed that including measures of fat free mass and sex improved the prediction of DLWAEE (explaining approximately 38% variance) using the IPAQ-LF, therefore more sophisticated adjustment approaches may need to consider these variables.

The use of correction factors for self-report measures of physical activity has implications for future research, particularly with the frequent use of these measures. Although a variety of methods are employed to assess physical activity behaviour, such as accelerometers and heart rate monitors, their use is often cost-prohibitive for epidemiological studies or public health surveillance. Both objective measures also have been shown to over-and under-estimate energy expenditure compared to DLW [28, 31]. Thus the preferred method for population-level purposes in adults is self-reported instruments [32].

Self-reported physical activity is subject to reporting bias and as this study has shown, physical activity questionnaires do not accurately quantify activity-related energy expenditure. The development of an appropriate correction factor, which could be applied to questionnaires such as the IPAQ would overcome systematic bias and provide a more reliable and valid measure of energy expenditure estimation. At the population level the IPAQ has potential to standardise physical activity measurement. Used in conjunction with an appropriate correction factor the IPAQ would be valuable for (1) international comparison and (2) estimating the burden of disease associated with physical inactivity.

Previous research findings have shown that physical activity recall questionnaires both under-and over-estimate energy expenditure, however physical activity records may be more accurate [25]. The inherent difficulties associated with recalling physical activities may limit those data to estimate the energy expenditure associated with free living physical activity. Rather, 7-day recall questionnaires may be best suited to national and international level surveillance by providing information on the nature (type and intensity) and duration (time) of physical activity.

Similar to other validations studies using the DLW approach this study was limited by the small sample size (n = 36). Both the IPAQ and NZPAQ were validated in healthy, free-living adults and therefore the findings cannot be generalised to other populations. Strengths however, include the prospective study design and the use of DLW and RMR to assess AEE. This is also the first study to conduct DLW validation of the IPAQ.

In conclusion, the findings of this DLW validation study support the use of the IPAQ and NZPAQ as measures of physical activity for the purpose of epidemiological research, such as periodic national surveys. Overall, both instruments have a demonstrated systematic bias toward underestimation of physical activity-related energy expenditure (particularly the NZPAQ) at higher levels of physical activity. Appropriate calibration factors could be used to correct for this measurement error and hence improve estimation of AEE. No single valid, reliable field measure of physical activity exists which is logistically feasible for use in all population settings. Therefore, an understanding of the strengths and limitations of each technique is required before choosing the appropriate assessment method for a specific research question. Although recall questionnaires are the most practical means of measuring PA in large population studies, they both over-and under-estimate physical activity-related EE and, if using them for this purpose, calibration factors should be considered to correct for this.




The authors wish to thank IsoTrace NZ Ltd for their technical assistance with the doubly-labeled water measurements. Thanks to Dr Joshua Salomon, Harvard University for contributing to the study design and methodology. We would also like to thank all the participants that took part in this study.

This was an investigator-initiated study conducted in collaboration with the Harvard School of Public Health (HSPH) and was funded as a subcomponent of an NIH research grant. HSPH provided advice regarding the study design and methodology but was not involved in analysis or interpretation of the data.

There were no professional relationships with companies or manufacturers and therefore, there is no conflict of interest. Results of the present study do not constitute endorsement of the product by the authors.

RM and CL are supported by National Heart Foundation of New Zealand Research Fellowships.

