Lucy A. Mead

Joint Injury in Young Adults and Risk for Subsequent Knee and Hip Osteoarthritis

Osteoarthritis is a major contributor to functional impairment and reduced independence in older adults (1-4). It is the leading cause of arthritis in the United States, affecting an estimated 21 million persons (5), and has substantial economic impact (6, 7). History of an injury to a joint, particularly at the knee and hip, is associated with an increased risk for osteoarthritis in cross-sectional and casecontrol studies (8-11). Such studies, however, may overestimate this relation because persons with symptomatic osteoarthritis may be more likely to recall a past injury or to interpret early symptoms of osteoarthritis as indicative of joint injury. A prospective cohort study can address this weakness by determining exposure status before the outcome develops. To date, prospective studies have examined the relation between history of joint injury and osteoarthritis in middle-aged persons and senior citizens (12, 13), but not in young adults. However, many athletic injuries occur in high school and college. In addition, joint trauma may be a more common cause of osteoarthritis than has been previously recognized (14). We performed a prospective cohort study of 1321 young adults to examine the risk for knee and hip osteoarthritis associated with joint injury during young adult life. Methods Study Participants The Johns Hopkins Precursors Study was designed by the late Caroline Bedell Thomas, MD, to identify precursors of the aging process (15). A total of 1337 medical students, members of the graduating classes of 1948 through 1964 at the Johns Hopkins University School of Medicine in Baltimore, Maryland, enrolled in the study. The cohort was 91% male and 97% white; the mean age was 22 years. At entry, participants underwent a standard history and examination, including assessment of musculoskeletal disorders, history of trauma, level of physical activity, and measurement of weight and height. Body mass index was calculated as weight in kilograms divided by height in meters squared. In addition, participants were asked to categorize their level of physical training during the past month as none, little, moderate, or much, as described elsewhere (16). Since graduation, participants have been followed prospectively with annual self-administered questionnaires to detect incident disease and to update risk factor status over time. At the time that baseline data were collected, it was not customary to obtain informed consent. After establishment of the Joint Committee on Clinical Investigation at Johns Hopkins, the follow-up protocol was reviewed and approved. Assessment of Injury Injury was defined as a report of trauma to the knee or hip joint, including internal derangement and fracture. During the baseline assessment, knee and hip injuries that occurred before graduation from medical school and the year of their occurrence were recorded. Postgraduation injuries were assessed by annual morbidity questionnaires. During every 5-year follow-up period, at least 86% of the living participants responded at least once to the questionnaires. Injuries were assigned diagnosis codes according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) (17). Diagnoses included torn meniscus (code 836.2); torn knee ligament (code 844.9); tibial fracture (code 823.8); femoral fracture (code 821.01); fibular fracture (code 823.81); patellar fracture (code 822); broken leg, site not otherwise specified (code 827.0); knee injury, not otherwise specified (code 959.7); gunshot wound (code 891); hemarthrosis (code 844.9); shin splints (code 844.9); and hip dislocation (code 836.5). Incidence of Osteoarthritis The methods used to determine osteoarthritis incidence in this cohort have been described elsewhere (18). Briefly, all participants were mailed annual morbidity questionnaires. Before 1985, respondents were queried about the development of physical and musculoskeletal disorders. Since 1985, follow-up morbidity questionnaires have contained the specific question, Have you ever had arthritis? and have asked those who respond affirmatively to provide the type of arthritis, year of onset, and treatment received. An event of osteoarthritis was defined as the self-report of osteoarthritis or degenerative arthritis in response to the questionnaire. The incidence date was determined by the reported year of osteoarthritis onset. Two rheumatologists reviewed each report of osteoarthritis, assigned ICD-9-CM codes (17) to the most specific arthritis site reported by the respondent, and excluded inflammatory arthritides. For this analysis, only osteoarthritis at the knee (code 715.96) and hip (code 715.95) were included as outcomes. In 1995, all surviving participants who had previously reported osteoarthritis were mailed a more detailed questionnaire. They were asked, Have you had pain in or around the knee, including back of the knee, on most days for at least 1 month? and Have you had pain in or around either hip joint, including the buttock, groin, and side of upper thigh, on most days for at least 1 month? These symptom-related questions were used to screen for knee and hip osteoarthritis, respectively, in the National Health and Nutrition Examination Survey (NHANES) (19). Respondents were also asked whether they had undergone radiographic evaluation of their knees and hips and whether radiography revealed osteophytes, joint space narrowing, subchondral cysts, or bony sclerosis (the radiographic hallmarks of osteoarthritis) (20). Statistical Analysis The association between baseline characteristics and injury status was assessed by using the Student t-test for continuous variables and the chi-square test for categorical variables. These data are presented both for the entire cohort and separately by sex. Injury was the independent variable in the survival analysis; age was the time variable. Knee and hip osteoarthritis were examined as separate outcomes. The relation of a joint injury at baseline to the incidence of osteoarthritis was examined by using KaplanMeier analysis (21). The log-rank statistic was used to test whether the cumulative incidence of osteoarthritis differed according to injury status at cohort entry (22). In addition, Cox proportional-hazards analysis was used to estimate the independent risk for osteoarthritis associated with a joint injury. Cox models were constructed for injury status at cohort entry and over follow-up, modeled as a time-dependent covariate. We adjusted for age at graduation, sex, body mass index, and level of physical activity, each of which was determined at baseline, to evaluate for possible confounding. In this cohort, body mass index at enrollment was more predictive of future osteoarthritis than average or most recent body mass index during follow-up (18). These analyses were performed for all reported events of knee and hip osteoarthritis and for cases confirmed by both characteristic symptoms and radiographic findings. Hazard ratios are reported as relative risks with 95% CIs. Statistical significance was defined as an level equal to 0.05 using a two-tailed test. We conducted a sensitivity analysis to assess the potential effect of an omitted covariate on the association between joint injury and subsequent osteoarthritis. Role of the Funding Sources The funding sources did not have a role in the collection, analysis, or interpretation of the data or in the decision to submit the study for publication. Results At graduation from medical school, the average age of the 1216 men and 121 women in the cohort was 26 years. Our analysis is based on events reported through 30 November 1995 and represents a median follow-up of 36 years. Information at baseline or during follow-up was available for 1321 participants, who form the basis of our longitudinal analysis. At the end of follow-up in 1995, the mean age of the cohort SD was 61.4 8.9 years. Overall, 141 participants had a joint injury (knee alone [n =111], hip alone [n =16], or injuries at both sites [n =14]) before or after graduation. Table 1 shows the baseline characteristics of the cohort according to injury status at the end of follow-up. Proportions of men and women with joint injury were similar (P >0.2, chi-square test). Participants were generally lean, 49% were physically active, and 54% smoked cigarettes. Men who had a joint injury were heavier at baseline than men who were injury-free, but both groups of men were similar in age, level of physical activity, and smoking status. Women with injury were, on average, 2 years older at baseline than those without injury. Table 1. Baseline Characteristics of 1321 Men and Women according to Joint Injury at Cohort Entry or during Follow-up At baseline, 47 men (3.9%) reported a knee injury and 12 men (1.0%) reported a hip injury. The mean age at which the injuries occurred was 16 years. Knee injuries included 10 tibial fractures, 3 fibular fractures, 4 femoral fractures, 2 patellar fractures, and 9 otherwise unspecified leg fractures. In addition, 8 men incurred trauma resulting in torn knee cartilage, 2 had torn knee ligaments, 2 had knee injuries resulting from gunshot wounds, 1 had traumatic hemarthrosis, 1 had joint dislocation, 1 had shin splints, and 4 had otherwise unspecified knee injuries. Of men with a knee injury, 4 reported that they were injured while playing football, 2 each reported that they were injured while playing basketball or participating in athletics, and 1 each reported that he was injured while horseback riding, bicycling, skiing, playing volleyball, wrestling, playing baseball, or playing tennis. Knee trauma also resulted from motor vehicle accidents (n =3) and falls (n =2). At baseline, 4 of the 121 women (3.3%) had a history of knee injuries and 1 (0.8%) had a history of hip injuries. Sixty-two men and 7 women developed knee osteoarthritis, and 27 men and 5 women developed hip osteoarthritis. The mean SD ages at onset of knee

