Spillover improved survival in non-invited patients of the colorectal cancer screening programme.

Objectives In colorectal cancer screening, randomized clinical trials have shown a 16% mean reduction in colorectal cancer mortality, but the Finnish randomized health services study showed no effect. We quantified spillover (the total indirect effect caused by the programme on the non-invited) and corrected the effectiveness estimate of the Finnish programme. Methods We retrieved from the Finnish Cancer Registry data on all non-invited colorectal cancer patients diagnosed in 1999-2013 in municipalities that adopted screening ( n = 18,948). Patients were stratified by three 5-year diagnostic periods and two calendar periods of programme adoption in the municipality of residence. Follow-up ended on 31 December 2013. We measured the spillover effect in patient survival, based on differences of adjusted estimates of the colorectal cancer-related hazard of death between pairs of consecutive diagnostic periods. Results The spillover effect was estimated as 9 percentage points (95% confidence interval: -1 to 19 percentage points). It was 13 percentage points in men (-1 to 26 percentage points) and 5 percentage points in women (-9 to 20 percentage points). The corrected effect estimate of implementing screening in Finland was 5 percentage points. Conclusions The corrected Finnish effectiveness estimate was consistent with estimates from randomized trials. Indirect effects (spillover) bias the invitee-control contrast. In this case, spillover was an inherent benefit of the Finnish programme.

Challenges in evaluation of screening for gastric cancer among men based on nonrandomized design.

BACKGROUND: Objective was to quantify biases in screening for gastric cancer when comparing attenders to nonattenders using serum pepsinogen I (SPGI) level as primary test. METHODS: In mid 1990s, all men aged 51-65 years from two Finnish cities were invited to SPGI screening. Mortality and premature mortality in attenders were compared to nonattenders. Efficacy of screening was studied by 15 years' follow-up of standardized mortality ratio (SMR) and potential years of life lost (PYLL) due to gastric cancer. Bias due to selective attendance was quantified using corrective coefficients based on total cancer incidence and mortality, and gastric cancer-specific incidence and mortality for total population and nonattenders. RESULTS: In 1994-1996, men aged 51-65 years (16,872) were invited to SPGI assay and 12,175 men (72%) attended. SPGI was 25 microg/l or less in 610 (5%) men, indicating severe atrophic gastritis (AG). Post-screening gastroscopy was performed to 435 men with low SPGI. Of these, 168 men were referred for treatment due to abnormal focal lesions. Attributable proportions in reductions of SMR and PYLL from gastric cancer due to screening were 59% and 67%. After correcting for selective participation, attributable proportions were reduced to 23% and 39%. CONCLUSIONS: Biomarker screening by low SPGI among middle-aged men followed by upper gastrointestinal endoscopy decreased long-term and premature mortality due to gastric cancer. However, in spite of methodological corrections done, the results do not justify any firm conclusions or recommend general screening programs. Randomized trials are warranted for this purpose.

Predicting the effectiveness of the Finnish population-based colorectal cancer screening programme.

Objective Because colorectal cancer (CRC) has a long natural history, estimating the effectiveness of CRC screening programmes requires long-term follow-up. As an alternative, we here demonstrate the use of a temporal multi-state natural history model to predict the effectiveness of CRC screening. Methods In the Finnish population-based biennial CRC screening programme using faecal occult blood tests (FOBT), which was conducted in a randomised health services study, we estimated the pre-clinical incidence, the mean sojourn time (MST), and the sensitivity of FOBT using a Markov model to analyse data from 2004 to 2007. These estimates were applied to predict, through simulation, the effects of five rounds of screening on the relative rate of reducing advanced CRC with 6 years of follow-up, and on the reduction in mortality with 10 years of follow-up, in a cohort of 500,000 subjects aged 60 to 69. Results For localised and non-localised CRC, respectively, the MST was 2.06 and 1.36 years and the sensitivity estimates were 65.12% and 73.70%. The predicted relative risk of non-localised CRC and death from CRC in the screened compared with the control population was 0.86 (95% CI: 0.79-0.98) and 0.91 (95% CI: 0.85-1.02), respectively. Conclusion Based on the preliminary results of the Finnish CRC screening programme, our model predicted a 9% reduction in CRC mortality and a 14% reduction in advanced CRC.

Design-corrected variation by centre in mortality reduction in the ERSPC randomised prostate cancer screening trial.

