EPPP Part 2 focuses on what a psychologist does with knowledge when a practice situation is incomplete, time sensitive, or open to more than one reasonable interpretation. You need to define the decision, weigh the quality and limits of available evidence, account for the person and setting, and explain a defensible next step. A test score may conflict with an interview; an effective treatment may be inaccessible; a supervisor may need to protect a client while helping a trainee learn. This guide develops those decisions across research use, assessment and intervention, relationships, professionalism, ethics, and work with other professionals. Each planned example ties several facts to a specific conclusion so you can see how the judgment follows.
Scientific orientation
1. Apply research to the person in front of you
A strong study can still be a poor match for a particular client. First identify the decision you need to make: which intervention, for which problem, under which conditions? Then compare the study’s participants, setting, treatment, and measured outcomes with the client’s circumstances. Random assignment can strengthen a causal conclusion within a trial, but it cannot by itself show that the result transfers to people excluded from that trial. Look for differences likely to change access, engagement, or the treatment mechanism, such as cognitive demands, language, safety, or housing. Use the finding as one source of evidence alongside assessment, client preferences, and practical supports. The ASPPB EPPP content areas place critical application of research within Part 2’s scientific orientation domain.
Worked example: A trial finds that brief anxiety treatment helps adults without major complicating needs. Your client has similar anxiety, cognitive impairment, and unstable housing. The trial supports considering the treatment, but its effect estimate may not transfer unchanged. Simplify written tasks, arrange feasible practice between sessions, and track anxiety and daily functioning. Revise the plan if access problems or comprehension limit progress.
Watch the distinction: A well-designed trial does not make every client interchangeable with its participants. Check whether differences affect how treatment works or whether the client can use it.
Reference: [1]
2. Read causal claims critically
A difference between treated and untreated groups is not automatically a treatment effect. Ask how people entered each group, what they were like before treatment, and whether outcomes were measured the same way. When clients choose a service, their motivation, schedules, symptom severity, or available support may influence both that choice and later improvement. Those factors confound a simple comparison. A credible comparison makes the groups similar at the outset; random assignment is one way to reduce selection differences, though missing outcomes and unequal follow-up can still bias results. Baseline measures, appropriate adjustment, and replication can strengthen an observational finding, but unmeasured confounding remains possible. Describe what the data show before deciding whether they justify a causal claim.
Worked example: A clinic reports that clients who selected group therapy improved more than clients who declined it. The two groups may have differed in motivation or availability before any session occurred. Their outcome difference shows an association, not the isolated effect of group therapy. Examine baseline severity, attendance, follow-up, and stronger comparative research before telling clients the group caused the improvement.
Watch the distinction: Improvement after choosing group therapy does not establish that the group caused it. The same personal circumstances may influence both the choice and the outcome.
Reference: [1]
3. Translate results into usable numbers
Translate a treatment result into risks or chances that a listener can picture. An absolute difference subtracts the comparison group’s outcome rate from the treated group’s rate. A relative change divides that difference by the comparison rate; it may sound dramatic even when the absolute change is small. For a desirable outcome, the reciprocal of a positive absolute difference gives the approximate number needed to treat for one additional favorable outcome, assuming the comparison is causal and relevant. State the outcome and time period, because improvement at one follow-up does not imply lasting recovery. Sampling uncertainty matters too: a point estimate from one study could differ from the underlying effect. If a confidence interval is available, explain its plausible range rather than presenting one number as certain.
Worked example: In a hypothetical study, 30 of 100 treated clients improve, compared with 20 of 100 controls. The rates are 30% and 20%; the absolute difference is 10 percentage points. The relative increase is 10 ÷ 20 = 50%. If the comparison supports a causal inference, 1 ÷ 0.10 = 10 people treated per additional improvement on average. No individual outcome is guaranteed.
Watch the distinction: A 50% relative increase here is not a 50-percentage-point gain. It describes a 10-point absolute gain relative to the comparison group’s 20% rate.
Reference: [1]
4. Share evidence without overstating it
Accurate communication preserves the shape of the evidence while changing the language for the audience. Name the measured outcome, population, setting, and follow-up period before offering an interpretation. A reduction in reports is not identical to elimination of behavior; it may also reflect changes in detection or reporting. Distinguish a controlled finding from a preliminary observation, and avoid extending a result from a few sites to all schools. When evidence is mixed, say what is consistent and what remains uncertain. For a public audience, concrete statements such as “fewer incidents were reported during the following month” are more useful than a technical effect size alone. For colleagues, include design limitations and alternative explanations so the finding can inform decisions without becoming a promise.
Worked example: Two schools report fewer bullying incidents shortly after a workshop. A request to announce that the workshop “prevents all bullying” exceeds those data. Describe the observed short-term drop in reported incidents at those two schools, explain that later outcomes and other schools were not assessed, and propose continued monitoring. The message remains useful without implying that every incident was prevented.
Watch the distinction: Clear public language need not erase uncertainty. Replacing a limited measured result with an absolute claim changes its meaning and can mislead planning.
Reference: [1]
Assessment and intervention
5. Turn a referral into an assessment question
An assessment begins with a decision, not a test battery. Translate a broad referral into questions whose answers could change support or treatment. Identify the specific difficulty, its onset and settings, the decisions facing the recipient, and any urgent concerns. Then choose information that separates plausible explanations: interviews for history and context, records for instruction and progress, observation for behavior in relevant settings, and measures for defined skills or symptoms. Each method has limits, so plan how sources will complement one another. A label can summarize a pattern, but it cannot substitute for a functional account of what the person needs. Candidates can use EPPP study guidance to connect assessment knowledge with applied decisions.
Worked example: A school asks whether poor grades mean a learning disorder. You clarify that reading comprehension has fallen while calculation remains steady. You review instruction, attendance, language history, attention, emotional concerns, and reading performance. If limited reading instruction explains much of the decline, that changes the interpretation and support plan. If a persistent skill weakness remains despite suitable instruction, targeted assessment becomes more informative.
Watch the distinction: Starting with a broad test battery can produce scores without answering the referral. Define the decision first, then collect evidence that could alter it.
Reference: [1]
6. Choose measures with suitable norms
Select an instrument by asking what construct it measures, whether it is dependable enough for the decision, and whether its scores mean what you intend. Reliability concerns consistency; validity concerns the interpretation and use of results. Neither property is universal across populations or purposes. Norms add a comparison group, so inspect who was sampled, when, and under what language or access conditions. A technically reliable measure may still misrepresent a client whose language experience differs from the norm group. Also consider administration demands, accessibility, and whether a criterion-referenced result would answer the question more directly. When no measure fits well, combine suitable alternatives and report the limitation rather than treating a precise-looking score as decisive.
Worked example: A reasoning measure has strong overall reliability, but its norms come from fluent speakers of a language the client is learning. Low scores on language-heavy items could reflect language exposure as well as reasoning. Choose measures with better construct and language fit, gather relevant history, and describe any result from the original measure as limited by its norm mismatch.
Watch the distinction: High reliability means scores are consistent under studied conditions. It does not establish that the test measures the intended ability fairly for this client.
Reference: [1]
7. Adapt the interview while preserving meaning
Interview adjustments should remove barriers to expressing an answer without supplying the answer itself. Establish the person’s preferred communication method, reduce distractions, use clear wording, and allow enough processing time. Ask one question at a time when complex questions cause confusion. Open prompts invite the person’s own account; specific follow-ups clarify details after that account is heard. Check understanding by asking the person to explain a question in their own words when necessary, rather than simply repeating it louder. Record any communication support because it affects how much confidence to place in apparent inconsistencies. Distinguish a hearing, language, or comprehension problem from a memory or thought problem through suitable follow-up before drawing a clinical conclusion.
Worked example: An older adult gives brief, inconsistent answers while a fan and hallway conversation are audible. You move to a quiet room, confirm which ear hears better, face the person, and ask one clear question at a time. Answers become fuller and more consistent. The first responses are weak evidence of memory loss because the listening conditions were poor; further assessment should address any remaining concern.
Watch the distinction: Repeating a question while adding suggested answers may improve apparent fluency by leading the person. Improve access first, then preserve the person’s own wording.
Reference: [1]
8. Interview within the referral context
The purpose of an interview changes the role, privacy expectations, and relevance of questions. In therapy, exploration may serve the client’s goals; in a consequential evaluation, the psychologist is answering a defined question for an identified recipient. Explain that distinction before gathering sensitive material. Clarify who requested the evaluation, who will receive findings, how information may be used, and the applicable limits of confidentiality. Then seek evidence tied to the referral question and test competing explanations rather than treating the interview as a therapy session. Collateral records may matter because self-report can be incomplete or shaped by the stakes. Fairness also requires documenting what was and was not established, especially when a conclusion could affect employment, legal interests, or access to services.
