|Year : 2021 | Volume
| Issue : 2 | Page : 56-63
Bottom-up and top-down cognitive rehabilitation following mild traumatic brain injury - Occupational therapists' perspective: An online survey study
Asha K Vas, Anna Luedtke, Eryn Ortiz, Marsha Neville
School of Occupational Therapy, Texas Woman's University, Dallas, TX, US
|Date of Submission||18-Jan-2021|
|Date of Acceptance||21-Jun-2021|
|Date of Web Publication||23-Jul-2021|
Asha K Vas
5500 Southwestern Medical Avenue, School of Occupational Therapy, Texas Woman's University, Dallas, TX 75235
Source of Support: None, Conflict of Interest: None
Background: Mild traumatic brain injuries (mTBIs) can result in long-lasting cognitive deficits. Comprehensive rehabilitation efforts need to integrate varied training approaches to address cognitive challenges and subsequently improve long-term functional outcomes. Objectives: The goal of this pilot study was to learn about occupational therapists' interpretation and practice of bottom-up and top-down cognitive rehabilitation activities for adults with mTBI. Study Design: An online survey study was conducted. Methods: A TBI expert team developed a case study with 30 functionally based treatment activities that closely align with the current rehabilitation practice. Of the 30 activities, the team identified 15 as bottom-up and 15 as top-down based on the definitions in the literature and clinical experience. The case study was sent in an online survey format to Occupational Therapists (OTs) across the country. Results: Out of 67 participants who agreed to participate, 30 completed the survey. The majority of the participants (87.5%) indicated that they were familiar with bottom-up and top-down treatment activities. Descriptive analyses found that the agreement between the authors and participants on bottom-up activities was 47.5% and on top-down activities was 77.56%; significant Intraclass Correlation Coefficient (0.939, 95% confidence interval [0.903, 0.966], P < 0.001) on all the items suggests that there was nearly perfect agreement in all item scores between the participants. These responses were consistent across OTs with varying levels of education, experience, and geographical locations. Conclusions: Increased knowledge and awareness of the distinction between bottom-up and top-down activities could equip clinicians with effective training tools to improve cognitive training. Future studies could address limitations of the study, including small sample size, descriptive statistical approach, and potentially biased view of the treatment activities by the authors.
Keywords: Bottom-Up, Cognitive Rehabilitation, Mild Traumatic Brain Injury, Top-Down
|How to cite this article:|
Vas AK, Luedtke A, Ortiz E, Neville M. Bottom-up and top-down cognitive rehabilitation following mild traumatic brain injury - Occupational therapists' perspective: An online survey study. Indian J Occup Ther 2021;53:56-63
|How to cite this URL:|
Vas AK, Luedtke A, Ortiz E, Neville M. Bottom-up and top-down cognitive rehabilitation following mild traumatic brain injury - Occupational therapists' perspective: An online survey study. Indian J Occup Ther [serial online] 2021 [cited 2022 Nov 30];53:56-63. Available from: http://www.ijotonweb.org/text.asp?2021/53/2/56/322175
| Introduction|| |
Nearly 75% of traumatic brain injuries (TBIs) are considered mild in nature. Although labeled mild, long-term sequelae are not necessarily mild. More than half of individuals with mild TBI (mTBI) report functional difficulties. Cognitive impairments include deficits in the areas of executive function, learning/memory, attention, and processing speed.,, Cognitive rehabilitation often focuses on compensatory and/or restorative approaches to remediate cognitive difficulties. Terms such as top-down and bottom-up approaches are also used to refer to compensatory and restorative approaches, respectively. “Bottom-up” interventions build or restore basic skills using rote practice, while “top-down” interventions use metacognitive skills, or “thinking about thinking,” to promote effective self-management of cognitive difficulties. Extensive clinical and empirical evidence supports use of top-down and bottom-up approaches to functional training.,,,, That is, “real-world” functioning draws upon both bottom-up and top-down cognitive skills. The complex top-down skills including self-monitoring, executive functions, and use of metacognitive strategies are all mediated by bottom-up foundational skills of attention, arousal, basic memory, and processing skills.
Occupational therapy practitioners (OT) create individualized intervention plans to help clients with mTBI, to resume participation in their daily life including areas such as activities of daily living, work, education, and social participation. OT interventions are designed to compensate or remediate memory, learning, self-awareness, executive functions, and attention in the context of how it impacts the client's daily occupations. Similar to physical injury issues, challenges in cognitive rehabilitation are deciding on the approach, such as if the interventions should be remedial or compensatory or task-specific to achieve a client's goal. In addition, there is the challenge of examining whether learning a specific task will generalize to other tasks with similar demands promoting improvement in cognitive function. One area of evidence requiring further understanding is OTs knowledge of top-down and bottom-up approaches. This could help guide OTs to design interventions that are best suited to help their clients' specific goals. Therefore, the aim of this pilot study was to (a) learn about OTs interpretation of top-down and bottom-up approaches in mTBI cognitive rehabilitation and (b) identify participants' agreement with authors' interpretation of top-down and bottom-up strategies.
| Methods|| |
The current research was conducted via an online survey study design, utilizing a questionnaire with both closed-ended and open-ended questions. The first and last author, faculty, and researchers in neurorehabilitation, each with over 15 years of experience in TBI rehabilitation, developed a case study-based questionnaire for the current project. There was no word limit on the length of the responses. If the participant was not familiar with bottom-up and top-down approaches in the context of mTBI cognitive rehabilitation, the participant did not have the opportunity to proceed further with the survey.