Authors’ Affiliations

Clinical Trials Research Unit, University of Auckland
Centre for Physical Activity and Nutrition, Auckland University of Technology


  1. Bull FC, Armstrong TP, Dixon T, Ham S, Neiman A, Pratt M: Physical inactivity. Comparative Quantification of Health Risks. Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Edited by: Essati M, Lopez AD, Rodgers A, Murray CJL. 2004, Geneva: World Health Organization, 729-881.Google Scholar
  2. Booth M: Assessment of physical activity: An International perspective. Research Quarterly for Exercise and Sport. 2000, 71 (2): 114-120.View ArticleGoogle Scholar
  3. Craig CL, Marshall AJ, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, Pekka OJA: International Physical Activity Questionnaire: 12-country reliability and validity. Medicine & Science in Sports & Exercise. 2003, 35 (8): 1381-1395. 10.1249/01.MSS.0000078924.61453.FB.View ArticleGoogle Scholar
  4. Brown WJ, Trost SG, Bauman A, Mummery K, Owen N: Test-retest reliability of four physical activity measures used in population surveys. Journal of Science and Medicine in Sport. 2004, 7: 205-215. 10.1016/S1440-2440(04)80010-0.View ArticleGoogle Scholar
  5. Fogelholm M, Malmberg J, Suni J, Santtila M, Kyröläinen H, Mäntysaari M, Oja P: International Physical Activity Questionnaire: Validity against fitness. Medicine & Science in Sports & Exercise. 2006, 38: 753-760. 10.1249/01.mss.0000194075.16960.20.View ArticleGoogle Scholar
  6. Hagstromer M, Oja P, Sjostrom M: The international physical activity questionnaire (IPAQ): a study of concurrent and construct validity. Public Health Nutrition. 2006, 9 (6): 755-762.View ArticleGoogle Scholar
  7. MacFarlane D, Lee C, Ho E, Chan KL, Chan D: Convergent validity of six methods to assess physical activity in daily life. Journal of Applied Physiology. 2006, 101: 1328-1334. 10.1152/japplphysiol.00336.2006.View ArticleGoogle Scholar
  8. Melanson E, Freedson PS: Physical activity assessment: a review of methods. Critical Reviews of Food Science and Nutrition. 1992, 36 (5): 385-396.View ArticleGoogle Scholar
  9. Starling RD: Use of doubly labeled water and indirect calorimetry to assess physical activity. Physical activity assessment for health-related research. Edited by: Welk GJ. 2002, Champaign, Ill: Human Kinetics, 197-209.Google Scholar
  10. Moy K, McFarlane K, Scragg R, Robinson S: Validation of MOH-short and SPARC-long Physical Activity Questionnaires. 2003, Wellington, NZ: Sport and Recreation New Zealand (SPARC), 1-142.Google Scholar
  11. Bassett DR: Validity and reliability issues in objective monitoring of physical activity. Research Quarterly for Exercise and Sport. 2000, 71 (2): 30-36.View ArticleGoogle Scholar
  12. Montoye HJ, Kemper HCG, Saris WHM, Washburn RA: Measuring Physical Activity and Energy Expenditure. 1996, Champaign, IL: Human KineticsGoogle Scholar
  13. Westerterp KR, Wouters L, van Marken Lichtenbelt W: The Maastricht protocol for the measurement of body composition and energy expenditure with labeled water. Obesity Research. 1995, 3 (Suppl 1): 49-57.View ArticleGoogle Scholar
  14. Schoeller DA, Ravussin E, Schutz Y, Acheson KJ, Baertschi P, Jequier E: Energy expenditure by doubly labeled water: validation in humans and proposed calculation. American Journal of Physiology. 1986, 250: R823-830.Google Scholar
  15. Lukaski H, Bolonchuk W, Hall C, Siders W: Validation of tetrapolar bioelectrical impedance method to assess body composition. Journal of Applied Physiology. 1986, 60 (4): 1327-1332.Google Scholar
  16. Swinburn B, Ley S, Carmichael H, Plank L: Body size and composition in Polynesians. International Journ. 1999, 23 (11): 1178-1183.Google Scholar
