Abstract
Background
Children are a particularly vulnerable population for tuberculosis (TB). The present study aims to describe epidemiological characteristics and temporal trends of paediatric TB in Spain.
Methodology
A retrospective study of paediatric TB in Spain was conducted using data from the National Epidemiological Surveillance Network and the l Basic Minimum Data Set of Hospitalizations from 2012 to 2023. Trends of annual notification rates (NR), hospitalization rates (HR) and annual percentage of change (APC) were calculated.
Results
Between 2012 and 2023 there were 3,472 reported cases and 2,768 hospitalizations. The mean NR and HR of the period were 4.2 and 3.3 per 100,000 inhabitants, respectively. The NR and the HR decreased from 5.0 to 3.5 and from 5.0 to 2.8, with significant APCs of −5.05 and −7.25, respectively. The most affected age group was 1–4 (39.8% of notifications and 43.4% of hospitalizations). Severe presentations (miliary and meningitis) were more common amongst cases less than 1-year-old.
Conclusions
A decreasing trend in paediatric TB in Spain has been observed since 2012, with a slight rebound since 2021. Younger children aged 1–4 had the highest incidence while children under 1 year had the highest percentage of severe TB. This highlights the importance of prioritizing and strengthening control strategies in this age group.
Keywords: Epidemiology, paediatric tuberculosis, Spain
Introduction
Paediatric tuberculosis (TB) has long remained a silent global epidemic, largely due to the challenges in quantifying its incidence and impact. According to the World Health Organization (WHO), in 2023 around 1.3 million children and adolescents under 15 years of age developed TB (known as paediatric TB), equivalent to about 12% of the total cases of TB in the world [1]. Since 2012, it has been included in various global TB control strategies [2].
A characteristic feature of paediatric TB surveillance is that children rarely transmit the disease, and cases of paediatric TB are considered a sentinel event reflecting recent transmission from an adult contact [3]. Despite this, children are particularly vulnerable in terms of disease prognosis and severity as they are at higher risk of rapid progression from tuberculous infection (TBI) to active TB than adults [4]. Children develop disseminated extrapulmonary forms more frequently than adults [5,6]. Under 5 years of age they are at even greater risk of progressing from TBI to severe forms such as miliary TB and tuberculous meningitis [4,7], with neonatal TB (congenital and postnatal) presenting the highest risk of disseminated forms and mortality [8]. Although deaths amongst children with TB are rare, extrapulmonary forms can lead to significant complications and other sequelae [9,10].
The symptoms of paediatric TB are non-specific and varied [4]. Confirmatory diagnosis is based on culture or PCR for Mycobacterium tuberculosis, but the low bacterial load in children reduces the sensitivity of these tests [9,11]. In addition, obtaining respiratory samples is difficult [12]. The treatment of paediatric TB has long been difficult due to the toxicity of the drugs and the lack of suitable formulations for children [11]. In 2022, updated WHO guidelines recommended shorter regimens for non-severe cases, a new treatment option for tuberculous meningitis, and the use of bedaquiline and delamanid in children with rifampicin-resistant (RR-TB) or multidrug resistant TB (MDR-TB) [13].
A recent increase in the incidence rate of paediatric TB has been observed across the European Union (EU). According to the latest Tuberculosis Surveillance and Monitoring Report published by the European Centre for Disease Prevention and Control (ECDC), paediatric TB cases increased by 10% in 2023 compared to the previous year, the third consecutive year where an increase has been observed [14]. The objective of this study was to describe reported cases and hospitalizations for TB amongst children aged 0 to 14 years in Spain during the period 2012–2023. Characterizing the epidemiology of this disease will support the development of public health strategies in order to improve the prevention and control of paediatric TB in Spain.
Methods
Study design and target population
We conducted a retrospective descriptive study of paediatric TB cases reported to the National Epidemiological Surveillance Network (RENAVE) and paediatric TB hospitalizations extracted from the National Basic Minimum Data Set of Hospitalizations (CMBD) in children aged 0 to 14 years in Spain from 2012 to 2023. As RENAVE and CMBD are independent systems and do not allow records to be linked at an individual level, the analyses of both sources were carried out separately. Consequently, the notification and hospitalization rates are interpreted independently and are not combined, thereby avoiding the risk of double counting.
Data sources
National epidemiological surveillance network (RENAVE)
RENAVE is the surveillance network that coordinates the detection and control of communicable diseases in Spain. It was established in 1995 with the aim of collecting and analysing data on communicable diseases and other events that may pose a public health threat [15]. TB data are reported to RENAVE on a weekly basis by Spain’s 17 autonomous communities (CCAA) and the two autonomous cities of Ceuta and Melilla through the computer platform via Web SiViES plus managed by the National Epidemiology Centre (CNE). For the classification of TB localization, the RENAVE protocol TB definition was used [16].
Sociodemographic, epidemiological and clinical variables were analyzed. ‘Time to diagnosis’ was created from the difference in the number of days between the date of symptom onset and the date of laboratory diagnosis, and ‘Time to initiation of treatment’ was computed from the difference between the date of symptom onset and the date of treatment initiation. Based on this classification, two binary variables were defined: ‘delayed diagnosis’ and ‘delayed treatment initiation’, establishing two groups according to the median number of days. For the first variable, a cut-off point of ≤24 days and >24 days was used; and for the second, a threshold of ≤19 days and >19 days. For the TB localization variable, extrapulmonary forms were grouped into six categories: ‘severe TB’ (miliary and meningitis), ‘lymph node TB’, ‘genitourinary TB’, ‘osteoarticular TB’, ‘abdominal TB’ and ‘other locations’. Both miliary tuberculosis and tuberculous meningitis are considered to be the most severe forms within the clinical spectrum of the disease, given their disseminated nature, the involvement of vital organs, and the need for urgent and prolonged treatment, as outlined in international guidelines and specialist reviews [17–19].
For antimicrobial susceptibility test results the most recent WHO and ECDC definitions were used for RR-TB, MDR-TB and extremely resistant TB (XDR-TB) [1,14]. The variable ‘treatment result’ was classified into three categories: ‘satisfactory’ if the result was cured or complete, ‘death’, and ‘other’ for all remaining outcomes (abandonment, transfer, loss to follow up, prolongation of treatment, and others).
