Breast cancer diagnosis and treatment wait times in specialized diagnostic units compared with usual care: a population-based study

Background

Breast assessment sites (bass) were developed to provide expedited and coordinated care for patients being evaluated for breast cancer (bca) in Ontario. We compared the diagnostic and treatment intervals for patients diagnosed at a bas and for those diagnosed through a usual care (uc) route.

Methods

This population-based, cross-sectional study of patients diagnosed with bca in Ontario during 2007-2015 used linked administrative data. "Diagnostic interval" was the time from the earliest cancer-related health care encounter before diagnosis to diagnosis; "treatment interval" was the time from diagnosis to treatment. Diagnosis at a bas was determined from the patient's biopsy and mammography institutions. Interval lengths for the bas and uc groups were compared using multivariable quantile regression, stratified by detection method.

Results

The diagnostic interval was shorter for patients who were bas-diagnosed than for those who were uc-diagnosed, with adjusted median differences of -4.0 days [95% confidence interval (ci): -3.2 days to -4.9 days] for symptomatic patients and -5.4 days (95% ci: -4.7 days to -6.1 days) for screen-detected patients. That association was modified by stage at diagnosis, with larger differences in patients with early-stage cancers. In contrast, the treatment interval was longer in patients who were bas-diagnosed than in those who were uc-diagnosed, with adjusted median differences of 4.2 days (95% ci: 3.8 days to 4.7 days) for symptomatic patients and 4.2 days (95% ci: 3.7 days to 4.8 days) for screen-detected patients.

Conclusions

Diagnosis of bca through a bas was associated with a shorter diagnostic interval, but a longer treatment interval. Although efficiencies in the diagnostic interval might help to reduce distress experienced by patients, the longer treatment intervals for patients who are bas-diagnosed remain a cause for concern.

Disparities in breast cancer diagnosis for immigrant women in Ontario and BC: results from the CanIMPACT study

BACKGROUND

In Canada, clinical practice guidelines recommend breast cancer screening, but there are gaps in adherence to recommendations for screening, particularly among certain hard-to-reach populations, that may differ by province. We compared stage of diagnosis, proportion of screen-detected breast cancers, and length of diagnostic interval for immigrant women versus long-term residents of BC and Ontario.

METHODS

We conducted a retrospective cohort study using linked administrative databases in BC and Ontario. We identified all women residing in either province who were diagnosed with incident invasive breast cancer between 2007 and 2011, and determined who was foreign-born using the Immigration Refugee and Citizenship Canada database. We used descriptive statistics and bivariate analyses to describe the sample and study outcomes. We conducted multivariate analyses (modified Poisson regression and quantile regression) to control for potential confounders.

RESULTS

There were 14,198 BC women and 46,952 Ontario women included in the study population, of which 11.8 and 11.7% were foreign-born respectively. In both provinces, immigrants and long-term residents had similar primary care access. In both provinces, immigrant women were significantly less likely to have a screen-detected breast cancer (adjusted relative risk 0.88 [0.79-0.96] in BC, 0.88 [0.84-0.93] in Ontario) and had a significantly longer median diagnostic interval (2 [0.2-3.8] days in BC, 5.5 [4.4-6.6] days in Ontario) than long-term residents. Women from East Asia and the Pacific were less likely to have a screen-detected cancer and had a longer diagnostic interval, but were diagnosed at an earlier stage than long-term residents. In Ontario, women from Latin America and the Caribbean and from South Asia were less likely to have a screen-detected cancer, had a longer median diagnostic interval, and were diagnosed at a later stage than long-term residents. These findings were not explained by access to primary care.

CONCLUSIONS

There are inequalities in breast cancer diagnosis for Canadian immigrant women. We have identified particular immigrant groups (women from Latin America and the Caribbean and from South Asia) that appear to be subject to disparities in the diagnostic process that need to be addressed in order to effectively reduce gaps in care.

Measuring the Quality of Personal Care in Patients Undergoing Radiotherapy for Prostate Cancer

AIMS

To describe the quality of the non-technical component of the care (personal care) of patients receiving radical radiotherapy for prostate cancer and to identify elements of personal care that should be priorities for quality improvement.

MATERIALS AND METHODS

One hundred and eight patients undergoing radiotherapy for localised prostate cancer completed a self-administered questionnaire that asked them to rate the importance of 143 non-technical elements of care and to rate the quality of their own care with respect to each element. The elements that a patient rated as both 'very important' and less than 'very good' were deemed to be his priorities for improvement. The priorities of the population were established by ranking the elements based on the percentage of patients who identified them as a priority (importance/quality analysis).

RESULTS

The response rate was 65%. The percentage of elements rated 'very good' varied from patient to patient: median 79% (interquartile range 69-92%). The percentage of elements rated either 'very good' or 'good' was higher: median 96% (interquartile range 86-98%). Nonetheless, almost every patient rated at least some elements of his care as less than optimal, regardless of the cut-off point used to define optimal quality. Patients assigned their lowest quality ratings to elements relating to the quality of the treatment environment and comprehensiveness of additional services available to them. However, patients rated most of these elements as relatively unimportant, and importance/quality analysis identified elements of care relating to communication of information about the disease and its treatment as the highest priorities for quality improvement.

CONCLUSIONS

Most patients rated most elements of their personal care as very good, but almost all were able to identify some elements that were less than optimal. When ratings of quality were integrated with ratings of importance, elements relating to communication emerged as the patients' highest priorities for quality improvement.

Symptom appraisal, help seeking, and lay consultancy for symptoms of head and neck cancer

OBJECTIVE

Early diagnosis is important in head and neck cancer (HNC) patients to maximize the effectiveness of the treatments and minimize the debilitation associated with both the cancer and the invasive treatments of advanced disease. Many patients present with advanced disease, and there is little understanding as to why. This study investigated patients' symptom appraisal, help seeking, and lay consultancy up to the time they first went to see a health care professional (HCP).

METHODS

We interviewed 83 patients diagnosed with HNC. The study design was cross sectional and consisted of structured telephone interviews and a medical chart review. We gathered information on the participant's personal reactions to their symptoms, characteristics of their social network, and the feedback they received.

RESULTS

We found that 18% of the participants thought that their symptoms were urgent enough to warrant further investigation. Participants rarely (6%) attributed their symptoms to cancer. Eighty-nine percent reported that they were unaware of the early warning signs and symptoms of HNC. Fifty-seven percent of the participants disclosed their symptoms to at least one lay consultant before seeking help from an HCP. The lay consultants were usually their spouse (77%), and the most common advice they offered was to see a doctor (76%). Lastly, 81% of the participants report that their spouse influenced their decision to see an HCP.

CONCLUSIONS

The results of this study suggest that patients frequently believe that their symptoms were nonurgent and that their lay consultants influence their decision to seek help from an HCP.

Use of physician services during the survivorship phase: a multi-province study of women diagnosed with breast cancer

INTRODUCTION

Oncologists have traditionally been responsible for providing routine follow-up care for cancer survivors; in recent years, however, primary care providers (pcps) are taking a greater role in care during the follow-up period. In the present study, we used a longitudinal multi-province retrospective cohort study to examine how primary care and specialist care intersect in the delivery of breast cancer follow-up care.

METHODS

Various databases (registry, clinical, and administrative) were linked in each of four provinces: British Columbia, Manitoba, Ontario, and Nova Scotia. Population-based cohorts of breast cancer survivors were identified in each province. Physician visits were identified using billings or claims data and were classified as visits to primary care (total, breast cancer-specific, and other), oncology (medical oncology, radiation oncology, and surgery), and other specialties. The mean numbers of visits by physician type and specialty, or by combinations thereof, were examined. The mean numbers of visits for each follow-up year were also examined by physician type.

RESULTS

The results showed that many women (>64%) in each province received care from both primary care and oncology providers during the follow-up period. The mean number of breast cancer-specific visits to primary care and visits to oncology declined with each follow-up year. Interprovincial variations were observed, with greater surgeon follow-up in Nova Scotia and greater primary care follow-up in British Columbia. Provincial differences could reflect variations in policies and recommendations, relevant initiatives, and resources or infrastructure to support pcp-led follow-up care.