Objectives To calculate design-corrected estimates of the effect of screening on prostate cancer mortality by centre in the European Randomised Study of Screening for Prostate Cancer (ERSPC). Setting The ERSPC has shown a 21% reduction in prostate cancer mortality in men invited to screening with follow-up truncated at 13 years. Centres either used pre-consent randomisation (effectiveness design) or post-consent randomisation (efficacy design). Methods In six centres (three effectiveness design, three efficacy design) with follow-up until the end of 2010, or maximum 13 years, the effect of screening was estimated as both effectiveness (mortality reduction in the target population) and efficacy (reduction in those actually screened). Results The overall crude prostate cancer mortality risk ratio in the intervention arm vs control arm for the six centres was 0.79 ranging from a 14% increase to a 38% reduction. The risk ratio was 0.85 in centres with effectiveness design and 0.73 in those with efficacy design. After correcting for design, overall efficacy was 27%, 24% in pre-consent and 29% in post-consent centres, ranging between a 12% increase and a 52% reduction. Conclusion The estimated overall effect of screening in attenders (efficacy) was a 27% reduction in prostate cancer mortality at 13 years' follow-up. The variation in efficacy between centres was greater than the range in risk ratio without correction for design. The centre-specific variation in the mortality reduction could not be accounted for by the randomisation method.

Effect of characteristics of women on attendance in blind and non-blind randomised trials: analysis of recruitment data from the EPHT Trial.

OBJECTIVES: To analyse the effect of women's characteristics on their willingness to join a blind or a non-blind subtrial or to be excluded by physicians. DESIGN: Primary prevention trial of postmenopausal hormone therapy (HT). A 2×2, randomised design with a non-blind HT arm or control arm and a blind HT arm or placebo arm. SETTING: 3 clinical centres in Estonia. METHODS: Interest in joining the trial was asked in a questionnaire together with demographic and health status data. Interested and eligible women were invited to a health examination that also informed whether they belonged to a blind or to a non-blind subtrial; the arm was not revealed. Trial physicians made further exclusions when checking the women's eligibility. Thereafter, informed consent was asked as detailed in the flow chart. Comparisons were made between non-blind and blind subtrials. Analyses were carried out for each of the background variables. OUTCOME MEASURES: The proportion of willingness, eligibility and attendance. RESULTS: Women randomised to the non-blind subtrial were more willing to join (relative risk (RR) 1.17) and more likely to be found eligible by physicians (RR 1.10) than women in the blind subtrial, resulting in larger attendance (RR 1.29). Women with higher education were differentially more willing to join the non-blind trial (RR 1.29) than those with basic education (RR 1.08); the differential willingness of never-smokers (RR 1.20) was larger than that of current smokers (RR 1.07). The differential exclusion by physicians by education and smoking were small. Some subjective symptoms (eg, diarrhoea/constipation, stomach pain) had reverse differential effects on attendance in the non-blind subtrial in comparison to the blind subtrial. Menopausal symptoms did not affect the differential interest, eligibility or attendance. CONCLUSIONS: Blinding in RCT reduces attendance, due to decisions of the women and the trial physicians. Differential attendance by blinding may affect the generalisability of the results from trials. TRIAL REGISTRATION NUMBER: ISRCTN35338757.

Risk of gastric cancer in Helicobacter pylori infection in a 15-year follow-up.

OBJECTIVE: We investigated the risk of gastric cancer among men with Helicobacter pylori (H. pylori) infection or atrophic gastritis (AG) in a 15-year follow-up. MATERIALS AND METHODS: Study population consists of 12,016 men aged 50-65 years at the beginning of the follow-up in 1994-1996. Serum levels of pepsinogen I (SPGI) and antibodies (IgG) to H. pylori (HpAb) were assayed from serums collected in 1994-1996. Incidence of gastric cancer in the study population was assessed in follow-up from 1994 to 2011 by data from the nationwide cancer registry. Based on SPGI and HpAb values, standardized incidence ratios (SIRs) of gastric cancer were calculated in three subgroups, that is, in those with a healthy stomach, those with H. pylori infection but without AG and those with AG. Risk ratios (RR) of gastric cancer were calculated using SIR of subgroups. RESULTS: During 15 years, seven gastric cancers appeared per 79,928 person years among men with healthy stomachs, 50 cancers per 92,533 person years in men with H. pylori infection but without AG, and 8 per 8658 person years in men with AG. Risk ratio (RR) of stomach cancer in men with H. pylori infection was 5.8 (95%CI: 2.7-15.3) compared to men with healthy stomachs, and 9.1 (95%CI: 2.9-30.0) in men with AG. There were no differences in cancer risk between cardia and distal stomach. CONCLUSIONS: Risk of gastric cancer is low in men with healthy stomachs. It is significantly increased in those with H. pylori infection and more in those with AG.

