Running head: CLINICAL RATING SCALE FOR CO-DEPENDENCY
The Development, Reliability, and Validity of a Clinical Rating Scale for Co-dependency(1)
Daniel Harkness, Ph.D., ACSW
Melaney Swenson, MSW
School of Social Work
Boise State University
Kathryn Madsen-Hampton, MSW
LDS Family Services
Richard Hale, MSW
I Think I Can Learning Center
Fairview Heights, Illinois
Keywords: Co-dependency scale; substance-abuse counselors.
This investigation examined the reliability and validity of a rating scale for co-dependency in substance-abuse treatment. The investigators developed an example-anchored rating scale to operationalize co-dependency as substance-abuse counselors construe it in practice, and recruited 27 counselors for a counterbalanced multiple-treatment experiment. Counselors were randomly assigned to one of four continuing-education workshops for rating-scale training, and asked to evaluate co-dependency in five video-taped cases. Semi-structured case interviews were videotaped with a male and a female from five adult populations to vary the gender and co-dependency of cases: (1) outpatients in treatment for addiction, (2) outpatient spouses, (3) members of Codependents Anonymous, (4) United States Bureau of Land Management Smoke Jumpers, and (5) college students majoring in business or economics. To control for gender effects, one workshop presented male cases, one workshop presented female cases, and two workshops presented cases of both genders. To control for order effects, the assignment of videotapes to workshops was randomized to counterbalance the order in which counselors viewed them. The findings suggest that the rating scale yields reliable and valid evaluations of co-dependency without appreciable gender bias.
The Development, Reliability, and Validity of a Clinical Rating Scale for Co-dependency
This is the second in a series of investigations that examine co-dependency through the eyes and ears of substance-abuse counselors (Harkness & Cotrell, 1997; Harkness, in press). Their perception of co-dependency is important for three reasons. First, many substance-abuse counselors believe that co-dependency plays a causal role in addictive behavior. Second, co-dependency is frequently a target for substance-abuse treatment. Third, substance-abuse counselors and the programs that employ them anecdotally seek third-party payments for co-dependency treatment. Benshoff and Janikowski (2000) identify the problem examined in this study.
[T]here is a growing, practical concern about developing a scientifically valid definition for codependency and its prevention, diagnosis, and treatment. Increasingly, both private and public funders of dependence treatment are demanding greater selectivity in treatment admission, diagnosis, and levels of care. Treatment must be based on an accurate diagnosis of client problems and planned in such a way to deliver an effective outcome in the most efficacious manner. Many insurers are limiting the number of available treatment days or episodes, and they are requiring that all treatment be based on diagnostic criteria from the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), the American Society of Addiction Medicine (ASAM) diagnostic and levels of care standards, or the guidelines of the International Classification of Diseases (ICD-10). However, codependency is not a recognized disorder under any of these classification systems. Some treatment centers have offered codependency treatment services to families by wrapping the costs of these services into the overall costs of individual care. Other centers have diagnosed family members, often inappropriately, with psychiatric diagnoses to bill for family codependency services. Adjustment disorders of adulthood or childhood and depressive disorder diagnoses have been used in this way. Unfortunately, neither one of these practices will hold up to either financial or clinical services audits, and both are ethically questionable. Still other facilities have opted to provide codependency family services without cost. While this is a laudable practice, it usually represents a significant financial and energy commitment by the facility, and it may not be affordable over the long term. (Benshoff & Janikowski, 2000, p. 160)
Defining Co-dependency: Multiple Meanings
What co-dependency means is widely disputed (O'Brien & Gaborit, 1992). Reviewing the professional literature for a core definition, Morgan (1991) found six conceptualizations of co-dependency, not consensus: (1) an emotional, psychological, and behavioral condition, (2) interpersonal reactivity and obsession with interpersonal control, (3) learned self-defeating behaviors, (4) suffering associated with attending to others, (5) an addictive disease, and (6) a preoccupation with others characterized by extreme dependency. To these definitions, Cermak (1986) added one more, conceptualizing co-dependency as a personality disorder that should be diagnosed on the basis of five signs and symptoms: (1) deriving one's own self-esteem from the feelings and behavior of others, (2) subordinating one's own welfare to the welfare of others, (3) impaired judgement and stress in close or distant interpersonal relationships, (4) interpersonal relationships with individuals that most persons seek to avoid, and (5) companion disorders of mood, anxiety, substance-abuse, and health status. Clearly, co-dependency has multiple meanings.
The Meaning of Co-dependency in Historical Context
The conceptualization of co-dependency as a putative mental disorder amenable to diagnosis and treatment emerged as an issue in the last quarter of the 20th Century. In the 60's, substance abuse appeared on the horizon as a gathering storm. In the 70's, the storm swept the nation. In response, the United States declared a social policy war on two fronts. On one front, the war addressed the supply of drugs entering the nation. On the second front, the war addressed demand in the form of human consumption. At the turn of the 21st Century, substance abuse is sometimes described as the nation's number one public health problem, and shows no sign of abating.
Three tactics were used to wage war against substance abuse on the front of consumption. The first tactic was one of moral persuasion; the nation mounted a series of campaigns designed to change public attitudes about the consumption of alcohol and other drugs. The second tactic was legislative and judicial; the nation criminalized substance abuse and sent consumers to prison. The third tactic was medical; substance abuse was defined as an illness needing medical treatment.
To marshal forces to serve in the war against drugs, the nation needed foot soldiers for the treatment campaign, but lacked the resources to train and to pay them. Public policy notwithstanding, the primary concentration of resources for medical treatment is held in the private sector. The best the nation could do was to recruit and credential inexpensive health-care providers and broker a mix of public and private payments for services rendered. It turned to Alcoholics Anonymous and the Twelve Step tradition of substance abuse treatment to recruit them.
Originally, co-dependency reflected a view of human behavior that the foot soldiers in the war against drugs brought with them. That view was formed in the Twelve-Step culture of Alcoholics Anonymous and Al-Anon (Staub & Kent, 1973; Stude, 1990; Tournier, 1979). It referred to a "recognizable pattern of behavior and attitudes characteristically found in family members of an alcoholic" (Cermak, 1986). At the heart of the pattern was enabling, which connotes aiding and abetting an addicted person's behavior through over-zealous helping (Asher & Brissett, 1988; Brown, 1988; Frank & Golden, 1992; Haaken, 1992; Haaken, 1993; Harper & Capdevial, 1990; Mulry, 1987). But as recruits from the Twelve-Step tradition became health-care providers, the meaning of co-dependency changed.
The growing health-care presence of the Twelve-Step tradition sparked the interest of popular culture. Talented speakers and writers found a surprisingly large market for co-dependency workshops and literature, and established a profitable cottage industry to address the demand. Rich narratives on the subject were published (e.g., Black, 1981; Friel, Subby, & Friel, 1984; Wegscheider-Cruse, 1985; Whitfield, 1984), and their authors achieved national prominence as co-dependency experts.
