| 【Objectives】(1)To investigate and analyse the epidemiological features of degenerative disc disease(DDD)in military pilots.(2)To study the risk factors and factors associated with the severity of symptoms in pilots.(3)To develop and validate a predictive model for DDD of pilots.【Methods】(1)Information on basic pilot information,flight-related characteristics,disease-related characteristics and personal habits was collected through questionnaires.The prevalence of DDD,cervical spondylosis(CS),lumbar disc herniation(LDH),degenerative lumbar spondylolisthesis,degenerative spinal stenosis,neck pain(NP),and low back pain(LBP)was calculated by combining the imaging data.The frequencies of NP and LBP,the visual analogue scale(VAS),the neck disability index(NDI),and the Oswestry disability index(ODI)were calculated to assess the severity of symptoms.The characteristics of the disease in different aircraft types were analysed.Differences in prevalence and severity between pilots and ground crews were compared and the relative risk(RR)for flight exposure were calculated.The chi-square(χ~2)test was used to analyse the nominal variables,and numerical variables were analysed using the t-test or Mann-Whitney U test.(2)Univariateχ~2 test analysis was performed to compare differences in the prevalence of DDD,CS,and LDH by characteristics of the pilots.Multivariate binary logistic regression analysis was then performed to identify the risk factors.The variables which correlated with VAS,NDI and ODI were analysed using Pearson linear correlation analysis for continuous data and Spearman rank correlation analysis for ranked data.Next,the correlation coefficient r was calculated.Variables that were significant for correlation analysis were included in the multiple linear regression model and adjusted R~2 was calculated.(3)Based on the results of the previous univariate analysis,variables with P<0.2 forχ~2 test were included to be screened.The akaike information criterion(AIC)was used to screen the models step by step,and goodness of fit of all combinations were compared.The model with the lowest AIC was selected as the best option.Finally,the model was visualised by plotting the nomogram.The receiver operating characteristic curve(ROC),calibration curve,and decision curve(DC)were plotted separately for the training and validation sets to evaluate and validate the model.【Results】(1)Overall prevalence:43.8%of DDD,25.4%of CS,29.3%of LDH,2.1%of lumbar spondylolisthesis,1.4%of lumbar spinal stenosis,64.7%of NP,and 74.2%of LBP.(2)In the analysis by aircraft type,the prevalence of DDD and CS among fighter pilots was 54.9%and 33.0%,both of which were significantly higher than those of bombers+transport+helicopter+others aircraft(P=0.007 and P=0.009).There was no significant difference in the prevalence of LDH,NP,and LBP between aircraft types.There was no significant difference in the frequency of NP and LBP between aircraft types.(3)The prevalence of DDD,CS,and NP was significantly higher in pilots than in ground crews(P<0.05),and there was no significant difference in the prevalence of LDH(P=0.554)and LBP(P=0.107).The RR and 95%confidence interval(CI)for prevalence between pilots(flight exposure)and ground crew(non-exposure)were:RR(DDD)=2.188(1.294-3.699),RR(CS)=3.143(1.404-7.036),RR(LDH)=1.308(0.670-2.553),RR(NP)=1.386(1.052-1.828),RR(LBP)=1.143(0.935-1.397).The percentage of frequency of occasional LBP was significantly higher in pilots(46.1%)than ground crews(22.1%)(P=0.002).The VAS of NP(Z=2.628,P=0.009)and NDI(Z=8.072,P<0.001)was significantly higher in pilots than ground crews.(4)Age>40 years(OR=2.022,P=0.041),overweight(OR=2.666,P=0.008),obesity(OR=5.330,P=0.065),high performance aircraft(OR=2.706,P=0.009),anxiety(OR=2.895,P=0.002),and smoking(OR=1.951,P=0.045)were independent risk factors for DDD of pilot.Age>40 years(OR=2.569,P=0.010),high performance aircraft(OR=2.466,P=0.017),and anxiety(OR=2.261,P=0.027)were independent risk factors for CS of pilot.Flight time>2000 hours(OR=2.747,P=0.012),anxiety(OR=2.174,P=0.029)and smoking(OR=2.075,P=0.035)were independent risk factors for LDH of pilots.(5)VAS of NP multiple linear regression results:stress frequency(B=0.778,P<0.001),insomnia level(B=0.300,P=0.045),helmet weight(B=0.275,P=0.014),and mean+Gz(B=0.193,P=0.034)were included in the model(F=10.826,P<0.001),and the adjusted R~2was 0.144.VAS of LBP multiple linear regression results:anxiety frequency(B=0.733,P=0.004),stress frequency(B=0.555,P=0.004),and insomnia level(B=0.419,P=0.012)were included in the model(F=19.045,P<0.001),and the adjusted R~2 was 0.183.The NDI multiple linear regression results:insomnia level(B=4.423,P<0.001),anxiety frequency(B=3.755,P<0.001),and stress frequency(B=2.468,P<0.001)were included in the final model(F=92.265,P<0.001),and the adjusted R~2 was 0.518.ODI multiple linear regression results:insomnia level(B=3.789,P<0.001),anxiety frequency(B=3.219,P<0.001),stress frequency(B=1.645,P=0.013),flight stress frequency(B=1.313,P=0.032),and age(B=0.126,P=0.008)were included in the final model(F=45.198,P<0.001),and the adjusted R~2 was 0.459.(6)The prediction model for pilot DDD contained five variables:age,body size,high performance aircraft,anxiety,and smoking.The prediction model for CS contained four variables:age,high performance aircraft,helmet weight,and anxiety.The prediction model for LDH contained four variables:body size,flight time,anxiety,and smoking.According to the ROC curve,calibration curve and DC,the predictive power of the DDD and LDH models was better than that the CS model,and the benefits of the DDD and CS models were better than that the LDH model.【Conclusions】(1)Military pilots had a high prevalence of DDD,CS,LBP,NP,and LBP,which were important problems affecting their health.(2)There were differences in the prevalence of DDD and CS between the fighter pilots and bombers+transport+helicopter+others aircraft pilots.Fighter pilots were more likely to be exposed to high+Gz and are at higher risk of DDD and CS.(3)Flight exposure increased the risk of DDD,CS,LBP,and NP.To some extent,flight exposure increased the frequency of occasional LBP and the intensity of NP.(4)Pilot DDD,CS,and LDH were influenced by a variety of risk factors,including body size,high performance aircraft,flight time,age,anxiety,and smoking.(5)Psychological factors such as anxiety,stress,and sleep status were important influences on the severity of neck and back dysfunction in pilots.(6)Pilot DDD,CS,and LDH prediction models can be used initially for the prediction and the implementation of interventions. |
PDF Full Text Request