Psychological Codependency

Abstract

Psychological codependency is a pattern of behavior: a relational style in which a person derives identity, worth, and purpose from meeting others' needs, to the chronic neglect of their own. The construct entered clinical language from the alcoholism-treatment field, where the spouse of a dependent person was seen to organize life around the other's illness. This article sets out what codependency is and how it broadened from the co-alcoholic into a general relational pattern; how its core feature of external focus regulates self-esteem; how it is measured, since the construct lives or dies on reliable assessment; its correlates in family of origin, shame, and parentification; a worked example of how a codependency scale is scored; and current work treating codependency as lived experience, a near-neighbor of enmeshment, and a predictor of relationship outcomes.

Keywords: codependency, external focus, self-sacrifice, relational pattern, family of origin

Psychological codependency denotes a relational pattern in which a person attempts to derive a sense of purpose and self-worth through caretaking of, and reliance on, another person, typically to the point of suppressing their own needs and feelings. The term arose in the 1970s and 1980s within the chemical-dependency treatment field to name what clinicians saw in the partners and families of people with substance use disorders — a complementary pattern of over-involvement, control, and self-neglect that seemed to sustain the very problem it was reacting to (Cermak, 1986). It matters to cognitive psychology because it is a candidate account of how the sense of self and the regulation of self-esteem can become contingent on another person: the codependent individual is proposed to monitor, predict, and manage another's states in place of their own, a stance with direct consequences for attention, appraisal, and emotion regulation (Wells et al., 1999).

What Psychological Codependency Is

Psychological codependency is a pattern of behavior in which a person's sense of self is organized around another person — most often around caring for, controlling, or rescuing someone whose own functioning is impaired. MeSH places it within the broad category of behavior and defines it as a relational pattern in which a person attempts to derive a sense of purpose through relationships with others. The clinical picture that the term was coined to capture has a recognizable shape: excessive caretaking, difficulty saying no, a tendency to derive worth from being needed, suppression of one's own emotions, and an anxious over-attunement to the moods and needs of the other (Cermak, 1986).

The concept has a specific history that explains both its appeal and its contested status. It emerged from the treatment of alcoholism, where the partner of a dependent person — then called the co-alcoholic — was observed to behave in ways that accommodated and inadvertently maintained the drinking. The influential popular treatment that followed generalized the pattern well beyond addiction, describing the codependent as anyone who has let another person's behavior affect them and who is obsessed with controlling that behavior (Beattie, 1987). That generalization — from a narrow observation about the families of addicts to a claimed personality style affecting, on some popular accounts, almost everyone — is exactly what later researchers had to discipline into something measurable, because a construct that applies to all of its subjects explains none of them.

Figure 1

The External-Focus Loop in Codependency

The External-Focus Pattern

Across the competing definitions, the feature most consistently placed at the center of codependency is external focus: the habitual direction of attention, concern, and self-evaluation outward, onto another person's states and onto that person's approval, rather than inward onto one's own feelings and standards (Dear & Roberts, 2005). On this account the codependent individual does not simply care about a partner, as any attached person does; their sense of worth becomes contingent on the partner, so that self-esteem rises and falls with the other's mood, approval, and need. Caretaking is then not only an expression of concern but a strategy for securing esteem, which is why it persists even when it is unreciprocated or self-defeating.

Two further features recur in the clinical and measurement literature and complete the picture. The first is self-sacrifice: the chronic subordination of one's own needs, preferences, and health to those of the other, often experienced as virtue rather than as cost. The second is reactivity, or an over-responsiveness to the other person that leaves the codependent individual with little autonomous regulation of their own emotional state (Dear & Roberts, 2005). The three together — external focus, self-sacrifice, and reactivity — describe a self whose regulatory center of gravity has shifted outside itself, and it is this shift, rather than caretaking as such, that distinguishes the codependent pattern from ordinary devotion. The demonstration below makes the contingency concrete: it contrasts a self-esteem signal that tracks another person's mood with one that is internally anchored.

