Every Day of Waiting Counts: Decade-Long Study Reveals Why Children Miss Specialist Appointments

When a child is referred to a specialist, the clock starts ticking long before the family ever reaches the clinic door. A new decade-spanning audit from Singapore’s largest children’s hospital has now quantified, with unusual statistical precision, exactly how much that waiting time matters — and the answer is striking. For every additional 30 days a child waits between referral and appointment, the odds that the family simply never shows up rise by roughly 70 percent. The finding, drawn from more than 156,000 pediatric referrals over ten years, offers one of the clearest quantitative portraits yet of the hidden attrition that erodes specialist outpatient care in a high-income health system.

The study, published in BMC Pediatrics by Xi Yar Ong and Bao Bei Kelly Low of the General Pediatrics Service at KK Women’s and Children’s Hospital, is a retrospective clinical audit of every consecutive referral to the hospital’s General Pediatrics Specialist Outpatient Clinic between 1 January 2015 and 31 December 2024. Because Singapore operates a primary care–led referral system — meaning children generally reach specialist services only after an assessment by a family doctor, a public polyclinic physician, or an emergency department — the referral letter is the gateway to nearly all specialist pediatric care. Understanding what happens after that gateway is opened, the authors argue, is essential for service planning and the efficient use of healthcare resources in a country with declining birth rates and rising demand for subspecialist expertise.

The raw numbers alone tell a story of scale. Over the ten-year window, 156,161 referrals were recorded at the single clinic. The median age of the referred children was 4.2 years, with an interquartile range stretching from 1.2 to 9.5 years — a reminder that pediatric referrals cluster heavily around toddlers and preschoolers, the ages when developmental concerns, recurrent infections, and growth issues most often prompt a doctor to seek a specialist opinion. Referral sources were far from evenly distributed: public healthcare polyclinics accounted for 43.9 percent of all referrals, while the children’s emergency department contributed 32.8 percent. Together, these two public-system channels generated more than three-quarters of the specialist workload, underscoring how tightly Singapore’s tertiary pediatric services are coupled to its front-line public infrastructure.

The median referral-to-appointment time was 21 days, with an interquartile range of 7 to 38 days. That figure may sound respectable by international standards, but the study’s central insight is that waiting time is not a neutral variable — it is an active driver of failure. Using multivariable logistic regression to disentangle the effects of waiting time, referral source, and patient characteristics, the researchers found that longer referral-to-appointment intervals were independently associated with non-attendance, with an adjusted odds ratio of 1.018 per day (95 percent confidence interval 1.017 to 1.019; p less than 0.001). Compounded over a month, that daily hazard translates into approximately 70 percent higher odds of a missed appointment for every 30-day increase in waiting time. In other words, the queue itself is quietly dissolving a substantial fraction of the demand it exists to serve.

Attendance overall was just 63.0 percent, meaning more than one in three referred children never arrived at the specialist clinic to which they had been directed. This is a figure with real clinical consequences. A child referred for suspected developmental delay, chronic allergy, or an undiagnosed cardiac murmur who fails to attend does not simply vanish from the system’s accounting — the underlying condition may progress undetected, the referring physician may assume specialist care is underway, and the appointment slot that went unused represents wasted diagnostic and clinical capacity that another child could have occupied. In a tertiary centre where specialist time is among the scarcest resources in the health system, a 37 percent non-attendance rate is not a statistical footnote; it is a structural leak in the pipeline of care.

The regression analysis also illuminated which patients were most at risk of falling through that leak. Referrals carrying subsidized status were associated with roughly double the odds of non-attendance (adjusted odds ratio 2.007, 95 percent confidence interval 1.929 to 2.087), and referrals originating from the children’s emergency department carried a similarly elevated risk (adjusted odds ratio 2.070, 95 percent confidence interval 2.018 to 2.214). By contrast, self-initiated referrals — families who sought the appointment on their own accord — were dramatically protective, cutting the odds of non-attendance to less than a third of baseline (adjusted odds ratio 0.304, 95 percent confidence interval 0.287 to 0.322). The pattern is intuitively coherent: emergency department referrals often capture transient, acute-phase concerns that resolve or feel less urgent once the crisis passes, while subsidized patients may face greater logistical, financial, or informational barriers to completing a journey through a multi-tiered system.

Beyond identifying risk factors, the team took the analysis a step further with receiver operating characteristic (ROC) analysis, treating referral-to-appointment time as a potential screening variable for predicting non-attendance. The model showed modest discriminatory performance, with an area under the curve of 0.669 (95 percent confidence interval 0.666 to 0.672). An exploratory threshold of 27.5 days emerged as a potential decision point: waiting times beyond this cutoff identified non-attendance with a sensitivity of 58.3 percent and a specificity of 71.3 percent. The authors are careful to frame this as an exploratory assessment rather than a validated clinical tool, but the implication is provocative — a hospital could, in principle, flag appointments predicted to be at high risk of no-show and deploy targeted reminders, rescheduling, or overbooking strategies before the slot is lost.

What makes the Singapore case scientifically valuable is its setting. In many health systems, long waiting lists and high non-attendance are entangled with underfunding, workforce shortages, or fragmented record-keeping, making it hard to isolate the causal contribution of waiting time itself. Singapore is a high-income country with a structured, digitally connected referral pathway and a single dominant tertiary children’s hospital, which allowed the researchers to audit a complete, consecutive ten-year cohort with administrative data of consistent quality. The consistency of the waiting-time effect — visible even after adjustment for referral source and subsidy status — suggests that the relationship between delay and disengagement is not merely a proxy for poverty or acuity, but a genuine behavioral and logistical phenomenon that operates even in a well-resourced system.

The study also carries a methodological honesty worth noting. Because it is a retrospective audit of administrative data, it can identify associations but not definitively explain them; the authors did not have access to the reasons families gave for missing appointments, nor to clinical outcomes for the children who did not attend. The exploratory ROC threshold of 27.5 days would require prospective validation before any operational deployment. The authors state no competing interests and received no external funding, and the work was conducted under SingHealth Central Institutional Review Board policy as a clinical audit and service evaluation, with formal ethics review and informed consent not required because the data were administrative in nature.

For health planners, the takeaways are concrete. Reducing referral-to-appointment intervals below the exploratory 27.5-day threshold could plausibly recover a meaningful share of lost attendance, and the identified high-risk profiles — subsidized referrals and emergency department origins — offer a rational basis for prioritizing reminder systems, SMS confirmations, caregiver education at the point of referral, or fast-track pathways for conditions that cannot safely tolerate a missed specialist review. As pediatric services worldwide grapple with post-pandemic backlogs and growing referral volumes, this Singaporean dataset delivers a deceptively simple message backed by 156,161 data points: in outpatient care, time is not just a convenience metric. Every day of waiting measurably increases the chance that a child who needs specialist care never receives it, and systems that treat waiting time as a modifiable risk factor — rather than an inevitable queue — stand to reclaim both appointments and health outcomes that are currently slipping through the cracks.

Subject of Research: Pediatric outpatient referral characteristics and clinic non-attendance at a Singapore tertiary children's hospital

Article Title: Pediatric outpatient referrals to a tertiary centre in Singapore: a 10-year review of referral characteristics and attendance patterns

Article References: Ong, X. Y., & Low, B. B. K. (2026). Pediatric outpatient referrals to a tertiary centre in Singapore: a 10-year review of referral characteristics and attendance patterns. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07714-5

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07714-5

Keywords: pediatrics, non-attendance, referral patterns, outpatient care, Singapore, waiting time, tertiary hospital, health services research, primary care, emergency department, logistic regression, healthcare access