Hospitalization minimum basic data set (CMBD and RAE-CMBD)
The CMBD was implemented in 1997 and includes a set of administrative and clinical data obtained from discharge reports of patients treated in hospitals of the National Health System. In 2016, a new dataset called RAE-CMBD was implemented, extending the registration to other alternative areas to hospitalization [20], with a transition in the disease coding system from the International Classification of Diseases in its ninth revision (ICD-9-MC), used until December 31, 2015, to the tenth revision (ICD-10-ES), implemented from January 1, 2016. This process involved internal harmonization and validation typically associated with the implementation of the new system. Furthermore, as the analysis covered a period during which different recording models and coding systems coexisted, it was necessary to standardize the data to ensure the consistency and comparability of the time series.
We analyzed all hospitalizations with a diagnosis of TB as the primary diagnosis. Only the patient’s first admission was analyzed, with any subsequent admissions of the same patient excluded. Sociodemographic and clinical variables were analyzed. A binary variable was created for the days of hospitalization, classifying the length of stay as less than or equal to the median, or greater than the median. As there are no clinical thresholds or guideline-based recommendations for defining prolonged hospital stays, this variable was classified into two categories using the median as the cut-off point, following an exploratory approach. The type of discharge was reclassified into 3 categories: ‘domicile’, ‘death’ and ‘others’, that was the result of the grouping of ‘voluntary discharge’, ‘unknown’, ‘transfer to socio-health centre’, ‘transfer to another hospital’ and ‘others’.
For the calculation of annual notification rate (NR) and hospitalization rate (HR), the census estimates provided by the National Institute of Statistics on 1 January of each corresponding year were used as population denominators [21].
Data analysis
For quantitative variables, mean and standard deviation (SD) or median and interquartile range (IQR) were calculated, depending on the normality of the distribution. For categorical variables, frequencies and percentages were calculated. Age was classified into 4 groups: <1 year, 1–4 years, 5–9 years and 10–14 years. Comparisons between variables were performed using the Pearson chi-square test or Fisher’s test and the Student’s t test or the non-parametric Mann–Whitney U test, as appropriate. Results were stratified by age group. Statistical significance was defined as p value <0.05.
To examine trends in reporting and hospitalization rates across the study period, a regression analysis was performed using the Joinpoint software version 5.0.2 [22]. This regression model was used to calculate the annual percentage change (APC), estimating the magnitude of increase or decrease over time, and to identify the years where significant trend changes occurred.
Tables and graphs were produced using Microsoft Excel 2024, whilst all statistical analyses were carried out using Stata version 18.
Results
Temporal evolution of reported and hospitalized cases in Spain, 2012–2023
Paediatric TB cases reported to RENAVE decreased from 354 in 2012 to 231 in 2023; a rate of 5 per 100,000 inhabitants in 2012 and 3.5 in 2023, with a significant APC of −5.05 (95% CI: −8.1 to −1.8) for the entire period. Disaggregating by age group, a significant decrease was observed throughout the study period in all age groups (p < 0.05), except in the age group from 10 to 14 years. The greatest decrease was observed in the 5 to 9-year group (APC:−6.3 [95% CI: −11.4 to −0.8]). Between 2021 and 2023, a non-significant increasing trend was observed when considering all age groups (APC: 15.3 [95% CI: −9.3 to 36.6]) (Table 1).
Table 1.
Joinpoint analysis of pediatric TB notification (surveillance) and hospitalization rates, Spain, 2012–2023.
| Reported pediatric TB cases (Renave) |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Annual incidence rate |
2012–2023 |
Trend 1 |
Trend 2 |
Trend 3 |
|||||||||
| 2012 | 2023 | APC | 95%CI | Period | APC | 95%CI | Period | APC | 95%CI | Period | APC | 95%CI | |
| 0 to 14 years old | 5.0 | 3.5 | −5.1 | (−8.1,−1.8) | 2012–2018 | −1.7 | (−11.7,15.4) | 2018–2021 | −16.0 | (−24.0, 1.7) | 2021–2023 | 15.3 | (−9.3, 36.6) |
| Age group | |||||||||||||
| < 1 year | 7.8 | 4.5 | −5.3 | (−9.3,−1.2) | 2012–2018 | −0.6 | (−20.6, 24.1) | 2018–2021 | −17.8 | (−27.3, 8.5) | 2021–2023 | 12.5 | (−15.6, 46.0) |
| 1 to 4 years | 8.5 | 5.7 | −5.5 | (−8.8,−2.1) | 2012–2018 | −3.3 | (−13.8, 15.5) | 2018–2021 | −15.9 | (−24.4, 0.5) | 2021–2023 | 18.8 | (−8.2, 41.7) |
| 5 to 9 years | 3.3 | 2.2 | −6.3 | (−11.4,−0.8) | 2012–2018 | 0.4 | (−37.0, 88.9) | 2018–2021 | −24.7 | (−45.5, 35.8) | 2021–2023 | 26.4 | (−30.7, 107.3) |
| 10 to 14 years | 3.0 | 3.3 | −1.4 | (−4.5,1.8) | 2012–2018 | 1.2 | (−25.2, 49.7) | 2018–2021 | −8.8 | (−26.0, 24.0) | 2021–2023 | 8.9 | (−26.0, 48.4) |
| HOSPITALIZED PEDIATRIC TB CASES (CMBD) | |||||||||||||
| Rates | 2012–2023 | Trend 1 | Trend 2 | Trend 3 | |||||||||
| 2012 | 2023 | APC | 95%CI | Period | APC | 95%CI | Period | APC | 95%CI | Period | APC | 95%CI | |
| 0 to 14 years old | 5.0 | 2.8 | −7.2 | (−11.1,−3.3) | 2012–2018 | −5.7 | (−19.3, 20.4) | 2018–2021 | −17.0 | (−27.9;2.4) | 2021–2023 | 19.3 | (−11.7;48.3) |
| Age group | |||||||||||||
| <1 year | 8.2 | 3.6 | −6.0 | (−10.2,−1.6) | 2012–2017 | −3.5 | (−26.3, 28.6) | 2017–2020 | −12.4 | (−26.8, 11.8) | 2020–2023 | 0.8 | (−21.6, 37.9) |
| 1 to 4 years | 8.3 | 5.3 | −7.0 | (−11.7,−2.1) | 2012–2018 | −5.4 | (−17.9, 16.3) | 2018–2021 | −18.1 | (−27.7, −3.8) | 2021–2023 | 24.6 | (−7.2, 53.7) |
| 5 to 9 years | 3.3 | 1.7 | −9.2 | (−14.5,−3.5) | 2012–2018 | −4.0 | (−35.3, 62.8) | 2018–2021 | −27.0 | (−45.4, 22.7) | 2020–2023 | 32.0 | (−25.4, 105.9) |
| 10 to 14 years | 3.1 | 2.2 | −4.4 | (−8.0,−0.5) | 2012–2016 | −10.1 | (−31.5, 24.9) | 2016–2021 | −3.8 | (−23.5, 15.6) | 2021–2023 | 11.0 | (−20.0, 45.7) |
The number of hospitalizations decreased from 354 in 2012 to 184 in 2023; a rate of 5 per 100,000 in 2012 and 2.8 in 2023, with a significant APC of −7.25 (95% CI: −11.1 to −3.3). Disaggregated by age group, there was a significant decrease in all age groups and throughout the period that was more pronounced among children aged 5–9 years (APC = −9.2 [95% CI: −14.5 to −3.5]).). An increasing trend was observed in 2021–2023 for the 4 age groups, although not significant (Table 1). Disaggregating by sex, during this last period 2021–2023 hospitalization rates increased in males of all age groups and girls aged 1–4 years (Figure 1).
