Anastomotic leakage after surgery for rectal cancer: a risk factor for local recurrence, distant metastasis and reduced cancer-specific survival?

AIM: The impact of anastomic leakage (AL) on the oncological outcome after anterior resection (AR) for rectal cancer is still controversial. We explored the impact of AL regarding local recurrence (LR), distant metastasis and overall recurrence (OAR). Overall and cancer-specific survival was analysed. METHOD: Patients undergoing AR for rectal cancer with a registered AL between 1995 and 1997 and a control group were identified in the Swedish Rectal Cancer Registry. The medical records were retrieved for additional data and validation. Differences in the oncological outcome at 5-year follow-up were analysed with multivariate methods. RESULTS: After validation, 114 patients with AL and 136 control patients with locally radical surgery for tumours in tumour-node-metastasis stages I-III were analysed. There was no difference detected between patients with AL and control patients regarding rates of LR [8% (9 of 114) vs 9% (12 of 136); P = 0.97], distant metastasis [18% (20 of 114) vs 23% (31 of 136); P = 0.37] and OAR [19% (22 of 114) vs 28% (38 of 136); P = 0.15]. The 5-year cancer-specific survival was almost 80% in both groups. In multivariate analysis, AL was not a risk factor of LR, distant metastasis or OAR and had no impact on 5-year overall or 5-year cancer-specific survival. Irrespective of the occurrence of AL, preoperative radiotherapy (P = 0.055) and rectal washout (P = 0.046) reduced the LR rate, but did not influence survival. CONCLUSION: Anastomotic leakage was not proved to be a risk factor of worse oncological outcome. Hence, additional adjuvant treatment or extended follow-up on the basis of the occurrence of AL after AR might not be justified.

Risk factors of rectal cancer local recurrence: population-based survey and validation of the Swedish rectal cancer registry.

AIM: Despite advances in rectal cancer treatment, local recurrence (LR) remains a significant problem. To select high-risk patients for different treatment options aimed at reducing LR, it is essential to identify LR risk factors. METHOD: Local recurrence and survival rates of 4153 patients registered 1995-1997 in the Swedish Rectal Cancer Registry were analysed. LR risk factors were analysed by multivariate methods. For LR patients the registry was validated and additional data retrieved. RESULTS: The 5-year overall and cancer-specific survival rates were 45% and 62% respectively. LR was registered in 326 (8%) patients. After R0-resections for tumours in TNM stages I-III, LR developed in 10% of tumours at 0-5 cm, 8% at 6-10 cm and 6% at 11-15 cm above the anal verge. Preoperative radiotherapy (RT) reduced the LR rate irrespective of height [0-5 cm: OR 0.50 (0.30-0.83), 6-10 cm: OR 0.42 (0.25-0.71), and 11-15 cm: OR 0.29 (0.13-0.64)]. Patients without preoperative RT had significantly higher LR risk after rectal perforation [OR 2.50 (1.48-4.24)], and almost significantly decreased LR risk when rectal washout was performed [OR 0.65 (0.43-1.00)]. Preoperative RT prolonged time to LR but did not significantly influence the survival among LR patients. LR was an isolated tumour manifestation in 103 (39%) patients with validated LR. CONCLUSION: Preoperative RT should be considered for rectal cancer also in the upper third of the rectum. Intraoperative perforation should be avoided, and rectal washout is indicated as valuable. Follow-up for the detection of isolated LR is important. Extended follow up should be considered for patients treated with RT.

Fifty-year follow-up of mortality among a cohort of iron-ore miners in Sweden, with specific reference to myocardial infarction mortality.

