Patients’ characteristics

Our sample included 151,440 patients who died of infectious diseases (mean age, 66.33 years; 62% male). The most common infectious diseases were pneumonia and influenza (n = 72,133), followed by parasitic diseases and other infectious diseases (n = 47,310), and septicemia (n = 31,119). (Table 1). Majority of the patients were white (n = 123,477) and had malignant tumors (n = 140,081). Approximately 78% of the patients were widowed. The average survival time was 65.31 months; moreover, pneumonia and influenza had the highest survival rate (Table 1). A significant difference was observed in the grade, laterality, race, behavior, and site between the different causes of infections. The most common type of cancer associated with infectious diseases was prostate cancer (n = 20,068), followed by breast cancer (n = 16,676), and Kaposi sarcoma (n = 13,046).

Table 1 Baseline characteristics of the included cancer patients who died due to infectious disease

Trend analysis of patients who died because of infections from 1973 to 2014

Based on the results of the join point analysis, from 1973 to 1984, there was an increase in the number of cases (APC = 5.4%, 95% CI = 3.9–6.8), followed by a significant increase in the rate of deaths from infectious diseases (APC = 10.4%, 95% CI = 8–12.8). However, from 1993 to 1998, a significant drop in the incidence of infectious diseases was observed (APC = 11.5%, 95% CI =  − 17.5 to − 5.7) (Fig. 1). In the next 3 years, a significant increase was observed in the number of deaths caused by cancer that overshot the previous years (APC = 26.74%, 95% CI = 3.7–54.9). From 2001 to 2012 and from 2012 to 2014, a significant decrease was noted in the number of cancer patients who died due to infectious diseases (Table 2). However, the overall trend from 1973 to 2014 showed an insignificant decrease (APC =  − 0.3, 95% CI =  − 2.2–1.7, P = 0.8).

Fig. 1

Trend analysis of cancer patients who died because of infectious diseases

Table 2 Annual percentage change of deaths caused by infectious diseases

Incidence of infectious diseases

The infectious diseases with the highest incidence were parasitic and other infectious diseases, including HIV (standardized incidence ratio [SIR] = 1.77, 95% CI = 1.69–1.84), followed by septicemia (SIR = 0.84, 95% CI = 0.81–0.88), tuberculosis (SIR = 0.72, 95% CI = 0.51–0.99), and pneumonia (SIR = 0.63, 95% CI = 0.61–0.64). Patients with blood vessel tumors including Kaposi Sarcoma had the highest incidence of parasitic disease and HIV infection (SIR = 88.83, 95% CI = 2.25–494.9) (Fig. 2).

Fig. 2

Incidence of reported infectious diseases in cancer patients based on its pathology

The highest SIR of tuberculosis was in complex epithelial neoplasms (SIR = 3.41, 95% CI = 0.09–19.01), followed by squamous cell neoplasms (SIR = 1.86, 95% CI = 0.96–3.25), and lymphoid leukemia (SIR = 1.42, 95% CI = 0.04–7.9) (Fig. 2).

For septicemia, patients with hematologic tumors other than leukemia, lymphoma, plasma cell tumors, and mast cell tumors had the highest incidence of septicemia, estimated to be 51.89% per 100 patients (SIR = 51.9, 95% CI = 1.31–289.16), followed by nerve sheath tumors (SIR = 3.91, 95% CI = 0.1–21.7), and mesothelial neoplasms (SIR = 3.21, 95% CI = 0.39–11.60). Meanwhile, for pneumonia and influenza, the mesonephromas had the highest incidence (SIR = 8.17, 95% CI = 0.21–45.6) (Fig. 2).

Survival analysis of infectious diseases

A significant difference was observed in survival between men and women (P < 0.0001), different organisms, race, and marital status (P < 0.0001). Furthermore, tumor characteristics, including behavior and grade, had significantly different survival according to its level (Fig. 3).

Fig. 3

Kaplan Meier survival analysis of infectious diseases

Based on the Cox regression analysis, old black men with intrahepatic tumor or acute leukemia of different grades, except the well-differentiated grade, had the highest risk of dying from infectious diseases (Additional file 3: Table 1). Basal cell neoplasms had the highest significant risk of mortality from infections (HR = 1.33, SE = 0.14, P = 0.04) (Additional file 3: Table 1).

Nomogram for predicting the 1-, 3-, and 5-year survival probability

A nomogram was constructed using significant variables in the Cox regression analysis. The C-index values for the nomogram were 0.85 (95% CI = 0.700–0.9) in the training dataset and 0.87 (95% CI = 0.7–0.9) in the validation set (Additional file 1: Fig. 1). The calibration plots revealed little or no difference between the nomogram prediction and actual observation for the 1-, 3-, and 5-survival years (Additional file 2: Fig. 2).

Identification of risk groups with prognostic median survival

The decision tree identified four risk groups for death from infectious diseases. The first group (blue in Fig. 4) included patients with pneumonia and influenza, septicemia, syphilis, and tuberculosis aged < 75.5 years, with a median survival of 2370 days. The second comprised patients aged > 75.5 years with a median survival of 1290 (red in Fig. 4). The third and fourth groups included cancer patients infected with parasitic, HIV, and other infectious diseases and were divided into subgroups based on marital status: married, separated, or widowed (median survival = 840 days) and single, unmarried, or domestic partner (median survival = 360 days) (Fig. 4).

Fig. 4

Survival decision tree identifying the four groups with their respective predicted survival. It achieved a brier score of 0.2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (