Public Online Search Behavior for Pediatric Urology Conditions: A Google Trends Infodemiology Analysis of Symptom Awareness versus Treatment-seeking Interest across Four Entities (2004–2024)

ABSTRACT

Background:

Pelviureteric junction obstruction (PUJO), vesicoureteric reflux (VUR), hypospadias, and undescended testis (UDT) (cryptorchidism) are among the most commonly encountered conditions in pediatric urology practice. However, caregiver online health-seeking behavior for these conditions remains poorly characterized.

Objective:

To compare symptom-seeking versus treatment-seeking online search behavior for these four conditions globally (2004–2024) and assess the impact of the coronavirus disease 2019 (COVID-19) pandemic.

Study Design:

Retrospective infodemiology study using Google Trends relative search volume data.

Methods:

Google Trends were queried globally at monthly resolution for 2004–2024. For each condition, 10 search terms – selected using prespecified criteria of clinical plausibility, lay vocabulary fidelity, and nonredundancy – were divided into symptom-seeking and treatment-seeking groups. A shared anchor term enabled cross-group normalization to yield a monthly symptom-to-treatment (S: T) ratio. Spearman correlation assessed temporal trends; the Kruskal–Wallis test with Bonferroni correction evaluated inter-entity differences.

Results:

Hypospadias showed the highest treatment-seeking volume and the lowest S: T ratio (0.003 ± 0.027), indicating persistent treatment-dominant searching throughout. UDT demonstrated a significant increasing S: T trend (ρ =0.845, P = 0.034). PUJO had the highest mean S: T ratio (0.214 ± 0.584), peaking at 1.75 in 2016. VUR returned only seven active months. COVID-19 suppressed PUJO searches while hypospadias remained stable. Inter-entity S: T differences were statistically significant (Kruskal–Wallis H = 7.89, P = 0.048).

Conclusions:

Significant differences exist in online caregiver health-seeking behavior across the four most common pediatric urological conditions, identifying condition-specific gaps in digital health literacy and targets for focused caregiver education.

INTRODUCTION

Pediatric urological conditions impose a substantial burden on children, families, and healthcare systems worldwide. Among the spectrum of conditions managed by pediatric urologists and pediatric surgeons, four stand out as the most consistently encountered in clinical practice: pelviureteric junction obstruction (PUJO), also termed ureteropelvic junction obstruction), vesicoureteric reflux (VUR), hypospadias, and undescended testis (UDT) (cryptorchidism). These four entities collectively account for the overwhelming majority of referrals to specialist pediatric urology outpatient clinics, representing the surgical and diagnostic workload that defines the day-to-day practice of the pediatric urologist.[1]

PUJO is the most common cause of neonatal hydronephrosis (incidence 1:500–1:1500 live births) and is increasingly detected prenatally.[2,3] VUR occurs in approximately 1% of children, rising to 30%–40% of those with urinary tract infection.[4] Hypospadias affects 1 in 200–300 male births and it's repair among the most frequently performed pediatric urological procedures.[5,6] UDT affects 2%–9% of full-term male neonates, with orchidopexy one of the most common operations in pediatric surgery.[7]

Despite the clinical significance of these conditions, the health-seeking behavior of caregivers in the digital domain remains incompletely characterized. In an era of ubiquitous internet access, caregivers increasingly turn to online search engines as a first step when confronted with a new diagnosis, an unfamiliar symptom, or a planned surgical procedure for their child.[8] Understanding how this online interest is distributed – whether predominantly symptom-oriented (“what is wrong with my child?”) or treatment-oriented (“what can be done?”) – has important implications for health education and the design of patient information resources.[9]

Infodemiology, the study of information and misinformation in the digital environment, has emerged as a validated approach to analyzing population-level health-seeking behavior. Google Trends, which reports relative search volume (RSV) as a normalized index (0–100), has been validated as a surrogate for real-world information-seeking in numerous healthcare domains.[10,11,12] The methodology of Mavragani and Ochoa, utilizing a shared anchor term to cross-normalize symptom-seeking and treatment-seeking query groups, permits computation of a symptom-to-treatment (S:T) ratio.[13]

Prior infodemiology studies in urology have examined single conditions, and multi-entity analyses exist for pediatric oncology and ophthalmology.[14,15] However, no study has simultaneously characterized and compared online search behavior across all four major pediatric urological conditions. Such a comparative approach identifies where public symptom-awareness lags treatment-knowledge – precisely the conditions where digital health education is most needed.

This study therefore aimed to: (i) quantify and compare symptom-seeking versus treatment-seeking RSV for all four conditions globally over 2004–2024; (ii) assess the impact of the Coronavirus disease 2019 (COVID-19) pandemic on search patterns; and (iii) determine whether statistically significant inter-entity differences in S: T ratios exist.

