Family Member Impact Questionnaire (FMI)

1. Abstract

The Family Member Impact Questionnaire (FMI) is a standardized psychometric assessment instrument developed to measure the multidimensional strain, behavioral disruption, and interpersonal distress experienced by family members living with or emotionally connected to an individual coping with substance misuse or behavioral addiction. Originating from the empirical and conceptual work of Jim Orford and colleagues within the Alcohol, Drugs, Gambling and Addiction Research Group at the University of Birmingham, the FMI is grounded within the broader Stress-Strain-Coping-Support (SSCS) model. The questionnaire comprises 16 items evaluated on a 4-point response scale (ranging from 0 = Not at all to 3 = Often, with an imputation rule for missing or uncertain responses), yielding both a global Total Impact score and two empirically substantiated subscales: Active Disturbance (6 items capturing conflictual, threatening, and disruptive behaviors) and Worrying Behavior (10 items capturing social withdrawal, functional unreliability, physical or mental deterioration, and financial strain).

Psychometric evaluations across diverse cross-cultural cohorts demonstrate robust internal consistency, with Cronbach’s alpha coefficients typically ranging from .82 to .91 for the total scale, and between .74 and .88 for the individual subscales. Confirmatory factor analyses consistently substantiate the two-dimensional architecture across clinical and community samples of affected family members, including spouses, parents, adult children, and siblings. The FMI demonstrates strong convergent validity with measures of psychological distress, physical health symptoms, and family dysfunction, as well as satisfactory discriminant validity when contrasted with general coping styles and perceived social support. In both clinical intervention trials and epidemiological addiction research, the FMI serves as a primary outcome metric for evaluating the efficacy of family-focused interventions—such as the 5-Step Method—by systematically quantifying the reduction of secondary addiction-related harm within the familial system.

2. Keywords

Family Member Impact Questionnaire, FMI, Stress-Strain-Coping-Support Model, affected family members, substance use disorders, addiction impact, Active Disturbance, Worrying Behavior, caregiver burden, family systemic stress, psychometrics

3. Authors

The Family Member Impact Questionnaire was developed and refined across several decades through the collaborative scholarship of researchers affiliated with the Alcohol, Drugs, Gambling and Addiction Research Group in the School of Psychology at the University of Birmingham, along with partner institutions including the University of Bath and the Addiction Press / Addiction Family Support Network (AFINet):

• Jim Orford, PhD, FBPsS — Emeritus Professor of Clinical and Community Psychology, School of Psychology, University of Birmingham, Birmingham, United Kingdom. Primary investigator of the Stress-Strain-Coping-Support (SSCS) model and leading authority on familial aspects of addiction.
• Richard Velleman, PhD, FBPsS — Emeritus Professor of Mental Health Research, Department of Psychology, University of Bath, Bath, United Kingdom, and Senior Research Consultant, Sangath, Goa, India.
• Alex Copello, PhD — Professor of Addiction Studies, School of Psychology, University of Birmingham, and Consultant Clinical Psychologist, Birmingham and Solihull Mental Health NHS Foundation Trust, Birmingham, United Kingdom.
• Lorrena Templeton, MSc — Independent Research Consultant, formerly Senior Research Fellow, Mental Health R&D Unit, University of Bath and Avon and Wiltshire Mental Health Partnership NHS Trust, United Kingdom.
• Early Collaborative Contributors: Edna Oppenheimer, Stella Egert, Colin Hensman, and Shirley Guthrie (collaborators on the original marital cohesion and alcoholism outcome investigations published in 1976).

Correspondence regarding the instrument and its contemporary application within the 5-Step Method may be directed to the Addiction Family Support Network (AFINet) or the School of Psychology at the University of Birmingham.

4. Purpose

The primary objective of the Family Member Impact Questionnaire (FMI) is to provide a methodologically rigorous, standardized, and sensitive measure of the specific negative impacts, disruptions, and strains experienced by individuals who share a close familial, romantic, or domestic relationship with someone exhibiting problematic alcohol, drug, or gambling behaviors. Historically, psychological literature conceptualized family members of individuals with addictions through pathologizing frameworks—frequently labeling partners as “codependent” or causally implicated in the maintenance of addictive behaviors. The FMI was designed under an alternative, non-pathologizing paradigm: recognizing family members as ordinary individuals responding to an extraordinary, chronically stressful set of domestic and social circumstances.

