The role of physical activity in the development of gallbladder disease has only recently been investigated. In addition, the mechanisms by which physical activity may influence gallbladder disease independently of its effect on body weight are intriguing but not clear. Suggested mechanisms beyond body weight maintenance include the influence of physical activity on gastrointestinal and gallbladder motility and on metabolic abnormalities such as hyperinsulinemia, elevated triglyceride levels, and low plasma HDL levels (4,5,18,19).
There have been four prospective studies that the authors are aware of that have investigated the role of physical activity in the development of gallbladder disease. In three of these studies, gallbladder disease was based on self-report via questionnaire (13,14,16). In the fourth study, cases were captured through surveillance of hospital records and were confirmed by histological or radiological examination (6). In all but one of these studies, physical activity was shown to decrease the risk of development of symptomatic gallbladder disease after controlling for the effect of potential confounders such as body mass index.
Because symptoms develop in only a fraction of persons with gallstones, an obvious major limitation of all of these prospective studies was the diagnosis of gallbladder disease. Despite the fact that all of these studies confirmed their self-reported cases of symptomatic gallbladder disease, none of these four prospective studies had the means to include subjects with asymptomatic gallstones in their case group. Further evaluation is warranted to investigate the relationship between physical activity and gallbladder disease in a cohort that has been screened by ultrasonography.
The Strong Heart Study (SHS), originally designed to investigate cardiovascular disease and its risk factors in a population-based sample of American Indian men and women, ages 45-74 yr, has provided this opportunity. This population has an extremely high prevalence of gallbladder disease (2,17). Past-year physical activity levels were assessed at baseline by a physical activity questionnaire that had been validated in American and Canadian Indians (7). Most importantly, gallbladder disease in this study was diagnosed by gallbladder ultrasonography performed on the entire cohort at the first follow-up examination (2).
The Strong Heart Study is the first multicenter, population-based ultrasonographic study of gallstone-related gallbladder disease (2). The purpose of this current effort is to investigate, in this high-risk cohort, the relationship between physical activity levels obtained at baseline (1989-1992) and the development of gallbladder disease determined by gallbladder ultrasonography at the first follow-up visit (1993-1995) as part of this multicenter effort.
MATERIALS AND METHODS
Study subjects are participants in the Strong Heart Study, a population-based study of cardiovascular risk factors and prevalent and incident cardiovascular disease in American Indian communities. The study design, survey methods, and laboratory techniques of the SHS have been previously reported (12). Briefly, the population for the SHS were resident tribal members, aged 45 to 75 yr, from central Arizona (the Gila River and Salt River Pima/Maricopa Indians, and Ak-Chin Pima/Papago Indians), Oklahoma (the seven tribes of Southwestern Oklahoma: Apache, Caddo, Comanche, Delaware, Fort Sill Apache, Kiowa, and Wichita), South Dakota (Oglala and Cheyenne River Sioux), and North Dakota (the Spirit Lake Sioux in the Fort Totten area). The study population was derived from tribal records and a review of the tribal lists by long time residents to define the eligible participants. The study protocol was approved by the National Indian Health Service institutional review board (IRB), by the IRB of the participating institutions, and by the participating tribes. Informed consent was obtained from all participants before their participation in any part of this study.
The baseline examination (1989-1992) was completed in 4549 individuals. Examinations were conducted in the morning after a 12-h fast. After informed consent was administered, fasting blood samples were obtained and a 2-h glucose tolerance test was administered. Exam components included medical history, medication use, electrocardiogram, and an interviewer-administered physical activity questionnaire.
Body size measurements included body mass index, waist/hip ratio, and percentage of body fat. Percent body fat was estimated using RJL impedance meter (Model B14101, RJL Equipment, Detroit, MI) and an equation based on total body water. To compute lean body mass, total body water was multiplied by 0.732. Anthropometric measurements were made with shoes removed and the participants wearing lightweight clothing. Waist circumference was measured at the level of the umbilicus, with the participant in a supine position. Hip circumference was measured over light clothing at the level of the widest protrusion of the hips, with the participant in a standing position.