Worked example: An employee arrives for a fitness evaluation believing it is confidential therapy. You explain the evaluator role, the work-related question, who receives the report, and privacy limits before proceeding. You ask about functioning relevant to that question and obtain authorized supporting information. You avoid promising treatment confidentiality or reporting unrelated intimate details merely because they arose during the interview.
Watch the distinction: A warm interviewing style does not make a consequential evaluation into therapy. Role and disclosure limits must be clear before the employee shares sensitive information.
Reference: [1]
9. Protect standardized administration
A standardized score gains meaning from comparison with people tested under specified conditions. Instructions, practice items, timing, permitted prompts, and scoring rules therefore matter. Extra help can change the task being measured; an altered score cannot automatically be placed on the original norm scale. When a departure occurs, identify exactly which items or conditions were affected and whether the manual or relevant evidence supports an adjusted interpretation. A valid unaffected portion may still be usable, while affected performance can sometimes be described qualitatively. If the result is central to the decision, consider appropriate readministration or a different method. Documentation should allow another psychologist to understand both what the client did and why the reported inference has limits.
Worked example: During a timed task, a trainee gives several hints beyond the standard instructions. The client finishes more items, but the score now combines ability with help that the norm group did not receive. You document the hints and affected items, withhold a norm-based conclusion for that task, and use suitable additional data or valid administration to answer the referral question.
Watch the distinction: Recording the extra hints does not restore score comparability. Documentation explains the departure; the interpretation must still reflect that the testing conditions changed.
Reference: [1]
10. Interpret necessary testing accommodations
An accommodation can improve access while changing what a score represents. Separate the barrier it addresses from the construct the test intends to measure. Translating instructions might allow a client to understand a nonverbal task; interpreting verbal items could alter difficulty, nuance, or the language skill being assessed. Review available guidance and validity evidence for the specific procedure, then identify affected scores. Record who provided support, what was translated or changed, and any observed uncertainty. If standard norms no longer apply, describe performance without implying an exact normative standing. Build conclusions from several suitable sources, including history and measures with better language fit. The aim is a fair decision with transparent limits, rather than a score that appears standardized on paper.
Worked example: A client needs an interpreter for a test standardized in one language. The interpreter relays instructions and some verbal items, but several terms have no close equivalent. You record which items were affected and avoid presenting their standard scores as directly comparable with the norm group. You integrate appropriately selected nonverbal tasks, interview information, and functional evidence before making recommendations.
Watch the distinction: An accommodation is not inherently invalid. Its effect depends on the task and construct, so assess score comparability item by item or scale by scale.
Reference: [1]
11. Compare applicable norm sets
Norms answer a comparison question: how does this performance stand relative to a specified group? The same raw performance can look different against the general population and a group with similar training. Both comparisons can be legitimate if the samples and testing conditions fit the decision. General norms may describe broad population standing; specialized norms may be more relevant to expected performance in a trained role. Inspect each group’s size, representativeness, recency, and eligibility rules before using it. Do not switch reference groups merely to obtain a preferred result. State the group each score uses and the practical question it answers, because a percentile without its reference population is easy to misread. Interpret either comparison alongside the person’s history and observed functioning.
Worked example: A trainee’s task performance is average relative to all adults but low relative to similarly trained peers. The first comparison indicates ordinary population standing; the second raises a question about readiness for duties requiring that trained skill. You report both, examine the relevant task demands and norm quality, and seek corroborating work samples before deciding what support or further evaluation is needed.
Watch the distinction: “Average” has no complete meaning without a comparison group. An average general-population score does not settle whether performance meets expectations for a specialized role.
Reference: [1]
12. Weigh evidence from several sources
Different sources observe different slices of behavior. Interviews reveal experience and meaning; records show history; tests sample performance under defined conditions; observations capture behavior in a particular setting; collateral reports add other viewpoints. Give each source weight according to its relevance, quality, and opportunity to observe the target behavior. Agreement increases confidence, but disagreement can reveal a setting effect rather than an error. Compare the timing, demands, incentives, and definitions behind each report. A brief observation during a calm period cannot rule out problems during transitions or demanding tasks. Instead of averaging unlike data into a false consensus, specify the working explanation and gather evidence where competing explanations make different predictions.
Worked example: A parent reports frequent disruptive behavior, a teacher reports little difficulty, and a psychologist observes a quiet class period. You ask when the behavior occurs and learn that the parent’s examples cluster around homework. You observe or obtain records from that period and compare task demands across settings. The quiet classroom observation limits claims about classroom behavior but does not disprove the parent’s report.
Watch the distinction: A direct observation is not automatically the strongest evidence. Its value depends on whether the relevant behavior had a fair chance to occur.
Reference: [1]
13. Explain discrepant findings
When findings conflict, generate explanations that can be checked. Differences may arise from setting, task length, structure, sleep, mood, effort, informant perspective, or measurement error. A structured test in a quiet room samples capacity under support; daily life may demand sustained self-management amid interruptions. Neither result should automatically cancel the other. Ask what each source actually measured and when the reported difficulty occurs. Then gather a discriminating observation, record, or targeted measure: for instance, performance across a full shift may test a workload explanation better than another short clinic task. State which hypotheses remain plausible and how the uncertainty affects recommendations. This approach turns discrepancy into useful information about conditions under which functioning changes.
Worked example: A client reports attention failures at work, while short structured tests are normal. The client works long shifts with frequent interruptions and sleeps poorly. You review shift patterns, sleep, errors, and reports from relevant settings. Normal clinic performance shows the client can focus under those conditions; it does not rule out difficulty sustaining attention at work. Recommendations address the identified conditions while assessment continues.
Watch the distinction: A normal test result does not prove the client’s report is false. Check whether the test recreated the duration and demands of the reported problem.
Reference: [1]
14. Use base rates when interpreting screens
A screen’s sensitivity is the proportion of true cases it identifies; specificity is the proportion of noncases it correctly clears. A positive result also depends on prevalence in the population being screened. When the condition is uncommon, even a reasonably specific screen can produce many false positives because there are many more noncases available to test positive. To reason clearly, imagine a fixed population, divide it into cases and noncases using the base rate, and apply sensitivity and specificity separately. Positive predictive value is true positives divided by all positive results. It is a property of the test in a particular population, not a permanent number printed on the instrument. Use a positive screen to justify appropriate follow-up, not to replace diagnosis.
Worked example: In a hypothetical group of 1,000, prevalence is 10%, so 100 are cases and 900 are noncases. At 80% sensitivity, 80 cases test positive and 20 negative. At 90% specificity, 810 noncases test negative and 90 positive. There are 170 positive results, of which 80 are true cases: 80 ÷ 170 ≈ 47%. A positive result needs follow-up assessment.
Watch the distinction: An 80% sensitive screen does not mean an individual positive result has an 80% chance of being a true case. Prevalence and false positives matter.
Reference: [1]
15. Build a differential diagnosis
Diagnosis requires a pattern, course, and functional context, not a single striking symptom. Establish when symptoms began, how long they persisted, their severity, and whether they caused impairment or distress. Compare the pattern with the criteria in the current applicable taxonomy, while considering medical conditions, substances, sleep, developmental history, and environmental stressors. A life event can explain why symptoms arose without automatically excluding a disorder; conversely, understandable distress should not be labeled solely because a symptom appears on a checklist. Prior episodes and collateral records may change the differential. Where evidence is incomplete, state the leading possibilities and what information would distinguish them. The diagnostic conclusion should support a useful plan and remain open to revision as new evidence emerges.
Worked example: A client reports low mood and poor concentration after a bereavement. Records also describe an earlier sustained episode with substantial impairment. You examine current duration, severity, functioning, sleep, medical factors, and the prior episode’s features. The bereavement context matters, but it neither proves nor rules out a depressive disorder. You apply current criteria only after comparing the full pattern with plausible alternatives.
Watch the distinction: Bereavement is neither an automatic diagnosis nor an automatic exclusion. The symptom pattern, impairment, history, and other explanations determine the differential.
Reference: [1]
16. Give integrated assessment feedback
Good feedback links each conclusion to the evidence that supports it and states what remains uncertain. Start with the recipient’s question, then explain the main findings in familiar language, including strengths and conditions that may have influenced performance. A low score describes performance on a particular task under particular conditions; it rarely establishes a disorder by itself. Integrate interview, records, observation, and relevant medical or functional information. Distinguish findings from hypotheses and recommendations, and choose actions that follow from the evidence. Check the client’s understanding and invite corrections to factual history. For referral sources, include enough explanation to support decisions while respecting the purpose and authorization for sharing. Feedback should leave the person knowing what the result means and what happens next.
Worked example: A client asks whether one low memory score proves dementia. You explain the task measured a specific kind of recall and that sleep, mood, effort, daily functioning, and medical findings affect interpretation. Other results show preserved abilities, so you describe both strengths and the concern. You recommend appropriate further evaluation and functional monitoring rather than claiming a diagnosis from that score alone.
Watch the distinction: A precise test score can sound conclusive, but precision does not equal diagnostic specificity. Explain what the task measured and what other evidence changes its meaning.
Reference: [1]
17. Form a changeable case formulation
A formulation is a testable account of why this person’s difficulty persists. Organize information into vulnerabilities, triggers, maintaining processes, and strengths or supports. Identify a sequence that can guide action: a cue evokes fear, a behavior brings immediate relief, and that relief makes the behavior more likely next time. Also ask whether the behavior serves a real protective function; a history of harm can make simple reassurance both inaccurate and ineffective. Link each proposed intervention to a mechanism, such as improving safety, testing threat predictions, or reducing avoidance. Treat the formulation as provisional. Repeated outcomes and the client’s account should tell you whether its predictions hold and whether another explanation fits better.
Worked example: A client leaves crowded buses after panic sensations, gaining immediate relief. That relief may reinforce avoidance, yet a past assault on transit makes safety concerns understandable. You assess present safety, agree on realistic travel goals, and plan gradual practice only under suitable conditions. If fear and functioning improve with practice, the formulation gains support; if they do not, you revisit its assumptions.
Watch the distinction: Calling a behavior “avoidance” does not explain the whole case. Ask what relief it provides and whether it also responds to a credible safety concern.
Reference: [1]
18. Match intervention to the problem
Choose an intervention by matching its proposed mechanism to the assessed maintaining process. If avoidance keeps fear from being tested, planned practice in feared situations may be more directly relevant than discussion that never changes behavior. Specify what the client will do, what prediction the activity tests, and how progress will be observed. Evidence for an approach matters, but so do the client’s goals, language, resources, access, and safety. Compare feasible options and explain why one is a reasonable starting point. Begin at a manageable level and adjust tasks using ongoing feedback rather than treating a manual as a fixed sequence. The Part 1 knowledge guide can help connect underlying treatment principles with Part 2 decisions.
Worked example: A client avoids speaking in meetings because they predict visible anxiety will bring ridicule. Insight into this pattern has not changed attendance. You agree on a graded practice task, such as asking one prepared question, and record the prediction and actual response. Later tasks increase difficulty as appropriate. You track participation and distress, then adjust if the task is inaccessible or the feared consequence occurs.
Watch the distinction: An evidence-supported treatment is not automatically a fit for every client. Match its active process and practical demands to the assessed problem and circumstances.
Reference: [1]
19. Include preference and readiness
Shared decisions combine research evidence with the client’s aims, values, readiness, and ability to carry out a plan. Ask what change the person wants now and what they are willing to try. Explain reasonable options, expected benefits, burdens, and uncertainties in language they can use. A goal can be measurable without demanding immediate agreement on the clinician’s preferred endpoint. In substance-use work, clarify how the proposed goal relates to safety and functioning, and discuss situations where additional care may be needed. Agree on an intervention that addresses the identified problem, set a review point, and revisit the plan using observed outcomes. Readiness is changeable information, not a fixed character trait or a reason to abandon collaborative care.
Worked example: A client wants to reduce alcohol use but will not commit to abstinence. You agree to track drinking days and drinks per day, identify high-risk situations, and practice a suitable change strategy. You discuss possible risks, alternatives, and support options. At review, compare the record with the agreed goal and revisit the plan if harm continues or the client’s readiness changes.
Watch the distinction: Refusing an abstinence goal does not mean the client refuses help. A measurable collaborative goal can create useful evidence and room for further change.
Reference: [1]
20. Track change with meaningful measures
Repeated measurement helps distinguish a promising impression from meaningful change. Use the same measure under comparable conditions, record its direction and scale, and pair symptom scores with functioning and the client’s own priorities. A raw difference shows numerical movement; a percentage change divides that difference by the starting value. Neither calculation alone establishes reliable improvement. Measurement error, normal fluctuation, and changes in how questions are answered can affect scores. A reliable-change estimate, when supported by the measure’s properties, asks whether the shift exceeds expected error; clinical meaning asks whether important symptoms or activities improved enough to matter. Review discordant indicators before changing treatment. Part 2 Skills practice questions can help rehearse this distinction in applied decisions.
Worked example: A hypothetical symptom score falls from 24 to 18. The decrease is 24 − 18 = 6 points, or 6 ÷ 24 = 25% of the starting score. Missed workdays remain unchanged. You report symptom improvement, check the measure’s reliability and timing, and ask what still prevents attendance. The data support further investigation, not a declaration of recovery.
Watch the distinction: A lower symptom score is useful evidence, but it is not automatically reliable or clinically meaningful change. Check measurement error and real-world functioning.
Reference: [1]
21. Respond systematically to nonresponse
A flat outcome measure tells you that improvement has not been detected; it does not tell you why. First check whether the measure reflects the client's main goal and whether enough time has passed to expect change. Then examine attendance, how the intervention was delivered, and whether the client could use its active elements between sessions. Ask about understanding, willingness, practical barriers, and changes in stress, health, or safety. If the intervention was delivered and used as intended, revisit the formulation: perhaps the target process is wrong or another condition is sustaining the problem. Agree on one plausible adjustment, define what improvement would look like, and measure again. The EPPP Part 2 skills topics include ongoing evaluation and modification of interventions.
Worked example: After five sessions, Mara's anxiety score remains 22 on the same measure. She completed exposure practice once because transport delays leave no time at the planned location. The psychologist checks that she understands the task, moves practice to a reachable setting, and agrees on three brief attempts before the next review. The unchanged score cannot yet establish that exposure has failed.
Watch the distinction: Changing treatments immediately may feel decisive, but a treatment's mechanism cannot be judged fairly when the client has had little opportunity to use it.
Reference: [1]
22. Make risk assessment actionable
Risk assessment should produce a decision about what happens next, not merely a list of risk factors. Ask directly about current thoughts, intent, plan, access to means, timing, recent actions, prior attempts, substance use, and factors that may increase or reduce immediate danger. Clarify whether the person can participate in a workable safety plan and who can provide support now. Protective factors matter, but their presence does not erase a specific, accessible plan or expressed inability to stay safe. Match the response to the combined picture: arrange an urgent evaluation or emergency response when immediate safety cannot be established, involve appropriate supports, and address access to means where feasible. Reassess as circumstances change and document the information, reasoning, actions, and follow-up. Local duties and available crisis pathways depend on the jurisdiction.
Worked example: During a session, a client reports suicidal thoughts, a specific plan, access to the means, and little confidence in remaining safe tonight. The psychologist clarifies timing and recent actions, stays engaged, and arranges an urgent response with appropriate supports under the applicable setting and jurisdiction. A routine appointment next week leaves the immediate concern unresolved; any later outpatient plan follows the urgent assessment.
Watch the distinction: A caring family member is a meaningful support, but it does not automatically offset current intent, ready access to means, and inability to maintain safety.
Reference: [1]
23. Check fit for remote services
Before using remote care, decide whether the service can be delivered meaningfully and safely through the available connection. Confirm the client's current location and a reliable way to reconnect, since a safety response may depend on where the client actually is. Check privacy at both ends, who might overhear, the stability of the technology, and whether sensory, language, or cognitive needs require support. For an assessment, determine whether remote administration preserves the task and whether available norms support the resulting score. Discuss how interruptions, emergencies, and records will be handled. If a private or valid encounter is impossible, change the setting, method, or timing while preserving continuity of care. A useful set of EPPP Part 2 practice questions should prompt this sequence of decisions.
Worked example: A client joins a trauma session from an open desk at work, wearing headphones while coworkers can hear their replies. The psychologist confirms the client's location, avoids sensitive exploration, and offers a private appointment or a suitable alternative connection. They agree how to reconnect if the call drops. Headphones alone do not make the conversation private.
Watch the distinction: A clear video signal does not establish that remote care fits the situation. Privacy, location, communication needs, and assessment validity can determine whether to proceed.