CONCLUSIONS

Optimizing the role of pcps in breast cancer follow-up care might require strategies to change attitudes about pcp-led follow-up and to better support pcps in providing survivorship care.

Multigene expression profile testing in breast cancer: is there a role for family physicians?

BACKGROUND

Family physicians (fps) play a role in aspects of personalized medicine in cancer, including assessment of increased risk because of family history. Little is known about the potential role of fps in supporting cancer patients who undergo tumour gene expression profile (gep) testing.

METHODS

We conducted a mixed-methods study with qualitative and quantitative components. Qualitative data from focus groups and interviews with fps and cancer specialists about the role of fps in breast cancer gep testing were obtained during studies conducted within the pan-Canadian canimpact research program. We determined the number of visits by breast cancer patients to a fp between the first medical oncology visit and the start of chemotherapy, a period when patients might be considering results of gep testing.

RESULTS

The fps and cancer specialists felt that ordering gep tests and explaining the results was the role of the oncologist. A new fp role was identified relating to the fp-patient relationship: supporting patients in making adjuvant therapy decisions informed by gep tests by considering the patient's comorbid conditions, social situation, and preferences. Lack of fp knowledge and resources, and challenges in fp-oncologist communication were seen as significant barriers to that role. Between 28% and 38% of patients visited a fp between the first oncology visit and the start of chemotherapy.

CONCLUSIONS

Our findings suggest an emerging role for fps in supporting patients who are making adjuvant treatment decisions after receiving the results of gep testing. For success in this new role, education and point-of-care tools, together with more effective communication strategies between fps and oncologists, are needed.

Adherence to and uptake of clinical practice guidelines: lessons learned from a clinical practice guideline on chemotherapy concomitant with radiotherapy in head-and-neck cancer

BACKGROUND

Clinical practice guidelines (cpgs) are systematically developed statements designed to assist practitioners and patients in making decisions about appropriate heath care interventions. Clinical practice guidelines are expensive and time-consuming to create. A cpg on concurrent chemotherapy with radiation therapy (ccrt) was developed in Ontario at a time when treatment approaches for head-and-neck cancer were changing significantly.

METHODS

An assessment of treatments and outcomes based on electronic and chart data obtained from a population-based study of 571 patients with oropharynx cancer treated in Ontario (2003-2004) was combined with a review of relevant knowledge transfer (publications and presentations at major meetings) to understand variation in adherence to a cpg.

RESULTS

In 9 Ontario cancer treatment centres, ccrt was used for 55% of all patients with oropharyngeal cancer; however, at the centres individually, that proportion ranged from 82% to 39%. Furthermore, there was no agreement on the chemotherapy regimen: 2-4 years later (a period during which newer regimens were emerging), only 4 of 9 centres were following the guideline for most patients. When outcomes of treated patients were compared for centres with "higher" and "lower" use of ccrt, no difference in survival was observed (p = 0.64).

CONCLUSIONS

At a time of treatment evolution, the new guideline was controversial, and there are many reasons for the mixed adherence. An estimation of adherence should be included during both development and review of guidelines.

Explaining socioeconomic status effects in laryngeal cancer

INTRODUCTION

People with lower socioeconomic status (SES) experience shorter survival times after a cancer diagnosis for many disease sites. We determined whether area-level SES was associated with the outcomes: cause-specific survival and local-regional failure in laryngeal cancer in Ontario, Canada. When we found an association we sought explanations that might be related to access to care including age, sex, rural residence, tumor stage, lymph node status, use of diagnostic imaging, treatment type, percentage of prescribed radiotherapy delivered, number of radiotherapy interruption days, treatment waiting time, and treating cancer center.

MATERIALS AND METHODS

The study population consisted of 661 glottic and 495 supraglottic stage-stratified randomly-sampled patients identified using the Ontario Cancer Registry. Area-level SES quintiles were assigned using adjusted median household income from the Canadian Census. Other data were collected from patient charts. Explanations for SES effects were determined by measuring whether the effect moved toward the null value by at least 10% when an access indicator was added to a the model.

RESULTS

Socioeconomic status was not related to either outcome for those with supraglottic cancer, but an association was present in glottic cancer. With the highest socioeconomic status quintile as the referent group, the relative risks for patients in the lowest socioeconomic quintile were 2.75 (95% CI 1.48, 5.12) for cause-specific survival and 1.90 (95% CI 1.24, 2.93) for local-regional failure. Disease stage as measured by T-category explained between 3% and 23% of these socioeconomic effects. None of the other access indicators met our 10% change criterion.

CONCLUSION

We question why people in lower socioeconomic quintiles were not diagnosed earlier in the disease progression. Having ruled out several variables that may be related to access to care, additional biologic and social variables should be examined to further understand socioeconomic status effects.

Radiotherapy patterns of practice: T1N0 glottic cancer in Ontario, Canada

AIMS

To describe the variation in the delivery of radiation therapy to patients with T1N0 glottic cancer who were diagnosed in Ontario, Canada, between 1982 and 1995.

MATERIALS AND METHODS

The patient population consisted of a random sample of 461 patients treated with curative intent from the nine cancer centres that administer radiation therapy in the province. Abstracted variables included prescribed dose (Gy) and fractionation (f), beam energy and arrangement, set-up, field size, beam modifiers, positioning and treatment interruptions.

RESULTS

Thirteen prescribed dose-fractionation schemes (> or = four cases each) were identified, including 50.0-53.0 Gy/20 f (54.5%), 55.0-61.0 Gy/25 f (30.3%), and 60.0-66.0 Gy/30-33 f (7.7%). All regimens used one fraction per day, 5 days per week. An isocentric set-up was used (94.3%), with megavoltage (MV) beam energies of Cobalt-60 (87.9%), 6 MV (6.1%) and 4 MV (6.1%). A lateral parallel-opposed pair of beams was the predominant technique (76.4%) versus an anterior oblique pair (17.2%) or angle-down pair (caudally directed fields to achieve shoulder clearance, 5.7%). Wedging (96.3%) and bolus (11.8%) were used as beam-modifying devices. Predominant field-width dimensions were 5.0-6.0 cm (43.4%) and 6.5-7.0 cm (43.1%), and field length dimensions were 5.0-6.0 cm (49.5%) and 6.5-7.0 cm (35.0%). Head, neck or chin immobilisation was used in 86.9% of the cases, with 94.6% of these being custom-made. We found that radiotherapy practice was stable over time, except for a trend of increasing field size and increasing use of immobilisation. In contrast, we found practice variations among the province's cancer centres. On the basis of our findings, we defined a predominant technical practice consisting of Cobalt-60 (reflecting machine availability during the period of the study), an isocentric set-up, a lateral parallel-opposed pair technique with wedging, and supine-head neutral positioning with custom immobilisation. Forty-two per cent of the cases had one or more components of treatment that differed from this definition.

CONCLUSIONS

Description of practice variation can provoke discussion about unrecognised differences in practice policies, perhaps identifying the need for better evidence, treatment guidelines, or both.

Time spent in hospital in the last six months of life in patients who died of cancer in Ontario

PURPOSE

To describe hospital bed utilization in the final 6 months of life in patients dying of cancer in Ontario, Canada.

PATIENTS AND METHODS

Hospital separation records were linked to a population-based cancer registry to identify factors associated with hospitalization in the 203,713 patients who died of cancer in Ontario between 1986 and 1998.

RESULTS

Between 1986 and 1998, 5.3% of all acute care beds in Ontario were devoted to the care of cancer patients in the last 6 months of life. The mean time spent in hospital in the last 6 months of life decreased from 34.3 days in 1986 to 22.7 days in 1998. Hospitalization rates increased exponentially during the last month of life. Patients younger than 50 years of age, women, and residents of poorer communities spent significantly longer in hospital than others. Hospitalization rates differed very little among the common solid tumors, but patients with CNS malignancies, the lymphomas, and the leukemias spent significantly longer in hospital than the other groups. There was significant interregional variations in hospitalization that were not explained by differences in case mix. There was a statistically significant inverse correlation between the rate of use of palliative radiotherapy and the hospital bed use in the county in which the patient resided.

CONCLUSION

The total time spent in hospital in the last 6 months of life has decreased over the last decade, but acute care hospitals continue to play a large role in the care of patients who are dying of cancer.

