Brazilian version of a lifestyle questionnaire: translation and validation for young adults

Abstracts

FUNDAMENTO: Estilos de vida inadequados têm sido as principais causas de morte no mundo. O questionário "Estilo de vida fantástico" tem por finalidade auxiliar médicos que trabalham com prevenção, a fim de estes possam melhor conhecer e medir o estilo de vida dos seus pacientes. OBJETIVO: Traduzir e validar o questionário "Estilo de vida fantástico" para adultos jovens. MÉTODOS: O processo de tradução e adaptação envolveu quatro etapas: 1) tradução, 2) correção e adaptação semântica por especialistas da área (juízes), 3) avaliação da clareza do conteúdo e 4) avaliação por amostra da população-alvo. Análise estatística: para a consistência interna, utilizou-se o alfa de Cronbach; a consistência externa foi verificada pela correlação intraclasse (R); para a validade de construto, usou-se o índice Kappa. RESULTADOS: O alfa de Cronbach que avalia as correlações entre os itens e o total foi de 0,69, o que é considerado razoável para um instrumento desenhado para medir uma variável latente que somente pode ser estimada e não medida. Quando agrupados os itens em domínios, verificou-se que todos contribuem de maneira semelhante para a estabilidade do instrumento (α = 0,60). A reprodutibilidade avaliada pela correlação intraclasse foi alta (R = 0,92; p = 0,2). A validade de construto testada pela capacidade classificatória do instrumento em quatro e três categorias foi de 75% e 80,7%, com um índice Kappa de 0,58 e 0,7, respectivamente, o que evidencia uma ótima capacidade discriminatória. CONCLUSÃO: O questionário "Estilo de vida fantástico" possui uma consistência interna e externa adequada para a avaliação do estilo de vida de adultos jovens. Recomenda-se a utilização do questionário na atenção primária e em estudos epidemiológicos.

Estilo de vida; questionários; qualidade de vida; saúde do adulto atividade motora

ORIGINAL ARTICLES

EPIDEMIOLOGY

Brazilian version of a Lifestyle Questionnaire: translation and validation for young adults

Ciro Romélio Rodriguez Añez^I; Rodrigo Siqueira Reis^I; Edio Luiz Petroski^II

^IPontifícia Universidade Católica do Paraná, Curitiba, PR - Brazil

^IIUniversidade Federal de Santa Catarina, Florianópolis, SC - Brazil

Mailing address

SUMMARY

BACKGROUND: Inappropriate lifestyles have been responsible for the leading causes of mortality. The purpose of the Fantastic Lifestyle Questionnaire is to help physicians involved in health prevention to identify and assess their patients' lifestyles.

OBJECTIVE: To translate and validate the Fantastic Lifestyle Questionnaire for use with young adults.

METHODS: The process of translation into Portuguese and validation consisted of four steps: 1) translation; 2) correction and semantic adaptation by professional experts (judges); 3) content validation; and 4) a final evaluation by the target population. The statistical analyses used were Cronbach's alpha to measure internal consistency reliability, intra-class correlation coefficient (R) for external consistency, and the Kappa index to test construct validity.

RESULTS: Cronbach's alpha coefficient measured the correlation between items and the total was 0.69, which is considered reasonable for an instrument designed to evaluate a latent variable that can only be estimated and not measured. When the items were grouped into domains, it was possible to verify that they all contributed equally to the stability of the questionnaire (α=0.60). The external consistency evaluated by intra-class correlation was R=0.92 (p=0.2), which demonstrates a high degree of reproducibility; the construct validity of 4 and 3 categories was 75 and 80.7% with Kappa indices of 0.58 and 0.7, respectively, which show an excellent discriminatory classificatory capacity.

CONCLUSION: The Fantastic Lifestyle Questionnaire has an adequate internal and external consistency for evaluating young adult lifestyles, and it can be recommended for primary care and epidemiological studies.

Introduction

There are many determinants of health, and they can be grouped into five categories¹ encompassing the elements that keep people healthy. The first category is the socioeconomic environment, which includes salary, having a job, social status, education, and social factors in the work place. The second category is the natural physical environment, or that constructed by man, such as climate and housing. The third takes into consideration personal attitudes, i.e., those behaviors that bring benefits or pose risks to human health. The fourth category represents individual capacities such as genetics, physiology, personal competence, sense of control, and the individual's own skills. The last group encompasses health promotion, maintenance, and recovery services¹.

