Epidemiological characteristics
Over time, the incidence increased in case numbers, with a trend towards gender balance from a predominance of men. The prevalence doubled in the observed period, from 5.967 patients by end of year 2005 to 10.394 patients by the end of 2015. The annual number of deaths was higher for males than for females, with a stable total number around 4,000 annually in spite of the increasing prevalence (Table 1).
Stage, pathology and comorbidity
Of the 44,291 patients diagnosed from 2006 through 2015 a detailed TNM stage was not reported for 8%. The stage distribution is presented in (Figure 1) where it is seen that the relative numbers of patients in stage classes IV and IIIB declined weakly whereas a relative increase in patients in stage classes 0 – IIIA, was observed. The increasing trend in the proportion of patients diagnosed with stage classes 0-IIIA was statistically significant (P<0.0001).
The number of patients without a specific morphology diagnosis decreased from 14% in 2006 to 8% in 2015. The proportion of patients diagnosed with adenocarcinoma increased statistically significant (P<0.0001) with a corresponding decrease in the number of patients with large cell carcinoma and other primary lung cancer (Figure 2). At high level morphology classification, the proportions of patients with squamous cell and small cell carcinomas have been relatively stable
The proportion of patients with comorbidity (CCI score>0) increased over time (Table 2). Thus, the proportion of patients without comorbidity (CCI score=0) decreased from 64 % in 2006 to 59 % in 2015. This trend was statistically significant (P<0.0001).
The statistical analysis of the incidence and mortality rates is summarized in (Table 3). During the study period, the incidence rate for females was 20% lower (P<0.0001) compared with the rate for males. For both sexes, the incidence rates increased statistically significantly (P<0.0001) for the age groups 60-69 and 70-79 years using the age group <60 years as the reference. The incidence rate level for the age group >80 years was interposed between those by the age groups 60-69 and 70-79 years. With the inclusion of age groups and sex as covariates the incidence rate decreased statistically significantly by 0.5% annually (P=0.005).
During the study period, the mortality rate for females was reduced by 19% (P<0.0001) compared with the rate for males. Compared with the age group <60 years, the mortality rate was not different for the age group 60-79 (6% increase, P=0.242), but was statistically significantly increased for the age groups 70-79 and >80 years by 21% (P<0.001) and 76% (P<0.0001), respectively. The mortality rate decreased significantly over time by 6%, P<0.000). Compared with no comorbidity (CCI=0) the mortality rate increased with increasing comorbidity by 44%, 34% and 86% (P<0.0001 for all) for CCI=1, CCI=2, and CCI=>2, respectively.
Epidemiological projections
The age-standardized summary rate is predicted to decrease slightly, but due to the ageing population the crude rate is modestly increasing (Figure 3 -top). The mortality in lung cancer is decreasing for all age-specific rates as well as for the crude (unadjusted) rate, and this trend is predicted to continue according to an exponential forecasting model (Figure 3 - bottom).
The observed and projected annual case numbers of incidence and mortality are showed in (Figure 4), with the corresponding annual growth. The age-standardized summary rate is predicted to decrease slightly, but due to the ageing population the crude rate is modestly increasing. The mortality in lung cancer is decreasing for all age-specific rates as well as for the crude rate, and this trend is predicted to continue according to an exponential forecasting model.
Using the annual numbers of new cases and deaths together with the number of prevalent cases by the end of 2005 as inputs to the ‘stock and flow’ model, annual prevalence numbers are estimated as shown in (Figure 5) together with the observed values for the period from the end of 2005 to the end of 2015. The modelled trend in prevalence numbers follows closely the observations. According to the model, the numbers of prevalent cases of lung cancer in Denmark will more than double from about 10,000 at the end of 2015 to about 23,000 at the end of 2030. In the reference scenario with constant sex and age specific incidence and mortality rates, the increase in prevalence is less marked with an estimated number of prevalent cases at about 12,500 at the end of 2030.
Using the modelled NORDCAN data from the core model of projection, the annual absolute number of incident cases increased by 13% from year 2006 through year 2015. Demographical changes would have increased the annual case number by 17%, but this is reduced by 4% due to the slight decreasing trend in the general incidence level. According to the core model, the annual number of incident cases will increase by 7% from 2016 through 2030. The expected demographic changes considered isolated would lead to an increase at 22%, but this is impaired negatively by 15% due to the expected decreasing trend in the incidence level.
During the period end of 2005 through end of 2015 the absolute number of prevalent cases increased by 76% according to the core model. The effect of epidemiological disequilibrium at the end of 2005 account for an increase at 8% and demographical changes account for additional 14%. However, the decreasing trend in incidence level has reduced the increase in prevalence by 6%. The remaining contribution to the increase in prevalence amounts to 60% and is accounted for by decreasing mortality level during the 10-years period.
For the period from the end of 2015 through the end of 2030, the core model predicts an increase in the absolute prevalent case number at 119%. The attributions to this increase are estimated as follows: Effect of epidemiological disequilibrium: 9%; demographical changes: 14%; changes in incidence level: -12%; changes in mortality level: 107%.