Anxiety and Intolerance of Uncertainty: Worry, Avoidance, and Decision-Making
Updated: Oct 8
Author: Ukrainian Psychological Hub · Published: October 1, 2026 · Editorial Policy
Uncertainty is unavoidable: a test result may be pending, a message may go unanswered, a decision may have several plausible outcomes, or a bodily sensation may have more than one possible explanation. For many people, that gap in knowledge is uncomfortable but manageable. Intolerance of uncertainty describes a stronger tendency to experience not knowing as aversive, threatening, or difficult to endure and to respond with worry, avoidance, repeated checking, reassurance seeking, delay, or other attempts to obtain certainty.
Intolerance of uncertainty, usually abbreviated IU, began as a construct developed in research on worry and generalized anxiety disorder. Three decades of research now show a broader pattern: IU is strongly related to worry and GAD, yet it also appears across other anxiety presentations and in conditions outside the anxiety-disorder chapter. A 2026 state-of-the-science review describes IU as both transdiagnostic and trans-situational while emphasizing that its definition and measurement remain active areas of debate. Dugas, Koerner, and Freeston (2026)
IU is not a diagnosis, and having difficulty with uncertainty does not mean that someone has an anxiety disorder. It is better understood as a psychological construct that may help explain why uncertainty becomes especially distressing for some people and why certain responses—such as persistent worry or repeated certainty-seeking—can become self-reinforcing.
Quick answer: what is intolerance of uncertainty?
Intolerance of uncertainty is a tendency to react negatively to situations in which important information is missing, outcomes are unclear, or certainty is impossible. The reaction can be emotional, such as anxiety or frustration; cognitive, such as repetitive “what if?” thinking; or behavioral, such as postponing a decision, checking repeatedly, avoiding an uncertain situation, or trying to obtain more reassurance.
The construct is closely related to worry but is not identical to worry. Worry is a chain of thoughts about possible future problems; IU describes how difficult the state of not knowing feels and how a person tends to respond to it. The latest large preregistered meta-analysis found a strong association between IU and worry (r = .59) across 138 independent samples, while also showing that the strength of IU’s association with anxiety differs across clinical phenotypes. Akbari et al. (2026)
Intolerance of uncertainty is a construct, not a DSM-5-TR or ICD-11 diagnosis
Neither DSM-5-TR nor ICD-11 lists “intolerance of uncertainty” as a standalone mental disorder. The American Psychiatric Association describes anxiety disorders as distinct diagnoses involving excessive fear, worry, or anxiety that cause clinically meaningful distress or interference, while the World Health Organization’s ICD-11 clinical manual provides disorder-specific diagnostic requirements. American Psychiatric Association and WHO ICD-11 CDDR
This distinction matters because a person can score high on an IU questionnaire without meeting criteria for any disorder, and a person can have an anxiety disorder without IU being the most important maintaining process. The diagnostic question depends on the full pattern of symptoms, duration, impairment, context, medical and substance factors, and the specific criteria of the disorder being considered.
For a broader clinical overview, see Anxiety Disorders: Symptoms, Types, Diagnosis, and Treatment.
What does intolerance of uncertainty look like in everyday life?
IU is not visible as one single behavior. Two people can feel equally uncertain and respond in opposite ways. One may overresearch every option and delay choosing; another may avoid the decision entirely. One may repeatedly ask other people for reassurance; another may privately rehearse worst-case possibilities. The common feature is that uncertainty itself becomes difficult to tolerate and behavior begins to organize around reducing, escaping, or controlling that uncertainty.
Worry and mental problem-solving
Worry can function like an attempt to solve a future problem before enough information exists to solve it. The mind generates scenarios, searches for warning signs, compares possibilities, and tries to reach a feeling of certainty. When the underlying question cannot yet be answered, more thinking may produce more branches rather than resolution. This is one reason uncertainty and worry can form a loop: uncertainty prompts worry, and worry keeps attention focused on uncertain threats.
Avoidance and postponement
Uncertainty may lead to avoiding situations in which an outcome cannot be guaranteed: sending an application, making a medical appointment, entering a social situation, traveling, beginning a difficult conversation, or making a choice that cannot be fully reversed. Postponement can bring immediate relief, which makes it more likely to be used again even when the long-term cost is growing.