Measuring Codependency

Because codependency began as a clinical impression and a popular label, its scientific standing has depended entirely on whether it can be measured as a coherent, reliable construct rather than a grab-bag of sympathetic traits. Several self-report instruments were built to do this, and comparing them is instructive. An early clinical attempt operationalized codependency as a multidimensional syndrome: the Codependency Assessment Tool identified factors including other-focus and self-neglect, low self-worth, hiding the self, and a medical or somatic dimension, and reported acceptable internal consistency (Hughes-Hammer et al., 1998). A second tradition sought a briefer, more clearly structured measure, and the Holyoake Codependency Index resolved into the three factors — external focus, self-sacrifice, and reactivity — described above, with each subscale showing adequate reliability and the expected pattern of correlations (Dear & Roberts, 2005).

The instruments do not agree perfectly, and the disagreement is the science. Factor analyses of different scales recover overlapping but non-identical structures, which is what one expects of a construct still negotiating its own boundaries, and a later revision was undertaken specifically to sharpen a measure whose psychometric properties had been questioned (Marks et al., 2012). The central methodological worry is discriminant validity: codependency correlates so strongly with low self-esteem, anxiety, and family dysfunction that critics have asked whether it names anything those established constructs do not already capture. The measurement work is the attempt to answer that empirically — to show that codependency scores predict something, or cohere in some way, that the neighboring constructs do not. Table 1 sets the principal instruments side by side.

Table 1

Principal Self-Report Measures of Codependency

The demonstration below lets the three HCI-style subscales be set independently, so the way a profile of responses becomes a set of subscale scores — and the way two people with the same total can have very different profiles — can be seen directly.

Origins and Correlates

If codependency is measurable, the next question is where it comes from, and here the evidence is correlational but consistent. The most robust finding is a link to the family of origin: codependency scores are elevated in people who grew up with parental substance abuse and interparental conflict, consistent with the construct's origins in the observation of addicted families (Knudson & Terrell, 2012). A proposed developmental route runs through parentification — the reversal of roles in which a child takes on caretaking responsibilities for a parent or for the household — which is thought to train precisely the external-focus, self-sacrificing stance that later defines the adult pattern (Wells et al., 1999).

The affective signature of codependency has also been mapped. It is associated with shame-proneness and low self-esteem, and the account that links them is developmental: a child who learns that worth must be earned by caretaking, and who is shamed for having needs of their own, may carry forward both the compensatory caretaking and the underlying sense of deficiency (Wells et al., 1999). The pattern is not confined to any one culture or to the original Western treatment context; characteristic features of codependency, including self-neglect and over-involvement, have been documented among the wives of men with substance use disorders in other societies, suggesting the relational dynamic generalizes beyond the setting in which it was named (Askian et al., 2016). What the correlational evidence cannot settle is direction: whether a dysfunctional family produces codependency, whether a shame-prone temperament predisposes a person to both, or whether the adult relationship itself maintains the pattern, remains open.

Worked Example

Codependency instruments are Likert scales, and interpreting them requires turning a set of item responses into subscale scores and then comparing those scores against a reference point. Making that arithmetic explicit shows both how the measurement works and why a single total can be misleading (Dear & Roberts, 2005).

Consider an illustrative self-sacrifice subscale of six items, each answered on a five-point scale from 1 (never true) to 5 (always true), so that a higher score means more of the trait. Suppose a respondent answers 4, 5, 5, 3, 4, and 5. The subscale score is the mean of the items: the sum is 4 + 5 + 5 + 3 + 4 + 5 = 26, and the mean is 26 / 6 ≈ 4.33. The natural reference point is the scale midpoint, 3.0, the value that separates the lower from the upper half of the response range; a subscale mean of 4.33 sits well above it, indicating an elevated self-sacrifice tendency. Expressed as a proportion of the usable range — the distance from the floor of 1 to the ceiling of 5, a span of 4 points — the score is (4.33 − 1) / 4 ≈ 0.83, meaning the respondent endorses the self-sacrifice items at about 83% of the maximum possible.

The instructive part is what a total score hides. Imagine a second subscale, reactivity, on which the same respondent is far more moderate, with a mean of 2.0. Averaging the two subscales gives an overall codependency index of (4.33 + 2.0) / 2 ≈ 3.17 — only just above the midpoint, and easy to read as mild. Yet the profile is not mild; it is lopsided, with a strong self-sacrifice component masked by low reactivity. This is why the measurement literature reports and interprets subscale scores rather than a single composite, and why the demonstration above lets each subscale move independently (Dear & Roberts, 2005). The worked arithmetic — mean within a subscale, comparison to the midpoint, and refusal to collapse distinct subscales into one number — is the whole of how these scales are read.

