Implementing low-dose computed tomography screening for lung cancer in Canada: implications of alternative at-risk populations, screening frequency, and duration.

BACKGROUND: Low-dose computed tomography (ldct) screening has been shown to reduce mortality from lung cancer; however, the optimal screening duration and "at risk" population are not known. METHODS: The Cancer Risk Management Model developed by Statistics Canada for the Canadian Partnership Against Cancer includes a lung screening module based on data from the U.S. National Lung Screening Trial (nlst). The base-case scenario reproduces nlst outcomes with high fidelity. The impact in Canada of annual screening on the number of incident cases and life-years gained, with a wider range of age and smoking history eligibility criteria and varied participation rates, was modelled to show the magnitude of clinical benefit nationally and by province. Life-years gained, costs (discounted and undiscounted), and resource requirements were also estimated. RESULTS: In 2014, 1.4 million Canadians were eligible for screening according to nlst criteria. Over 10 years, screening would detect 12,500 more lung cancers than the expected 268,300 and would gain 9200 life-years. The computed tomography imaging requirement of 24,000-30,000 at program initiation would rise to between 87,000 and 113,000 by the 5th year of an annual nlst-like screening program. Costs would increase from approximately $75 million to $128 million at 10 years, and the cumulative cost nationally over 10 years would approach $1 billion, partially offset by a reduction in the costs of managing advanced lung cancer. CONCLUSIONS: Modelling various ways in which ldct might be implemented provides decision-makers with estimates of the effect on clinical benefit and on resource needs that clinical trial results are unable to provide.

Using the Cancer Risk Management Model to evaluate the health and economic impacts of cytology compared with human papillomavirus DNA testing for primary cervical cancer screening in Canada.

BACKGROUND: In Canada, discussion about changing from cytology to human papillomavirus (hpv) dna testing for primary screening in cervical cancer is ongoing. However, the Canadian Task Force on Preventive Health Care has not yet made a recommendation, concluding that the evidence is insufficient. METHODS: We used the cervical cancer and hpv transmission models of the Cancer Risk Management Model to study the health and economic outcomes of primary cytology compared with hpv dna testing in 14 screening scenarios with varying screening modalities and intervals. Projected cervical cancer cases, deaths, colposcopies, screens, costs, and incremental cost-effectiveness were evaluated. We performed sensitivity analyses for hpv dna test costs. RESULTS: Compared with triennial cytology from age 25, 5-yearly hpv dna screening alone from age 30 resulted in equivalent incident cases and deaths, but 55% (82,000) fewer colposcopies and 43% (1,195,000) fewer screens. At hpv dna screening intervals of 3 years, whether alone or in an age-based sequence with cytology, screening costs are greater, but at intervals of more than 5 years, they are lower. Scenarios on the cost-effectiveness frontier were hpv dna testing alone every 10, 7.5, 5, or 3 years, and triennial cytology starting at age 21 or 25 when combined with hpv dna testing every 3 years. CONCLUSIONS: Changing from cytology to hpv dna testing as the primary screening test for cervical cancer would be an acceptable strategy in Canada with respect to incidence, mortality, screening and diagnostic test volumes.

First do no harm: extending the debate on the provision of preventive tamoxifen.

The Breast Cancer Prevention Trial (BCPT-P-1) demonstrated that tamoxifen could reduce the risk of invasive breast cancer in high-risk women by 49%, but that it could also increase the risk of endometrial cancer, vascular events and cataracts. This paper provides an estimate of the net health impacts of tamoxifen administration on high-risk Canadian women with no prior history of breast cancer. The results of the BCPT-P-1 were incorporated into the breast cancer and other modules of Statistics Canada's microsimulation POpulation HEalth Model (POHEM). While the main intervention scenario conformed as closely as possible to the eligibility criteria for tamoxifen in the BCPT-P-1 protocol, 3 additional scenarios were simulated. Predicted absolute risks of breast cancer at 5 years of 1.66%, 3.32% and 4.15% were calculated for women 35 to 70 years of age. When the BCPT-P-1 results were incorporated into the simulation model, the analysis suggests no increase in life expectancy in this risk group. Tamoxifen appeared to be beneficial for women with a 5-year predicted risk of 3.32% or greater. The results of these simulations are particularly sensitive to the reduction in mortality observed in the BCPT-P-1, as well as being sensitive to other characteristics of the simulation model. Overall, the analysis raises questions about the use of tamoxifen in otherwise healthy women at high risk of breast cancer.