A comparison of published head and neck stage groupings in carcinomas of the tonsillar region

BACKGROUND

The combination of T, N, and M classifications into stage groupings was designed to facilitate a number of activities including: the estimation of prognosis and the comparison of therapeutic interventions among similar groups of cases. The authors tested the UICC/AJCC 5th edition stage grouping and seven other TNM-based groupings proposed for head and neck cancer to determine their ability to meet these expectations in a specific site: carcinoma of the tonsillar region.

METHODS

The authors defined four criteria to assess each stage grouping scheme: 1) The subgroups defined by T and N comprising a given group within a grouping scheme have similar survival rates (hazard consistency); 2) The survival rates differ across the groups (hazard discrimination); 3) The prediction of cure is high (outcome prediction); and 4) The distribution of patients among the groups is balanced. The authors identified or derived a measure for each criterion and the findings were summarized using a scoring system. The range of scores was from 0 (best) to 7 (worst). Data were from a retrospective chart review on 642 cases of carcinoma of the tonsillar region treated with radiotherapy for cure at the Princess Margaret Hospital from 1970-1991. None of the patients had distant metastases.

RESULTS

The scheme proposed by Synderman and Wagner, which was published in Otolaryngology Head and Neck Surgery in 1995 (vol.112, pages 691-4), scored best at 1.2. The UICC/AJCC scheme scored worst at 6.1. The hazard consistency ranged from a 3.1% average survival difference to 6.7% across the 8 schemes. The hazard discrimination measure varied by 28% from the best to worst scheme. Prediction varied by up to almost twofold across the schemes assessed. The distribution of patients varied from expected by between 0.13% and 0.57%.

CONCLUSION

UICC/AJCC stage groupings were defined without empirical investigation. When tested, this scheme did not perform as well as any of seven empirically-derived schemes the authors evaluated. The results of the current study suggest that the usefulness of the TNM system can be enhanced by optimizing the design of stage groupings through empirical investigation.

A comparison of published head and neck stage groupings in carcinomas of the oral cavity

BACKGROUND

The combination of T, N, and M classifications into stage groupings is meant to facilitate a number of activities, including the estimation of prognosis and the comparison of therapeutic interventions among similar groups of cases. We tested the UICC/AJCC 5th edition stage grouping and seven other TNM-based groupings proposed for head and neck cancer for their ability to meet these expectations in a specific site: carcinomas of the oral cavity.

METHODS

We defined four criteria to assess each grouping scheme: (1) the subgroups defined by T, N, and M that make up a given group within a grouping scheme have similar survival rates (hazard consistency); (2) the survival rates differ among the groups (hazard discrimination); (3) the prediction of cure is high (outcome prediction); and (4) the distribution of patients among the groups is balanced. We identified or derived a measure for each criterion, and the findings were summarized by use of a scoring system. The range of scores was from 0 (best) to 7 (worst). The data are population based from a prospectively gathered series in Southern Norway, with 556 patients diagnosed from 1983 through 1995. Clinical stage assignment was used, and the outcome of interest was cause-specific survival.

RESULTS

Summary scores across the eight schemes ranged from 1.66 for TANIS-3 to 6.50 for UICC/AJCC-5. The TANIS-7 staging scheme performed best on the hazard consistency criterion. The Kiricuta scheme performed best on the hazard discrimination criterion. Synderman predicted outcome best overall and Berg produced the most balanced distribution of cases among its groups.

CONCLUSIONS

UICC/AJCC stage groupings were defined without empirical investigation. When tested, this scheme did not perform as well as any of seven empirically derived schemes we evaluated. Our results suggest that the usefulness of the TNM system could be enhanced by optimizing the design of stage groupings through empirical investigation.

Glottic cancer in Ontario, Canada and the SEER areas of the United States. Do different management philosophies produce different outcome profiles?

We compared the management and outcome of glottic cancer in Ontario, Canada to that in the Surveillance, Epidemiology and End Results (SEER) Program areas in the United States to determine whether the greater use of primary radiotherapy with surgery reserved for salvage in Ontario was associated with similar survival and better larynx retention rates than the U.S. approach where primary surgery is used more often. Electronic, clinical and hospital data were linked to cancer registry data and supplemented by chart review where necessary. Initial treatment and survival in patients diagnosed in the SEER areas from 1988 through 1994 were compared to patients from Ontario diagnosed from 1982 through 1995. Actuarial laryngectomy rates were compared for patients over 65 at diagnosis in the two regions. Analyses were conducted over all cases and stratified by disease stage. In localized disease (T1 or T2), conservative treatment was the most common initial treatment in both regions, although total laryngectomy was used more often in SEER than Ontario (6.2% vs. 0.2%, respectively, P <.001). In advanced disease (T3 or T4), total laryngectomy was more commonly used as initial treatment in SEER (62.9% vs. 21.0% in Ontario, P < or =.001). Over all cases, the relative survival rate was 80% in Ontario at 5 years compared to 78% in SEER (P =.33). In localized disease, the relative survival rates were 4 to 5% higher in Ontario from the second year on, while in advanced disease 2 to 3% higher rates in SEER did not approach statistical significance. Actuarial laryngectomy rates at 3 years differed between the two regions, with a 4% higher rate in SEER (P =.01). In localized disease, 12.6% of Ontario patients had a laryngectomy by 3 years postdiagnosis compared to 17.9% in SEER (P =.05). In advanced disease, the rates were 63.3% and 79.2%, respectively (P =.07). There are large differences in the management of glottic cancer between the SEER areas of the U.S. and Ontario and no evidence that a policy emphasizing radiotherapy with surgery reserved for salvage is associated with worse survival. Ultimate laryngectomy rates are lower in Ontario for localized disease and may be lower for advanced disease. Conservation treatment should be used for localized disease while the treatment decision in advanced disease may be especially sensitive to patient values for voice retention versus initial cure.

The role of computed tomography in the T classification of laryngeal carcinoma

BACKGROUND

The objectives of this study were 1) to describe patterns of use of computed tomography (CT) in laryngeal carcinoma, and 2) to characterize the contribution of CT to the T classification of laryngeal carcinoma.

METHODS

The study population comprised 1195 patients with laryngeal carcinoma diagnosed from 1982 through 1995 chosen randomly from the Ontario provincial cancer registry. A chart review was conducted to obtain data on each case. Patient-related, tumor-related, and health-system-related factors were analyzed to identify factors associated with the use of CT. Descriptions of clinical exams and CT reports were reviewed to see how CT information modified T classification. Actuarial local control and cause specific survival curves were plotted by clinical T classification without and with CT to evaluate stage migration. The percentage of the variance in outcome explained by T classification in a Cox analysis was used to evaluate whether the prognostic accuracy of T classification was improved with the use of information from CT.

RESULTS

Patients with glottic (20.1%) and supraglottic (41.7%) carcinoma underwent CT. The use of CT increased over time in glottic and supraglottic carcinoma combined from 17.2% in 1982-5 to 33.9% in 1991-5. Computed tomography was used less often in older patients with a 16% (95% confidence interval, 5-27%) decrease in the odds of having CT with each 10-year age increment. Computed tomography use varied considerably across the cancer center regions in Ontario. Computed tomography altered the T classification in 20.2% of those patients who had CT, with most being "upstages." Stage migration due to CT was demonstrated. Using information from CT in the assignment of T classification for 27.8% of this study population did not make a significant contribution to the ability of T classification to predict outcome over the entire group.

CONCLUSIONS

There is large variation in the use of CT among different age groups and regions. The ability to compare outcomes by stage across geographic areas is compromised when the use of CT varies.

Time to first relapse as an outcome and a predictor of survival in patients with squamous cell carcinoma of the head and neck

OBJECTIVE

To identify patterns and predictors of relapse and to determine whether the time to relapse influences survival.

STUDY DESIGN

Multivariate anal. ysis of prospective database.

METHODS

We present findings in 446 consecutive prospective patients from a regional cancer center with invasive squamous cell carcinoma who were treated for cure and were declared disease free. Time to relapse, site of relapse, and survival are reported by stage and site using bivariate and multivariate analysis.

