Therapeutic Relationship Utilizing the HAQ-2

Therapeutic Relationship

Utilizing the HAQ-2 to Examine the Therapeutic Alliance

The therapeutic relationship, sometimes referred to as the helping alliance or working alliance, is one of the most important predictors of psychiatric patient outcome. This relationship is used as a tool to help promote positive patient outcomes. Relationships are a two-way street, resulting either transference or a lack of connection between the therapist and the patient. This study will explore the ability of clinical psychologists to the health of the therapeutic relationship. The ability to accurately assess the therapeutic relationship plays an important role in the ability to adjust treatment regimes accordingly. The ability to repair an ailing relationship is an important part of the healing process. This research will explore the appropriateness of the Health Alliance Questionnaire-2 (HAQ-2) as an instrument for measuring psychologist ability to assess the therapeutic relationship in an upcoming study.

Part 1: Therapeutic Relationship or Client-Therapist Attachment?

Understanding the Elements of the Therapeutic Relationship

Although many other factors can affect ability of a client to reach therapeutic goals, the therapeutic relationship is one of the most important predictors of client outcome. The therapist must be able to motivate the patient to participate in the prescribed treatment, take their medications, and make the necessary lifestyle changes for their success. A healthy therapeutic relationship can be an excellent tool in achieving goals. In a positive relationship, the therapist acts as a coach for the patient. However, in an unhealthy relationship, the therapist can cause irreparable damage to the patient. Therefore, it is important for the therapist to understand and be able to assess their relationships with clients.

In order to better understand the dynamics of the therapeutic relationship, one must attempt to understand as much as possible about the factors that influence the relationship. The therapeutic relationship is an important predictor in individual outcomes during psychotherapy (Horvath & Bedi, 2002). The therapeutic relationship has also been shown to affect the outcome of couples in therapy under a number of circumstances, accounting for as much as 5-22% of the therapeutic outcome (Knobloch-Fedders, 2007). These sources indicate that inconsistency exists between studies that measured different phases of the treatment process. However, this does not undermine the importance of the therapeutic relationship in the final success of the therapy.

Many variables affect treatment outcome, and it may be difficult to determine how much of the outcome can be directly attributed to the therapeutic alliance. Llgen, Finney & Moos (2006) found that the therapeutic alliance could counteract the effect of low self-esteem in patients being treated for alcohol abuse. This study found that in patients with high self-esteem, the therapeutic relationship had little impact on outcomes. However, in patients with low self-esteem, the therapeutic relationship had the ability to change the outcome from a negative one to a positive one.

Several factors affect the ability to form a strong therapeutic relationship. For instance, the attitude of the patient and their internal motivation can have an impact on the ability to form a strong therapeutic relationship. Patients that are highly motivated and ready for change are more likely to form a positive therapeutic relationship than those that are resistant to change (BCR, 2006). Perfectionists also have problems with the therapeutic relationship (BCR, 2006). Anxiety and cognitive impairment can also affect the ability to form a positive therapeutic relationship (BCR, 2006). Patients that are better adjusted in the beginning of the therapy sessions are more likely to build a positive therapeutic relationship than those that have a poor relationship from the start (BCR, 2006). Patients bring certain things with them into the relationship and these things have a direct impact on the ability to form a positive therapeutic relationship.

Therapists must realize that the patient will bring with them some things that affect the ability to build a positive relationship. However, it was also found that therapists must bring certain things with them in order to make the relationship a success. Therapists who bring expertise and empathy with them are more likely to influence the relationship in a positive manner (BCR, 2006; Feller & Cottone, 2003). A therapist with these qualities will be able to make adjustments throughout the treatment in order to build a relationship.

The therapeutic alliance occurs in a number of treatment settings, even in the absence of face-to-face contact. Children who received cognitive-behavioral treatment via telephone formed a therapeutic alliance with their specialists, even though they had no face-to-face contact (Lingley-Pottie & McGrath, 2008). Both the therapist and the patient share in the responsibility for developing a positive therapeutic alliance (Meissner, 2006). The therapeutic alliance begins to develop in the assessment stage of treatment and carries through the entire treatment process (Hilsenroth, Peters, & Ackerman, 2004).

