Haodf Scraper: 好大夫在线 Doctor & Hospital Reviews

Real patient reviews from 好大夫在线, structured and ready to use Pull doctor profiles, per-condition patient ratings, and hospital stats from China's largest doctor review platform. No login, no account, no API key, just clean structured data.

🩺 Real per-condition ratings Effect, attitude, and skill scores broken out by the exact condition each patient was treated for, not one blended average.

🔎 Two ways to target Auto-discover doctors by medical department, or point the Actor at specific doctor URLs or IDs.

🔔 Monitor mode Track a shortlist of doctors across repeat runs and get only the reviews that are genuinely new since last check.

🏥 Optional hospital data Add hospital grade, doctor and faculty counts, and service-patient volume without slowing down a base run.

好大夫在线 (Haodf) is China's largest doctor and hospital review platform, covering hundreds of thousands of doctors across every province and specialty. This Actor turns its public doctor profiles and patient reviews into a clean, structured dataset: no login, no account, and no API key required to run it.

🙋 Why use this Actor

• I am a hospital marketing manager benchmarking my own doctors' patient recommend scores and review volume against competing hospitals in the same department, so I know where we're actually losing reputation ground.
• I am a pharma or medical device field rep identifying the highest-volume, highest-rated specialists treating a given condition, so I can prioritize which doctors are worth an outreach visit.
• I am a medical tourism agency coordinator matching international patients to specialists with a track record of real, patient-reviewed outcomes, instead of relying on a hospital's own marketing copy.
• I am an academic researcher pulling structured patient review text and ratings at scale for sentiment or health-outcomes research, the same kind of public review corpus that has already been used in published studies analyzing patient sentiment in China.

🚀 How to use it

STEP 1 Choose your doctors Pick a department to auto-discover doctors, or paste specific doctor URLs or IDs.

STEP 2 Run the Actor It pulls each doctor's profile, reputation stats, and their most-reviewed conditions.

STEP 3 Get three ready-to-use views Doctor profiles, individual patient reviews, and department-level breakdowns.

Want to track the same doctors over time instead of a one-off pull? Turn on Only new reviews since last check and rerun the Actor later with the same State name. Only reviews that weren't there last time come back.

📥 Input

Field	Type	Required	Description
doctorUrls	array of URLs	No	Direct Haodf doctor profile links or their bare numeric IDs, one per line. The cheapest, most reliable way to target specific doctors. Leave empty to auto-discover doctors from department instead.
department	enum	No	Which Haodf department directory to auto-discover doctors from, only used when doctorUrls is empty. Defaults to Respiratory Medicine (呼吸内科). Any valid Haodf department slug works, not only the suggested list.
proxyConfiguration	object	No	Apify Proxy settings. Defaults to the cheaper datacenter proxy, which was confirmed to reach Haodf's doctor, review, and hospital pages without issue.
includeHospitalDetails	boolean	No	Fetch each doctor's hospital page once (deduplicated per hospital) to add hospital grade, total doctor and faculty counts, and service-patient volume. Defaults to false to keep runs fast and cheap.
maxDoctors	integer	No	Maximum number of doctors to scrape in this run, whether sourced from doctorUrls or department auto-discovery. Defaults to 15.
maxDiseaseTagsPerDoctor	integer	No	Caps how many of a doctor's most-reviewed conditions get fetched, each one is a separate set of aggregate stats and up to 10 reviews. Defaults to 5.
onlyNewReviews	boolean	No	Turns the run into a monitor: only reviews not seen before for this doctor and condition under stateName are pushed to the Reviews dataset. Defaults to false.
stateName	string	No	Name for this tracked doctor set's saved monitor history, so multiple independent trackers (one per client or department, for example) don't overwrite each other. Only relevant when onlyNewReviews is on. Defaults to default.
resetState	boolean	No	Clears the saved seen-review history for stateName before the run starts, so every review currently on the page counts as new again. Defaults to false.

📊 Data table

Doctors (default view)

Field	Type	Description
doctorId	string	The doctor's numeric Haodf ID.
doctorUrl	string	Link to the doctor's public Haodf profile.
name	string	Doctor's name.
title	string	Professional title, for example 主任医师 (Chief Physician) or 副主任医师 (Associate Chief Physician).
hospitalId	string	The affiliated hospital's Haodf ID.
hospitalName	string	Affiliated hospital name.
hospitalUrl	string	Link to the hospital's public Haodf profile.
departmentName	string	Department name within the hospital.
departmentUrl	string	Link to the department's Haodf page.
specialty	string	The doctor's stated specialty focus.
recommendScore	number	Patient recommend score out of 5, as shown on the doctor's own profile.
totalPatientsHelped	number	Cumulative patients-helped count reported on the profile.
totalArticles	number	Number of educational articles the doctor has published.
totalThanksAndVotes	number	Combined count of patient thank-you notes and votes.
awards	array of strings	Doctor of the Year (年度好大夫) awards, one entry per year won.
diseaseTagCount	number	Total number of condition tags listed on the doctor's profile.
diseaseTagsFetched	array of strings	Which of those condition tags this run actually fetched reviews for.