Participants were interviewed to elicit information on leisure-time and occupation-related physical activities. Almost all participants were fluent in English, so the questions were read in English, with further explanation when needed. The physical activity questionnaire has been validated in numerous populations, including Pima Indians (9,10). The questionnaire assesses the combination of reported leisure and occupational activity during the past year as an estimate of "usual" activity levels. Leisure activities were assessed by asking participants to report, from a list of common local leisure activities, all those activities that they had participated in during the past 12 months. They were then asked to estimate the frequency and duration of each activity positively identified.
For occupational activity, individuals were asked to list all jobs held during the past 12 months. For each job entry, data were collected for minutes per day walking or cycling to work, as well as the number of months per year, days per week, and hours per day spent working in that job. Job activity was determined by the number of hours spent sitting at work and the most frequent activities performed when not sitting. Housewives, unemployed, physically challenged, and retired individuals were questioned about activities during a normal "40-h work week" (9).
Estimates of leisure and occupational activity were calculated separately as hours per week (h·wk−1) averaged for the past year. Each activity was also weighted by its relative metabolic cost, referred to as an MET, thereby deriving MET-hours per week (MET·h·wk−1) as the final unit of expression. One MET represents the energy expenditure for an individual at rest, whereas a 10-MET activity requires 10 times the resting energy expenditure (1). Total physical activity was the combination of leisure and occupational activity averaged for the past year.
The first SHS follow-up examination (1993-1995) was completed on 89% of the surviving cohort and included gallbladder imaging. The gallbladder ultrasonography protocol was adapted from the third National Health and Nutrition Examination Survey (NHANES). Imaging studies were performed by trained sonographers using the Acuson 128 series ultrasound machine with a 3.5-MHz probe. After training, the field sites performed approximately 20 ultrasounds, which were then sent to the ultrasound reading center at George Washington University. Videotapes were read, and agreement on the diagnosis between the ultrasonographer, the reading center, and the radiologist's evaluation determined if further training was needed. A kappa statistic of at least 0.8 was the standard set for each site to maintain between the readings of the ultrasonographer and the radiologist.
All study participants undergoing gallbladder ultrasonography were asked to fast for at least 6 h before the ultrasonography, but they were not excluded if they had not fasted. The examinations were performed by registered sonographers with the patient in both the supine and the left lateral decubitus positions. Cholecystectomy was based on self-report or findings of an appropriate surgical scar and absent gallbladder on ultrasound. Ultrasonographers completed a data-collection form with their findings of the examination/scanning and sent this form, along with the videotape, to the reading center. In addition to the criteria used for reading gallbladders examinations in the NHANES survey, the radiologist estimated the proportion of gallbladder volume being displaced by any gallstones.
On the basis of the individual's medical history and the results of the gallbladder ultrasonography, individuals were classified into one of two categories, gallbladder disease or no disease. Those individuals who were considered positive for gallbladder disease either had gallbladder stones according to the ultrasonography findings, or they reported having gallbladder surgery and had the surgical scars to match.
All statistical analyses were conducted using SAS version 6.12 (1990) and were performed separately for each sex. The Wilcoxon rank-sum test was used to compare median physical activity levels by gallbladder status (yes/no).
Multiple logistic regression analysis was used to determine the contribution of physical activity levels at baseline to the development of gallbladder disease at the first follow-up visit. Each model was adjusted for age, sex, and body mass index. For the independent variables, the square root of total physical activity and the natural logarithm of body mass index were employed to improve the fit of all models. We tested for statistical interactions between activity and age, sex, or body mass index. The importance of these interactions were determined by examining the individual P values of the interaction coefficients and by assessing the impact of the addition of these terms on the overall fit of the models.
