Subnational Heterogeneity in Respiratory Syncytial Virus Seasonality in Tropical Colombia, 2022–2024

Abstract

Background

Respiratory syncytial virus (RSV) transmission in tropical settings is often assumed to occur year-round, and preventive strategies are frequently recommended accordingly. However, sub-national variation in epidemic timing remains incompletely characterized. Granular local surveillance data may help refine timing of prevention, particularly in resource-limited settings prioritizing high-risk populations.

Method

We analyzed aggregated RSV surveillance data from the Viral Respiratory Infection Surveillance Network (VIRAL Network LATAM–Colombia) across epidemiological weeks 1–52 during 2022–2024. Data were obtained from 15 centers in 7 climatically diverse cities spanning sea level to 2640 m altitude. City-specific percentile thresholds were used to define transmission intensity. Epidemic patterns were categorized as concentrated, prolonged, or bimodal, and inter-city synchrony was assessed using pairwise Spearman correlations.

Results

Among 39 505 multiplex respiratory panels, 6699 (17.0%) were RSV-positive. RSV was detected year-round in all cities, with off-season detections accounting for 16.2%–64.5% of annual cases. Five cities exhibited a synchronized epidemic corridor between March and July with high temporal concordance (all pairwise ρ ≥ .71, P < .01). Cartagena showed delayed coastal transmission, while Cúcuta demonstrated bimodal circulation with attenuated seasonality. Overall, 2 cities met criteria for concentrated transmission, 4 for prolonged transmission, and 1 for bimodal patterns.

Conclusions

Although RSV circulated throughout the year, epidemic activity converged temporally across most Colombian cities, with notable sub-national heterogeneity. These findings suggest that improved understanding of local epidemic timing could support geographically tailored implementation of RSV preventive strategies.

Graphical Abstract

Graphical Abstract.

Tidbit Graphical Abstract: https://tidbitapp.io/institutional-portal/open-forum-infectious-diseases/tidbits/subnational-heterogeneity-in-rsv-seasonality-in-tropical-colombia-2022-2024/update

This multicity molecular surveillance study demonstrates substantial sub-national heterogeneity in RSV seasonality within tropical Colombia, with synchronized epidemic corridors across most cities. Findings support geographically tailored timing of maternal vaccination and long-acting monoclonal antibody strategies to optimize prevention impact.

Respiratory syncytial virus (RSV) is a leading cause of acute lower respiratory tract infection and hospitalization in infants worldwide, with a disproportionate burden in low- and middle-income countries [1, 2]. The recent availability of maternal prefusion RSV F protein (RSVpreF) vaccination and long-acting monoclonal antibodies has expanded prevention options [3, 4], but both confer time-limited protection, making alignment with local circulation patterns critical [5–7].

RSV is widely regarded as a year-round pathogen in tropical countries, although sub-national circulation patterns remain poorly defined. RSV seasonality is well characterized in temperate regions, where predictable winter epidemics facilitate seasonal immunization strategies [8]. In tropical and subtropical settings, circulation patterns are more variable and less predictable, posing substantial challenges for prevention planning [3, 8, 9]. Prior reviews have classified tropical countries as exhibiting “year-round” seasonality, but this characterization may obscure important sub-national heterogeneity. Analyses from China [10], Brazil [8, 11], and Guatemala [12] have demonstrated significant within-country variation linked to latitude, altitude, and climate, suggesting that national-level classifications may be insufficient to guide regional immunization strategies. Nevertheless, the evidence base characterizing RSV seasonality in Latin America and the Caribbean remains limited [13].

Colombia offers a unique setting for examining this heterogeneity: 7 cities spanning sea level to above 2600 m altitude, 3 distinct precipitation regimes, and temperatures from 15°C to above 28°C—all within a single national health system. In November 2025, Colombia incorporated maternal RSVpreF vaccination into its national immunization program, while Bogotá, the country's most populous city, introduced nirsevimab for infants as a complementary component of a hybrid prevention strategy, highlighting the need for sub-national data to optimize prevention. We aimed to characterize RSV seasonality across 7 Colombian cities using year-round molecular surveillance, identify distinct transmission patterns, and assess their implications for regional prevention planning and hospital preparedness.

