ORIGINAL INVESTIGATION. Use of Medications and Dietary Supplements in Later Years Among Male Former Top-Level Athletes

Transcription

1 ORIGINAL INVESTIGATION Use of Medications and Dietary Supplements in Later Years Among Male Former Top-Level Athletes Urho M. Kujala, MD; Seppo Sarna, PhD; Jaakko Kaprio, MD Background: The association between sports participation and later need of medications and dietary supplements is unknown. Subjects and Methods: Male athletes (N=2026) who had represented Finland in international events from 1920 through 1965 and 1401 control subjects who had been classified healthy at the age of 20 years participated in this population-based cohort study. Main Outcome Measures: The main outcome measures were reimbursable medications for hypertension, cardiac insufficiency, coronary heart disease, diabetes, and asthma identified from the national registry from 1970 through 1998 as well as the use of nonsteroidal antiinflammatory drugs, antacids, and specific vitamin and mineral supplements for at least 60 days during the past year reported by questionnaire in Results: Among former top-level athletes compared with controls, the probability of initiating medication was decreased for cardiac insufficiency (age-adjusted hazard ratio [HR], 0.61; 95% confidence interval [CI], ; P.001), coronary heart disease (age-adjusted HR, 0.72; 95% CI, ; P=.002), and asthma (age-adjusted HR, 0.47; 95% CI, ; P.001). Furthermore, the risk of initiation of treatment with regular medication for hypertension (age-adjusted HR, 0.73; 95% CI, ; P=.046) and diabetes (age-adjusted HR, 0.38; 95% CI, ; P=.004) was reduced for endurance athletes but not for power athletes. In 1985, compared with control subjects, athletes used fewer nonsteroidal antiinflammatory drugs (age-adjusted odds ratio [OR], 0.48; 95% CI, ; P.001) and antacids (age-adjusted OR, 0.49; 95% CI, ; P=.002) but more vitamin A (age-adjusted OR, 1.87; 95% CI, ; P=.003), vitamin B (age-adjusted OR, 2.26; 95% CI, , P.001), vitamin C (age-adjusted OR, 1.96; 95% CI, ; P.001), selenium (age-adjusted OR, 1.62; 95% CI, ; P=.006), and iron (age-adjusted OR, 2.35; 95% CI, ; P=.003) supplements. Conclusion: The need for long-term therapy for cardiac disease and asthma as well as for treatment with nonsteroidal anti-inflammatory drugs and antacids is reduced among former top-level athletes, but the use of dietary supplements is increased. Arch Intern Med. 2003;163: From the Unit for Sports and Exercise Medicine, Institute of Clinical Medicine (Dr Kujala), and the Department of Public Health (Drs Sarna and Kaprio), University of Helsinki, Helsinki, Finland. The authors have no relevant financial interest in this article. REGULAR PHYSICAL activity is widely thought to improve a number of health outcomes. Observational studies have indicated that there is a dose-response relationship between baseline physical activity level and reduced risk of many diseases, such as coronary heart disease 1 and type 2 diabetes mellitus. 2 However, epidemiological studies based on samples drawn from the general population have not demonstrated conclusively whether extended periods of very vigorous exercise actually produce these benefits, most likely because there are few individuals who exercise very vigorously and also because of the variability of the exercise patterns of those who do. These problems can be overcome, to some extent, by studying subjects who are former athletes, such as the large Finnish male former top-athlete cohort. 3 The members of this cohort are persons who usually have started very vigorous physical activity during childhood or adolescence and who commonly have continued to participate in specific types of vigorous physical activity over very long periods during their adult life. 3,4 The training of top athletes has changed over time, and it may be that the training of persons who currently participate regularly and actively in recreational sports is similar, in many respects, to the way former top athletes trained. When we compared the Finnish male former top-level athletes with agematched control subjects who were healthy when they were young, we found that the 1064