In the past decades, the effects of individual behavior on health have been well established^2,3. Although there is positive evidence that lifestyle and physical activity play a role in health, a large percentage of people do not lead healthy lifestyles. The rates of physical inactivity are high, and degenerative chronic diseases are still the leading causes of death¹.

Lifestyle is characterized by identifiable behavioral patterns that can have a marked effect on an individual's health, and it is related to several aspects that reflect the attitudes, values, and opportunities in a person's life⁴.

Physical activity and eating patterns are two lifestyle elements that play a significant role in promoting health and preventing diseases⁵. Other lifestyle elements such as avoiding cigarette smoking, having good relationships with family and friends, avoiding alcohol consumption, practicing safe sex, controlling stress, besides maintaining an optimistic and positive outlook on life are also important for promoting good health and well-being¹.

Generally speaking, instruments that use information provided by individuals (questionnaires, interviews, and diaries) are inexpensive, easy to apply, and can be used to evaluate large numbers of people, making them convenient for epidemiological studies as they yield quick answers for topics that would be time-consuming if evaluated any other way.

The objective of the Fantastic Lifestyle Questionnaire is to measure the main elements that characterize a healthy lifestyle¹. The questionnaire is an ancillary tool used by primary care physicians to know and assess their patients' lifestyles⁵. The instrument consists of a questionnaire with 25 closed-end questions across nine domains of the physical, psychological, and social components of lifestyle. There are several versions of the instrument in English and Spanish suitable for application with students^6,7, workers⁸, family clinic patients⁹, hypertensive patients¹⁰, and patients with type-2 diabetes⁵.

However, the appropriate use of a questionnaire requires that it first be translated and evaluated as to its adequacy with the culture of the people to be assessed. It is also necessary to test its psychometric characteristics in order to verify if it has validity allowing conclusions to be drawn from the results^6,11,12.

The objective of the study was to translate and validate the self-applied Fantastic Lifestyle Questionnaire to assess the lifestyles of young adults.

Methods

The Fantastic Lifestyle Questionnaire is a generic instrument developed in 1984⁸, by Wilson and Ciliska of the Department of Family Medicine at the McMaster University in Canada, aimed at helping physicians who deal with disease prevention to know and assess their patients' lifestyles.

The form used in this study is the one suggested in 1998 by the Canadian Society for Exercise Physiology and constitutes part of the standardized battery of tests known as the Canadian Physical Activity Fitness & Lifestyle Appraisal. The questionnaire covers a wide range of issues that have a subtle but powerful influence on health. The lifestyle survey supplements the assessment of health-related physical fitness and allows a more comprehensive view of the individual¹.

The acronym FANTASTIC represents the first letters of the nine domains (in English) in which the 25 questions or items are distributed:

F= Family and Friends

A= Activity (Physical activity)

N= Nutrition

T= Tobacco & Toxics

A= Alcohol Intake

S= Sleep, Seatbelts, Stress, and Safe sex

T= Type of behavior (Type A or Type B behavior pattern)

I= Insight

C= Career (Work, satisfaction with profession)

The Fantastic Lifestyle Questionnaire is a self-administered instrument that addresses the behavior of individuals during the preceding month. Its results allow the determination of the association between lifestyle and health. The instrument has 25 questions across the 9 following domains: 1) family and friends, 2) physical activity, 3) nutrition, 4) tobacco and toxics, 5) alcohol intake, 6) sleep, seat belt, stress, and safe sex, 7) behavior patterns, 8) insight, and 9) career.

The questions are distributed on a Likert scale; 23 of them have multiple-choice questions (five answers) and two are dichotomous. The alternatives are presented in columns in order to facilitate coding; the left-hand column is always the one with the lowest value or that bears the least relationship with a healthy lifestyle. Questions are coded by points as follows: zero for the first column, 1 for the second, 2 for the third, 3 for the fourth, and 4 for the fifth column. For questions with just two alternative answers, the score is zero for the first column and 4 points for the last column.

The sum of all points yields a total score that classifies individuals in five categories, as follows: "Excellent" (85 to 100 points), "Very good" (70 to 84 points), "Good" (55 to 69 points), "Regular" (35 to 54 points), and "Needing improvement" (0 to 34 points).