Several factors can help patients feel as if the therapeutic relationship is a positive one. They are getting little extras from the treatment, looking for common ground, feeling like the practitioner sincerely cares, practitioner availability, practitioner flexibility and the opportunity for the patient to have input into the treatment (Ware, Tugenber, & Dickey, 2004). Communication skills by the therapist cannot be emphasized enough in the development of a positive patient perception. Communication by the medical practitioner was found to be a facto in patient's overall satisfaction, their willingness to adhere to treatment plans, treatment outputs, and the likelihood of filing a malpractice claim (Cruz & Pincus, 2002). The study found that doctors could be more responsive to patient needs without lengthening visits (Cruz & Pincus, 2002).

These studies demonstrate the importance of developing a positive therapeutic relationship early in the treatment process; both the patient and the therapist play an important role in the therapeutic relationship. The patient can carry past emotional baggage with them, which can have a negative impact on the therapeutic relationship. The therapist must actively work to provide positive communication and make adjustments to promote a positive therapeutic relationship. These are the key elements in the therapeutic relationship that can affect treatment outcomes.

Understanding Client-Therapist Attachment

One can understand the importance of developing a positive therapeutic relationship, but sometimes, too much attachment can develop from a client standpoint. The therapeutic relationship requires that certain boundaries be set to define where the therapeutic relationship begins and when a taboo boundary has been crossed. Every person has certain boundaries in their lives that help to define their personal space. Often these boundaries are not well defined and it can be difficult to judge where another person's boundaries lie. However, these boundaries can play in important role in helping the therapeutic

The technical term for our personal space is the intersubjective field. These boundaries fall into the categories of propriety, space, behavioral, verbal and energetic (Rand, 2008). When boundaries are crossed unintentionally, both the therapist and the client can be emotionally harmed (Rand, 2008). In order to avoid this dilemma, the therapist must understand these various types of boundaries in themselves and in their clients. They must attend to them and make certain that they are not crossed. Making boundaries concrete can play an important role in making certain that undesirable attachments are not crossed (Rand, 2008).

The inability to adhere to behavioral boundaries begins in childhood. Crossing behavioral boundaries can result in resentment of the client or the therapist (Rand, 2008). Verbal boundaries are closely connected to social space (Rand, 2008). One of the more common examples of verbal boundaries occurs when the client asks the therapist a question. The therapist must decide whether to answer the client honestly, or to answer in a way that will best suit the client's needs. Physical boundaries mean the space around us and the proximity in which we feel comfortable with another person. Sometimes one person may feel uncomfortable and the other is unaware of it (Rand, 2008). The therapist must be aware of these various boundaries and make certain that they are not violated.

Client attachment to the therapist can stem from a memory of parental care giving (Woodhouse, Schlosser, & Crook, et al., 2003). Clients are more apt to develop an attachment to their therapist after a longer or ongoing therapy plan, than during treatment plans that are short duration (Woodhouse, Schlosser, & Crook, et al., 2003). Clients engaged in more transference when client-therapist attachment developed (Woodhouse, Schlosser, & Crook, et al., 2003).

The nature of therapy is such that the therapist seeks to develop relationships with clients that promote the sharing of intimate feelings (Parish, & Eagle, 2003). The therapist must seek to understand each patient's attachment style. In doing so, they can take steps to prevent the client from forming an unhealthy relationship (Parish, & Eagle, 2003). One must remember that therapy will not continue forever in most cases. At some point, the relationship will end. The client that has formed an unhealthy relationship with their therapist may experience the separation as a loss, causing more trauma in the long-term.