Figure 1.
Notification (surveillance) and hospitalization rates per 100,000 population by age group and sex, 2012–2023, Spain.
Characteristics of paediatric TB cases
During the period 2012–2023, a total of 3,472 TB cases were reported in children aged 0–14 years in RENAVE, of which 5.4% were imported (n = 186). The average annual notification rate during the study period was 4.2 per 100,000 inhabitants.
The number of cases reported to RENAVE was slightly higher in boys (52.5%, n = 1,824) compared to girls (47.5%, n = 1,648), with male to female ratio of 1.1:1. The reporting rate for the entire period was 4.3 and 4.1 per 100,000 inhabitants in boys and girls, respectively. The mean age was 5.8 years (SD = 4.5) and 5.9 years in girls (SD = 4.5). The age group with the highest number of reported cases was 1 to 4 years (n = 1,383; 39.8%), with a rate of 6.7 per 100,000 inhabitants (Table 2).
Table 2.
Characteristics of reported and hospitalized pediatric TB cases in Spain by age group between 2012 and 2023.
| Reported pediatric TB cases (Renave) N = 3,472 | ||||||
|---|---|---|---|---|---|---|
| Age groups | ||||||
| Total | <1 year | 1–4 years | 5–9 years | 10–14 years | ||
| Variables n (%)a | n (%) | n (%) | n (%) | n (%) | n (%) | p value |
| 3,472 (100) | 298 (8.6) | 1,383 (39.8) | 867 (25.0) | 924 (26.6) | ||
| Sex n = 3,472 (100) | 0.69 | |||||
| Boys | 1,824 (52.5) | 160 (53.7) | 721 (52.1) | 468 (54.0) | 475 (51.4) | |
| Girls | 1,648 (47.5) | 138 (46.3) | 662 (47.9) | 399 (46.0) | 449 (48.6) | |
| Imported n = 3,472 (100) | <0.001 | |||||
| Yes | 186 (5.4) | 8 (2.7) | 55 (4.0) | 46 (5.3) | 77 (8.3) | |
| No | 3,286 (94.6) | 290 (97.3) | 1,328 (96.0) | 821 (94.7) | 847 (91.7) | |
| Country of birth n = 3,177 (91.5) | <0.001 | |||||
| Spain | 2,507 (78.9) | 243 (91.7) | 1,113 (87.9) | 618 (78.1) | 533 (62.3) | |
| Outside Spain | 670 (21.1) | 22 (8.3) | 153 (12.1) | 173 (21.9) | 322 (37.7) | |
| Location n = 3,347 (96.4) | <0.001 | |||||
| Pulmonary | 2,577 (77.0) | 253 (86.1) | 1,081 (80.9) | 627 (75.0) | 616 (69.9) | |
| Extrapulmonary | 770 (23.0) | 41 (13.9) | 255(19.1) | 209 (25.0) | 265 (30.1) | |
| Diagnostic delay n = 681 (19.6) | <0.002 | |||||
| ≤24 days | 349 (51.3) | 49 (64.5) | 127 (51.4) | 88 (56.8) | 85 (41.9) | |
| >24 days | 332 (48.7) | 27 (35.5) | 120 (48.6) | 67 (43.2) | 118 (58.1) | |
| Delay in treatment initiation | ||||||
| n = 1,267 (36.5) | 0.11 | |||||
| ≤19 days | 656 (51.8) | 70 (59.3) | 245 (52.8) | 161 (51.8) | 180 (47.4) | |
| >19 days | 611 (48.2) | 48 (40.7) | 219 (47.2) | 144 (47.2) | 200 (52.6) | |
| Hospitalization n = 2,106 (60.6) | <0.001 | |||||
| Yes | 1,387 (65.9) | 162 (88.5) | 574 (72.0) | 321 (61.5) | 330 (54.6) | |
| No | 719 (34.1) | 21 (11.5) | 223 (28.0) | 201 (38.5) | 274 (45.4) | |
| Directly observed treatment n = 1,175 (33.8) | 0.23 | |||||
| Yes | 327 (27.8) | 24 (32.4) | 132 (31.1) | 83 (26.8) | 88 (24.0) | |
| No | 743 (63.2) | 44 (59.5) | 263 (61.9) | 198 (63.9) | 238 (65.1) | |
| Unknown | 105 (9.0) | 6 (8.1) | 30 (7.0) | 29 (9.3) | 40 (10.9) | |
| HIV result n = 2,445 (70.4) | 0.22 | |||||
| Positive | 39 (1.6) | 6 (2.8) | 14 (1.4) | 6 (1.0) | 13 (1.9) | |
| Negative | 1,983 (81.1) | 177(82.7) | 807 (82.5) | 468 (81.3) | 531 (78.6) | |
| NP | 423 (17.3) | 31 (14.5) | 158 (16.1) | 102 (17.7) | 132 (19.5) | |
| Treatment outcome n = 2,888 (83.2) | <0.001 | |||||
| Satisfactory | 2,695 (93.3) | 209 (92.1) | 1,090 (93.6) | 693 (95.0) | 703 (91.8) | |
| Death | 13 (0.5) | 7 (3.1) | 1 (0.1) | 1 (0.1) | 4 (0.5) | |
| Other | 180 (6.2) | 11 (4.8) | 74 (6.3) | 36 (4.9) | 59 (7.7) | |
| Hospitalised pediatric TB cases (CMBD) N = 2,768 | ||||||
| Age groups | ||||||
| Total | <1 year | 1–4 years | 5–9 years | 10–14 years | ||
| n (%) | n (%) | n (%) | n (%) | n (%) | p value | |
| 2,768 (100) | 272 (9.8) | 1,202 (43.4) | 657 (23.7) | 637 (23.1) | ||
| Sex | 0.92 | |||||
| Boys | 1,459 (52.7) | 138 (50.7) | 638 (53.1) | 347 (52.8) | 336 (52.7) | |
| Girls | 1,309 (47.3) | 134 (49.3) | 564 (46.9) | 310 (47.2) | 301 (47.3) | |
| Days of hospitalization | <0.001 | |||||
| ≤5 days | 1,517 (55) | 95 (34.9) | 684 (56.9) | 425 (64.7) | 313 (49.1) | |
| >5 days | 1,251 (45) | 177 (65.1) | 518 (43.1) | 232 (35.3) | 324 (50.9) | |
| Location | 0.024 | |||||
| Pulmonary | 1,897 (68.5) | 193 (71.0) | 852 (70.9) | 442 (67.3) | 410 (64.4) | |
| Extrapulmonary | 871 (31.5) | 79 (29.0) | 350 (29.1) | 215 (32.7) | 227 (35.6) | |