METHODS

Study Design and Setting

We conducted a descriptive time-series analysis of aggregated RSV surveillance data from the Viral Respiratory Infection Surveillance Network (VIRAL Network LATAM–Colombia) spanning epidemiological weeks 1–52 for 2022–2024 (156 weeks total). The network comprises referral hospitals performing year-round molecular surveillance among patients hospitalized with acute respiratory illness. Testing was directed at patients with moderate-to-severe respiratory distress, escalating oxygen requirements, or need for intensive care. Participating institutions serve both pediatric and adult populations; based on clinical profiles, an estimated >90% of panels were performed in pediatric patients.

Seven cities spanning 3 altitudinal zones contributed data (Supplementary figure 1): highland (>2000 m: Bogotá, 2640 m; Pasto, 2527 m), mid-altitude (1000–2000 m: Medellín, 1495 m; Cali, 1018 m), and lowland (<1000 m: Neiva, 442 m; Cúcuta, 320 m; Cartagena, sea level), classified per Colombia's Caldas-Lang system [14]. Fifteen centers participated: 5 in Bogotá, 3 each in Medellín and Cali, and 1 each in Cúcuta, Neiva, Pasto, and Cartagena. Complete 3-year coverage was achieved for Bogotá, Cali, and Medellín; partial coverage was available for remaining cities.

Laboratory Testing

Respiratory specimens were tested using multiplex molecular panels capable of detecting RSV alongside influenza A/B, parainfluenza, rhinovirus/enterovirus, adenovirus, and human metapneumovirus, among others. Most sites employed the BioFire FilmArray Respiratory Panel (bioMérieux); 1 site used the Seegene Allplex Respiratory Panel. Weekly data extraction included absolute RSV-positive counts and positivity rates. Analyses used epidemiological weeks 1–52; weeks with zero panels were excluded from percentile calculations but retained in visualizations. Because this was a sentinel rather than population-based surveillance network, specimen numbers reflected pediatric catchment population, institutional complexity, local practices, and availability of molecular diagnostics. Testing volumes varied across sites and did not depend on population size. Nevertheless, standardized molecular platforms and harmonized surveillance procedures ensured comparability of temporal circulation patterns across cities.

Percentile-Based Classification Framework

Fixed RSV positivity thresholds (eg, 3%) are poorly discriminative in tropical settings where weekly positivity frequently exceeds such cutoffs year-round. We therefore adopted a city-specific percentile framework: very low (<P25), low (P25–P50), moderate (>P50–P75), high (>P75–P90), and peak-intensity (>P90). Weeks > P50 were classified as epidemic-level activity, consistent with recent analyses [13] and thresholds applied in a 10-year Hong Kong RSV study [15]. This approach is analogous to the WHO Average Curves Method [16]; however, no standardized approach exists for defining RSV epidemic timing [17].

Seasonality Metrics

All seasonality metrics were derived from 52 pooled weekly positivity values. Sustained transmission was defined as ≥2 consecutive weeks > P50, consistent with standard surveillance practice [18]. Runs separated by ≤2-week gaps below P50 were merged. Season onset and end were defined as the first and last weeks of the merged run containing the annual peak. Transmission was classified as temporally concentrated when epidemic burden was ≥75% within ≤22 weeks, following Li et al.'s framework for passive immunization planning in low- and middle-income countries [7]. Cities not meeting both criteria were classified as prolonged. Cities with 2 distinct epidemic periods ≥6 weeks apart were classified as bimodal. Inter-city synchrony was assessed using pairwise Spearman correlations of mean monthly positivity (ρ ≥ .70 = high synchrony) [19, 20]. Two-sided P values were used. Weekly series were normalized to each city's maximum for timing comparisons [21]. Data management and statistical analyses were performed using Python 3.11 (NumPy, Pandas, and SciPy, Python Software Foundation, USA).

Precipitation Seasonality

Rainfall patterns were characterized using official climatological descriptions from IDEAM [22–24]. Three operational precipitation regimes were identified among the 7 study cities: Andean bimodal with 2 wet peaks bracketing a mid-year dry period (Bogotá, Medellín, Cali, Cúcuta), transitional bimodal with an asymmetric prolonged dry season (Neiva, Pasto), and unimodal Caribbean with a single May-to-November wet season (Cartagena) [25]. The temporal relationship between precipitation and RSV timing was examined descriptively.

Ethical Considerations

The protocol received ethics committee approval from each participating institution. Data were de-identified and aggregated; informed consent was waived. The study was conducted in accordance with the Declaration of Helsinki.

