Why this would work
Connection is not a feeling. It is a health factor, and it has been measured.
Every number on this page has a named source, and the link goes to the paper. Where there is no evidence, it says so.
50%
greater likelihood of surviving the follow-up period, for people with strong ties
Holt-Lunstad, PLoS Medicine, 2010
1 in 6
people worldwide affected by loneliness
World Health Organization, 2025
871,000
deaths a year linked to loneliness, about 100 an hour
World Health Organization, 2025
The problem, measured
The landmark review is by Holt-Lunstad, Smith and Layton, published in the journal PLoS Medicine in 2010. It pooled 148 studies covering 308,849 people, followed for 7.5 years on average.
The finding: people with stronger social ties had a 50% greater likelihood of surviving the follow-up period. The result held across age, sex, health at the start and cause of death. The authors compare it with quitting smoking, and say it is larger than physical activity or not being obese.
Strong social ties
Quitting smoking
Physical activity
Comparable to quitting smoking, and larger than physical activity, in the authors’ words. The figure shows the order, not a scale.
- Design
- A review that pooled 148 studies
- Size
- 308,849 people, followed for 7.5 years on average
- Published in
- Holt-Lunstad, Smith and Layton, PLoS Medicine, 2010
- What it found
- 50% greater likelihood of surviving the period, for people with stronger ties
In May 2025 the World Health Assembly passed its first resolution on social connection. In June, the World Health Organization (WHO) Commission on Social Connection published its report: 1 in 6 people worldwide is affected by loneliness, and it is linked to about 100 deaths an hour, more than 871,000 a year.
What research says works
The Harvard Study of Adult Development has followed the same people since 1938: 724 men at the start, and later about 1,300 of their children. Robert Waldinger, who directs it, puts it this way: the people most satisfied in their relationships at 50 were the healthiest at 80, and it was not their middle-age cholesterol that predicted it.
John Gottman and Robert Levenson watched couples talk and followed what happened to them. What pointed to separation was what you can see in a conversation, not personality: more negative than positive emotion, defensiveness, stubbornness and withdrawal (73 couples, 1992). Over 14 years, the lack of warmth in the conversation about the day predicted the separations that came later (2000).
The findings point the same way: what changes a relationship is not one big conversation a year. It is what happens on ordinary days, including the conversation about how the day went.
How Confidey was built from this
What repeats in each relationship
Confidey keeps the context of each relationship and shows what it has been noticing in it: what helps, what gets in the way, and how trust in each person moves with what actually happens. It is an offer to talk, not a verdict: if it got something wrong, you correct it.
A starting point, and then another
Five areas, from your point of view, answered in two minutes. Not a test and not a diagnosis: it is a measure of where you are today, so that later on there is something to compare against. Progress with no starting measure is activity by another name.
Exercises drawn from known tools
From the goal you choose come exercises built out of tools that exist outside this product and have names: the GROW coaching model (goal, reality, options and will), the scaling question from solution-focused brief therapy, the plan "if this happens, then I will do that", Nonviolent Communication, force-field analysis and values clarification. None of them was invented here.
No exercise repeats
An exercise that comes back stops being an exercise and becomes a form. Each one arrives once, and the order follows what practice shows: look before acting, act before talking, talk before sustaining. Asking for a hard conversation in week one is asking someone to fail.
Both sides, each in their own space
Every relationship has two people. When the other accepts the invitation, each talks in private, both answer the same question of the day, and an exercise can be done with up to five people close to you. Nobody sees anyone else’s answer until everyone is done: reading first changes what you write, and then the two readings stop being independent, which was the whole point.
None of it is public
The World Health Organization records that many people living with a mental disorder face stigma and discrimination. That is why nothing here is published, notifications never reveal the subject, and you can see and delete what Confidey has kept about you, whenever you want.
What holds up each day’s exercise
The exercise is not a question someone invented here. Each one comes from a tool with a named author and known use outside this product: Kurt Lewin’s force field, Marshall Rosenberg’s Nonviolent Communication, the Center for Creative Leadership’s situation, behaviour and impact model, John Whitmore’s GROW model, the wheel of life, Peter Gollwitzer’s "if, then" plan. You see the exercise; the professional who follows you also sees its name and author.
One of them has been measured many times, and it is the one the product uses to turn intention into a plan: "if this happens, then I will do that". Gollwitzer and Sheeran pooled 94 tests in 2006: people who made the plan got closer to their goal than people who only had the intention, an effect the authors themselves call medium to large. It is the exact format of the step Confidey asks for after the person describes the hard moment.
- Design
- A review that pooled 94 independent tests
- Size
- the "if, then" plan against the intention alone
- Published in
- Gollwitzer and Sheeran, 2006
- What it found
- With the plan, the typical person went further than about 3 in 4 people without it
What this does NOT mean: that Confidey inherits the methodology’s effect. An instrument adapted to a screen, answered in four minutes, is not the same as the instrument applied by a professional in a session. What the origin guarantees is that the question has provenance, not that the outcome is measured here.
