Abstract
Objective
To evaluate the degree of agreement of dietary intake reported by the patient subject with the dietary intake reported by a respondent (a next-of-kin or a caregiver), collected by a validated Food Frequency Questionnaire (FFQ).
Methodology
126 adults, both sexes, the average age was 65.9 years for patients and 54.4 years for respondents. They were recruited from the General Practice Clinic at the Clinical Hospital of São Paulo (AGD-FMUSP). The agreement between the responses given by patients and respondents was assessed using Spearman, weighted Kappa and Bland Altman tests.
Results
The analysis for accuracy between responses (Spearman test) showed a moderate degree of agreement (0.31-0.39) for Energy, Total fat, Total Saturated Fatty Acids (SFA), Total Monounsaturated Fatty Acids (MUFA). Regarding food groups a moderate agreement was found for the majority of the foods (fruits (0.30), dairy products (0.50), natural juices (0.45), beans (0.48), butter/margarine (0.55), coffee (0.41) and soda (0.45), with the exception of vegetables (0.12) and rice (0.63). The ingestion differences did not exceeded the limit of the two standard deviations for the majority of the pairs (Bland Altman). A respondent subsample composed only of husband/wives (N = 36) revealed a moderate agreement concordance for most macronutrients studied (0.30 - 0.58), except polyunsaturated fats (0.25).
Conclusion
The results of this study show that, the FFQ may be used in cases where is impossible to get the answers directly from the patients.
Key words: Dietary assessment, food frequency questionnaire, food Intake, concordance, older adults
Introduction
There is an increasing awareness of the importance of maintaining the health of older people. Maintaining good health means better independence; safety and productivity in subsequent years, but the non-communicable diseases (NCDs) can decrease the quality of life of older people, and increase the workload of family members who are responsible for their care (1). Estimates predict that by 2020, the Americas will have 200 million older adults, almost double the number in 2006, and over half of them will be living in Latin America and the Caribbean (1). The costs of treating seniors with chronic conditions are high and growing, and these costs are endured by governments, families, and the elderly themselves. Out-of-pocket health costs are highest for people with chronic health conditions and functional impairment (2, 3).
Dietary habits are one of the important modifiable factors that can affect the maintenance of a healthy ageing (4). Several epidemiological studies that analyse the relationship between food intake and the development of diseases use a food frequency questionnaire (FFQ) as an instrument (5, 6). However, for studies involving participants in late life stages, subjects often experience difficulties in providing responses to questions due to physical or cognitive issues; in such cases, the subject's caregiver responds to the questionnaire. To our knowledge, the feasibility of obtaining reliable information of nutritional habits of patients using information gathered from relatives has not been hitherto defined.
Accordingly, the present study sought to analyse the level of agreement between dietary intake reported on an FFQ by the subjects themselves and by proxy respondents who were close companions of these subjects, as part of the “Health Aging” project of the São Paulo University Medical School.
Methods
The total convenience sample of 126 adults of both genders was studied, including 63 patients and 63 respondents who were close to these subjects. All participants were recruited in the waiting room in the services of the General Practice Clinic at the Clinical Hospital of São Paulo (AGD-FMUSP), that they were waiting for the medical appointment. Adults aged 60 or more who were waiting with a companion for medical consultations in various specialties were identified and invited to participate in the study.
The participants were recruited from March to December of 2014. Participant pairs (subject and companion) were approached by nutritionists and appropriately trained nutrition students who rotated among the patients and their respondents as they administered the questionnaire. The inclusion criteria specified that participants and their respondents, without cognitive impairment, were required to live together, share at least three meals per day for at least 5 days per week. The most frequent diseases were non-communicable chronic disease and autoimmune disease. Patients and their respondents were interviewed simultaneously in the separated space; so, one could not interpose in the other answers.
Table 1.