METHODS

Study design and ethical considerations

This retrospective infodemiology study used publicly available, anonymized aggregate data from Google Trends. No individual patient data were collected; ethical approval was not required.

Data source and extraction

Google Trends (trends.google.com) was queried using pytrends (v4.9.2; open-source Python library, available at: https://github.com/GeneralMills/pytrends, accessed 02-03-2026) in Python 3.x (Python Software Foundation, Beaverton, OR, USA) via Google Colaboratory (Google LLC, Mountain View, CA, USA). RSV data were extracted for January 2004–December 2024 at monthly resolution with worldwide geographic scope. Google Trends expresses RSV on a normalized scale of 0–100, where 100 represents peak search interest within the selected timeframe and all other values are scaled proportionally. Monthly rather than daily resolution was selected to minimize normalization artifacts associated with shorter timeframes. Mandatory inter-query delays of 45–90 s were observed between API calls in compliance with Google’s terms of service.

Search term selection and anchor normalization

For each of the four conditions, 10 search terms were selected a priori using three prespecified criteria: (i) clinical plausibility – the term reflects a symptom or treatment genuinely associated with the condition; (ii) lay vocabulary fidelity – the term reflects language a caregiver rather than a clinician would realistically use; and (iii) nonredundancy with the anchor term. Terms were independently proposed by each author and cross-validated against existing infodemiology studies in urology before final selection by consensus. All 40 search terms are listed in Table 1.

Table 1.

Search terms used for symptom-seeking and treatment-seeking query groups for each of the four pediatric urology conditions. The anchor term (bold) was included in both groups to enable cross-group normalization

Condition	Group	Search Term	Role
PUJO	Symptom	hydronephrosis child symptoms	Anchor
	Symptom	pelviureteric junction obstruction symptoms
	Symptom	kidney blockage baby signs
	Symptom	flank pain child kidney
	Symptom	PUJ obstruction symptoms child
	Treatment	hydronephrosis child symptoms	Anchor
	Treatment	pyeloplasty child
	Treatment	hydronephrosis treatment baby
	Treatment	pelviureteric junction obstruction surgery
	Treatment	laparoscopic pyeloplasty child
VUR	Symptom	vesicoureteral reflux child symptoms	Anchor
	Symptom	kidney reflux baby signs
	Symptom	urinary tract infection child recurrent
	Symptom	VUR symptoms child
	Symptom	kidney reflux urine warning signs
	Treatment	vesicoureteral reflux child symptoms	Anchor
	Treatment	vesicoureteral reflux treatment child
	Treatment	ureteral reimplantation child surgery
	Treatment	VCUG test child
	Treatment	deflux injection VUR child
Hypospadias	Symptom	hypospadias symptoms baby	Anchor
	Symptom	hypospadias signs newborn
	Symptom	urethral opening bottom penis baby
	Symptom	bent penis newborn abnormal
	Symptom	chordee baby penis curved
	Treatment	hypospadias symptoms baby	Anchor
	Treatment	hypospadias repair surgery
	Treatment	hypospadias surgery baby age
	Treatment	urethroplasty child
	Treatment	hypospadias surgery complications
UDT	Symptom	undescended testicle symptoms baby	Anchor
	Symptom	undescended testis signs infant
	Symptom	testicle not in scrotum baby
	Symptom	cryptorchidism symptoms child
	Symptom	empty scrotum baby signs
	Treatment	undescended testicle symptoms baby	Anchor
	Treatment	orchidopexy surgery child
	Treatment	undescended testicle surgery age
	Treatment	orchiopexy baby recovery
	Treatment	undescended testis treatment options

PUJO: Pelviureteric junction obstruction; VUR: Vesicoureteric reflux; UDT: Undescended testis

To enable quantitative comparison between symptom and treatment search volumes – independently normalized by Google within each query – a shared anchor term was included as the first element of both groups for each entity. This cross-group normalization method[13] allows the two groups to be referenced to a common scale. The symptom score was computed as the mean RSV of the four nonanchor symptom terms, and the treatment score as the mean RSV of the four nonanchor treatment terms. Months in which both scores were zero were excluded from the analysis.

Outcome measures

The primary outcome was the monthly S: T ratio, computed as: S: T ratio = symptom score/treatment score. Both scores were derived from Google Trends RSV data normalized to a common scale via the shared anchor term. The S: T ratio quantifies where public online interest is concentrated on the symptom–treatment spectrum: a ratio above 1.0 indicates that symptom-seeking searches predominate (the public is primarily searching for “what is wrong”); below 1.0 indicates treatment-seeking predominance (primarily searching for “what can be done”); and 1.0 represents equilibrium. Months in which the treatment score was zero were assigned a null S: T value and excluded from ratio-based analyses.