2 Table 1. Number of Subjects at Study Entry and Still Alive and Identified on January 1, 1970, as Well as Identified and Responding to the 1985 Questionnaire Sport No. of Subjects at Entry No. of Subjects Alive and Identified in 1970 Mean Age (Range) in 1970, y No. of Subjects Alive and Address Identified in 1985 No. (%) Responding to the 1985 Questionnaire Mean Age (Range) of Responders in 1985, y Endurance sports (27-82) (85.4) 61.6 (43-90) Long-distance running (27-82) (82.4) 60.1 (43-89) Cross-country skiing (32-79) (90.0) 63.8 (46-90) Mixed sports (21-80) (85.0) 55.0 (36-89) Soccer (23-76) (82.4) 55.4 (38-84) Ice hockey (23-72) (85.5) 51.9 (38-79) Basketball (21-55) (88.3) 49.3 (36-68) Track and field* (24-80) (85.5) 57.1 (39-89) Power sports (23-85) (83.3) 57.5 (38-94) Boxing (23-72) (82.9) 57.5 (38-82) Wrestling (25-85) (85.5) 58.4 (40-94) Weight lifting (24-75) (80.7) 56.9 (39-76) Track-and-field throwing (26-80) (82.5) 56.6 (40-88) All sports (21-85) (84.7) 56.8 (36-94) Controls (24-86) (76.9) 55.0 (38-86) *Jumping (n = 201), sprinting (n = 132), hurdling (n = 91), middle distance running (n = 95), and decathlon (n = 54). athletes who participated in endurance-type sports had a clearly reduced risk of metabolic syndrome diseases when they were older 5 ; we also found a reduced risk of pulmonary diseases 6 and an increased risk of osteoarthritis of the lower limbs among all athletes. 7 On the one hand, these reduced or increased risks of various diseases should lead to different needs and costs of specific medical drug therapies during later years. On the other hand, competitive athletes are generally known to use a lot of vitamin and mineral supplements, 8,9 with the belief that sufficient nutrition will help them achieve toplevel performances, and this habit may endure long after their athletic careers are over. In this study, we investigated the use of most common reimbursable medications (for cardiovascular diseases, diabetes, and pulmonary diseases) and the use of nonsteroidal anti-inflammatory drugs and antacids, as well as dietary supplements in later years, among male Finnish former top-level athletes and controls. METHODS SUBJECTS Male athletes who had represented Finland at least once at the Olympic games, world or European championships, or other international competitions in selected sports between 1920 and 1965 were identified. 3 The full name, place, and date of birth were traced for 98% of the athletes from selected sports. Control subjects were selected from among Finnish men who at about 20 years of age had been classified as completely healthy (military class AI, fully fit for ordinary military service) at the medical examination preceding their conscription. 3 They were drawn from the public archives of the register of men eligible for military service and were matched for age and area of residence with the athletes. The original cohort of athletes included 2401 men and the reference group 1712 men. 3 We grouped the selected sports based on their ranking in terms of average maximal oxygen uptake 10 as follows: endurance sports (highest maximal oxygen uptake; long-distance runners and cross-country skiers), mixed sports (medium maximal oxygen uptake; soccer, ice-hockey, basketball, track-and-field jumpers, and short-distance runners), and power sports (lowest maximal oxygen uptake; weight-lifters, wrestlers, boxers, and trackand-field throwers) (Table 1). 3 In 1985, a total of 1282 athletes (response rate, 80%-90% by sport of those alive in 1985) and 777 controls (response rate, 77%) responded to our questionnaire study (Table 1). 5 All residents of Finland since January 1, 1967, have a unique personal identifier that is used in all main population and medical registers. Personal identifiers and possible dates of emigration or death for every cohort member were obtained from the Population Register Center in Helsinki. Data on reimbursable medication were obtained from the register of the Social Insurance Institution, the public agency responsible for basic social security covering all residents of Finland. 4 The follow-up of reimbursable medication started January 1, 1970, when reimbursements granted earlier to surviving subjects were coded, and ended December 31, During our study, reimbursable medication granted the patients a total or partial (at least 75%) preferential refund from the cost of prescription drugs approved for the treatment of the studied diseases. To qualify for the refund, the patient had to submit a medical certificate (issued by a physician) describing the illness and the need for drug therapy. The qualification of each patient was then determined by the Social Insurance Institution according to uniform minimum criteria, and each case was reviewed by their expert specialists. Consequently, the register does not cover all mild disease cases sensitively, but the validity of