Back to Details

icd10-cm-skill

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

ICD-10-CM Coding from Clinical Notes

从临床记录提取ICD-10-CM编码

Turn a clinical note into the diagnosis codes a professional coder would submit on the claim for that encounter. This happens in two distinct steps: first decide which conditions belong on the claim, then find the exact code for each. Both steps cause errors: coders miss claims by listing the wrong conditions, and by coding the right condition at the wrong specificity.
将临床记录转换为专业编码员会在本次就诊索赔单中提交的诊断代码。这一过程分为两个明确步骤：首先确定哪些病症应纳入索赔单，然后为每个病症找到精确的代码。这两个步骤都可能出错：编码员可能因列出错误病症或对正确病症编码的特异性不足而导致索赔失误。

Step 1: Decide what belongs on the claim

步骤1：确定索赔单应包含的内容

A claim reflects the encounter, not the patient's chart. Per the ICD-10-CM Official Guidelines for outpatient coding:
Code, in this order:
  1. The reason for the visit (first-listed diagnosis). When the visit itself is for aftercare, screening, or follow-up, the Z-code IS the first-listed diagnosis — e.g. orthopedic aftercare/hardware removal (Z47.x), suture removal (Z48.02), a scheduled wellness exam (Z00.0x).
  2. Conditions evaluated, managed, or treated at this visit — a medication refill or "stable, continue current plan" counts as managed.
  3. Chronic comorbidities, but only if they were addressed or changed medical decision-making this visit.
Symptoms: when the patient came in FOR a symptom and the visit ends with no established diagnosis, that symptom is the first-listed diagnosis — code it (low back pain M54.5x, joint pain M25.5xx). That is the only time a symptom is coded. A symptom that accompanies a coded diagnosis (headache with a coded neck injury, dizziness with coded vertigo, fatigue with coded anemia) is part of that diagnosis and never coded separately.
Leave off the claim:
  • Uncertain diagnoses — "probable", "suspected", "rule out". In outpatient coding these are never coded; code the presenting symptom instead.
  • Conditions mentioned only as history and not treated today.
  • Wellness-exam codes (Z00.0x) on a problem-focused visit. They belong only when the encounter is an actual scheduled physical.
  • Status, lifestyle, and counseling codes — nicotine dependence (F17.x), alcohol use (F10.x), long-term medication (Z79.x), device/stent status (Z95.x), counseling (Z71.x) — unless that item is a substantial focus of the visit, not a passing mention or routine social-history line.
  • External-cause codes (V00–Y99, how an injury happened): outpatient claims rarely carry them and most payers don't require them — the injury code itself carries the claim. Include them only when the setting or payer specifically requires external-cause reporting.
A correctly coded outpatient encounter is short — usually 1 to 4 codes. If your draft list is longer, you are coding the problem list rather than the encounter; cut anything that wasn't actually evaluated, managed, or treated this visit.
索赔单反映的是本次就诊情况，而非患者的全部病历。根据《ICD-10-CM门诊编码官方指南》：
编码顺序如下：
  1. 就诊原因（首要诊断）。当就诊目的是术后护理、筛查或随访时，Z-code即为首要诊断——例如骨科术后护理/内固定取出（Z47.x）、拆线（Z48.02）、预约健康检查（Z00.0x）。
  2. 本次就诊中评估、管理或治疗的病症——药物续方或“病情稳定，维持当前方案”属于管理范畴。
  3. 慢性合并症，但仅当本次就诊中对其进行了处理或改变了医疗决策时才纳入。
症状编码规则： 当患者因某一症状就诊且就诊结束未确诊时，该症状即为首要诊断——需进行编码（如腰痛M54.5x、关节痛M25.5xx）。这是唯一需要对症状编码的情况。伴随已编码诊断的症状（如已编码颈部损伤伴随头痛、已编码眩晕伴随头晕、已编码贫血伴随乏力）属于该诊断的一部分，无需单独编码。
索赔单需排除的内容：
  • 不确定的诊断——“可能”“疑似”“待排查”。门诊编码中从不对此类情况编码，而是对主诉症状进行编码。
  • 仅作为病史提及且本次未治疗的病症。
  • 以问题为导向的就诊中使用健康检查代码（Z00.0x）。该代码仅适用于实际预约的体检就诊。
  • 状态、生活方式和咨询类代码——尼古丁依赖（F17.x）、饮酒（F10.x）、长期用药（Z79.x）、器械/支架状态（Z95.x）、咨询（Z71.x）——除非该内容是本次就诊的重点，而非顺带提及或常规社会史记录。
  • 外部原因代码（V00–Y99，即受伤原因）：门诊索赔单很少包含此类代码，且大多数付款方不要求提供——损伤代码本身已足够支撑索赔。仅当就诊场景或付款方明确要求报告外部原因时才纳入。
正确编码的门诊就诊通常只有1至4个代码。如果你的草稿列表较长，说明你是在对问题列表而非本次就诊进行编码；需剔除所有本次就诊未实际评估、管理或治疗的内容。