OBJECTIVES: This study investigates both general mortality and mortality from myocardial infarction among men employed in iron-ore mines in Sweden. METHODS: The mortality of employees (surface and underground workers) at the iron-ore mines in Malmberget and Kiruna, Sweden was investigated. The study cohort comprised men who had been employed for at least 1 year between 1923 and 1996. The causes of death were obtained from the national cause of death register from 1952 to 2001. Indirect standardised mortality ratios (SMR) were calculated for four main causes. Mortality specifically from myocardial infarction was also analysed. RESULTS: 4504 deaths in the cohort gave an SMR for total mortality of 1.05 (95% CI 1.02 to 1.09). Mortality was significantly higher for lung cancer (SMR 1.73, 95% CI 1.52 to 1.97). There was an increased risk of injuries and poisonings (SMR 1.34, 95% CI 1.24 to 1.46) and respiratory diseases (SMR 1.14, 95% CI 1.00 to 1.28). There were 1477 cases of myocardial infarction, resulting in an SMR of 1.12 (95% CI 1.07 to 1.18). SMR was higher (1.35, 95% CI 1.22 to 1.50) for men aged 60 years of age (1.06, 95% CI 1.00 to 1.13). CONCLUSIONS: Mortality from myocardial infarction was higher than expected. There was also an increased risk of death from injuries and poisonings, lung cancer and respiratory diseases, as well as higher general mortality. Our findings support the results of previous studies that there is an association between working in the mining industry and adverse health outcomes.

The Swedish rectal cancer registry.

BACKGROUND: An audit of all patients with rectal cancer in Sweden was launched in 1995. This is the first report from the Swedish Rectal Cancer Registry (SRCR). METHODS: Between 1995 and 2003, 13 434 patients treated for adenocarcinoma of the rectum were registered with the SRCR; there were approximately 1500 new patients annually. RESULTS: Approximately half had an anterior resection, a quarter an abdominoperineal resection and 15 per cent a Hartmann's procedure. The median 30-day postoperative mortality rate was 2.4 per cent and the overall postoperative morbidity rate was 35.0 per cent. The 5-year cancer-specific survival rate was 62.3 per cent. The 5-year relative survival rate was 70.1 per cent after anterior resection, 59.8 per cent after abdominoperineal resection and 39.8 per cent after a Hartmann's procedure. The crude 5-year local recurrence rate was 9.5 per cent overall, 6.1 per cent after preoperative radiotherapy and 11.4 per cent after surgery alone. For 3868 patients who had a locally curative procedure the local recurrence rate was 7.4 per cent overall, 5.9 per cent for those who had radiotherapy and 10.2 per cent for those who did not. The local recurrence rate was 2.9 per cent (28 of 968) for stage I disease, 7.9 per cent (112 of 1418) for stage II, 13.9 per cent (188 of 1357) for stage III and 8.5 per cent (45 of 532) for stage IV. CONCLUSION: These good population-based results are due, in part, to the nationwide prospective quality assurance registration.

Does prophylactic breast irradiation prevent antiandrogen-induced gynecomastia? Evaluation of 253 patients in the randomized Scandinavian trial SPCG-7/SFUO-3.

OBJECTIVES: To examine the development of antiandrogen-induced gynecomastia and breast tenderness in the first 253 patients in a randomized Scandinavian trial (SPCG-7/SFUO-3) with a 12-month complete follow-up evaluation performed by both doctors and patients. METHODS: In this study, the treating doctor and patient decided whether prophylactic irradiation (RT) of the breast should be given to prevent antiandrogen-induced gynecomastia. At each visit, the doctor evaluated the occurrence of gynecomastia and breast tenderness. Questions about gynecomastia and breast tenderness were also included in the study quality-of-life questionnaire (Prostate Cancer Symptom Scale). RESULTS: Mammary RT with mostly single fraction (12 to 15 Gy) electrons was given to 174 (69%) of the 253 evaluated patients. At the 1-year follow-up visit, the doctor evaluations indicated some form of gynecomastia in 71% and 28% (P <0.001) of the nonirradiated (no-RT) and irradiated (RT) patients, respectively. The patient evaluations at 1 year showed some form of breast enlargement in 78% and 44% (P <0.001) of the no-RT and RT patients, respectively. The doctors reported some form of breast tenderness at 1 year in 75% and 43% (P <0.001) of the no-RT and RT patients, respectively. The patient evaluations of breast tenderness show an expected significant increase in the RT arm at the 3-month follow-up, which was probably due to skin reactions. At 1 year, significantly more patients who marked "very much" on the Prostate Cancer Symptom Scale were seen in the no-RT group. A weak correlation between the doctors' and patients' detection of breast problems was observed. CONCLUSIONS: The results show that, with high significance, prophylactic RT of the breast decreases the risk of antiandrogen-induced gynecomastia and breast tenderness.

Residential radon and lung cancer among never-smokers in Sweden.