the register is very high and the existence of wrong positive cases is unlikely. We studied in detail those categories with the highest number of cases (chronic hypertension, n=701; chronic cardiac insufficiency, n=428; chronic coronary heart disease, n=351; diabetes mellitus, n=223; and chronic asthma and similar obstructive pulmonary diseases, n=142), thereby giving our analyses adequate statistical power. The questionnaire mailed to surviving cohort members in 1985 included questions on the use of medicines for specific diseases and supplements. Using a structured questionnaire with 5 alternative answers regarding the frequency of use, the members were asked about their use of each medicine and/or supplement during the past year, and those who reported using the medicine and/or supplement for at least 60 days during the past year were classified as users. 1065

3 Table 2. Crude Cumulative Incidence and Age-Adjusted Hazard Ratios Compared With Controls of Most Common Reimbursable Medications Before 1999 Among Cohort Members Alive in 1970 According to Type of Sport Medication Indication Endurance Sports (n = 279) Mixed Sports (n = 983) Power Sports (n = 764) All Sports (n = 2026) Controls (n = 1401) Hypertension No. (%) 48 (17.2) 190 (19.3) 174 (22.8) 412 (20.3) 289 (20.6) HR (95% CI) 0.73 ( ) 0.88 ( ) 1.09 ( ) 0.93 ( ) 1.0 P value* Chronic cardiac insufficiency No. (%) 35 (12.5) 77 (7.8) 115 (15.1) 227 (11.2) 201 (14.3) HR (95% CI) 0.49 ( ) 0.47 ( ) 0.83 ( ) 0.61 ( ) 1.0 P value* Chronic coronary heart disease No. (%) 26 (9.3) 87 (8.9) 77 (10.1) 190 (9.4) 161 (11.5) HR (95% CI) 0.62 ( ) 0.67 ( ) 0.84 ( ) 0.72 ( ) 1.0 P value* Diabetes mellitus No. (%) 10 (3.6) 45 (4.6) 72 (9.4) 127 (6.3) 96 (6.9) HR (95% CI) 0.38 ( ) 0.61 ( ) 1.27 ( ) 0.82 ( ) 1.0 P value* Chronic asthma and similar pulmonary disorders No. (%) 11 (3.9) 24 (2.4) 25 (3.3) 60 (3.0) 82 (5.9) HR (95% CI) 0.60 ( ) 0.38 ( ) 0.55 ( ) 0.47 ( ) 1.0 P value* Abbreviations: CI, confidence interval; HR, hazard ratio. *P value for statistical difference compared with controls by 2-tailed Wald test. Occupational data were collected partly from the Central Population Register and partly from the 1985 questionnaire study. The main occupational groups were as follows: executives, clerical staff, skilled workers, unskilled workers, and farmers. 11 Each person was classified according to occupation. STATISTICAL ANALYSIS In the whole cohort, we used Cox proportional hazards model to calculate hazard ratios and their 95% confidence intervals for the start of treatment with specific reimbursable medications. The follow-up of reimbursable medication started January 1, 1970, when reimbursements granted earlier to surviving subjects were coded, and ended December 31, Cohort members who died before January 1, 1970, were thus excluded from the study. Follow-up ended at emigration, at death, or at the beginning of treatment with reimbursable medication (month and year), or on December 31, 1998, whichever occurred first. In the subgroup of individuals who responded to our questionnaire in 1985, we used logistic regression models to calculate the odds ratios and their 95% confidence intervals for the use of specific medicines or dietary supplements for at least 60 days during the past year for different athletic groups compared with controls. In both analyses, age and occupational group were included as confounding factors. For all statistical analyses, we used SPSS for Windows software (Version 10.0; SPSS Inc, Chicago, Ill). RESULTS Based on data on reimbursable medications, among all groups of former top-level athletes, compared with controls, the age-adjusted risk of initiating long-term treatment with medication was lower for chronic cardiac insufficiency (P.001), chronic coronary heart disease (P=.002), and chronic asthma and similar pulmonary disorders (P.001) (Table 2). Also, compared with controls, former endurance-sports athletes had a lower need of medications for hypertension (P=.046) and diabetes (P=.004), and former mixed-sports athletes had a lower need of medications for diabetes (P=.007) (Table 2). The hazard ratios were very similar after occupational group (results not shown) was adjusted for. Based on questionnaire data from 1985, the ageadjusted use of nonsteroidal anti-inflammatory drugs (P.001) and antacids (P=.002) for at least 60 days per year was lower for all former athletes compared with controls (Table 3). Also, compared with controls, former athletes used more vitamin A (P =.003), vitamin B (P.001), vitamin