RESULTS

Thirty-six percent of patients relapsed. The median time to relapse was 8.3 months, and 95% of relapses occurred within 3 years. T stage in the TNM staging system was associated with relapse, and N stage was not. In patients who relapsed before the median time to relapse, prognosis was worse than in those who relapsed after that time.

Associations between community income and cancer incidence in Canada and the United States

BACKGROUND

Associations between socioeconomic status (SES) and the incidence of cancer have been reported previously in the U.S. Canada has more comprehensive health care and social programs than the U.S. The purpose of this study was to compare the strength of associations between SES and cancer incidence in Canada and the U.S.

METHODS

The regions studied were the Canadian province of Ontario and the areas of the U.S. covered by the Surveillance, Epidemiology, and End Results (SEER) program. The populations at risk were defined using the 1991 Canadian Census and the 1990 U.S. Census. The populations of Ontario and of the SEER areas of the U.S. were each divided into deciles on the basis of median household income. Population-based cancer registries were used to identify incident cases. Age-standardized incidence rates for all major groups of malignant diseases were calculated for each SES decile in Ontario and in the U.S. Income-associated incidence gradients observed in Ontario and the U.S. were compared.

RESULTS

The incidence of most types of cancer was similar in Ontario and the U.S. In both countries, there were moderately strong, inverse associations between income level and the incidence of carcinomas of the cervix, the head and neck region, the lung, and the gastrointestinal tract. In both Ontario and the U.S., several of these diseases were twice as common in the bottom income decile than they were in the top decile. In contrast, carcinoma of the female breast and carcinoma of the prostate were more common among higher income communities in both countries, but the observed associations were weaker in Ontario.

CONCLUSIONS

Despite Canada's universal health insurance and more comprehensive social security system, the association between lower socioeconomic status and the incidence of many common cancers is just as strong in Ontario as it is in the U.S. The mechanisms responsible for these associations require further investigation.

The impact of comorbidity on the survival of patients with squamous cell carcinoma of the head and neck

BACKGROUND

In North America, cigarette smoking and/or alcohol consumption not only cause head and neck cancer, they also cause many of the other diseases, illnesses, and conditions, also known as comorbidities, frequently found in our patients. Comorbidities can influence treatment decision making and treatment outcome. The aim of this study is to quantify the increased risk of comorbidity in our patients.

METHOD

The survival of 655 consecutive patients with squamous cell carcinoma from a regional cancer center is analyzed. We compare the survival curves for all-cause death, death from cancer, and death from noncancer causes to the expected survival of age/sex-matched populations of Ontario residents, Canadian smokers, and Canadian nonsmokers.

RESULTS

Of those patients who had not survived 5 years, 59% died of their index tumor, 23% would have been expected to die if they did not have head and neck cancer, and 18% died of the increased comorbidity associated with being a patient with head and neck cancer.

DISCUSSION

Comorbidity, and specifically the increased comorbidity found in patients with head and neck cancer, is an important factor in overall survival.

The management and outcome of bladder carcinoma in Ontario, 1982-1994

BACKGROUND

To the authors' knowledge no previous study has described the management and outcome of bladder carcinoma on a population-based level. The objective of the current study was to describe the characteristics, treatment, and outcome of newly diagnosed invasive bladder carcinoma (n = 20,822) reported in Ontario, Canada between 1982-1994.

METHODS

Electronic records of invasive bladder carcinoma (International Classification of Diseases code 188) from the Ontario Cancer Registry were linked to surgical and radiotherapy (RT) records. Bivariate and multivariate techniques were used to assess variations in the use of initial cystectomy and pelvic RT. The authors modeled the likelihood of death after diagnosis and the probability of cystectomy free survival. All analyses were controlled for age, gender, histology, and year of diagnosis.

RESULTS

The most common histologic type was papillary transitional cell carcinoma. Maximum initial treatment was comprised of total cystectomy (5.1%), partial cystectomy or open excision (3. 5%), pelvic RT (5.9%), transurethral resection of the bladder (66. 7%), or lesser or no procedures (18.7%). The use of total cystectomy and pelvic RT varied among the regions of Ontario. Overall 5-year survival was 58.8%, and was 86.5% for patients with papillary histology. In multivariate analysis, although survival was similar among the regions, the relative risk of cystectomy conditional on survival varied.

CONCLUSIONS

Papillary tumors portend a better survival than nonpapillary tumors. Variations in the use of total cystectomy and in the use of pelvic RT among the regions of Ontario did not appear to be associated with variations in survival. However, cystectomy free survival appeared to vary among the regions. These results suggest that patients can be managed safely using a bladder-preserving approach.

Association between age and the utilization of radiotherapy in Ontario

PURPOSE

The purpose of this study was to assess whether: (i) radiotherapy (RT) utilization varies with age in Ontario cancer patients; (ii) age-associated differences in the use of RT (if they exist) vary with cancer site and treatment intent; (iii) the age-associated variation in RT utilization is comparable to the decline in functional status in the general population; and (iv) the variation with age is due to differences in referral to a cancer center or to subsequent decisions.

METHODS AND MATERIALS

Details for several cancer sites diagnosed between 1984-1994 were obtained from the Ontario Cancer Registry (OCR). RT records from all treatment centers were linked to the OCR database. Information about the functional status of the Canadian population was obtained from the 1994 National Population Health Survey conducted by Statistics Canada.

RESULTS

The rate of RT use declined with age, particularly for adjuvant and palliative indications. The relative decline in RT with age exceeded the relative decline in functional status with age in the general population. Most of the decline in RT use was related to a decline in referral to cancer centers.

CONCLUSIONS

The referral for, and use of, palliative and adjuvant RT decreases more with age than can be explained by age-associated decline in functional status observed in the general population.

Management and outcome of glottic cancer: a population-based comparison between Ontario, Canada and the SEER areas of the United States. Surveillance, Epidemiology and End Results

OBJECTIVE

We compared treatment practice and outcome in glottic cancer in Ontario, Canada to that in the Surveillance, Epidemiology and End Results (SEER) program areas in the United States to determine whether the Ontario emphasis on the use of delayed combined therapy was associated with similar survival and better laryngectomy-free survival than the U.S. approach, which emphasizes greater use of surgery.

METHODS

Electronic, clinical, and hospital data were linked to cancer registry data. The study groups compared on survival comprised all patients diagnosed from 1982 to the end of 1991 in Ontario (2324 patients) and in the SEER areas (5715 patients). Comparisons on initial treatment, laryngectomy rates, and laryngectomy-free survival were limited to subsets of these study populations due to data availability. Initial treatment data were provided by the SEER registries in the U.S. and by the cancer clinic and hospitalization data in Ontario. Information about laryngectomies performed subsequent to initial treatment was available from Medicare hospitalization data in the U.S. and from Canadian Institute for Health Information hospitalization data in Ontario.

RESULTS

Although radiotherapy was the most common initial treatment in both areas, it was used more often in Ontario (84.4% versus 63.2% in the U.S. [p < 0.001]). Relative survival was not statistically different with a relative risk comparing SEER to Ontario of 1.09, 95% confidence interval (CI) (0.93, 1.29). Laryngectomy rates were similar with a relative risk of 1.01, 95% CI (0.67, 1.52), and it follows from the survival and laryngectomy rate comparisons that the laryngectomy-free survival was not statistically different (p = .95).

CONCLUSIONS

There are large differences in the management of glottic cancer between the U.S. and Ontario and no corresponding differences in survival or laryngectomy-free survival. This work highlights a need for more clinical investigation into the relative merits of differing management policies in glottic cancer.

Cancers of the upper aerodigestive tract in Ontario, Canada, and the United States

BACKGROUND

Squamous cancers of the upper aerodigestive tract (UADT) are related to the use of tobacco and/or alcohol, and in North America they are more common among the poor. They are usually locoregionally confined at diagnosis, and local treatment with surgery and/or radiation therapy is often curative. This study compares the incidence and survival of this group of diseases in Canada and the U.S., two North American neighbors with many cultural similarities but significant differences in their health care and social programs.