Before long, co-dependency checklists designed to help those who might be afflicted diagnose and treat their own illness began to appear (e.g., Engs & Anderson, 1988; Friel, 1985; Friel & Friel, 1987; Friel & Friel, 1988; Potter-Efron & Potter-Efron, 1989). The scope and prevalence of co-dependency mushroomed in concert. Like Sherwin-Williams paint, co-dependency seemed to be spreading, covering the earth.
Academic interest may have made matters worse. The academic community developed an heuristic interest in co-dependency as the construct flourished in the popular media. Social scientists sought to fit the construct they found in the popular literature to academic theories of human behavior (e.g., Bruss & Glickauf-Hughes, 1997; Collins, 1993; Gemin, 1997; Inclan & Hernandez, 1992; Loring & Cowan, 1997; Messner, 1997; Rice, 1992; Wells, Glickauf-Hughes, & Jones, 1999); feminist scholars sought to deconstruct co-dependency as a page in the text of oppression authored by men to subordinate women (e.g., Asher, 1992; Asher & Brissett, 1988; Collins, 1993; Finagrette, 1991; Frank & Golden, 1992; Haaken, 1992; Haaken, 1993; Harper & Capdevial, 1990; Katz & Aimee, 1991; Krestan & Bepko, 1990; Longino, 1993; Nelson-Zlupko, Kauffman & Dore, 1995; Swigonski, 1994); and methodologists designed self-report measures to operationalize co-dependency for empirical research (e.g., Clark & Stoffel, 1992; Cowan, Bommersbach, & Curtis, 1995; Crothers & Warren, 1996; Fischer & Crawford, 1992; Fischer, Spann, & Crawford, 1991; Fischer, Wampler, Lyness, & Thomas, 1992; Hinklin & Kahn, 1995; Irwin, 1995; Lindley, Giordano, & Hammer, 1999; Martin & Piazza, 1995; Meyer, 1997; O'Brien & Gaborit, 1992; Pidcock & Fischer, 1998; Roehling & Gaumond, 1996; Teichman & Basha, 1996; Wright & Wright, 1990; Wright & Wright, 1991). Two academic examples from the empirical literature illustrate how divergent conceptualizations of co-dependency yield different findings in research.
In one series of studies, Fischer, Spann, and Crawford (1991); Fischer and Crawford (1992); and Fischer, Wampler, Lyness, and Thomas (1992) defined co-dependency as a dysfunctional pattern of relating to others with "an extreme focus outside oneself, lack of open expression of feelings, and attempts to derive a sense of purpose [exclusively] through relationships" with others. The investigators developed a self-rating scale to operationalize that definition and used it to examine the association among co-dependency, family substance abuse, family functioning, risk taking, parent-child relationships, and offspring alcoholism in samples drawn from self-help groups and university students. Thus defined, the phenomenon of co-dependency was gender-free, and associated with neither chemical dependency nor dysfunctional family patterns.
In another study, Lyon and Greenberg (1991) operationalized co-dependency as women finding exploitive men attractive and offering them help. Their experiment drew daughters of persons with alcoholism or sober parents from a sample of university students and exposed them to requests for help by a male confederate under exploitive and neutral conditions. Daughters of parents with alcoholism found the exploitive man significantly more attractive, and offered him help at twice the rate of their counterparts from sober families. Thus defined, co-dependency has gender and signals family dysfunction.
In our view, the multiple meanings of co-dependency suggest that the phenomenon is largely a social construction, as much perceiver as perceived. Therefore, our investigation sought to unravel the social construction of co-dependency in substance abuse practice, where constructed meaning shapes the treatment that consumers receive.
To address the issues raised by Benshoff and Janikowsi (2000), we believe that academic research must examine co-dependency in the context of substance-abuse practice. But, thus far, we've seen little evidence of two-way communication between applied and heuristic concerns. Few substance-abuse counselors read the research published in scientific journals (Miller, 1985), and most of the samples used in co-dependency research have been drawn from college-student populations, not from populations of substance-abuse patients, family members, and other care takers. As a result, the systematic advancement and dissemination of clinically-relevant knowledge about co-dependency in the treatment of substance abuse has shown modest progress.
As non-partisan scientists, our interest in co-dependency had an applied clinical bias. However, disinterested in advancing one conceptualization of co-dependency over another, in this study we merely sought to operationalize the social construction of co-dependency as it is commonly used in clinical practice, responding to an unanswered challenge by Cermak (1986):
Efforts to define codependency have taken a range of perspectives. . . What is common to all of these approaches is the conviction that codependency exists independently within all members of a drug dependent family, and that many of its manifestations become more overt in the context of a committed relationship. . . Also common to all these definitions is that little attention has been paid to them by professionals outside the drug dependence field. To date, these definitions have failed to coalesce into an integrated conceptual framework that is able to be empirically tested. . .While the diagnosis of codependence has been of pragmatic value in the alcoholism field, it has not been integrated into the standard nomenclature set forth in DSM-III (American Psychiatric Association 1980). Such an inclusion would be premature until the following have been accomplished: (1) a definition of codependency with objective criteria for the diagnosis is developed on a level of sophistication at least equal to other diagnostic categories in DSM III; and (2) research using adequate diagnostic criteria is undertaken to verify the existence of codependency as a reliable and valid entity. . . There is no reason to expect exportation of the concept to the rest of the mental health field until it can be packaged into well-defined bodies of data and not merely described by eloquent metaphors. (Cermak, 1986, p. 16)
Cermak (1986) encouraged substance-abuse counselors to diagnose co-dependency as a Mixed Personality Disorder (APA, 1980) to energize co-dependency research. We agree that clinical co-dependency research is needed, but we have little faith that operationalizing co-dependency as a personality disorder will energize research. Fifteen years after Cermak's proposal, little research examining co-dependency as a personality disorder has appeared in the literature, and perhaps for good reasons. First, personality disorders are defined as pervasive and inflexible (APA, 1994), whereas anecdotal evidence from clinical practice suggests that co-dependency is responsive to treatment. Second, the diagnosis of personality disorders is unreliable (Heumann & Morley, 1990; Zimmerman, 1994), and unreliable measures are an impediment to empirical research. Third, the limited empirical evidence does not support Cermak's description of co-dependency as a personality disorder (Loughead, Spurlock, & Ting, 1998). Finally, anecdotal evidence from clinical practice suggests that the treatment of personality disorders is rarely reimbursed in the contemporary managed health-care environment.
Therefore, our goal was limited; we sought to operationalize and test co-dependency as an expert consensus in substance-abuse practice. Eschewing reliance on secondary sources, we made the assumption that the foot soldiers in the war against drugs are the experts who count in substance-abuse treatment, and our research was designed to examine the reliability and validity of their clinical judgement. To address the concerns raised by Benshoff and Janikowski (2000), however, we employed research methods used in the development of DSM-IV (APA, 1994). We began by asking substance-abuse counselors for rich descriptions of the putative mental disorder in clinical practice in order to operationalize a professional consensus and test inter-rater agreement (Francis, First, Widiger, Miele, Tilly, Davis, & Pincus, 1991; Widiger, Francis, Pincus, Davis, & First, 1991). Then we designed an experiment to validate their clinical judgement (Robins & Guze, 1970).