There has not been a significant amount of literature published on the unhealthy client-therapist relationship. It is generally considered a social taboo. Often the client is unable to take steps to avoid the undesirable emotional attachment. The therapist must take the initiative in maintaining proper distance and personal space. However, it is important to be aware that a positive therapeutic relationship could become too much of a good thing. When it does, a positive relationship can become toxic to the therapeutic outcome.

Comparing and Contrasting the Therapeutic Relationship and Client-Therapist Attachment

The therapeutic relationship and client-therapist attachment have many common elements, but the are major differences as well. Both the therapeutic relationship and the client-therapist attachment develop from the relationship between a therapist and their client. Research cited earlier, tells us that the development of a relationship is necessary for the success of the treatment plan. The more intimate the relationship becomes, the more likely it is to result in the type of shared secrets that result in positive therapeutic outcomes. However, it appears that this relationship can go too far and cause more harm than good.

When one talks about the therapeutic relationship, it is mentioned in a positive light. It is a natural part of the therapy process and should be encouraged. However, as we learned in our discussion of client-therapist attachment, one must be careful to keep proper distance and to avoid overt closeness. It would appear that there is an obscure line that cannot be crossed in the intimacy of the client-therapist relationship. When the relationship changes from a positive interaction to an attachment, it can be damaging to both parties and can have a negative, rather than a positive impact on the treatment process.

There are many differences that distinguish a therapeutic relationship from client-therapist attachment. The first is that studies have demonstrated that the therapeutic relationship develops early in the treatment cycle. It begins to develop during the initial assessment phase of the treatment continues to grow throughout the treatment process (Hilsenroth, Peters, & Ackerman, 2004). This differs from client-therapist attachment that typically only develops after a long-term treatment plan where a significant level of intimacy develops between the client and therapist (Woodhouse, Schlosser, & Crook, et al., 2003).

The most significant difference between a therapeutic relationship and client-therapist attachment is that the therapeutic relationship can be nurturing for the client. A positive therapeutic relationship has a positive outcome on the treatment outcome. A client-therapist attachment stems from this nurturing relationship, but has a negative impact on the treatment outcome. In the end, the relationship will end, which can lead to more trauma for the client and potentially even damage to the therapist as well.

Client-therapist attachment develops from a positive therapeutic relationship. During the course of treatment client-therapist attachment may have a positive influence on the treatment progress. It may lead to greater intimacy and ability to share with the therapist. It may lead to increased feelings of trust and a willingness to follow treatment prescriptions. This may be the result of feelings that the client does not wish to disappoint the therapist, so they follow the treatment as prescribed. These are all positive affects that stem from client-therapist attachment. However, these affects are short-lives and in the end, client-therapist attachment can lead to feelings of mistrust, anger, and resentment on both sides of the couch.

It is important to understand the differences between a positive therapeutic relationship and client-therapist attachment. It can be generally agreed that when the line is crossed and the relationship evolves into attachment, a good thing turns bad. However, the most difficult aspect of the differences in these relationships is knowing exactly when this line has been crossed. Often the transition from relationship to attachment progresses slowly, as feelings gradually increase. Due to this gradual increase in feelings, it is sometimes difficult, even for a seasoned professional, to recognize the signs that too much intimacy has developed in the relationship.

Research Rationale

The ability to maintain professional distance from clients is an important ethical consideration for therapists. They must be able to recognize the signs of too much intimacy from their clients and must be able to devise a way to transition the relationship back into a healthier state without harming the client. The most difficult aspect of this professional skill is to be able to recognize the early signs of a relationship gone wrong.

This study will explore the ability to therapists accurately assess the status of their relationship with their clients. It will explore their ability to recognize a negative relationship from a positive relationship. It will also examine their ability to determine when the client is transitioning from a positive therapeutic relationship into a more harmful client-therapist attachment. If therapists are unable to accurately distinguish these elements of the client relationship, they could cause more harm than good by continuing the therapy.