  17. Larsson PU, Wadell KM, Jakobsson EJ, Burlin LU, Henriksson-Larsen KB: Validation of the MetaMax II portable metabolic measurement system. International Journal of Sports Medicine. 2004, 24: 115-123.Google Scholar
  18. Medbo JI, Mamen A, Welde B, Von Heimburg V, Stokke R: Examination of the Metamax I and II oxygen analysers during exercise studies in the laboratory. Scandinavian Journal of Clinical Laboratory Investigation. 2002, 62: 585-598. 10.1080/003655102764654321.View ArticleGoogle Scholar
  19. Weir J: New methods for calculating metabolic rate with special reference to protein metabolism. Journal of Physiology. 1949, 109: 1-9.View ArticleGoogle Scholar
  20. Ainsworth BE, Montoye HJ, Leon AS: Methods of assessing physical activity during leisure and work. Measuring Physical Activity and Energy Expenditure. Edited by: Montoye HJ, Kemper HCG, Saris WHM, Washburn RA. 1996, Champaign Illinois: Human Kinetics, 146-159.Google Scholar
  21. Altman DG, Bland JM: Measurement in Medicine: the analysis of method comparison studies. The Statistician. 1983, 32: 301-317. 10.2307/2987937.View ArticleGoogle Scholar
  22. Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinical measurement. The Lancet. 1986, 320 (February): 307-310.View ArticleGoogle Scholar
  23. World Health Organization: Obesity: preventing and managing the global epidemic. 1998, Geneva, Switzerland: World Health OrganisationGoogle Scholar
  24. Russell DG, Parnell WR, Wilson NC, Faed J, Ferguson E, Herbison P, Horwath C, Nye T, Reid P, Walker R, Wilson B, Tukuitonga l: NZ Food: NZ People. Key results of the 1997 National Nutrition Survey. 1999, Wellington: Ministry of HealthGoogle Scholar
  25. Conway JM, Seale JL, Jacobs DR, Irwin ML, Ainsworth BE: Comparison of energy expenditure estimates from doubly labeled water, a physical activity questionnaire, and physical activity records. American Journal of Clinical Nutrition. 2002, 75: 519-525.Google Scholar
  26. Conway JM, Irwin ML, Ainsworth BE: Estimating energy expenditure from the Minnesota Leisure Time Physical Activity and Tecumseh Occupational Activity questionnaires - a doubly labeled water validation. J Clin Epidemiol. 2002, 55 (4): 329-399. 10.1016/S0895-4356(01)00497-8.View ArticleGoogle Scholar
  27. Mahabir S, Baer DJ, Giffen C, Clevidence BA, Campbell WS, Taylor PR, Hartman TJ: Comparison of energy expenditure estimates from 4 physical activity questionnaires with doubly labeled water estimates in postmenopausal women. American Journal of Clinical Nutrition. 2006, 84 (1): 230-236.Google Scholar
  28. Ainslie PN, Reilly T, Westerterp KR: Estimating human energy expenditure. Sports Medicine. 2003, 33 (9): 683-698. 10.2165/00007256-200333090-00004.View ArticleGoogle Scholar
  29. Irwin ML, Ainsworth BE, Conway JM: Estimation of energy expenditure from physical activity measures: Determinants of accuracy. Obesity Research. 2001, 9 (9): 517-525.View ArticleGoogle Scholar
  30. Ainsworth BE, Haskell WL, Leon AS, Jacobs DR, Montoye HJ, Sallis JF, Paffenbarger RS: Compendium of physical activities: classification of energy costs of human physical activities. Medicine & Science in Sports & Exercise. 1993, 25: 71-80. 10.1249/00005768-199301000-00011.View ArticleGoogle Scholar
  31. Livingstone MB, Coward WA, AM P, Davies PSW, Strain JJ, McKenna PG, Mahoney CA, White JA, Stewart CM, Kerr MJ: Daily energy expenditure in free-living children: comparison of heart-rate monitoring with doubly labeled water method. American Journal of Clinical Nutrition. 1992, 56 (2): 343-352.Google Scholar
  32. Sallis J, Saelens B: Assessment of physical activity by self-report: status, limitations, and future directions. Research Quarterly for Exercise and Sport. 2000, 71: S1-S14.View ArticleGoogle Scholar


© Maddison et al; licensee BioMed Central Ltd. 2007

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.