RESULTS

Study Population

Between January 2022 and December 2024, 39,505 multiplex panels were performed across 15 centers: 11 708 (29.6%) in 2022, 14,251 (36.1%) in 2023, and 13 546 (34.3%) in 2024. Bogotá contributed 16 235 tests (41.1%), followed by Cali (9106; 23.1%), Medellín (5196; 13.2%), Cartagena (4247; 10.8%), Cúcuta (2584; 6.5%), Neiva (1362; 3.4%), and Pasto (775; 2.0%). Table 1 summarizes city characteristics.

Table 1.

RSV Surveillance Summary by Altitudinal Zone Across Colombian Cities, 2022–2024

City	Alt. Zone	Alt. (m)	Centers	Tests	RSV Cases	Pos. (%)	Median Weekly RSV (IQR)a
Bogotá	Highland	2640	5	16 235	3114	19.2	40 (28.0–87.2)
Pasto	Highland	2527	1	775	161	20.8	2 (1.0–6.0)
Medellín	Mid-alt.	1495	3	5196	765	14.7	12.5 (9.8–19.2)
Cali	Mid-alt.	1018	3	9106	977	10.7	16 (6.8–28.5)
Neiva	Lowland	442	1	1362	314	23.1	3.5 (1.0–9.8)
Cúcuta	Lowland	320	1	2584	465	18.0	8 (6.8–11.0)
Cartagenab	Lowland	0	1	4247	903	21.3	6.5 (2.0–31.5)
Total	—	—	15	39 505	6699	17.0	—

Abbreviations: IQR, interquartile range; m, meters above sea level; RSV, respiratory syncytial virus.

^aMedian (IQR) of RSV cases summed across years per epidemiological week (n = 52).

^b71 weeks (May 2023−December 2024, intermittent gaps).

RSV Burden and Temporal Patterns

RSV was detected in 6699 of 39 505 samples (overall positivity 17.0%), stable across years (17.5%, 16.0%, and 17.4% in 2022–2024, respectively). Bogotá accounted for 3114 cases (46.5%), followed by Cali (977; 14.6%), Cartagena (903; 13.5%), Medellín (765; 11.4%), Cúcuta (465; 6.9%), Neiva (314; 4.7%), and Pasto (161; 2.4%). Weekly positivity ranged from 0% to 63.2% (Pasto, week 22). Transmission onset varied by up to 18 weeks across cities (Figure 1). Four cities initiated their epidemic season in March—Bogotá and Pasto (week 10), Medellín (week 11), and Neiva (week 12)—while Cali exhibited the earliest onset in February (week 5). Cúcuta (week 22) and Cartagena (week 23) showed delayed onset in late May–June. Five cities (71%) reached peak intensity within a 4-week window in May–June (weeks 20–23), with peak positivity ranging from 30.7% to 63.2%. Cali peaked earlier in late March (week 13) and Cartagena later in early August (week 31).

Figure 1.

RSV temporal circulation patterns across Colombian cities, 2022–2024. Heatmap displaying normalized weekly RSV positivity (0%–100%, relative to each city’s maximum) across epidemiological weeks. Cities are ordered by median epidemic peak week, and white circles indicate the median peak week for each city. Most cities reached peak RSV activity between epidemiological weeks 20 and 23, whereas Cartagena exhibited a substantially delayed peak and Cúcuta showed more prolonged circulation, illustrating marked sub-national heterogeneity in RSV seasonality. Abbreviation: RSV, respiratory syncytial virus.

Escalation from baseline to peak was gradual in 6 cities (12–18 weeks, 1.4%–4.5%/week). Cúcuta was the exception, with rapid escalation (4 weeks, 7.2%/week) driven by its elevated baseline positivity (9.7%), the highest observed. Baseline positivity ranged from 0.3% (Pasto) to 9.7% (Cúcuta). Despite these differences, all 7 cities exhibited concurrent elevated activity during weeks 23–30 (June–August).

Epidemic Season and Case Concentration

The defined epidemic period (>P50) captured 35.5%–83.8% of total RSV cases (Figure 2). Two cities exhibited concentrated transmission: Pasto (82.0% within 21 weeks) and Cartagena (80.7% within 19 weeks). Four showed prolonged transmission: Neiva (83.8%, 23 weeks), Cali (79.2%, 26 weeks), Bogotá (71.6%, 22 weeks), and Medellín (58.6%, 23 weeks). Cúcuta was the only city with a bimodal pattern, with 35.5% of cases during the primary epidemic period. Across all cities, between 51.8% and 73.7% of cases occurring within the city-specific epidemic period were concentrated during high- or peak-intensity weeks (≥P75) (Table 2). Off-season detections ranged from 16.2% (Neiva) to 64.5% (Cúcuta).