Part 2: Assessing Measurement Instruments

The purpose of this research is to develop a means to measure the factors that influence a therapist's ability to accurately assess the status of their client relationships. This study will lead to the development of criteria that therapists can use as a guideline to assess their client relationships and make certain that they remain healthy ones for all parties involved. The development of a meaningful measurement procedure is paramount to the ability of these guidelines. The following will explore various factors that are involved in the development of an appropriate measurement device for the study.

In the first part of this research study, we explored studies that addressed the therapeutic relationship. We explored the impact of the relationship on therapeutic outcomes and various factors that can affect its development. Over fifteen scales have been used in studies to assess the therapeutic relationship (McCabe & Priebe, 2004). However, it was found that no single scale was widely used in psychiatric research. These scales assessed various components of the therapeutic relationship and few used the same scale (McCabe & Priebe, 2004). This makes it difficult to measure the reliability and validity of the scale in research. This is the problem that will be confronted in the remainder of this research.

Relationship Measurement Instruments

As the importance of the client-therapist relationship became better understood, it led to the development of several early measures to assess the quality of this relationship. One of the first instruments was the Barrett-Lennard Relationship Inventory (BLRI) (Barrett-Lennard, 1962)

This instrument measures three factors that are important in the development of relationships. These factors are empathy, regard, and congruence (Ganley, 1989). This instrument is typically used to measure family relationships and marriage relationships (Ganley, 1989). It does not specifically address the issues that are specific to the development of the client-therapist relationship. However, these elements are reflected in the HAQ-2. One of the key drawbacks of the BLRI is its length.

Another instrument frequently used in the assessment of relationships is the California Psychotherapy Alliance Scales (CALPAS) (Gaston, 1991). This instrument is often used instead of the BLRI as a shorter alternative. The scale contained five subscales: patient working capacity, patient commitment, goal consensus, working strategy consensus, and therapist understanding and involvement (Gaston, 1991). This instrument consists of a 7-poitn scale in which the subject rates various items in Likert-type responses. This instrument is administered to both patients and therapists after every few sessions. This scale can be used to measure changes in the client-therapist relationship and has wider possibilities for application than the BLRI.

These scales are useful for the measurement of relationship factors in various situations, but they have drawbacks in relation to the proposed research. For instance, the BLRI is lengthy, which may deter potential clients from participating in the study. In addition, it does not specifically address the topic of client-therapist relationships. The CALPAS more closely addresses the topic of the study and is shorter in length. However, this scale is weighted towards factors that influence patient satisfaction. The focus of the proposed research study focuses on therapist perceptions, rather than patient perceptions. These are the primary reasons for deciding not to use either of these studies for the proposed research studies.

The Working Alliance Inventory (WAI) (Horvath & Greenberg, 1989) is another widely used questionnaire that measures the strength of the therapeutic alliance. It bears significant resemblance to the HAQ-2, only it has 12 items and uses a 7-point Likert scale. However, the HAQ-2 is more widely used. This means that much more is known about the reliability and validity of the HAQ-2.

The HAQ-I was one of the first scales to assess the therapeutic alliance between client and therapist. This early version of what would later evolve into the HAQ-2 included some question that reflects how much the patient benefited from the therapy (Luborsky, Crits-Christoph, & Alexander et al., 1983). These questions were later dropped from the scale, as some researchers did not feel that this represented a good condition from which to assess the relationship. These questions were more likely to address the quality of the treatment, rather than the relationship between the client and therapist. The HAQ-1 was rated on a 7-poitn scale and the newer HAQ-2 is rated on a 6-poitn scale. The key differences between the HAQ-1 and the HAQ-2 are in content changes.

The Helping Alliance Questionnaire (HAQ-2)

After an examination of various research instruments that are available for this research study, the Helping Alliance Questionnaire (HAQ-2) holds promise as a suitable instrument. This Instrument has been used extensively in a variety of research environments. It has been used under a number of research constraints as well. The HAQ-2 provides a quick overview of the patient's perceptions of the quality of the therapeutic alliance with their therapist (DeWeert-Van, DeJong, Jorg, & Schrijver, 1999).

