Fitness and Body Mass Index During Adolescence and Disability Later in Life

Abstract

Noncommunicable diseases are global public health issues that lead to premature death and disability (1, 2). Psychiatric and musculoskeletal disorders, cardiovascular diseases, cancer, and injuries are a major burden on societies (1, 35). Thus, identification of early and potentially modifiable risk factors for later chronic disease is of great public health importance. An objective approach to studying severe chronic diseases is to use receipt of a disability pension as a health outcome. In many countries, disability pensions are granted to working-aged persons who are likely to never work full-time again because of a chronic disease or injury diagnosed by a physician. In addition to serving as an important indicator of chronic disease, disability pensions are associated with high societal costs (6) and thus have not only clinical but also economic consequences. Low cardiorespiratory fitness in childhood and adolescence has been associated with increased risk for death and disability later in life (79). However, few studies have examined associations of cardiorespiratory fitness in youth with later receipt of a disability pension. A study of 49321 Swedish military conscripts aged 18 to 20 years showed that low cardiorespiratory fitness was associated with higher risk for receipt of a disability pension due to all causes, although it did not examine specific causes (10). Therefore, large studies that are powered to examine the association of cardiorespiratory fitness with receipt of a disability pension for specific causes are needed. Obesity in youth has been found to be related to impaired health and premature death (11). Although previous studies have reported that obesity is associated with increased risk for receipt of a disability pension (6, 12), the association between severe obesity (body mass index [BMI] 40.0 kg/m2) in youth and later chronic disability (as indicated by receipt of a disability pension) is unknown. High cardiorespiratory fitness has been shown to potentially attenuate the negative effects of obesity (13, 14). However, the combined association of cardiorespiratory fitness and obesity with receipt of a disability pension has yet to be examined. The aim of this study was to examine the associations of cardiorespiratory fitness and obesity with receipt of a disability pension in a large sample (>1 million) of Swedish male adolescents. Given the high power of the study, we examined individual and combined associations of cardiorespiratory fitness and obesity with receipt of a disability pension due to all and specific causes. Methods Study Design and Population This prospective cohort study used data from the Swedish Military Service Conscription Registry, which was linked to several other national registries using the men s unique personal identification numbers as described previously (12). The Cause of Death Register and the Registry of the Total Population were used to identify men who died or emigrated during follow-up. The study was approved by the Regional Ethical Review Board, Stockholm, Sweden. The cohort comprised male adolescents who were born in Sweden between 1951 and 1976, were conscripted into the military between 1972 and 1994, and were followed until 31 December 2012. Military conscription was mandatory for the study participants; only 2% to 3% of Swedish adolescents were exempted from conscription because of incarceration or severe medical conditions (15). In this study, we included male adolescents who were aged 16 to 19 years at conscription and had complete data on the exposures, outcomes, and covariates. In accordance with previous studies (12), we excluded males with extreme values for height (150 or 210 cm), weight (40 or 150 kg), or BMI (15 or 60 kg/m2). The overall cohort consisted of 1125739 adolescents, of whom 1083561 had complete data on cardiorespiratory fitness (the primary exposure). Another 4433 had missing or extreme values for the other exposures, outcomes, and covariates, leaving 1079128 adolescents (95.9%) for the analyses. Cardiorespiratory Fitness, BMI, and Covariates Cardiorespiratory fitness was measured in conscripts with a normal electrocardiogram using an electrically braked ergometer cycle test, as described previously (16). The initial resistance was determined by weight and was increased by 25 W for men who reported regular exercise. After a 5-minute warm-up period with a pulse between 120 and 170 beats/min, the resistance was further increased by 25 W/min until volitional exhaustion. The final work rate was retained and used in the analysis. Weight was measured using analogue or digital scales, and height was assessed using wall-mounted stadiometers according to standardized procedures (7). Body mass index was calculated as weight in kilograms divided by height in meters squared and was classified according to World Health Organization criteria (17). Year of conscription, conscription center (6 