Example output

{
  "doctorId": "11908",
  "doctorUrl": "https://www.haodf.com/doctor/11908.html",
  "name": "王建国",
  "title": "主任医师",
  "hospitalId": "5678",
  "hospitalName": "北京协和医院",
  "hospitalUrl": "https://www.haodf.com/hospital/5678.html",
  "departmentName": "呼吸内科",
  "departmentUrl": "https://www.haodf.com/keshi/xxxxx.html",
  "specialty": "支气管哮喘、慢性阻塞性肺疾病诊疗",
  "recommendScore": 4.5,
  "totalPatientsHelped": 9459,
  "totalArticles": 44,
  "totalThanksAndVotes": 1213,
  "awards": ["2023年度好大夫", "2024年度好大夫"],
  "diseaseTagCount": 19,
  "diseaseTagsFetched": ["哮喘", "慢性阻塞性肺疾病", "肺结节"]
}

Reviews view

Field	Type	Description
doctorId	string	The reviewed doctor's Haodf ID.
doctorName	string	The reviewed doctor's name.
hospitalName	string	The doctor's affiliated hospital.
diseaseKey	string	Internal slug for the condition this review was left under.
diseaseName	string	Condition name in Chinese, for example 哮喘 (Asthma).
reviewId	string	Stable review ID, safe to use as a unique key across runs.
content	string	The patient's full written review.
effect	string	Patient rating of treatment effect.
attitude	string	Patient rating of the doctor's attitude.
skill	string	Patient rating of the doctor's skill.
remedy	string	Treatment or remedy context the patient mentioned.
patientProvince	string	Patient's province, as pre-anonymized by Haodf itself.
patientCity	string	Patient's city, as pre-anonymized by Haodf itself.
reviewDate	string	When the review was posted.
isNew	boolean	True when this review was not seen in a prior run under the same monitor state.

Example output

{
  "doctorId": "11908",
  "doctorName": "王建国",
  "hospitalName": "北京协和医院",
  "diseaseKey": "xiaochuan",
  "diseaseName": "哮喘",
  "reviewId": "48213092",
  "content": "医生非常耐心，详细解释了病情和用药方案，复诊后症状明显改善。",
  "effect": "显效",
  "attitude": "很好",
  "skill": "很好",
  "remedy": "吸入用药调整",
  "patientProvince": "北京",
  "patientCity": "朝阳",
  "reviewDate": "2026-08-02",
  "isNew": true
}

Department breakdown view

Field	Type	Description
dimension	string	What this row summarizes, for example department or hospital.
label	string	The department or hospital name for this row.
doctorCount	number	Number of doctors scraped under this dimension in the run.
reviewCount	number	Number of reviews fetched under this dimension in the run.
avgRecommendScore	number	Average patient recommend score across the doctors in this row.
hospitalGrade	string	Official hospital grade, for example 三甲 (Tier 3A), only present when hospital enrichment is on.
totalDoctorsInHospital	number	Hospital-wide doctor count, only present when hospital enrichment is on.
totalFacultiesInHospital	number	Hospital-wide faculty count, only present when hospital enrichment is on.
servicePatientCount	string	Hospital-wide cumulative service-patient count, as reported by the hospital's own page.

Example output

{
  "dimension": "department",
  "label": "呼吸内科",
  "doctorCount": 15,
  "reviewCount": 63,
  "avgRecommendScore": 4.4,
  "hospitalGrade": "三甲",
  "totalDoctorsInHospital": 312,
  "totalFacultiesInHospital": 42,
  "servicePatientCount": "1,200,000+"
}

💰 Pricing

Pricing is pay per result, billed on the doctor and review records actually saved to your dataset. Empty runs cost nothing. There are no fixed monthly subscriptions or hidden maintenance fees.

⭐ Enjoying Haodf Scraper?

⭐ ⭐ ⭐ ⭐ ⭐ If this saved you hours of manually checking doctor pages one by one, let us know. A 5-star rating takes 10 seconds and helps other hospital marketing teams and field researchers find this Actor. Your feedback also tells us what to build next.

★  Rate this Actor on Apify

💡 Tips

• Start with a low maxDoctors and maxDiseaseTagsPerDoctor to see the shape of the data before scaling up to a full department.
• Turn on includeHospitalDetails only when you actually need hospital-level context. It adds one extra request per unique hospital, not per doctor.
• For ongoing tracking, give each tracked doctor set its own stateName so a hospital-benchmarking tracker and a KOL-monitoring tracker never overwrite each other's history.

❓ FAQ

好大夫在线怎么爬取医生评价数据?

You can pull doctor profiles and patient reviews from 好大夫在线 with this Actor. Point it at a department or specific doctor links, run it, and get structured doctor, review, and hospital data back with no login required.

如何查询医生的患者推荐度?

Each doctor record includes recommendScore, the patient recommend score shown on the doctor's own Haodf profile, along with total patients helped and total patient thanks and votes for further context.

Does this Actor need a Haodf account or login?

No. Every page this Actor reads is public. It never logs in, uses a cookie, or requires an API key from Haodf or from you.

Does it extract patient names or private contact details?

No. Haodf pre-anonymizes patient reviews itself, showing only province and city, never a name or contact detail. This Actor passes that anonymization through unchanged and never attempts to identify a patient.

How fresh is the data?

Every run reads live pages at request time. There is no cached or pre-scraped snapshot, so results reflect what is on Haodf right now.

Can I track new reviews instead of re-pulling everything each time?

Yes. Turn on onlyNewReviews and rerun the Actor later against the same doctors and stateName. Only reviews that are new since the last run are returned.

🔗 Other actors

• CNKI Scraper: 中国知网 Citations, Rankings & Academic Search ↗ - pulls citation counts, rankings, and academic search results from China's largest academic database.
• Zhaopin Scraper 智联招聘: China Jobs & Salary API ↗ - scrapes job listings and salary data from one of China's largest recruitment platforms.
• Lagou Tech Jobs Scraper: China IT Recruitment Data ↗ - collects tech job postings and compensation data from China's leading IT recruitment site.
• China Recall Scraper: 中国缺陷产品召回 ↗ - monitors official Chinese government product recall notices by brand and category.
• Employee Reviews Scraper for TeamBlind ↗ - pulls anonymous employee reviews and company ratings for reputation research.