The First Investigation
In a prior study, we asked two questions germane to the present research: (1) what do substance-abuse counselors mean by co-dependency, and (2) to what extent do they agree? Assuming that co-dependency was a social construction that substance-abuse counselors use to evaluate cases for treatment, and that co-dependency described a continuum of human behavior that counselors have learned to perceive, we recruited counselors to describe and rank-order examples of co-dependency in their practice. Although Harkness and Cotrell (1997) have described that study in detail, a brief recapitulation may be helpful.
We invited 274 substance-abuse counselors associated with 40 organizations in urban and rural settings to participate in the study. In three waves, we drew our sample from lists of state-certified substance-abuse counseling programs and providers. In the Midwest, we asked the first 20 counselors to compose three 100-word vignettes describing high-, moderate-, and low-levels of co-dependency in clinical practice. We transcribed their vignettes to 5"X 7" printed note cards and sorted them into three 20-card decks, one deck for each degree of co-dependency.
We recruited a second wave of 29 counselors from the Midwest and Intermountain regions of the nation. Counselors were asked to rank order the cases in each deck from the most- to the least-co-dependent. This procedure formed the basis for calculating the coefficient of concordance among counselor rank-order judgements for each deck, and allowed us to establish the range-of-ranks for each vignette. We used the coefficient of concordance to measure inter-rater agreement, the degree to which substance-abuse counselors perceived co-dependency in the same light. The range of ranks for each vignette allowed us to determine which ones enjoyed the greatest inter-judge agreement.
The range of ranks was used to construct a full-spectrum continuum of co-dependency. The most distinguished vignettes from the three decks of cards were retained, and the least distinguished vignettes were discarded, on the basis of their rank-order range. The 14 most distinguished high-, moderate-, and low-co-dependency vignettes that survived were used to compile a full-spectrum deck of cards that reflected the social construction of co-dependency by the first 49 counselors.
A third wave of counselors was recruited from Inter-mountain and Pacific North West. We asked eleven counselors to rank-order the vignettes in the fourth deck of cards, and evaluated the reliability of their clinical judgement by calculating W, Kendall's coefficient of concordance (Kendall, 1955), across counselors, for a vignette-by-rank-order matrix of ranks. Across the full co-dependency spectrum, the coefficient of concordance among counselors was .78 (X2 = 111.48, p = .00). A second measure of reliability was the average Spearman rank-order correlation between paired judges (rpj = .75). Finally, we calculated the estimated reliability of pooled ranks (rpr = .97), a variant of the Spearman-Brown Prophecy formula, to predict the expected agreement between one set of pooled judgements and another (Taylor, 1968).
Across the full spectrum, the agreement among substance-abuse counselors about the meaning of co-dependency was reliable and stable, enjoying a statistically-significant degree of order. In view of the confusion about the meaning of co-dependency that exists in the literature, we found it astonishing that co-dependency might have a uniform meaning, across counselors with varied educational backgrounds, employed in public and private practice settings, as far as two-thousand miles apart, in urban, rural, and frontier environments.
The Development of the Idaho Co-dependency Scale
The promising inter-rater reliability of counselor judgements suggested a challenging replication. Whether counselors can reliably rank-order a deck of cards imprinted with brief case vignettes is one question; whether they can reliably rank-order richly-nuanced human cases is another question altogether. To conduct a more rigorous test, we used counselors' rank-ordered vignettes to operationalize their professional consensus and test it in clinical practice. We constructed an example-anchored rating scale that counselors could use measure co-dependency by quantifying the rank-order of cases.
The procedure we followed was developed by Taylor (1968), and refined by Taylor, Haefele, Thompson, and O'Donoghue (1970), to measure and enhance the reliability of clinical judgement. Taylor and his colleagues (1970) developed the procedure to improve inter-judge reliability, reduce response-set bias, reduce sources of error variance found in clinical practice, and mitigate halo effects, especially "when the raters are untrained, highly heterogenous, and base their judgement on different samples of behavior observed under varying circumstances" (Taylor, Haefele, Thompson, & O'Donoghue, 1970, p. 309). See Schwab and Heneman (1975) and Murphy and Constans (1987) for critical reviews of related methods.
The Idaho Co-dependency Scale (Idaho scale) is a vertical 100-point thermometer anchored by seven case vignettes. The scale was constructed with seven vignettes because no more would fit on an 8.5" X 14" sheet of paper. For this purpose, vignettes with the smallest range of ranks were selected from the full co-dependency spectrum. An analogue of the scale is shown in Figure 1, and a full-size Idaho scale is available from the first author.
Insert Figure 1 about here
Instructions for using the scale are printed on the back of the page. Cases are rated in three steps. First, the counselor is instructed to scan all seven vignettes, reading up and down from the mid-point vignette anchoring a co-dependency score of 50. In roughly one hundred words, the midpoint describes the case of a 38 year old white male who reports "going crazy" since his break-up with his girlfriend of eight years.
Second, the counselor is instructed to "sandwich" the case between two vignettes, selecting the pair of descriptions the case most nearly resembles. The first vignette should be more "co-dependent" than the counselor's case, and the second vignette should be less "co-dependent." For extraordinary cases, the "sandwich" may be formed by an extreme vignette and one end of the scale.
The final step requires the counselor to draw a horizontal line across the scale between the anchoring vignettes to position the case. The line is drawn to show how much closer the case "feels" to one vignette, relative to the other. The co-dependency score for the case is quantified by reading the number on the thermometer.
Questions for Research
Having operationalized what co-dependency meant to the rank-and-file substance abuse counselors who developed the construct and use it day-in-and-day out, we questioned the reliability and validity of their judgement in the evaluation of cases for treatment.
Materials and Methods
We tested their judgement by asking substance-counselors to use the Idaho scale to evaluate co-dependency in ten video-taped cases. A counterbalanced multiple-treatment experiment (Kazdin, 1992) was designed to vary the five cases that each counselor viewed. First, cases were recruited to participate in semi-structured videotaped interviews. Second, videotapes were selected to vary the gender of cases by workshop. Third, the order in which videotapes were viewed was randomized for each workshop. Finally, counselors were recruited and randomly assigned to one of the workshops. The materials and methods used in the study were reviewed and approved for human-subjects research by independent review boards in university and medical settings.
Cases were recruited from five adult populations: (1) spouses of adult outpatients receiving substance-abuse treatment were recruited to serve as traditional cases; (2) unrelated adult outpatients in substance-abuse treatment were also recruited as traditional cases; (3) members of Codependents Anonymous were recruited as recovering cases, (4) Bureau of Land Management Smoke Jumpers were recruited as normals for contrast, and (5) university students majoring in business or economics were recruited as normals as well.