Reference: [1]
24. Consult when a case exceeds familiar practice
Competence includes knowing when a familiar treatment problem has acquired an unfamiliar cause or level of complexity. Identify what has changed, when it began, and whether the change creates a safety or medical concern. A psychologist can reassess behavior and cognition, but should not infer that a new symptom is psychological merely because the person is already in therapy. Seek consultation matched to the question: a qualified peer may help revise the psychological formulation, while a medical professional addresses possible medication or health effects. With appropriate authorization, coordinate findings and agree on responsibilities and follow-up. Document the uncertainty and any interim modification to treatment. Knowledge from EPPP Part 1 informs the assessment; Part 2 asks how that knowledge guides action in practice.
Worked example: A client treated for anxiety begins having brief episodes of confusion two days after a medication change. The psychologist asks about onset, severity, and immediate safety, and coordinates timely medical review with appropriate consent. A qualified colleague helps consider how the episodes affect the therapy plan. Until clarified, labeling the confusion as anxiety would narrow the inquiry prematurely.
Watch the distinction: A known anxiety diagnosis cannot explain every later symptom. Treating new confusion as proof of worsening anxiety may delay assessment of a medical or medication-related cause.
Reference: [1]
Relational competence
25. Notice how identity shapes the encounter
The psychologist's identities, status, and authority can shape what a client feels free to say. A shared background may help rapport, yet it cannot establish that two people hold the same values or want the same family roles. Notice assumptions before acting on them, then ask questions that allow the client to define their own preferences. Attend to how age, language, disability, race, community ties, and professional power intersect in this encounter rather than treating any one category as a complete explanation. Watch for signs that a client is agreeing to avoid disappointing the psychologist. Offer real choices about goals, participation, and disclosure, and check how the client understood the options. Reflect on feedback and alter your behavior when an assumption has constrained the work.
Worked example: A psychologist assumes a younger client wants parents included in every treatment decision because family consultation is common in the psychologist's community. The client hesitates when an invitation is proposed. The psychologist asks whom the client wants involved, what may be shared, and what should remain private. The plan follows the client's stated preferences within applicable consent rules.
Watch the distinction: Shared cultural identity may offer context, but it is not consent for family involvement or evidence that the client follows the psychologist's preferred decision style.
Reference: [1]
26. Ask about culture without stereotyping
Cultural context changes the meaning of an experience, but it does not settle its clinical significance by itself. Invite the person to describe what happened, the words they use for it, how family or community members understand it, and whether it is expected in this situation. Then examine distress, control, frequency, associated symptoms, functional effects, and any risk. Ask how the person wants help and whether spiritual or community supports matter to them. This approach avoids imposing a diagnosis from an unfamiliar expression while still allowing assessment when an experience is frightening, impairing, or changing. Compare the account with the individual's history and other available information. State conclusions as specific observations and hypotheses, making clear which parts remain uncertain.
Worked example: After her grandmother's death, a client says she sometimes hears her grandmother call her name during mourning rituals. She finds it comforting, retains control, and reports no disruption at work or home. The psychologist asks about cultural meaning and other symptoms, then records the experience without treating it alone as a disorder. New distress or impairment would warrant reassessment.
Watch the distinction: Calling every unusual perception a symptom ignores context; assuming every culturally meaningful experience is harmless ignores distress, impaired functioning, and possible changes over time.
Reference: [1]
27. Use language support effectively
Language support should let the client communicate their own meaning, especially when questions concern trauma, risk, or consent. Ask which language the client prefers for this task and whether they want an interpreter. Choose a qualified interpreter able to maintain accuracy, confidentiality, and appropriate boundaries. Brief the interpreter on the purpose, explain their role to the client, and address the client directly in short, clear segments. Avoid asking the interpreter to summarize, advise, or answer for the client. Where an idiom lacks a direct equivalent, explore its meaning with the client rather than forcing a literal translation. Record relevant language arrangements and any limits they place on interpretation, particularly for standardized measures. Check understanding by inviting the client to explain important choices in their own words.
Worked example: A client's relative offers to interpret a trauma interview. The psychologist arranges a qualified interpreter so the client need not filter disclosures through a family member. When the client uses an idiom translated as 'my heart fell,' the psychologist asks what that phrase means in this context. They record the language support used and avoid scoring answers as though language had no effect.
Watch the distinction: A bilingual relative may be willing and fluent, yet family roles can affect privacy, accuracy, and what the client feels able to disclose.
Reference: [1]
28. Repair a rupture in the working relationship
A sudden change in participation can signal a rupture in the working relationship. Notice the shift and ask about it without demanding agreement. Listen for how the psychologist's words affected the client, including whether a label sounded blaming or overlooked a genuine threat. Acknowledge any part the psychologist contributed and clarify the intended meaning. Then test the formulation together: what does the behavior protect against, what does it cost, and does the client want it to change? Agree on a revised goal or pace that preserves collaboration. A repair is judged by whether the client can again express disagreement and participate in decisions, not by whether they accept the psychologist's original wording. Follow up later because a polite response in the moment may hide continuing distrust.
Worked example: After the psychologist calls leaving crowded places 'avoidance,' a client becomes quiet. Asked about the change, the client says leaving once prevented an assault. The psychologist acknowledges that protective history, explains the current question about costs and safety, and asks which situations the client wants to approach. They agree to distinguish realistic danger from feared discomfort before planning practice.
Watch the distinction: Repeating the clinical label more firmly may deepen the rupture. The immediate task is to understand what the comment meant to the client.
Reference: [1]
29. Use empathy to deepen understanding
Empathy conveys that the psychologist is trying to understand the client's experience without claiming to know more than the client has said. A useful reflection names the apparent feeling or meaning and leaves room for correction. Clarifying questions then turn broad statements into information about people, settings, and consequences. Premature reassurance can shut this process down: it may dispute a painful belief before its basis is understood. Agreement can also be misleading if the psychologist appears to confirm an untested conclusion. Attend to tone and timing as well as words, especially when the client is ashamed or unsure whether disclosure is safe. The aim is a more accurate shared understanding that can guide assessment and action. Relational skills in the EPPP Part 2 scope include respectful, empathic communication.
Worked example: A client says, 'Everyone thinks I am failing.' The psychologist replies, 'It sounds painful to feel judged. Whose reactions have stood out?' The client describes a supervisor's criticism and a partner's silence. The psychologist checks what each response meant to the client. This separates workplace feedback from uncertainty at home and gives them two concrete concerns to explore.
Watch the distinction: Saying 'They do not think that' may be comforting briefly, but it asserts facts the psychologist lacks and can bypass the client's actual concern.
Reference: [1]
30. Work with differing family accounts
When family members give opposing accounts, treat each as evidence about a perspective and setting, not as a vote on the correct diagnosis. Clarify the purpose of the meetings, who is participating, how information will be shared, and whose decisions require consent under the applicable rules. Ask each person for observable examples: what happened, when, with whom, and what followed. Hear the young person's account directly and consider whether behavior differs across homes, school, or other contexts. Identify any safety concerns before trying to reconcile interpretations. Then formulate shared goals in behavioral terms, such as improving school attendance or reducing conflict at bedtime. The family may disagree about causes while still collaborating on an initial plan. Revisit the formulation as new observations test competing explanations.
Worked example: One parent calls a teenager's withdrawal defiance; the other calls it fear. Both report that school mornings are hardest, while the teenager describes panic on the bus after harassment. The psychologist checks the safety concern, asks about other settings, and proposes a shared goal of safer attendance. Neither parent's label alone explains the pattern.
Watch the distinction: Averaging conflicting family accounts can erase the context that explains them. Specific episodes and the teenager's own account may reveal different mechanisms.
Reference: [1]
31. Build trust across groups and systems
Trust with a group or organization depends partly on whether participation has a real effect. Map who can propose, decide, fund, and implement changes before asking people for input. Observe whose voices enter the discussion and whether meeting times, language, or procedures exclude stakeholders. When members disengage, ask about the process itself rather than assuming lack of interest. Give people enough information and time to influence a decision while it is still open, then report what was accepted, changed, or declined and why. Be honest about limits to the group's authority. The psychologist's role is to support a workable relationship across power differences, not to promise control that partners do not have. Track whether the revised process actually changes attendance and participation.
Worked example: A community partner stops attending meetings because agency staff settle plans before members arrive. The psychologist confirms the sequence, asks staff which decisions can still change, and moves those decisions to a joint agenda. Members receive materials in advance. At the next meeting, the chair records which recommendations will be adopted and who has authority to implement them.