Figure 2.

RSV case distribution by epidemic period across Colombian cities, 2022–2024. Stacked bars show the proportion of RSV cases occurring during epidemic (>P50), high-intensity (P75–P90), peak (>P90), and off-season (<P50) periods. Cities are ordered by epidemic burden in descending order. Pasto and Cartagena exhibited concentrated epidemics with most cases occurring during the epidemic period, whereas Cúcuta displayed a bimodal transmission pattern with most cases occurring outside the primary epidemic period. Definitions of epidemic intensity categories and transmission pattern classifications are provided in the Methods section. Abbreviation: RSV, respiratory syncytial virus.

Table 2.

RSV Epidemic Characteristics Across Colombian Cities, 2022–2024

Transmission Dynamics										Case Distribution Within Epidemic Period (%)
		Onset	End	Duration	Burden (%)	Base (%)	Peak Week	Peak (%)	Vel (%/wk)	Moderate	High	Peak	High + Peak
Cali	Prolonged	Week 5	Week 30	26	79.2	2.5	13	27.0	2.1	36.8	29.0	34.3	63.2
Bogotá	Prolonged	Week 10	Week 31	22	71.6	7.1	21	42.9	2.6	29.4	30.7	40.0	70.6
Pasto	Concentrated	Week 10	Week 30	21	82.0	0.3	22	63.2	3.6	33.1	30.3	36.6	66.9
Medellín	Prolonged	Week 11	Week 33	23	58.6	7.2	23	30.7	1.4	35.7	32.1	32.2	64.3
Neiva	Prolonged	Week 12	Week 34	23	83.8	2.1	20	56.8	3.3	26.3	38.7	35.0	73.7
Cúcutaa	Bimodal	Week 22	Week 34	13	35.5	9.7	22	41.5	7.2	48.2	26.8	25.0	51.8
Cartagenab	Concentrated	Week 23	Week 41	19	80.7	0.7	31	57.5	4.5	28.6	34.3	37.1	71.4

Cities ordered by epidemic onset. Onset/End = epidemic period boundaries (>P50). Dur = duration in weeks. Burd = proportion of total RSV cases within epidemic period. Base = baseline positivity. Vel = escalation velocity (%/wk).

Case distribution within epidemic period: Mod = Moderate (P50−P75); High = P75−P90; Peak = >P90; H + P = High + Peak combined (≥P75).

Pattern classifications: Concentrated = ≥ 75% epidemic burden AND ≤22 weeks duration; Prolonged = not meeting both criteria; Bimodal = 2 epidemic periods ≥6 weeks apart. Neiva exceeded the burden threshold (83.8%) but was classified as prolonged due to duration (23 weeks) exceeding ≤22-week criterion.

^aSecondary high-activity period weeks 41–47.

^bSurveillance initiated May 2023 (71 weeks).

Precipitation and RSV Timing

Epidemic onset coincided with rainy season transitions in 6 of 7 cities (Figure 3; Supplementary Figure 1). In Bogotá, Medellín, and Pasto, onset occurred during dry-to-wet transitions; in Neiva, during the first rainy season; and in Cúcuta and Cartagena, during established wet seasons. Cali was the exception: onset preceded the first rains, although peak RSV activity coincided with peak precipitation.

Figure 3.

Consolidated RSV circulation patterns by epidemic timing classification, Colombia 2022–2024. (A) Composite weekly RSV positivity across the 5 synchronized cities (Bogotá, Cali, Medellín, Neiva, and Pasto), illustrating the shared epidemic corridor, pooled median positivity, and inter-city variability. (B) Weekly RSV positivity in Cartagena, representing a delayed, highly concentrated epidemic with rapid escalation and a narrow transmission window. (C) Weekly RSV positivity in Cúcuta, illustrating sustained baseline transmission with bimodal epidemic activity and year-round RSV circulation. Summary metrics below each panel compare epidemic timing, duration, baseline positivity, epidemic burden, and peak characteristics across transmission patterns. Definitions of epidemic intensity categories and transmission pattern classifications are provided in the Methods section. Abbreviation: RSV, respiratory syncytial virus.

