This observational survey was completed to assess epidemiological and clinical trends in CRC over a 37-year period, and to estimate future changes in the patient population. The overall incidence rate of CRC increased by 90% during the study period. Of this observed increase, 28% was attributed to changes in the population (age and sex), whereas 72% was related to other factors. According to our estimates, the number of new CRC patients, particularly octogenarians, will continue to rise in the coming years. We shall expect a 40% increase in 2030 and a 70% increase in 2040, compared with mean incidence rates the past 10 years.
The local incidence rate in our catchment area was somewhat below the national level in 1980–1984, but increased to the national level during the last 5 year period of the study . Our county, as well as other rural areas of Norway, has undergone some urbanisation throughout this period. Differences in lifestyle among Norwegian citizens living in the cities and in the countryside are diminishing, and the population is to an increasing extent exposed to the same risk factors. Global patterns show a marked increase in the incidence of CRC in countries adopting modern Western living habits. . Norway has enjoyed rapid social and economic development since the 1970s, in great extent due to the oil industry. There has been an increase in the rates of obesity and diabetes in our county [14,15], as well as in the rest of the country. Only 30% of the Norwegian population fulfil the recommended level of daily physical activity. On the other hand, there has been a decrease in daily smokers, from 36% in 1980 to 12% in 2018 .
Other reports have findings comparable to ours, attributing a large proportion of the increase in CRC incidence to preventable risk factors . In the United Kingdom, one-third of all cancers are attributed to smoking, and one third to diet, nutrition, and physical activity . Despite public initiatives to reduce the exposure to known risk factors – for example, advice regarding physical activity, smoking and diet – incidence levels have increased. From the present report, it seems that the effect of preventable risk factors on the incidence of CRC reached a peak around 2000–2010, with a more stable incidence in later years. Whether this is an effect of increased knowledge of risk factors and consequent behavioural changes in the population or indicates a maximum steady-state level of exposure to these risk factors in the population is disputable.
CRC is a disease with a multifactorial genesis primarily affecting the population in a sporadic manner, with a peak incidence in persons older than 70 years of age. The proportion of elderly patients has increased throughout our observation period, and this trend will continue in the future. Especially noticeable is the increasing number of patients above 85 years of age. According to the Norwegian national guidelines on CRC, a 33% increase in incidence is expected by 2024–2028, mainly due to ageing of the population . Our predictions coincide with the numbers presented in those guidelines.
Among the OECD countries, Norway is fourth in life expectancy. Other countries at the top of this list are also high HDI countries with high incidences of CRC (e.g., Switzerland, Japan, Australia, and Sweden) . Norwegian life expectancy has increased by 7.5 years since the 1980s, and we found that 28% of the increased incidence in CRC could be attributed to increased age.
The Norwegian health care system is fully funded by the government. Hence, every Norwegian citizen has access to state-of-the-art medical services, and can seek medical help at any time, regardless of income. Colonoscopy and CT are nowadays, in contrast with the 1980s, considered low-threshold examinations. General practitioners can refer patients for these examinations within 9 calendar days (fast-track examination), if cancer is suspected. This may contribute to the high incidence levels, earlier stages detected, and decrease in the number of perforations at presentation observed in Norway recently.
Decreasing incidences of CRC are observed in countries with established screening programs [19,20]. A national Norwegian screening program is currently being planned, enrolling patients at the age of 55 years. An increase in incidence rates must be expected before the incidence rates decline. Implementation of this screening program will not affect incidence among patients aged above 55 years at the time of implementation. During the first years after the Second World War, Norway experienced all-time-high birth rates. As life expectancy continues to increase in Norway, these large cohorts of elderly citizens not undergoing screening will result in an increased number of elderly CRC patients. In combination, these two factors will contribute to a peak in CRC incidence in the coming years. In a longer time-frame, however, we might observe falling incidence rates as the result of screening. Declining birth rates in Norway may augment this change in an even longer perspective.