Avoidance is a broad behavior rather than a diagnosis. For the learning processes that can make avoidance self-reinforcing, see Anxiety and Avoidance: How Avoidance Maintains Fear.
Checking, reassurance, and information seeking
Some responses to uncertainty are approach-like rather than avoidant: checking a source again, asking another person, comparing one more review, searching symptoms, rereading a message, or collecting more information before deciding. These behaviors are not automatically problematic. Their function and scale matter. Information gathering is useful when new information can meaningfully improve a decision; it becomes less useful when the search continues mainly because certainty still does not feel complete.
Experimental work supports a relationship between prospective IU and information-seeking under uncertainty, although behavior depends on how relevant the information is to the perceived threat. Bartoszek et al. (2022)
Why uncertainty can become an anxiety amplifier
Anxiety is oriented toward possible future threat. Uncertainty leaves open several questions that anxious prediction cares about: Will something bad happen? How likely is it? When will it happen? Could I cope? Would I notice it in time? The less tolerable the unanswered state feels, the stronger the pressure can become to predict, control, avoid, or resolve it.
That does not mean uncertainty itself is dangerous. It means that the appraisal of uncertainty can change the emotional and behavioral response to an ordinary feature of life. A person who treats incomplete information as unacceptable may experience more threat even when objective risk is unchanged.
The broader clinical boundary is important. The World Health Organization notes that anxiety disorders involve fear or worry that is intense, excessive, difficult to control, persistent, and impairing; ordinary anxiety can occur without a disorder. WHO anxiety disorders fact sheet (2026)
How strong is the evidence linking intolerance of uncertainty to anxiety?
Established evidence: the association is robust
The association between IU and anxiety-related symptoms has been replicated extensively. A 2019 meta-analysis synthesized 181 studies with 52,402 participants and found moderate associations between IU and symptom dimensions spanning generalized anxiety, social anxiety, panic, agoraphobia, OCD, depression, and eating disorders. Methodological factors changed some estimates, but the overall transdiagnostic pattern was substantial. McEvoy et al. (2019)
The 2026 preregistered meta-analysis provides a more differentiated picture. IU was most strongly associated with GAD (r = .54) and nonspecific anxiety (r = .48), followed by PTSD (r = .46), social anxiety (r = .45), OCD (r = .41), and panic disorder (r = .27). Those estimates describe associations, not diagnostic equivalence and not proof that IU causes each condition. Akbari et al. (2026)
Established evidence: IU is especially closely linked with worry and GAD
IU was originally developed in models of excessive worry, and its relationship with GAD remains especially strong. That does not make IU specific to GAD, but it helps explain why uncertainty-focused models and treatments have been developed most extensively for GAD.
An earlier meta-analysis already challenged a GAD-only interpretation: with the GAD-oriented definition of IU, mean correlations were .57 with GAD symptoms, .53 with major depression symptoms, and .50 with OCD symptoms. Gentes and Ruscio (2011)
For the diagnostic territory of generalized anxiety disorder itself, see Generalized Anxiety Disorder: Symptoms, Causes, Diagnosis, and Treatment.
More tentative evidence: IU as a causal mechanism
Association is easier to establish than causation. A systematic review focused specifically on experimental manipulation and temporal precedence found the strongest causal support for anxiety-related difficulties, but the evidence base was small and inconsistent enough that firm transdiagnostic causal conclusions were not justified. Rosser (2019)
Experimental work has shown that manipulating IU can change worry in controlled settings, which supports a mechanistic role, but laboratory effects do not prove that IU is the sole or primary cause of chronic anxiety in everyday life. Ladouceur et al. (2000)
How intolerance of uncertainty and worry can reinforce each other
A useful model is a sequence rather than a label. An uncertain situation appears. The person evaluates the lack of certainty as threatening or unacceptable. Worry starts in an effort to anticipate possibilities and prevent surprise. The worry generates more hypothetical outcomes, which creates more uncertainty to solve. The person may then seek reassurance, check, avoid, overprepare, or postpone. Relief follows temporarily, but the underlying capacity to remain with uncertainty may receive little practice.
The 2026 meta-analytic structural model is consistent with this architecture: IU had a strong relationship with worry, and worry partially mediated the relationship between IU and anxiety outcomes. The authors describe IU as a more distal factor and worry as a more proximal cognitive process. Because the data are meta-analytic and largely drawn from observational measures, “mediation” here should not be read as proof of a single causal chain in every person.