| Type of discharge | <0.001 | |||||
| Home | 2,661 (96.1) | 248 (91.2) | 1,159 (96.4) | 638 (97.1) | 616 (96.7) | |
| Death | 8 (0.3) | 4 (1.5) | 1 (0.1) | 0 (0.0) | 3 (0.5) | |
| Other | 99 (3.6) | 20 (7.3) | 42 (3.5) | 19 (2.9) | 18 (2.8) | |
NP: Not performed.
a% of data completeness in RENAVE (n = 3.472; data from CMBD was 100% complete).
Regarding hospitalization, between 2012 and 2023, a total of 2,768 children under 15 years of age were hospitalized, with a mean of HR of 3.3 per 100,000 inhabitants. There were no significant differences by sex (52.7% were boys), and the boy/girl ratio was 1.09:1. The mean age was 5.4 years (SD = 4.4), for both girls and boys. The highest percentage of hospitalizations occurred in the group from 1 to 4 years (n = 1,203, 43.4%) with a mean annual hospitalization rate of 5.8 per 100,000 inhabitants (Table 2).
Information on country of birth was recorded in 91.5% of reported cases. There were 670 cases born outside Spain (21.1%) from 51 different nationalities. The countries in which most cases were born were Morocco (n = 146), Romania (n = 50), and Pakistan (n = 44). The age group with the highest percentage of cases born in another country was 10 to 14 years old (37.7%). Mother country of birth was recorded in only 6.4% records, of which 38.5% had mothers born in Spain and 61.5% outside Spain. The main country of origin amongst mothers of cases was Morocco with 58 cases, followed by Romania with 17.
Regarding TB location, 77% (n = 2,577) and 68.5% (n = 1,897) of reported and hospitalized cases, respectively, had pulmonary forms of TB. This location was significantly more common in children under 1 year (86.1% and 71.0% for reported and hospitalized cases, respectively). Extrapulmonary forms were more common with increasing age group, for both data sources (Table 1). Considering only extrapulmonary forms of TB, severe presentations (miliary and meningitis) were more common amongst cases less than 1-year-old (n = 20 and n = 52, in RENAVE and CMBD, respectively). For all other age groups of reported cases, the lymph nodes were the most common site of extrapulmonary TB (Figure 2).
Figure 2.
Locations of extrapulmonary forms in reported cases (n = 770) and hospitalizations (n = 871) by age group, Spain, 2012–2023.
Complete data to calculate ‘Time to diagnosis′ and ‘Time to initiation of treatment’ were only available for 681 (19.6%) and 1,267 (36.5%) reported cases. The median time to diagnosis and initiation of treatment was 24 (IQR: 11–54) and 19 days (IQR: 9–41), respectively. The group aged 10 to 14 years had the greatest delay in diagnosis and initiating treatment, with a median of 33 days (IQR: 14–72) and 22 days (IQR 10–60), respectively (Supplementary Figure 1).
A non-significant decrease was observed in both ‘Time to diagnosis′ and ‘Time to initiation of treatment’ in 2020 during the onset of the COVID-19 pandemic, with a median of 15 and 22 days for each variable, respectively. In 2021, both figures increased, to 27.5 days for time to diagnosis and 23 days for time to initiation of treatment, both higher than the median of the entire study period (Figure 3).
Figure 3.
Median delay in diagnosis and treatment initiation in pediatric TB cases 2012–2023, Spain.
Children under 1-year-old most frequently required directly observed treatment and hospitalization (32.4% and 88.5%, respectively). For hospitalized cases, the median number of days of admission for all children was 5 days (IQR: 3–8), rising to 7 days amongst children under 1-year-old (IQR: 4–10).
Regarding the causative agent, there were 796 cases of M. tuberculosis complex unspecified, 471 cases of M. tuberculosis sensu stricto, 36 cases of M. bovis, and one case of M. caprae. Cases of M. bovis were mainly detected in lymphatic locations (22 out of 36; 64.7%).
The highest percentage of HIV co-infection was observed in children under 1-year-old (6 out of 214; 2.8%) (Table 2). Amongst hospitalizations, only three cases of HIV co-infection were identified. In total, there were 515 cases (13.8% of the total) where an antibiogram was performed. The anti-tuberculosis drug with the highest resistance was isoniazid, with 33 cases (n = 453;7.3%), followed by pyrazinamide and streptomycin, in 31 reported cases (n = 423;7.3% and n = 418;7.4% respectively). Amongst all reported cases, four cases of RR-TB and nine cases of MDR-TB were detected throughout the study period, with no cases of XDR-TB reported.