Means and standard deviations (SD) of daily intake of food groups (g/mL), Spearman’s rank and Kappa correlation, with FFQ research for total patients and his respondent, and between couples. São Paulo, Brazil, 2015
| Total’s Sample | Couples’s Sample | |||||||
|---|---|---|---|---|---|---|---|---|
| Food Groups (g/mL) | Patients | Respondent | Spearman | Kappa | Patients | Wife/Husband- | Spearman | Kappa |
| Mean (SD) | Mean (SD) | Coeficient | Mean (SD) | Mean (SD) | Coeficient | |||
| Vegetables | 128.6 (76.3) | 153.5 (85.5) | 0.12 | 0.13 | 114.8 (60.6) | 143.4 (89.6) | 0.40* | 0.39* |
| Fruits | 495.0 (307.1) | 497.9 (341.2) | 0.30* | 0.31* | 514.4 (298.6) | 509.5 (328.2) | 0.35 | 0.36 |
| Dairy Products | 253.5 (163.4) | 239.9 (169.9) | 0.50* | 0.48* | 259.9 (154.8) | 261.7 (162.3) | 0.56* | 0.57* |
| Nature Juice | 89.9 (196.8) | 100.2 (158.6) | 0.45* | 0.45* | 77.8 (170.2) | 110.6 (143.9) | 0.39 | 0.38* |
| Rice | 164.2 (76.2) | 170.8 (76.1) | 0.63* | 0.66* | 170.6 (73.6) | 179.1 (69.3) | 0.53* | 0.53* |
| Bean | 88.6 (54.9) | 104.6 (61.1) | 0.48* | 0.43* | 90.6 (52.9) | 110.2 (61.0) | 0.39 | 0.43* |
| Red Meat | 168.8 (90.8) | 179.6 (96.2) | 0.32* | 0.32* | 178.9 (78.1) | 182.2 (81.4) | 0.38 | 0.42* |
| Sausages | 6.3 (10.7) | 6.1 (10.5) | 0.30* | 0.25* | 6.7 (11.8) | 5.7 (8.5) | 0.53* | 0.49* |
| Butter/Margarine | 12.9 (9.6) | 12.3 (9.2) | 0.55* | 0.67* | 11.7 (8.7) | 13.1 (9.6) | 0.66* | 0.71* |
| Coffee | 163.8 (149.5) | 148.0 (133.4) | 0.41* | 0.44* | 162.5(136.4) | 166.0 (155.2) | 0.39 | 0.43* |
| Soft Drink |
61.1 (118.7) |
70.8 (136.4) |
0.45* |
0.51* |
70.1 (126.5) |
91.9 (153.5) |
0.53* |
0.58* |
Total Sample: N=126; Couples Sample: N=72;
p<0.01
A FFQ validated for the adult population of both genders in São Paulo was the instrument used to assess dietary intake (7, 8), referring to the year prior to the application of this instrument.
Frequencies of intake were expressed in terms of daily intake (in grams) for each food. For each evaluated item, this daily intake (in grams) was multiplied by the nutritional composition of the item to transform the intake into its corresponding energy and nutrient values. All statistical analyses were performed using Stata software (version 10, 2007, Stata Corp LP, College Station, Texas). A significance level of 5% was considered.
For this study, we defined as healthy foods the following items: vegetables, fruits, natural juices, dairy products, rice and beans. On the other hand, the following classes of foods were considered as unhealthy: red meat, sausage, coffee, margarine/ butter and soft drinks. This food groups were selected because they are often studied in the literature for their relationship to health (9). The macronutrients consisted of calories, protein, fibre, total fat, saturated fat and monounsaturated fat. Values adjusted by energy intake, using the residual method, were used to analyse nutrient intakes (5).
The agreement between the food intake responses obtained from the patients and their respondents was analysed using Spearman correlations (significance at 0.05) (10). Coefficient values of less than 0.29 and between 0.30 and 0.59 are regarded as weak and moderate correlations, respectively.
Weighted kappa was used to assess the precise agreement between responses that had been categorised into quartiles of the two samples (11). The values greater than 0.80 are regarded as optimal agreement and the values (0.60 to 0.79); (0.59 to 0.40); (0.39 to 0.20); and less than 0.19 were considered as indicative of good agreement, moderate agreement, fair agreement and poor agreement, respectively (12).
Bland-Altman was used to not only determine whether systematic differences exist between the intakes reported by patients and the intakes reported by respondents but also assess the level of agreement between these intakes (13). In this approach, the mean total difference indicates whether a response tends to overestimate or underestimate the range defined by two standard deviations.
Ethical Approval
The present study is part of the project “Correlation between abdominal visceral fat and thoracic with the degree of systemic atherosclerosis”, research of the Medical School/University of São Paulo, protocol number 412/11 from 09/28/2011”. This study comprised a separate sample of convenience, in order to validate the proposal to the application of instrument and methodology. All participants agreed to participate in the project and signed free and written informed consent forms (FICFs)
Financial Support
This study is funded from the National Council of Scientific and Technological Development (CNPq), and Foundation Support of São Paulo State Research (FAPESP), grant number (2008/57717-6).
Table 2.