For a symptom-level discussion of worry, doubt, attention, and concentration, see Cognitive Symptoms of Anxiety: Worry, Threat Monitoring, Doubt, and Concentration.
Intolerance of uncertainty and avoidance
Avoidance can reduce contact with uncertainty before it is experienced. A person may avoid making a decision, entering a setting, asking a question, opening a message, or starting a task because doing so creates an uncertain outcome. This can produce short-term relief while leaving the feared uncertainty unresolved.
The relationship can also operate through subtle safety behaviors. A person may technically enter the uncertain situation but rely on repeated checking, extensive preparation, a companion, a backup plan, constant monitoring, or reassurance to make the situation feel tolerable. The behavior may be sensible in one context and excessive in another; its meaning depends on realistic risk, frequency, flexibility, and whether it is becoming necessary for ordinary functioning.
IU and avoidance should therefore be kept conceptually separate. IU concerns the response to uncertainty. Avoidance is one possible behavior used in response to threat, discomfort, or uncertainty. Avoidance also occurs in many contexts that have nothing to do with IU.
Intolerance of uncertainty and decision-making
Decision-making is an especially clear setting for IU because decisions often require action before certainty is available. A choice can be reasonable without being guaranteed. For someone high in IU, the residual possibility of being wrong may feel like evidence that the decision is unfinished.
Indecisiveness and repeated deliberation
A 2024 experience-sampling study followed 247 people as they reported real-life decisions repeatedly during the day. Higher dispositional and situational IU were associated with more indecisiveness and more decision-related safety behavior. Problematic safety behavior earlier in the sampling period also predicted later indecisiveness through IU. Appel et al. (2024)
A 2026 preregistered experiment in people with anxiety or obsessive-compulsive disorders manipulated situational IU and found a significant indirect effect on indecisiveness through situational IU. The effects were small, and the mixed-diagnosis sample limits disorder-specific conclusions, so the study is best read as preliminary causal evidence rather than a settled account of clinical indecision. Appel, Mattes, and Gerlach (2026)
Confidence, speed, and changing a decision
IU does not simply make every decision slower. In a 2014 high-risk decision task, participants high in IU became less confident across repeated decisions, while response time did not significantly differ between high- and low-IU groups. High-IU participants also showed a nonsignificant tendency to be less likely to revise a decision after new information appeared. Jensen et al. (2014)
That pattern matters because “decision difficulty” can mean several things: taking longer, feeling less confident, collecting more information, avoiding commitment, changing choices repeatedly, or becoming rigid once a choice is made. IU may relate differently to each component.
Information seeking can help—and can become a certainty strategy
Good decisions require information. The clinical question is whether the next search is likely to improve the decision or is mainly an attempt to remove the normal residue of uncertainty. When no realistic amount of information can provide a guarantee, continued searching can become part of the problem rather than a path to resolution.
Prospective and inhibitory intolerance of uncertainty
The widely used 12-item Intolerance of Uncertainty Scale (IUS-12) was developed with two factors often described as prospective and inhibitory IU. Prospective IU reflects a desire for predictability and distress about uncertain future events; inhibitory IU reflects difficulty acting or functioning when uncertainty is present. Carleton, Norton, and Asmundson (2007)
The distinction can be clinically intuitive: one person may respond to uncertainty by searching, planning, and asking questions, while another may freeze, postpone, or avoid. But the factor structure of IU measures remains an active psychometric question, and the 2026 state-of-the-science review cautions against treating any single measurement model as the final definition of the construct.
The IUS-12 is a measure of a psychological construct. It is not a diagnostic test for GAD, an anxiety disorder, OCD, PTSD, or any other clinical condition. A score can describe relative levels of IU in a research or assessment context; it cannot by itself establish a diagnosis.
Is intolerance of uncertainty specific to generalized anxiety disorder?
No. The strongest contemporary evidence supports a transdiagnostic interpretation with meaningful differences in strength across conditions. GAD and worry show particularly strong relationships with IU, but elevated IU is also found in social anxiety, panic-related presentations, depression, OCD research, trauma-related research, and other areas.