A satisfactory treatment outcome was recorded in 93.3% of notified cases. The group aged 10 to 14 years had the lowest proportion of satisfactory results (91.8%, p < 0.001). During the period analysed, no significant variations (p = 0.631) were observed in the percentages of satisfactory results, reaching a maximum of 95.4% in 2020 and a minimum of 91.4% in 2023. There were 13 and 8 deaths in RENAVE and CMBD. The mortality rate in paediatric age was 0.02 and 0.01 per 100,000 inhabitants in RENAVE and CMBD, respectively. By age group, the highest mortality rate was observed in children under 1 year of age (0.15 per 100,000 for reported cases and 0.08 for hospitalizations) (Table 2).
Discussion
This study is the first comprehensive analysis of paediatric TB cases and hospitalizations from two national databases, RENAVE and CMBD, in Spain. Through examining the evolution of trends over time and characteristics of cases, the findings deepen the epidemiological and clinical knowledge of this disease.
Our study found that overall notification and hospitalization rates have decreased across the study period, with a slight increase from 2021 in all age groups. A similar trend has been observed in adults in Spain and other European countries [14,23], which may be explained by several factors. During the pandemic, many countries diverted public health resources towards COVID-19, leading to poorer control and monitoring of other diseases such as TB [24]. Lockdowns and difficulty in accessing medical care may have led to longer delays in diagnosis and initiation of treatment [25], potentially contributing to an increase in transmission of the disease and the observed rebound in incidence starting from 2021 [25,26]. Other explanatory factors are the pandemic’s impact on poverty determinants, delays in diagnosis and/or notification, and changes in migration patterns following the COVID-19 pandemic [26].
We did not observe significant differences between sexes in the incidence rate of reported cases or hospitalizations across age groups. In contrast, amongst adult TB cases men are more affected, with almost twice as many cases reported amongst men compared with women in 2023, according to RENAVE data [23]. Factors such as immunological differences between sexes and risk behaviors like tobacco use, alcohol consumption or drug use which are more often associated with adult men, may explain these differences [27].
Regarding the differences by age group, for both reported cases and hospitalizations, the highest incidence rate occurred in the group of 1 to 4 years, followed by children under 1 year. Previous studies have shown that, after a primary infection, children under 5 years of age are more likely to progress to tuberculous disease than older children [28]. Therefore, it is important that for cases of pulmonary TB in adults, particular attention should be given to contacts under 5 years of age to ensure treatment is started as soon as possible.
Pulmonary TB was more common than extrapulmonary forms for all age groups, and most severe forms (meningitis and miliary TB) were more frequent in children under 1 year. These results align with previous studies [29,30]. This age group appears to be especially vulnerable, with the highest rate of hospitalization amongst notified cases, and longest hospital stays compared to older age groups, as well as the highest mortality rates amongst both notified and hospitalized cases. Although the absolute number of deaths was low, the mortality rate was found to decrease with each successive increase in age group.
The percentage of paediatric cases born outside Spain was highest in the 10–14 age group, decreasing with each descending age group. This has also been observed in other European countries [26]. In 2023, 34.3% of all TB cases reported in Spain were foreign-born [23], a percentage significantly higher than in that of the paediatric population (21%) [23]. This could be hypothesized by the fact that although their parents may not be originally from Spain, many of these children were born in the country where levels of circulating TB may be lower than their parent′s country of origin. In addition to country of birth, other social determinants such as overcrowding, unequal access to health services, and parents’ educational level can significantly influence vulnerability to TB in childhood [31]. These structural factors, especially in migrant families or those experiencing social exclusion, may facilitate transmission and hinder diagnosis and follow-up [31]. Further research is needed to better understand how such factors may affect paediatric TB in Spain, which could be supported by incorporating social and economic indicators into existing surveillance systems.
There were 36 TB paediatric cases caused by M. bovis. This is not unexpected, given that Spain is the second EU country in terms of the number of zoonotic TB cases, and this species is considered endemic in the country [32]. Moreover, an increase in M. bovis and M. caprae cases has been observed in recent years, which may reflect improvements in genomic surveillance and molecular diagnostics, as well as a growing recognition of transmission risks at the human-to-livestock interface [33,34].
In 2020, the year the COVID-19 pandemic began, there was a decrease in both time to diagnosis and time to treatment initiation for paediatric TB. However, this data should be interpreted with caution, given the low data completion. There are currently few studies analyzing this variable, and those that do exist use different methodologies making comparison difficult [35–37]. Paradoxically, the median time to starting treatment is shorter than the median time to diagnosis, which can be explained by the fact that treatment for probable and suspected cases may be started prior to diagnostic confirmation, particularly amongst children where obtaining sputum samples is more challenging [38]. Delays in diagnosis and in the initiation of treatment were defined using the median of the distribution observed in our cohort, given that the study includes both intrathoracic and extrapulmonary forms of tuberculosis, for which there are no universally applicable clinical cut-off points. This approach allows for a more accurate representation of the variability in diagnostic pathways across all disease localisations and facilitates internal comparison between age groups, particularly given the substantial proportion of missing values and the fact that this variable was used to explore potential differences in accessibility rather than to assess transmission risk. In younger children, especially those under 5 years of age, a delay in the diagnosis and/or the treatment initiation can have more serious consequences, including higher mortality rates, an increased risk of disseminated disease (miliary spread), and life-threatening complications like TB meningitis, due to their developing immune systems [39]. This fact underscores the need for flexible care pathways and specific protocols in primary and hospital care for the early identification of TB paediatric cases.
HIV co-infection was infrequent. In countries with adequate antiretroviral treatment coverage and low TB transmission such as Spain, paediatric TB-HIV co-infection is rare but associated with significant morbidity [40].
The limited availability of information on drug susceptibility made it difficult to characterize drug resistant TB including detecting trends in resistance over the study period. As the management of drug-resistant TB involves more complex and potentially toxic treatment regimens, particularly for children, it is critical to improve the recording of resistance data and enhance diagnostic methods to better estimate the true burden of drug-resistant TB in children [41].