C (P.001), selenium (P=.006), and iron (P=.003) supplements (Table 3). These findings persisted after occupational group was adjusted for. COMMENT The need for chronic medications for cardiac disease and asthma, as well as the use of nonsteroidal antiinflammatory drugs and antacids, is reduced among former top-level athletes. Also, former endurance athletes but not power athletes need less medications for diabetes and hypertension. The use of vitamin and mineral supplements is increased among former athletes. To be included in our study the athletes had to have participated in international competitions at young age and the selected age-matched controls had to have beenhealthy when they were young adults. In later years, the volume and intensity of leisure physical activity among our former athletes has been clearly greater compared with controls. 3,4,12 Participation in competitive sports is also associated with a distribution of other lifestyle habits that 1066

4 Table 3. Crude Prevalence and Age-Adjusted Odds Ratios Compared With Controls of Use of NSAIDs, Antacids, and Vitamin and Mineral Supplements for at Least 60 Days During the Previous Year Among Cohort Members Responding to the 1985 Questionnaire* Endurance Sports Mixed Sports Power Sports All Sports Controls NSAIDs % (No.) 9.2 (14) 5.4 (32) 8.7 (34) 7.0 (80) 12.2 (86) OR (95% CI) 0.49 ( ) 0.41 ( ) 0.58 ( ) 0.48 ( ) 1.0 P value Antacids % (No.) 2.1 (3) 3.2 (18) 4.8 (17) 3.6 (38) 6.6 (41) OR (95% CI) 0.23 ( ) 0.46 ( ) 0.64 ( ) 0.49 ( ) 1.0 P value Vitamin A % (No.) 10.2 (14) 9.3 (51) 8.7 (31) 9.2 (96) 5.0 (33) OR (95% CI) 1.89 ( ) 1.96 ( ) 1.73 ( ) 1.87 ( ) 1.0 P value Vitamin B % (No.) 11.8 (17) 17.9 (105) 16.6 (62) 16.7 (184) 8.0 (54) OR (95% CI) 1.45 ( ) 2.49 ( ) 2.23 ( ) 2.26 ( ) 1.0 P value Vitamin C % (No.) 15.9 (23) 19.1 (110) 16.7 (61) 17.9 (194) 10.0 (67) OR (95% CI) 1.71 ( ) 2.12 ( ) 1.80 ( ) 1.96 ( ) 1.0 P value Selenium % (No.) 12.3 (17) 12.5 (68) 11.8 (42) 12.3 (127) 7.9 (51) OR (95% CI) 1.58 ( ) 1.67 ( ) 1.55 ( ) 1.62 ( ) 1.0 P value Iron % (No.) 6.3 (8) 6.6 (33) 4.2 (14) 5.7 (55) 2.5 (16) OR (95% CI) 2.72 ( ) 2.68 ( ) 1.72 ( ) 2.35 ( ) 1.0 P value Abbreviations: CI, confidence interval; NSAIDs, nonsteroidal anti-inflammatory drugs; OR, odds ratio. *Cases with missing responses to specific questionnaire items were excluded from specific analyses. P value for statistical difference compared with controls by 2-tailed Wald test. is different from that of the control population. In particular, athletes smoke less commonly. 3,13 We do not have data on these concomitant factors of our cohorts at baseline. However, we studied the prevalence of metabolic syndrome diseases and pulmonary diseases in a crosssectional manner in 1985, adjusting for main confounding factors recorded by the 1985 questionnaire, and the results were in agreement with those of the current study. 5,6 In the subgroup of individuals who responded to the 1985 questionnaire, occupational group was associated with smoking status (P.001) and some other disease risk factors. Therefore, adjustment for occupational group partially adjusts for smoking and some other risk factors, but our results persisted after occupational group was adjusted for. Our results are in accordance with the documented dose-response relationship between high physical activity and reduced metabolic syndrome disease morbidity. 1,2 Because of the lack of randomized trials, it is difficult to determine the extent of genetic selection and continuity of physical activity behind the associations between physically active lifestyle and coronary heart disease. Based on our more detailed analyses, we found that after main confounding factors were adjusted for, both natural selection to endurance-type sports and continuity of physical activity in later years contribute to the prevention of coronary heart disease, but natural selection to power sports does not provide protection. 12 Therefore, power athletes may genetically be more prone to metabolic syndrome diseases. Concerning type 2 diabetes mellitus, there is one randomized clinical trial in which the effect of physical activity was specifically investigated, 14 and, in accordance with our observational data, the results showed that physical activity, particularly endurance training, delays the onset of diabetes. The finding concerning asthma is interesting, because young athletes who participate in vigorous sports often experience asthmalike symptoms during very vigorous activity. It seems that a physically active lifestyle does not cause severe chronic asthma among adults, although selection may contribute to our finding of less chronic asthma among former top-level athletes. Our previous study showed that osteoarthritis of the weight-bearing joints of the lower limbs is more common among athletes than among controls. 