Four thousand forty-eight individuals (89% of the baseline cohort of 4549) participated in the follow-up visit. Among these 4048 individuals, 3143 (78%) successfully completed both a physical activity questionnaire at baseline and a gallbladder ultrasound or assessment of gallbladder surgery at the first follow-up clinic visit. Baseline characteristics (mean values) of these individuals are presented in Table 1, stratified by gallbladder disease status determined at the follow-up visit. Individuals who were considered positive for gallbladder disease at the follow-up visit either had gallstones according to the ultrasonography findings, or they reported having gallbladder surgery and had the surgical scars to match. Body mass indexes, along with the prevalence rates of diabetes in these men and women from the various SHS American Indian communities, were high, consistent with earlier reports (7,11).
As also can be seen in Table 1, the prevalence rates of gallbladder disease at the follow-up visit in the men and women from the various SHS American Indian communities are substantially high, as reported previously (2). Specifically, the prevalence of gallbladder disease in these 3143 individuals at the first follow-up visit was 51% (1610/3143). In addition, the prevalence of gallbladder disease was substantially higher in women compared with men (64% vs 30%), with women more likely to have a cholecystectomy than men.
Median physical activity levels obtained at baseline are presented by sex and gallbladder status as determined by ultrasound or reported surgery at the first follow-up clinic visit. As shown in Figure 1, both men and women without gallbladder disease were significantly (P < 0.01) more active at baseline than those that either had known gallbladder disease or tested positively for gall stones during the ultrasound procedure (Fig. 1).
To focus on cases of gallbladder disease that developed during follow-up, individuals at baseline that reported that a physician had told them that they had gallbladder disease or surgery along with individuals with missing values (N = 50) were eliminated from the next series of analyses. Among the 2130 participants (1067 women and 1063 men) that did not report gallbladder disease at baseline, 650 individuals (403 women and 247 men) were found to have gallbladder disease at the first follow-up clinic visit on the basis of ultrasound or reported surgery. As can be seen in Figure 2, median baseline physical activity levels were lower both in women (P < 0.01) and men (P < 0.10) who developed gallbladder disease by the first follow-up clinic visit.
Controlled for age, sex, and body mass index, a high level of reported total physical activity at baseline was significantly and inversely associated with the development of gallbladder disease at follow-up (P = 0.028), as determined by multiple logistic regression analysis (Table 2). Adding smoking status, alcohol intake, study site, or waist circumference to the model did not alter this significant association between physical activity and the development of gallbladder disease (data not shown). The significant relationship between physical activity and the development of gallbladder disease was maintained when the data were stratified by sex. In fact, there were no significant interactions between reported physical activity levels and age, sex, or body mass index in this data set.
The relationship between physical activity and the development of gallbladder disease was then examined in terms of whether the individual had diabetes at baseline. Stratifying by baseline diabetes status and controlling for age, sex, and body mass index, a high level of reported total physical activity at baseline remained significantly and inversely associated with the development of gallbladder disease at follow-up (P = 0.02) in those without diabetes at baseline (Table 2). However, neither physical activity level nor body mass index were significantly related to the development of gallbladder disease in those with baseline diabetes.
This investigation was the first to examine the relationship between physical activity and gallbladder disease, partially evaluated by ultrasonography, in a cohort of individuals recruited from a population at high risk for gallbladder disease. Gallbladder disease status was determined at baseline by self-report or by having gallbladder surgery, whereas gallbladder disease status at follow-up was evaluated by ultrasonography. These findings suggest that physical activity is associated with a decreased risk of development of gallbladder disease (13,15,17). The relationship between activity and gallbladder disease development remained after controlling for the effect of potential confounders such as sex, age, and body mass index. In addition, this relationship was more robust in women than men, and it may reflect that twice as many women had gallbladder disease than men. However, when the analyses were stratified by diabetes status, the significant relationship between activity and gallbladder disease development only remained in those without diabetes.
