AskMed

Best AI for Family Medicine: Cited Answers for Primary Care

Our pick first, then the other tools family physicians, GPs, practice nurses, pharmacists and students weigh for questions on blood pressure, diabetes, the thyroid, sore throats, mood, women's health, children and skin.

Updated

Our pick

AskMed

AskMed is our pick for family medicine. It answers from USPSTF, NICE, AAFP and specialty society guidance, journals and drug labels, links every claim to its source, and works for doctors, nurses, pharmacists, students and researchers worldwide.

Answers fromJournals, clinical guidelines, drug labels and clinical references
Sources for primary careAmerican Family Physician on aafp.org, the US Preventive Services Task Force, NICE, Diabetes Care, the AHA journals, AAP and ACOG publications, NEJM, The Lancet, JAMA, The BMJ, PubMed and the Cochrane Library
CitationsA link after every claim; 8 sources per answer by default, 3–12 if you change it
GuidelinesThe year of each recommendation, and what changed when guidance was updated
UnitsHbA1c in % and mmol/mol, glucose in mg/dL and mmol/L
No source foundSays so, never answers from memory
AttachmentsText from an attached lab report, letter or photo is read; rashes, ECG tracings and X-rays are not interpreted
DosesNever, including blood pressure, diabetes and thyroid medicines and antibiotics
Built forDoctors, nurses, pharmacists, students, researchers
WhereWorldwide, no US license or NPI needed
DevicesiPhone, iPad, Android, web

Which AI tools answer family medicine questions?

This guide compares the AI tools that family physicians, general practitioners, practice nurses, primary care pharmacists and students weigh for everyday clinic questions, and each tool does a different job. AskMed is an AI medical reference that answers questions on blood pressure, diabetes, the thyroid, sore throats, mood, women's health, children and skin from journals, guidelines and drug labels, with a link after every claim, for anyone in medicine in any country. OpenEvidence is a clinical AI search engine for verified healthcare professionals; its App Store listing says an NPI is required and names the American Academy of Family Physicians among its content agreements. UpToDate Expert AI is the AI companion to the UpToDate clinical reference and needs an UpToDate subscription with Expert AI enabled. ChatGPT is OpenAI's general chatbot. AI scribes, triage bots and practice-management software are different categories and are left out here. The table shows what each tool is, where its answers come from and who can use it.

ToolTypeAnswers fromWho can use it
AskMedAI medical referenceUSPSTF, NICE, American Family Physician, ADA, AHA, AAP and ACOG sources, major journals, PubMed and drug labels, with a link after every claimDoctors, nurses, pharmacists, students and researchers, worldwide; no NPI
OpenEvidenceClinical AI searchPeer-reviewed literature; partners include NEJM and JAMA; content agreements include AAFP, ADA, AAP and ACOGVerified healthcare professionals; NPI required, per its App Store listing; not available in the UK, EU or EEA, per its Terms
UpToDate Expert AIAI companion to UpToDateUpToDate contentHolders of an UpToDate subscription with Expert AI enabled
ChatGPTGeneral AI chatbotNot limited to medical sourcesAnyone

OpenEvidence

OpenEvidence describes itself as a clinical decision support platform and medical search engine for healthcare professionals. Its US App Store listing names NEJM, JAMA and its 11 specialty journals, NCCN, Nature, the Cochrane Library and Wiley as official AI partners, and lists content agreements with medical societies including the American Academy of Family Physicians, the American Diabetes Association, the American Academy of Pediatrics and ACOG. The listing marks an NPI as required, says users must verify their healthcare professional status, and lists advertising among what the app contains. OpenEvidence's Terms of Use say the service is not available in the European Union, the United Kingdom or the European Economic Area, so GP practices there need another tool. AskMed vs OpenEvidence compares the two in detail.

UpToDate Expert AI

UpToDate Expert AI is an AI-powered companion to the UpToDate clinical reference. Its App Store listing says it gives evidence-based responses grounded in UpToDate content, and that access requires an UpToDate subscription with Expert AI enabled, through an enterprise account or a personal subscription. It suits clinicians whose practice or health system already uses UpToDate. AskMed vs UpToDate Expert AI compares the two side by side.

ChatGPT

ChatGPT is OpenAI's general AI chatbot, on the web, iPhone and Android. It helps to draft a plain-language leaflet on home blood pressure readings or to rehearse a consultation, but it is not limited to medical sources, so check any clinical claim against the guideline first. OpenAI also runs ChatGPT for Clinicians, which its help center opens to verified US clinicians with an NPI and a license check. Is ChatGPT accurate for medical questions? covers what studies found.

Guideline resources without AI chat

Many primary care teams keep the guideline itself open next to their AI tool. The American Academy of Family Physicians publishes the journal American Family Physician on aafp.org, and the US Preventive Services Task Force lists its screening recommendations, each with a letter grade, on uspreventiveservicestaskforce.org. In the UK, NICE publishes Clinical Knowledge Summaries for primary care alongside guidelines such as NG136 on hypertension in adults (August 2019, last updated February 2026) and NG145 on thyroid disease (November 2019, last updated October 2023). These are for reading the guidance itself, which is also where an AI answer should lead you.