In this study, we attempted to reduce existing uncertainty about the relative risk of lung cancer from residential radon exposure among never-smokers. Comprehensive measurements of domestic radon were performed for 258 never-smoking lung cancer cases and 487 never-smoking controls from five Swedish case-control studies. With additional never-smokers from a previous case-control study of lung cancer and residential radon exposure in Sweden, a total of 436 never-smoking lung cancer cases diagnosed in Sweden between 1980 and 1995 and 1,649 never-smoking controls were included. The relative risks (with 95% confidence intervals in parentheses) of lung cancer in relation to categories of time-weighted average domestic radon concentration during three decades, delimited by cutpoints at 50, 80, and 140 Bq m(-3), were 1.08 (0.8--1.5), 1.18 (0.9--1.6), and 1.44 (1.0--2.1), respectively, with average radon concentrations below 50 Bq m(-3) used as reference category and with adjustment for other risk factors. The data suggested that among never-smokers residential radon exposure may be more harmful for those exposed to environmental tobacco smoke. Overall, an excess relative risk of 10% per 100 Bq m(-3) average radon concentration was estimated, which is similar to the summary effect estimate for all subjects in the main residential radon studies to date.

[Quality registry for better treatment of rectal cancer established. Economical support is required to use the benefits of the information]. / Kvalitetsregister etablerat för bättre rektalcancerbehandling. Nu krävs ekonomiskt stöd för att informationen skall komma till nytta.

The treatment of rectal cancer has changed significantly during the last 30 years. With improved surgical technique and the introduction of preoperative radiotherapy sphincter preserving surgery is now predominant and the rate of local recurrence has been reduced substantially. However, new therapy concepts may also introduce an increased risk of complications. A register to monitor quality control in rectal cancer treatment in Sweden was established in 1995. It covers over 95 per cent of the patients with rectal cancer reported to the Swedish National Cancer Registry. Collection of data and validation are done by six regional oncology centres under supervision of surgeons appointed by the hospitals involved. The results are then collated to a nationwide quality register, enabling regions to compare themselves with other regions, and hospitals with other hospitals.

Cancers in concrete workers: results of a cohort study of 33,668 workers.

OBJECTIVES: To study cancer morbidity patterns in concrete workers. METHODS: A cohort of 33,503 concrete workers was enrolled in the study from 1971-86. The average duration of follow up was 19.4 years (582,225 person-years). The workers' cancer morbidity was compared with the morbidity of the general population. RESULTS: A total of 3572 incident cancers were observed. Significantly increased standardised incidence ratios (SIRs) were found for all malignant neoplasms (SIR 107; 95% confidence interval (95%CI) 103 to 110), cancer of the lip (SIR 179; 95%CI 134 to 234), cancer of the stomach (SIR 139; 95%CI 122 to 158), cancer of the lung (SIR 125; 95%CI 114 to 137), and cancer of the prostate (SIR 108; 95%CI 101 to 116). Reduced risk was found for cancer of the large intestine (SIR 80; 95%CI 69 to 93) and cancer of the testis (SIR 50; 95%CI 26 to 87). Smoking was more prevalent among the concrete workers than in the general population (50% v 35%). CONCLUSION: The study has shown a slightly increased overall risk of cancer among concrete workers. The increased risk of lung cancer could entirely be due to differences in smoking habits between concrete workers and the general population. There is a possibility that the smoking also has contributed to the increased risks of stomach cancer and lip cancer, but occupational factors may have contributed to these cancer sites.

Characteristics of pregnancy and birth and malignancy in the offspring (Sweden).

OBJECTIVES: The aim of the study was to investigate whether factors of pregnancy and birth influence the risk of malignancy in the offspring. METHODS: Data on all deliveries (248,701 births) in two counties in Sweden 1955-90 were extracted from two birth registries. The follow-up period closed at the end of 1994 and the subjects were followed up to early middle-age at most (39 years). Incidence rates of malignancy were obtained from the Cancer Register 1958-1994. Standardized incidence ratios (SIR) and relative risks (RR) were calculated. RESULTS: Overall, few associations were detected. A significantly increased standardized incidence ratio (SIR) of 50.00 (95% CI = 13.45-99.99) was found for the relationship between Down's syndrome and lymphatic leukaemia. Elder maternal age (> or =35 years) and lymphatic leukaemia were associated with a significantly enhanced risk (SIR = 2.00; 95% CI, 1.16-3.20). Maternal age 25-34 years, compared to younger age, was associated with a reduced risk of cervical cancer (RR = 0.47; 95% CI = 0.26-0.86). CONCLUSIONS: Although some associations, the consistent pattern of non-association indicated a low impact of intrauterine environment or changed genetic material on the future development of malignancy in the offspring up to early middle-age.