Statistical analysis

Descriptive statistics (mean ± standard deviation) were computed for symptom score, treatment score, and S: T ratio per entity. Entities with fewer than 12 nonzero monthly data points were excluded from Spearman temporal trend analysis; VUR (n = 7) was therefore excluded from this analysis. Temporal trends in annual mean S: T ratio were assessed by Spearman rank correlation (ρ) with year as an ordinal variable for the remaining three entities. The COVID-19 pandemic was operationalized as three periods: pre-COVID (2017–2019), COVID (2020–2021), and post-COVID (2022–2024). Inter-entity differences in S: T ratios were evaluated by the Kruskal–Wallis H-test with pairwise post hoc comparisons using the Mann–Whitney U-test with Bonferroni correction; all available nonzero months were included for each entity, as this nonparametric test does not require equal group sizes. All analyses were performed in Python 3.x using the scipy.stats library. A two-sided P < 0.05 was considered statistically significant.

Geographic analysis

Country-level regional search interest data were extracted using the anchor term for each condition over the full study period. The top-ranking countries by RSV were identified for each of the four entities.

RESULTS

Data extraction and active search months

Over the 21-year observation period, Google Trends returned nonzero RSV data for a total of 173 condition-months across all four entities (PUJO: n = 21 months; VUR: n = 7 months; hypospadias: n = 122 months; UDT: n = 23 months). The marked differences in active data months reflect fundamental differences in global search volume: hypospadias demonstrated by far the most consistent and sustained online interest. VUR returned extremely sparse data despite being one of the most clinically prevalent conditions, suggesting either low public search interest under the queried clinical terminology or substantial geographic variability in lay search vocabulary (e.g., “kidney reflux” rather than the acronym VUR). These temporal patterns are illustrated in Figure 1.

Figure 1.

Annual mean symptom-seeking and treatment-seeking relative search volume (RSV) for each of the four pediatric urology conditions (2004–2024). Each panel shows symptom-seeking RSV (blue) and treatment-seeking RSV (red). Grey shading indicates the coronavirus disease 2019 pandemic period (2020–2021). Note the sustained treatment-seeking dominance for hypospadias and the early symptom-seeking peak with subsequent decline for PUJO and undescended testis. UDT: Undescended testis, COVID-19: Coronavirus disease 2019

Condition-specific findings

Pelviureteric junction obstruction (PUJO)

PUJO had a mean S: T ratio of 0.214 ± 0.584, nominally below equilibrium. However, PUJO demonstrated the highest peak S: T ratio of all four conditions, reaching 1.75 in 2016 – the only year across all entities where the S: T ratio substantially exceeded equilibrium – indicating a transient but marked symptom-dominant spike. The peak global RSV for symptom-seeking searches occurred in 2005 (RSV = 25), corresponding to the early adoption phase of internet search for medical information. Treatment-seeking interest was detectable from 2004 onward and remained at low but sustained levels through 2024 (RSV ~ 0.75). The Spearman trend analysis did not reveal a statistically significant change in the annual S: T ratio (ρ =0.068, P = 0.861).

Vesicoureteric reflux

VUR returned the fewest active months (n = 7) of all four conditions, with a mean symptom score of 3.75 ± 9.38 and mean treatment score of 4.93 ± 9.00. The mean S: T ratio was 0.000 ± 0.000, reflecting that symptom scores were zero in all months where treatment scores were nonzero. The highest VUR treatment-seeking RSV occurred in 2008 (RSV = 25), coinciding with publication of major VUR management guidelines. This pattern suggests that lay search volume for VUR-specific terms is intrinsically low globally, with caregivers defaulting to broader terms such as “recurrent UTI child” that fall outside the query scope. Spearman trend analysis was not applicable due to insufficient data points (n = 7, below the prespecified threshold of 12 months).

Hypospadias

Treatment-seeking dominance defines the entire online search history of hypospadias across the 21-year observation period – a pattern unlike any of the other three conditions studied. The mean Treatment Score (8.42 ± 3.34) substantially exceeded the Symptom Score (0.66 ± 3.43), yielding a markedly low S: T ratio of 0.003 ± 0.027. By 2011–2024, treatment searches consolidated at approximately RSV 7–15 per month while symptom searches fell to near zero. The Spearman trend test did not demonstrate a statistically significant change (ρ = −0.211, P = 0.450).