Watch the distinction: Inviting people to a meeting does not create collaboration if decisions have already been made or nobody can act on what they say.
Reference: [1]
32. Clarify opposing viewpoints
An objection often names a conclusion, such as 'unsafe,' while leaving its reason unstated. Ask what the person expects might happen and what experience informs that expectation. Separate concerns about confidentiality, interpersonal harm, accessibility, timing, and the proposed intervention itself because each calls for a different response. Reflect the concern before offering information, and check whether your interpretation is accurate. If a practical change could address it, describe that option and let the person evaluate it. If the concern points to a genuine limit of the service, consider another approach. This sequence respects the person's viewpoint while preserving a clear clinical recommendation. It also prevents a false choice between persuading the client and abandoning the plan before understanding the barrier.
Worked example: A client refuses a therapy group, saying it feels unsafe. The psychologist asks what feels risky. The client fears that another member who knows their employer might repeat a disclosure. They discuss group confidentiality limits and whether a different group or individual work fits better. Offering a later meeting time would not address this privacy concern.
Watch the distinction: Treating every refusal as resistance misses the reason behind it. An accurate response depends on whether the barrier is privacy, past harm, access, or preference.
Reference: [1]
33. Manage direct interpersonal conflict
Direct professional conflict is easier to address when the discussion starts with observable conduct and its effect on the shared task. Choose a private, timely conversation. Describe what happened without assigning motives, explain how it affected the client or work, and ask for the colleague's view. Listen for legitimate constraints, such as information they believed was missing. Then propose a specific change in roles or procedure and agree on how to check whether it helps. If the behavior creates an immediate client risk, take the necessary protective step while addressing the conflict. Document significant decisions through ordinary professional channels. The goal is a workable arrangement that serves the client, even if the colleagues continue to differ in style or opinion.
Worked example: During two feedback meetings, a colleague interrupted test explanations before the client could ask questions. The psychologist speaks privately, describes the interruptions and the client's confusion, and asks what the colleague needed to contribute. They agree that the psychologist will finish each finding before the colleague adds treatment implications. At the next meeting, they check whether the client understood both parts.
Watch the distinction: Calling a colleague disrespectful may invite a debate about character. Naming the interruptions and their effect creates a specific problem both can address.
Reference: [1]
34. Examine personal reactions in conflict
Strong personal reactions can supply useful information, but they should not silently determine a professional decision. Notice the trigger, bodily response, thoughts, and urge to act. Ask whether the reaction fits the client's actual conduct or echoes earlier experiences, status concerns, or fatigue. Separate what the client needs from what would relieve the psychologist's discomfort. A qualified peer can help test that distinction, using only the information needed for consultation. After reflecting, return to observable facts, the treatment goals, and any boundary or safety issue. If the relationship can continue, discuss the relevant interaction with the client in a way that serves their care. If a transfer is indicated, plan it for clinical reasons and protect continuity.
Worked example: A client asks whether the psychologist has enough experience, and the psychologist feels defensive and wants to end care. After consulting a peer without unnecessary identifiers, the psychologist recognizes that the question is reasonable given a recent treatment change. They answer accurately, invite the client's concerns, and review progress together. Referral remains an option if the client's needs exceed competence.
Watch the distinction: Ending care to escape feeling criticized mistakes the psychologist's discomfort for a clinical reason. Reflection helps distinguish a true competence limit from a defensive reaction.
Reference: [1]
Professionalism
35. Define the edge of competence
Competence is specific to the service, population, methods, and decisions at stake. Experience with adult therapy may support interviewing and rapport, but it does not establish skill in specialized pediatric testing, developmental interpretation, or school and medical coordination. Clarify the referral question and the consequences of the proposed opinion. Compare those demands with actual training, supervised experience, current knowledge, and access to suitable measures. If there is a gap, decide whether a qualified referral, joint work, or a supervised learning path can meet the client's needs in time. Describe your role accurately to the referral source and client. A license permits practice within applicable rules; it is not evidence of competence in every specialty. The EPPP Part 2 topics explicitly include recognizing limits of competence.
Worked example: An adult therapist is asked to give a pediatric neuropsychological opinion for a child's education plan. The psychologist identifies the need for age appropriate measures, developmental interpretation, and specialized experience they lack. They explain that limitation and arrange a qualified referral. If they pursue the specialty later, they obtain training and supervised experience before making such opinions independently.
Watch the distinction: General assessment experience does not automatically transfer to a specialized population and purpose. The decisive question is whether the psychologist can support this particular opinion.
Reference: [1]
36. Expand practice responsibly
Expanding scope is a staged process. Begin by identifying what the new service requires: theory, assessment, intervention skills, recognition of complications, and decisions about when to refer. A workshop can introduce a method, but listening to instruction differs from delivering it competently with varied clients. Seek supervised practice with a qualified person, use cases appropriate to the current level of skill, and request feedback on observable work. Track outcomes and errors, then expand complexity as competence is demonstrated. Explain training and experience accurately in referrals, consent discussions, and public descriptions. Arrange consultation for cases that exceed the developing skill set. The point is to build a reliable service while clients receive care suited to their needs, rather than treating a certificate of attendance as proof of independent ability.
Worked example: A clinician attends a short workshop on a specialized trauma treatment and wants to advertise it immediately. They instead obtain supervised practice, review case selection and safety issues, and begin with suitable clients while monitoring progress. Public descriptions state the training completed and current experience accurately. Complex presentations are referred or managed with qualified oversight until the clinician's competence develops.
Watch the distinction: Completing a workshop shows exposure to a method, not demonstrated skill in selecting clients, responding to complications, or delivering it independently.
Reference: [1]
37. Keep skills current
Keeping skills current requires comparing established habits with evidence from both research and one's own practice. Look for patterns that suggest a procedure is underperforming: repeated inconclusive results, uneven outcomes across groups, client feedback, or changes in available methods. Check whether the problem lies in selection, administration, interpretation, or the question being asked. Review relevant evidence and seek training where the gap is substantive. Revise the procedure, state any limits during the transition, and monitor whether the revision improves the decisions clients need. A familiar method may remain useful for some people even if it performs poorly for others. Current practice means evaluating fit and consequences, not adopting every new tool because it is new or keeping an old tool because it is comfortable.
Worked example: A clinic's usual assessment battery often yields inconclusive reports for multilingual clients. The team reviews which tests depend heavily on the testing language, how norms were used, and what decisions remain unanswered. After training, they add suitable methods and clearer limitations. They then compare whether later reports answer referral questions more effectively, rather than treating fewer inconclusive labels alone as success.
Watch the distinction: An assessment procedure can be familiar and still be poorly matched to some clients. Review its decision value and fairness before defending it by habit.
Reference: [1]
38. Measure your own practice quality
Practice quality has more than one denominator. Good outcomes among clients who complete treatment do not describe people who leave early. Define the service goal, then review attendance, early engagement, symptoms, functioning, and client feedback together. Look for patterns by referral route or practical barriers, while guarding against conclusions from a small or selective group. With appropriate consent and safeguards, review session recordings or notes to examine specific behaviors such as whether goals were agreed on and options explained. Ask colleagues for feedback tied to observable conduct. Choose one development target, make a change, and see whether the relevant measure improves. This is a cycle of self-assessment, not a search for a flattering average. Clients who leave may have improved, disengaged, or faced barriers; their outcomes need separate inquiry.
Worked example: Of 20 new clients, 8 stop after one session. Among the remaining 12, symptom scores improve, but that says nothing certain about the 8. The psychologist reviews first-session feedback and permitted recordings, finding unclear goal agreements and recurring scheduling problems. They test clearer goal summaries and appointment options, then track whether early departures and reported fit change.
Watch the distinction: Reporting only the results of treatment completers can hide an access or alliance problem. Include early departures when judging the service clients actually receive.
Reference: [1]
39. Correct a professional error
When a professional error is found, the first question is what decisions it may have affected. Preserve the original record and trace the mistake through calculations, interpretation, recommendations, and distribution. Redo the affected work using the appropriate method, then explain the correction plainly to authorized recipients and the client as appropriate. If a decision is pending, alert those who need the corrected information promptly. Document the error, corrected analysis, communications, and any revised recommendation without concealing the original conclusion. Consider whether another qualified reviewer is needed. Finally, identify why the error passed review and improve that step. Accountability includes limiting harm now and reducing recurrence, while avoiding unnecessary disclosure to people who were not authorized to receive the report.
Worked example: A report compared a trainee's score with all adults instead of similarly trained peers. The psychologist recalculates it, finds that the result changes from average to low relative standing for the referral purpose, and revises the interpretation and recommendation. They send a clearly marked correction to authorized recipients, discuss the effect on decisions, and add a norm-set check to report review.
