Evaluating the effectiveness of cervical cancer screening invitation letters.

OBJECTIVE: The objective of this study was to evaluate the effectiveness of an invitation letter on cervical screening participation among unscreened women 30 to 69 years of age. METHODS: A cluster randomized trial design was used in which unscreened women (n=31,452) were randomized by the forward sortation area (FSA) of their postal code to an intervention group that was sent an invitation letter (n=17,068) or a group that was not sent an invitation letter (n=14,384). RESULTS: Six months after the letters were mailed, 1,010 women in the intervention group (5.92%) and 441 women in the control group (3.06%) had a Pap test. After adjusting for variables that have previously shown to influence screening participation, women who were sent an invitation letter were significantly more likely to have had a Pap test in the next 6 months compared with women who were not sent an invitation letter (odds ratio [OR]=2.60, 95% confidence interval [CI] 2.09-3.35, p<0.001). Overall, the effectiveness of the invitation letter improved with increasing age (p=0.02). CONCLUSIONS: Sending invitation letters increased cervical screening participation but because the overall effect was small, additional strategies that remove barriers to screening for unscreened women are also necessary.

Concordance with ASCO guidelines for surveillance after colorectal cancer treatment: a population-based analysis.

PURPOSE: Intensive surveillance after curative treatment of colorectal cancer (CRC) is associated with improved overall survival. This study examined concordance with the 2005 ASCO surveillance guidelines at the population level. METHODS: A cohort of 250 patients diagnosed with stage II or III CRC in 2004 and alive 42 months after diagnosis was identified from health administrative data in Manitoba, Canada. Colonoscopy, liver imaging, and carcinoembryonic antigen (CEA) testing were assessed over 3 years. Guidelines were met if patients had at least one colonoscopy in 3 years and at least one liver imaging test and three CEA tests annually. Multivariate logistic regression assessed the effect of patient and physician characteristics and disease and treatment factors on guideline concordance. RESULTS: Guidelines for colonoscopy, liver imaging, and CEA were met by 80.4%, 47.2%, and 22% of patients, respectively. Guideline concordance for colonoscopy was predicted by annual contact with a surgeon, higher income, and the diagnosis of colon (rather than rectal) cancer. Adherence was lower in those older than 70 years and with higher comorbidity. For liver imaging, significant predictors were annual contact with an oncologist, receipt of chemotherapy, and diagnosis of colon cancer. Concordance with CEA guidelines was higher with annual contact with an oncologist and high levels of family physician contact, and lower in urban residents, in those older than 70, and in those with stage II disease. CONCLUSION: Completion of recommended liver imaging and CEA testing fall well below guidelines in Manitoba, whereas colonoscopy is better provided. Addressing this gap should improve outcomes for CRC survivors.

Papanicolaou test utilization and frequency of screening opportunities among women diagnosed with cervical cancer.

BACKGROUND: Although the importance of Papanicolaou (Pap) smear test screening in reducing the incidence of cervical cancer is well established, in 1994-95 one in 4 women in Manitoba aged 18 to 69 years reported never having had a Pap test or not having had a Pap test in the last 3 years. The objectives of this study were to examine the screening history of women in Manitoba diagnosed with invasive cervical cancer and to explore whether opportunities for screening were missed. METHODS: In this case-control study women aged 18 years and older who resided in Manitoba and were diagnosed with invasive cervical cancer between 1989 and 2001 were each matched by age and area of residence to 5 controls, (N = 4009). Conditional logistic regression analyses were used to examine the association between Pap test utilization and the likelihood of diagnosis with invasive cervical cancer. Generalized linear models using the negative binomial distribution were used to assess the association between cancer status and rates of prior Pap testing and of opportunities to be screened. Logistic generalized estimating equation models were used for the analysis of physician characteristics. RESULTS: Forty-six percent of women in Manitoba diagnosed with invasive cervical cancer and 67% of the control group had received a Pap test in the 5 years before the case's diagnosis. After adjustment for age, income and residence, the rate of Pap testing was significantly higher in the control group (rate ratio [RR] = 1.57, 95% confidence interval [CI] 1.44-1.73). Conversely, when cervical cancer was the outcome, women who had not had Pap tests were more likely to be diagnosed with invasive cervical cancer (odds ratio [OR] = 2.77, 95% CI 2.30-3.30) than women who did have a Pap test. Although women diagnosed with invasive cervical cancer had fewer Pap tests, they had had as many opportunities to be screened as controls (RR = 1.04, 95% CI 0.96-1.12). Compared with urban family physicians, rural family physicians were less likely to provide Pap tests (OR = 0.68, 95% CI 0.58-0.80) and specialists were more likely to provide Pap tests (OR = 1.70, 95% CI 1.30-2.22). CONCLUSIONS: Women who were diagnosed with invasive cervical cancer in the province of Manitoba, Canada, had fewer Pap tests but the same frequency of opportunities to be screened as matched controls. These results reinforce the need to educate women about cervical cancer screening and the importance of receiving Pap tests.

Influence of direct referrals on time to diagnosis after an abnormal breast screening result.

This study examined the influence of a direct referral process implemented by a population-based provincial breast screening program on the time from screening to first procedure, first procedure to diagnosis, and screening to diagnosis following an abnormal breast screening result. The direct referral process shifted the responsibility for the coordination of diagnostic follow-up procedures from the family physician to the screening program. Three cohorts of women were included: a control cohort (screened prior to the initiation of a direct referral process, n = 1347), a usual care cohort (screened after the initiation of a direct referral process but for whom permission to refer was denied by the family physician, n = 1225), and a direct referral cohort (screened after the initiation of the direct referral process and for whom permission to refer was given by the family physician, n = 1232). The direct referral cohort was subdivided into women referred to a breast health centre (BHC group) (n = 606) and women referred to a diagnostic facility (diagnostic group) (n = 626). The direct referral cohort completed all three time intervals significantly faster than the other two cohorts (P < 0.0001). The diagnostic group experienced a significantly lower time from screening to the first procedure than the other cohorts or the BHC group (P < 0.0001). However, the BHC group had a significantly lower time from first procedure to diagnosis than the other cohorts or the diagnostic group (P < 0.0001). The control and the usual care cohorts were not significantly different from each other (P = 0.6250). The direct referral process significantly reduced the time to diagnosis after an abnormal screening result.