The HAQ-2 examines the therapeutic alliance using a short Likert scale. It has a patient version and a therapist version. The purpose of this study is to determine the accuracy of the perceptions of the therapist, as they correlate with the perceptions of the client. This instrument allows a direct comparison between the perceptions of the client and the perceptions of the therapist.

Use of the HAQ-2 provides a clear method for conduct of the intended research study. The instrument would be administered to pairs of therapists and clients engaged in a therapeutic relationship. The method of administration could be modified to examine various stages of the therapeutic process. For instance, the HAQ-2 could be administered at the beginning of the therapeutic process, again in the middle of the process, and in the end of the process. This technique would allow the research to track changes in both client and therapist perceptions of the therapeutic alliance. An administration scheme such as this would allow the researcher to investigate the therapist's ability to accurately assess the therapeutic alliance as it undergoes changes through time. It would enable the ability to examine the ability of therapists to notice subtle changes that occur throughout the therapeutic process.

The HAQ-2 addresses a specific client-therapist relationship in which a questionnaire is given to both the client and the therapist. Both versions of the questionnaire mirror each other. For instance, the patient version states, "I felt that the therapist wants me to achieve my goals." The therapist version states, "The patient feels that I want him/her to achieve the goals." (Luborsky, Barber, & Siquiland et al., 1996). This makes comparison of patient responses and therapist responses easy.

Statistical analysis is easily achieved through side-by-side correlation of individual responses of a particular client-therapist pair. The ability of the therapist to properly assess the client-therapist relationship is easily scored by the number of correlates achieved on the scale. The higher number of correlates the therapist achieves, the more accurate their assessment of the relationship. This feature of the scale also makes comparison of responses over time easy to achieve.

Reliability and Validity of the HAQ-2

The selection of an appropriate research instrument must consider several measures of the quality of the instrument. Test-retest reliability is achieved by comparing the results for a certain individual given at two different times. The length of time between test periods has an affect on the outcome of the test-retest measure of reliability. Tests taken closer together often have higher test-retest reliabilities than two test re-administered over longer periods of time. Values of 0.81-1.0 are considered to be desirable,

The HAQ-2 is considered to have an excellent internal validity and test-retest reliability. It also demonstrates convergent validity with the CAPLAS (Luborsky et al., 1996). Le Bloc and associates (2006) compared the reliability and validity of the first and second versions of the HAQ-2. This study compared versions 1 and 2 of the HAQ with those of the WAI. The study also found that the French version of the HAQ-2 had similar construct validity to the English version.

Cronbach's alpha is a coefficient that ranged from 0 to 1, with a 0 indicating that the scale is not at all consistent. A Cronbach's coefficient of 1 would indicate perfect consistency. The closer, the scale comes to a 1, the more consistent it is. An acceptable Cronbach's alpha is one that ranges from between 0.80 to 0.95. Shorter scales have t tendency towards lower consistency than longer measurement instruments. However, this is not the case with the HAQ-2. This scale demonstrates exceptionally high internal consistency, particularly considering its brief length.

When compared with the English version of the instrument, the French versions had similar Cronbach's Alphas, ranging between 0.85 to 0.91 (Le Bloc, de Roten, & Drapeau et al., 2006). Luborsky et al. (1996) reported the internal consistency coefficient (Cronbach's alpha) of between 0.90 to 0.93. Test-retest reliability was reported at r=0.78 for the patient version of the questionnaire. The excellent internal validity of the HAQ-2 was compared to other therapeutic alliance measures. It was found that the HAQ-2 was similar to other measures (Le Bloc, de Roten, & Drapeau et al., 2006).

One of the key problems associated with the HAQ-1 was that it had items that appeared to assess the pre-treatment attitudes of the patient. One of the key goals of the development of the HAQ-2 was to reduce the ability of the scale to measure early clinical improvement. The other key factor in the revision was to incorporate various aspects of the therapeutic alliance that are directly related to the efforts of the client and therapist to collaborate efforts (Le Bloc, de Roten, & Drapeau et al., 2006).

