The confidant gap: why men have fewer people to talk to

What is a confidant?

In social-network research, a confidant is a person to whom an individual discloses personal matters — health concerns, relationship difficulties, fears, regrets, and other emotionally significant content. The presence of at least one confidant is one of the most consistent predictors of psychological resilience and physical health in adulthood. The category is distinct from friends or acquaintances who perform other roles, including spouses who serve primarily as romantic partners or co-parents.

The friendship recession has been steeper for men

In its 2021 friendship survey, the Survey Center on American Life documented a striking shift in men's social networks. The share of American men reporting no close friends rose from roughly 3 percent in 1990 to about 15 percent in 2021 — a fivefold increase. The share reporting six or more close friends fell from 55 percent to 27 percent. Men were also less likely than women to report having recently shared personal feelings with a friend, and a growing share identified their spouse or partner as the only person they confide in. The pattern is sometimes labelled the friendship recession in policy discussion; men have absorbed a disproportionate share of it. (See the male loneliness epidemic for the broader pattern.)

A friend is not always a confidant

The distinction matters because the structural problem is specifically a confidant gap, not a friend gap. Male friendships are more often built around shared activity — a team, a job, a project — than around mutual disclosure. This pattern produces real connection and survives well as long as the activity continues, but does not necessarily produce a disclosure relationship. A man may be surrounded by people he would call close and still have nowhere to bring a difficult personal subject.

When the only person a man confides in is his spouse, the configuration concentrates a category of relational labor that historically was distributed across several relationships. This is a structural load on the marriage — not a moral failing of either partner — and is associated with reduced marital satisfaction in longitudinal research.

The measurable cost of withholding

The absence of a confidant is not a soft problem. James Pennebaker's research program, beginning in the mid-1980s and consolidated in a frequently-cited 1997 review, demonstrated that putting an emotional experience into words — naming and ordering it for an audience — is associated with measurable improvements in psychological and physical health, including immune function. The counterpart finding is that chronic withholding is a low-grade physiological cost that the body continues to pay.

Holt-Lunstad and colleagues (2015), in a meta-analysis of roughly 3.4 million participants, found that social isolation was associated with a 26 percent increase in risk of premature mortality. The U.S. Surgeon General's 2023 advisory used this and related findings to argue that the mortality effect of chronic loneliness is comparable to smoking roughly fifteen cigarettes a day. Men, on average, pay this cost more quietly and for longer.

The barrier to formal help

For many men, the gap between recognizing a need to talk and entering a therapeutic relationship is wide — cost, stigma, time, scheduling, and culturally absorbed beliefs about self-reliance all contribute. This is not strictly irrational: not every difficult period rises to clinical significance, and not every concern requires professional treatment. The unmet need sits in an intermediate category — heavier than a casual conversation, lighter than clinical intervention — and has historically had no standard infrastructure.

What actually helps rebuild a confidant network

1. Initiate, on a low-stakes cadence. Two-person standing calls every two weeks outperform reunion-scale gatherings every two years for relationship maintenance. Frequency beats intensity.
2. Convert side-by-side to face-to-face. Friendships built around an activity become confidant relationships only when at least some interaction occurs outside the activity. A walk together is a different category from a game together.
3. Practice disclosure in lower-stakes contexts. The skill of putting an experience into words can be developed independently of the most important conversations. Writing, voice memos to oneself, or conversation with a neutral listener all rehearse the same capacity.
4. Use professional support when appropriate. A therapist or counselor is the appropriate venue for persistent symptoms, processing trauma, or working on patterns that affect multiple relationships. The boundary is not perfectly defined, but the presence of persistent symptoms is a useful marker.

Where voice-based AI fits, honestly

Voice-based AI conversation tools — including CallByrd, a phone-based AI designed for unstructured conversation — can serve as one venue for the disclosure practice described above. The structural appeal for users hesitant to enter therapy is that the conversation occurs with no audience, no appointment, no recurring cost commitment, and no record retained beyond what the service explicitly stores. This is not equivalent to a human confidant. It is one outlet, appropriate for some uses, not for crisis or treatment.

The bottom line

The confidant gap is a measured, structural feature of contemporary American social networks, and the trend is steeper for men. Rebuilding the network is slow, unglamorous, and effective. Anyone experiencing thoughts of self-harm should contact the 988 Suicide and Crisis Lifeline. Anyone experiencing persistent symptoms should consult a clinician. For the in-between — needing to put something into words without entering a clinical relationship — the appropriate response is to identify and use any available outlet, human or otherwise.