Screening and diagnosis are different jobs
Screening asks one question: is there enough here to look at properly? It is deliberately broad. It is designed to catch more people than it needs to, because missing someone costs more than a false alarm.
Diagnosis is the opposite. It asks what is actually going on, and it is done by a qualified professional using your history, your physical health, your medications, your circumstances, and criteria such as those the American Psychiatric Association describes — five or more symptoms, most of the day, nearly every day, for at least two weeks, causing real difficulty in daily life.
The US Preventive Services Task Force recommends screening adults for depression precisely because screening is useful. It also frames screening as a step toward accurate diagnosis, effective treatment and follow-up. Nobody in that chain treats the screen as the answer.
What a questionnaire genuinely gives you
The PHQ-9 is the instrument most apps use, and it is a good one. It is public domain, it has been validated for decades, and in the original 2001 study by Kroenke, Spitzer and Williams a cut-off of 10 gave roughly 88 percent sensitivity and 88 percent specificity for major depression.
Read that number carefully, because it is where the honesty lives. At 88 percent specificity, a meaningful share of people who score 10 or more do not have major depression. That is not a fault; it is what a screen is for. It sorts people into look further and probably no need, and then a human being does the rest.
So a score is real evidence. It is not a verdict, and a screen cannot see the things that most often explain a high score — a thyroid problem, anaemia, a medication, bereavement, a hard six weeks.
What an app can actually do
Quite a lot, once it stops pretending to be a clinician.
- Give you a private way to check, at 11pm, without booking anything or telling anyone
- Turn a vague feeling into a specific sentence you can say to a GP
- Keep a record over weeks, so you are not reconstructing two months from memory in a ten-minute appointment
- Deliver small daily steps of the kind used in behavioural activation, which NICE lists among first-line options for less severe depression
- Put crisis resources one tap away and keep them free
What an app cannot do
It cannot diagnose you. It cannot rule anything out, either — a low score does not mean nothing is wrong, and plenty of people with a real problem score in the mild band.
It cannot examine you or order a blood test, so it cannot see the physical causes that mimic depression closely enough to fool people who are not looking for them.
It cannot prescribe, and it cannot replace therapy. And it is not an emergency service: no app monitors you, knows where you are, or calls anyone on your behalf. If you are in danger tonight, that is a phone call to 988 in the US, Samaritans on 116 123 in the UK and Ireland, a line from findahelpline.com, or your local emergency number.
Claims that should make you close the tab
There is a lot of software in this category and not all of it is careful. A few things are reliable warning signs:
- "Diagnose your depression" — no app can, and saying so is the tell
- Analysing your face, voice or selfies to detect depression — treat with heavy scepticism, and check what happens to the recording
- A cure, or a promised outcome in a set number of days
- A score with no explanation of the instrument behind it or where the bands come from
- No mention of crisis resources anywhere, or crisis resources locked behind a subscription
- A privacy policy that lets your answers be shared with advertisers or data brokers
Where DepreScan sits
DepreScan screens, it never diagnoses. The PHQ-9 check inside the app is free, it produces a screening score with the standard published bands, and it points you toward a health professional rather than standing in for one.
The rest of the app is deliberately not a questionnaire. It asks you a short series of questions about how you have been feeling, sleeping, thinking, socialising and functioning lately — less like a medical form and more like someone helping you take inventory of your life. Then it gives you a snapshot, and a practical plan: small daily steps aimed at whichever parts of your life have slipped. Scan, understand, reclaim.
The daily check-in, the daily steps, the rescue action and every crisis resource are free and stay free. Journey data lives on your iPhone; there is no account and no server holding it. That is the whole claim, and it is a smaller one than most apps in this category make.
Questions people ask
Can an app diagnose depression?
No. Apps can screen, which means flagging that there may be enough symptoms to look at properly. Diagnosis requires a qualified health professional who can take your history, consider your physical health and medications, rule out other causes, and apply diagnostic criteria. Any app claiming to diagnose you is making a claim it cannot support.
What is the difference between screening and diagnosis?
Screening is a broad first pass designed to catch anyone who might need a closer look, so it deliberately flags more people than actually have the condition. Diagnosis is the closer look: a clinician gathering history, ruling out physical causes such as thyroid problems or anaemia, considering medications and circumstances, and applying formal criteria. The screen produces a number; the clinician produces an answer.
Is the PHQ-9 accurate?
It is well validated for what it is. In the original 2001 study, a cut-off of 10 gave around 88 percent sensitivity and 88 percent specificity for major depression, which is why most services use 10 as the point to follow up. But those figures mean a real share of people scoring 10 or more do not have major depression, and some people who do will score below it. That is normal for a screening instrument and is exactly why a human makes the final call.
Should I take a screening result to my doctor?
Yes — it is one of the most useful things you can bring. A score, a date, and a rough sense of how long it has been going on saves you having to find words in a short appointment, and gives the clinician something concrete to work from. Take the record of the last few weeks too if you have one. You are handing over evidence, not a conclusion.
Are depression apps private?
It varies enormously, so check rather than assume. Look for whether answers leave the device, whether there is an account, whether any analytics or advertising SDK is present, and what the privacy policy permits sharing with third parties. DepreScan keeps journey data on the iPhone, has no account and no server holding it, and does not perform face recognition, emotion detection or photo upload.
Sources
- US Preventive Services Task Force — Depression and Suicide Risk in Adults: Screening (2023)
- Kroenke K, Spitzer RL, Williams JBW. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613.
- American Psychiatric Association — What Is Depression?
- NICE guideline NG222 — Depression in adults: treatment and management