Longitudinal biobanks-based study on the joint effects of infections, nutrition and hormones on risk of prostate cancer.

Background To evaluate the individual and combined effects of enterolactone, vitamin D, free testosterone, Chlamydia trachomatis and HPV-18 on the risk of prostate cancer in a large population-based biochemical material that combined three Nordic serum sample banks. Material and methods A joint cohort of 209 000 healthy men was followed using cancer registry linkages. From this cohort altogether 699 incident cases of prostate cancer were identified. Four controls were selected by incidence density sampling and matching for country, age and date of the blood sampling. Complete data for all investigated exposures was available for 483 eligible cases and 1055 eligible controls. Multivariate regression analyses were performed to investigate the solitary and combined effects. Results The solitary effects were small. Significantly increased risk [rate ratio 1.6 (95% CI 1.0-2.5)] was found in those seronegative for C. trachomatis infection. The joint effect in risk levels of enterolactone and vitamin D was antagonistic [observed rate ratio (RR) 1.4 (1.0-2.1), expected RR 2.0 (1.0-4.1)] as well as that of HPV-18 and C. trachomatis [observed RR 1.9 (0.8-4.5), expected RR 9.9 (1.1-87.0)]. Conclusion A large follow-up study combining data from several previously investigated exposures to investigate joint effects found no evidence that exposure to two risk factors would increase the risk of prostate cancer from that expected on basis of exposure to one risk factor. If anything, the results were consistent with antagonistic interactions.

Effect of Screening on Variation in Cervical Cancer Survival by Socioeconomic Determinants--a Study from Rural South India.

BACKGROUND: Socioeconomic factors are associated with screening in terms of reducing the risk of cervical cancer. This study aimed to clearly establish the effect of screening on variation in socio-economic factor-specific survival estimates. MATERIALS AND METHODS: Survival estimates were calculated using the life table method for 165 women from the routine care control arm and 67 from the visual inspection with acetic acid screening arm diagnosed with cervical cancer during 2000-2006 in rural south India. Kaplan-Meier survival curves were plotted to compare the variation in survival by socioeconomic factors. RESULTS: Whereas there was a significant variation in survival estimates of the different categories of age at diagnosis among the screen-detected cancers with women aged<50 years having an improved survival, no significant variation was noted among women diagnosed with cervical cancer from the control arm. Compared to the variation among the cancer cases detected in the unscreened control group, screening widened the variation in survival estimates by age and type of house, and reduced the variation by education. The direction of the magnitude of the survival estimates was reversed within the different categories of occupation, marital status and household income in the screen-detected cancer cases compared to control group cancer cases. Also, women diagnosed with stage 1 disease had a very good survival. CONCLUSIONS: Screening changed the pattern of survival by socio-economic factors. We found improved survival rates in screened women aged <50 years, with no formal education, manual workers and married women.

Risk of invasive cancer among women visually screened and colposcopy triaged by trained nurses in rural South India.

OBJECTIVE: To estimate the long-term risk of cervical cancer among women screened by visual inspection with acetic acid (VIA) and to evaluate the benefit of additional colposcopy triage in rural south India. METHODS: A retrospective analysis was conducted among 31 343 women who had undergone VIA at Dindigul district, India between January 1, 2000, and August 5, 2003, as part of a randomized screening trial. Women with positive VIA test results were offered colposcopy triage by trained nurses. Cervical cancer incidence data during follow-up (January 1, 2000, to December 31, 2012) were obtained from a regional cancer registry. RESULTS: Among 3021 screen-positive women free of cancer at baseline, 2974 women underwent colposcopy; colposcopic abnormalities suggestive of precancerous lesions were detected among 2792 of these women (93.9%). Compared with the women with negative VIA screening results, the hazard ratio (HR) of cervical cancer during follow-up among the VIA-positive women without colposcopic abnormalities was 6.5 (95% confidence interval [CI], 1.6-27.1). The risk was similar among VIA-positive women with colposcopic abnormalities but without histological confirmation (HR5.2; 95% CI, 1.9-14.6). CONCLUSION: The high risk of cancer among women without colposcopic abnormalities who tested positive by VIA suggested that screening without triage is potentially effective.

The Finnish prostate cancer screening trial: analyses on the screening failures.

