Invasive Pneumococcal Disease and Long-Term Mortality Rates in Adults, Alberta, Canada

Kristen A. Versluys; Dean T. Eurich; Thomas J. Marrie; Gregory J. Tyrrell


Emerging Infectious Diseases. 2022;28(8):1615-1623. 

In This Article

Abstract and Introduction


The relationship between increased short-term mortality rates after invasive pneumococcal disease (IPD) has been frequently studied. However, the relationship between IPD and long-term mortality rates is unknown. IPD patients in Alberta, Canada, had clinical data collected that were linked to administrative databases. We used Cox proportional hazards modeling, and the primary outcome was time to all-cause deaths. First IPD events were identified in 4,522 patients, who had a median follow-up of 3.2 years (interquartile range 0.8–9.1 years). Overall all-cause mortality rates were consistently higher among cases than controls at 30 days (adjusted hazard ratio [aHR] 3.75, 95% CI 3.29–4.28), 30–90 days (aHR 1.56, 95% CI 1.27–1.93), and >90 days (aHR 1.43, 95% CI 1.33–1.54). IPD increases risk for short, intermediate, and long-term mortality rates regardless of age, sex, or concurrent conditions. These findings can help clinicians focus on postdischarge patient plans to limit long-term effects after acute IPD infection.


Despite introduction and recommendation of the capsular polysaccharide pneumococcal vaccine in Canada during 1989 to persons ≥65 years of age, Streptococcus pneumoniae is still a cause of major illness and death in Canada and worldwide.[1–3] The most serious manifestation of infection is invasive pneumococcal disease (IPD), which is characterized by bacteria invading normally sterile body sites, such as blood, lungs, or cerebrospinal fluid. In Canada, the incidence of IPD is around 8.8–9.9 cases/100,000 persons[4,5] and consistently highest in persons >60 years of age (27.1 cases/100,000 men and 20.2 cases/100,000 women).[6]

Short-term 30-day mortality rates after IPD have been frequently studied (estimated case-fatality rate within 30 days ranging from 13% to 21%).[7,8] Increasing age and concurrent conditions are associated with increased case-fatality rates.[7,8] However, studies on long-term mortality rates after IPD have been largely deficient, despite IPD being a reportable disease in Canada since 2000.[9]

Two studies conducted in Norway and the Netherlands investigated 1-year and 5-year mortality rates after IPD compared with those for age- and sex-matched controls in the general population.[7,10] In persons who survived initial hospitalization or survived 30 days after acute infection, IPD mortality rates were higher for cases than for controls (1-year mortality rate 10%–30% for cases vs. 1%–3% for controls; 5-year mortality rate 35%–42% for cases vs. 7%–15% for controls).[7,10] However, the study noted that, in the Netherlands, most deaths occurred within the first 30 days (case-mortality rate 17%).[7] Thus, it remains unclear whether IPD increases long-term mortality rates. Moreover, these were highly selected samples because both studies used data from only 1 hospital in a large urban center, which are unlikely to be representative of the broader IPD population.[7,10]

Widespread pneumococcal vaccination has seen major success.[11] However, with an aging population at risk for IPD, and pneumococcal serotypes changing to evade current vaccinations, IPD remains a disease of public health concern.[12] We have shown a change in serotypes and associated potential increases in severity of disease among IPD patients in Alberta, Canada.[13] To determine how IPD is affecting mortality rates, we investigated short, intermediate, and long-term mortality outcomes for persons who had IPD compared with age- and sex-matched controls in Alberta over a 20-year period.