No difference in survival between sporadic, familial and hereditary prostate cancer.

OBJECTIVE: To estimate the survival of men with familial prostate cancer and compare them with prostate cancer cases unselected for family history. PATIENTS AND METHODS: The overall and prostate cancer-specific survival was calculated in two large (249 and 304 men, respectively) population-based cohorts of men with familial prostate cancer. The tumour grade at diagnosis was also obtained in one of the cohorts. RESULTS: There were no significant differences in either overall or prostate cancer-specific survival between familial and sporadic cases. The spectrum of tumour grades at diagnosis in familial cases did not differ from that in a population with prostate cancer unselected for family history. CONCLUSION: No differences in treatment between men with or without a positive family history of prostate cancer are justified, based on the result from this study.

Familial prostate cancer and possible associated malignancies: nation-wide register cohort study in Sweden.

There is a familial aggregation of prostate cancer, and 5 to 10% of all prostate cancers are estimated to be inherited in an autosomal-dominant mode. A population-based cohort study was performed in order to study familial prostate cancer and associated malignancies. A nation-wide register cohort study was conducted using an unselected study population. The cohort of 5,595 sons and 5,089 daughters of Swedish men found to have prostate cancer between 1959 and 1963 was identified. All types of cancer reported between 1958 and 1992 in this cohort were identified through linkage to the Swedish Cancer Registry. The expected number of different cancers was calculated using incidence rates obtained from the Registry. A highly significant increased overall standardized incidence ratio (SIR) of 1.65 (95% CI, 1.49-1.83) was obtained for prostate cancer, with 370 observed cases compared with 224 expected prostate cancers. The SIR was 3.18 among cases 45 to 49 years old at diagnosis, with the risk gradually decreasing to a SIR of 1.45 among cases over 80 years of age. Among sons and daughters with a father whose prostate cancer was diagnosed at an early age (<70 years), an increased risk for colorectal cancer SIR 1.48 (1.10-1.95) was observed. No significant difference in cancer risk for other sites was observed among the daughters and sons of men with prostate cancer. This cohort study confirms earlier studies that a positive family history of prostate cancer is an important risk factor for developing this disease. Though increased risk was found for all ages, it was more pronounced in younger men. Since no other malignancy was significantly associated with prostate cancer, it is most likely that familial prostate cancer is "site-specific".

Segregation analysis of prostate cancer in Sweden: support for dominant inheritance.

Several epidemiologic studies have shown familial aggregation of prostate cancer. To assess the nature of familial clustering of prostate cancer, a complex segregation analysis was performed on a population-based sample of 2,857 nuclear families ascertained through an affected father diagnosed with prostate cancer in Sweden during 1959-1963. The segregation analysis, using a large, unselected population of prostate cancer families, revealed that the observed clustering of prostate cancer was best explained by a high risk allele inherited in a dominant mode, with a high population frequency (1.67%) and a moderate lifetime penetrance (63%). This study confirms the result from one earlier published segregation analysis and provides the context for interpreting the recently published linkage of hereditary prostate cancer families to chromosome 1q 24-25 (HPC1).

Neoadjuvant chemotherapy with cisplatin and 5-fluorouracil in advanced squamous cell carcinoma of the head and neck: a randomized phase III study.