The HAQ-2 incorporates only 5 items from the 11 items on the HAQ-1. It added 14 new items, with 5 of them worded negatively (Le Bloc, de Roten, & Drapeau et al., 2006). According to this study, the HAQ-2 successfully predicted the patient's satisfaction with the treatment, but it did not assess improvement in symptoms. It would appear that the revised version of the study met its goals in this respect. This improvement in the HAQ make it better suited for objectives of the proposed research study. The newer version of the HAQ directly related to the therapeutic alliance. It removes the confounding variable associated with early symptomatic improvement. This improvement makes it a more reliable instrument for measurement of the therapeutic alliance than its earlier version.

Comparison with Other Instruments

When compared to other research instruments that measure therapeutic alliance, the HAQ-2demonstrated a high correlation with the CALPAS (Le Bloc, de Roten, & Drapeau et al., 2006). However, the HAQ-2 was better at predicting outcome than the CALPAS (Le Bloc, de Roten, & Drapeau et al., 2006). The CALPAS was better at predicting retention than the HAQ-2 (Le Bloc, de Roten, & Drapeau et al., 2006). However, this is not expected to be a factor in the proposed research study. The purpose of the proposed study is to measure current therapeutic alliance and is not concerned with the future of these events.

Overlap between the CALPAS and the HAQ-2 varied from 12-76% (Le Bloc, de Roten, & Drapeau et al., 2006). Both instruments are an excellent choice for the measurement of the therapeutic alliance. One of the key difficulties in research design to investigate therapeutic alliance is choosing which instrument to use. They are close in terms of validity and reliability. The HAQ-2 is often the instrument of choice, as it has greater ease of use and is relatively short.

The HAQ-1 appeared to be more predictive of outcomes than the HAQ-2 (Le Bloc, de Roten, & Drapeau et al., 2006). However, one must question what was being measured in the HAQ-1, as 5 of the items were related to symptom improvement. It is logical that patients whose symptoms were improving would report that they were happier than those that were not improving. It is not difficult to see where the relationship between early improvement and improved final outcome could have an impact on one another. Patients that demonstrate early improvement of symptoms are more likely to have an improved outcomes. This leads one to believe that the HAQ-2 is a better instrument for measuring therapeutic alliance, as opposed to improvements in symptoms.

To illustrate the differences between the HAQ-1 and the HAQ-2, one example of the questions that were eliminated from the HAQ-2 is, "I have been feeling better recently" or " I believe that the treatment is helping me" (Le Bloc, de Roten, & Drapeau et al., 2006). These questions do not examine causality of the improvement. The patient may be feeling better because the medication is working, or because external circumstances in their life have changed in a positive manner. They do not necessarily reflect positive changes in the therapeutic relationship. The HAQ-2 was designed to measure the therapeutic alliance. It is more focused on this alliance than this earlier version.

If the HAQ-2 were a better measure of the therapeutic alliance than the earlier version, then one would expect to find similar scores as other research instruments. When compared to the WAI, a high degree of variance was found with the HAQ-2 (Le Bloc, de Roten, & Drapeau et al., 2006). Correlations were lower when the WAI was compared with the HAQ-1 (r=0.17 to 0.34) (Le Bloc, de Roten, & Drapeau et al., 2006). This lack of correlation with the HAQ-1 may be a result of the symptomology related questions that were dropped from the HAQ-2.

This finding supports the appropriateness of the HAQ-2 as the research instrument of choice for the proposed research study. High correspondence of the WAI with the HAQ-2, but not with the HAQ-1 may indicate that the content of the HAQ-2 is more closely related to the therapeutic alliance. Had this finding been reversed or absent from the analysis, one would have to question the appropriateness of the content in the HAQ-2. However, these differences in comparison of the two instruments with a similar instrument support the argument that the content changes in the HAQ-2 reflect a closer approximation of the therapeutic alliance.