Step 2: Code at the documented specificity

步骤2：按文档记录的特异性编码

This is where most miscoding happens. Two rules:
Don't hedge on a diagnosis. Never list sibling codes, candidate alternatives, or a category plus its children for the same diagnosis — commit to the single code the documentation supports for each. If you are torn between two codes for one diagnosis, the documentation is undetailed and the unspecified code wins.
Code exactly what the note documents — never above it, never below it.
  • Default to unspecified when the note doesn't subtype. "Asthma" with no severity → J45.909. "Psoriatic arthritis" with no subtype → L40.50. "Type 2 diabetes" with no complication linked in the note → E11.9. "Hepatitis C" without documented chronicity → B19.20. Unspecified (.9, .50, .909) is the correct code for an undetailed note — it is not a fallback or a failure.
  • Do not infer chronicity, severity grades, laterality, episode type, or diabetes-complication links that the note doesn't state. An ulcer coded with a severity character (L97.x1x "limited to breakdown of skin") requires the note to actually stage the depth; otherwise use unspecified severity (L97.x19).
  • A complication or subtype must be linked by the clinician, not assembled from data. Lab values, vitals, and imaging findings in the note do not by themselves make a complication codable — an elevated A1c does not establish "T2DM with hyperglycemia," and an echo finding does not establish the heart-failure subtype, unless the clinician's own assessment states it. Code from the assessment wording; if the assessment names the condition without the complication, code it unspecified.
  • Add-on codes accompany their base code, never replace it. Resistant hypertension I1A.0 is assigned in addition to I10; if you use an add-on, the base code stays on the claim.
  • "Other specified" (.8, .59) is not "unspecified." Use it only when the note names a specific subtype that has no code of its own. No subtype documented → unspecified, not "other."
  • Never output a bare category. J45, F32, E11, N20 alone are not billable codes. Every code must be carried to its full billable length.
  • Use the documentation when it IS specific. "Acute on chronic systolic heart failure" → I50.23, not I50.9. Under-coding documented detail loses exactly as much as over-inferring.
这是大多数编码错误发生的环节。需遵循两条规则：
诊断编码不模糊。 对于同一诊断，切勿列出同级代码、候选替代代码或一个类别及其子类别——需确定文档支持的单一代码。如果对同一诊断的两个代码存在疑问，说明文档不够详细，此时应选用未特指代码。
严格按照文档记录编码——绝不高估，绝不低估。
  • 文档未明确亚型时默认选用未特指代码。 “哮喘”未提及严重程度→J45.909。“银屑病关节炎”未提及亚型→L40.50。“2型糖尿病”文档未提及并发症→E11.9。“丙型肝炎”未记录慢性病程→B19.20。未特指代码（.9、.50、.909）是文档不够详细时的正确编码——并非备选方案或编码失败。
  • 切勿推断文档未说明的慢性病程、严重程度分级、侧别、发作类型或糖尿病并发症关联。编码带有严重程度字符的溃疡（如L97.x1x“仅皮肤破损”）需文档明确分期深度；否则使用未特指严重程度代码（L97.x19）。
  • 并发症或亚型必须由临床医生明确关联，而非从数据中拼凑。 文档中的实验室值、生命体征和影像学结果本身不足以作为并发症编码依据——糖化血红蛋白升高不能直接认定为“2型糖尿病伴高血糖”，超声结果也不能直接认定为心力衰竭亚型，除非临床医生的评估明确说明。需根据评估措辞编码；如果评估仅提及病症未提及并发症，则编码未特指类型。
  • 附加代码需伴随基础代码，绝不能替代基础代码。 顽固性高血压I1A.0需与I10一同编码；如果使用附加代码，基础代码必须保留在索赔单中。
  • “其他特指”（.8、.59）不等于“未特指”。 仅当文档提及的特定亚型无对应代码时才使用该类代码。未记录亚型→选用未特指代码，而非“其他特指”。
  • 绝不能输出单纯的类别代码。 J45、F32、E11、N20本身不是可计费代码。每个代码必须完整到可计费的长度。
  • 文档明确时按具体内容编码。 “慢性收缩性心力衰竭急性发作”→I50.23，而非I50.9。编码时遗漏文档明确的细节与过度推断的错误程度相同。