Inter-City Synchrony

Five cities—Bogotá, Cali, Medellín, Neiva, and Pasto—shared a common high-activity window during weeks 12–30 (March–July), defining a coherent epidemic corridor (Figure 3A) that peaked at week 21 (median positivity 39.0%) and spanned weeks 10–34. Spearman correlations confirmed high temporal synchrony among these 5 cities (all 10 pairs ρ ≥ 0.71, all P < .01). Cartagena exhibited a delayed unimodal pattern peaking at week 31 (57.5%), consistent with Caribbean rainfall seasonality (Figure 3B). Cúcuta showed bimodal circulation with a primary peak at week 22 (41.5%) and a secondary period during weeks 41–47 (Figure 3C). Neither city reached the high-synchrony threshold with any of the other 5 (ρ range −0.41 to 0.55). Core epidemic months (≥5/7 cities with epidemic-level activity) spanned April through July.

DISCUSSION

We present a multicity characterization of RSV seasonality in a tropical country spanning wide altitudinal and climatic diversity within a single health system. RSV circulated year-round in all cities, but its temporal structure varied, with 3 distinct patterns—concentrated, prolonged, and bimodal—and epidemic-period burden ranging from 83.8% to 35.5% of annual cases. Despite this heterogeneity, 5 cities shared a synchronized epidemic corridor with peak activity between March and July, while 2 diverged in timing and intensity. These findings refine the prevailing view of “year-round” RSV circulation in tropical settings by demonstrating both actionable convergence and clinically relevant sub-national variation [3, 8, 9].

Despite differences spanning nearly 2200 m in altitude and more than 20-fold in baseline positivity, 5 cities—Bogotá, Cali, Medellín, Neiva, and Pasto—shared overlapping March-through-August seasons, consistent with Pan-American analyses indicating that South American RSV activity concentrates between March and August [9, 13, 26]. These cities also exhibited synchrony in peak timing: 4 reached peak intensity within a 4-week window (weeks 20–23), with pairwise Spearman correlations confirming high temporal concordance (all ρ ≥ .71, P < .01). Cali was the exception, peaking 7 weeks earlier (week 13), for reasons that remain unclear, as Londoño-Avendaño et al. previously reported no consistent association between precipitation and RSV activity in this city [27]. The shared corridor identifies a predictable window of peak RSV activity. Whether this corridor extends to other Colombian cities not yet included in the network—particularly in underrepresented regions such as the Pacific lowlands, the Orinoquía, and the Amazon—remains unknown, and expanding molecular surveillance to these areas could reveal additional transmission patterns that further refine sub-national prevention planning.

Cartagena and Cúcuta diverged from the 5-city corridor in both timing and epidemic structure. Cartagena exhibited a delayed unimodal season (onset week 23, peak week 31) consistent with Caribbean rainfall seasonality and with the south-to-north progression of RSV epidemics across the Americas, in which Southern Cone countries initiate RSV seasons around March while Caribbean countries begin around late June [13]. This pattern also aligns with the latitudinal gradient in epidemic duration described across Chinese provinces (r = −0.69, P < .0007) [10], confirmed globally in a pre-pandemic systematic review [28]. Although Bogotá, Medellín, Cali, and Neiva exhibited variable low-intensity activity during weeks 46–52 (November–December), these late-year increases were inconsistent across years and did not meet sustained epidemic criteria. In contrast, Cúcuta showed true bimodal epidemic-level activity (>P50) with the highest baseline positivity (9.7%), the lowest epidemic-period burden (35.5%), and a secondary high-activity period during weeks 41–47 not observed elsewhere. Cúcuta was the only city where RSV positivity remained elevated year-round, with little difference between epidemic and non-epidemic periods. Multimodal or year-round RSV epidemics have been described in Caribbean island nations including Barbados, Saint Lucia, Surinam, and Trinidad and Tobago [13]. The elevated baseline in Cúcuta may reflect attenuated seasonality driven by high humidity in a tropical lowland setting [29]. Local contextual factors, including the city’s high level of cross-border population mobility with Venezuela, may also contribute by facilitating repeated viral introductions and sustained community transmission, although these mechanisms were not directly evaluated in the present study. These divergent profiles illustrate that no single seasonal window captures RSV risk across all Colombian regions, posing a distinct challenge for nationally uniform immunization calendars.

