Human brucellosis: Widespread information deficiency hinders an understanding of global disease frequency

Abstract

Background: For decades, human brucellosis has been recognized worldwide as a significant cause of morbidity, yet the annual incidence of this disease remains unknown. We analyzed this frequency, using international reports (2005-2019), identifying information gaps, and distinguishing a possible path forward.

Methodology/principal findings: A novel approach to estimating the incidence of this disease was explored. We utilized annual health data extracted from the World Organization for Animal Health (OIE)-World Animal Health Information System (WAHIS) database, assessing the dataset completeness and representativeness of the data for the world population. Additionally, we assessed the reported country level human brucellosis case counts and the factors that influenced the observed changes over time. Our analysis revealed incomplete and unrepresentative information, preventing the estimation of annual human brucellosis case incidence at the global level. In the OIE-WAHIS database, only 48.4% of the required reports have been submitted as of 2019, with approximately 47.3% of the world population represented. Additionally, geographic regions were disproportionate in completeness, representativeness, and actual reported case counts. Africa and Asia constituted the majority of reported cases, while simultaneously submitting the lowest percentage of reports as well as covering the lowest percentage of their populations within those reports, when compared to the rest of the world.

Conclusions/significance: The global annual frequency of human brucellosis cases remains elusive. Furthermore, there exists great heterogeneity in diagnostic, surveillance, and reporting systems worldwide, calling into question the validity of available information. This study reveals that the Neglected Zoonotic Disease priority status for brucellosis should be restored.

Fig 1. Percentage of OIE annual reports providing case counts of human brucellosis (informed reports), by region (2005–2019).

A: Percentage of reports submitted to the OIE classified as informative, separated by region and compared to the entire world (2005–2019). Each point on the plot denotes the percentage of reports within a region that included a case count of human brucellosis during a single year. Bars signify mean and standard deviation. Statistical significance compared to the worldwide mean are denoted by asterisks (* through **** → low through high significance). p-value for Africa is < 0.0001, the Americas = 0.0009, Asia = 0.0363, Europe < 0.0001, and Oceania = 0.6566. B: Proportion of reports submitted to the OIE classified as informed, worldwide and separated by region, plotted over time (2005–2019). Linear regression indicates that the Asian region has the only significant change with an increase (p = 0.038). Worldwide, African, American, European, and Oceanian regression lines are all stable with slopes not significantly diverting from zero (p = 0.26; 0.086; 0.47; 0.53; 0.20 respectively).

Fig 2. Percentage of populations represented by OIE-WAHIS reports on human brucellosis (informative reports), by region (2005–2019).

A: Proportion of each region’s population that are represented within the informed reports, compared to the world as a whole (2005–2019). Each point on the plot denotes the percentage of a regional population that is represented during a single year. Bars signify mean and standard deviation, and level of significance when deviating from the worldwide mean are characterized by asterisks (* through **** → low through high significance). p-value for Africa < 0.0001, the Americas < 0.0001, Asia = 0.0031, Europe < 0.0001, and Oceania < 0.0001. B: Proportion of each region’s population that are represented within informed reports, plotted over time (2005–2019). Linear regression indicates that there was no significant change in representation over time whether worldwide (p = 0.22), Africa (p = 0.26), Asia (p = 0.12), the Americas (p = 0.32), Europe (p = 0.25), or Oceania (p = 0.15).

Fig 3. Case counts of human brucellosis reported to the OIE, by region (2005–2019).

A: Number of reported case counts (RCC) worldwide and by region (2005–2019). Each point on the scatter plot denotes the actual number of human brucellosis cases reported during a single year. Bars signify mean and standard deviation. B: Number of reported case counts (RCC) worldwide and by region plotted over time (2005–2019). Linear regression indicates that significant change only occurred in the reported case counts worldwide (p = 0.0004), in Africa (p < 0.0001), and in Oceania (p < 0.0001). The solid line arrow signifies the year Kenya began routinely submitting informative reports. The dashed line arrow signifies the year that China discontinued reporting.

Fig 4. Association between reporting and case counts, globally (2005–2019).

Reported case counts worldwide are plotted in comparison with the number of countries submitting informed reports. Arrows indicate progression of time. The plot was then analyzed for correlation (r = 0.49) and fitted with a line of best fit which isn’t statistically significant (p = 0.061). This indicates the number of countries submitting informed reports doesn’t significantly influence the total number of reported cases worldwide.

Fig 5. Influence of Kenya on the global and regional reported case counts of human brucellosis (2005–2019).

A: Reported case counts compared to the remainder of Africa (2005–2019). B: Reported case counts compared to the remainder of the world (2005–2019). C: Population compared to the remainder of Africa (2005–2019). D: Population compared to the remainder of the world (2005–2019).