7 Interestingly, this condition did not lead to increased use of nonsteroidal anti-inflammatory drug or antacids. Randomized clinical trials on patients with osteoarthritis have shown that exercise therapies alleviate pain. 15 As expected, the use of vitamin and mineral supplements was also high among former athletes in later years. This may or may not have contributed to the finding of the lower need for other medications. The athletes also have continuous monetary costs as a result of supplement use. 1067

5 According to our results, former Finnish athletes live a healthier life than controls. This reduced morbidity agrees with our earlier finding that they do not require as much hospital care in later years. 13 Together, the reduction in need for reimbursable medications and hospital care among athletes means lower costs for the public health care system. The training regimens of today s top athletes differs from that of our cohort; therefore, our results may be more relevant for persons who are actively involved in regular exercise or recreational sports at present. Accepted for publication August 16, This study was supported in part by the Finnish Ministry of Education. Corresponding author: Urho M. Kujala, MD, Unit for Sports and Exercise Medicine, Mannerheimintie 17E (Töölö Sports Hall), FIN Helsinki, Finland ( Urho.Kujala@helsinki.fi). REFERENCES 1. Kohl HW III. Physical activity and cardiovascular disease evidence for a dose response. Med Sci Sports Exerc. 2001(suppl);33:S472-S Kelley DE, Goodpaster BH. Effects of exercise on glucose homeostasis in type 2 diabetes mellitus. Med Sci Sports Exerc. 2001(suppl);33:S495-S Sarna S, Sahi T, Koskenvuo M, Kaprio J. Increased life expectancy of world class male athletes. Med Sci Sports Exerc. 1993;25: Kujala UM, Kettunen J, Paananen H, et al. Knee osteoarthritis in former runners, soccer players, weight lifters, and shooters. Arthritis Rheum. 1995;38: Kujala UM, Kaprio J, Taimela S, Sarna S. Prevalence of diabetes, hypertension, and ischemic heart disease in former elite athletes. Metabolism. 1994;43: Kujala UM, Sarna S, Kaprio J, Koskenvuo M. Ashma and other pulmonary diseases in former élite athletes. Thorax. 1996;51: Kujala UM, Kaprio J, Sarna S. Osteoarthritis of the weight-bearing joints of the lower limbs in former elite male athletes. BMJ. 1994;308: Sobal J, Marquart LF. Vitamin/mineral supplement use among athletes: a review of the literature. Int J Sport Nutr. 1994;4: American College of Sports Medicine, American Dietetic Association, and Dietitians of Canada. Joint position statement: nutrition and athletic performance. Med Sci Sports Exerc. 2000;32: Åstrand P-O, Rodahl K. Textbook of Work Physiology: Physiological Bases of Exercise. 3rd ed. New York, NY: McGraw-Hill Co; 1986: Finland Central Statistical Office. Alphabetical List of Occupations and Classification of Social Class. Helsinki, Finland: Central Statistical Office; Kujala UM, Sarna S, Kaprio J, Tikkanen HO, Koskenvuo M. Natural selection to sports, later physical activity habits, and coronary heart disease. Br J Sports Med. 2000;34: Kujala UM, Sarna S, Kaprio J, Koskenvuo M. Hospital care in later life among former world-class Finnish athletes. JAMA. 1996;276: Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance: the Da Qing IGT and Diabetes Study. Diabetes Care. 1997;20: Baker K, McAlindon T. Exercise for knee osteoarthritis. Curr Opin Rheumatol. 2000;12: Correction Error in Table. In the Original Investigation titled Distrust, Race, and Research, published in the November issue of the ARCHIVES (2002;162: ), labels in Table 4 for the variable on employment status were listed incorrectly. Table 4 is reprinted correctly here. Adjusted Prevalence Odds Ratio (OR) and 95% Confidence Interval (CI) for the Association Between Race and Distrust, Controlling for Sociodemographic Confounders* Predictor OR (95% CI) Race African American 4.7 ( ) White Education (ordinal) 0.9 ( ) Employment No 1.7 ( ) Yes Region Northeast 2.7 ( ) South 2.2 ( ) West 3.3 ( ) Midwest *Each variable is adjusted for confounding by other variables using multiple logistic regression. Six observations were deleted because of missing values for 1 or more of the model variables. 1068