METHODS

To describe and compare the case mix, incidence, and outcome of squamous cancers of the UADT in Ontario, Canada, and the U.S., we used the Ontario Cancer Registry (OCR) and the Surveillance, Epidemiology, and End Results (SEER) registries in the U.S. to identify all cases of cancer with International Classification of Disease (ICD) codes 141, 143-9, 160-1, and a subset of 140, which were diagnosed between 1982 and 1994. ICD-O histology codes were placed into clinically relevant groupings, and ICD-9 site codes were grouped into sites as defined by the International Union Against Cancer and the American Joint Committee on Cancer. Age-adjusted incidence rates were calculated for each site. For the SEER registry, race specific incidence rates were also calculated. Observed and expected survival were plotted by site and registry, and from these, relative survival was calculated. Survival was compared during the first 5 years after diagnosis and during the next 5 years among patients who had survived the first 5 years.

RESULTS

Of the 16,577 and 42,990 cases identified in the OCR and SEER registries, respectively, squamous cancer was by far the most common histology (94.1% in OCR, 94.6% in SEER) and will form the main subject of this report. The distribution of squamous cancers by site, subsite, age, and gender were remarkably similar in the two populations. Overall, the incidence was about 17% higher in the U.S. than in Ontario, and this difference was seen for all sites except the nasopharynx, which was more common in Ontario. The higher incidence in the U.S. in part reflects the much higher rate for African Americans than for Americans of other ethnic backgrounds. During the first 5 years after diagnosis, when most deaths from UADT cancer occur, there was a significant relative survival difference in favor of the U.S. for cancer of the supraglottis, and in favor of Ontario for cancer of the oral cavity. There was a nonsignificant trend in favor of Ontario for cancer of the nasopharynx. Within the SEER population, for all sites except the nasopharynx, 5-year relative survival was considerably worse for African Americans than for Americans of other ethnic backgrounds. Examination of survival beyond 5 years after diagnosis for patients who had survived the first 5 years revealed that for all sites, the observed survival continued to diverge markedly from the expected survival. The excess mortality ranged from less than 20% for glottic and nasopharyngeal cancers to about 30-40% for oropharyngeal and supraglottic cancers.

CONCLUSIONS

Despite remarkable similarities in case mix between the two countries, UADT cancers were more frequent in the SEER population of the U.S. than in Ontario, and this was partly attributable to the much higher incidence among African Americans. Significant differences between the registries in 5-year survival were seen for several sites. African Americans with UADT cancers had much worse prognoses than did Americans of other ethnic backgrounds. Patients who survive their UADT cancer remain at a higher-than-expected risk of death even after they have been cured.

A population-based study of the effectiveness of breast conservation for newly diagnosed breast cancer

PURPOSE

Our objective was to evaluate the effectiveness of breast conservation for newly diagnosed breast cancer. Effectiveness was operationalized as two outcomes within 5 years of the diagnosis of breast cancer: the probability of mastectomy-free survival (either death or mastectomy count as event, whichever comes first), and the probability of mastectomy conditional on survival (mastectomy counts as event, observations censored at death).

METHODS AND MATERIALS

We linked records of 46,687 new cases of breast cancer from 1982 to 1991 in the Ontario Cancer Registry to records of surgery from 1982 to 1995, radiotherapy (RT) from 1982 to 1992, and median household income from the 1986 census. We labeled breast surgery within 4 months and postoperative RT within 12 months of diagnosis as treatment for newly diagnosed breast cancer. Surgery was categorized as mastectomy, lumpectomy plus RT, lumpectomy alone, or no surgical procedure. Among cases that did not undergo mastectomy within 4 months of diagnosis, we labeled mastectomy subsequent to 4 months after diagnosis as treatment failure. We performed life-table analysis and Cox proportional hazards regression, to describe the probability of mastectomy conditional on survival and the probability of mastectomy-free survival.

RESULTS

A total of 16,279 cases underwent lumpectomy as the maximum procedure on the breast within 4 months of diagnosis, and 49.7% of these received postoperative RT. Compared to the provincial mean, regions with higher rates of lumpectomy plus RT have higher probability of mastectomy-free survival and lower probability of mastectomy conditional upon survival 5 years after diagnosis of breast cancer.

CONCLUSIONS

These findings are consistent with a hypothesis that breast conservation is effective in the overall breast cancer population of Ontario within the first 5 years after diagnosis.

A population-based study of the use and outcome of radical radiotherapy for invasive bladder cancer

PURPOSE

The objective of this study is to describe the use and outcome of radical radiotherapy for bladder cancer in the province of Ontario, Canada, between 1982 and 1994.

METHODS

Electronic records of invasive bladder cancer (ICD code 188) from the Ontario Cancer Registry were linked to surgical records from all Ontario hospitals and radiotherapy (RT) records from all Ontario cancer centers. We identified cases receiving radical RT by selecting RT records containing "bladder" or "pelvis" anatomic region codes and a radical or curative intent code (or dose >39.5 Gy if intent missing). We identified cases receiving salvage total cystectomy by selecting total cystectomy procedure codes occurring at any time beyond 4 months from the start of radical RT. We used life table methods to compute the following: the time from diagnosis to radical RT, the time from radical RT to salvage cystectomy, overall and cause-specific survival from radical radiotherapy to death, and overall and cause-specific survival from salvage cystectomy to death. We modeled the factors associated with time to death, time to cystectomy conditional on survival, and time to cystectomy or death, whichever came first, using Cox proportional hazards regression.

RESULTS

From the 20,906 new cases of bladder cancer diagnosed in Ontario from 1982 to 1994, we identified 1,372 cases treated by radical radiotherapy (78% male, 22% female; mean age 69.8 years). The median interval to start of radical RT from diagnosis was 13.4 weeks. Ninety-three percent of patients were treated on high-energy linacs, and the most common dose/fractionation scheme was 60 Gy/30 (31% of cases). Five-year survival rates were as follows: bladder cancer cause-specific, 41%; overall, 28%; cystectomy-free, 25%; bladder cancer cause-specific following salvage cystectomy, 36%; overall following salvage cystectomy, 28%. Factors associated with a higher risk of death and a poorer cystectomy-free survival were histology (squamous or nonpapillary transitional cell carcinoma [TCC]) and advanced age.

CONCLUSION

This population-based study confirms previous institutional studies and clinical trials and shows that radical RT has a curative role in the management of invasive bladder cancer and allows about one-quarter of patients receiving radiotherapy to survive 5 years while retaining the bladder. Salvage cystectomy following RT provides a chance of cure at the time of bladder relapse.

A population-based study of rectal cancer: permanent colostomy as an outcome

PURPOSE

The objectives of this study are to describe the utilization of surgery and of radiotherapy in the treatment of newly diagnosed rectal cancer in Ontario between 1982 and 1994, and to describe the probability of permanent colostomy at any time after the diagnosis of rectal cancer, as an outcome of the treatment of newly diagnosed rectal cancer.

METHODS AND MATERIALS

Electronic records of rectal cancer (International Classification of Diseases code 154) from the Ontario Cancer Registry (n = 18,695, excluding squamous, basaloid, cloacogenic, and carcinoid histology) were linked to surgical records from all Ontario hospitals, and radiotherapy (RT) records from Ontario cancer centers. Procedures occurring within 4 months of diagnosis, or within 4 months of another procedure for rectal cancer, were considered part of initial treatment. Multivariate analyses controlled for age, sex, and year of diagnosis.

RESULTS

Resection plus permanent colostomy was performed in 33.1% of cases, whereas local excision or resection without permanent colostomy was performed in 38.2%. Multivariate logistic regression demonstrated higher odds ratios (OR) for resection plus permanent colostomy in all regions of Ontario relative to Toronto. The OR for postoperative RT following local excision or resection without permanent colostomy varied among the regions relative to Toronto (e.g., OR Ottawa = 0.59, OR Hamilton = 0.76, OR London = 1.25). The relative risk (RR) of colostomy conditional upon survival within 5 years from diagnosis varied among regions relative to Toronto (e.g., RR Ottawa = 1.21, RR Hamilton = 1.20).

CONCLUSIONS

There is regional variation in the utilization of resection with permanent colostomy, and in the utilization of postoperative RT among cases not undergoing permanent colostomy. Regions with higher initial rates of resection plus permanent colostomy continue to experience higher probability of permanent colostomy 5 years after diagnosis of rectal cancer. Higher initial rates of permanent colostomy may be malleable to interventions aimed at improving overall outcomes.