Serum Cholesterol in Young Men and Subsequent Cardiovascular Disease

BACKGROUND The increased risk of cardiovascular disease associated with higher serum cholesterol levels in middle-aged persons has been clearly established, but there have been few opportunities to examine a potential link between serum cholesterol levels measured in young men and clinically evident premature cardiovascular disease later in life. METHODS We performed a prospective study of 1017 young men (mean age, 22 years) followed for 27 to 42 years to quantify the risk of cardiovascular disease and total mortality associated with serum cholesterol levels during early adult life. The mean serum cholesterol level at entry was 192 mg per deciliter (5.0 mmol per liter). RESULTS During a median follow-up of 30.5 years, there were 125 cardiovascular-disease events, 97 of which were due to coronary heart disease. The serum cholesterol level at base line was strongly associated with the incidence of events related to coronary heart disease and cardiovascular disease, as well as to total mortality and mortality due to cardiovascular disease. The risks were similar whether the events occurred before or after the age of 50. In a proportional-hazards analysis adjusted for age, body-mass index (the weight in kilograms divided by the square of the height in meters), the level of physical activity, coffee intake, change in smoking status, and the incidence of diabetes and hypertension during follow-up, a difference in the serum cholesterol level at base line of 36 mg per deciliter (0.9 mmol per liter)--the difference between the 25th and 75th percentiles of cholesterol level in the study population at base line--was associated with an increased risk of cardiovascular disease (relative risk, 1.72; 95 percent confidence interval, 1.39 to 2.14), coronary heart disease (relative risk, 2.01; 95 percent confidence interval, 1.59 to 2.53), and mortality due to cardiovascular disease (relative risk, 2.02; 95 percent confidence interval, 1.23 to 3.32). A difference in the base-line serum cholesterol level of 36 mg per deciliter was significantly associated with an increased risk of death before the age of 50 (relative risk, 1.64; 95 percent confidence interval, 1.03 to 2.61), but not with the overall risk of death (relative risk, 1.21; 95 percent confidence interval, 0.93 to 1.58). CONCLUSIONS These findings indicate a strong association between the serum cholesterol level measured early in adult life in men and cardiovascular disease in midlife.