Watch the distinction: Correcting the number alone may leave an inaccurate conclusion in circulation. Recheck interpretation, recommendations, recipients, and decisions affected by the original report.
Reference: [1]
40. Respond to impaired functioning
Personal illness or stress becomes a professional concern when it disrupts judgment, availability, documentation, or follow-through. Use observable signs rather than waiting for certainty about the cause: missed urgent messages, delayed reports, or repeated lapses in concentration. Assess which clients may have unmet needs and which tasks require immediate coverage. Arrange qualified support, redistribution, or a temporary reduction in duties suited to the risk. Review affected cases and communicate necessary continuity information through authorized channels. Seek appropriate personal care and professional consultation, then set concrete conditions for resuming work, such as reliable response coverage and a manageable caseload. Monitor whether those conditions hold. Taking responsibility does not require sharing private health details with clients; they need accurate information about their care and access to services.
Worked example: After a long illness, a psychologist misses two urgent messages and falls behind on records. They arrange qualified coverage for time-sensitive contacts, review the affected clients for unanswered needs, and inform them of the care pathway. With support, they reduce the caseload and establish message monitoring before resuming full duties. Simply promising to catch up would not resolve the access risk.
Watch the distinction: Good intentions cannot substitute for reliable service. When functioning is impaired, assess the effect on clients and put coverage and a workable recovery plan in place.
Reference: [1]
Ethical practice
41. Make consent an ongoing conversation
Informed consent is a process for making a real choice about a defined service. At the outset, explain the purpose and likely activities, who will provide them, foreseeable benefits and burdens, reasonable alternatives, fees where relevant, and how participation can end. Describe confidentiality in usable terms: who normally receives information, what limits may apply, and how questions about disclosure will be handled. Check understanding instead of treating a signature as proof that the client understood. Consent needs renewed attention when the service, participants, or use of information changes. A move from individual therapy to a group changes both the setting and the privacy risks because other members hear disclosures. Explain the psychologist's confidentiality duties and the limits of control over members' conduct, then invite an informed decision. The EPPP Part 2 Skills content areas include applying ethical and legal standards in practice.
Worked example: A client attending individual therapy is offered a group at the same practice. The psychologist describes the group's aims, facilitator, fees, participation expectations, and alternatives. They explain that members will hear personal information and that the practice cannot guarantee what members will say outside meetings. After asking about these differences, the client chooses to join. The psychologist documents the discussion and choice.
Watch the distinction: A signed intake form does not automatically cover every later service. A material change in participants, purpose, or privacy calls for a fresh explanation and an informed choice.
Reference: [1]
42. Limit disclosures to their purpose
Before disclosing information, identify the recipient, the purpose, the information needed for that purpose, and the authority to share it. A request by a third party is not itself authorization from the client. Examine what the client's authorization actually covers, including any limits on content or recipient, and consider applicable legal duties or exceptions. Then select a disclosure that answers the legitimate question without exposing unrelated clinical material. This is a reasoning rule, not a promise that one document type is always sufficient: a specific lawful requirement may call for more, while a narrower authorized confirmation may satisfy an ordinary attendance inquiry. Tell the client what is proposed when appropriate, record the basis and scope of disclosure, and use a secure channel suited to the recipient. Separate facts such as dates attended from opinions about diagnosis or work capacity.
Worked example: An employer asks for a client's complete therapy record to verify attendance. The psychologist checks the client's authorization and finds that it permits attendance confirmation to the employer. A dated confirmation answers the request, so the psychologist sends that document through an appropriate channel and records the disclosure. Therapy notes, diagnoses, and treatment details are outside the stated purpose and authorization.
Watch the distinction: A broad request does not establish a broad need. Sending the whole record because it is easier can reveal sensitive material that the recipient neither requested for a valid purpose nor was authorized to receive.
Reference: [1]
43. Resolve law and ethics conflicts
A legal demand for information must be classified before it is answered. Identify the issuing authority, jurisdiction, recipient, deadline, requested material, and whether the document compels production or merely requests it. Determine whether client authorization, privilege, confidentiality duties, or other protections apply under the law governing that practice. Ethical duties to protect privacy and comply with valid law can coexist, but a broad demand may require clarification, objection, or a narrower response through the proper process. Consultation with qualified legal or professional advisers helps resolve unfamiliar procedural questions; it does not transfer the psychologist's responsibility to understand and carry out the final decision. Preserve the records while the issue is resolved, notify the client when appropriate and lawful, and document the basis, consultation, and exact disclosure. The relevant rules differ across U.S. states and Canadian provinces.
Worked example: A psychologist receives a demand for every therapy note, although its stated purpose concerns attendance dates. They verify its authority and deadline, examine applicable privilege and disclosure rules, and seek legal advice about narrowing the request. If the proper process permits a limited response, they provide attendance information. If a valid order requires more, they comply with its actual scope and document what was released.
Watch the distinction: Neither automatic refusal nor automatic release resolves a law and ethics conflict. The operative questions are what the demand legally requires, what protections apply, and whether a narrower lawful response is available.
Reference: [1]
44. Work through an ethical decision
An ethical decision begins with a precise description of the proposed act and everyone it may affect. Identify the relevant duties, including client welfare, autonomy, confidentiality, fairness, and avoidance of exploitation, alongside applicable law and professional standards. Generate feasible options rather than treating the first apparent choice as inevitable. For each option, examine likely benefit, possible harm, the parties' power differences, and whether a less risky route meets the same goal. Past therapeutic relationships can leave knowledge and influence that persist after treatment ends, so elapsed time alone cannot settle a proposed business relationship. Seek consultation when uncertainty is material, while limiting identifying details, then decide and document the facts and reasoning. Revisit the decision if circumstances change. A sound record shows how competing interests were weighed, not merely that a rule was mentioned.
Worked example: A former client proposes a joint business venture. The psychologist considers whether therapy disclosures could confer unfair influence, whether the former client might feel obliged to agree, and whether business conflict could cause harm. They review applicable standards and seek confidential consultation. Finding substantial unresolved power and privacy concerns, the psychologist declines the partnership and offers a brief, respectful explanation without drawing on therapy material.
Watch the distinction: Ending therapy does not erase the prior power relationship or confidential knowledge. Treating a former client exactly like a stranger can hide risks that require case-specific analysis.
Reference: [1]
45. Address a colleague's concerning conduct
Concern about a colleague calls for a response proportionate to what is known and to the risk. First distinguish a direct observation from rumor, and specify the conduct, setting, and potential effect on clients. Check whether an immediate protective action is needed. When direct discussion is suitable and likely to address the issue, describe the behavior without speculation about motives and request a concrete change. If the conduct continues, is serious, or cannot safely be handled informally, use the relevant supervisory, organizational, professional, or legal route. Applicable reporting duties vary by jurisdiction and circumstance. Protect client information while seeking advice or making a report, and document objective observations and actions. The purpose is to stop harm and meet obligations, rather than to win an interpersonal dispute or silently assume someone else will intervene.
Worked example: A psychologist twice hears a colleague discuss identifiable clients in a public hallway. After confirming the concern, they privately identify the location, disclosures, and privacy risk and ask the colleague to move such discussions to a secure setting. When the behavior recurs, they document what they observed and use the practice's appropriate escalation route, sharing only the information needed to address it.
Watch the distinction: A single uneasy impression may warrant clarification; repeated, specific privacy breaches warrant action. Avoiding a difficult conversation does not protect the clients whose information is being exposed.
Reference: [1]
46. Keep records that support continuity
A useful clinical record lets another authorized professional understand what happened, why a decision was made, and what must happen next. Record relevant reports and observations separately from interpretations, so uncertainty remains visible. Include significant assessment findings, consent and disclosure discussions, interventions, consultation, the clinical reasoning behind consequential decisions, and follow-up responsibilities. A crisis note needs enough detail to reconstruct the risk assessment and response, including changes during contact and any unresolved concerns. Write promptly, use factual language, and correct errors transparently under the applicable record rules rather than silently rewriting history. Documentation should support continuity without collecting irrelevant intimate detail. Local law, organizational requirements, and the service setting may specify additional record duties; these are checked against the clinical purpose. A vague conclusion such as “stable” is especially weak because it conceals its evidentiary basis.
Worked example: After a crisis call, a psychologist records the client's reported thoughts, stated access to means, protective supports, and willingness to follow the agreed plan. The note identifies the consultation obtained, actions taken during the call, the client's response, and who will make the next contact and when. A covering clinician can now assess what was known and carry out follow-up.