Representation from diverse populations was desired to ensure that co-dependency varied. Cases were recruited from spouses and outpatients because these groups traditionally serve as case exemplars in the co-dependency literature. Although the term co-dependency was originally applied to women married to men with substance-abuse problems, recent constructions of co-dependency include persons of both genders and substance-abusers. Members of Codependents Anonymous were recruited because they accept the diagnosis as valid and describe themselves as recovering. Icons of rugged individualism in the Inter-mountain West, Smoke Jumpers earn their living by parachuting from airplanes to extinguish forest fires. Although this might invite playful contrasts and comparisons with traditional views of co-dependent behavior, we had a hunch that co-dependency would be low in this group, because Fischer, Wampler, Lyness, and Thomas (1992) have argued that co-dependency moderates risk-taking behavior. Business and economics majors were recruited on the hunch that goal-oriented students seeking employment in those market sectors would be less likely to manifest co-dependent behavior than others. Males and females were recruited because many observers have suggested that the diagnosis of co-dependency may be influenced by the gender of putative cases (Asher, 1992; Asher & Brissett, 1988; Collins, 1993; Finagrette, 1991; Fischer, Spann & Crawford, 1991; Frank & Golden, 1992; Haaken, 1992; Haaken, 1993; Harper & Capdevial, 1990; Katz & Amee, 1991; Krestan & Bepko, 1990; Longino, 1993; Nelson-Zlupko, Kauffman & Dore, 1995; Swigonski, 1994).
Cases ranged in age from 18 to 65 years ( = 41), and reported three ethnic backgrounds. Given the small size of our case sample, no purpose is served by providing additional descriptions that could identify individuals. We recruited cases for variation, not social representation.
A variety of methods was used to recruit cases for video-taped interviews. Treatment agencies were asked to refer outpatients and spouses, posters were tacked up in agency waiting rooms, and a press release was issued as well. The investigators attended open meetings of Co-dependents Anonymous to distribute fliers advertising the research. Smoke Jumpers were recruited with the help of a supervisor in the United States Bureau of Land Management, who agreed to distribute fliers by hand. Students were recruited with an advertisement posted beside a bank of elevators in a business-school building, and those volunteers were promised a $1 certificate toward a meal at McDonald's for completing the study. Recruitment continued until a male and a female was obtained from each group.
A 45-minute battery of paper-and-pencil tests was administered before each interview was videotaped. The Spann-Fischer Co-dependency Scale (Fischer, Spann, & Crawford, 1991) was administered to assess the convergent validity of the Idaho Scale. The Spann-Fisher Scale is a 16-item self-report instrument. Individual items are rated on a 6-point Likert scale, and then summed with two reversed items to describe co-dependency on a scale from a high of 96 to a low of 16. Spann-Fischer scores have been associated with membership in Codependents Anonymous, gender, self-esteem, locus of control, depression, relationship with parents, and anxiety (Fischer, Spann & Crawford, 1991); narcissism (O'Brien & Gaborit, 1992); parental co-dependency (Irwin, 1995); age (Crothers & Warren, 1996); treatment outcomes and education (Teichman & Basha, 1996); parenting style (Fischer & Crawford, 1992); powerlessness in relationships (Cowan, Bommersbach & Curtis, 1995); and risk-taking (Fischer, Wampler, Lyness & Thomas, 1992) - but not with parental chemical dependency (Crothers & Warren, 1996); the number of family addictions, the severity of dysfunction in the family-of-origin, or alcoholism (Fischer, Wampler, Lyness & Thomas, 1992); childhood trauma (Irwin, 1995); or family cohesion and adaptability (Teichman & Basha, 1996). The Spann-Fischer Scale has enjoyed good test-retest reliability ® > .80), and acceptable internal consistency (.62 < < .92), across studies.
A protocol was developed to standardize the interview, using questions suggested by the theoretical and empirical co-dependency literature, to elicit co-dependent behavior from subjects. The protocol was pilot-tested by videotaping several interviews and inviting consultants from the addictions-treatment community to review and critique it. Short of crossing the line between interviewing for research and providing clinical service, the protocol was revised and expanded to incorporate counselor suggestions.
Interviews were conducted by the first author in a small, plush university studio designed for videotaping distance-education lectures. An electronic device was used to disguise the identity of each case by introducing mild distortion to the visual image recorded. Cases were invited to read the printed protocol before the interview began, and seated to enable them to turn the recording equipment on and off at will. During the interview, each female was called "Mary," and each male was called "John." Interviews ranged from 15 to 35 minutes in length ( = 22.2 minutes).
The interview opened by asking each volunteer, "Can you say just a few things about yourself so that we can get some idea of who you are?", "Could you say a bit about the important people in your life?", and "Could you speak for a few minutes about any part that alcohol or prescription or illegal drugs may have played in your household when you were growing up?" A script of yes-or-no follow-up questions probed for substance abuse in the family-of-origin and intimate relationships, with requests for narrative descriptions of significant relationships.
The second phase of the interview was designed to elicit descriptions of how each case interacted with others in commercial, social, and intimate relationships. For example, each subject was asked to provide a brief example of a time when "you had to tell someone NO!", and of a relationship in which "other people are there for you as much as you're there for them."
The third phase of the interview asked subjects to describe themselves by endorsing or rejecting a series of adjectives (e.g., worried or afraid, empty or bored, bossy or unyielding), and by answering yes-or-no questions about their own substance-abuse history.
After reminding subjects how to turn off the recording equipment, the interview closed with debriefing. The interviewer offered to address any unanswered questions about the research, and observed that sometimes the interview raised personal questions. Each subject was offered a list of certified substance-abuse facilities and counselors to address any latent clinical issues that the interview raised.
A pilot study was conducted to estimate how many videotaped cases a counselor could observe before experiencing fatigue. Based on our findings, we asked counselors to view excerpts from three training tapes and evaluate co-dependency in five videotaped case interviews.
Counselors were randomly assigned to receive training and then evaluate five videotaped cases in one of four workshop conditions. One workshop was exposed to male cases, a second workshop was exposed to female cases, and the third and fourth workshops were exposed to cases of both genders. A coin-tossing procedure was used to assign the male or female from each of the three co-dependency groups and the two groups of normals to one or the other of the mixed-gender workshops. After those assignments were made, the order of cases was randomized within workshops.
Each workshop began with 1.5 hours of data collection and training in the use of the Idaho scale. First, counselors were asked to complete a five-page questionnaire, an instrument adapted from the last national survey of substance-abuse counselors (NIAAA, 1984). After the questionnaires were completed, each counselor received a set of self-report co-dependency measures and several copies of the Idaho Scale.
Training began with a review of the co-dependency measurement literature, including a description of the psychometric properties of the distributed measures. Then the investigators described the purpose of the present research, the procedures used to develop the Idaho scale, and how to use it for the evaluation of cases.