Cardiovascular reactivity to the cold pressor test as a predictor of hypertension.

Cardiovascular reactivity to stress is hypothesized to be a marker for subsequent neurogenic cardiovascular disease, but few prospective studies of this hypothesis are available. We studied 910 white male medical students who had their blood pressure and pulse rate measured before and during a cold pressor test in the years 1948–1964. Hypertensive status (requiring drug treatment) was ascertained by annual questionnaires in the 20- to 36-year follow-up period. An association was observed between maximum change in systolic blood pressure and later hypertension, with a cumulative incidence of hypertension by age 44 of 6.7%, 3.0%, and 2.4% for a change in systolic blood pressure in the upper, middle two, and lowest quartiles, respectively (Kaplan-Meier, p < 0.02). After adjustment for study entry age, Quetelet Index, cigarette smoking, pretest systolic blood pressure, and paternal or maternal history of hypertension in a Cox model, the association persisted. The excess risk associated with systolic blood pressure reactivity was not apparent until the population aged some 20 years and was most apparent among those in whom hypertension developed before age 45 (relative risk=2.5, 95% confidence intervals=1.47, 4.71 for a 20 mm Hg change). Diastolic blood pressure and heart rate changes were not associated with later hypertension. These data suggest that persons prone to later hypertension manifest an altered physiology at a young age.

Coffee Consumption and the Incidence of Coronary Heart Disease

We conducted a prospective investigation of the effect of coffee consumption on coronary heart disease in 1130 male medical students who were followed for 19 to 35 years. Changes in coffee consumption and cigarette smoking during follow-up were examined in relation to the incidence of clinically evident coronary disease in comparisons of three measures of coffee intake--base-line intake, average intake, and most recent intake reported before the manifestation of coronary disease. Clinical evidence of coronary disease included myocardial infarction, angina, and sudden cardiac death. In separate analyses for each measure of coffee intake, the relative risks for men drinking five or more cups of coffee per day, as compared with nondrinkers, were approximately 2.80 for all three measures in the univariate analyses (maximum width of 95 percent confidence intervals, 1.27 to 6.51). After adjustment for age, current smoking, hypertension status, and base-line level of serum cholesterol, the estimated relative risk for men drinking five or more cups of coffee per day (using the most recent coffee intake measure), as compared with those drinking none, was 2.49 (maximum width of 95 percent confidence interval, 1.08 to 5.77). The association between coffee and coronary disease was strongest when the time between the reports of coffee intake and the coronary event was shortest. These findings support an independent, dose-responsive association of coffee consumption with clinically evident coronary heart disease, which is consistent with a twofold to threefold elevation in risk among heavy coffee drinkers.

Exercise and depression in midlife: a prospective study.

OBJECTIVES This study examined the relationship of self-reported physical activity with subsequent depression and psychiatric distress. METHODS Physical activity was assessed in medical school and midlife in 973 physicians as part of a prospective observational study. Outcome measures were the incidence of self-reported clinical depression and psychiatric distress on the General Health Questionnaire. RESULTS The risk of depression was similar for nonexercisers and exercisers. No relationship was observed between physical activity level and subsequent psychiatric distress. CONCLUSIONS This study found no evidence that exercise reduces risk for depression or psychiatric distress.

The psychosocial work environment of physicians. The impact of demands and resources on job dissatisfaction and psychiatric distress in a longitudinal study of Johns Hopkins Medical School graduates.

This study examines the relationship between the psychosocial work environment and cross-sectional job dissatisfaction and prospective psychiatric distress in a cohort of Hopkins Medical School graduates in midcareer. An instrument was constructed consisting of five scales: psychological job demands, patient demands, work control, physician resources, and coworker support. The results of scale reliability and factor analysis are presented. Higher job demands were found to be associated with increases in job dissatisfaction and psychiatric distress and greater resources were associated with decreased levels of dissatisfaction and distress. In multiple-regression analysis, only work control and social support were found to be independently associated with dissatisfaction and distress. These results suggest that the presence of control and social support at work protects physicians from developing job dissatisfaction and psychiatric distress.