The repeatability of three methods for measuring prospective patients' values in the context of treatment choice for end-stage renal disease

In the context of the choice of treatment for end-stage renal disease (ESRD), three approaches to value assessment were examined for their repeatability over time within subjects. If formal decision analyses are to be used to advise patients about treatment choice, then repeatable value assessment methods, an essential component of such analyses, are needed. The methods assessed were standard gamble (SG), time trade-off (TTO), and visual analogue (VA). Sixty-six nephrology clinic patients were interviewed on two occasions, 10 days apart, by one of two interviewers. An information session was conducted 1 week before the first interview. Subjects were taught about the treatments using an information package developed expressly for the study and a video produced by a pharmaceutical company for use in this decision context. Patients differed widely in the values provided for the various treatments of ESRD, with responses that ranged across the entire scale (0 to 100). The repeatability of the three methods was poor, with the coefficients of repeatability (95% range of differences from one occasion to the next) observed as +/- 27.4 for SG, +/- 38.4 for TTO, and +/- 36.5 for VA. When subsets defined by characteristics that may have improved the repeatability were analyzed, the magnitude of the error did not vary substantially. The poor repeatability of these methods raises questions about their use for decision analyses applied to the individual context.

Associations between community income and cancer survival in Ontario, Canada, and the United States

PURPOSE

The objectives of this study were as follows: (1) to compare the magnitude of the association between socioeconomic status (SES) and cancer survival in the Canadian province of Ontario with that in the United States (U.S.), and (2) to compare cancer survival in communities with similar SES in Ontario and in the U.S.

METHODS

The Ontario Cancer Registry provided information about all cases of invasive cancer diagnosed in Ontario from 1987 to 1992, and the Surveillance, Epidemiology and End Results Registry (SEER) provided information about all cases diagnosed in the SEER regions of the U.S. during the same time period. Census data provided information about SES at the community level. The product-limit method was used to describe cause-specific survival. Cox proportional hazards models were used to describe the association between SES and the risk of death from cancer.

RESULTS

There were significant associations between SES and survival for most cancer sites in both the U.S. and Ontario, but the magnitude of the association was usually larger in the U.S. In the poorest communities, there were significant survival advantages in favor of cancer patients in Ontario for many disease groups, including cancers of the lung, head and neck region, cervix, and uterus. However, in upper- and middle-income communities, there were significant survival advantages in favor of the U.S. for all cases combined and for several individual diseases, including cancers of the breast, colon and rectum, prostate, and bladder.

CONCLUSION

The association between SES and cancer survival is weaker in Ontario than it is in the U.S. This is due to a combination of better survival among patients in the poorest communities and worse survival among patients in the wealthier communities of Ontario relative to those in the U.S.

Mortality from myocardial infarction following postlumpectomy radiotherapy for breast cancer: a population-based study in Ontario, Canada

PURPOSE

To compare the risk of mortality from myocardial infarction (MI) after left-sided postlumpectomy radiotherapy (RT) to the risk after right-sided postlumpectomy RT.

METHODS

We conducted a population-based cohort study of cases of invasive female breast cancer in Ontario, diagnosed between January 1, 1982 and December 31, 1987 (n = 25,570). Records of the Ontario Cancer Registry (OCR) were linked to hospital procedure and discharge abstracts and to RT records from Ontario cancer centers. A case was labelled as lumpectomy if this was the maximum breast surgery within 4 months of diagnosis. Postlumpectomy RT occurred up to 1 year postdiagnosis. Laterality was assigned from the laterality descriptor of the RT records. A case was labelled as having had a fatal MI if ICD code 410 (myocardial infarction) was recorded as the cause of death in the OCR. We used logistic regression to compare the likelihood of utilization of: 1. Dose per fraction > 2.00 Gy; 2. cobalt vs. linac; and 3. boost RT. We used life table analysis and the log rank test comparing the time to fatal MI from diagnosis of breast cancer between women who received left-sided postlumpectomy RT and women who received right-sided. We used Cox proportional hazards models to study the relative risk for left-sided cases overall, and stratified by age, RT characteristics, and among conditional survival cohorts.

RESULTS

Postlumpectomy RT was received by 1,555 left-sided and 1,451 right-sided cases. With follow-up to December 31, 1995, 2% of women with left-sided RT had a fatal MI compared to 1% of women with right-sided RT. Comparison of the time to failure between women who had left-sided RT and women who had right-sided RT showed the left-sided RT group to be associated with a higher risk of fatal MI (p = 0.02). Adjusting for age at diagnosis, the relative risk for fatal MI with left-sided postlumpectomy RT was 2.10 (1.11, 3.95).

CONCLUSION

Among women who received postlumpectomy RT for breast cancer in Ontario between 1982-1987, left-sided postlumpectomy RT was associated with a higher risk of fatal MI compared to right-sided.

Using TNM staging to predict survival in patients with squamous cell carcinoma of head & neck

BACKGROUND

There is a need for a classification system for prognosis based on the TNM system for patients with squamous cell carcinoma of the head and neck such that patient groupings are homogeneous within and heterogeneous between.

METHODS

Six hundred fifty-five consecutive patients with invasive squamous cell carcinoma of the head and neck followed prospectively are split into a training set and a test set. Using the training set, the Cox Proportional Hazards Model and the outcome of dead with disease, we created homogeneous prognostic levels (PLs) based on RR. Using the test set, we compare our model to others in the literature.

RESULTS

Using the training set, we identified five PLs, and using the test set, we demonstrated that our model is superior to others for both within-group homogeneity and between-group heterogeneity.

CONCLUSIONS

A simple classification system can be used to group patients for survival and is valid for future study in prognostication.

Using the TNM system to predict survival in squamous cell carcinoma of the head and neck

PURPOSE

To create a classification system for prognosis based on the TNM system for patients with squamous cell carcinoma of the Head and Neck such that patient groupings are homogeneous within and heterogenous between.

METHOD

655 consecutive patients with invasive squamous cell carcinoma of the Head and Neck followed prospectively are split into a training set and a test set. Using the training set, the Cox Proportional Hazards Model and the outcome of dead with disease, we create homogenous prognostic levels based on relative risk. Using the test set, we compare our model to others in the literature.

RESULTS

Using the training set we identify 5 prognostic levels and using the test set we demonstrate that our model is superior to others for both within group homogeneity and between group heterogeneity.

CONCLUSIONS

A simple classification system can be used to group patients for survival and is valid for future study in prognostication.

Mortality from myocardial infarction after adjuvant radiotherapy for breast cancer in the surveillance, epidemiology, and end-results cancer registries

PURPOSE

To compare the risk for fatal myocardial infarction (MI) after adjuvant radiotherapy (RT) for left-sided breast cancer with the risk for MI after adjuvant RT for right-sided breast cancer.

METHODS

We studied women with local- and regional-stage breast cancer diagnosed from 1973 to 1992 from the Surveillance, Epidemiology, and End-Results (SEER) cancer registries. We performed life-table analysis, the log-rank test, and Cox proportional hazards regression to compare the time to fatal MI from diagnosis between left-sided and right-sided cases, censoring deaths from other causes.

RESULTS

Among irradiated patients, the relative risk (RR) for fatal MI in women with left-sided breast cancer was 1.17 (95% confidence interval [CI], 1.01 to 1.36), controlling for age, compared with those with right-sided breast cancer. The RR for fatal MI among left-sided cases was increased for those under the age of 60 years (RR = 1.98; 95% CI, 1.31 to 2.97) compared with right-sided cases, but not at age 60 years or older. Among women with irradiated regional-stage cancer who were younger than 60 years of age, the risk was significantly increased (RR = 2.24; 95% CI, 1.38 to 3.64) for those with left-sided compared with right-sided breast cancer, but not among patients aged 60 years or older. Among irradiated local-stage cases, the risk for those with left-sided breast cancer was not significantly elevated in either age category. Analysis of 5-year conditional survival cohorts showed an increased risk for irradiated left-sided cases among women younger than 60 years of age in the 10- to 15-year conditional survival cohort (RR = 5.28; 95% CI, 1.82 to 15.3).

CONCLUSION

Adjuvant RT for left-sided breast cancer diagnosed in women younger than 60 years of age is associated with a higher risk for fatal MI 10 to 15 years later compared with adjuvant RT for right-sided cases.