Limitations and conclusions
Our study has several important limitations. First, the incompleteness of surveillance data; there were several key variables with a high proportion of missing information, such as date of onset, date of diagnosis, microbiological confirmation, HIV coinfection and treatment resistance. Furthermore, there might be underreporting in RENAVE. Another limitation of the study is that with the available surveillance data it is not possible to distinguish whether the reported cases correspond to index cases or secondary cases. Finally, data available is limited in terms of social and economic determinants. Reporting this information could contribute to create a clearer picture of the epidemiological situation of TB and inform health policies addressing TB in children. Despite these limitations, our work has several strengths. Combining two complementary national databases allowed us to obtain a more comprehensive view of paediatric TB in Spain. Furthermore, using more than 10 years of data allows us see how paediatric TB has evolved over time in Spain.
From the start of the study period in 2012, paediatric TB showed a downward trend until the beginning of the COVID-19 pandemic in 2020, when incidence rates of both reported cases and hospitalizations began to increase, although not yet to pre-pandemic levels. Given the disruption of the COVID-19 pandemic on TB control, it is necessary to consolidate and strengthen surveillance systems to support a return to the previous trajectory of a declining trend. The disruption associated with 2020 may have influenced surveillance activities, and the patterns observed in subsequent years should therefore be interpreted with caution rather than as evidence of a structural shift. Improvements in data collection, advanced data interoperability (including social determinants) and establishing specific protocols for the study of TB contacts, enhanced diagnosis and treatment in children, will be key to strengthening paediatric TB prevention and control.
Supplementary Material
Acknowledgments
This work is based on two national sources of information, in which a great team of people are committed. Therefore, we would like to thank all the professionals involved in data collecting, from clinicians, coders, administrative staff, computer scientists, epidemiologists, microbiologists, and especially to the CMBD from the Ministry of Health and the community participating in the National Surveillance Network. Finally, we would also like to acknowledge the important work of the Spanish Working Group on Tuberculosis.
List of membership of the Spanish Working Group on Tuberculosis:
Consejería de Salud de Andalucía: Verónica Álamo; Departamento de Sanidad de Aragón: Cristina Navarro; Consejería de Salud de Asturias: Ana Fernandez Ibañez; Consejería de Salud de Baleares: Jaume Giménez Duran; Consejería de Salud de Canarias: Laura García Hernández; Consejería de Salud de Cantabria: Marta Pacheco Gorostiaga; Consejería de Salud de Castilla y León: María del Henar Marcos; Consejería de Salud de Castilla y la Mancha: Pilar Peces; Consejería de Salud de Cataluña: Laura Gavalda Maestre; Consejería de Salud de la Comunidad Valenciana: Empar Giner Ferrando; Consejería de Salud de Extremadura: María del Mar López-Tercero Torvisco; Consejería de Salud de Galicia: Elena Cruz Ferro; Consejería de Salud de La Rioja: Carmen Quiñones; Consejería de Salud de Madrid: Mercedes Rumayor; Consejería de Salud de Murcia: Blanca Andreu Ivorra; Instituto de Salud Pública de Navarra: Jesus Castilla; Consejería de Salud del País Vasco: Patricia Sancho; Consejería de Salud del Ceuta: Violeta Ramos; Consejería de Salud del Melilla: Daniel Castrillejo. María Fontana: Conceptualisation, data curation, formal analysis, methodology, writing–original draft, writing–review and editing. Juan Miguel Guerras: Conceptualisation, data curation, formal analysis, methodology, writing–original draft, writing–review and editing. Allegra Chatterjee: formal analysis, methodology, writing–review and editing. Alvaro Roy: formal analysis, methodology, writing–review and editing. Christine Schwarz: Conceptualisation, writing–review and editing. Begoña Santiago-Garcia: Conceptualisation, writing–review and editing. Alfredo Tagarro: Conceptualisation, writing–review and editing. Laura Herrera-León: Conceptualisation, writing–review and editing. Zaida Herrador: Conceptualisation, data curation, formal analysis, methodology, project administration, writing–original draft, writing–review and editing. All the authors have approved this version, and they all agree to be accountable for all aspects of the work.
Disclosure statement
The authors have nothing to disclose.
Ethical statement
This study used anonymized medical data from CMBD and RENAVE in Spain. Both databases are maintained by the Ministry of Health (MoH). Formal applications were submitted to the relevant organisations to obtain CMBD and RENAVE data. In these applications, a signed confidentiality commitment is required. All data is anonymized and de-identified. All databases comply with Spain’s personal data protection regulations (https://www.sanidad.gob.es/en/proteccionDatosPersonales/home.htm). Data from the National Statistics Institute (INE) were obtained through open data access. Thus, no ethical approval or informed consent was required by the Spanish Human Research Act (https://www.isciii.es/documents/d/guest/spanishlawonbiomedicalresearchenglish) to conduct this data analysis. This research study on humans was performed in accordance with the principles stated in the Declaration of Helsinki.
Data availability statement
This study involves the use of patient data owned by third-party organizations. The CMBD is a set of clinical–administrative data of hospitalizations that collects sociodemographic, clinical and administrative patient data from the National Health System (NHS) discharge reports. It is hosted by the Ministry of Health (https://www.sanidad.gob.es/estadEstudios/portada/home.htm). The National Epidemiological Surveillance Network (RENAVE) provides data on cases reported through the national reporting electronic platform (in Spanish Sistema de Vigilancia en España – SiViEs) and it is hosted by the National Centre of Epidemiology (https://cne.isciii.es/es/servicios/enfermedades-transmisibles/protocolos-renave). Formal applications were submitted to the relevant organisations to obtain CMBD, RENAVE data. In these applications, a signed confidentiality commitment is required. Datasets cannot be shared due to security concerns but they can be requested through the provided links. All data is anonymized and de-identified. All sources follow the mandate of Spanish and international legislation (https://www.sanidad.gob.es/en/proteccionDatosPersonales/home.htm). Data will be made available on reasonable request.
References
- 1.Global Tuberculosis Report . 2024. [Internet]. [cited 2025 May 21]. Available from: https://www.who.int/teams/global-programme-on-tuberculosis-and-lung-health/tb-reports/global-tuberculosis-report-2024.