Before the current study, four other prospective studies had examined the role of physical activity in the development of gallbladder disease (6,13,14,16). Three of these four prospective studies suggested a significant inverse relationship between physical activity and gallbladder disease, defined as either cholecystectomy (14), symptomatic gallstones (diagnosed by ultrasonography or radiography), or cholecystectomy (13), or radiologically confirmed cases of gallbladder disease, captured by continued surveillance of all possible hospitals in the area (6). In contrast, the fourth study ascertained gallbladder disease by self-report, with validation on a small subsample of their cohort; they did not find a significant relationship between activity and gallbladder disease (16). The results of the present study, which examined the entire cohort by ultrasonography, confirmed the findings of the former three studies.
In addition to being the first study to measure gallbladder disease in the entire cohort by ultrasonography, this is the first study to examine the role of physical activity in the development of gallbladder disease in a high-risk population. The previous four prospective studies in this area had recruited their subjects from Harvard University alumni (16), U.S. male health professionals (13), female registered nurses (14), and American men of Japanese ancestry (6).
It is intriguing to speculate as to the possible reasons for the discrepancy in results between the studies that have demonstrated a significant relationship between physical activity and gallbladder disease development, and the one prospective study that contradicted this finding (16). The most obvious differences in these studies are that the later investigation had a younger cohort at baseline (15-24 yr of age compared with a much older age group of 40+ or 45+ yr of age in the other studies) and had a much longer follow-up period compared with the other studies. Misclassification of exposure to physical activity is more likely to occur in an observational study of 15- to 24-yr-old individuals observed for an extended period of time than in an older cohort observed for a much shorter time period. In fact, both studies by Leitzmann et al. (13,14) report decent tracking of the physical activity variable over time and demonstrate similar results when the activity variable was adjusted every 2 yr in contrast to using the baseline activity value alone. Length of time from exposure to outcome was not as much of an issue in the current study, because the follow-up occurred only 2-4 yr after baseline.
A significant relationship between physical activity level and gallbladder disease development was found in those without diabetes, but not in those with diabetes. Part of the rationale for these findings may be that physical activity levels were substantially lower for both males and females with diabetes in this data set (data not shown) as well as in other published reports of the U.S. (3) and Native populations (8). The narrow range of activity levels in individuals with diabetes is likely joined by physiological changes that occur with diabetes to explain the lack of an activity/gallbladder relationship in this group. The specifics have yet to be determined.
A limitation to this current effort is that the ultrasound assessment was only completed at the first follow-up visit and was not performed at baseline. To identify new cases that have developed since the baseline visit in this investigation, individuals at baseline who reported that a physician told them they had gallbladder disease or who had surgery were eliminated from the prospective analyses. To examine the possibility of a potential bias introduced by a reduction in physical activity levels among individuals with undetected preclinical gallbladder disease at baseline, Leitzmann et al. (13,14), in both of their studies, excluded the first 2-4 yr of follow-up. Their findings did not change significantly when these first 2-4 yr were excluded. It was not possible for us to carry out the same types of examinations, because the length of our follow-up period was so short. A second limitation to this manuscript is that diet information was not available, although it is unlikely, on the basis of the current literature, that adjusting for diet would have significant impact on the relationship between activity and gallbladder disease. A final limitation is that the activity questionnaire used in this effort assessed leisure, occupational, and walking/biking activity and ignored other types of mostly lower-intensity physical activities, such as housework and child care, that are common in older adults and women. Because the accuracy of recall for lower-intensity activities is notoriously poor when determined subjectively, these types of activities are not assessed by the current activity questionnaire.
A review of physical activity in the treatment and prevention of obesity-related conditions (15) suggests that future priorities in gallbladder disease research should include an extension of the work in this area to that of more diverse populations as well as clarification as to the type and intensity of physical activity recommended. This current effort has certainly moved the field forward regarding diversity of the populations examined. Clarification of the patterns of physical activity that are the most beneficial for gallbladder disease prevention will likely require the use of objective measures of physical activity such as accelerometers and pedometers in addition to activity questionnaires, because the recall of lower-intensity physical activities in the latter is questionable.
The authors acknowledge the assistance and cooperation of the participating tribes and the Indian Health Service facilities that serve those tribes. The authors want to thank the study participants and the SHS staff for their important contribution to this effort. Finally, AMK wants to thank Dr. Peter Bennett and Dr. William Knowler for their unending support and guidance in earlier discussions of this manuscript.