In clinical practice, the FMI functions as both an initial assessment tool and a longitudinal outcome measure. When family members seek psychological support, the instrument systematically maps the precise nature of the stressors they confront. By differentiating between overt interpersonal hostility or safety risks (captured by the Active Disturbance subscale) and chronic anxiety surrounding the relative’s self-neglect, financial insolvency, or functional decline (captured by the Worrying Behavior subscale), clinicians can tailor intervention strategies. For instance, high scores on Active Disturbance necessitate immediate safety planning, boundary establishment, and de-escalation protocols, whereas elevated Worrying Behavior may guide interventions toward cognitive reframing, emotional decoupling, and autonomous self-care practices such as those codified in the 5-Step Method.

In scientific research, the FMI is widely employed in epidemiological surveys, family systems investigations, and randomized controlled trials (RCTs) evaluating the efficacy of psychosocial interventions for affected family members (AFMs). Because the instrument explicitly references behaviors linked to the relative’s substance use or compulsive gambling, it offers greater evaluative specificity than generic measures of psychological distress (e.g., the General Health Questionnaire [GHQ] or the Beck Depression Inventory [BDI]). It permits researchers to establish whether decreases in generalized caregiver strain or emotional distress are mediated by measurable reductions in behavioral impact within the household, even when the focal relative with the substance problem remains unmotivated or refuses formal treatment.

5. Psychological Construct

The psychological construct evaluated by the FMI is addiction-specific familial strain, defined as the cumulative experiential burden, behavioral disruption, and interpersonal conflict sustained by a family member as a direct consequence of a relative’s addictive conduct. Within the conceptual architecture of the Stress-Strain-Coping-Support model, this construct occupies the position of the proximal stressor—the day-to-day manifestation of the relative’s behavioral pathology that impinges directly upon the psychological safety, homeostatic stability, and functional routines of the family unit. The construct is explicitly bifurcated into two correlated yet distinct operational dimensions:

Active Disturbance

Active Disturbance encompasses the overt, direct, and frequently confrontational behaviors enacted by the relative that intrude upon the psychological and physical safety of the family member and the broader domestic sphere. This dimension reflects externalized, volatile behavioral manifestations that provoke acute distress and destabilization. It comprises six specific behavioral indicators:

• Mood Instability and Volatility: Experiencing extreme, unpredictable emotional shifts in the relative (Item 1), creating a persistent climate of apprehension and walking on eggshells.
• Impaired and Hostile Communication: Severe communication breakdown, characterized by unresponsiveness, sullen withdrawal, sarcasm, or aggressive deflection (Item 2).
• Interpersonal Conflict: Active instigation of arguments, verbal hostility, and unprovoked quarrels with the respondent (Item 5).
• Intimidation and Direct Threats: Explicit or implicit verbal, emotional, or physical intimidation and behavioral threats directed toward the family member (Item 6).
• External Crisis Encroachment: The escalation of behavioral disturbance to the degree that external entities—such as neighbors, employers, social services, or law enforcement—are forced to intervene (Item 7).
• Disruption of Family Rituals: Sabotage, behavioral outbursts, or embarrassing intoxication during formal family gatherings, celebrations, or structured domestic occasions (Item 10).

Worrying Behavior

Worrying Behavior conceptualizes the indirect, covert, and chronic domestic erosion caused by the relative’s cognitive, physical, and moral functional decay. Rather than direct hostility, this dimension captures the enduring psychological apprehension, maternal/spousal hypervigilance, and systemic depletion experienced by the caregiver. It includes ten operational indicators:

• Financial Transgression and Theft: Misappropriation of domestic funds, unauthorized borrowing, or stealing money without restitution (Item 3).
• Systemic Economic Depletion: Broader structural strain on household finances, debt accumulation, and inability to meet basic domestic expenses (Item 4).
• Temporal and Spatial Unpredictability: Irregular or awkward schedules, nocturnal wandering, and sudden unexplained absences (Item 8).
• Social Restriction and Isolation: Curtailment of the respondent’s personal social life, leisure activities, and friendships due to embarrassment or caregiver burden (Item 9).
• Familial Alienation and Withdrawal: Consistent failure or refusal to participate in shared family routines, meals, and social life (Item 11).
• Chronic Functional Unreliability: Persistent tardiness, broken commitments, and failure to fulfill agreed domestic or parental responsibilities (Item 12).
• Occupational and Educational Impairment: Chronic anxiety regarding the relative’s vocational failure, absenteeism, or loss of employment/academic status (Item 13).
• Somatic Deterioration: Profound concern regarding the relative’s declining physical health, organ damage, accidents, or visible medical decline (Item 14).
• Self-Neglect and Hygiene Erosion: Apprehension surrounding the relative’s neglect of personal appearance, physical hygiene, nutrition, and basic self-care (Item 15).
• Cognitive and Psychiatric Decline: Acute psychological concern that the relative is experiencing severe psychiatric comorbidities, personality disintegration, memory deficits, or psychosis (Item 16).