The within-country heterogeneity observed in Colombia is consistent with previous reports from other geographically diverse settings but extends current evidence from Latin America by demonstrating the coexistence of multiple transmission profiles within a single national surveillance system. Guo et al. identified 4 RSV transmission zones across 21 Chinese provinces using a moving-interval immunization framework, with epidemic duration showing a significant latitudinal gradient [10]. In Brazil, the Global Epidemiology of RSV in Community and Hospitalized Care study documented marked regional variation, with median season onset ranging from week 10 in the Midwest to week 16 in the South [11]. Similarly, Hamid et al. reported onset differences of 2–10 weeks between highland and lowland regions in Guatemala [30]. In contrast to Southern Cone countries like Chile or Argentina, where RSV seasonality is generally more homogeneous, our findings demonstrate the coexistence of synchronized, delayed, and bimodal transmission profiles within a relatively small geographic area. Our findings further suggest that the observed altitudinal and climatic gradients may contribute to these differences, with tropical lowland cities exhibiting more attenuated seasonality and sustained transmission than cooler highland settings. Together, these observations reinforce that national epidemic curves may mask clinically meaningful sub-national differences and support the need for locally tailored surveillance and prevention strategies.

The distinction between concentrated and prolonged transmission has direct implications for RSV prevention strategies. In low- and middle-income countries, seasonal monoclonal antibody approaches can achieve similar effectiveness to year-round strategies with substantially greater efficiency per dose [7]. Applying a ≤22-week framework, 2 cities met criteria for concentrated transmission (Pasto, Cartagena), 4 showed prolonged patterns (Neiva, Cali, Bogotá, Medellín), and 1 exhibited bimodal circulation (Cúcuta). Neiva marginally exceeded the duration threshold, highlighting the sensitivity of categorical classifications and suggesting that continuous metrics may better capture transmission dynamics. Similar sub-national variation has been described in China, reinforcing the need for within-country profiling to guide prevention strategies. Confirmation over additional seasons will be important to inform cost-effectiveness analyses of seasonally intensified approaches.

The consistent temporal association between rainy season onset and RSV epidemic initiation in 6 of 7 cities aligns with prior observations of respiratory infection seasonality coinciding with rainy periods in Bogotá [31–33] and with broader tropical surveillance implicating rainfall as an RSV seasonality driver [9, 34]. Proposed mechanisms include indoor crowding during rainy periods, humidity-mediated effects on viral stability, and altered population mixing patterns. Haynes et al. reported that RSV peaked during wet months in countries with high annual precipitation (Bangladesh, Guatemala, and Thailand), whereas RSV peaked during cooler months in moderately hot (China) and arid (Egypt) regions [12]. Baker et al. found that the estimated effects of precipitation and humidity on RSV transmission mirror prior results described for influenza [29]. Cali was the sole exception in our data, where epidemic onset preceded the first rains, consistent with the observation by Londoño-Avendaño et al. that RSV cases in Cali peak in the first semester without a clear association with pluviosity [27], although peak RSV activity coincided with peak precipitation, suggesting rainfall may amplify rather than trigger transmission in this setting. The 3 precipitation regimes provide a descriptive ecological framework linking climate to RSV transmission profiles. Future studies incorporating continuous meteorological variables (temperature, relative humidity, air pollution, and population density) are needed to develop predictive models for epidemic timing.

The detailed quantification of baseline positivity and off-season burden adds an operational dimension rarely captured in RSV surveillance. Bogotá maintained 7.1% baseline positivity with 28.4% of cases occurring outside the epidemic period—a volume exceeding the entire annual burden of smaller cities such as Pasto—demonstrating that substantial clinical workload persists beyond the main epidemic window. Off-season detections ranged from 16.2% (Neiva) to 64.5% (Cúcuta). In the United States, RSV seasons defined by PCR detections lasted a median of 31 weeks nationally, with some regions showing activity extending well beyond seasonal peaks [35]. During the COVID-19 pandemic, unusual inter-seasonal RSV activity was documented across the European Region during the summer of 2021 [36], and in Australia RSV epidemics shifted markedly later than historically observed after easing of restrictions [37]. Although these observations reflect disrupted epidemiology, they demonstrate that RSV transmission can persist outside traditional seasonal windows, consistent with the substantial off-season burden observed in our network.