It is desirable that individuals completing the questionnaire be classified as "Good." The lower the score, the greater the need for change. Generally speaking, results can be interpreted as follows: "Excellent" indicates that the individual's lifestyle represents an optimal influence for health; "Very good" indicates that the lifestyle represents an adequate influence for health; "Good" indicates that the lifestyle represents many benefits for health; "Regular" indicates that the lifestyle represents some benefit for health, although it also poses risks; "Needing improvement" indicates that the individual's lifestyle poses many risk factors.

Translation

The questionnaire was translated by the authors and subjected to review by four specialists, all of them knowledgeable in the area and fluent in the English language, who were in charge of evaluating the translation and cross-cultural adaptation of the terms into Portuguese. To check for clarity, the translated questionnaire was applied to 18 subjects who were asked to point out if the question was not clear or if they had difficulties identifying a response option.

Attachment we shows the final version of the questionnaire after the inclusion of suggestions by the specialists and assessment for clarity and the original questionnaire.

Validation

The sample was non-probabilistic and consisted of 62 young adults of both genders, recruited among graduate and postgraduate students at the Universidade Federal de Santa Catarina, who voluntarily agreed to participate.

Data collection

To check for the instrument's stability, two data collections were conducted with a one-week interval.

The internal consistency of the questionnaire was calculated using the item-total correlation and Cronbach's alpha. Intra-class correlations (R) among items, among domains, and the overall score were used to evaluate reproducibility. The construct validity of the instrument was also examined as to its capacity of reclassification between two applications, concordance rate, and Kappa index (a coefficient of agreement for nominal scales).

This study was approved by the Human Research Ethics Committee of the Universidade Federal Santa Catarina (UFSC), Project No. 059-2003.

Statistical analysis

The instrument's internal consistency was assessed by Cronbach's alpha, the external consistency by intra-class correlation (R), and the construct validity by Kappa test.

The internal consistency of the questionnaire was measured by Cronbach's alpha using the sum of all items of the questionnaire (25) as the representative value of the latent "lifestyle" variable. The test of Cronbach's alpha is sensitive to the number of items in the instrument and to the size of the sample, and is the result of the correlations between the items and the total score (item-total correlation); it can be interpreted as the mean value of all possible split-half reliability estimates. To calculate Cronbach's alpha, it is necessary that all items be coded as variables with numerical values and preferably have a normal distribution since alpha is sensitive to variances of the set.

The reproducibility of the questionnaire, external consistency, was assessed using intra-class correlation (R). The reproducibility was tested item-by-item, domain-by-domain, and by the FLS overall score (pre- and post- Fantastic Lifestyle).The construct validity was tested by examining the instrument's classification capacity in two ways: first, the concordance between the first and the second assessments taking into consideration the scores obtained in four categories (Regular, Good, Very Good, and Excellent); second, considering the scores obtained in three categories (Low, Medium and High) divided by the first and second terciles. For analysis of the classification in four and three groups, data were placed on contingency tables and the concordance rate (CR) and Kappa index were considered.

Results

The descriptive characteristics of the sample are displayed on Table 1. Of the 62 individuals evaluated, 34 (54.8%) were male and 28 (45.2%) were female. The mean age of the sample subjects was 21.3 years (SD 3.5).

As to lifestyle, no single individual scored less than 34 points; one (1.6%) scored between 35 and 54; 13 (21%) scored between 55 and 69; 38 (61.3%) scored between 70 and 84, and 10 (16.1%) scored between 85 and 100.

The Kolmogorov-Smirnov test for normality of the FLS (Fantastic Lifestyle) variable indicated a normal distribution (p=0.20), whereas the Shapiro-Wilk test was p=0.91. This allows testing the instrument for internal and external consistency as per the selected procedures. The instrument's internal consistency assessed by Cronbach's alpha was 0.69.

Table 1 shows the mean score per item, the item-total score correlation, Cronbach's alpha, and the intra-class correlation (R) obtained with the FLS questionnaire applied to young adults.

Item 7 (body image) displayed a low item-total correlation, and two other items, number 4 (moderate physical activity) and 9 (use of toxics), yielded a negative item-total correlation. Nevertheless, the exclusion of any of these items would not raise the alpha value to an extent that would justify any elimination.

When items were grouped into domains (family and friends, physical activity, nutrition, use of tobacco and toxics, alcohol intake, sleep, seat belt, stress, and safe sex, type of behavior, insight, and career), the alpha value was 0.60. Although the alpha value had diminished, virtually all domains equally contribute to the stability of the instrument (Table 2).