Watch the distinction: A polished conclusion is no substitute for the facts and reasoning behind it. “Client stable” alone leaves a covering clinician unable to judge risk or continue the plan.
Reference: [1]
47. Bill for the service actually provided
A billing claim is a representation of the service actually delivered. Compare the claim with the clinical record: the type of work, provider, participants, date, duration when relevant, and any conditions attached to the billing category. Record contemporaneously so that later financial pressure does not reshape the account. If a claim is wrong, correct it through the payer or organization's proper process and keep a clear audit trail. The psychologist should not infer that a client's need, a scheduled appointment length, or a desired reimbursement permits describing work that did not occur. Financial incentives can also affect clinical decisions, such as extending care without a defensible purpose; separate the treatment rationale from the payment outcome. Specific codes and coverage rules depend on the payer and jurisdiction, but the stable principle is accurate representation of work performed.
Worked example: A client attends a short check-in after a canceled therapy appointment. A manager asks the psychologist to bill for the full scheduled session. The psychologist documents the check-in's actual content and duration, selects only a billing description that fits it, and corrects any submitted inaccurate claim. The client's record and the claim therefore describe the same encounter.
Watch the distinction: A reserved appointment slot is not the service provided. Billing for the planned full session after only a brief check-in misstates the work, even under revenue pressure.
Reference: [1]
48. Report research without selective distortion
Research reporting must preserve the distinction between observed results and claims about everyone who entered a study. Define the analysis population and report how many participants lack outcome data, when they left, and whether their known characteristics differ from those who remained. A result among completers can describe those completers; it may overstate or understate the program's effect for all enrollees if leaving is related to outcome. Show denominators with percentages, describe the outcome measure and timing, and avoid selective presentation of only favorable analyses. Consider plausible missing-data explanations and use sensitivity analyses when the design allows them. Statistical precision cannot repair biased follow-up. Ethical reporting also means acknowledging uncertainty even when it complicates a preferred conclusion. The related EPPP study guide provides broader context for research interpretation.
Worked example: In a hypothetical pilot, 60 clients enroll, 36 provide follow-up data, and 27 of those improve. Improvement is 27/36, or 75% of completers, but only 27/60, or 45% of all enrollees, have observed improvement. Outcomes for the other 24 are unknown. The report presents both figures and examines why follow-up is missing before claiming a program-wide effect.
Watch the distinction: Calling the 75% completer rate the program's success rate for all 60 enrollees silently treats missing outcomes as favorable. Attrition changes what the observed percentage can establish.
Reference: [1]
49. Credit scholarly contributions fairly
Scholarly credit should track actual intellectual and writing contributions, with expectations discussed early and revised as the work develops. Distinguish contributions to the question, design, analysis, interpretation, and manuscript from routine access approval, administrative support, or seniority alone. Authorship and acknowledgment serve different purposes, and the applicable journal or institution may define detailed criteria. Ask contributors what they did, agree on roles and order through a fair process, and ensure each named author can review and take appropriate responsibility for the work. Supervisory power makes the discussion especially sensitive: a trainee may hesitate to challenge an inaccurate allocation of credit. Preserve a record of substantive contributions and address disputes before submission. Credit should neither omit a trainee's work nor grant authorship automatically to a senior person whose only role was granting permission.
Worked example: A trainee designs the analysis, interprets results, and drafts major manuscript sections. A senior colleague authorizes access to existing data but makes no substantive intellectual or writing contribution. The team discusses the journal's criteria, credits the trainee for their actual work, and considers an acknowledgment for the access contribution if appropriate. They agree on the final roles before submission.
Watch the distinction: Rank does not establish authorship. Automatically placing a supervisor first while treating substantial trainee analysis and writing as routine assistance can misrepresent who produced the scholarship.
Reference: [1]
50. Handle organizational pressure ethically
An organization's limits are real practice conditions, but they do not erase the psychologist's duty to make defensible clinical decisions. Identify exactly what the rule requires and where it conflicts with the assessed need. Determine which questions can be answered within the available service, which remain open, and whether safety or continuity requires action now. Present the clinical rationale for an exception or an alternative pathway in terms decision makers can use, such as the unanswered referral question and foreseeable consequence of premature closure. Coordinate with appropriate staff, avoid promising resources that are unavailable, and document both the constraint and the response. If the case must transfer, communicate what has and has not been established so the next provider can continue. A single visit can sometimes be sufficient, but a blanket limit cannot substitute for case-specific judgment.
Worked example: A clinic requires closing assessments after one visit. One client's interview reveals unresolved safety concerns and conflicting diagnostic evidence. The psychologist documents the open questions, arranges immediate safety follow-up, and requests an authorized additional assessment or suitable referral. The report states which conclusions remain provisional and gives the receiving clinician the information needed to continue, rather than presenting the visit as complete.
Watch the distinction: A productivity rule does not turn incomplete evidence into a complete assessment. Closing the file with an unsupported diagnosis hides the uncertainty and may interrupt needed care.
Reference: [1]
Collaboration, consultation, and supervision
51. Map barriers within a service system
Access problems often arise from the service system rather than a client's willingness. Map the path from referral to first visit: eligibility decisions, handoffs, contact attempts, available appointment times, language access, transportation, and confirmation procedures. At each step, ask who is lost, how the loss is measured, and whether the recorded reason reflects the client's account or a staff assumption. Compare groups and time periods carefully because referral mix can change. Choose a change aimed at the observed bottleneck, such as adding an agreed contact channel or time window, then track whether contact and attendance improve without creating new privacy problems. Involve frontline staff and clients in identifying workable changes. The unit of analysis is the pathway, not just the missed appointment, so a useful solution targets the step that actually fails.
Worked example: A clinic labels many first-visit absences as low motivation. A review shows that most affected clients work during the hours when scheduling calls are made, and few receive a completed call. The team offers authorized text or evening contact options, records successful contact and booked visits, and compares subsequent attendance. The initial finding supports a contact barrier, while later data test whether the change helps.
Watch the distinction: Missed visits do not reveal motivation by themselves. If clients never receive the scheduling call, counseling them about commitment targets the wrong cause and leaves the access barrier intact.
Reference: [1]
52. Coordinate with other disciplines
Interdisciplinary coordination starts with a shared client goal and clear professional roles. Ask what each discipline is addressing, where recommendations appear inconsistent, and what information each person needs to act. Obtain appropriate authorization before sharing client details, then communicate the relevant functional facts and agreed language rather than transmitting an entire record. In chronic pain care, for example, a psychologist may address fear, coping, and activity avoidance while a physical therapist directs movement assessment and physical rehabilitation. A shared pacing plan can reduce contradictory advice, but each professional remains accountable for decisions within their competence. Include the client in defining a realistic goal and explain any differences that remain. Document the agreed plan and who will review progress. Coordination is successful when the client receives a coherent route toward the goal, not merely when professionals exchange messages.
Worked example: A client with chronic pain hears “increase activity” from physical therapy and “avoid setbacks” from psychology, then stops exercising. With the client's authorization, both clinicians clarify that the goal is gradual, tolerable activity. The physical therapist specifies the movement plan; the psychologist addresses fear and coping during practice. They give the client consistent pacing language and agree to review functional progress.
Watch the distinction: Collaboration does not mean one discipline takes over another's decisions. A psychologist can align behavioral goals with rehabilitation while leaving physical treatment decisions to the qualified professional.
Reference: [1]
53. Make consultation requests answerable
A consultation request is useful when the consultant can identify the decision needing help. State the referral question in one or two sentences, then provide the facts that bear on it: relevant history, current findings, competing explanations, and limits of the available data. Distinguish what has been observed from what has merely been reported or inferred. Ask a focused question that matches the consultant's expertise, such as how a language accommodation affects interpretation of a cognitive result. Obtain the needed authorization and disclose only information relevant to the consultation. Specify the time frame if a decision is urgent, and clarify whether the consultant is advising on interpretation or assuming direct clinical responsibility. After receiving input, integrate it with the whole case rather than treating it as a verdict. Record the recommendation, its rationale, and the resulting decision.
Worked example: A psychologist asks a neuropsychologist whether a low verbal score can support a cognitive conclusion. The request includes the referral purpose, language history, medical findings, other cognitive results, and the interpreter's role during testing. It asks specifically how much weight to give the score and what additional assessment might resolve uncertainty. The psychologist shares only authorized, relevant material and incorporates the answer into the final formulation.
Watch the distinction: “Please review this difficult case” leaves the consultant to guess the decision. A focused question and the data that could change its answer make consultation more useful.