Counselors were asked to practice using the scale by rating excerpts from a common set of three training interviews. These included an interview with a spouse of an adult outpatient in substance abuse treatment, a member of co-dependents anonymous, and a student. After each tape was rated by the workshop participants, the counselors were invited to report, explain, and discuss their ratings with their colleagues. All workshops were exposed to the same set of training interviews and excerpts.
Three hours were reserved to obtain counselor ratings of the five research cases shown in each workshop. Each counselors received a new set of Idaho scale forms that were coded to cross-reference counselors, cases, and ratings. Counselors rated cases immediately after each tape was shown, and were asked to refrain from further case discussion until all counselors had rated all cases. Time was reserved at the end of each workshop to allow counselors to compare and discuss their ratings.
A state-wide list of 88 certified substance-abuse counselors was obtained from their professional organization. An advertisement for a series of four half-day continuing-education workshops in the assessment of co-dependency was mailed to each counselor. In bold letters, the advertisement declared "Your knowledge is your tuition," explained that volunteers would be randomly-assigned to workshops on one of four scheduled date and earn 4.5 continuing-education credits for their participation.
Thirty-one percent of the 88 counselors we recruited enrolled in the workshop (N = 27), nine males and 18 females. The ethnicities counselors used to describe themselves included Native American, Hispanic, and Caucasian. Counselors ranged from 28 to 70 years in age ( = 49.48), and had between 1 and 20 years of substance-abuse counseling experience ( = 8.7). Thirty percent of the counselors had earned baccalaureate degrees, and 52% had earned graduate degrees. Their disciplines included counseling (N = 14), rehabilitation (N = 1), social work (N = 7), psychology (N = 2), nursing (N = 1), and other (N = 2). Six counselors endorsed the Minnesota model of treatment, two reported a cognitive orientation, two reported a family-systems orientation, and the remaining counselors described their orientation as eclectic. Seventy-five percent of the counselors agreed that they were "in recovery;" 78% acknowledged that a close family member or "significant other" was currently abusing alcohol or other drugs; and 63% were self-described adult children of alcoholics.
The mean length of time required to view the five case interviews presented in each workshop was 111 minutes. The amount of time required to view videotapes was not significantly different across workshops (F = 1.40, p = n.s.).
The mean Spann-Fischer co-dependency score of the videotaped cases across workshops was 52.60, approximating the midpoint between a "high" score of 67.2 and a "low" score of 37.3 suggested by Fischer, Spann, and Crawford (1991). No significant between-workshop differences were found for Spann-Fischer measures of co-dependency in the cases presented (F = .042, p = n.s.).
Because the vignettes used to anchor the Idaho scale were derived from rank-ordered cases, we evaluated the scale with conservative two-tailed non-parametric statistics. Each of the 27 counselors in our sample observed 5 videotaped cases, producing 135 ratings. The highest Idaho score that counselors assigned was 93.5, and the lowest was 0. The mean Idaho scale rating was 34.17 across cases, counselors, and workshops. These values should be interpreted as mean ranks.
The reliability of the Idaho scale was assessed by calculating Kendall's coefficient of concordance (Kendall, 1955) across counselors on a case-by-rank-order matrix of ranks. The coefficient of concordance among the 27 substance- abuse counselors was W = .963 over 135 ratings (X2 = 130.03, p = .000). The counselors' rank-ordering of cases on the Idaho scale was very reliable.
The convergent validity of the Idaho scale was determined by calculating the Spearman rank-order correlation between counselors' ratings and case self-reports of co-dependency measured by the Spann-Fischer test. We found a robust and significant correlation between the two measures ( = .542, p = .01). This finding supports the convergent validity of the Idaho scale as an operational definition of counselor judgement.
The discriminant validity of co-dependency measures is usually established by calculating the association between self-reported co-dependency scores and indices of socio-economic status drawn from large samples of university students. That procedure assumes that (1) co-dependency varies in student samples, (2) that socio-economic status varies in student samples, and (3) that the co-dependency of students should not covary with their socio-economic status. Although the first assumption is theoretically reasonable, we questioned the second and third. To the extent that university students are socio-economically homogenous, it is not clear how widely their socio-economic status varies. And to the extent that co-dependency is associated with anxiety and depression (Roehling & Gaumond, 1996), negative self-confidence (Lindley, Giordano, & Hammer, 1999), and self-defeating characteristics (Wells, Glickauf-Hughes, & Bruss, 1998), it is hard to imagine how those attributes would not be associated with socio-economic status, even among relatively well-functioning university students, unless their socio-economic status was relatively invariable.
In contrast, the present study included self-supporting adults whose socio-economic status varied widely; our sample included students, seasonal workers, and professional people, for example. Moreover, our study included clinically-significant cases, among whom higher levels of co-dependency and other forms of pathology should be expected (Prest, Benson, & Protinsky, 1998). Consequently, we evaluated the discriminant validity of the Idaho scale by calculating associations between co-dependency ratings and the characteristics of the counselors who made them.
Neither counselors' age, gender, ethnicity, treatment setting, experience, professional discipline, educational achievement, model of practice, recovery status, nor their own past or present relationships with substance-abusing parents or significant others, were associated with their clinical judgement across 135 Idaho-scale ratings of case co-dependency. This lends support to the discriminant validity of the Idaho scale; it appears to be measuring case co-dependency, not the attributes of the counselors conducting the ratings..
Evidence for concurrent validity was obtained by performing a Kruskal-Wallis test to contrast Idaho scores across case groupings. Table 1 shows post-hoc pair-wise Kruskal-Wallis case-group comparisons, which found significant differences among counselor ratings of spouses, outpatients, members of Codependents Anonymous, Smoke Jumpers, and students (X2 = 69.68, p < .001). This lends support to the concurrent validity of the Idaho scale.
Insert Table 1 about here
Although post-hoc comparisons did not find spouses significantly more co-dependent than outpatients, and although Smoke Jumpers were not found significantly more co-dependent than students, the scatter plot of co-dependency scores shown in Figure 2 describes a trend consistent with co-dependency theory in clinical practice. Counselors described untreated spouses as more co-dependent than outpatients in treatment for addiction, outpatients more co-dependent than recovering members of Codependents Anonymous, those in recovery more co-dependent than Smoke Jumpers, and risk-taking Smoke Jumpers more co-dependent than students of balance sheets and profits. In theory, co-dependency increases as compulsive interpersonal helping behavior and substance abuse are coupled in the family millieu (Cermak, 1986).