Relation between income inequality and mortality in Canada and in the United States: cross sectional assessment using census data and vital statistics.

OBJECTIVE: To compare the relation between mortality and income inequality in Canada with that in the United States. DESIGN: The degree of income inequality, defined as the percentage of total household income received by the less well off 50% of households, was calculated and these measures were examined in relation to all cause mortality, grouped by and adjusted for age. SETTING: The 10 Canadian provinces, the 50 US states, and 53 Canadian and 282 US metropolitan areas. RESULTS: Canadian provinces and metropolitan areas generally had both lower income inequality and lower mortality than US states and metropolitan areas. In age grouped regression models that combined Canadian and US metropolitan areas, income inequality was a significant explanatory variable for all age groupings except for elderly people. The effect was largest for working age populations, in which a hypothetical 1% increase in the share of income to the poorer half of households would reduce mortality by 21 deaths per 100 000. Within Canada, however, income inequality was not significantly associated with mortality. CONCLUSIONS: Canada seems to counter the increasingly noted association at the societal level between income inequality and mortality. The lack of a significant association between income inequality and mortality in Canada may indicate that the effects of income inequality on health are not automatic and may be blunted by the different ways in which social and economic resources are distributed in Canada and in the United States.

Relation between income inequality and mortality: empirical demonstration.

Objective To assess the extent to which observed associations between income inequality and mortality at population level are statistical artifacts. Design Indirect "what if" simulation using observed risks of mortality at individual level as a function of income to construct hypothetical state-level mortality specific for age and sex as if the statistical artifact argument were 100% correct. Method Data from the 1990 census for the 50 US states plus Washington, DC, were used for population distributions by age, sex, state, and income range; data disaggregated by age, sex, and state from the Centers for Disease Control and Prevention were used for mortality; and regressions from the national longitudinal mortality study were used for the individual-level relation between income and risk of mortality. Results Hypothetical mortality, although correlated with inequality (as implied by the logic of the statistical artifact argument), showed a weaker association with the level of income inequality in each state than the observed mortality. Conclusions The observed associations in the United States at the state level between income inequality and mortality cannot be entirely or substantially explained as statistical artifacts of an underlying individual-level relation between income and mortality. There remains an important association between income inequality and mortality at state level above anything that could be accounted for by any statistical artifact. This result reinforces the need to consider a broad range of factors, including the social milieu, as fundamental determinants of health.

The economics of lung cancer management in Canada.

Because lung cancer is a major health care problem in Canada, it is imperative to understand how resources are used to diagnose and treat this disease. This paper describes a method of modelling the direct patient care costs for lung cancer from the perspective of the government as payer in a universal health care system. Clinical algorithms were developed to describe the management of non-small cell (NSCLC) and small cell (SCLC) lung cancer. Patients were allocated to the treatment algorithms in the model based on a knowledge of their distribution by cell type and stage in Canadian cases. A microsimulation model developed by Statistics Canada was used to integrate the data on type of lung cancer, extent of disease, clinical management, survival and health care resource utilization. The direct care costs for diagnosis and treatment of NSCLC ranged from $Cdn 17 889 for the surgery/post-operative radiotherapy treatment of Stages I and II to $Cdn 6333 for supportive care for patients with Stage IV disease. The costs of determining relapse for NSCLC were estimated to be $Cdn 1528 and terminal care costs, made up largely of hospitalization charges and some palliative radiotherapy, were $Cdn 10 331. Direct care costs for the diagnosis and initial treatment of SCLC ranged from $Cdn 18 691 for management of limited stage disease to $Cdn 4739 for the supportive care of patients with extensive disease. The cost of determining relapse for SCLC was estimated to be $Cdn 1590 and terminal care costs averaged $Cdn 9966. For all 15 624 cases of lung cancer diagnosed in Canada in 1988, it was estimated that the total cost of providing treatment and follow-up, and managing relapse over 5 years was $Cdn 328 million. Despite the large total cost of lung cancer management, estimates of cost effectiveness of therapy showed that the cost per life year gained was approximately $Cdn 11 000 for NSCLC and $Cdn 19 560 for SCLC. These estimates of the direct health care costs assume that all patients have access to care, treatment is uncomplicated and practice is standard, and must be viewed as an idealized assessment of the cost of lung cancer management. The microsimulation model, however, does provide a useful framework for evaluating the costs of new diagnostic procedures, treatment strategies and new drugs.

Health-adjusted life expectancy.

