ICU Pain Was Common and Persistent in a 711-Patient Cohort

Key Takeaways

• Among mixed adult ICU patients in four ICUs across two tertiary university hospitals in Finland, pain at rest was commonly documented during ICU stay.
• Average daily pain prevalence was 43.7%, and the burden did not decline during follow-up.
• Female sex, surgical admission, and higher opioid dose during ICU stay were associated with higher odds of documented pain.
• Continuous sedation and higher SAPS II scores were associated with lower odds of documented pain.

Recognizing pain in critically ill adults remains a bedside challenge because some patients can self-report, others fluctuate with delirium, and others are too deeply sedated for direct communication. In mixed medical and surgical intensive care unit (ICU) populations, clinicians still face the question of how often pain at rest persists across admission rather than clustering around procedures or early stabilization. Repeated pain assessments and analgesic-use tracking across participating ICUs in Finland were used to examine that burden over time.

In the prospective multicenter cohort, investigators enrolled 711 critically ill adults in four mixed intensive care units (ICUs) at two tertiary university hospitals in Finland between 2018 and 2020 and followed them during ICU stay for up to 14 days. The primary outcome was daily prevalence of at least moderate pain at rest, defined by the highest pain score recorded that day and measured with the Numeric Rating Scale (NRS), Verbal Rating Scale (VRS), or Critical Care Pain Observation Tool (CPOT) according to each patient's ability to communicate. At least moderate pain meant NRS 4 or higher, VRS moderate pain or higher, or CPOT 3 or higher, and assessments were performed at minimum every 2 hours while missing assessments remained in the prevalence denominator for a conservative estimate.

At least moderate pain at rest during ICU stay occurred in 76.4% of patients at some point during admission, with higher rates in surgical than medical patients. Average daily prevalence was 43.7%, peaked at 55% on ICU day 2, and did not decline during follow-up. In adjusted analyses, female sex, surgical admission, and higher opioid dose during ICU stay were associated with higher odds of documented pain, whereas continuous sedation and higher Simplified Acute Physiology Score II (SAPS II) scores were associated with lower odds. Opioids and paracetamol were the most commonly used analgesics, and opioid use remained frequent across ICU days.

Recruitment stopped early during the COVID-19 pandemic before the planned sample size was reached, and exclusion of neurosurgical patients and those not expected to survive more than 48 hours may limit generalizability to sicker ICU populations. Variation in pain-assessment frequency and incomplete documentation could have affected prevalence estimates, and pooling NRS, VRS, and CPOT into a dichotomized pain outcome may have reduced measurement precision. The observed link between opioid administration and pain was observational and should not be interpreted as cause and effect. Because the cohort came from tertiary ICUs in Finland, the findings describe that setting rather than directly establishing applicability to other health systems.

In this mixed adult ICU cohort, pain at rest remained common across ICU stay, and investigators identified patient and treatment variables associated with higher or lower odds of documented pain. The findings describe prevalence patterns and correlates within ICU pain monitoring and documentation rather than why pain persisted. They also do not show whether any analgesic strategy changed outcomes.

Clinician Questions

How was at least moderate pain defined in critically ill adults during ICU stay?

At least moderate pain at rest was defined as an NRS score of 4 or higher, a VRS rating of moderate pain or higher, or a CPOT score of 3 or higher, with the tool selected according to each patient's ability to communicate. Pain was assessed at minimum every 2 hours, and missing assessments remained in the denominator for prevalence estimates so that documented pain burden reflected the full observation window.

Which ICU patients were included in this pain cohort, and which groups were not represented?

The cohort included adults needing ICU care for critical illness or after major elective surgery in four mixed ICUs at two tertiary university hospitals in Finland. Groups not represented included patients who were pregnant, had major traumatic brain injury or other neurosurgical conditions, were estimated to survive less than 48 hours, had major cognitive or psychiatric disorders that limited communication, had a permanent inability to consent, or did not speak Finnish or Swedish.

Why might continuous sedation be linked to lower odds of documented pain in ICU patients?

Continuous sedation was associated with lower odds of documented pain in ICU patients, but the study treated that as an observational finding tied to documented pain rather than proof that sedation reduced pain itself. As the authors noted, sedation can make pain harder to assess and may alter how pain is expressed or recorded, which offers one explanation for the lower documented burden.