- 2.World Health Organization, Centers for Disease Control (U.S.), Stop TB Partnership, United States. Agency for International Development, United Nations Children’s Fund (UNICEF) . Roadmap for childhood tuberculosis: towards zero deaths. [Internet]. Geneva: World Health Organization; 2013. p 38. [cited 2025 May 27]. Available from: https://iris.who.int/handle/10665/89506 [Google Scholar]
- 3.Tuberculosis pulmonar en Pediatría | Pediatría integral . [Internet]. 2021. Apr 6 [cited 2025 Apr 23]. Available from: https://www.pediatriaintegral.es/publicacion-2021-03/tuberculosis-pulmonar-en-pediatria/.
- 4.Perez-Velez CM, Marais BJ.. Tuberculosis in children. N Engl J Med. 2012;367(4):348–361. doi: 10.1056/NEJMra1008049. [DOI] [PubMed] [Google Scholar]
- 5.Basu Roy R, Whittaker E, Seddon JA, et al. Tuberculosis susceptibility and protection in children. Lancet Infect Dis. 2019;19(3):e96–108–e108. doi: 10.1016/S1473-3099(18)30157-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Carvalho ACC, Cardoso CAA, Martire TM, et al. Epidemiological aspects, clinical manifestations, and prevention of pediatric tuberculosis from the perspective of the end TB strategy. J Bras Pneumol. 2018;44(2):134–144. doi: 10.1590/S1806-37562017000000461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lamb GS, Starke JR.. Tuberculosis in infants and children. Microbiol Spectr. 2017;5(2). doi: 10.1128/microbiolspec.TNMI7-0037-2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rodríguez-Molino y García – Tuberculosis pediátrica y tuberculosis congénita.pdf . [Internet]. [cited 2025. May 19]. Available from: https://www.aeped.es/sites/default/files/documentos/22_tuberculosis.pdf.
- 9.Baquero-Artigao F, del Rosal T, Falcón-Neyra L, et al. Actualización del diagnóstico y tratamiento de la tuberculosis. An Pediatría. 2023;98(6):460–469. doi: 10.1016/j.anpedi.2023.03.011. [DOI] [PubMed] [Google Scholar]
- 10.Piñeiro Pérez R, Santiago García B, Mellado Peña MJ, et al. Red Española de Estudio de Tuberculosis Pediátrica. A la vanguardia de Europa. Rev Esp Salud Públ. [Internet]. 2017; [cited 2023 May 12]; 91. Available from: https://scielo.isciii.es/scielo.php?script=sci_abstract&pid=S1135-57272017000100100&lng=es&nrm=iso&tlng=es [PMC free article] [PubMed] [Google Scholar]
- 11.Carvalho I, Goletti D, Manga S, et al. Managing latent tuberculosis infection and tuberculosis in children. Pulmonology. 2018;24(2):106–114. doi: 10.1016/j.rppnen.2017.10.007. [DOI] [PubMed] [Google Scholar]
- 12.Jain SK, Andronikou S, Goussard P, et al. Advanced imaging tools for childhood tuberculosis: potential applications and research needs. Lancet Infect Dis. 2020;20(11):e289–97–e297. doi: 10.1016/S1473-3099(20)30177-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.WHO consolidated guidelines on tuberculosis: module 5: management of tuberculosis in children and adolescents . [Internet]. [cited 2025. May 12]. Available from: https://www.who.int/publications/i/item/9789240046764. [PubMed]
- 14.European Centre for Disease Prevention and Control, World Health Organization . Tuberculosis surveillance and monitoring in Europe 2025: 2023 data. [Internet]. LU: Publications Office; 2025. [cited 2025 May 12]. Available from: https://data.europa.eu/doi/10.2900/2650058 [Google Scholar]
- 15.Ministerio de Sanidad y Consumo . Real Decreto 2210/1995, de 28 de diciembre, por el que se crea la red nacional de vigilancia epidemiológica. [Internet]. Sect. 1, Real Decreto 2210/1995; p 2153–8. [1996. Jan 24]. Available from: https://www.boe.es/eli/es/rd/1995/12/28/2210.
- 16.Protocolos de la Red Nacional de Vigilancia Epidemiológica. Ponencia de Vigilancia Epidemiológica: 9 de abril de 2013 . Comisión de salud Pública: 19 de junio de 2013. Consejo Interterritorial del Sistema Nacional de Salud: 23 de julio de 2013. [Internet]. Red Nacional de Vigilancia Epidemiológica. Instituto de Salud Carlos III; 2015. [cited 2025 Feb 19]. Available from: http://hdl.handle.net/20.500.12105/5345. doi: 10.4321/repisalud.5345. [DOI]
- 17.Muzanyi G, Bark CM.. Severe Tuberculosis. Med Clin North Am. 2025;109(3):641–650. doi: 10.1016/j.mcna.2024.12.001. [DOI] [PubMed] [Google Scholar]
- 18.GOV.UK . [Internet]. Tuberculosis in children aged 0 to 17 years, England. 2023. [cited 2026 Mar 26]. Available from: https://www.gov.uk/government/publications/tuberculosis-in-england-2024-report/tuberculosis-in-children-aged-0-to-17-years-england-2023.
- 19.Guidance for national tuberculosis programmes on the management of tuberculosis in children . 2nd ed. [Internet]. [cited 2026. Mar 26]. Available from: https://www.who.int/publications/i/item/9789241548748.
- 20.Ministerio de Sanidad – Sanidad en datos – Registro de Altas de los Hospitales del Sistema Nacional de Salud. CMBD . [Internet]. [cited 2025. Apr 24]. Available from: https://www.sanidad.gob.es/estadEstudios/estadisticas/cmbdhome.htm.
- 21.INE. [Internet]. [cited 2025 Feb 19]. INEbase/Demografía y población. Available from: https://ine.es/dyngs/INEbase/es/categoria.htm?c=Estadistica_P&cid=1254734710984.
- 22.Joinpoint Regression Program . Internet]. [cited 2025. May 19]. Available from: https://surveillance.cancer.gov/joinpoint/.