Means and standard deviations (SD) of daily macronutrient intake and fat fractions (adjusted energy), Spearman’s rank and Kappa correlation, with FFQ research total patients and his respondent, and between Couples. São Paulo, Brazil, 2015
| Total’s Sample |
Couples’s Sample |
|||||||
|---|---|---|---|---|---|---|---|---|
| Nutrients | Patient Mean(SD) | Respondent Mean(SD) | C. Spearman | Kappa | Patients Mean(SD) | Wife/Husband Mean(SD) | C. Spearman | Kappa |
| Energy (Kcal) | 2502.5 | 2587.8 | 0.31 | 0.31* | 2466.2 | 2762.0 | 0.44* | 0.43* |
| (933.7) | (1045.5) | (713.5) | (1092.1) | |||||
| Total fat (g) | 88.9 (44.7) | 87.7 (36.5) | 0.36* | 0.36* | 85.0 (28.6) | 92.2 (35.7) | 0.50* | 0.49* |
| Carbohydrate (g) | 327.2 (113.9) | 342.2 (131.7) | 0.23 | 0.23 | 327.5 (101.7) | 362.0 (134.8) | 0.30 | 0.29 |
| Protein (g) | 109.5 (50.8) | 107.7 (39.0) | 0.26 | 0.25 | 107.8 (31.2) | 113.4 (35.9) | 0.36 | 0.36 |
| Fiber (g) | 33.5 (13.8) | 34.3 (15.0) | 0.21 | 0.21 | 33.0 (11.7) | 35.7 (15.3) | 0.33 | 0.35 |
| SFA (g) | 29.4 (17.5) | 28.0 (12.6) | 0.39* | 0.38* | 28.0 (11.2) | 29.9 (13.1) | 0.58* | 0.56* |
| MUFA (g) | 31.1 (17.1) | 30.7 (13.4) | 0.34* | 0.35* | 29.0 (9.8) | 31.9 (12.7) | 0.50* | 0.52* |
| PUFA (g) |
20.6 (7.8) |
21.2 (8.6) |
0.24 |
0.26 |
20.4 (6.9) |
22.3 (8.1) |
0.25 |
0.27 |
Total’s Sample: N=126; Couples’s Sample: N=72;
p<0.01
Discussion
The current study has shown an acceptable agreement for most food groups and nutrients selected in this analysis between dietary intake estimated by a validated FFQ reported by the subject, an older adult, and a proxy respondent. Several studies evaluated the agreement among various dietary assessment tools for adults, including the FFQ (14., 15., 16., 17.). However, the errors present in the measurements of the questionnaire depend on the respondent's memory and ability to recall and report the foods consumed (18). Knowing that error is inherent in food intake measurements, methodological strategies have been used such: the team involved in applying the questionnaire was composed by group of nutritionists with large experience in FFQ questionnaires, and adjustments by energy intake to analysed nutrient intakes (18).
This study compared the agreement between the food intake reported by patients and his/her intake reported by a proxy respondent, a caregiver or next-of-kin. A factor that motivated this study was the need to assess dietary habits among patients with a functional impairment (19, 20) that would prevent them from understanding questions or providing useful answers (21). The importance of being able to accurately assess food consumption in this late stage of life is currently acknowledged.
Although in this study the people interviewed have no cognitive impairment, for ageing studies specific strategies are required for the analysis and interpretation of the results (22, 23). One such strategy is the use of respondents who are close to study subjects. This approach can be necessary because the presence of comorbidities (including cognitive, functional or metabolic comorbidities) can impair or prevent elderly subjects' direct participation in a study; in these situations, an individual who is close to the patient must provide the requisite information, and the internal validity of the study must be verified (24., 25., 26.).
The results of the current study demonstrated that there was moderate agreement between the FFQ responses provided by respondents and the responses provided by the patient. Acceptable agreement values between two tools vary between 0.40 and 0.70. Evaluations of the coefficients (Spearman and Kappa) obtained for the food groups indicated that 64% of these coefficients presented values > 0.40. By another, when analysing the data of the couples we observed a better degree of agreement among the food groups analysed. Literature comparing subject´s responses and next of kin is scarce. The results of the present investigation show that is possible to obtain useful information from relatives of dietary habits for most of food groups and nutrients considered in the present study.