Insert Figure 2 about here
Our experiment was designed to determine whether the gender of cases influenced Idaho ratings. To examine the issue, we conducted a Mann-Whitney test to compare the co-dependency ratings of the "male" and "female" workshops, and found that the co-dependency ratings of the two workshops were equivalent (U = 178, p = n.s.). A second Mann-Whitney test compared case ratings in gender-segregated workshops with ratings in workshops that integrated gender. The co-dependency ratings of the integrated and segregated workshops were also equivalent (U = 1984.5, p = n.s.). Finally, we conducted a third Mann-Whitney test to contrast counselors' ratings of male and female cases across workshops. In our study, counselors described male cases as significantly more co-dependent than female cases (U = 1693.5, p = .14).
To estimate the effect size of gender on Idaho scale ratings, we used parametric statistics to calculate the correlation between the gender of cases and counselors' ratings and obtain the coefficient of determination. Although the correlation was highly significant ® = - .23, p = .007), the effect size was modest (r2 = .05). The size of the gender effect appears too small to warrant clinical significance, and the finding may be spurious because it counters conventional wisdom about the gender of co-dependency. We are cautiously optimistic that the Idaho scale is unbiased for gender.
As an alternative to self-rating scales, the Idaho scale appears to operationalize substance-abuse counselors clinical judgement. Scientists, counselors, and programs may find the Idaho scale useful for quantifying co-dependent behavior to describe and compare pre-treatment and post-treatment levels of client functioning. Nevertheless, several caveats are in order. First, our findings reflect the behavior and judgement found in small non-random samples that may not represent cases and counselors at large; it remains to be seen whether our findings can be generalized to other cases, counselors, and settings. Second, we suspect that conducting semi-structured case interviews enhanced the reliability and validity of the scale; a different interview format may produce different findings. Third, the brief period of counselor training may have enhanced the reliability and validity of co-dependency ratings as well; untrained ratings may be less reliable and valid. Fourth, the Idaho scale cannot be used to discriminate between "normals" and clinically-significant cases; the scale is not normed. Fifth, ethnic minorities were under-represented among the anchoring vignettes of the Idaho scale and the counselors in our study; the scale may be biased for ethnicity. Subject to those limitations, copies of the Idaho Co-dependency Scale and the companion structured interview are available from the first author.Discussion
The first paper in this series reported that co-dependency was reshaping popular views of human behavior, social problems, and the interpersonal meaning of helping (Harkness & Cotrell, 1997). Among substance-abuse counselors, co-dependency has sometimes been viewed as the distal cause or prime mover of addictive behavior. Anecdotally, co-dependency has become one of the targets of substance-abuse treatment. On the other hand, co-dependency has also been viewed with suspicion by outside observers as a fanciful term that lay counselors socialized in the narratives of 12-step culture have adopted to appear more "professional," and sometimes as a term coined by men to pathologize women. We conceptualized co-dependency as a social construction, because the same putative phenomenon serves as a beacon of hope for unhappy persons, as a symbol of faith for those in recovery, as a target of derision for professional rivals, and as a symbol of oppression in the politics of gender.
As other investigators have done, we wondered whether co-dependency could be measured. However, we turned to substance-abuse counselors to examine the question, because counselors invented the term and use it to address what many believe to be the outstanding health problem in the nation. We sought to bridge the growing divide between clinical practice and academic science in substance-abuse research.
Among many substance abuse counselors, co-dependency denotes a mental disorder. When we began interviewing counselors for this study, six years after Cermak (1986) described co-dependency as a personality disorder, a number f our respondents believed that co-dependency would appear in the forthcoming edition of the Diagnostic and Statistical Manual, expanding the revenue stream of third-party payments for substance-abuse treatment services. Many counselors understood Cermak's challenge as a forecast. They appeared to over-estimate the clinical significance of co-dependency, and underestimate the scientific and political scrutiny that putative mental disorders undergo before the Diagnostic and Statistical Manual adopts them (Widiger, et al, 1991). Moreover, they seemed unaware that few health insurers are willing to pay for the treatment of Axis II diagnoses.
If we found them naive, they found us ignorant about substance-abuse counseling practice. Some counselors suggested that we were sequestered in an "ivory tower," completely "out of the loop," and uninformed about a phenomenon (co-dependency) and forthcoming event (the appearance of co-dependency in DSM-IV) of monumental clinical significance. Perhaps a few may have viewed us as inter-professional saboteurs or spies in competition for scarce treatment dollars. At times, our skepticism and their faith made us distrust one another.
When the fourth edition of the Diagnostic and Statistical Manual was published, co-dependency was not included among the mental disorders (APA, 1994). To the best of our knowledge, listing co-dependency as a mental disorder was never considered. Paradoxically, substance-abuse counselors may be more reliable judges of co-dependency than psychiatrists and psychologists are of the Axis II personality disorders that appear in DSM-IV. The diagnostic reliability of the antisocial personality disorder, for example, was examined in one of the 12 national field trials conducted in the development of DSM-IV (Widiger, et al, 1996). The procedures used in the field trial and the present research were similar; both studies employed semi-structured protocols to interview cases, conducted pilot training, asked clinicians to diagnose videotaped case interviews, and measured the inter-rater reliability of their judgements. The coefficient of inter-rater agreement for the diagnosis of anti-social personality disorder was r = .50 for persons in drug-treatment facilities and homeless shelters, r = .56 for inmates incarcerated in medium-security penal institutions, r = .38 for outpatients from methadone clinics, and r = 1.00 for psychiatric inpatients, using DSM-III-R criteria (Widiger, et al, 1996, p. 8). Heumann and Morley (1990) reported that the coefficient of inter-rater reliability for the diagnosis of borderline personality disorder was only r = .17 in a study that asked clinicians to evaluate standardized vignettes. And a recent review of the literature suggests that inter-rater agreement on Axis II diagnoses is modest at best, when corrected for agreement by chance (Zimmerman, 1994).
In contrast, it appears that the "diagnosis" of co-dependency by substance-abuse counselors may be reliable and valid. With minimal training in the use of the Idaho scale, counselors were able to distinguish adult outpatients and their spouses from members of Co-dependents Anonymous and normals, to distinguish between students and Smoke Jumpers, and to distinguish between students and those in recovery. Thus, when substance abuse counselors "diagnose" co-dependency, the distinction appears to have clinical significance.
Clinical significance notwithstanding, it is premature to argue that co-dependency should appear in future revisions of DSM-IV. The most recent revision of the Diagnostic and Statistical Manual relied on systematic reviews of the empirical literature, the reanalysis of large data sets, and diagnostic field trials to guide medical judgement (Widiger, et al, 1991). Because relatively few substance-abuse counselors have acquired the training and resources to mount sophisticated clinical research, we encourage those with a working investment in pursuing the issue to form research alliances with colleagues who do.