Step 3: Find the exact code via the ICD-10 connector

步骤3：通过ICD-10连接器查找精确代码

Look up every diagnosis with the ICD-10 Codes connector's tools — including diagnoses you're sure you know. Code sets change every October and your memory of common codes can be stale; for example, "depression, unspecified" has been F32.A (not F32.9) since 2022. The connector has the current set; trust it over recall.
  • Use 
    search_codes
    with 
    code_type="diagnosis"
    , building the query from the note's own wording plus the specificity decision from Step 2 — if you decided "unspecified," put "unspecified" in the search terms.
  • Take the first result whose description matches the note's wording. If the first result's type, laterality, or complication status contradicts the note (e.g. "Type 1" when the note says "type 2"), it is not a match — move to the next result or refine the query once. Don't page through more than the top few results.
  • Confirm the chosen code with 
    lookup_code
    or 
    validate_code
    — every code on the claim must be valid and billable.
  • The connector returns complete codes, including 7th characters (A/D/S) and X placeholders for injury codes. Copy the code exactly as the connector returns it — if it includes a dot, keep the dot; if not, don't add one. Do not reformat, strip, or extend what the connector gave you.
If the connector's tools are not available, stop. Tell the user the ICD-10 Codes connector needs to be installed or enabled, and do not produce codes from memory — codes recalled without verification are exactly where stale-code-set errors come from.
使用ICD-10 Codes连接器工具查找每个诊断——包括你确定自己熟知的诊断。代码集每年10月都会更新，你对常用代码的记忆可能已过时；例如，“未特指抑郁症”自2022年起编码为F32.A（而非F32.9）。连接器拥有最新代码集；请信任工具而非记忆。
  • 使用
    search_codes
    并设置
    code_type="diagnosis"
    ，根据文档措辞和步骤2确定的特异性构建查询条件——如果确定选用“未特指”，则在搜索词中加入“unspecified”。
  • 选择描述与文档措辞匹配的第一个结果。如果第一个结果的类型、侧别或并发症状态与文档矛盾（例如文档说明是“2型”但结果显示“1型”），则不匹配——需查看下一个结果或调整一次查询条件。无需查看前几个结果之外的内容。
  • 使用
    lookup_code
    或
    validate_code
    确认所选代码——索赔单中的每个代码必须有效且可计费。
  • 连接器会返回完整代码，包括第7位字符（A/D/S）和损伤代码的X占位符。完全按照连接器返回的代码复制——如果包含点号则保留，否则不要添加。请勿重新格式化、删减或扩展连接器返回的内容。
如果连接器工具不可用，请停止操作。 告知用户需要安装或启用ICD-10 Codes连接器，切勿凭记忆生成代码——未经验证的记忆代码正是过时代码集错误的来源。

Working style

工作方式

Work through Steps 1–3 using tool calls only. Don't write explanatory text between searches — no running commentary, no candidate-by-candidate analysis in prose. Your reply is consumed by a claims pipeline that reads every code string in it, so the only prose you produce is the final answer itself, and the only code strings in it are the claim.
仅通过工具调用完成步骤1至步骤3。搜索过程中无需编写解释性文本——无需运行说明，无需逐候选分析的散文内容。你的回复会被索赔流程读取，因此你只需生成最终答案的散文内容，且回复中唯一的代码字符串就是索赔单所需的代码。

Step 4: Final check

步骤4：最终检查

Walk your draft list once before answering. For each code ask:
  1. Does the note show this condition was evaluated, managed, or treated at this visit?
  2. Is the specificity exactly what the note documents — unspecified if undetailed, detailed if documented?
  3. Is this the ONLY code on the list for this diagnosis, at full billable length, taken from a lookup result, dots removed?
Keep the code only if all three hold. If two codes describe the same diagnosis, delete one before answering.
在给出答案前，再次检查草稿列表。对每个代码询问：
  1. 文档是否显示该病症在本次就诊中被评估、管理或治疗？
  2. 编码特异性是否完全符合文档记录——文档不够详细则用未特指代码，文档明确则用详细代码？
  3. 该代码是否是列表中针对该诊断的唯一代码，且为完整可计费长度，来自查询结果，已去除点号？
只有当三个条件都满足时才保留代码。如果两个代码描述同一诊断，给出答案前需删除其中一个。

Answer format

答案格式

End with the codes on their own labeled lines so the first-listed diagnosis is unambiguous:
First-listed: E11.65
Secondary: I10, Z79.4
Codes appear exactly as returned by the connector — dots included. Any reformatting for a specific claims system happens downstream, not here.
最终答案需将代码单独列在带有标签的行中，确保首要诊断明确：
First-listed: E11.65
Secondary: I10, Z79.4
代码需完全按照连接器返回的格式呈现——包含点号。针对特定索赔系统的格式调整需在后续环节进行，而非在此处操作。