Mohammad Alì Mansournia

Effects of treatment or other exposure on outcome events are commonly measured by ratios of risks, rates, or odds. Adjusted versions of these measures are usually estimated by maximum likelihood regression (eg, logistic, Poisson, or Cox modelling). But resulting estimates of effect measures can have serious bias when the data lack adequate case numbers for some combination of exposure and outcome levels. This bias can occur even in quite large datasets and is hence often termed sparse data bias. The bias can arise or be worsened by regression adjustment for potentially confounding variables; in the extreme, the resulting estimates could be impossibly huge or even infinite values that are meaningless artefacts of data sparsity. Such estimate inflation might be obvious in light of background information, but is rarely noted let alone accounted for in research reports. We outline simple methods for detecting and dealing with the problem focusing especially on penalised estimation, which can be easily performed with common software packages.

OBJECTIVE: To report the global, regional, and national burden of chronic obstructive pulmonary disease (COPD) and its attributable risk factors between 1990 and 2019, by age, sex, and sociodemographic index. DESIGN: Systematic analysis. DATA SOURCE: Global Burden of Disease Study 2019. MAIN OUTCOME MEASURES: Data on the prevalence, deaths, and disability adjusted life years (DALYs) of COPD, and its attributable risk factors, were retrieved from the Global Burden of Disease 2019 project for 204 countries and territories, between 1990 and 2019. The counts and rates per 100 000 population, along with 95% uncertainty intervals, were presented for each estimate. RESULTS: In 2019, 212.3 million prevalent cases of COPD were reported globally, with COPD accounting for 3.3 million deaths and 74.4 million DALYs. The global age standardised point prevalence, death, and DALY rates for COPD were 2638.2 (95% uncertainty intervals 2492.2 to 2796.1), 42.5 (37.6 to 46.3), and 926.1 (848.8 to 997.7) per 100 000 population, which were 8.7%, 41.7%, and 39.8% lower than in 1990, respectively. In 2019, Denmark (4299.5), Myanmar (3963.7), and Belgium (3927.7) had the highest age standardised point prevalence of COPD. Egypt (62.0%), Georgia (54.9%), and Nicaragua (51.6%) showed the largest increases in age standardised point prevalence across the study period. In 2019, Nepal (182.5) and Japan (7.4) had the highest and lowest age standardised death rates per 100 000, respectively, and Nepal (3318.4) and Barbados (177.7) had the highest and lowest age standardised DALY rates per 100 000, respectively. In men, the global DALY rate of COPD increased up to age 85-89 years and then decreased with advancing age, whereas for women the rate increased up to the oldest age group (≥95 years). Regionally, an overall reversed V shaped association was found between sociodemographic index and the age standardised DALY rate of COPD. Factors contributing most to the DALYs rates for COPD were smoking (46.0%), pollution from ambient particulate matter (20.7%), and occupational exposure to particulate matter, gases, and fumes (15.6%). CONCLUSIONS: Despite the decreasing burden of COPD, this disease remains a major public health problem, especially in countries with a low sociodemographic index. Preventive programmes should focus on smoking cessation, improving air quality, and reducing occupational exposures to further reduce the burden of COPD.

OBJECTIVE: To use data from the Global Burden of Disease Study between 1990 and 2017 to report the rates and trends of point prevalence, annual incidence, and years lived with disability for neck pain in the general population of 195 countries. DESIGN: Systematic analysis. DATA SOURCE: Global Burden of Diseases, Injuries, and Risk Factors Study 2017. MAIN OUTCOME MEASURES: Numbers and age standardised rates per 100 000 population of neck pain point prevalence, annual incidence, and years lived with disability were compared across regions and countries by age, sex, and sociodemographic index. Estimates were reported with uncertainty intervals. RESULTS: Globally in 2017 the age standardised rates for point prevalence of neck pain per 100 000 population was 3551.1 (95% uncertainty interval 3139.5 to 3977.9), for incidence of neck pain per 100 000 population was 806.6 (713.7 to 912.5), and for years lived with disability from neck pain per 100 000 population was 352.0 (245.6 to 493.3). These estimates did not change significantly between 1990 and 2017. The global point prevalence of neck pain in 2017 was higher in females compared with males, although this was not significant at the 0.05 level. Prevalence increased with age up to 70-74 years and then decreased. Norway (6151.2 (95% uncertainty interval 5382.3 to 6959.8)), Finland (5750.3 (5058.4 to 6518.3)), and Denmark (5316 (4674 to 6030.1)) had the three highest age standardised point prevalence estimates in 2017. The largest increases in age standardised point prevalence estimates from 1990 to 2017 were in the United Kingdom (14.6% (10.6% to 18.8%)), Sweden (10.4% (6.0% to 15.4%)), and Kuwait (2.6% (2.0% to 3.2%)). In general, positive associations, but with fluctuations, were found between age standardised years lived with disability for neck pain and sociodemographic index at the global level and for all Global Burden of Disease regions, suggesting the burden is higher at higher sociodemographic indices. CONCLUSIONS: Neck pain is a serious public health problem in the general population, with the highest burden in Norway, Finland, and Denmark. Increasing population awareness about risk factors and preventive strategies for neck pain is warranted to reduce the future burden of this condition.