The reproducibility of the instrument was tested item-by-item, domain-by-domain, and taking into consideration the overall score. The intra-class correlation values are displayed on Table 2. The intra-class correlation between the pre- and post-test total score was R=0.92 (p=0.200), which corroborates the high reproducibility of the instrument.

The construct validity was tested by examining the instrument's classification capacity in two ways; first, by assessing the concordance between the first and the second assessments taking into consideration the scores obtained by each individual during pre- and post-test recoded in four categories (Regular, Good, Very Good, and Excellent); second, by coding the scores obtained in three categories (Low, Medium, and High) divided by the first and second terciles.

When four classification categories are considered, the concordance rate (CR) for classification and reclassification of subjects evaluated in the first and second application of the instrument was 75%. Five individuals had a negative classification, that is, they descended one level in the ordinal classification of the first evaluation, and 10 individuals had positive classifications, i.e., they ascended one level in the ordinal classification of the first evaluation. The concordance rate for nominal scales (Kappa) was 0.58

The concordance of lifestyle classifications between two applications of the Fantastic Lifestyle Questionnaire indicated that the classification ratio in three groups was 80.7%, whereas the concordance rate (Kappa) was 0.70.

Discussion

Measuring the lifestyle construct is a challenging task due to the multiple dimensions involved and the difficulties posed by the attempt to perform a direct and objective assessment.

The Fantastic Lifestyle Questionnaire was incorporated into the Canadian Physical Activity Fitness & Lifestyle Appraisal Plan in 1996 with the objective of expanding the reach of the protocol with the inclusion of measurements of the main elements that characterize a healthy lifestyle¹. Several studies have concluded that the Fantastic Lifestyle instrument has apparent validity, content validity, and a good level of consistency for determining the lifestyle of healthy subjects^8,9, family clinic patients¹², hypertensive patients¹³, and patients with type-2 diabetes^7,14.

The instrument's internal consistency, measured by Cronbach's alpha, was 0.69, a value lower than that obtained by Rodriguez-Moctezuma et al⁷ in type-2 diabetes patients, which was 0.80. However, the reproducibility obtained in the present study, assessed by the test-retest method, was R=0.92 (p=0.200), whereas in Rodriguez-Moctezuma et al⁷ study it was 0.84.

According to Hill and Hill¹¹, an alpha value between 0.7 and 0.8 can be considered reasonable; the instrument's internal validity was designed to measure a latent variable that can only be estimated and not measured. To assess reliability it is necessary to use a set of empirical data drawn from a sample of subjects. Since there are no perfect samples, the reliability coefficients, as is the case with all descriptive statistics, vary from sample to sample and must be considered as estimates only. An alpha value greater than 0.5 has been considered as acceptable in questionnaire validation studies¹².

Although some questions yielded a low item-total correlation, as is the case with questions 4, 7 and 9, their elimination would not increase the instrument's alpha value; therefore, there is no justification to do this; moreover, it would deprive the instrument of its characteristics.

The analysis of questions grouped into domains (family and friends, physical activity, nutrition, tobacco and toxics, alcohol intake, sleep, seat belt, stress, and safe sex, behavior type, insight, and career) shows that all domains equally contribute to the instrument's stability. The inter-domain alpha value was 0.60 (Table 2).

The instrument's reproducibility was tested item-by-item, domain-by-domain, and taking into consideration the total score, and the result (R=0.92) confirms that the reproducibility was optimal for use with young adults. "R" represents the level of association between pre-test scores and post-test scores.

When four categories of classification are considered (Regular, Good, Very Good and Excellent), the concordance rate (CR) for classification and reclassification of subjects evaluated was 75%. Five individuals had negative classifications, i.e. they descended one level in the ordinal classification of the first evaluation, and 10 individuals had positive classifications, i.e., they ascended one level in the ordinal classification of the first evaluation. The concordance rate for nominal scales (Kappa) was 0.58.

The classification ratio in three groups was 80.7% and the concordance rate (Kappa) was 0.7. A Kappa (a coefficient of concordance for nominal scales) value between 0.5 and 0.75 can be considered good, and a result above 0.75 is excellent.

The results of this study suggest that the instrument has a good level of classification capacity. That is, if there are no changes in the variables that determine the latent lifestyle variable, the individuals will be appropriately reclassified. This is especially important for intervention programs aimed at promoting lifestyle changes as a way to improve health and quality of life. Positive changes in the classification will indicate that the program is attaining the expected results.