Discussion

Codependency occupies an unusual position in clinical psychology: a construct with enormous popular reach and a genuinely contested scientific status. The strongest case against it as a distinct entity is Cermak's own ambition for it. He proposed that codependence be recognized as a mixed personality disorder with specifiable diagnostic criteria — continued investment of self-esteem in the ability to control self and others, assumption of responsibility for others' needs, anxiety and boundary distortions around intimacy, and enmeshment with personality-disordered or substance-dependent individuals (Cermak, 1986). That proposal was never adopted into the diagnostic manuals, and the reason is instructive: the criteria describe features already distributed across recognized diagnoses and personality dimensions, so codependency as a disorder risked relabeling rather than discovering.

What survived the failure to achieve diagnostic status is codependency as a dimensional relational construct — a measurable way in which some people organize their close relationships — and on that reading the research program is coherent and ongoing. The construct names a real and recognizable configuration of external focus, self-sacrifice, and reactivity; that configuration can be measured with acceptable reliability; and it carries predictable correlates in family history, shame, and self-esteem. The live scientific task is discriminant validity — demonstrating that codependency scores add explanatory value beyond the established constructs they resemble. The honest summary is that codependency is better supported as a description of a relational style than as a diagnostic category, and that its measurement, not its clinical folklore, is where its scientific future lies.

Current Directions

Three strands organize the recent work. The first is qualitative and phenomenological: rather than scoring the construct, interpretative phenomenological analysis has been used to describe codependency from the inside, yielding accounts of a self experienced as absent or defined wholly through others, of compulsive caretaking, and of a gradual loss of one's own boundaries (Bacon et al., 2020). This work reframes codependency less as a checklist of traits than as a disturbance in the sense of self, which connects it to the cognitive-psychological questions about self-representation and self-esteem that the construct has always implied.

The second strand is conceptual housekeeping. As codependency has been studied alongside neighboring relational constructs, researchers have argued that it overlaps substantially with enmeshment — the family-systems notion of boundaries so diffuse that members cannot function as separate individuals — and have proposed treating the two as closely related, perhaps fused, rather than independent (Bacon & Conway, 2023). The third strand is predictive and dyadic: codependency has been examined for its consequences within couples, where higher codependency is associated with poorer dyadic coping, a more distorted perception of the relationship, and lower life satisfaction (Happ et al., 2023). Together these lines move the field past the old argument over whether codependency is real toward the more productive questions of what it is a disturbance of and what, measurably, it predicts.

Common Misconceptions

Codependency is just caring too much about someone.

Ordinary care is compatible with an intact, internally regulated sense of self; the codependent pattern is defined by external focus, in which self-worth becomes contingent on the other person and caretaking is a strategy for securing esteem rather than an expression of it (Dear & Roberts, 2005).

Codependency is a recognized psychiatric disorder.

A proposal to recognize codependence as a mixed personality disorder with formal diagnostic criteria was advanced but never adopted into the diagnostic manuals, because its criteria largely relabel features already covered by established diagnoses; it is better supported as a dimensional relational construct than as a diagnosis (Cermak, 1986).

Codependency only happens in the families of addicts.

The construct originated in the alcoholism-treatment field but was generalized to a broader relational pattern, and characteristic features have since been documented across different cultures and relationship contexts, not only in addicted families (Askian et al., 2016).

Glossary

Behavior.

The observable actions and responses of an organism; the broad MeSH category under which codependency is classified.

Co-Alcoholic.

The original term for the partner of a person with alcohol dependence, whose accommodating behavior was seen to sustain the drinking; the historical seed of the codependency concept.

Codependency.

A relational pattern in which a person derives identity and self-worth from caretaking of and reliance on another, to the neglect of their own needs.

Discriminant Validity.

The degree to which a measure captures something distinct from related constructs; the central empirical challenge for codependency, which overlaps with low self-esteem and anxiety.

Enmeshment.

A family-systems condition of boundaries so diffuse that members cannot function as separate individuals; argued to overlap closely with codependency.

External Focus.