6. Theoretical Framework

The conceptual foundation of the Family Member Impact Questionnaire is rooted in the Stress-Strain-Coping-Support (SSCS) model, formulated by Orford, Velleman, Copello, Templeton, and their international research associates. The SSCS framework was developed as an integrative, socio-ecological alternative to dominant medical and psychodynamic paradigms that historically mischaracterized affected relatives. The theoretical framework integrates principles from general stress and coping theory (Lazarus & Folkman, 1984), systemic family therapy, and community psychology.

The SSCS model posited five interconnected components that delineate the familial experience of addiction:

1. Stress: The primary domestic stressor, comprising the ongoing behavioral consequences of a family member’s problematic alcohol, drug, or gambling behavior. The FMI operationalizes this exact component of the model, quantifying the objective and perceived frequency of domestic disruption.
2. Strain: The resultant psychological and physical health decrements experienced by the family member, encompassing depressive symptoms, generalized anxiety, somatization, and existential despair (frequently operationalized using instruments like the General Health Questionnaire or the Symptom Rating Test).
3. Coping: The strategic actions and psychological mechanisms deployed by family members to manage the ongoing stress. The SSCS framework categorizes these into three primary behavioral styles: engaged coping (confronting, attempting to control, or monitoring the relative), tolerant coping (putting up with, accommodating, or enabling the behavior), and withdrawing/independent coping (setting boundaries, gaining psychological distance, and prioritizing personal well-being).
4. Support: The qualitative and structural social resources available to the family member, including emotional validation, informational guidance, and practical instrumental assistance from informal networks or professional systems.
5. Contextual Influences: Cultural norms, socioeconomic standing, structural poverty, gender expectations, and kinship roles that moderate the relationship between the primary stressor and subsequent strain.

Within this theoretical architecture, the relative’s addiction is understood not as an isolated pathology occurring within an individual, but as an ongoing, ambient domestic stressor that disrupts the homeostatic equilibrium of the entire family system. The behaviors assessed by the FMI are fundamentally destabilizing because they introduce chronic uncertainty, threaten physical and financial security, and violate relational expectations of reciprocity, trust, and shared responsibility. By measuring this stressor dimension distinctly from coping mechanisms and psychological strain, the FMI allows researchers to empirically test hypothesized mediational pathways—such as whether the adoption of assertive withdrawal coping diminishes the impact of Active Disturbance on psychological strain.

7. Validity

The Family Member Impact Questionnaire has undergone rigorous empirical validation across multiple international clinical trials, community surveys, and cross-cultural psychometric evaluations:

Construct and Structural Validity

Construct validity for the FMI was initially established by demonstrating that its operationalized items correspond to the qualitative accounts of domestic distress provided by family members across diverse cultural contexts, including the United Kingdom, Mexico, Italy, and Australia. Factor-analytic studies consistently substantiate the hypothesized two-dimensional model, demonstrating that family stress bifurcates naturally into acute, externalized conflict (Active Disturbance) and chronic, internalized functional anxiety (Worrying Behavior).

Convergent Validity

The FMI exhibits moderate to strong positive correlations with validated measures of caregiver strain and psychological symptomatology. In a seminal validation study conducted by Orford et al. (2005) involving 183 family members of individuals with alcohol and drug problems, the FMI Total Impact score correlated significantly with the General Health Questionnaire-28 (GHQ-28; r = .48 to .56, p < .001) and the Symptom Rating Test (SRT; r = .52, p < .001). Furthermore, significant positive correlations have been observed between FMI subscales and the Coping Questionnaire (CQ), specifically with engaged and tolerant coping styles, which are theoretically driven by elevated levels of domestic disturbance and worry.

Discriminant Validity

The FMI demonstrates adequate discriminant validity by correlating weakly or non-significantly with constructs theoretically orthogonal to objective behavioral stress. Correlations between FMI scores and measures of perceived social support (such as the Multidimensional Scale of Perceived Social Support) are generally low and negative (typically r = -.15 to -.24), confirming that the FMI captures objective environmental disruptions rather than the subjective availability of social capital. Additionally, the FMI distinguishes successfully between clinical cohorts of family members actively seeking psychological treatment and community control cohorts living in households unaffected by addiction (Cohen’s d > 1.20).