Coffee intake and coronary heart disease.

We examined the risk of coronary heart disease (CHD) associated with coffee intake in 1040 male medical students followed for 28 to 44 years. During the follow-up, CHD developed in 111 men. The relative risks (95% confidence interval) associated with drinking 5 cups of coffee/d were 2.94 (1.27, 6.81) for baseline, 5.52 (1.31, 23.18) for average, and 1.95 (0.86, 4.40) for most recent intake after adjustment for baseline age, serum cholesterol levels, calendar time, and the time-dependent covariates number of cigarettes, body mass index, and incident hypertension and diabetes. Risks were elevated in both smokers and nonsmokers and were stronger for myocardial infarction. Most of the excess risk was associated with coffee drinking prior to 1975. The diagnosis of hypertension was associated with a subsequent reduction in coffee intake. Negative results in some studies may be due to the assessment of coffee intake later in life or to differences in methods of coffee preparation between study populations or over calendar time.

Youthful precursors of alcohol abuse in physicians

PURPOSE This study was designed to determine youthful precursors of alcohol abuse in physicians. SUBJECTS AND METHODS We analyzed data from an ongoing prospective study of 1,014 male medical students enrolled in the graduating classes of 1948-1964 at the Johns Hopkins School of Medicine. The cohort, now physicians aged 52 to 68 years, has been contacted regularly since medical school to identify major disease. In 1986, the CAGE alcoholism screening questionnaire was administered. Alcohol abuse was defined as self-admitted alcoholism, excessive consumption of four or more beverages per day on average, or a score of 2 or higher on the CAGE questionnaire. RESULTS By these criteria, 131 of 1,014 (12.9%) patients abused alcohol. Medical school precursors associated (p less than 0.05) with subsequent alcohol abuse were as follows: non-Jewish ancestry (relative odds [RO] = 3.1), lack of religious affiliation (RO = 4.1), cigarette use of one pack or more per day (RO = 2.6), regular use of alcohol (RO = 3.6), anxiety (RO = 1.8) or anger (RO = 1.8) as a reaction to stress, frequent use of alcohol in nonsocial settings (RO = 1.6), past history of alcohol-related difficulty (RO = 3.1), and maternal alcoholism or mental illness (RO = 1.9). Precursors found not to be associated with alcohol abuse included sleep habits, use of sedatives or amphetamines, interest in athletics or hobbies, and parental relationship. CONCLUSION Our results suggest that there are several identifiable medical school precursors of alcohol abuse in physicians.

CARDIOVASCULAR DISEASE IN AFRICAN AMERICAN AND WHITE PHYSICIANS: THE MEHARRY COHORT AND MEHARRY-HOPKINS COHORT STUDIES

Differences in cardiovascular disease (CVD) were studied in a cohort of medical students from the classes of 1958-65 of Meharry Medical College (n = 435), all African Americans, and the classes of 1957-64 of Johns Hopkins University (n = 580), all white. At baseline, African Americans were older (27 vs. 24 years, p = 0.001), more likely to smoke (71 vs. 47 percent), had greater body mass index (24 vs. 23 kg/m), and had higher systolic blood pressure (120 vs. 116 mmHg). At follow-up (23-35 years later), African American physicians had higher CVD risk (RR = 1.65, 95% CI = 1.3-2.41), higher incidence of coronary artery disease (1.4 times), and much higher case fatality (51.5 vs. 9.4 percent). Risk factor levels in youth can predict CVD events several years later; predictors may differ between racial groups. Best predictors were cigarette smoking, cholesterol, and paternal history in white physicians, and blood pressure in African American physicians.

Temperament as a potential predictor of mortality: Evidence from a 41-year prospective study

Psychological factors were hypothesized to influence mortality, in particular, early versus later mortality. To explore the relationship between temperament, a psychological factor, and mortality in a prospective study of 1337 medical students, we constructed a measure portraying three temperament types, using latent class analysis. Death occurred in 113 subjects over 25–41 years of follow-up. In univariate survival analysis, subjects tending to direct tension “inward” when under stress (“Tension-In”) had a higher risk of mortality than “Tension-Out” or “Stable” types. These associations persisted after adjustment for age, smoking, cholesterol level, and Quetelet Index. The relative risk (RR) of mortality for Tension-In was 1.56 (95% confidence interval, 1.00–2.44) compared with the Stable group. The risk was due entirely to the excess risk in persons under 55 years of age (RR, 2.59; 95% confidence interval, 1.46–4.62); the corresponding risk of death in older persons was 0.66 (0.30–1.48). Thus temperament is a significant risk factor for mortality, in particular, premature death.