Radiotherapy for breast cancer in Ontario: rate variation associated with region, age and income

OBJECTIVE

To describe the variation in the use of radiotherapy (RT) in women in Ontario within 1 year of diagnosis of breast cancer, from 1982 to 1991, and to identify factors associated with these variations.

DESIGN

Retrospective, population-based cohort study.

SETTING

Ontario.

POPULATION

All women registered by the Ontario Cancer Registry (OCR) with a diagnosis of invasive breast cancer between Jan. 1, 1982, and Dec. 31, 1991.

INTERVENTIONS

RT to any anatomic site within 1 year of the diagnosis of breast cancer.

OUTCOME MEASURES

Odds of receiving RT within 1 year of diagnosis (from RT files from all radiotherapy departments in Ontario) associated with year and with geographic, age-related and socioeconomic factors.

RESULTS

Use of RT within 1 year of diagnosis increased from 21.1% (95% confidence interval [CI] 19.8-22.4) in 1982 to 44.7% (95% CI 43.4-46.0) in 1991 (p < 0.0001). Among the regions of Ontario, the use of RT varied from 24.5% (95% CI 23.5-25.6) to 44.4% (95% CI 43.0-45.9) (p < 0.0001). Increasing age was associated with decreasing likelihood of receiving RT (test for trend p < 0.0001), as was decreasing income (test for trend p < 0.0001).

CONCLUSIONS

The use of RT within 1 year of the diagnosis of breast cancer in women in Ontario varies by region, age and income. Despite universal and comprehensive health insurance coverage, women with breast cancer in some populous regions of Ontario were less likely to receive RT within 1 year of their diagnosis than women in other populous regions.

Socioeconomic status and cancer survival in Ontario

BACKGROUND AND PURPOSE

It is known that the socioeconomic status (SES) of the patient is associated with cancer survival in the United States. The purpose of this study was to determine whether the association between SES and survival is also present in Canada, a society with a comprehensive, universal, health insurance program.

METHODS

A population-based cancer registry was used to identify the 357,530 cases of invasive cancer diagnosed in the Canadian province of Ontario between 1982 and 1991. Information from the 1986 Canadian census was linked to the registry and used to describe the SES of the area in which each patient resided. Cox regression was used to describe the association between median household income and survival while controlling for age, sex, and the region in which the patient resided. The Cox model was fitted in a competing risk framework to assess the association between income and the probability of specific causes of death.

RESULTS

Lung cancer and cancers of the head and neck region were relatively more common in poor-income communities, and cancers of the breast, CNS, and testis were relatively more common in richer communities. A strong and statistically significant association between community income and survival was observed in cancers of the head and neck region, cervix, uterus, breast, prostate, bladder, and esophagus. Smaller, but significant associations were seen in cancers of the lung and rectum. No significant association between community income and survival was observed in cancers of the stomach, colon, pancreas, or ovary. Analysis of the cause of death showed that community income is associated both with the probability of death from cancer and with the probability of death from other causes.

CONCLUSION

Although Canada's health care system was designed to provide equitable access to equivalent standards of care, it does not prevent a difference in cancer survival between rich and poor communities.

Decision analysis in locally advanced non-small-cell lung cancer: is it useful?

PURPOSE

The optimal management of locally advanced non-small-cell lung cancer (NSCLC) has not been established. While combined-modality treatments have been shown to increase the survival of patients with this illness, the appropriate balance between the benefit of increased quantity of life and the quality-of-life costs of the more toxic treatment combinations remains unresolved. Decision analysis has been promoted as useful when medical decisions must be made under conditions of uncertainty. We consider the potential of this method to guide therapy in locally advanced NSCLC.

METHODS

We developed two types of decision models that addressed the choice between radiation alone and combined chemotherapy-radiation therapy in locally advanced NSCLC. The models were constructed using the principles of decision analysis.

RESULTS

The models successfully replicated results of relevant clinical trials published in the literature. The analyses of both models showed that the treatment decision was sensitive to patients' values, despite significant increases in survival rates. The models clarified a need for further validation of the three fundamental components: structuring the decision, determining the probabilities of events, and assigning utilities to treatment outcomes.

CONCLUSION

In the setting of NSCLC, the models suggest that quality-of-life considerations are important in the treatment choice. Further research is required to identify the health states critical to the decision, the probabilities for occurrence of these health states, and valid measures of their utility.

Does a centralized radiotherapy system provide adequate access to care?

PURPOSE

In the Canadian province of Ontario, all radiotherapy is provided by a centrally managed provincial network of nine cancer centers. The primary goal of this study was to determine whether this highly centralized radiotherapy system provides adequate and equitable access to care for the province's dispersed population.

METHODS

The Ontario Cancer Registry (OCR) was used to identify 295,386 cases of invasive cancer, excluding nonmelanoma skin cancer, which were diagnosed in Ontario between 1984 and 1991. Electronic radiotherapy records from each of the province's radiotherapy centers were linked to the registry at the level of the individual case.

RESULTS

The proportion of incident cases treated with radiotherapy was 18.8% at 4 months after diagnosis, 23.7% at 1 year, 25.8% at 2 years, 28.2% at 5 years, and 29.1% at 8 years. These rates of radiotherapy use are much lower than the accepted national and international targets, and lower than rates reported from other jurisdictions. The rate of radiotherapy use at 1 year varied significantly from county to county across Ontario (range, 18.6% to 32.4%; P < 10(-6)), and the highest rates were recorded in communities close to radiotherapy centers. There was a common geographic pattern of rate variations among several disease groups, including breast cancer, lung cancer, the genitourinary malignancies, and the gastrointestinal malignancies.

CONCLUSION

The low and uneven rates of radiotherapy use across the province indicate that Ontario's centralized radiotherapy system does not, at present, provide adequate or equitable access to care.

Does TANIS (T And N Integer Score) help predict survival?

OBJECTIVE

This study was conducted to validate the TANIS model as a predictor of survival.

DESIGN

Using Cox Proportional Hazards model, we examined the subgroups of the TANIS score levels to see if they were homogeneous on survival.

SETTING

The Departments of Otolaryngology and Oncology and the Radiation Oncology Research Unit at Queen's University in Kingston, Ontario.

METHODS

Five hundred and ninety-four patients with cancer of the head and neck classified by TNM and TANIS were analyzed for the outcome dead of disease using estimates of relative risk.

RESULTS AND CONCLUSION

The stages of TNM and the levels of TANIS contain subgroups of patients with different survival rates.

Necrosis following radiotherapy for carcinoma of the pinna

PURPOSE

Radiation therapy is often the preferred modality of treatment for carcinoma of the pinna because it avoids the cosmetic defect of surgery. However, radiation oncologists are sometimes reluctant to irradiate the ear because of the risk of subsequent necrosis. The goal of this study was to establish the long-term disease control and necrosis rates following irradiation of the external ear.

METHODS AND MATERIALS

A retrospective analysis was undertaken of 138 courses of curative radiotherapy given to 128 patients for biopsy-proven basal (70 courses), squamous (62 courses), or mixed (6 courses) tumors of the pinna between January 1, 1982, and December 31, 1991, at the Kingston Regional Cancer Center.

RESULTS

The median age of the patients was 73 (range 43-94) and the median size of the tumors was 12 mm (range 3-50 mm). Treatment was given using orthovoltage X rays (79) or electrons (59). The most common dose prescription was 35 Gy/5 fractions; total doses ranged from 17.50 to 64 Gy. The median follow-up is 58 months (range 6-149). The actuarial 5-year local control rate is 93%; the actuarial necrosis rate at 5 years is 13%. Most necroses healed with conservative management; only two patients required surgery for necrosis. We analyzed the following factors as possible predictors of radiation necrosis: patient age, size of lesion, histology, fraction size, total dose, overall time, and beam energy. Only daily fraction sizes > 6 Gy (p = 0.0093) and treatment times < 5 days (p = 0.0053) were significantly associated with an increased risk of necrosis.

CONCLUSION

To reduce the risk of necrosis, radiation therapy for external ear cancer should be given using protracted fractionation.

Does N stage predict survival?

OBJECTIVE

To examine the relationship between N stage and survival in head and neck cancer to see if N stage has prognostic value.