The adoption of a new diagnosis begins with the emergence of professional consensus. To the best of our knowledge, substance abuse counselors have not been surveyed to assess systematically whether they believe that a new diagnosis is needed. Another issue is whether co-dependency should be viewed as a personality disorder, as argued by Cermak (1986). The distinction between the personality disorders and Axis I disorders is "accompanied by troubling conceptual inconsistencies," and the Axis II disorders have become a target of discrimination in reimbursement decisions, as Frances et al (1991, p. 407) predicted. Moreover, the purpose of diagnosis is the differentiation of disorders in the service of treatment matching, and Morgan (1991) has argued persuasively that co-dependency speaks to a global malaise that overlaps with other disorders, which defeats the purpose of differential diagnosis. Furthermore, empirical evidence that explains the relationship between co-dependency and substance abuse is lacking. Is co-dependency a free-standing disorder that deserves treatment in its' own right, or is it an antecedent, a corollary, or a sequel of substance-abuse that co-responds to substance-abuse treatment? In that light, one of the anonymous reviewers of this manuscript asked whether "co-dependency" might be better understood as a set of behaviors elicited by stressful life condition than as a mental disorder. Finally, co-dependency connotes a relational problem between two or more people, and the "DSM provides diagnoses only for mental disorders in individual persons, not for relational problems" (Frances, et al, 1991, p. 410).
Alternatively, perhaps co-dependency is a "psychosocial condition manifested through a dysfunctional pattern of relating to others . . . characterized by . . . extreme focus outside of self, lack of open expression of feelings, and attempts to derive a sense of purpose through relationships, " as Fischer, Spann, and Crawford (1991, p. 2) have reasoned. After all, the social construction of co-dependency by substance-abuse counselors in this study was associated with the Spann-Fischer measure of that "trait-like condition;" we used the Spann-Fisher measure as an index of the concurrent validity of counselor judgements.
On the other hand, the Spann-Fischer scale explained less than 30% of substance-abuse counselors' judgement. We are not prepared to say what part of the remainder was due to error variance, nor what part is an indication that the clinical view of co-dependency should be distinguished from Fischer, Spann, and Crawford's (1991) theory and measurement model. This is an interesting question that deserves further study, and one that we will address in forthcoming research reports.
We are decidedly non-partisan on the issue of what co-dependency means, but our findings suggest that substance abuse counselors know it when they see it. Therefore, we are cautiously optimistic that our findings will stimulate challenging replications and encourage new lines of research. If replications confirm our finding that substance-abuse counselor assessments of co-dependency are reliable and valid, then perhaps diagnostic criteria for co-dependency can be formulated and tested as well. The next steps in that enterprise should distinguish co-dependency as a putative mental disorder from established disorders in DSM-IV, laboratory studies, follow-up research, and family studies (Robinson & Guze, 1970). We encourage investigators to enlist substance-abuse counselors as partners in co-dependency research. If our methods have successfully captured co-dependency as substance-abuse counselors perceive it, then there remains much to learn about what co-dependency is, what it is not, and the role that it plays in the process and outcomes of substance-abuse treatment.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th. ed.). Washington, DC: Author.
Asher, R. (1992). Women with alcoholic husbands: Ambivalence and the trap of codependency. Chapel Hill: The University of North Carolina Press.
Asher, R., & Brissett, D. (1988). Codependency: A view from women married to alcoholics. The International Journal of the Addictions 23: 331-350.
Benshoff, J., and Janikowski, T. (2000). The rehabilitation model of substance abuse counseling. Belmont, CA: Wadsworth.
Black, C. (1981). It will never happen to me. Denver: M.A.C. Publications.
Brown, S. (1988). Treating adult children of alcoholics: A developmental perspective. N.Y.: John Wiley & Sons.
Bruss, K., & Glickauf-Hughes, C. (1997). Co-dependency issues in the termination of lesbian relationships. Family Therapy 24 (1): 9-18.
Cermak, T. (1986). Diagnostic criteria for codependency. Journal of Psychoactive Drugs 18: 15-20.
Clark, J., & Stoffel, F. (1992). Assessment of codependency behavior in two health student groups. The American Journal of Occupational Therapy 46 (9): 821-828.
Collins, B. (1993). Reconstruing codependency using self-in-relation theory: A feminist perspective. Social Work 38: 470-476.
Cowan, G., Bommersbach, M., & Curtis, S. (1995). Codependency, loss of self, and power. Psychology of Women Quarterly 19(2): 221-236.
Crothers, M., & Warren, L. (1996). Parental antecedents of adult codependency. Journal of Clinical Psychology 52(2): 231-239.
Engs, R., & Anderson, D. (1988). The CODE Manual: Using the CODE checklist for determining co-dependency among university students (Unpublished research report). Washington, DC: Integrates Substance Abuse Consultations.
Finagrette, H. (1991). Alcoholism: The mythical disease. In D. Pittman & H. White, (Eds.). Society, culture, and drinking patterns reexamined (417- 438). New Brunswick, NJ: Rutgers Center of Alcohol Studies.
Fischer, J., & Crawford, D. (1992). Codependency and parenting styles. Journal of Adolescent Research 7: 352-363.
Fischer, J., Spann, L., and Crawford, D. (1991). Measuring codependency. Alcoholism Treatment Quarterly 8: 87-99.
Fischer, J., Wampler, R., Lyness, K., and Thomas, E. (1992). Offspring codependency: Blocking the impact of the family of origin. Family Dynamics of Addiction Quarterly 2: 1-12.
Frank, P., & Golden, G. (1992). Blaming by naming. Social Work 37: 5-6.
Friel, J. (1985). Codependence assessment inventory: Preliminary research tool. Focus on the Family and Chemical Dependency 8: 20-21.
Francis, A., First, M., Widiger, T., Miele, G., Tilly, S., Davis, W., & Pincus, H. (1991). An a to z guide to DSM-IV conundrums. Journal of Abnormal Psychology 100(3): 407-412.
Friel, J., & Friel, L. (1987). Uncovering our frozen feelings: The iceberg model of codependency. Focus on the Family and Chemical Dependency 10: 10-13.
Friel, J., & Friel, L. (1988). Adult children: The secrets of dysfunctional families. Deerfield Beach, FL: Health Communications.
Friel, J., Subby, R., & Friel, L. (1984). Co-dependency and the search for identity. Pompano Beach, Florida: Health Communications.
Gemin, J. (1997). Manufacturing codependency: Self-help as discursive formation. Critical Studies in Mass Communication 14: 249-266.
Haaken, J. (1992). A critical analysis of the co-dependence construct. Psychiatry 53: 396
Haaken, J. (1993). From Al-anon to ACTA: Codependence and the reconstruction of care giving. Signs 18: 321-345.
Harkness, D., & Cotrell, G. (1997). The social construction of co-dependency in the treatment of substance abuse. Journal of Substance Abuse Treatment 14(5): 473-480.
Harkness, D. (in press). Testing Cermak's hypothesis: Is dissociation the mediating variable that links substance abuse in the family of origin with offspring co-dependency? Journal of Psychoactive Drugs.
Harper, J., & Capdevial, C. (1990). Codependency: A critique. Journal of Psychoactive Drugs 22: 285-292.
Heumann, B., & Morley, L. (1990). Reliability of categorical and dimensional judgments of personality disorder. American Journal of Psychiatry 147(4): 498-500.