“Transdiagnostic” does not mean that these conditions are interchangeable. Social anxiety disorder is organized around social-evaluative fear; panic disorder centers on recurrent unexpected panic attacks and their consequences; agoraphobia has its own situational fear structure; OCD and PTSD belong to different diagnostic groupings. IU may be one process that appears across several problems while each condition retains its own diagnostic criteria and treatment evidence.
For GAD specifically, the Hub’s etiological overview places IU alongside other risk and maintenance processes rather than treating it as a single cause: What Causes Generalized Anxiety Disorder? Risk Factors and Mechanisms.
OCD and PTSD require separate diagnostic boundaries
IU research often includes obsessive-compulsive and trauma-related symptoms because uncertainty can be relevant to checking, doubt, threat appraisal, and attempts to feel safe. That research does not reclassify OCD or PTSD as anxiety disorders. DSM-5-TR and ICD-11 organize them in separate diagnostic families, and their core symptom structures differ from those of anxiety disorders.
This distinction is especially important when uncertainty leads to checking or reassurance seeking. Those behaviors can occur in ordinary anxiety, GAD, health concerns, OCD, trauma-related conditions, and other contexts. Their presence does not identify a diagnosis without the larger pattern.
Normal anxiety about uncertainty versus a clinical problem
Uncertainty is often genuinely stressful. Waiting for a medical result, an immigration decision, an examination grade, a job offer, or news about a loved one can produce substantial worry without implying a mental disorder. The intensity of emotion must be understood in context.
Clinical concern rises when fear or worry is persistent or excessive for the situation, difficult to control, and associated with significant distress or interference. Avoidance, repeated reassurance, checking, or indecision may become clinically relevant when they consume substantial time, restrict important activities, or make ordinary decisions increasingly dependent on certainty rituals.
For the boundary between ordinary worry and GAD, see Generalized Anxiety Disorder vs Normal Worry.
How clinicians assess intolerance of uncertainty
IU may be explored through conversation, behavioral examples, and self-report measures. A clinician might ask what kinds of unknowns are most difficult, what predictions arise, what the person does to obtain certainty, how long relief lasts, and what those strategies cost. The assessment is functional: it looks at patterns and consequences rather than treating one behavior as proof of a disorder.
The original English Intolerance of Uncertainty Scale showed that IU includes reactions such as finding uncertainty stressful, feeling unable to act, viewing uncertain events negatively, and wanting to avoid them. Buhr and Dugas (2002)
Questionnaire results should be interpreted in context. They can support formulation or research, but they do not replace a diagnostic interview, medical review when indicated, assessment of impairment, or disorder-specific criteria.
For the wider diagnostic process, see How Is Anxiety Diagnosed? Assessment, History, and Differential Diagnosis.
Can intolerance of uncertainty change?
Yes, IU can change, but the treatment literature needs to be read carefully. Evidence that a therapy reduces IU does not automatically prove that IU was the causal mechanism of symptom improvement, and findings from GAD should not be generalized without evidence to every anxiety disorder.
Meta-analytic evidence across anxiety-related treatments
A 2023 meta-analysis of 28 randomized controlled trials found a large treatment effect on IU compared with control conditions (g = .89). Improvements in IU corresponded with improvements in symptom severity and statistically accounted for 36% of the variance in symptom improvement. Miller and McGuire (2023)
A 2024 systematic review and meta-analysis of 22 studies also found a large pooled effect versus passive controls, but effects versus active controls were not significant and study quality moderated results. The authors concluded that how reliably psychotherapy changes IU—and which treatment components are responsible—remains unresolved. Näsling et al. (2024)
The strongest disorder-specific evidence is in GAD
For adults with GAD, a 2023 systematic review and meta-analysis of 26 studies (1,199 participants) found large pre-to-post and follow-up improvements in IU and worry. CBT that directly targeted IU outperformed general CBT for IU and worry immediately after treatment, although that advantage was not maintained at follow-up. Wilson, Abbott, and Norton (2023)
A 2026 open-access study tested a brief two-session videoconference intervention using behavioral experiments focused on uncertainty in adults with GAD and reported reductions in IU, worry, tension, safety behaviors, and avoidance. It is promising primary evidence, not a replacement for the broader treatment literature. Wilson et al. (2026)
For a full disorder-specific treatment review, see CBT for Generalized Anxiety Disorder: How It Works and What Evidence Shows.
