Reviews of the places we train in

Where you spend a rotation shapes what you take out of a healthcare degree more than almost anything else in it, and most students choose one on rumour. A senior tells you a department is worth asking for. A friend of a friend says another one is six weeks of standing in a corner. None of it is written down anywhere, so every cohort rediscovers the same things from scratch — and the departments that teach well get no credit for doing it.

This site is an attempt to write it down. One review is an opinion. Forty reviews of the same hospital, from students in different fields across several years, is evidence.

Every review carries one overall rating and a few sentences of testimony. Six optional sub-ratings cover the things students actually compare placements on:

• Supervision and teaching — were you taught, or left to watch?
• Hands-on experience — did you get to do things yourself?
• How students are treated — as a colleague, or as an inconvenience?
• Workload and hours — reasonable, and predictable?
• Facilities and equipment — somewhere to sit, eat and change, and equipment that works.
• Safety and wellbeing — protective equipment, incident reporting, and what happened when someone raised a concern.

A review also records the field you were training in and what you were there as — student, summer trainee, intern, resident, fellow or observer — which read together as “Pharmacy · Intern”. Naming the department is optional and free text, because departments are called different things everywhere.

Timing is recorded as a year and nothing finer, and on a facility with fewer than five reviews even the year is widened to a five-year band. An exact month in a small department can point at exactly one person.

Nothing here is behind a login. Browsing facilities, reading reviews, following a sort link, sharing a page — none of it asks who you are, and none of it ever will.

Posting a review, commenting or voting needs an account, because otherwise the ratings are worth nothing. An account is a username and a password. There is no email field, no phone number, no real name and no school anywhere in the database, so there is nothing to connect a review back to a rotation list. The privacy page describes exactly what is stored, mechanism by mechanism.

Highest rated does not sort by plain average, because a plain average lets one enthusiastic five-star review outrank a hospital with forty reviews averaging 4.6. Each rating is pulled toward the site-wide mean in proportion to how little evidence stands behind it — a Bayesian average with a confidence constant of eight reviews. A facility earns its position by accumulating agreement, not by being new.

Most helpful sorts reviews by the lower bound of a 95% confidence interval on their like rate, rather than by likes minus dislikes. Subtracting would let a 600/400 review beat a 20/0 one; a raw ratio would put a single like at the top of the page.

Neither number can be bought. Facilities cannot pay to rank higher, to add a review, or to have one taken down.

The directory itself is imported from OpenStreetMap: the facility names, kinds, coordinates and whatever address, phone number and website the map happens to hold. Records are keyed on their OpenStreetMap identity, so a re-import updates a place rather than duplicating it.

The data is uneven, and it shows. Some facilities are named only in Arabic, some only in transliteration, and an address is the exception rather than the rule. A name that reads oddly here almost certainly reads oddly in the map it came from.

If the place you trained at is missing, you can add it, and it joins the directory after a light duplicate check. If a facility’s details are wrong here they are probably wrong in OpenStreetMap too — correcting them there fixes them for everything else built on that map, not just for us.

Facility data © OpenStreetMap contributors, available under the Open Database Licence.

• Not a patient review site. Nothing here is about whether a hospital is a good place to be treated. The reviewers are describing a workplace they were taught in, which is a different question with different answers.
• Not a complaints channel. Nobody at a facility is notified when it is reviewed. If something unsafe happened, it needs an incident report inside your institution as well — a review is not a substitute for one, and it will not reach anyone who can act.
• Not unmoderated. Anonymous does not mean consequence-free. Naming staff or patients gets a review removed, and doing it repeatedly costs the account. The guidelines spell out where the line is.

Tellyrate is an independent project, not affiliated with any hospital, university, ministry or professional body. There is no advertising, nothing is sponsored, no listing is paid for, and no data is sold or shared with anyone — there is barely any to sell.

The way to raise something is the report control on the review, comment or facility in question. It goes to a moderation queue that a person reads.