BACKGROUND AND PURPOSE: In 1986 a prospective, randomized, multi-centre trial for evaluation of neoadjuvant chemotherapy with cisplatin and 5-fluorouracil in the treatment of advanced squamous cell carcinoma of the head and neck was initiated. As survival in this group of patients is poor the purpose was to find a possible survival benefit of the chemotherapy in addition to radiotherapy compared to radiotherapy only. METHODS: Four-hundred sixty-one patients from Denmark, Norway and Sweden with tumors in oral cavity, oropharynx, hypopharynx and larynx were randomized to receive either standard treatment (radiotherapy or radiotherapy followed by surgery) or neoadjuvant chemotherapy followed by standard treatment. Chemotherapy included three courses of cisplatin 100 mg/m2 i.v. infusion on day 1 followed by 5-fluorouracil 1000 mg/m2 per day continuous i.v. infusion for 120 hours. Radiotherapy 64-70 Gy in 2 Gy per fraction, 5 times/week, was given to patients in both treatment arms. RESULTS: Response rate was 71% for patients randomized to chemotherapy-radiotherapy and 66% for patients randomized to standard treatment (not statistically significant). Residual tumors were excised if possible. After surgery 62% of the patients randomized to chemotherapy-radiotherapy and 60% of the patients in the standard treatment group were clinically tumor free. CONCLUSIONS: No statistically significant benefit in survival was observed for patients treated with neoadjuvant chemotherapy followed by radiotherapy. Nor was there any impact of chemotherapy on the number of patients achieving loco-regional tumor control after primary treatment.

Prostate cancer mortality in northern Sweden, with special reference to tumor grade and patient age.

OBJECTIVES: This study is designed to estimate the prostate cancer-specific mortality according to patient age and tumor grade in an unselected population of patients with prostate cancer who mostly received deferred or direct hormonal treatment as initial treatment. METHODS: The study population was composed of 6514 patients diagnosed with prostate cancer during 1971 to 1987 in northern Sweden. For those who died during follow-up, the cause of death was determined from the comprehensive Swedish registry data (population registries and causes of death registry). RESULTS: About 85% of these patients died during the 7 to 23 years of follow-up, and the prostate cancer-specific mortality was estimated to be 55%. Age at diagnosis was found to be a strong predictor of prostate cancer death. Patients diagnosed before the age of 60 had an 80% risk of dying of prostate cancer, whereas those over 80 years of age at diagnosis had less than a 50% risk of prostate cancer-related death. CONCLUSIONS: The prostate cancer mortality is high but decreases with older age at diagnosis. We found, using data from the causes of death registry, that the relative survival and the cause-specific survival of these patients were compatible with each other.

Residential radon and lung cancer in Sweden: risk analysis accounting for random error in the exposure assessment.

A large epidemiologic study on residential radon exposure and lung cancer has been conducted in Sweden. An attempt is now made to quantify the impact of random error in the exposure assessment on the risk estimate for lung cancer in this study. The study included 1,360 lung cancer cases, diagnosed from 1980 to 1984, and 2,847 population controls. Radon measurements were performed in 8,992 dwellings occupied by the study subjects some time since 1947. Questionnaires provided information on smoking and other risk factors. Imprecision in the retrospective exposure assessment for radon was estimated from a Monte Carlo technique modeling Swedish conditions. Adjusted risk estimates were obtained from regression analyses based on expected values for true time-weighted average residential radon concentration (TWA), conditional on observed TWA. Without adjustment for random error in the TWA estimates, the linear excess relative risk coefficient was 0.10 per 100 Bq m(-3), but an excess relative risk of about 0.15 to 0.20 per 100 Bq m(-3) was suggested following adjustment. The potentially significant consequences of errors in the retrospective radon exposure assessment should be taken into consideration in the risk estimation as well as in comparisons of results of different studies and in future pooled analyses.

Total food consumption and body mass index in relation to prostate cancer risk: a case-control study in Sweden with prospectively collected exposure data.

PURPOSE: Prostate cancer is the most common cancer among men and accounts for most cancer related deaths in Sweden today. To find or confirm exogenous risk factors for prostate cancer a population based case-control study was performed. MATERIALS AND METHODS: By linking the Swedish Cancer Registry with the Swedish Twin Registry 406 cases of prostate cancer were selected from the twin register. As controls 1,218 men without prostate cancer were randomly selected from the same register. The selection procedure ensured that no cases or controls were related to each other. Questionnaire concerning height, weight, dietary habits, and alcohol and tobacco consumption were mailed in 1967 and 1970 to members in the twin register and the collected information was used in this study. RESULTS: There was a positive trend for prostate cancer risk seen for total food consumption to (p < 0.001) with an odds ratio of 2.22 (95% confidence interval 1.23 to 3.99) for those who consumed somewhat more and 3.89 (1.09 to 13.96) for those who consumed much more than people in general. An increased trend was also seen for body mass index (BMI) (p = 0.015), with an odds ratio of 1.44 (0.98 to 2.11) for 26 to 29 kg./m.2 and 1.80 (1.07 to 3.04) for BMI greater than 29 kg./m.2 compared with BMI less than 23 kg./m.2. Total food consumption and BMI remained independent risk factors in a multivariate analysis. All specific food items studied, as well as tobacco and alcohol consumption, were unrelated to prostate cancer risk. CONCLUSIONS: This study suggests that high BMI and total food consumption are independent risk factors for prostate cancer and that dietary habits are important in the development of this tumor. It is also unlikely that neither tobacco nor alcohol use substantially changes the risk of prostate cancer.