How therapy can target intolerance of uncertainty
Uncertainty-focused CBT does not try to make the future certain. It works on the relationship to uncertainty: the beliefs attached to not knowing, the behaviors used to neutralize uncertainty, and the capacity to act when a reasonable amount of information is available.
Behavioral experiments
Behavioral experiments test a prediction in a planned and sufficiently safe situation. Examples might examine whether making a low-stakes decision without exhaustive checking leads to the feared consequence, whether uncertainty remains permanently unbearable, or whether anxiety can rise and fall without additional reassurance. The purpose is to gather information from experience, not to force risk-taking.
Reducing certainty-seeking when it is maintaining the cycle
Treatment may examine checking, repeated information gathering, overpreparation, reassurance seeking, or postponement when those behaviors are serving mainly to eliminate normal uncertainty. The goal is not to remove sensible verification. It is to distinguish proportionate problem solving from repeated safety behavior that no longer improves the decision.
Problem solving versus hypothetical worry
A practical distinction in GAD-focused work is whether a concern concerns a current, solvable problem or a hypothetical future possibility. Solvable problems can be translated into concrete actions. Hypothetical worries often cannot be resolved now because the information required does not exist yet. Recognizing that difference can reduce the temptation to treat every uncertain possibility as a problem that must be solved immediately.
Acting with incomplete information
Decision-making work can focus on what level of evidence is sufficient for the stakes involved, rather than on achieving a feeling of total certainty. Low-stakes choices may tolerate a lower information threshold; high-stakes medical, legal, financial, or safety decisions appropriately require more information and professional input. Tolerance of uncertainty is not carelessness.
Practical ways to respond to uncertainty without turning the exercise into another certainty ritual
For ordinary low-risk situations, several principles from uncertainty-focused CBT can be translated into general psychological skills. They are educational strategies rather than a diagnosis or individualized treatment plan.
Name the unanswered question
Instead of responding to a global feeling that “something is wrong,” identify the uncertainty precisely: “I do not know whether they will reply today,” “I cannot know which option will be perfect,” or “I have enough information to choose, but I cannot guarantee the outcome.” A precise uncertainty is easier to evaluate than diffuse threat.
Separate information that changes the decision from information that only soothes
Ask whether one more search, check, or conversation is likely to add decision-relevant evidence. If the answer is no, the next step may be an attempt to feel certain rather than an attempt to become better informed.
Use proportional decision rules
A reversible, low-cost decision does not need the same evidence threshold as a high-stakes, irreversible decision. Matching the amount of deliberation to the real stakes can reduce the tendency to treat every choice as if one mistake would be catastrophic.
Allow a residue of doubt
Many good decisions end with some uncertainty. Practicing action while a small amount of doubt remains can be more realistic than waiting for a complete internal feeling of certainty that may never arrive.
Keep real safety rules intact
Tolerance of uncertainty never requires ignoring medical red flags, bypassing professional advice, violating safety procedures, accepting coercion, or entering objectively dangerous situations. The target is excessive certainty-seeking in situations that are safe enough, not sensible risk management.
Intolerance of uncertainty in children and adolescents
IU is also studied in younger populations, but developmental context matters. Children and adolescents have different levels of autonomy, cognitive development, dependence on caregivers, and exposure to uncertainty. A young person’s repeated questions or reluctance to decide may reflect normal development, anxiety, family accommodation, neurodevelopmental differences, stress, or other factors.
A recent clinical study of adolescents found higher IU in those with GAD than in low-anxiety peers and higher levels than in a smaller group with other anxiety disorders, supporting a meaningful relationship while not establishing IU as a diagnostic marker. Cognitive aspects of GAD in adolescents (2025)
Assessment in young people therefore needs the same distinction used in adults: IU is a construct that may contribute to a formulation; diagnosis depends on the full disorder-specific clinical picture.
Common misunderstandings about intolerance of uncertainty
“If I dislike uncertainty, I have an anxiety disorder”
Disliking uncertainty is common. A clinical disorder requires a broader pattern of symptoms and impairment. IU can be high without a diagnosis and can be relevant in multiple conditions.