centers), and age at conscription were obtained from the Military Service Conscription Registry, and information on the occupations of participants parents was retrieved from the Population Housing Censuses. Childhood socioeconomic status was defined as the highest-level occupation of either parent. Disability Pension Data on the date and cause of receipt of a disability pension were retrieved from the Social Insurance Agency for 1972 to 2012. During the study years, the cause of receipt of a disability pension had to be confirmed by a certificate from a physician using International Classification of Diseases (ICD) codes (ICD-8, ICD-9, or ICD-10), and the work capacity had to be reduced by at least 25%. The codes were then used to classify the cause as psychiatric (for example, affective and nonaffective disorders), musculoskeletal (for example, dorsalgia and soft tissue disorders), nervous system (for example, multiple sclerosis), circulatory (for example, cerebrovascular and ischemic heart diseases), injuries, or tumors (Appendix Table 1). Appendix Table 1. ICD Codes Used in the Study Statistical Analysis We used Cox proportional hazards regression models to calculate hazard ratios (HRs) with 95% CIs and to estimate cumulative incidences. Men were followed until they were granted a disability pension, died, or emigrated or until the end of follow-up (31 December 2012), whichever occurred first. We first analyzed the associations of cardiorespiratory fitness and BMI with later receipt of a disability pension due to all and specific causes. We created 2 models: the first included basic covariates (conscription year, conscription center, age at conscription, and childhood socioeconomic status), and the second included these covariates plus BMI or cardiorespiratory fitness. Analyses involving receipt of a disability pension due to all and psychiatric causes were also adjusted for any mental hospitalization before conscription and any psychiatric diagnosis at conscription. In the analyses that included BMI categories, we examined the association of severe obesity with receipt of a disability pension due to all causes as well as psychiatric and musculoskeletal causes (which were most prevalent) because there were too few cases for the remaining causes. We also assessed the combined association of cardiorespiratory fitness and BMI categories with later receipt of a disability pension, with adjustment for the aforementioned basic covariates. In these analyses, we investigated receipt of a disability pension due to all, psychiatric, and musculoskeletal causes. Adolescents in the first 2 deciles were classified as unfit, and fit adolescents were further divided into moderately fit (deciles 3 to 8) and highly fit (deciles 9 and 10) categories. The proportional hazards assumption for the exposures (fitness, BMI, and both combined) in the Cox regression was examined using a log-minus-log plot, and we found no evidence that the assumption was violated. Statistical analyses were conducted using SPSS Statistics, version 22 (IBM). Role of the Funding Source Karolinska Institutet had no role in the design or conduct of the study, management of the data, interpretation of the results, or the decision to submit the manuscript for publication. Results Descriptive Statistics The Table shows descriptive data on the 1079128 participants. Over a median follow-up of 28.3 years (30.6 million person-years), 54304 men were granted a disability pension. Figure 1 shows the unadjusted cumulative incidences of receipt of a disability pension due to all causes by cardiorespiratory fitness level (left panel) and BMI category (right panel). The cumulative incidence was consistently higher in adolescents with lower cardiorespiratory fitness and a higher BMI during follow-up. Additional data on cumulative incidences by cardiorespiratory fitness level and BMI category are shown in Appendix Figures 1, 2, 3, 4, 5 and 6 and Appendix Tables 2 and 3. Table. Descriptive Data on Study Participants (n= 1079128) Figure 1. Unadjusted cumulative incidences of receipt of a disability pension due to all causes, by cardiorespiratory fitness level (left) and BMI category (right). Cumulative incidences were estimated from unadjusted Cox regression models. BMI= body mass index; D= decile. Appendix Figure 1. Unadjusted cumulative incidences of receipt of a disability pension due to psychiatric causes, by cardiorespiratory fitness level (left) and BMI category (right). Cumulative incidences were estimated from unadjusted Cox regression models. BMI = body mass index. Appendix Figure 2. Unadjusted cumulative incidences of receipt of a disability pension due to musculoskeletal causes, by cardiorespiratory fitness level (left) and BMI category (right). Cumulative incidences were estimated from unadjusted Cox regression models. BMI = body mass index. Appendix Figure 3. Unadjusted cumulative incidences of receipt of a disability pension due to injuries, by cardiorespiratory fitness level (left) and BMI catego