Prostate cancer (PC) screening with prostate-specific antigen (PSA) has been shown to decrease PC mortality in the European Randomized Study of Screening for Prostate Cancer (ERSPC). However, in the Finnish trial, which is the largest component of the ERSPC, no statistically significant mortality reduction was observed. We investigated which had the largest impact on PC deaths in the screening arm: non-participation, interval cancers or PSA threshold. The screening (SA) and control (CA) arms comprised altogether 80,144 men. Men in the SA were screened at four-year intervals and referred to biopsy if the PSA concentration was ≥ 4.0 ng/ml, or 3.0-3.99 ng/ml with a free/total PSA ratio ≤ 16%. The median follow-up was 15.0 years. A counterfactual exclusion method was applied to estimate the effect of three subgroups in the SA: the non-participants, the screen-negative men with PSA ≥ 3.0 ng/ml and a subsequent PC diagnosis, and the men with interval PCs. The absolute risk of PC death was 0.76% in the SA and 0.85% in the CA; the observed hazard ratio (HR) was 0.89 (95% confidence interval (CI) 0.76-1.04). After correcting for non-attendance, the HR was 0.78 (0.64-0.96); predicted effect for a hypothetical PSA threshold of 3.0 ng/ml the HR was 0.88 (0.74-1.04) and after eliminating the effect of interval cancers the HR was 0.88 (0.74-1.04). Non-participating men in the SA had a high risk of PC death and a large impact on PC mortality. A hypothetical lower PSA threshold and elimination of interval cancers would have had a less pronounced effect on the screening impact.

Use of hospital resources in the Finnish colorectal cancer screening programme: a randomised health services study.

OBJECTIVE: To estimate the difference in use of hospital resources in the Finnish Colorectal Cancer (CRC) screening programme between those invited and controls, within the year of randomisation and the next year. DESIGN: CRC screening was implemented in Finland in 2004 as a population-based randomised design using biennial faecal occult blood test (FOBT) for men and women aged 60-69 years. Those randomised to screening and control groups during years 2004-2009 were included in this analysis and use of hospital resources was estimated. Data were collected from the national register on hospital discharges. Outpatient visits, inpatient episodes and colonoscopies were compared between the two groups. RESULTS: The screening group comprised of 123 149 and control group of 122 930 people. Most people in both groups had not used hospital resources at all. More people in the screening group than in the control group had at least one hospital-based outpatient visit (7.8% vs 7.4%), inpatient episode (3.9% vs 3.8%) and colonoscopy (1.5% vs 1.3%). In total, the screening group had 31 975 and control group 27 061 cumulative outpatient visits, 9260 and 7903 inpatient episodes, and 2686 and 1756 hospital colonoscopies, respectively. The proportion of those with a positive FOBT result with at least one outpatient visit, one inpatient episode or one colonoscopy, was 3.7 times, 2.5 times or 9 times that of those with a negative FOBT result, respectively. CONCLUSIONS: CRC screening using the FOBT slightly increased the volume of hospital outpatient visits, inpatient episodes and hospital colonoscopies in Finland.

Sensitivity, effect and overdiagnosis in screening for cancers with detectable pre-invasive phase.

Studies on cancer screening often evaluate the performance by indirect indicators. In case the screening detects pre-invasive lesions, they may be a mixture of benefit of sensitivity and effect as well as of harm of overdiagnosis. Here, we develop the formulae for the sensitivity, the effect and overdiagnosis in screening for pre-invasive lesions of cancer. Sensitivity is the ability of screening to identify a progressive lesion at the level of test (relevant for the laboratory), episode (relevant in the clinic) and programme (relevant at the population level). Effect is reduction of cancer incidence in those screened (efficacy) and in the target population (effectiveness). The sensitivity is estimated by interval cancers between two consecutive screens (incidence method) and the effect by interval cancers and cancers detected at the subsequent screen. Overdiagnosis is estimated as the detection rate of pre-invasive lesions minus the rate of invasive cancer prevented by screening in one screening round. All the indicators are corrected for nonattendance and selective attendance by disease risk. The population to be followed and the period of follow-up are defined for each indicator separately. Data on cervix cancer screening with Papnet® automation device are given as an example. Estimation of sensitivity and effect are consistent with the purpose of the screening to prevent invasive disease. We further define the purpose at the level of laboratory, clinical medicine and public health and derive six estimators corresponding to the specific purposes considered in our article.