Familial prostate cancer in Sweden. A nationwide register cohort study.

BACKGROUND: Although prostate carcinoma is not widely recognized as a familial cancer, familial aggregation of this disease has been shown in some retrospective case-control studies. To study familial prostate cancer in Sweden, a population-based cohort study was performed, that attempted to avoid possible bias connected with some earlier studies of familial prostate cancer. METHODS: A nationwide register cohort study was conducted using an unselected study population. The study cohort of 5496 sons of Swedish men found to have prostate cancer between 1959 and 1963 was identified through parish offices. All prostate cancer patients reported between 1958 and 1990 in this cohort were identified through linkage to the Swedish Cancer Register. The expected number of prostate cancer patients was calculated using incidence rates obtained from the same register. RESULTS: A highly significant increased overall standardized incidence ratio (SIR) of 1.70 (95% confidence interval, 1.51-1.90) was obtained for prostate cancer in this cohort, with 302 observed cases compared with 178 expected prostate cancers. The SIR was 3.38 among patients aged 45-49 years at diagnosis, with the risk gradually decreasing to a SIR of 1.35 among patients older than 80 years (trend, P = 0.013). Among sons with a father whose prostate cancer was diagnosed at an early age (< 70 years), a significant trend (P = 0.01) for prostate cancer risk was observed due to early onset of the disease. CONCLUSIONS: This cohort study provides further evidence that a positive family history of prostate cancer is a risk factor for developing the disease in an unselected population. The increased risk was found for all ages, but was more pronounced in younger men.

A cohort study with regard to the risk of haematological malignancies in patients treated with x-rays for benign lesions in the locomotor system. I. Epidemiological analyses.

Roentgen treatment for painful benign conditions in the locomotor system as arthrosis and spondylosis was in Sweden very common up to the beginning of the 1960s. The mode of treatment differed from the British ankylosing spondylitis series as smaller parts of the red bone marrow were exposed and smaller doses were applied. A cohort of 20,024 such patients treated 1950-1964 at two hospitals in northern Sweden was analysed with regard to the risk of haematological malignancies. Average factors for conversion of prescribed skin doses to mean absorbed red bone marrow doses were estimated on random samples of the different treatment sites and then applied on the cohort in its whole. The standard incidence ratio (SIR) for leukaemia was 1.18 (95% CI: 0.98-1.42) and the standard mortality ratio (SMR) 1.25 (0.99-1.45). In the highest dose group (mean absorbed red bone marrow dose > 0.5 Gy) the corresponding values were 1.40 (1.00-1.92) and 1.50 (1.08-2.04). In the mortality analysis also a slightly increased myeloma risk was noted with SMR = 1.20 (0.99-1.56). Extension of the cohort and nested case-control studies are under progress.

A cohort study with regard to the risk of haematological malignancies in patients treated with x-rays for benign lesions in the locomotor system. II. Estimation of absorbed dose in the red bone marrow.

A cohort study with regard to the risk of haematological malignancies was performed on about 20,000 patients who in 1950-1964 received roentgen treatment for benign conditions in the locomotor system. In order to estimate the mean absorbed red bone marrow dose the treatments were classified as concerning 10 sites (cervical spine, thoracic spine, lumbar spine, sacral region, shoulder, hip, elbow, wrist, knee and ankle). The four last-mentioned sites do not normally contain red bone marrow in adults and their contribution to the mean absorbed dose was regarded as zero. For the other 6 sites random samples consisting of 30 patients for each site were drawn from the cohort. By use of the treatment records and data from the literature on some physical parameters and red bone marrow distribution in normal adult persons, average conversion factors were calculated by which the subscribed surface dose could be converted into mean absorbed dose in red bone marrow. These conversion factors were then applied on the whole cohort and used for stratification of it according to different levels of exposure.