We found few studies that use caregivers/spouse's responses for the evaluation of dietary practices (27., 28., 29.). A case control study of lung cancer in New Mexico, where the wives were asked to report their husbands' usual diet, showed small agreement with Kappa values around 0.20 for any food (27). In the Chinese study that evaluated the degree of concordance among couples about lifestyle and diet, obtained better results to those found in our study, where more than 70% of the couples had kappa statistics exceeding 0.61. Food items exhibiting high agreement between the spouses include rice, apples, tomatoes, and pork chops (28). Another study from Australia demonstrated that the levels of agreement (kappa statistics) between couples regarding the consumption of vegetables and milk were 0.35 and 0.77, respectively (29). Thus, the present study obtained levels of agreement within the same order of magnitude of those reported in previous studies, reinforcing the possibility of evaluating feeding habits on the basis of questionnaires applied to relatives.
With respect to food groups, a study conducted in south Brazil (16) found Kappa values close to the values obtained in the current study for fruits and red meat. In comparison with the current study, this prior study found higher values for vegetables and lower values for the remaining food groups (dairy products, fats, legumes, sausage and cereals). Higher Spearman correlation coefficients were obtained in this prior study than in the present study.
Figure 1.
Bland Altman plots related to macronutrients studied. SP, 2015
Our results indicated that regardless of the testing approach, certain variables exhibited similar coefficients. Among the examined foods, rice featured the highest correlations and vegetables had the worst correlations. The high level of agreement in rice-related responses may relate to the fact that rice is the food consumed by the highest percentage of Brazilians (84%), according to data from the 2008-2009 Consumer Expenditure Survey; thus, consumption of rice was easily remembered by participants when the FFQ was administered. In contrast, the difficulties associated with reporting vegetable consumption during the prior year may relate to the fact that vegetables are the food group that is least consumed by the Brazilian population (16%) (30).
There are also limitations to consider. Validation and repeatability studies of FFQ showed that overestimation/ underestimation of certain foods could occur due to inaccurate subject reporting of frequency and/or mean portion size (31). In addition, educational level may be a significant source of misinformation mainly at the lower levels of schooling (31) and, in our case; population's educational level was not evaluated. Finally, the relatively small sample size of the present study may affect the statistical power of the analysis, but reduced sample sizes are common in this type of study (31, 32).
Conclusion
The results of the present study indicated that in cases for which the FFQ cannot be directly administered to a patient to investigate his/her eating habits the subject's partner/respondent may provide estimators of feeding habits.
Acknowledgements
The authors would like to thank the students Larissa Vitoriano Moraes and Camilla Vieira Floriano for their collaboration in data collection and data entry.
Conflict of interest
The authors declare that they have no conflicts of interest.
Contributions
Saldiva SRDM contributed to the design, coordination and analysis of the study; the interpretation of study results; and the writing of the paper. Bassani L, Castro ALS, Gonçalves IB and Sales CRO contributed to the design, data collection and data entry of the study; study-related discussions; and the writing of the paper. Marchioni DML contributed to the conception of the study; discussed and critically reviewed study content; and approved the final version of the paper.
References
- 1.PAHO-Pan American Health Organization 2008. Basic Indicators of the Pan American Health Organization, data available from 30 countries out of 46 in the Americas.