The programmatic relevance of off-season RSV burden is illustrated by Argentina's maternal RSVpreF vaccine experience. A March–August 2024 campaign was associated with a 33.6% reduction in RSV-LRTI hospitalizations among infants [38], whereas extending immunization to January–October in 2025 increased effectiveness to 59.7% and reduced the number needed to immunize (to prevent one hospitalization) from 83.9 to 36.8 [39]. These findings suggest that aligning program duration with local circulation patterns can substantially enhance impact—particularly in areas where RSV activity extends well beyond 6 months in multiple regions, such as Colombia, where 4/7 cities showed this, and where rigid semester-based campaigns may miss a substantial proportion of at-risk infants.

Overall, these findings are consistent with year-round immunization as the appropriate baseline strategy for Colombia [40], while demonstrating that sub-national circulation profiles can meaningfully inform its implementation. The recent introduction of nirsevimab for infants in Bogotá alongside the national maternal RSVpreF vaccination strategy provides a timely context in which local seasonality data may help optimize this hybrid approach. The synchronized Andean corridor (March–July) identifies a predictable window for seasonal intensification: in resource-limited settings where year-round universal coverage is not yet achievable, concentrating monoclonal antibody deployment before and during the corridor could expand reach to catch-up cohorts, high-risk infants who missed initial dosing, and those requiring protection ahead of secondary epidemic periods. The delayed Caribbean coastal pattern argues for a shifted intensification window beginning in June, while sustained year-round delivery appears most appropriate for border cities with bimodal or attenuated seasonality where no single window captures the majority of transmission. Argentina’s experience extending its immunization window from 6 to 10 months—with effectiveness nearly doubling [38, 39]—provides early real-world evidence that matching program duration to local circulation patterns substantially enhances impact. As newly available but resource-constrained prevention tools reach Latin America, this type of sub-national profiling may prove essential for maximizing the public health return on each dose administered.

Strengths and Limitations

Strengths include 3 post-pandemic years of continuous multiplex PCR surveillance from 15 centers across 7 climatically diverse cities, a standardized percentile framework enabling comparison across settings with differing RSV intensity, and the incorporation of precipitation seasonality as an ecological dimension to epidemic timing. Limitations include hospital-based surveillance focused on severe illness, non-standardized testing indications across sites, reliance on aggregated positivity rather than population-based incidence or individual-level data on age, comorbidities, or RSV subtype, and a 3-year window—including shorter coverage for Cartagena (71 weeks), which may not capture stable long-term seasonality. Although multiplex molecular testing detected multiple respiratory pathogens, this analysis focused exclusively on RSV circulation and did not evaluate viral co-detections, temporal overlap with other respiratory viruses, or potential viral interference. These interactions, particularly with influenza, human metapneumovirus, parainfluenza viruses, and SARS-CoV-2 may influence RSV epidemic timing and intensity and warrant dedicated future analyses. The estimated >90% pediatric testing reflects clinical profiles of participating institutions but could not be verified with individual-level age data. The percentile-based thresholds, while internally consistent, have not been validated against external gold standards, and the categorical classification of transmission patterns is sensitive to cutoff selection, as illustrated by Neiva's borderline classification. Precipitation associations were qualitative and descriptive, not formally assessed with time-series or regression methods. Some cities were represented by single institutions, limiting generalizability within those settings. Large regions of Colombia were not included, and the transmission patterns described here should be viewed as hypothesis-generating for future prospective studies incorporating individual-level clinical data, continuous meteorological variables, and viral genomic surveillance.

CONCLUSIONS

RSV circulates year-round across climatically diverse Colombian cities, but temporal concentration of epidemic activity varies at the sub-national level. We identified 3 transmission patterns—a synchronized Andean corridor (Bogotá, Cali, Medellín, Neiva, Pasto) with peak activity March–July, a delayed Caribbean coastal pattern in Cartagena (June–October), and a bimodal border pattern in Cúcuta with attenuated seasonality. This heterogeneity is consistent with altitude, precipitation regime, and geographic location, and it has direct implications for RSV prevention. Although year-round immunization remains the appropriate baseline strategy, these country profiles can identify predictable windows for strategic intensification of monoclonal antibody deployment and maternal vaccination, extending coverage to catch-up cohorts and secondary epidemic periods that would otherwise remain unprotected. Expanding sub-national surveillance will support locally adapted prevention strategies as Latin America implements new RSV immunization tools. VIRAL Network LATAM continues to develop real-time analytical capacity to translate multi-city surveillance data into actionable guidance for national immunization programs across the region.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.