SHS is conducted as a collaborative agreement supported by grants U01 HL41642, U01 HL41652, and U01 HL4164. MD-000-207-03. Additional support was provided by the National Institute of Diabetes and Digestive and Kidney Diseases. AMK is supported through the EXPORT Health Project at the Center for Minority Health, Graduate School of Public Health, University of Pittsburgh, NIH/NCMHD, grant no. P60.
The opinions expressed are those of the authors and do not necessarily reflect the views of the Indian Health Service.
1. American College of Sports Medicine. Guidelines for Graded Exercise Testing and Exercise Prescription
, 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2000.
2. Everhart, J. E., F. Yeh, M. C. Hill, et al. Prevalence of gallbladder disease in American Indian populations: findings from the Strong Heart Study. Hepatology
3. Ford, E. S., and W. H. Herman. Leisure-time physical activity patterns in the U.S. diabetic population. Findings from the 1990 National Health Interview Survey-health promotion and disease prevention supplement. Diab. Care
4. Heaton, K. W., P. M. Emmett, C. L. Symes, and F. E. M. Braddon. An explanation for gallstones in normal-weight women: slow intestinal transit. Lancet
5. Hofmann, A. Recreational physical activity and the risk of cholecystectomy in women: correspondence. N. Engl. J. Med.
6. Kato, I., A. Nomura, G. N. Stemmerman, and P. H. Chyou. Prospective study of clinical gallbladder disease and its association with obesity, physical activity, and other factors. Dig. Dis. Sci.
7. Knowler, W. C., D. J. Pettitt, M. Saad, et al. Obesity in the Pima Indians: its magnitude and relationship with diabetes. Am. J. Clin. Nutr.
8. Kriska, A. M., R. E. LaPorte, D. J. Pettitt, et al. The association of physical activity with obesity, fat distribution and glucose intolerance in Pima Indians. Diabetologia
9. Kriska A. M., and C. J. Caspersen. Introduction to the collection of physical activity questionnaires in a collection of physical activity questionnaires for health-related research. Med. Sci. Sports Exerc.
10. Kriska, A. M., W. C. Knowler, R. E. Laporte, et al. Development of questionnaire to examine relationship of physical activity and diabetes in Pima Indians. Diab. Care
11. Lee, E. T., B. V. Howard, P. J. Savage, et al. Diabetes and impaired glucose tolerance in 3 American Indian populations aged 45-74 years. The Strong Heart Study. Diab. Care
12. Lee, E. T., T. K. Welty, R. R. Fabsitz, et al. The Strong Heart Study. A study of cardiovascular disease in American Indians: design and methods. The Strong Heart Study. Am. J. Epidemiol.
13. Leitzmann, M. F., E. L. Giovannucci, E. B. Rimm, et al. The relation of physical activity to risk of symptomatic gallstone disease in men. Ann. Intern. Med.
14. Leitzmann, M. F., E. B. Rimm, W. C. Willett, et al. Recreational physical activity and the risk of cholecystectomy in women. N. Engl. J. Med.
15. Rissanen, A., and F. Fogelholm. Physical activity in the prevention and treatment of other morbid conditions and impairments associated with obesity: current evidence and research issues. Med. Sci. Sports Exerc.
16. Sahi, T., R. S. Paffenbarger, C. Hsieh, and I. M. Lee. Body mass index, cigarette smoking, and other characteristics as predictors of self-reported, physician-diagnosed gallbladder disease in male college alumni. Am. J. Epidemiol.
17. Sampliner, R. E., P. H. Bennett, L. J. Comess, F. A. Rose, and T.A. Burch. Gallbladder disease in Pima Indians: demonstration of high prevalence and early onset by cholecystography. N. Engl. J. Med.
18. Utter, A., and F. Goss. Exercise and gallbladder function. Sports Med.
19. Vega, K. J., and D. E. Johnston. Exercise and the gallbladder. N.Engl. J. Med.