It is important to remember, that the way the questionnaire was applied and validated, it was meant for young adults with no apparent health problems and university students, and this is the major limitation of the study. However, in the Canadian Physical Activity Fitness & Lifestyle Appraisal Plan, the use of the questionnaire is suggested for use from 15 years of age on¹.

The repercussion of this study is the validation of an instrument that aims to assess the main elements that characterize a healthy lifestyle. The Fantastic Lifestyle Questionnaire is a tool that helps primary healthcare professionals to get to know their patients and provide guidance for introducing lifestyle changes.

Conclusion

The Portuguese version of the Fantastic Lifestyle instrument has met all criteria of internal and external consistency, and construct validity. There is evidence that the questionnaire is an adequate tool for assessing the lifestyles of young adults and should be used in primary care and epidemiological studies.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

There were no external funding sources for this study.

Study Association

This article is part of the thesis of doctoral submitted by Ciro Romélio Rodriguez Añez, from Universidade Federal de Santa Catarina.

References

• 1. CSEF - Canadian Society for Exercise Physiology. The Canadian Physical Activity, Fitness and Lifestyle Appraisal: CSEP's guide to health active living. 2nd ed. Ottawa: CSEF, 2003.
• 2. USDHHS - U.S. Department of Health and Human Services. Physical Activity and Health: a Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Center for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996.
• 3. U.S. Department of Health and Human Services. Health-related quality of life and activity limitation - eight states. MMWR. 1998; 47 (7): 135-40.
• 4. World Health Organization (WHO). Health promotion glossary. Geneva; 1998.
• 5. Blair SN, Horton E, Leon AS, Lee I-MIN, Drinkwater BL, Dishman RK, et al. Physical activity, nutrition, and chronic disease. Med Sci Sports Exerc. 1996; 28 (3): 335-49.
• 6. Fleck MP, Fachel O, Louzada S, Xavier M, Chachamovich E, Vieira G, et al. Desenvolvimento da versão em português do instrumento de avaliação de qualidade de vida da OMS (WHOQOL-100). Rev Bras Psiquiatr. 1999; 21 (1):19-28.
• 7. Rodriguez-Moctezuma R, López-Carmona JM, Munguia Miranda C, Hernández-Santiago JL, Bermúdez-Martínez M. Validez y consistencia del instrumento "FANTASTIC" para medir estilo de vida en diabéticos. Rev Med IMSS. 2003; 41 (3): 211-20.
• 8. Wilson DM, Ciliska D. Lifestyle assessment: testing the FANTASTIC instrument. Can Fam Physician. 1984; 30: 1863-6.
• 9. Sharratt JK, Ylamko V, Smith DM, Howell ML, Davenport L. FANTASTIC lifestyle survey of University of Waterloo employees. Can Fam Physician. 1984; 30: 1869-72.
• 10. Kason Y, Ylamko V. FANTASTIC life-style assessment: part 5: measuring lifestyle in family practice. Can Fam Physician. 1984; 30: 2379-83.
• 11. Hill MM, Hill A. Investigação por questionário. Lisboa: Silabo, 2000.
• 12. Gutierrez BAO. Tradução para o português e descrição do processo de validação do "Seattle Obstrutiva Lung Disease Questionnaire". [Dissertação de Mestrado]. São Paulo: Escola de Enfermagem; Universidade de São Paulo; 2000.
• 13. López-Carmona JM, Rodríguez-Moctezuma JR, Munguia Miranda C, Hernándzez-Santiago JL, Casas de La Torre E. Validez y fiabilidad del instrumento FANTASTIC para medir el estilo de vida de pacientes mexicanos con hipertensión arterial. Aten Primaria. 2000; 26 (8): 542-9.
• 14. López-Carmona JM, Ariza-Andraca CR, Rodríguez-Moctezuma JR, Monguíra-Miranda C. Construcción y validación inicial de un instrumento para medir el estilo de vida en pacientes con diabetes mellitas tipo 2. Salud Pública Mex. 2003; 45 (4): 259-68.

Publication Dates

• Publication in this collection
  08 Aug 2008
• Date of issue
  Aug 2008

History

• Received
  09 Nov 2008
• Reviewed
  08 Jan 2008
• Accepted
  08 Jan 2008