یک تجزیه و تحلیل تطبیق تکنیک های رفتاردرمانی شناختی برای سبک های یادگیری
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|30049||2012||6 صفحه PDF||سفارش دهید||4140 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Behavior Therapy and Experimental Psychiatry, Volume 43, Issue 4, December 2012, Pages 1039–1044
Background and objectives To optimize the effectiveness of cognitive-behavior therapy (CBT) for each individual patient, it is important to discern whether different intervention techniques may be differentially effective. One factor influencing the differential effectiveness of CBT intervention techniques may be the patient's preferred learning style, and whether this is ‘matched’ to the intervention. Method The current study uses a retrospective analysis to examine whether the impact of two common CBT interventions (thought records and behavioral experiments) is greater when the intervention is either matched or mismatched to the individual's learning style. Results Results from this study give some indication that greater belief change is achieved when the intervention technique is matched to participants' learning style, than when intervention techniques are mismatched to learning style. Limitations Conclusions are limited by the retrospective nature of the analysis and the limited dose of the intervention in non-clinical participants. Conclusions Results suggest that further investigation of the impact of matching the patient's learning style to CBT intervention techniques is warranted, using clinical samples with higher dose interventions.
While cognitive-behavior therapy (CBT) has demonstrated efficacy for a variety of disorders (Butler, Chapman, Forman, & Beck, 2006), there remains room for improvement – a significant proportion of patients do not benefit from CBT and the mean improvement among responders may only be 20–50% (Westbrook & Kirk, 2005). Furthermore, the limited resources in routine clinical practice (White, 2008) and high drop out rates early in therapy (e.g., Bados, Balaguer, & Saldana, 2007) mean that there is a need to optimize the effectiveness of CBT for each individual patient, at the earliest opportunity. Recent research suggests that a variety of different single-sessions interventions (e.g., solution focused, exposure, motivational interviewing, CBT) can lead to clinically and statistically significant improvements (e.g., Perkins, 2006) to the extent that more than one-third of patients do not require any further intervention, and are satisfied with the intervention (see Bloom, 2001; Zlomke & Davis III, 2008 for reviews). As a route to increased therapy effectiveness, research has endeavored to match patients to particular kinds of therapy (Allen, Babor, Mattson, & Kadden, 2003; Giovazolias & Davis, 2005). Patient-treatment matching can be defined as a method of choosing between alternative treatment options based on particular patient characteristics that interact differentially with interventions to produce a more favorable outcome (Mattson et al., 1994). Patient-treatment matching has shown some promising results in matching patients' characteristics, such as personality traits, and coping style to different substance abuse treatments (e.g., Conrod et al., 2000; Karno & Longabaugh, 2007) and stress management interventions (e.g., Martelli, Auerbach, Alexander, & Mercuri, 1987). However, no research has looked at the impact of matching therapy technique to patients' learning style, a characteristic more commonly identified in educational environments. In the last three decades, the proposition that students learn in different ways has emerged as a prominent pedagogical issue within the field of education (Hawk & Shah, 2007). The individual's ‘learning style’ is their preferred mode of receiving and processing information, such as a preference for theoretical or practical methods of learning. Matching teaching methods to students' (Ford & Chen, 2001; Nor-Azan, 2009), supervisors' (Wolfsfeld & Haj-Yahia, 2010) and medical patients' (Arndt & Underwood, 1990) learning styles has been shown to maximize learning. While there are a number of conceptualizations of learning styles one of the most influential has been Kolb's (1984) theory of experiential learning and conceptualization of four modes of the learning process. Rainey and Kolb (1995) describe the four different learning styles, two of which are directly relevant to the data we report here. ‘Abstract Conceptualization’ indicates an analytical approach to learning that relies heavily on logical thinking and rational evaluation, with less benefit from ‘discovery’ learning approaches such as exercises and role-plays. In contrast ‘Active Experimentation’ indicates an active, ‘doing’ orientation to learning that relies heavily on experimentation, with more learning occurring when the recipient engages in relevant tasks. There are clear parallels with the broader fields of learning and education because CBT can be conceptualized as a process in which the patient learns (i.e., discovers new information in relation to existing beliefs or learns techniques to change beliefs or manage emotions) and the clinician teaches (Lightburn & Black, 2001) and educational principles are consistent with the overall didactic goal of CBT (Riess, 2002). Hence, we set out to investigate the effects of matching patients' learning styles with interventions in CBT. The matching hypothesis in psychotherapy research suggests that patients benefit more from therapeutic approaches and techniques that are similar to their specific cognitive or attitudinal styles (Babor, 2008). This implies that outcomes will be better when the intervention utilizes methods consistent with a patient's preferred learning style, because that is their natural, and therefore most efficient, way of processing information. If corrective information is encountered using the preferred mode, then processing load is reduced, with corresponding facilitation of acquisition and consolidation of the relevant information (Nor-Azan, 2009). Thus we hypothesized that patients may achieve more change in targeted beliefs and associated behaviors and symptoms when CBT interventions were matched to their preferred learning style, than when they were not matched. The current study set out to test this hypothesis using retrospective analysis of the data reported by McManus, Van Doorn, and Yiend (2011). We examined whether the impact of two common CBT interventions, behavioral experiments or thought records, was greater when participant's learning style was matched (i.e., favored active experimentation or abstract conceptualization, respectively) than when it was mismatched.