Undescended testis (cryptorchidism)

Of the four conditions, UDT showed the most striking temporal shift: it is the only entity to demonstrate a statistically significant increase in S: T ratio over the 21-year study period (Spearman ρ =0.845, P = 0.034). The mean treatment score (12.41 ± 8.15) substantially exceeded the mean symptom score (2.23 ± 5.29), yielding a low overall mean S: T ratio of 0.029 ± 0.081. Treatment-seeking searches showed a rising trend in the post-2019 period (peak treatment RSV 21.00 in 2019, followed by the consistent values of 12–21 through 2024). The significant S: T increase was driven by a concurrent rise in symptom-seeking searches in 2023–2024 relative to the early years of the study period.

Coronavirus disease 2019 pandemic impact

The COVID-19 pandemic (2020–2021) differentially affected search patterns across all four conditions. PUJO showed the most dramatic impact, with near-complete suppression of searchable data during 2020–2021 (S:T pre-COVID: 0.750; COVID period not calculable; post-COVID: 0.179), consistent with widespread disruption of antenatal hydronephrosis follow-up and cessation of nonurgent outpatient referrals. Hypospadias treatment-seeking searches were notably resilient, remaining stable across all three periods (S:T pre-COVID: ~0.003, COVID: ~0.008, post-COVID: ~0.000), consistent with sustained parental engagement with planned surgical conditions despite service disruption.[16] UDT showed a modest progressive S: T decline across periods (pre-COVID: 0.179, COVID: 0.173, and post-COVID: 0.034), driven by rising treatment scores rather than falling symptom scores. VUR generated insufficient data in all three periods for meaningful COVID-period analysis.

Inter-entity comparison and spearman trends

Kruskal–Wallis analysis of monthly S: T ratios across the four entities revealed statistically significant inter-entity differences (H = 7.89, P = 0.048). The inter-entity comparison is shown in Figure 2. Pairwise Bonferroni-corrected post hoc analysis confirmed that PUJO had a significantly higher S: T ratio distribution than hypospadias. The violin plot [Figure 3] demonstrates the considerably wider distribution of PUJO’s S: T ratio, driven by several outlier months with high S: T values. Spearman trend analysis confirmed UDT as the only condition with a statistically significant temporal change in S: T ratio (ρ =0.845, P = 0.034); no significant trend was identified for PUJO (ρ =0.068, P = 0.861) or hypospadias (ρ = −0.211, P = 0.450); VUR was excluded from trend analysis (n = 7, below threshold).

Figure 2.

Inter-entity overlay of annual mean symptom-to-treatment (S:T) ratio for all four conditions (2004–2024). The dashed horizontal line represents S:T = 1.0 (equilibrium). Grey shading: coronavirus disease 2019 pandemic (2020–2021). PUJO (blue) demonstrates the highest and most variable S:T trajectory; vesicoureteric reflux (purple), hypospadias (orange), and undescended testis (green) remain near the treatment-dominant zone throughout. S:T: Symptom-to-treatment, COVID-19: Coronavirus disease 2019

Figure 3.

Violin plot of monthly symptom-to-treatment (S:T) ratio distributions for each of the four conditions (2004–2024). Individual monthly data points are superimposed (jittered). The dashed horizontal line represents S:T = 1.0 (equilibrium). The Kruskal–Wallis test result is annotated (H = 7.89, P = 0.048). PUJO demonstrates the widest distribution; hypospadias, vesicoureteric reflux, and undescended testis are tightly clustered near zero. VUR: Vesicoureteric reflux, UDT: Undescended testis, RSV: Relative search volume, S:T: Symptom-to-treatment, PUJO: Pelviureteric junction obstruction

DISCUSSION

This is, to our knowledge, the first study to simultaneously characterize and compare online health-seeking behavior across all four major pediatric urological conditions using a standardized infodemiology methodology over a 21-year global observation period. Our principal findings are threefold: first, statistically significant inter-entity differences in S: T search ratios exist (H = 7.89, P = 0.048); second, hypospadias stands out as the condition with the most established treatment-seeking online character, while PUJO is the most prone to symptom-dominant search spikes; and third, UDT is the only condition demonstrating a significant long-term increase in S: T ratio, suggesting an emerging shift in caregiver attention toward symptom recognition.