Screening and prostate cancer mortality: results of the European Randomised Study of Screening for Prostate Cancer (ERSPC) at 13 years of follow-up.

BACKGROUND: The European Randomised study of Screening for Prostate Cancer (ERSPC) has shown significant reductions in prostate cancer mortality after 9 years and 11 years of follow-up, but screening is controversial because of adverse events such as overdiagnosis. We provide updated results of mortality from prostate cancer with follow-up to 2010, with analyses truncated at 9, 11, and 13 years. METHODS: ERSPC is a multicentre, randomised trial with a predefined centralised database, analysis plan, and core age group (55-69 years), which assesses prostate-specific antigen (PSA) testing in eight European countries. Eligible men aged 50-74 years were identified from population registries and randomly assigned by computer generated random numbers to screening or no intervention (control). Investigators were masked to group allocation. The primary outcome was prostate cancer mortality in the core age group. Analysis was by intention to treat. We did a secondary analysis that corrected for selection bias due to non-participation. Only incidence and no mortality data at 9 years' follow-up are reported for the French centres. This study is registered with Current Controlled Trials, number ISRCTN49127736. FINDINGS: With data truncated at 13 years of follow-up, 7408 prostate cancer cases were diagnosed in the intervention group and 6107 cases in the control group. The rate ratio of prostate cancer incidence between the intervention and control groups was 1·91 (95% CI 1·83-1·99) after 9 years (1·64 [1·58-1·69] including France), 1·66 (1·60-1·73) after 11 years, and 1·57 (1·51-1·62) after 13 years. The rate ratio of prostate cancer mortality was 0·85 (0·70-1·03) after 9 years, 0·78 (0·66-0·91) after 11 years, and 0·79 (0·69-0·91) at 13 years. The absolute risk reduction of death from prostate cancer at 13 years was 0·11 per 1000 person-years or 1·28 per 1000 men randomised, which is equivalent to one prostate cancer death averted per 781 (95% CI 490-1929) men invited for screening or one per 27 (17-66) additional prostate cancer detected. After adjustment for non-participation, the rate ratio of prostate cancer mortality in men screened was 0·73 (95% CI 0·61-0·88). INTERPRETATION: In this update the ERSPC confirms a substantial reduction in prostate cancer mortality attributable to testing of PSA, with a substantially increased absolute effect at 13 years compared with findings after 9 and 11 years. Despite our findings, further quantification of harms and their reduction are still considered a prerequisite for the introduction of populated-based screening. FUNDING: Each centre had its own funding responsibility.

Effect of a pre-screening survey on attendance in colorectal cancer screening: a double-randomized study in Finland.

OBJECTIVES: To explore effects of a pre-screening life style survey on the subsequent attendance proportion in colorectal cancer screening. SETTING: Finnish colorectal cancer screening programme in 2011. DESIGN: Double randomized and controlled follow-up design. METHODS: The study population comprised of 31,951 individuals born in 1951. In 2010 to a random sample of every sixth (n = 5,312) person we sent a 7-paged life style questionnaire, and to another random sample of every sixth person (n = 5,336) a 10-paged life style and quality of life questionnaire. One year later, in 2011, 31,484 individuals of the original cohort were independently randomized (1:1) for colorectal cancer screening (n = 15,748) or control group (n = 15,736). Of those who were invited for screening, 5185 had received a questionnaire during the previous year. RESULTS: 5870 individuals (55.1 %) responded to the questionnaire in 2010. The overall attendance at screening in 2011 was 59.0 % in those born in 1951 (i.e. the 60-year-olds). In those who had been sent the survey the attendance in screening was 56.6% (57.3% for the short and 56.0% for the long questionnaire) and in those who had not received the questionnaire it was 60.2% (P < 0.001). CONCLUSIONS: We believe that the observed reduction in attendance in those who had been sent a questionnaire earlier is generally true. Thus, if any survey is enclosed in the screening invitation, this finding should be taken into account when planning the programme. Any extra effort requested may reduce the attendance proportion for screening, reducing the population level impact of screening.

Effect of intervention on decision making of treatment for disease progression, prostate-specific antigen biochemical failure and prostate cancer death.