Reference: [1]
54. Test whether consultation helped
Consultation should produce a testable working hypothesis and an observable next step. Define the target behavior and the circumstances in which it occurs, then explain how the recommendation is expected to change it. Agree on who will implement the change, how consistently, and what will be measured. Collect a baseline when feasible, compare it with later observations using the same definition and similar opportunities, and check for other changes that might explain the result. If the expected improvement does not appear, examine implementation before rejecting the hypothesis; a good idea delivered inconsistently has not been fairly tested. If it was delivered as planned, reconsider the formulation and alternative supports. Feedback from the people using the plan matters alongside counts because a numerical improvement that disrupts other important goals may require revision.
Worked example: A school team counts disruptive transitions in a hypothetical 20 observed transitions and finds 12 with disruption. After introducing a predictable cue, it records 5 disrupted transitions in 20 comparable observations. The observed rate falls from 60% to 25%, a 35-percentage-point difference. The psychologist checks whether staff used the cue consistently and whether classroom conditions changed before attributing the improvement to it.
Watch the distinction: A recommendation is not validated just because it sounds plausible or staff like it. Specify the outcome and check delivery and context before claiming that consultation caused improvement.
Reference: [1]
55. Design a program evaluation
A program evaluation begins by stating the program's purpose and the decision the results should support. Define who is eligible, what service is intended, and what change would count as success. Separate implementation measures, such as attempted and successful contacts, from outcomes, such as first-visit attendance. Establish the denominator and observation period in advance so an apparent improvement cannot arise from counting different groups. Choose a comparison that is feasible and informative, while recording concurrent changes in staffing, appointment supply, referral mix, or reminders. Ask clients and staff whether the process works as intended and whether it creates burdens. Protect privacy in data collection and reporting. An evaluation can show that performance changed without proving why; stronger causal claims require a design that separates the program from competing explanations. Practice applying these distinctions with EPPP Part 2 Skills practice questions.
Worked example: A clinic starts outreach to improve first-visit attendance. It defines eligible new referrals, records contact attempts and successful contacts, and tracks the fraction who attend their first scheduled visit. It compares periods before and after launch while recording new evening appointments and any referral changes. If attendance rises, the data show improvement; the concurrent scheduling change remains a possible explanation.
Watch the distinction: Counting outreach calls alone measures activity, not access or attendance. A credible evaluation connects delivery to the intended outcome and records changes that could account for the result.
Reference: [1]
56. Interpret program results over time
When program results change over time, start with the same outcome definition and denominator in each period. Calculate both rates and their absolute difference, then examine whether the clients, service opportunities, or data completeness changed. Case mix matters: a later group with fewer access barriers may attend more often even if the intervention has no effect. Missing records can also bias the rate if absent outcomes are treated differently across periods. Implementation fidelity matters because an outreach program cannot plausibly explain a change among clients it never reached. Plotting successive periods can reveal whether improvement began before the program or coincided with another change. Interpret the observed association at the strength the design supports. Use additional comparisons or stratified results when available, and revise the program according to what the combined evidence shows.
Worked example: In hypothetical periods, missed visits fall from 40 of 100 to 30 of 100. Rates are 40% and 30%; the absolute reduction is 10 percentage points, and the relative reduction is 10/40, or 25%. The later period also offers evening appointments. The clinic reports better attendance, then compares clients with similar scheduling access and checks outreach delivery before assigning credit.
Watch the distinction: A 10-percentage-point decrease is not a 10% relative decrease. It is a 25% reduction from the original 40% rate, and neither calculation establishes which service change caused it.
Reference: [1]
57. Set a supervision plan before risk arises
A supervision plan must make responsibility usable during ordinary work and urgent events. Define the trainee's permitted duties, what requires prior review, how cases and records are reviewed, meeting frequency, documentation, and how the supervisor can be reached. Specify a backup when the primary supervisor is unavailable and an escalation sequence for high-risk situations. The sequence should identify when the trainee pauses independent action, contacts supervision, activates local emergency resources, and records the response, with details adapted to the setting and jurisdiction. Review it before high-risk work begins and rehearse how contact will happen outside ordinary hours. Supervision is an active clinical safeguard, not simply a later signature. The supervisor also needs enough information about caseload and setting to detect patterns of risk, while the trainee needs a route to disclose uncertainty without fear of punishment for asking.
Worked example: A trainee is assigned evening crisis appointments while the supervisor works off-site. Before scheduling them, the supervisor defines which decisions require live contact, tests a reliable phone route, names qualified backup coverage, and records the local urgent-response pathway. They set case-review and documentation expectations. During a later crisis, the trainee knows whom to contact immediately and which actions cannot await the next supervision meeting.
Watch the distinction: A supervisor's nominal availability is insufficient if the trainee cannot reach them when a crisis occurs. The plan must specify a working contact and backup route before exposure to that risk.
Reference: [1]
58. Keep trainees within authorized scope
Supervision does not create competence or legal authority by itself. Before assigning work, check the trainee's training and demonstrated skills, the supervisor's own competence and license, the setting's rules, and the jurisdiction's requirements. Define which parts of a service the trainee may perform and which require direct observation, review, or a qualified specialist. Clients should understand the trainee's status, the supervisor's role, and how supervision may involve review of their information as part of informed consent. When a proposed assessment exceeds the supervisor's expertise, adding ordinary supervision meetings does not repair the gap; obtain appropriately qualified oversight or arrange referral. Continue monitoring actual work, because an initially suitable assignment can become more complex. The supervisor remains responsible for keeping the supervised service within authorized and competent bounds and for correcting problems promptly.
Worked example: A trainee proposes a specialized pediatric neuropsychological assessment, but their supervisor practices adult therapy and lacks that assessment expertise. The supervisor does not authorize the evaluation under general supervision. They arrange oversight by a qualified professional if the setting permits it, or refer the case. For any service the trainee provides, the client receives a clear explanation of the trainee's and supervisor's roles.
Watch the distinction: A supervisor's license is not a blanket authorization for every specialty. If the supervisor cannot competently oversee the procedure, the trainee's enthusiasm and general training do not make it suitable.
Reference: [1]
59. Give observable supervision feedback
Supervision feedback works best when it identifies behavior the trainee can observe and change. Describe what happened in a specific encounter, its effect on the client or task, and the skill to practice next. Invite the trainee's account before deciding why it happened; interruptions may reflect anxiety, a mistaken interview strategy, or pressure to finish on time. Model or rehearse a replacement behavior, agree on a small target, and choose evidence for review, such as an observed interview or recording where permitted. Note strengths as well as growth areas so feedback is accurate rather than merely corrective. At the next review, assess both the targeted conduct and the client's response. Separate developmental coaching from any formal performance decision, while communicating concerns honestly to the trainee and to the organization or jurisdiction when required.
Worked example: During one observed interview, a trainee interrupts the client three times while the client describes a loss. The supervisor names those moments and explains that the account was cut short. They model a reflection followed by a pause, then rehearse it with the trainee. At the next permitted observation, they count interruptions and assess whether the trainee accurately summarizes the client's account.
Watch the distinction: “Be more empathic” names a judgment, not an action the trainee can practice. Specific observations, a replacement response, and a follow-up measure make improvement assessable.
Reference: [1]
60. Protect the supervisory relationship
Supervision has an unavoidable power difference: the supervisor evaluates work and may also control opportunities, pay, or recommendations. That power can suppress disclosure of mistakes unless the process for raising concerns is clear and fair. Discuss evaluation criteria, confidentiality limits within supervision, how client safety concerns are handled, and who can receive a concern if the usual relationship is strained. Identify multiple roles that could bias decisions, then separate functions where feasible and document safeguards. Invite disagreement and respond to disclosure by first addressing the client's needs, then examining the trainee's conduct and training needs. Accountability remains necessary; a supportive environment does not promise freedom from consequences. It does give the trainee a reliable path to report uncertainty early and a transparent basis for remediation. Review whether the arrangement is working, especially after an error or conflict.
Worked example: A supervisor controls a trainee's paid shifts and performance rating. The trainee hesitates to disclose a documentation error. The supervisor invites a factual account, corrects any client-care consequences, and explains how the error will be reviewed. Where possible, another manager handles shift decisions during remediation. The supervisor records the correction, coaching, and review criteria, and gives the trainee a route to raise concerns about fairness.
Watch the distinction: Telling trainees to “speak up” is insufficient when disclosure may affect their income or evaluation. Clear procedures and separation of conflicting roles make candor more realistic.
Reference: [1]