The habitual direction of attention and self-evaluation outward onto another's states and approval; the feature most consistently placed at the center of codependency.

Family of Origin.

The family in which a person was raised; parental substance abuse and conflict within it are the most robust correlates of adult codependency.

Holyoake Codependency Index.

A self-report measure resolving codependency into three factors — external focus, self-sacrifice, and reactivity — with adequate subscale reliability.

Internal Consistency.

The extent to which the items of a subscale measure the same thing; a basic reliability criterion a codependency scale must meet.

Parentification.

A role reversal in which a child takes on caretaking of a parent or household; proposed as a developmental route into the adult codependent stance.

Personality Disorder.

An enduring, pervasive maladaptive pattern of inner experience and behavior; the diagnostic status once proposed for codependence but never adopted.

Reactivity.

An over-responsiveness to another person that leaves little autonomous regulation of one's own emotional state; one of the three HCI factors.

Self-Sacrifice.

The chronic subordination of one's own needs and health to another's, often experienced as virtue rather than cost; one of the three HCI factors.

Shame-Proneness.

A dispositional tendency to experience shame about the self; associated with codependency and proposed to share its developmental origin in a shaming family environment.

Key Researchers

Ingrid Bacon

. Mental health researcher at Kingston University and Kingston and St George's University of London; her interpretative phenomenological work describes codependency from the inside as a disturbance in the sense of self and has argued for its overlap with enmeshment.

Melody Beattie

(1948-2025). Author whose book Codependent No More generalized the concept from the families of addicts to a broad relational pattern and brought the term into wide public use.

Wikipedia - Wikidata

Timmen L. Cermak

(b. 1945). Psychiatrist and addiction-medicine specialist; he proposed that codependence be recognized as a diagnosable mixed personality disorder with formal criteria, the most ambitious clinical claim made for the construct.

Wikipedia - Wikidata

Greg E. Dear

. Psychologist at Edith Cowan University in Australia; he developed and validated the Holyoake Codependency Index and its three-factor structure of external focus, self-sacrifice, and reactivity.

Google Scholar

Donald W. Hine

. Psychologist and Head of the School of Psychology, Speech and Hearing at the University of Canterbury, New Zealand; he contributed to the psychometric revision of codependency measurement.

Faculty Page - Google Scholar

Anthony D. G. Marks

. Psychologist at the University of New England, Australia; he led the development and validation of a revised measure of codependency addressing earlier psychometric criticism.

Google Scholar

Frequently Asked Questions

What is psychological codependency?

Psychological codependency is a relational pattern in which a person derives identity, worth, and a sense of purpose from caretaking of and reliance on another, typically neglecting their own needs; MeSH classifies it as a form of behavior (Cermak, 1986).

Where did the concept of codependency come from?

It emerged in the alcoholism-treatment field to describe the partner of a dependent person, then called the co-alcoholic, and was later generalized into a broad relational pattern by influential popular writing (Beattie, 1987).

Is codependency an official psychiatric diagnosis?

No; a proposal to recognize codependence as a mixed personality disorder with formal criteria was never adopted into the diagnostic manuals, so it is better treated as a dimensional relational construct than a diagnosis (Cermak, 1986).

What is the core feature of codependency?

The feature most consistently central to it is external focus, the direction of attention and self-worth outward onto another person's states and approval rather than being regulated from within (Dear & Roberts, 2005).

How is codependency measured?

By multi-item self-report scales such as the Codependency Assessment Tool and the Holyoake Codependency Index, which resolve the construct into subscales and report subscale rather than single composite scores (Hughes-Hammer et al., 1998).

What are the main causes or correlates of codependency?

It is most robustly correlated with a dysfunctional family of origin, including parental substance abuse and conflict, and with childhood parentification, shame-proneness, and low self-esteem, though causal direction is not established (Knudson & Terrell, 2012).

Is codependency the same as enmeshment?

They are closely related; recent work argues that codependency overlaps substantially with the family-systems concept of enmeshment and proposes treating the two as fused rather than independent (Bacon & Conway, 2023).

Does codependency affect a relationship's outcomes?

Yes; higher codependency within couples is associated with poorer dyadic coping, a more distorted perception of the relationship, and lower life satisfaction (Happ et al., 2023).

References