National health system characteristics, breast cancer stage at diagnosis, and breast cancer mortality: a population-based analysis

Background In some countries, breast cancer age-standardised mortality rates have decreased by 2-4% per year since the 1990s, but others have yet to achieve this outcome. In this study, we aimed to characterise the associations between national health system characteristics and breast cancer age-standardised mortality rate, and the degree of breast cancer downstaging correlating with national age-standardised mortality rate reductions. Methods In this population-based study, national age-standardised mortality rate estimates for women aged 69 years or younger obtained from GLOBOCAN 2020 were correlated with a broad panel of standardised national health system data as reported in the WHO Cancer Country Profiles 2020. These health system characteristics include health expenditure, the Universal Health Coverage Service Coverage Index (UHC Index), dedicated funding for early detection programmes, breast cancer early detection guidelines, referral systems, cancer plans, number of dedicated public and private cancer centres per 10 000 patients with cancer, and pathology services. We tested for differences between continuous variables using the non-parametric Kruskal-Wallis test, and for categorical variables using the Pearson chi(2) test. Simple and multiple linear regression analyses were fitted to identify associations between health system characteristics and age-standardised breast cancer mortality rates. Data on TNM stage at diagnosis were obtained from national or subnational cancer registries, supplemented by a literature review of PubMed from 2010 to 2020. Mortality trends from 1950 to 2016 were assessed using the WHO Cancer Mortality Database. The threshold for significance was set at a p value of 0.05 or less. Findings 148 countries had complete health system data. The following variables were significantly higher in high income countries than in low-income countries in unadjusted analyses: health expenditure (p=0.0002), UHC Index (p<0.0001), dedicated funding for early detection programmes (p=0.0020), breast cancer early detection guidelines (p<0.0001), breast cancer referral systems (p=0.0030), national cancer plans (p=0.014), cervical cancer early detection programmes (p=0.0010), number of dedicated public (p<0.0001) and private (p=0.027) cancer centres per 10 000 patients with cancer, and pathology services (p<0.0001). In adjusted multivariable regression analyses in 141 countries, two health system characteristics were significantly associated with lower age-standardised mortality rates: higher UHC Index levels (beta=-0.12, 95% CI -0.16 to -0.08) and increasing numbers of public cancer centres (beta=-0.23, -0.36 to -0.10). These findings indicate that each unit increase in the UHC Index was associated with a 0.12-unit decline in age-standardised mortality rates, and each additional public cancer centre per 10 000 patients with cancer was associated with a 0.23-unit decline in age-standardised mortality rate. Among 35 countries with available breast cancer TNM staging data, all 20 that achieved sustained mean reductions in age-standardised mortality rate of 2% or more per year for at least 3 consecutive years since 1990 had at least 60% of patients with invasive breast cancer presenting as stage I or II disease. Some countries achieved this reduction without most women having access to population-based mammographic screening. Interpretation Countries with low breast cancer mortality rates are characterised by increased levels of coverage of essential health services and higher numbers of public cancer centres. Among countries achieving sustained mortality reductions, the majority of breast cancers are diagnosed at an early stage, reinforcing the value of clinical early diagnosis programmes for improving breast cancer outcomes. Copyright (C) 2021 World Health Organization. Published by Elsevier Ltd. All rights reserved.