The Value of Questionnaires in Assessing Physical Activity, Fitness, and Quality of Life

Abstract

The measurement of functional cardiac capacity by means of ergospirometry, i.e., oxygen consumption and its kinetics, yields an accurate estimate of the ability of patients to achieve functional activity. After an initial “familiarization” test, ergospirometry is not only accurate but also reproducible [1,2]. However, it is not a simple task to perform an exercise functional capacity test. It requires advanced apparatus utilization and time, and it is costly. Moreover, a certain percentage of patients cannot perform physical exercise due to heart failure. According to our experience, 5 out of 40 patients with clinical severe heart failure and an ejection fraction below 30% were not able to attain an exercise level that could be adequately evaluated [3]. To circumvent these dif~culties, questionnaires that can yield a reliable estimation of physical activity have been developed during the last 20 years. The ~rst questionnaire of its kind was the Minnesota Leisure Time Physical Activity (LTPA) questionnaire, developed by Taylor et al. in 1978 [4]. Here it must be stressed that most questionnaires assess leisure time, including sports, occupation, and home and household activity. The time of recall of these questionnaires is from the past week to the past 12 months. Time frames of 1 to 3 days can also be used [5]. The questionnaires are either interviewer administered or self-administered, usually with supervision. In a very recent supplement of Medicine and Science in Sports and Exercise, directions are given for interviewers [5]. These directions stress that during an interview, special attention must be given to limiting any bias and preventing the interview from becoming too cumbersome. The need for skillful interrogation in order to assess a whole year’s average activity is especially emphasized, as is the need for establishing rapport with the person being interviewed. Also, the need to avoid unnecessary details and as the importance of clarity and accuracy are given special emphasis. A detailed account of how to administer the questionnaires has recently been published [5]. Poor quality can yield unreliable results and confuse meaningful correlations. In a methodologic critique of previous studies, it was stressed that only 20% of estimates of physical activity were good; 40% were satisfactory, and 40% were judged as unsatisfactory. The improvement in quality from unsatisfactory to good led to an increase in statistically signi~cant associations among studies from 50% to 88% [5–7]. The Minnesota LTPA questionnaire includes 63 items pertaining to sports, recreational, yard, and household activities. Activity over the last 12 months is evaluated. Very speci~c rules for estimating activity have been set. The year is considered as comprising 240 work days and the month 22 work days, with 100 weekend days per year [8]. Also, estimated times for various activities are given: climbing one _ight of stairs is estimated to take half a minute, one bowling game 10 minutes, a tennis singles set 20 minutes, and a tennis doubles set 30 minutes. Speci~c attention is given to details such as the difference between socializing and actually being involved in a sporting activity, especially with reference to swimming, tennis, and bowling. Finally, a total activity metabolic index is computed by differentiating between light, moderate, and heavy activities. To create the index, every activity is given a score of intensity units, multiplied by the minutes dedicated to this activity per week, month, and year; e.g., walking for pleasure is alloted 3.5 intensity units, while tennis is given 8 intensity units. This questionnaire has been widely used and validated. Correlations for peak VO2 range from 0.19 [9] to 0.47 [10]. The best correlations are found when heavy activity is assessed. Most subsequent questionnaires have used a similar format, although some questionnaires assign MET values to activity levels. Another expression of METs is given in calories. Thus, 1 MET-h/wk can be expressed as 1 kcal/kg/wk. Many questionnaires use kcal/kg directly [11], while other questionnaires only differentiate among activity levels, i.e., inactive, very low, low, moderately and highly active [12].