In 1991, the National Task Force on Health Information recommended that in order to assess the health of Canadians, the health information system should include an aggregate index of population health. This article presents such an index--Health-Adjusted Life Expectancy (HALE)--as one possibility in a range of indicators. In contrast to conventional life expectancy, which considers all years as equal, to calculate HALE, years of life are weighted by health status. To measure health status, the Health Utility Index, obtained from 1994-95 National Population Health Survey data, was used. Traditional life expectancy and HALE figures are compared to estimate the burden of ill health. The societal burden of ill health is higher for women than for men, and is highest among those in "early" old age, not among the most elderly. The data further indicate that sensory problems and pain comprise the largest components of the burden of ill health, and that higher socioeconomic status confers a dual advantage--longer life expectancy and a lower burden of ill health.

Estimating the cost of lung cancer diagnosis and treatment in Canada: the POHEM model.

Because lung cancer is a major health care problem in Canada, it would be useful to identify the direct health care costs of diagnosing and treating this disease and to create an analytic framework within which diagnostic and therapeutic options can be assessed. This paper describes a method of modelling the costs of care for lung cancer. The perspective of the costing model is that of the government as payer in a universal health care system. Clinical algorithms were developed to describe the management of non-small cell (NSCLC) and small cell (SCLC) lung cancer. Patients were allocated to the treatment algorithms in the model, based on a knowledge of the stage distribution of cases within provincial cancer registries and an estimate of the use of therapeutic modalities, according to lung cancer experts. A microsimulation model (POHEM) developed at Statistics Canada was used to integrate data on risk factors, disease onset and progression, health care resource utilization and direct medical care costs. The model incorporates survival data on patients, according to cell type and stage, based on published studies. Relapse and terminal care costs were assigned during the year of death, in order to determine the cost of continuing care and the cumulative cost of lung cancer management over time. Patients surviving five years were assumed to be cured. The model estimates that the total five year cost to provide care to the 15,624 cases of lung cancer diagnosed in Canada in 1988 was in excess of $328 million. Over 82% of this total was spent in the first year for diagnostic tests, therapy (surgery, chemotherapy, radiation therapy, or combinations of these), hospitalization and follow-up costs. The average five year cost per case was $21,000, and ranged from a high of $29,860 for limited disease SCLC, to a low of $16,500 for Stage IV NSCLC. The actual cost of providing care, including the management of complications, is unknown and our estimates should be regarded as an idealized estimate of the cost of lung cancer management. However, the POHEM model has a level of sophistication which, we believe, reasonably reflects the cost per case and total costs of treating lung cancer by stage and therapeutic modality in Canada.

Diagnostic and therapeutic approaches to lung cancer in Canada and their costs.

Escalating health care costs have made it imperative to evaluate the resources required to diagnose and treat major illnesses in Canadians. For Canadian men, lung cancer is not only the most common malignancy, but also the major cancer killer. As of 1994, lung cancer is expected to overtake breast cancer as the leading cause of cancer deaths in women. This paper presents a detailed description of the methodology used to determine the direct health care costs associated with 'standard' diagnostic and therapeutic approaches for lung cancer in Canada in 1988. Clinical algorithms were developed for each stage of non-small-cell lung cancer (NSCLC) and small-cell lung cancer (SCLC). The algorithms were designed to take the form of decision trees for each clinical stage of lung cancer. The proportion of patients assigned to each branch was based upon questionnaire responses obtained from thoracic surgeons and radiation oncologists when presented with clinical scenarios, and information from provincial cancer registries. Direct care costs were derived primarily from one provincial fee schedule (Ontario), and costing information obtained during the conduct of several Canadian clinical trials in lung cancer. Direct costs for diagnosis and initial treatment of NSCLC (excluding relapse and terminal care costs) ranged from $17,889 for the surgery/post-operative radiotherapy arm of stages I and II to $6,333 for the supportive care arm (stage IV). The cost of determining relapse for NSCLC was estimated to be $1,528, and terminal care costs, which included palliative radiotherapy and hospitalisation, were $10,331. Direct costs for diagnosis and initial treatment of SCLC ranged from $18,691 for limited stage disease to $4,739 for the supportive care arm of extensive disease. The cost of diagnosing relapse for SCLC was estimated to be $1,590, and terminal care costs averaged $9,966. This report provides an estimate of the Canadian costs of managing lung cancer by stage and treatment modality. Because the actual costs of all components of care are not available from any combination of sources, these cost estimates must be viewed as an idealised estimate of the cost of lung cancer management. However, we believe that the lung cancer costing model that we have developed provides a level of sophistication which gives a reasonable estimate of the cost per case of treating NSCLC and SCLC.