- 2.Kass-Bartelmess B.L., Preventing Disability in the Elderly With Chronic Disease, 2002
- 3.Crystal S., Johnson R.W., Harman J., Sambamoorthi U., Kumar R. Out-of-pocket health care costs among older Americans. J Gerontol B Psychol Sci Soc Sci. 2000;55(1):S51–S62. doi: 10.1093/geronb/55.1.s51. 10.1093/geronb/55.1.S51 PubMed PMID: 10728130. [DOI] [PubMed] [Google Scholar]
- 4.Mathers J.C. Nutrition and ageing: knowledge, gaps and research priorities. Proc Nutr Soc. 2013;72(2):246–250. doi: 10.1017/S0029665112003023. 10.1017/S0029665112003023 PubMed PMID: 23312402. [DOI] [PubMed] [Google Scholar]
- 5.Anderson A.L., Harris T.B., Tylavsky F.A., Perry S.E., Houston D.K., Lee J.S., Sahyoun N.R. Dietary patterns, insulin sensitivity and inflammation in older adults. European Journal of Clinical Nutrition. 2012;66(1):18–24. doi: 10.1038/ejcn.2011.162. 10.1038/ejcn.2011.162 PubMed PMID: 21915138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Corley J., Starr J.M., McNeill G., Deary I.J. Do dietary patterns influence cognitive function in old age? International Psychogeriatrics. 2013;25:1393–1407. doi: 10.1017/S1041610213000793. 10.1017/S1041610213000793 PubMed PMID: 23732046. [DOI] [PubMed] [Google Scholar]
- 7.Fisberg R.M., Colucci A.C.A., Morimoto J.M., Marchioni D.M.L. Food frequency questionnaire for adults from a population-based study questionnaire for adults from a population-based study. Rev Saúde Pública. 2008;42(3):550–554. doi: 10.1590/s0034-89102008005000020. 10.1590/S0034-89102008005000020 PubMed PMID: 18425293. [DOI] [PubMed] [Google Scholar]
- 8.Selem S.S., Carvalho A.M., Verly-Junior E., Carlos J.V., Teixeira J.A., Marchioni D.M., Fisberg R.M. Validity and reproducibility of a food frequency questionnaire for adults of São Paulo, Brazil. Rev Bras Epidemiol. 2014;17(4):852–859. doi: 10.1590/1809-4503201400040005. 10.1590/1809-4503201400040005 PubMed PMID: 25388486. [DOI] [PubMed] [Google Scholar]
- 9.Vrieling A., Buck K., Seibold P., Heinz J., Obi N., Flesch-Janys D., Chang-Claude J. Dietary patterns and survival in German postmenopausal breast cancer survivors. British Journal of Cancer. 2012;108:188–192. doi: 10.1038/bjc.2012.521. 10.1038/bjc.2012.521 PubMed PMID: 23169282, PMCID 3553521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brown H., Benedetti J. Sample behavior of tests of correlation in two ways contingency tables. J Am Estat Assoc. 1977;72:309–315. [Google Scholar]
- 11.Szklo M., Nieto F.J., Epidemiology: beyond the basics, 2000
- 12.Landis J.R., Koch G.G. An Application of Hierarchical Kappa-type Statistics in the Assessment of Majority Agreement among Multiple Observers. Biometrics. 1977;33(2):363–374. 10.2307/2529786 PubMed PMID: 884196. [PubMed] [Google Scholar]
- 13.Bland G.M., Altman D.G. Statistical method for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–310. 10.1016/S0140-6736(86)90837-8 PubMed PMID: 2868172. [PubMed] [Google Scholar]
- 14.Kaaks R., Ferrari P. Dietary intake assessment in epidemiology: Can we know what we are measuring? Ann. Epidemiol. 2006;16:377–380. doi: 10.1016/j.annepidem.2005.06.057. 10.1016/j.annepidem.2005.06.057 [DOI] [PubMed] [Google Scholar]
- 15.Zanolla A.F., Olinto M.T.A., Henn R.L., Wahrlich V., Anjos L.A. Assessment of reproducibility and validity of a food frequency questionnaire in a sample of adults living in Porto Alegre, Rio Grande do Sul State, Brazil. Cad Saúde Pública. 2009;25:840–848. doi: 10.1590/s0102-311x2009000400015. 10.1590/S0102-311X2009000400015 PubMed PMID: 19347210. [DOI] [PubMed] [Google Scholar]
- 16.Pope S.K., Kritchevsky S.B., Morris M.C., Block G., Tylavsky F.A., Lee J.S., Stewart S., Harris T., Rubin S.M., Simonsick E.M. Cognitive ability is associated with suspected reporting errors on food frequency questionnaires. J Nutr Health Aging. 2007;11(1):55–58. PubMed PMID: 17315081. [PubMed] [Google Scholar]
- 17.Pakseresht M., Sharma S. Validation of a culturally appropriate quantitative food frequency questionnaire for Inuvialuit population in the Northwest Territories, Canada. J Hum Nutr Diet. 2010;23(1):75–82. doi: 10.1111/j.1365-277X.2010.01105.x. 10.1111/j.1365-277X.2010.01105.x PubMed PMID: 21158965. [DOI] [PubMed] [Google Scholar]