In this study there was a trend towards earlier stages at diagnosis in recent decades. This might reflect more awareness of the disease among both patients and primary care physicians, better access to colonoscopy, and a more widespread use of CT with improved quality. These findings are contrary to other studies, which have reported unchanged or increasing rates of advanced stages with time [21-23]. Screening-detected cancer patients present with earlier stages of disease compared with non-screening-detected patients [24-28]. The patients in this study were all diagnosed before the introduction of systematic screening for CRC, indicating that the shift towards earlier stages at presentation will continue in the future. Distal localizations had earlier stages compared with proximal tumours, in accord with previous reports [29,30].
Colorectal obstruction and perforation
Previous reports found emergency presentation of colorectal cancer in 9–32% of the patients, primarily due to colorectal obstruction and bowel perforation [31-37]. The incidence of complete obstruction has been reported as 8.3 to 22.9%, and the perforation rates from 2.3 to 3.6% [31,34,36-42]. We found comparable rates, of 10.5% and 3.1% of the patients, respectively. Neither colorectal obstruction nor spontaneous perforation was associated with age in the present study, contrary to findings in previous reports . Primary tumour localization to the left flexure had the highest rate of obstruction, at 34%. Two other studies found that almost half of the tumours with this localization resulted in obstruction [42,43]. The rate of spontaneous perforation diminished significantly during the study period. This might be due to a more effective health care system with shorter waiting times prior to surgery in patients presenting with obstructive symptoms or stenotic tumours at the time of colonoscopy.
Strengths and weaknesses
This study included a complete cohort of patients diagnosed with CRC over 37 years at a single institution serving a catchment area that remained unchanged throughout the study period. All patients with suspected CRC in our region were referred to our hospital for diagnostic work-up. Data were accessible at an individual level, and completed with data from the Norwegian Cancer Registry. Preoperative examinations, treatment and follow-up followed local guidelines (standardized policies) throughout the period, and similar guidelines were implemented at a national level in 2009. As all patients were included, we avoided selection bias. The population in our county is a stable population, suitable for epidemiologic studies . The study reflects the epidemiology of elective as well as emergency admission of patients with colorectal cancer on a population basis.
The retrospective design implies certain weaknesses. The quality of the database was dependent on the quality of the individual records of the patients. By combining the data from the Norwegian cancer registry with our own database, we believe that the data used to calculate incidences were nearly complete. We may have missed some old, frail patients with symptoms of CRC who were treated at home or in nursing homes, without further investigation. The incidence in very old persons might thus be higher than reported.
Predictions of future cancer incidence depend upon a number of uncertain factors, and numbers must be interpreted with caution . The numbers of CRC cases predicted to occur by 2030 and by 2040 in the present study assumed the same age- and gender-specific incidence rates as the means of the rates that were observed during 2007–2016.
The most striking results of predicting future CRC cases occurring by 2030 and by 2040 were the continuous increase in CRC cases in our catchment area and the high numbers of octogenarians, the latter reflecting the impact of increased life expectancy in Norway in the coming years. Awareness of risk factors and systematic screening may reduce the incidence rates. Measures to also reduce the risk of CRC in the elderly non-screened parts of the population should be considered.
In the coming years, the Norwegian health care system must prepare for an increasing number of patients diagnosed with CRC. A large proportion of these patients will be 80–90 years of age. The planned national screening program will not have an impact on CRC incidence among inhabitants aged above 55 years. In the screened part of the population, an initial increase in incidence and a shift towards earlier stages of CRC at presentation should be expected. In the long run, both screening and changes in the population may result in a decline in CRC incidence. Knowledge of these changes in patient volume and characteristics is imperative in order to establish a rational and effective organization of health services to accommodate these patients.
The current study demonstrates that a substantial number of cancer cases can be attributed to preventable causes. Increased knowledge concerning these causes is imperative to complete the puzzle regarding risk factors and disease development. The adverse development regarding obesity and lifestyle-related diseases accentuates the reality that current primary preventive strategies lack effectivity. Given the fact that more than two-thirds of CRC cases might be preventable, a key question is whether changes in these factors can be expected, and what impact this might have on disease development.