- 23.RENAVE_informe_Vigilancia_TB__2023.pdf . [Internet]. [cited 2025. Sep 25]. Available from: https://www.sanidad.gob.es/ciudadanos/enfLesiones/enfTransmisibles/sida/TB/docs/RENAVE_informe_Vigilancia_TB__2023.pdf.
- 24.Starke JR, Erkens C, Ritz N, et al. Strengthening tuberculosis services for children and adolescents in low endemic settings. Pathogens. 2022;11(2):158. doi: 10.3390/pathogens11020158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Barros M, Silva JB, Sousa M, et al. COVID-19 pandemic and tuberculosis: how to ensure adequate care in pediatric age. Pulmonology. 2022;28(6):421–422. doi: 10.1016/j.pulmoe.2022.06.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Cristea V, Ködmön C, Dias JG, Group the paediatric T expert ., et al. Increase in tuberculosis among children and young adolescents, European Union/European Economic Area, 2015 to 2023. Euro Surveill. 2025;30(11):2500172. doi: 10.2807/1560-7917.ES.2025.30.11.2500172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Klein SL, Flanagan KL.. Sex differences in immune responses. Nat Rev Immunol. 2016;16(10):626–638. doi: 10.1038/nri.2016.90. [DOI] [PubMed] [Google Scholar]
- 28.Martinez L, Cords O, Horsburgh CR, Villaorduna C, Desai Ahuja S ., et al. The risk of tuberculosis in children after close exposure: a systematic review and individual-participant meta-analysis. Lancet. 2020;395(10228):973–984. doi: 10.1016/S0140-6736(20)30166-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hussey G, Chisholm T, Kibel M.. Miliary tuberculosis in children: a review of 94 cases. Pediatr Infect Dis J. 1991;10(11):832–836. doi: 10.1097/00006454-199111000-00008. [DOI] [PubMed] [Google Scholar]
- 30.Marais BJ, Gie RP, Schaaf HS, et al. The clinical epidemiology of childhood pulmonary tuberculosis: a critical review of literature from the pre-chemotherapy era. Int J Tuberc Lung Dis. 2004;8(3):278–285. [PubMed] [Google Scholar]
- 31.Santos VS, Holanda J D, Gois-Neto RDS, et al. Social and health factors associated with unfavourable treatment outcomes in children and adolescents with drug-sensitive tuberculosis in Brazil: a national retrospective cohort study. Lancet Reg Health Am. 2024:40:100938. doi: 10.1016/j.lana.2024.100938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.The European Union One Health 2024 Zoonoses Report . [Internet]; 2025. [cited 2026 Mar 26]. Available from: https://www.ecdc.europa.eu/en/publications-data/european-union-one-health-2024-zoonoses-report.
- 33.Roy Á, Gómez-Barroso D, Cruz-Ferro E, et al. Spatiotemporal distribution and clinical characteristics of zoonotic tuberculosis, Spain, 2018–2022. Emerg Infect Dis. [Internet]. 2022;5–31 Volume July (7):1344–1352. doi: 10.3201/eid3107.250031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Pérez de Val B, Vidal E, Stuber T, et al. Zoonotic tuberculosis in Catalonia, Spain: Phylogenetic insights into Mycobacterium bovis and M. caprae transmission at the human–livestock interface. One Health. 2025;20:100993. doi: 10.1016/j.onehlt.2025.100993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Nofuentes-Pérez E, Zamora-Molina L, Grau-Delgado J, et al. Missed opportunities for detecting tuberculosis: an analysis of emergency department visits before the diagnosis. Rev Clin Esp (Barc). 2019;219(7):390–393. doi: 10.1016/j.rce.2018.11.010. [DOI] [PubMed] [Google Scholar]
- 36.Lewis KE, Stephens C, Shahidi MM, et al. Delay in starting treatment for tuberculosis in east London. Commun Dis Public Health. 2003;6(2):133–138. [PubMed] [Google Scholar]
- 37.Feiterna-Sperling C, Thoulass J, Krüger R, et al. Pilot study to identify missed opportunities for prevention of childhood tuberculosis. Eur J Pediatr. 2022;181(9):3299–3307. doi: 10.1007/s00431-022-04537-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Seminario A, Anibarro L, Sabriá J, et al. Estudio del retraso diagnóstico de la tuberculosis en España. Arch Bronconeumol. 2021;57(6):440–442. doi: 10.1016/j.arbres.2020.09.003. [DOI] [PubMed] [Google Scholar]
- 39.Jenkins HE, Tolman AW, Yuen CM, et al. Incidence of multidrug-resistant tuberculosis disease in children: systematic review and global estimates. Lancet. 2014;383(9928):1572–1579. doi: 10.1016/S0140-6736(14)60195-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.López-Medina EM, Sainz T, de Ory SJ, et al. Tuberculosis in a Spanish cohort of children living with HIV: the CHOTIS study (Childhood HIV and TB study). Int J Tuberc Lung Dis. 2020;24(3):303–309. doi: 10.5588/ijtld.19.0237. [DOI] [PubMed] [Google Scholar]
- 41.Basu S, Chakraborty S.. A comprehensive review of the diagnostics for pediatric tuberculosis based on assay time, ease of operation, and performance. Microorganisms. 2025;13(1):178. doi: 10.3390/microorganisms13010178. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
This study involves the use of patient data owned by third-party organizations. The CMBD is a set of clinical–administrative data of hospitalizations that collects sociodemographic, clinical and administrative patient data from the National Health System (NHS) discharge reports. It is hosted by the Ministry of Health (https://www.sanidad.gob.es/estadEstudios/portada/home.htm). The National Epidemiological Surveillance Network (RENAVE) provides data on cases reported through the national reporting electronic platform (in Spanish Sistema de Vigilancia en España – SiViEs) and it is hosted by the National Centre of Epidemiology (https://cne.isciii.es/es/servicios/enfermedades-transmisibles/protocolos-renave). Formal applications were submitted to the relevant organisations to obtain CMBD, RENAVE data. In these applications, a signed confidentiality commitment is required. Datasets cannot be shared due to security concerns but they can be requested through the provided links. All data is anonymized and de-identified. All sources follow the mandate of Spanish and international legislation (https://www.sanidad.gob.es/en/proteccionDatosPersonales/home.htm). Data will be made available on reasonable request.