Hypospadias’s dominance of treatment-seeking searches throughout the observation period is clinically coherent. Unlike PUJO or VUR – which are often detected prenatally or incidentally, leaving parents in a passive informational role at diagnosis – hypospadias is typically identified at the newborn examination, after which parents are actively counseled and directed toward surgical planning. This clinical pathway naturally channels parents toward treatment-oriented searching. The finding that treatment-seeking RSV for hypospadias remained stable through COVID-19 (treatment score 7.73 in 2020–2021 vs. 7.90 pre-COVID) corroborates this, consistent with sustained family commitment to elective pediatric urology procedures throughout the pandemic, with waiting lists continuing to accumulate despite service disruption.[16]

The comparatively high S: T ratio of PUJO, and particularly the 2016 spike (S:T = 1.75), warrants discussion. The mid-2010s saw growing international attention to the management of antenatally detected hydronephrosis, including updated grading frameworks and postnatal surveillance protocols, which may have generated broader public awareness of kidney obstruction in children and prompted caregiver symptom-oriented searching.[3] The complete suppression of PUJO-related searches during COVID-19 is consistent with global reports of the pandemic’s impact on antenatal care and elective pediatric urology: antenatal hydronephrosis follow-up was severely disrupted, and patients with stable PUJO were actively discouraged from attending outpatient appointments.[17,18]

The significant increasing S: T trend for UDT (ρ = 0.845, P = 0.034) is a potentially important public health observation. While treatment-seeking searches for orchidopexy increased substantially over the study period, symptom-seeking searches also grew in the post-2020 period, driving the S: T increase. This may reflect growing parental concern about correctly identifying an UDT at home and distinguishing it from a retractile testis – a clinically important distinction that remains a source of confusion for many caregivers.[19,20] Targeted digital content clarifying these clinical signs may therefore represent a high-value health literacy intervention.

The sparse data returns for VUR (n = 7 active months over 21 years) reflect a fundamental limitation of keyword-based infodemiology: when a condition lacks a well-established lay vocabulary distinct from its generic symptom descriptors, Google Trends returns minimal signal. Caregivers are far more likely to search “recurrent UTI child” or “kidney infection child” than “vesicoureteric reflux.” This terminology gap is itself a clinically significant finding, indicating that digital outreach for VUR must be redesigned around lay vocabulary. Future analyses should include lay synonym terms rather than clinical nomenclature as primary queries.

Several methodological considerations merit emphasis. Google Trends returns RSV – a normalized relative index (0–100) – rather than absolute search volumes, precluding direct comparison of total query volumes across conditions. The pytrends library is a third-party wrapper and not an official Google product; its output may vary slightly between successive identical queries due to internal sampling, and it is subject to rate-limiting. Monthly rather than daily resolution was selected to minimize normalization artifacts. Search term selection was guided by prespecified criteria but retained an element of clinical judgment; future studies should validate caregiver search vocabulary empirically. A minimum data threshold was not prespecified, and VUR’s sparse data (n = 7) limits its statistical comparability, with findings interpreted descriptively. Google Trends does not capture searches on other platforms (Bing, Baidu) or voice-based queries, and data reflect individuals with internet access – introducing digital access bias. The rapid adoption of generative AI tools for health information-seeking from 2023 represents an evolving dimension of digital health literacy not captured in this analysis; future studies should examine how artificial intelligence-mediated information-seeking compares with traditional search behavior. Finally, conditions of high long-term morbidity burden such as posterior urethral valves and neurogenic bladder were not included; they represent important targets for future infodemiology study. Conditions such as phimosis, inguinal hernia, and hydrocele – which may generate higher absolute search volumes – were not included as they are managed across both general pediatric surgery and pediatric urology and represent a separate analytical focus.

The clinical implications of this work are practical. The findings support a condition-specific approach to digital health literacy: PUJO and UDT have identifiable symptom-recognition gaps that could be addressed through targeted parent-facing information resources, while hypospadias may benefit more from resources addressing surgical decision-making and timing. VUR warrants a redesigned digital outreach strategy based on lay terminology. These data can directly inform patient information leaflet development and social media health education priorities for pediatric urology societies.

CONCLUSIONS

This study demonstrates that online health-seeking behavior differs significantly across the four most common pediatric urological conditions. Hypospadias shows a stable, treatment-dominant search pattern throughout the observation period. PUJO shows the highest vulnerability to symptom-dominant search spikes and was most severely suppressed during the COVID-19 pandemic. UDT uniquely exhibits a significant long-term increase in its S: T ratio, possibly reflecting growing parental uncertainty around symptom recognition. VUR generates an insufficient signal under standard clinical nomenclature to be reliably characterized. These condition-specific findings should inform caregiver-facing digital health literacy initiatives by pediatric urology services.

AI use declaration

During the preparation of this work, the author(s) used Claude (Anthropic, claude.ai) to assist with manuscript drafting and structural editing and language refinement. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

Ethical approval

Not required. This study used publicly available, anonymized aggregate data from Google Trends and did not involve human participants, patient records, or any individual-level data.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.