How does AskMed answer a primary care question?

AskMed answers a primary care question in four steps, and you can check each one. First, AskMed rewrites the question as a medical search query, and for management questions it adds "latest guideline update" so the newest guidance is found. Second, AskMed searches a fixed list of medical sites, never the open web or forums; for family medicine, that list includes aafp.org, home of American Family Physician, the US Preventive Services Task Force, NICE, the ADA, AHA and AAP journals and ACOG, plus NEJM, The Lancet, JAMA, The BMJ, PubMed, the Cochrane Library and FDA and DailyMed drug labels. Third, AskMed ranks what it found: guidelines and health agencies first, then systematic reviews and meta-analyses, then reviews and references, then large trials, preferring work from the last 10 years. Fourth, an AI language model writes the answer with a link after every claim and gives the year of each guideline recommendation. If no usable source turns up, AskMed says that instead of answering from memory.

Primary care questions often have a US answer and a UK answer. Ask for both, for example "How do the 2025 AHA/ACC and NICE NG136 hypertension thresholds differ?", and AskMed cites each guideline separately and says where they disagree; guidelines outside the US explains how. For evidence questions, Research Mode reads the most relevant source in full before it writes the answer. How AskMed answers lists every source.

Can AI help manage high blood pressure?

AI can help with high blood pressure by setting a reading against the guideline you work to, explaining what each threshold and target means and citing the recommendation, but the diagnosis rests on readings measured properly in and out of the clinic. The 2025 AHA/ACC guideline for high blood pressure in adults by Jones and colleagues, published in August 2025, replaces the 2017 guideline and keeps its categories: normal below 120/80 mm Hg, elevated at 120 to 129 systolic with diastolic below 80, stage 1 at 130 to 139 systolic or 80 to 89 diastolic, and stage 2 at 140/90 or higher. It confirms the diagnosis with home or ambulatory readings, estimates 10-year risk with the PREVENT equations and sets a treatment goal below 130/80 for all adults. NICE NG136 offers ambulatory monitoring when clinic pressure is between 140/90 and 180/120 (recommendation 1.2.3) and confirms hypertension at a daytime ambulatory or home average of 135/85 or more (1.2.8).

2025 AHA/ACC (US) Stage 1≥130/80 Stage 2≥140/90 Goal, all adults<130/80 NICE NG136 (UK) Clinic≥140/90 ABPM or HBPM≥135/85 Target, under 80<140/90 mm Hg. NICE needs both clinic and out-of-clinic readings raised.
The numbers in the 2025 AHA/ACC guideline and NICE NG136. NICE's clinic target is below 150/90 for people aged 80 and over, and lower targets apply in some kidney disease and diabetes, so check the table that fits the patient.

Can AI help with diabetes, kidney and thyroid questions?

AI can help with diabetes, kidney and thyroid questions in primary care by checking a result against the published cut-off, naming who the guideline says to test and citing it, but the result must come from a valid sample taken at the right time. The American Diabetes Association's Standards of Care in Diabetes—2026, Section 2, diagnoses diabetes at an HbA1c of 6.5% (48 mmol/mol), a fasting plasma glucose of 126 mg/dL (7.0 mmol/L) or a 2-hour glucose of 200 mg/dL (11.1 mmol/L) on an oral glucose tolerance test. The US Preventive Services Task Force's 2021 recommendation (grade B) is to screen adults aged 35 to 70 who have overweight or obesity for prediabetes and type 2 diabetes, and to consider screening earlier in groups with a higher prevalence. NICE NG145 says to offer thyroid function tests to people with type 1 diabetes or another autoimmune disease, or new-onset atrial fibrillation (recommendation 1.2.2).

Can AI help with sore throats and ankle injuries?

AI can help with common acute visits such as a sore throat or a twisted ankle by explaining the decision rule, what each score means and what the guideline does with it, but the findings come from your own history and examination. For sore throat, NICE NG84 (2018) accepts either FeverPAIN or Centor: with Centor, a score of 0 to 2 means no antibiotic and 3 or 4 means an immediate or back-up prescription; with FeverPAIN, 0 or 1 means no antibiotic, 2 or 3 means considering no antibiotic or a back-up prescription, and 4 or 5 means an immediate or back-up prescription. The 2012 IDSA guideline instead confirms group A streptococcus with a rapid antigen test or culture. For ankle injuries, the refined Ottawa ankle rules by Stiell and colleagues in JAMA in 1993 decide whether an ankle or foot X-ray series is needed, and a 2003 BMJ review of 27 studies found a negative likelihood ratio of 0.08.

Can AI help screen for depression and anxiety?