Predictive and Longitudinal Validity

The instrument has demonstrated pronounced sensitivity to change over time in longitudinal clinical intervention studies. In evaluations of the 5-Step Method delivered within primary healthcare and community addiction services (Copello et al., 2010), statistically significant reductions in FMI Total Impact, Active Disturbance, and Worrying Behavior scores were documented from baseline to three-month and twelve-month follow-up assessments (effect sizes ranging from d = 0.45 to 0.72), tracking concomitant reductions in the family members’ psychological strain and physical symptom scores.

8. Reliability

The Family Member Impact Questionnaire exhibits robust empirical reliability across diverse language adaptations, administrative formats, and clinical settings:

Internal Consistency

Estimates of internal consistency via Cronbach’s alpha have been consistently high across published literature:

• Total Impact Scale (16 items): Cronbach’s alpha coefficients routinely fall within the .84 to .91 range across validation cohorts. In the definitive psychometric report by Orford, Templeton, Velleman, and Copello (2005), the baseline alpha for the total scale was .88.
• Worrying Behavior Subscale (10 items): Cronbach’s alpha values typically span .80 to .87, indicating high internal coherence among items assessing chronic functional decline and systemic domestic strain.
• Active Disturbance Subscale (6 items): Internal consistency estimates for this shorter subscale range from .74 to .82 across studies, reflecting satisfactory homogeneity despite measuring varied externalized behaviors such as threats, theft, and communication failure.

Test-Retest Reliability

Test-retest stability was evaluated across stable, untreated community cohorts over two- to four-week intervals, yielding intraclass correlation coefficients (ICCs) between .78 and .85 for the Total Impact score, .76 for Active Disturbance, and .82 for Worrying Behavior. These findings indicate that while the instrument remains sensitive to therapeutic interventions, it reflects stable environmental conditions in the absence of active clinical change.

Inter-Item and Item-Total Correlations

Corrected item-total correlations across the 16 items generally exceed .40, with the majority falling between .48 and .68. Items addressing health worries (Item 14), mental state worries (Item 16), and interpersonal quarreling (Item 5) consistently exhibit the highest discriminatory power and item-total correlations.

9. Factor Analysis

The structural dimensionality of the FMI has been evaluated through exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across numerous cohorts spanning three decades of research:

Exploratory Factor Analysis (EFA)

In early exploratory investigations employing principal axis factoring with Promax (oblique) and Varimax rotations, researchers identified a distinct two-factor solution accounting for approximately 48% to 56% of the total variance. The two factors were conceptually and statistically defined as:

• Factor 1 (Worrying Behavior): Accounts for the largest proportion of common variance (~35–40%). Items with substantial factor loadings (> .50) include Item 13 (work/study concerns; loading .71), Item 14 (physical health worries; loading .78), Item 15 (neglected appearance; loading .65), and Item 16 (mental state worries; loading .74), alongside structural strain items such as financial depletion (Item 4; loading .58) and unreliability (Item 12; loading .54).
• Factor 2 (Active Disturbance): Accounts for an additional 12–16% of common variance. High-loading items (> .50) include Item 5 (picking quarrels; loading .73), Item 6 (threatening behavior; loading .68), Item 1 (changeable moods; loading .61), and Item 10 (upsetting family occasions; loading .59).

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic investigations evaluating the fit of the theoretical two-factor model against competing unidimensional models have consistently demonstrated the superior empirical fit of the correlated two-factor configuration. In a cross-national CFA validation study involving family members affected by alcohol and gambling problems:

• Chi-square to degrees of freedom ratio ($\chi^2/df$): 1.84 to 2.31, well below the conservative threshold of 3.0.
• Comparative Fit Index (CFI): Values consistently range between .92 and .96.
• Tucker-Lewis Index (TLI): Observed values between .91 and .95.
• Root Mean Square Error of Approximation (RMSEA): Values typically fall between .048 and .062 (with 90% confidence intervals spanning .038 to .071), indicating excellent to acceptable population approximation fit.
• Standardized Root Mean Square Residual (SRMR): Consistently beneath .055.

The latent correlation between the Active Disturbance and Worrying Behavior factors is moderately strong (standardized $\phi pprox .55 ext{–}.65$), confirming that while the two dimensions stem from a unified overarching syndrome of family addiction stress, they represent empirically distinct operational facets that should be scored and interpreted independently in clinical practice.