The cost of managing lung cancer in Canada.

The POpulation HEalth Model (POHEM) lung cancer microsimulation model has provided a useful framework for calculating the cost of managing individual cases of lung cancer in Canada by stage, cell type, and treatment modality, as well as the total economic burden of managing all cases of lung cancer diagnosed in Canada. These data allow an estimation of the overall cost effectiveness of lung cancer therapy. the model also provides a frame-work for evaluating the cost effectiveness of new therapeutic strategies, such as combined modality therapy for stage III disease or new chemotherapy drugs for stage IV disease. By expressing the cost of lung cancer treatment as cost of life-years gained, such analyses allows useful comparisons of the cost effectiveness of these treatments with those of other costly but accepted medical therapies.

POHEM--a framework for understanding and modelling the health of human populations.

A variety of developments have come together to serve as both an impetus to and foundation for the development of a new POpulation HEalth Model (POHEM) at Statistics Canada. Part of the impetus is statistical and derives from weaknesses in Canada's health statistics programme--particularly the lack of balance between information on health outcomes and health care resource consumption, and the absence of a coherent statistical structure. The other major impetus is the need for rational processes for managing and allocating resources to improve the health of Canadians. The foundation for the development of this model has come from the revolution in computing. Dramatic improvements have opened up new methodological opportunities, particularly sophisticated simulation modelling and detailed analyses of large volumes of microdata. POHEM is designed to build on these increasingly powerful methods in order to meet health statistical and policy needs. At this time, POHEM is like a partially-completed building. This article reviews its motivation, the overall architectural plan, and the portion of the structure already completed. A major portion of POHEM is devoted to the explicit modelling of chronic disease processes, using monte carlo microsimulation methods. The article concludes with illustrations of a few recent applications, focusing on the joint patterns of smoking, cholesterol and heart disease, osteoarthritis and lung cancer morbidity. While POHEM has been developed in a Canadian context, work is under way to create a version that can be used in other countries.

Modelling the population health impact of musculoskeletal diseases: arthritis.

OBJECTIVE: A model adjusting for reductions in quality and quantity of life was developed to estimate the population health impact of musculoskeletal diseases. METHODS: Using arthritis as the prototype, prevalence, mortality, and severity data from a variety of sources were combined to model a hypothetical cohort of 1,000 individuals through life. Quality adjusted life years and population health expectancy were calculated for those with arthritis and compared to the general population. RESULTS: Without adjusting for quality of life, a cohort of 1,000 women and 1,000 men at age 15 years could expect 65,010 and 58,735 life years, respectively. Adjusting for quality of life, women with arthritis could expect 61,719 life years, and men 57,123 life years. The unadjusted population health expectancy was 65.0 for girls and 58.7 for boys (at age 15 years). Adjusting for quality of life, the population health expectancies were 61.7 and 57.1, years for women and men, respectively. CONCLUSION: Using this model, the typical adult woman with arthritis can expect to lose 3.3 healthy years of life, and a man, 1.6 healthy years of life. Overall, the model provided a general methodology for determining the population health impact of musculoskeletal diseases. In addition, it is hoped that the methodology will stimulate further research into this area, raise awareness about the uses and limitations of currently available data, and provide a useful model for monitoring the impact of interventions.

A template for health information.

The Template for Health Information is a software package for microcomputers developed by the National Task Force on Health Information of Statistics Canada. The Template software provides a conceptual framework for the collection, analysis and utilization of health information for policy planning and management uses. This framework involves interactions between a population of individuals, their external environment, and sets of potential health-affecting interventions. The Template is seen as a pedagogical device for incorporating the many perspectives of different health constituencies. A floppy diskette is provided with the article to demonstrate the potential of the Template.

When the baby boom grows old: impacts on Canada's public sector.

The authors discuss the dependency burden that is expected to result from demographic aging in Canada. "The estimated size of the burden depends on projections of demographic change, economic growth, and structural aspects of the major age-sensitive public-sector programmes. The burdens are analysed for 2016 and 2036, the period when demographic aging may be expected to have its most adverse impacts on old-age dependency ratios and public-sector programme costs. Contrary to many popularly expressed concerns, demographic aging is not the most important factor in determining future public-sector costs and revenues. Rather, aspects of the design and management of public-sector programmes represent the greatest area of uncertainty."