DESIGN

Database analysis using the Cox Proportional Hazards Model.

METHOD

Five hundred and twenty-three consecutive patients with squamous cell carcinoma of the head and neck prospectively entered into a clinical database are analyzed for the outcome of dead with disease.

RESULTS

There is no difference in survival between some increments of N stage, specifically N1 and N2, controlling for T stage and site.

CONCLUSION

A modified N stage system consisting of N0, N(limited), and N(extended) appears to have a better prognostic value.

Outmigrant ascertainment for bias assessment in environmental epidemiology

BACKGROUND

In the summer of 1985, the McGill Epidemiology Rapid Response Unit undertook a study comparing two areas exposed to emissions of 'sour gas' refineries to an unexposed area. One operational objective of the project was the identification and survey of all the people who had lived in the study area but had since moved (outmigrants).

METHODS

We estimated the number of outmigrants (people who had ever lived in the area during the period 1957 to 1985) to be 3363 by using information obtained from our cross-sectional survey and from population statistics for the area of interest. Ten different methods combined lead to the identification of approximately 87% of all the outmigrants who left the study area during that period. We used vital statistics to identify the outmigrants who had died and mailed questionnaires to obtain the necessary information from the others.

RESULTS

We confirmed the vital status of approximately 46% of them (1532/3363). The results from the outmigrant survey showed that they were younger than area residents, that they experienced lower rates of heart disease and hypertension and that they had moved for health reasons in only 1.3% of the cases. These findings were similar across comparison areas.

CONCLUSION

We concluded that there was no effect due to migration bias on the cross-sectional study results. This evidence considerably strengthened the conclusions regarding the effects of exposure, a benefit that largely justified the cost of identifying and surveying the outmigrants.

Quality adjusted life years added by treatment of obstructive sleep apnea

We assessed the impact of treatment with nasal continuous positive airway pressure (nCPAP) on the quality of life of 19 patients with obstructive sleep apnea (OSA). We measured the utility for the patients' health states before and with treatment using the standard gamble approach. The study group had an average age of 57 years and had been on treatment for a mean of 9.5 months. For all the patients, the polysomnographic indicators of OSA disease severity improved markedly with treatment. For nine of the 12 symptoms most commonly associated with OSA, the patients reported improvement during treatment. The mean utility and the standard deviation obtained with the standard gamble method were 0.87 +/- 0.17 on treatment and 0.63 +/- 0.29 pretreatment. The difference in utility between treatment and pretreatment health states was combined with the life expectancy of each patient to generate the number of quality adjusted life years (QALYs) considered equivalent to the impact of treatment. This resulted in an average gain of 5.4 QALYs. When we related this impact to the cost of treatment, we obtained a cost-utility ratio between Can $3,397 and Can $9,792 per QALY added. These costs are relatively small when compared to the cost per QALY for many other clinical interventions. Hence, nCPAP clearly offers the prospect for a well-tolerated therapy with a very favorable cost-utility ratio.

Content of a decision analysis for treatment choice in end-stage renal disease: who should be consulted?

For an illness whose treatment options lead to similar survival, quality-of-life issues determine the choice. If the patient has not yet experienced the treatments, then the information provided regarding the options plays a crucial role in the patient's choice. The authors interviewed 43 people who are intimately involved with the treatments for end-stage renal disease (ESRD) (ten physicians, 11 nurses, and 22 patients) with a view to determining what information needs to be considered when choosing between the various treatment options for this illness. They compared the three groups based on the items obtained from the interviews to determine whether the inclusion of patients in the process changed the content of the treatment descriptions. They were also interested in determining whether the use of the frequency with which an item is mentioned in the interviews is a valid measure of its relative importance to other items. 1,269 relevant items were obtained from the interviews and categorized into 51 areas of concern (subject domains). The health professionals (physicians and nurses) were found to have mentioned seven subject domains more often than the patients, and no domain was mentioned more often by the patients than by the health professionals. The frequency that an item was mentioned was correlated (r = 0.55) with direct measures of its importance. These results imply that careful consultation with health professionals to determine the content of a decision analysis or informational materials that address treatment choice is sufficient to address patient concerns and that items can be chosen for inclusion based on the relative frequencies with which they are mentioned.

Atrial fibrillation and anticoagulation: from randomised trials to practice

Randomised trials confirm that anticoagulants reduce the risk of emboli in atrial fibrillation. To apply this evidence to practice, we developed an expression relating all relevant factors. Trial-based estimates of the risks of emboli and haemorrhage, and of the effects of anticoagulants on these risks were used to derive the extent to which haemorrhage has to be seen to be more detrimental than emboli to justify not using anticoagulants. Information from other studies was used to assess the risks for the types of patients not included in the trials. Haemorrhage needs to be assessed as being at least six times more detrimental than emboli to warrant withholding anticoagulants from patients like those in the trials. Only in patients with lone atrial fibrillation and in those with features suggesting a bleeding risk six times higher than the trials' average would a perception of equal detriment risk justify not giving anticoagulation.

The morbidity of abdominal hysterectomy

OBJECTIVE

To determine the morbidity of abdominal hysterectomy.

DESIGN

Descriptive. Physician billings to the Quebec Health Care Plan in the 1-month period after abdominal hysterectomy were examined.

SETTING

Operations performed in 102 hospitals in the Province of Quebec between Jan. 1, 1989, and Mar. 31, 1989, were selected.

PATIENTS

The study group included 3322 patients who had abdominal hysterectomy. Patients who had vaginal or abdominal hysterectomy for invasive cancer or pregnancy-related complications were excluded. A patient was considered to be morbid if the physician's intervention indicated concern for the patient's well-being.

MAIN OUTCOME MEASURES

Multiple logistic regression analysis to determine the adjusted rate ratio for inclusion in a categorical list of morbid patients among different subsets of surgeons, hospitals and patients.

RESULTS

Postoperative morbidity occurred in 646 patients (19.5%), who spent an average of 1.7 days longer in the hospital than patients with no postoperative morbidity. There were two postoperative deaths (0.1%). Forty-nine patients (1.5%) had postoperative surgical intervention. On 119 occasions (3.6%), patients were treated in the intensive care unit. A consultation was given by a medical specialist in 303 cases (9.1%). The rate ratio for postoperative morbidity was not significantly affected by years in practice or specialty of the surgeon but was increased for operations performed in mid-sized hospitals. The strongest predictor of postoperative morbidity was pre-existing medical disorder (RR).

CONCLUSION

The major causes of morbidity in patients who undergo abdominal hysterectomies are medical rather than surgical.

Factors relevant to preventing embolic stroke in patients with non-rheumatic atrial fibrillation

Whether or not to treat patients with non-rheumatic atrial fibrillation with anticoagulants to prevent embolic stroke is a dilemma for physicians. If randomized trials, currently underway, demonstrate a beneficial effect, the dilemma will not be solved because not all of the relevant factors can be addressed by trials. We used current knowledge about non-rheumatic atrial fibrillation and a method of obtaining patient-derived weights for avoiding stroke from eight medically trained subjects, to determine the overall benefit of anticoagulants and to see what factors were relevant and what effect each might have in deciding whether to use anticoagulant therapy. Using standard assumptions, anticoagulants gave an expected benefit for all subjects. The expected benefit (expressed in terms of lives per 1000 saved due to anticoagulants) varied between 5.4 and 46.7. This benefit remained for all subjects when we did a sensitivity analysis for different rates of stroke prevented by anticoagulants and different rates of intracranial hemorrhage caused by anticoagulants. When we used different baseline rates of stroke and different impacts of major hemorrhagic complications the benefit disappeared for 3 and 4 subjects respectively. We found the factors that were most crucial to the decision will not be included in randomized trials; the weight that an individual would place on avoiding embolic stroke vs the risk of intracranial bleeding from anticoagulant therapy; and the rate of embolic stroke that could be expected for the subject at risk. Factors which will be measured in randomized trials, will change results less substantially: the increased risk of major hemorrhages; the proportion of strokes that could be prevented by treatment; the increase in risk of intracranial hemorrhage. This method of analysis suggests that for most patients anticoagulants are beneficial and that the most important factor in determining this result is the value that subjects put on different outcomes.