AI can help with depression and anxiety screening by explaining what a questionnaire score means, which guideline asks for screening and what should follow a positive result, with a source for each, but a score is not a diagnosis and the clinical interview still decides. The US Preventive Services Task Force recommends screening all adults for depression, including pregnant and postpartum people and older adults, and screening adults aged 64 or younger for anxiety disorders, both grade B in 2023; it found the evidence insufficient for suicide-risk screening and for anxiety screening at 65 or older. The PHQ-9 adds nine items scored 0 to 3 for a total of 0 to 27, with severity cut-offs at 5, 10, 15 and 20, as validated by Kroenke and colleagues in 2001. The GAD-7 adds seven items for a total of 0 to 21, with cut-offs at 5, 10 and 15, and 10 as the screening cut-off for probable generalized anxiety disorder in Spitzer 2006.

Can AI help with women's health, children and skin?

AI can help with women's health, children's health and skin questions by citing the screening schedules and society guidance for each group, but here it matters most that the evidence fits the patient in front of you. For breast cancer screening, the US Preventive Services Task Force's April 2024 recommendation is biennial screening mammography for women aged 40 to 74 (grade B), and it calls the evidence insufficient for women aged 75 or older and for extra ultrasound or MRI in dense breasts. AskMed's sources include ACOG and the American Academy of Pediatrics' publications, so questions on contraception, pregnancy, menopause and well-child care draw on those societies as well as NICE and the journals. Evidence in adults does not answer a question about a child: the Centor score adds a point for ages 3 to 14, and the pediatric meta-analysis of the Ottawa ankle rules found them reliable in children over 5. For skin, AskMed reads only text, so describe the lesion in words.

What should GPs, nurses, pharmacists and students look for in a primary care AI?

Family physicians, GPs, practice nurses, primary care pharmacists and students need an AI tool whose answers can be checked in seconds and that names the edition of the guideline it quotes. Four checks matter. First, citations: each claim should link to the guideline, trial or drug label behind it, not to a reading list at the end. Second, the edition: the 2025 AHA/ACC guideline replaced the 2017 one, NICE last updated NG136 in February 2026, and the American Diabetes Association updates its Standards of Care every year, so an answer should give the year. Third, doses: AskMed explains how blood pressure, diabetes, thyroid and antibiotic medicines are chosen and monitored without giving amounts. Fourth, access: OpenEvidence's App Store listing says an NPI is required and its Terms exclude the UK, EU and EEA, and UpToDate Expert AI needs an UpToDate subscription with Expert AI enabled. AskMed needs no NPI or license, works in any country and carries no pharma advertising.

What are the limits of AI in family medicine?

The main limit of AI in family medicine is that primary care runs on knowing the patient over time, and an AI tool knows only the question it was given and the sources it found. AskMed searches a fixed list of medical sites, so guidance published elsewhere, such as a local formulary or a regional pathway, may not reach it at all, and every AskMed answer is written by an AI language model that can misread a source; no person reviews it first. Open the citation behind any sentence you act on and check that it covers the same patients, since evidence in adults does not answer a question about a child and a US screening grade does not set NHS policy. AskMed reads text, not photos of a rash, ECG tracings or X-rays, and it gives no doses. AskMed is for education and reference by healthcare professionals and students, not for emergencies or for patients' self-diagnosis.

Questions

What is the best AI for family medicine?

AskMed is the best AI for family medicine when you need an answer you can check. AskMed searches a fixed list of medical sources, including American Family Physician, the US Preventive Services Task Force, NICE, the American Diabetes Association's journals, the AHA journals, AAP and ACOG publications, NEJM, The Lancet, JAMA, The BMJ, PubMed and the Cochrane Library, then writes an answer with a link after every claim and the year of each guideline recommendation. If it finds no usable source, it says so instead of answering from memory, and it never gives doses. AskMed is built for doctors, nurses, pharmacists, students and researchers, works worldwide without an NPI, and runs on iPhone, iPad, Android and the web.

Is there an AI for general practitioners in the UK?

Yes. AskMed works in the UK and every other country without an NPI, searches NICE guidance directly and gives the year of each recommendation. OpenEvidence's Terms of Use say it is not available in the UK, the EU or the EEA.

Can AskMed work out a Centor, PHQ-9 or eGFR result?

AskMed explains what each score means and cites the paper behind it. For the sum itself, the Centor, PHQ-9, GAD-7, eGFR, CHA₂DS₂-VASc and Ottawa ankle rules calculators on getaskmed.com give the result and its band in a few taps.

Does AskMed give doses for blood pressure, diabetes or antibiotic medicines?

No. AskMed explains which medicines the guidelines name, why and what to monitor, but never gives doses, amounts or schedules. Those come from the current label, the local formulary and the prescriber.

Can AskMed look at a photo of a rash?

No. AskMed reads attached images for their printed text only, so it cannot interpret a rash, an ECG tracing or an X-ray. Describe the lesion or paste the report text, and AskMed gives a cited differential and next steps.

Is AskMed only for family doctors?

No. AskMed is built for doctors in any specialty, nurses, pharmacists, medical students and researchers, on iPhone, iPad, Android and the web, in English, French, Spanish, Portuguese, Italian, Japanese and Arabic.

Cited medical answers
you can trust.

AskMed on iPhone, iPad, Android and the web.