نتیجه گیری انگلیسی
Preliminary analyses revealed that assumptions of normality were not violated hence parametric analyses are used throughout and all tests are two-tailed. 3.1. Intervention checks The participants categorized as AE learning style (n = 24) scored significantly higher than those categorized as AC learning style (n = 35) on the AE subscale of the LSI (means [SDs] 36.67 [6.08] vs. 26.09 [0.96] t (57) = −6.82, p < 0.001) and significantly lower on the AC subscale (means [SDs] 26.38 [5.54] vs. 39.06 [0.80], t (57) = 9.45, p < 0.001). There was no difference in the proportion of participants receiving the TR or BE interventions in the ‘matched’ (14 BE: 17 TR) and ‘mismatched’ groups (17 BE: 11 TR) χ2 (57) = 1.43, p = 0.23. Nor was there any difference in the alliance (STAR means = 36.00 [6.50] vs. 34.44 [8.10] t = 0.80 p = 0.43) and credibility of the interventions (CEQ means = 31.42 [7.41] vs. 30.89 [6.41] t = 0.32 p = 0.75) between the matched and mismatched group. 3.2. Participant characteristics Twenty-eight participants (the “mismatched group”) received an intervention mismatched to their learning style i.e., those that scored higher on the AC learning style but received the BE intervention and those that scored higher on the AE learning style but received the TR intervention. The remaining 31 participants (the “matched group”) received an intervention matched to their learning style i.e., BE intervention and AE learning style or TR intervention and AC learning style. To compare the matched and mismatched participants' characteristics χ2 tests and independent t tests were used (see Table 1 for means and standard deviations). T-tests showed that the differences between the matched and mismatched groups' pre-intervention scores approached significance on the IBI, OBQ-44, and Belief 1 (ps = 0.07–0.15). Hence, analysis of change score is the preferred method of analysis ( Maris, 1998; Oakes & Feldman, 2001). Change scores were calculated by subtracting the post-intervention or follow-up score from the pre-intervention score so that positive change scores are indicative of improvement in belief ratings, anxiety, avoidance or symptoms and are reported in Table 2. Table 1. Participants' characteristics and mean pre-intervention scores (standard deviations in parentheses) for the ‘mismatched’ and ‘matched’ group. Measure (scale) ‘Mismatched’ ‘Matched’ Statistic p-value Characteristic Gender (frequency) χ2(57) = 0.27 0.60 Women 22 26 Men 6 5 Age (years) 24.85(9.91) 22.39(6.91) t(56) = 1.11 0.27 Ethnic group (frequency) χ2(57) = 1.52 0.68 Caucasian 19 24 Other 9 7 Highest education (frequency) χ2(57) = 2.37 0.67 BSc or more 10 13 A-levels or less 18 18 Previous therapy (frequency) χ2(57) = 1.33 0.25 Yes 8 5 No 20 26 Pre-intervention ratings Belief 1 ‘Not washing your hands after going to the toilet will make you ill’ (1–9) 6.75(1.14) 7.17(1.02) −1.47 0.15 Belief 2 ‘Not washing my hands after going to the toilet will make others ill’ (1–9) 6.00(1.98) 6.10(1.77) −0.20 0.84 Situational anxiety rating (1–9) 5.43(2.06) 5.55(2.00) −0.23 0.82 Situational avoidance rating (1–9) reversed 8.11(0.74) 7.97(1.62) 0.42 0.68 Pre-intervention scores on standardized measures Obsessional beliefs questionnaire (OBQ-44) 166.25(44.44) 149.45(32.84) 1.61 0.1 Irrational beliefs inventory (IBI) 152.43(20.05) 143.39(17.18) 1.87 0.07 Table options Table 2. A comparison of the mean amount of change achieved by the ‘mismatched’ and ‘matched’ groups (standard deviations in parentheses). Measure (scale) Time point Group t (53)= p-value Effect size Cohen's d ‘Mismatched’ ‘Matched’ Ratings Belief 1 ‘Not washing my hands after going to the toilet makes me ill.’ (1–9) Pre-post 1.71(1.46) 2.37(1.38) −1.75 0.09 0.46 Pre-follow 1.62(1.50) 2.34(1.20) −1.99 0.05 0.53 Belief 2. ‘Not washing my hands after going to the toilet makes others ill.’ (1–9) Pre-post 1.17(1.87) 1.70(1.76) −1.09 0.28 ns. Pre-follow 1.38(1.58) 2.03(1.27) −1.69 0.09 0.45 Situational anxiety rating. (1–9) Pre-post 0.61(1.47) 0.29(2.12) 0.66 0.51 ns. Pre-follow 0.96(1.48) 0.67(2.04) 0.61 0.54 ns. Situational avoidance rating. (1–9) reversed Pre-post 0.89(1.60) 0.55(1.21) 0.94 0.35 ns. Pre-follow 0.46(1.33) 0.90(1.77) 1.03 0.31 ns. Standardized measures OBQ-44 Pre-post 11.11(17.58) 8.71(14.44) 0.58 0.57 ns. Pre-follow 13.34(20.07) 11.47(17.64) 0.41 0.68 ns. Irrational beliefs: IBI Pre-post 4.14(8.83) 0.90(6.11) 1.65 0.1 ns. Table options 3.3. Associations between learning styles and change scores Within the BE condition (n = 31) the AE learning style (matched) was significantly positively correlated with change in one of the belief ratings (Belief 2) from pre- to post-intervention (r = 0.36, p = 0.05), and pre- to follow-up (r = 0.38, p < 0.05). In contrast, the AC learning style (mismatched) was negatively correlated with change in this belief rating from pre- to follow-up, although this correlation narrowly missed significance(r = −0.36, p = 0.06). Within the TR condition (n = 28) the AC learning style (matched) was significantly positively correlated with belief change (Belief 1) from pre- to post-intervention (r = 0.50, p < 0.01), and pre- to follow-up (r = 0.59, p < 0.001). In contrast, the AE learning style (mismatched) was not significantly correlated (r = 0.26, p = 0.18) with outcome for participants that received the TR intervention. There were no significant correlations between learning styles and change in anxiety and avoidance ratings. 3.4. Comparison of the change achieved by ‘matched’ and ‘mismatched’ groups T-tests were used to compare the size of change achieved by the matched and mismatched groups. For the primary outcome measure (belief rating: Belief 1) the matched group achieved significantly more change than the mismatched group from pre-intervention to follow-up (t = 1.99, p < 0.05, d = 0.53). Similarly, the difference between the change achieved by the matched and mismatched groups from pre-intervention to post-intervention approached significance in favor of the matched group (t = −1.75, p = 0.09, d = 0.46) for Belief 1. For Belief 2 there was a trend that approached significance for the matched group to achieve more change than the mismatched group from pre-intervention to follow-up (t = 1.69, p = 0.09, d = 0.45). There were no significant differences between the matched and mismatched groups in the change achieved in ratings of anxiety and avoidance, or on the IBI or OBQ-44.