BACKGROUND: Patient preference for the choice of treatment modality for prostate cancer has increasingly gained attention. OBJECTIVE: To assess the impact of client-oriented decision on long-term mortality, disease progression and biochemical failure compared with standard treatment protocol (TP). METHODS: With data from a Finnish multicentre, randomized controlled trial with two arms [104 in the enhanced patient participation (EPP) arm and 106 in the TP arm], disease-specific and disease-free survival, biochemical failure with elevated prostate-specific antigen (PSA) level and disease progression were compared between the two arms using Wilcoxon test and also Cox proportional hazards regression model. RESULTS: Patients in the EPP arm had a higher risk of death by 37% [HR, 1.37 (0.87-2.17)] compared with those in the TP arm. Patients in the EPP arm were at increased risk of having biochemical failure by 14% [HR, 1.14 (0.72-1.79)] and for having disease progression by 2% [HR, 1.02 (0.61-1.70)] compared with those in the TP arm. All the differences were non-significant. CONCLUSIONS: Patients actively involved in the choice of treatment had higher risk of prostate cancer death but only slightly increased risk of biochemical failure and clinical disease progression. These findings would provide a good reference when patient autonomy for the choice of treatment modality is addressed.

Prostate cancer mortality in the Finnish randomized screening trial.

BACKGROUND: Prostate cancer (PC) screening with prostate-specific antigen (PSA) has been shown to decrease PC mortality by the European Randomized Study of Screening for Prostate Cancer (ERSPC). We evaluated mortality results in the Finnish Prostate Cancer Screening Trial, the largest component of ERSPC. The primary endpoint was PC-specific mortality. METHODS: A total of 80 144 men were identified from the population registry and randomized to either a screening arm (SA) or a control arm (CA). Men in the SA were invited to serum PSA determination up to three times with a 4-year interval between each scan and referred to biopsy if the PSA concentration was greater than or equal to 4.0 ng/mL or 3.0 to 3.99 ng/mL with a free/total PSA ratio less than or equal to 16%. Men in the CA received usual care. The analysis covers follow-up to 12 years from randomization for all men. Hazard ratios (HRs) were estimated for incidence and mortality using Cox proportional hazard model. All statistical tests were two-sided. RESULTS: PC incidence was 8.8 per 1000 person-years in the SA and 6.6 in the CA (HR = 1.34, 95% confidence interval [CI] = 1.27 to 1.40). The incidence of advanced PC was lower in the SA vs CA arm (1.2 vs 1.6, respectively; HR = 0.73, 95% CI = 0.64 to 0.82; P < .001). For PC mortality, no statistically significant difference was observed between the SA and CA (HR = 0.85, 95% CI = 0.69 to 1.04) (with intention-to-screen analysis). To avoid one PC death, we needed to invite 1199 men to screening and to detect 25 PCs. We observed no difference in all-cause mortality between trial arms. CONCLUSIONS: At 12 years, a relatively conservative screening protocol produced a small, non-statistically significant PC-specific mortality reduction in the Finnish trial, at the cost of moderate overdiagnosis.

Effect of screening on the risk estimates of socio demographic factors on cervical cancer - a large cohort study from rural India.

BACKGROUND: Prospective cohort studies to determine cofactors with oncogenic HPV- infections for cervical cancer are very rare from developing countries and such data are limited to the few screening trials. Large screening trials provide such data as a by product. Some of the cases are prevented by screening and do not surface as invasive cancers at all. Also, pre-invasive lesions are detected almost entirely by screening. Screening causes selection bias if attendance in or effectiveness of screening is correlated with the risk factors. The aim of this study was to quantify the influence of screening on risk factors for cervical cancer. MATERIALS AND METHODS: Our material stems from a rural cohort of 80,000 women subjected to a randomised screening trial. The effect of screening on the incidence of cervix cancer was estimated with reference to socio-demographic and reproductive risk factors of cervical cancer. We compared these risks with the incidence of cancer in the randomised control population by the same determinants of risk. RESULTS: The results in the screening arm compared to the control arm showed that the women of low SES and young age were benefitting more than those of high SES and old age. The relative risk by age (30-39 vs 50-59) was 0.33 in the control arm and 0.24 in the screening arm. The relative risk by education (not educated vs educated) was 2.8 in the control arm and 1.8 in the screening arm. The previously married women did not benefit (incidence 113 and 115 per 100,000 women years in control vs screening arms) whereas the effect was substantial in those married (86 vs 54). CONCLUSIONS: The results in controls were consistent with the general evidence, but results in attenders and nonattenders of the screening arm showed that screening itself and self-selection in attendance and effectiveness can influence the effect estimates of risk factors. The effect of cervical cancer screening programmes on the estimates of incidence of cervical cancer causes bias in the studies on etiology and, therefore, they should be interpreted with caution.