“Worry gives me certainty”
Worry can create a sense of active preparation, but thinking through every possible future outcome cannot guarantee what will happen. In some situations it improves planning; in others it multiplies hypothetical problems that cannot yet be solved.
“Avoidance proves the situation is dangerous”
Relief after avoidance shows that the behavior reduced distress. It does not, by itself, show that the avoided situation was objectively dangerous. The actual risk still has to be assessed independently.
“The IUS-12 tells me whether I have GAD”
It does not. The IUS-12 measures intolerance of uncertainty. It is not a GAD diagnostic instrument and cannot replace clinical assessment.
“The goal is to stop caring about outcomes”
The goal in uncertainty-focused work is psychological flexibility: gathering enough information for the stakes, making reasonable plans, and remaining able to act when some uncertainty remains. Caring about outcomes and tolerating uncertainty can coexist.
When to consider professional assessment
Professional assessment may be useful when worry, avoidance, checking, reassurance seeking, or indecision is persistent, difficult to control, time-consuming, or interfering with work, study, relationships, health care, sleep, travel, or ordinary responsibilities. Assessment is also useful when it is unclear whether the problem is best explained by an anxiety disorder, OCD, trauma-related symptoms, depression, a neurodevelopmental condition, a medical issue, substance or medication effects, or another factor.
A clinician can evaluate the disorder-specific pattern rather than treating IU as a diagnosis. Treatment can then target the processes that are actually relevant to the person’s presentation.
What the current evidence supports
The strongest conclusion is that intolerance of uncertainty is a well-supported psychological construct with a robust relationship to worry and anxiety, especially GAD. It is also meaningfully related to several other forms of psychopathology, making a purely GAD-specific interpretation too narrow.
The next strongest conclusion is that IU can change during psychological treatment and that directly targeting uncertainty can be useful in GAD. Evidence across broader diagnostic groups is promising but more heterogeneous.
The more cautious conclusion concerns causality. Experimental, longitudinal, and treatment studies support the possibility that IU can help maintain anxiety and indecision, yet correlations remain more abundant than causal evidence, effects vary by diagnosis and measure, and bidirectional relationships are plausible. Current science supports IU as an important mechanism to assess—not as a universal explanation for anxiety.
Frequently asked questions
What is intolerance of uncertainty in anxiety?
It is a tendency to experience not knowing as especially aversive or threatening and to respond with worry, distress, avoidance, checking, reassurance seeking, indecision, or other attempts to obtain certainty.
Is intolerance of uncertainty a mental disorder?
No. It is a psychological construct, not a standalone DSM-5-TR or ICD-11 diagnosis.
Is intolerance of uncertainty the same as anxiety?
No. Anxiety is an emotional and physiological response to anticipated threat. IU describes how a person tends to react to uncertainty, which can influence anxiety but is not identical to it.
Why does uncertainty cause worry?
When uncertainty is appraised as threatening or unacceptable, worry can become an attempt to predict outcomes, prevent surprise, and mentally solve possibilities that do not yet have enough information to be solved.
Can intolerance of uncertainty lead to avoidance?
It can contribute to avoidance or postponement when uncertain outcomes feel too difficult to tolerate. Avoidance also has many other causes, so the relationship should not be assumed in every case.
Why can uncertainty make decisions difficult?
Most decisions leave some possibility of error. Higher IU is associated with indecisiveness, reduced confidence in some tasks, and more decision-related safety behavior, although effects differ across studies and forms of decision-making.
Does the IUS-12 diagnose generalized anxiety disorder?
No. The IUS-12 measures intolerance of uncertainty. A diagnosis of GAD requires a clinical assessment of the full symptom pattern, duration, impairment, and differential diagnosis.
Is intolerance of uncertainty only a GAD problem?
No. Its association is especially strong with worry and GAD, but meta-analyses show relationships with social anxiety, panic-related symptoms, depression, OCD, trauma-related symptoms, and other difficulties.
Can therapy reduce intolerance of uncertainty?
Yes. Meta-analyses show that psychological treatments can reduce IU, with particularly developed evidence in GAD. How much direct IU targeting adds over other active treatments remains an active research question.
Should I try to eliminate uncertainty?
No. Uncertainty is an ordinary property of real life. The clinically useful goal is to respond more flexibly: gather proportionate information, tolerate what cannot be known yet, and remain able to act without requiring a guarantee.
Related Articles