And on the professional’s side
Between-session homework is one of the few things in this field with many studies behind it. Kazantzis, Whittington and Dattilio pooled 46 studies in 2010. In the ones that compared the SAME therapy with and without homework, people with homework ended better off: the typical person was ahead of about 2 in 3 people without it. That is not the difference between therapies: it is the difference homework makes inside one.
And it is not enough for homework to exist: how much of it gets done relates to the result. Kazantzis and colleagues, in 2016, found across 15 studies with 1,537 people that the more homework a person did, the better the result at the end of treatment, a strong relation. Doing the homework is not an administrative detail. It is part of the treatment.
That is exactly the part that stays invisible. Days pass between one session and the next, and the professional only finds out what happened by asking at the start of the following one, on that session’s time. Confidey for professionals gives the assignment a place to live in that gap, and gives back what got done and where the person got stuck. With their explicit consent, item by item, and never sharing conversation content.
- Design
- A review that pooled 46 studies
- Size
- 1,072 people
- Published in
- Kazantzis, Whittington and Dattilio, 2010
- What it found
- In the same therapy, with homework between sessions, the typical person ended better off than about 2 in 3 people without it
- Design
- A review of studies on how much homework got done
- Size
- 15 studies, 1,537 people
- Published in
- Kazantzis and colleagues, 2016
- What it found
- The more homework a person did, the better the result at the end of treatment: a strong relation
How we read the reviews’ numbers: an effect of 0.48 means the typical person in the homework group was ahead of 68% of the group without it, and 0.65 means 74%. This is Cohen’s reading (1988), which assumes a normal distribution. The 0.79 of the second review measures the relation between how much was done and the result, and sits close to what Cohen calls large (0.8).
What this literature does NOT say: that Confidey improves clinical outcomes. It shows that between-session homework matters and that compliance relates to outcome. No study has measured this product, and we are not going to write that one has.
See Confidey for professionalsWhat we do not claim
The literature on this page supports that social connection matters and that certain behaviours predict outcomes. It does not support the claim that this product improves those outcomes. Turning evidence about the problem into evidence about the solution is the most common error in this category, and we are not going to make it.
Confidey does not replace therapy and does not diagnose anything. The starting point and the product’s other measures are our own instruments, for you to track your own perception over time, and not validated clinical scales. The loneliness figures on this page describe the size of the problem in the world, not our results.
About this page and about what is ours
The sources cited here are public, and that is why they are linked: anyone can check every number. What is not public is the construction: the product, the method, the texts, the interface, the names and the visual identity of Confidey are original work and are protected by copyright.
Citing public literature does not place what we built in the public domain. Reproduction, automated extraction, and total or partial copying of this site to train, replicate or recreate the product are not authorised. Reading, learning and citing with attribution is welcome; cloning is not the same thing.
Confidey is protected work. All rights reserved.
Sources
- Holt-Lunstad J, Smith TB, Layton JB (2010). Social relationships and mortality risk: a meta-analytic review. PLoS Medicine 7(7): e1000316.
- World Health Organization (2025). From loneliness to social connection: charting a path to healthier societies. Report of the WHO Commission on Social Connection.
- World Health Organization (30 June 2025). Social connection linked to improved health and reduced risk of early death.
- World Health Organization. Mental disorders, fact sheet (updated 11 September 2026).
- Mineo L (2017). Good genes are nice, but joy is better. The Harvard Gazette, 11 April 2017.
- Harvard Study of Adult Development (1938 to today).
- Gottman JM, Levenson RW (1992). Marital processes predictive of later dissolution: behavior, physiology, and health. Journal of Personality and Social Psychology 63(2): 221-233.
- Gottman JM, Levenson RW (2000). The timing of divorce: predicting when a couple will divorce over a 14-year period. Journal of Marriage and Family 62(3): 737-745.
- Kazantzis N, Whittington C, Dattilio F (2010). Meta-analysis of homework effects in cognitive and behavioral therapy: a replication and extension. Clinical Psychology: Science and Practice 17(2): 144-156.
- Kazantzis N, Whittington C, Zelencich L, Kyrios M, Norton PJ, Hofmann SG (2016). Quantity and quality of homework compliance: a meta-analysis of relations with outcome in cognitive behavior therapy. Behavior Therapy 47(5): 755-772.
- Gollwitzer PM (1999). Implementation intentions: strong effects of simple plans. American Psychologist 54(7): 493-503.
- Gollwitzer PM, Sheeran P (2006). Implementation intentions and goal achievement: a meta-analysis of effects and processes. Advances in Experimental Social Psychology 38: 69-119.
- Whitmore J (1992). Coaching for Performance. Nicholas Brealey.
- de Shazer S, Berg IK, Lipchik E, Nunnally E, Molnar A, Gingerich W, Weiner-Davis M (1986). Brief therapy: focused solution development. Family Process 25(2): 207-221.
- Rosenberg MB (2015). Nonviolent Communication: A Language of Life, 3rd edition. PuddleDancer Press.
- Center for Creative Leadership. Use Situation-Behavior-Impact (SBI) to understand intent versus impact.
No card. Cancel anytime.