- 18.Willett W.C. Nutritional Epidemiology. Oxford University Press; Oxford: 1998. 10.1093/acprof:oso/9780195122978.001.0001 [Google Scholar]
- 19.Rosner B., Willett W.C., Spiegelman D. Correction of logistic regression relative risk estimates and confidence intervals for systematic within-person measurement error. Stat. Med. 1989;8:1051–1069. doi: 10.1002/sim.4780080905. 10.1002/sim.4780080905 PubMed PMID: 2799131. [DOI] [PubMed] [Google Scholar]
- 20.Kiefte-de Jong J.C., Mathers J.C., Franco O.H. Nutrition and healthy ageing: the key ingredients. Proceedings of the Nutrition Society. 2014;73:249–259. doi: 10.1017/S0029665113003881. 10.1017/S0029665113003881 PubMed PMID: 24503212. [DOI] [PubMed] [Google Scholar]
- 21.Eysteinsdottir T., Thorsdottir I., Gunnarsdottir I., Steingrimsdottir L., Assessing validity of a short food frequency questionnaire on present dietary intake of elderly Icelanders. Nutr J, 2012, 11–12 [DOI] [PMC free article] [PubMed]
- 22.McNeill G., Winter J., Jia X. Diet and cognitive function in later life: a challenge for nutrition epidemiology. Eur J Clin Nutr. 2009;63(1):S33–S37. doi: 10.1038/ejcn.2008.62. 10.1038/ejcn.2008.62 PubMed PMID: 19190641. [DOI] [PubMed] [Google Scholar]
- 23.Palmer E.C., David A.S., Fleming S.M. Effects of age on metacognitive efficiency. Conscious Cogn. 2014;28:151–160. doi: 10.1016/j.concog.2014.06.007. 10.1016/j.concog.2014.06.007 PubMed PMID: 25064692, PMCID 4154452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wallace R.B., Woolson R.F. The Epidemiologic study of the elderly. Oxford University Press; New York: 1992. [Google Scholar]
- 25.Ebrahim S. In: Epidemiology in old age. Ebrahim S., Kalache A., editors. BMJ Publishing Group; London: 1996. Principles of epidemiology in old age; pp. 12–21. [Google Scholar]
- 26.Vries J.H., Groot L.C., van Staveren W.A. Dietary assessment in elderly people: experiences gained from studies in the Netherlands. Eur J Clin Nutr. 2009;63(1):S69–S74. doi: 10.1038/ejcn.2008.68. 10.1038/ejcn.2008.68 PubMed PMID: 19190649. [DOI] [PubMed] [Google Scholar]
- 27.Nelson L.M., Longstreth W.T., Koepsell T.D., Van Belle G. Proxy respondents in epidemiologic research. Epidemiologic Review. 1990;12:71–86. doi: 10.1093/oxfordjournals.epirev.a036063. 10.1093/oxfordjournals.epirev.a036063 [DOI] [PubMed] [Google Scholar]
- 28.Lerchen M.L., Samet J.M. An assessment of the validity of questionnaire responses provided by a surviving spouse. Am J Epidemiol. 1986;123(3):481–489. doi: 10.1093/oxfordjournals.aje.a114263. 10.1093/oxfordjournals.aje.a114263 PubMed PMID: 3946394. [DOI] [PubMed] [Google Scholar]
- 29.Liang W., Binns C., Lee A.H., Huang R., Hu D. The reliability of dietary and lifestyle information obtained from spouses in an elderly chinese population. Asia Pac J Public Health. 2008;20(2):87–93. doi: 10.1177/1010539507311183. 10.1177/1010539507311183 PubMed PMID: 19124302. [DOI] [PubMed] [Google Scholar]
- 30.Parekh S., King D., Owen N., Jamrozik K. Spousal concordance and reliability of the ‘Prudence Score’ as a summary of diet and lifestyle. Australian and New Zealand Journal of Public Health. 2009;33(4):320–324. doi: 10.1111/j.1753-6405.2009.00402.x. 10.1111/j.1753-6405.2009.00402.x PubMed PMID: 19689591. [DOI] [PubMed] [Google Scholar]
- 31.POF 2008-2009. IBGE, Diretoria de Pesquisas, Coordenação e Trabalho e Rendimento, Pesquisa de Orçamentos Familiares 2008-2009. http://www.ibge.gov.br/home/estatistica/populacao/condicaodevida/pof/2008_2009_analise_consumo/tabelas_pdf/tab_1_2.pdf.
- 32.Barrat E., Aubineau N., Maillot M., Derbord, Barthes P., Lescuyer J.-F., Peltier S. L., Repeatability and relative validity of a quantitative foodfrequency questionnaire among French adults. Food & Nutrition Research, 2012, 56 [DOI] [PMC free article] [PubMed]
Uncited references
- 33.Turconi G., Bazzano R., Roggi C., Cena H. Reliability and relative validity of a quantitative food-frequency questionnaire for use among adults in Italian population. Int J Food Sci Nutr. 2010;61(8):846–862. doi: 10.3109/09637486.2010.495329. 10.3109/09637486.2010.495329 PubMed PMID: 20666692. [DOI] [PubMed] [Google Scholar]