Hinklin, C., & Kahn, M. (1995). Psychological symptomatology in spouses and adult children of alcoholics: An examination of the hypothesized personality characteristics of codependency. The International Journal of the Addictions 30(7): 843-861.
Inclan, J., & Hernandez, J. (1992). Cross-cultural perspectives and codependence: The case of poor Hispanics. American Journal of Othopsychiatry 62(2): 244-255.
Irwin, H. (1995). Codependence, narcissism, and childhood trauma. Journal of Clinical Psychology 51(5): 658-665.
Katz, S., & Aimee, L, (1991). The codependency conspiracy. NY: Warner Books.
Kendall, M. (1955). Rank correlation methods. London: Charles Griffen & Company.
Krestan, J., & Bepko, C. (1990). Codependency: The social reconstruction of female
experience. Smith College Studies in Social Work 60: 217-232.
Lindley, N., Giordano, P., & Hammer, E. (1999). Codependency: Predictors and psychometric issues. Journal of Clinical Psychology 55(1): 59-64.
Longino, H. (1993). Feminist standpoint theory and the problems of knowledge. Signs:
Journal of Women in Culture and Society 19: 201-212.
Loring, S., & Cowan, G. (1997). Codependency: An interpersonal phenomenon. Sex Roles 36(1/1): 115-123.
Loughead, T., Spurlock, V., & Ting, Y. (1988). Diagnostic indicators of codependence: An investigation using the MCMI-II. Journal of Mental Health Counseling 20(1): 64-76.
Lyon, D., and Greenberg, J. (199 1). Evidence of codependency in women with an
alcoholic parent: Helping out mr. wrong. Journal of Personality and Social Psychology 61: 435-439.
Martin, A., & Piazza, N. (1995). Codependency in women: Personality disorder or popular descriptive term? Journal of Mental Health Counseling 17(4): 428-440.
Messner, B. (1997). Archetypal evolution and "new birth" from codependency. Communication Studies 48: 76-92.
Meyer, D. (1997). Codependency as a mediator between stressful events and eating disorders. Journal of Clinical Psychology 53(2): 107-116.
Morgan, J. (1991). What is codependency? Journal of Clinical Psychology 47: 720-729.
Mulry, J. (1987). Codependency: A family addiction. American Family Physician 35: 215-219.
Murphy, K., & Constans, J. (1987). Behavioral anchors as a source of bias in rating. Journal of Applied Psychology 72(4): 573-577.
National Institute on Alcohol Abuse and Alcoholism. (1984). Development of model professional standards for counselor credentialing (DHHS Publication No. ADM 281-82-0001).
Washington, DC: U.S. Government Printing Office.
Nelson-Zlupko, L., Kauffman, E., & Dore, M. (1995). Gender differences in drug
addiction and treatment: Implications for social work intervention with substance-abusing women. Social Work. 40: 45-54.
O'Brien, P., and Gaborit, M. (1992). Codependency: A disorder separate from chemical
dependency. Journal of Clinical Psychology 48: 129-136.
Pidcock, B., & Fischer, J. (1998). Parental recovery as a moderating variable of adult offspring problematic behaviors. Alcoholism Treatment Quarterly 16(4): 45-57.
Potter-Efron, R., & Potter-Efron, P. (1989). Assessment of co-dependency with individuals from alcoholic and chemically dependent families. Alcohol Treatment Quarterly 6(1): 37-57.
Prest, L., Benson, M., & Protinsky, H. (1998). Family of origin and current relationship influences on codependency. Family Process 37(4): 513-528.
Rice, J. (1992). Discursive formation, life stories, and the emergence of co-dependency:
"Power/knowledge" and the search for identity. The Sociological Quarterly 33: 337-364.
Robins, E., & Guze, S. (1970). Establishment of diagnostic validity in psychiatric illness: Its application in schizophrenia. American Journal of Psychiatry 126: 983-987.
Roehling, P., & Gaumond, E. (1996). Reliability and validity of the codependent questionnaire. Alcoholism Treatment Quarterly 14(1): 85-95.
Schwab, D., & Heneman, H. (1975). Behaviorally anchored rating scales: A review of the literature. Personnel Psychology 28: 549-562.
Staub, G., & Kent, L. (Eds.). (1973). The para-professional in the treatment of alcoholism
[a new profession]. Springfield, Ill.: Charles C. Thomas.
Stude, E. (1990). Professionalization of substance abuse counseling. Journal of Applied
Rehabilitation Counseling 21: 11-14.
Swigonski, M. (1994). The logic of feminist standpoint theory and social work research. Social Work 39: 387-393.
Taylor, J. (1968). Rating scales as measures of clinical judgment: A method for increasing scale reliability and sensitivity. Educational and Psychological Measurement 28: 747-766.
Taylor, J., Haefele, E., Thompson, P., O'Donoghue, C. (1970). Rating scales as measures of clinical judgement II: The reliability of example-anchored scales under conditions of rater heterogeneity and divergent behavior sampling. Educational and Psychological Measurement 36: 301-310.
Teichman, M., & Basha, U. (1996). Codependency and family cohesion and adaptability: Changes during treatment in a therapeutic community. Substance Use & Misuse 31(5): 599-615.
Tournier, R. (1979). Alcoholics Anonymous as treatment and as ideology. Journal of Studies on Alcohol 40: 230-239.
Wegscheider-Cruse, S. (1985). Choicemaking. Pompano Beach, Florida: Health Communications.
Wells, M., Glickauf-Hughes, C., & Bruss, K. (1998). The relationship of co-dependency to enduring personality characteristics. Journal of College Student Psychotherapy 12(3): 25-38.
Wells, M., Glickauf-Hughes, C., & Jones, R. (1999). Codependency: A grass roots construct's relationship to shame-proneness, low self-esteem, and childhood parentification. The American Journal of Family Therapy 27: 63-71.
Whitfield, C.(1984). Co-alcoholism: Recognizing a treatable illness. Family and Community Health 7: 16-25.
Widiger, T., Cadoret, R., Hare, R., Robins, L., Rutherford, M., Zanarini, M., Alterman, A., Apple, M., Corbitt, E., Forth, A., Hart, S., Kultermann, J., Woody, G., & Francis, A. (1996). DSM-IV antisocial personality disorder field trial. Journal of Abnormal Psychology 105(1): 3-16.
Wright, P., & Wright, K. (1990). Measuring codependents close relationships: A preliminary study. Journal of Substance Abuse 2: 335-344.
Wright, P., & Wright, K. (1991). Codependency: Addictive love, adjustive relating, or both? Contemporary Family Therapy 13(5): 435-454.
Zimmerman, M. (1994). Diagnosing personality disorders: A review of the issues and research methods. Archives of General Psychiatry 51: 225-245.
1. This investigation was supported by a faculty research grant to the first author from the Boise State University College of Social Sciences and Public Affairs.