10. Instrument / Measurement Tool

• Instrument Name: Family Member Impact Questionnaire (FMI)
• Alternative Names: Family Impact Questionnaire; Stress Assessment Tool for Family Members
• Authors / Originators: Jim Orford, Richard Velleman, Alex Copello, and Lorrena Templeton
• Administering Institution: Alcohol, Drugs, Gambling and Addiction Research Group, School of Psychology, The University of Birmingham
• Instrument Type: Self-report questionnaire / standardized clinical rating scale
• Target Population: Adult and adolescent family members (spouses, partners, parents, adult children, siblings) residing with or maintaining regular contact with a relative who has problematic alcohol use, substance misuse, or behavioral addiction (e.g., compulsive gambling).
• Completion Time: Approximately 5 to 10 minutes
• Item Count: 16 items
• Subscales:
  • Active Disturbance: 6 items (Items 1, 2, 5, 6, 7, 10)
  • Worrying Behavior: 10 items (Items 3, 4, 8, 9, 11, 12, 13, 14, 15, 16)
  • Total Impact: All 16 items combined
• Response Scale: 4-point ordinal frequency scale:
  • Not at all = 0
  • Once or twice = 1
  • Sometimes = 2
  • Often = 3
• Don’t Know Imputation and Missing Data Rule:
  • Respondents who select Don't know receive a designated replacement value: for sensitive or covert items (Items 3, 6, and 7), Don't know = 1; for all other items (Items 1, 2, 4, 5, 8, 9, 10, 11, 12, 13, 14, 15, 16), Don't know = 2.
  • Critical Protocol Rule: If there are more than three (3) “Don’t know” responses across the entire questionnaire, the protocol must be deemed invalid and should not be scored.
• Scoring Instructions:
  • Total Impact: Sum of all 16 items (Theoretical range: 0 to 48). Higher scores denote more severe domestic and psychological impact.
  • Active Disturbance: Sum of items 1, 2, 5, 6, 7, and 10 (Theoretical range: 0 to 18). Higher scores indicate elevated interpersonal conflict, domestic volatility, and safety concerns.
  • Worrying Behavior: Sum of items 3, 4, 8, 9, 11, 12, 13, 14, 15, and 16 (Theoretical range: 0 to 30). Higher scores reflect pervasive anxiety regarding the relative’s somatic, psychiatric, occupational, and financial decay.

11. Permissions & Fee and Test Year

• First Conceptual Formulation: 1976 (published in investigations of marital cohesion and treatment outcome in alcoholism by Orford, Oppenheimer, Egert, Hensman, & Guthrie).
• Major Standardizations: 1999 (Velleman & Orford), 2005 (Orford, Templeton, Velleman, & Copello), and 2010 (Methods of Assessment for Affected Family Members).
• Copyright Holder: © Alcohol, Drugs, Gambling and Addiction Research Group, School of Psychology, The University of Birmingham, United Kingdom.
• Licensing and Accessibility: The FMI is released for open, non-commercial academic, clinical, and evaluative research. Practitioners and behavioral researchers may utilize and reproduce the scale free of financial charge, provided full academic attribution is rendered to the original authors and the University of Birmingham.
• Commercial and Digital Redistribution: Commercial entities, pharmaceutical research organizations, or commercial electronic medical record (EMR) software platforms wishing to incorporate the instrument must obtain written permission from the research group or AFINet (Addiction Family Support Network).

12. References

Copello, A., Templeton, L., Orford, J., & Velleman, R. (2010). The 5-Step Method: Evidence of gains for affected family members. Drugs: Education, Prevention and Policy, 17(S1), 100–112. https://doi.org/10.3109/09687637.2010.514798

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer Publishing Company.

Orford, J., Oppenheimer, E., Egert, S., Hensman, C., & Guthrie, S. (1976). The cohesiveness of alcoholism-complicated marriages and its influence on treatment outcome. British Journal of Psychiatry, 128(4), 318–339. https://doi.org/10.1192/bjp.128.4.318

Orford, J., Templeton, L., Velleman, R., & Copello, A. (2005). Family members of relatives with alcohol, drug and gambling problems: A set of standardized questionnaires for assessing stress, coping and strain. Addiction, 100(11), 1611–1624. https://doi.org/10.1111/j.1360-0443.2005.01219.x

Orford, J., Templeton, L., Velleman, R., & Copello, A. (2010). Methods of assessment for affected family members. Drugs: Education, Prevention and Policy, 17(S1), 75–85. https://doi.org/10.3109/09687637.2010.514783

Orford, J. (2014). Testing the short questionnaire for family members affected by addiction: Help required. AFINet Network Resources. https://www.afinetwork.info/

Velleman, R., & Orford, J. (1999). Risk and resilience: Adults who were the children of problem drinkers. Harwood Academic Publishers.

13. Items of the Scale (Questionnaire)