Participants included occupational therapy practitioners in the United States. The target population was occupational therapy practitioners who were affiliated with the American Occupational Therapy Association and Occupational Therapy-Neuroscience group of the American Congress of Rehabilitation Medicine. The authors believed that this target group could reach the practitioners in the field of TBI rehabilitation.
A questionnaire was developed by the authors (first and last authors) with feedback from a team of clinicians, researchers in the field of TBI rehabilitation, and biostatisticians (see Appendix A). This team included four occupational therapists (including first and last authors) and two speech language pathologists, all with clinical expertise in TBI cognitive rehabilitation and research expertise in neuroplasticity. The goal of the study and the case scenario were presented to the team. The authors met with the clinical team multiple times, both individually and as a group to refine and finalize the case study and the intervention strategies. Based on the case study, the team identified six target domains to address that could enhance function, all of which could overlap to some degree. The six domains included family activities (with wife and children), household chores, organizational strategies (e.g., to-do lists), information processing, faith-based activities, work, leisure (e.g., pet care, baseball coaching), and therapeutic exercises (e.g., memory games for fun, that could be used as preparatory activities to improve cognition). Several intervention strategies were discussed with the team. Finally, the team agreed upon 30 strategies in total (15 bottom-up and 15 top-down) that addresses all domains of the case study. Furthermore, the authors also factored in the length of the questionnaire and time commitment from potential participants (i.e., clinicians).
The questionnaire is based on a case study of a 41-year-old man with an mTBI sustained 6 months ago. Details including demographics, roles and routines at home, work, and community participation were elaborated for the reader. The survey begins with open-ended questions asking the participant to elaborate on their interpretation of top-down and bottom-up approaches.
Followed by this, the authors provided definitions of the approaches, as follows:
- Bottom-up approaches in cognitive rehabilitation target domains such as (a) memory, (b) processing speed, and (c) attention. Therapeutic/functional activities may involve “repetition/recall exercises,” “drill and practice,” and “attention training” exercises
- Top-down approaches in cognitive rehabilitation target domains such as (a) abstract thinking, (b) problem-solving, and (c) inhibition. Therapeutic/functional activities may include “strategy training” and “metacognition” exercises.
Next, the survey in no specific order introduced 30 functionally relevant intervention strategies/activities that would normally be considered part of rehabilitation interventions. Participants rated the intervention strategies/activities on a 4-point Likert scale with the following choices: “bottom-up,” “somewhat bottom-up,” “somewhat top-down,” and “top-down.” A sample of the intervention strategies are listed below:
- Bottom-up strategies: “Identify 'who,' 'what,' and 'when' from a newspaper article”
- Top-down strategy: “Identify the 'take home message' of the newspaper article.”
The study was conducted in accordance with the ethics committee regulation and approval from the institutional review board (IRB). The survey was created and distributed via PsychData, an online web-based data collection program. The online written informed consent was included in the questionnaire; the participant got to read the consent before taking the survey. PsychData is approved by the IRB and is used extensively for web-based social science research. The study was approved by the IRB in November 2018 (Protocol # 20300). Once the survey was agreed upon by all team members and experts, the survey was distributed via AOTA's CommunOT forum and ACRM's OT-Neuroscience distribution list. The survey was posted on PsychData for 3 months. A reminder was sent to the distribution list after 5–6 weeks of posting the survey. Since the survey was posted/sent to a large distribution list, we do not have an exact number of participants who received the survey and or initiated the survey. A total of 30 participants completed the survey, whose data were analyzed for the current study.
Data Analyses: Likert Scale Questionnaire
Given the small sample size and to establish moderate power, the four-point Likert scale was reduced to a dichotomous scale of bottom-up and top-down. “Completely bottom-up” and “somewhat bottom-up” ratings were combined into “bottom-up category;” and “completely top-down” and “somewhat top-down” was combined in “top-down category.” Percentage agreement and disagreement between authors and participants interpretation of bottom-up and top-down strategies were calculated. The level of significance was set at P < 0.05 at the outset of the study. All analyses were conducted in IBM Corporation. (2017). IBM SPSS Statistics for Windows, version 25.0 [Computer software]. Armonk, NY: Author.
| Results|| |
The demographics of the participant sample (n = 30) is presented in [Table 1].
Findings: Open-Ended Questions: Definitions of Bottom-Up and Top-Down
The majority of the participants (87.5%) indicated that they were familiar with bottom-up and top-down approaches. Participants' definition of each bottom-up and top-down approaches, were on average two sentences in length (range of 1–4 sentences). Word count of the definitions of bottom-up (mean ± standard deviation [SD] = 30.11 ± 17.61) and top-down (mean ± SD = 27.33 ± 16.10) was comparable between domains (95% confidence interval [CI] [11.64, 30.15], P = 0.54). Below is a sample of the narrative responses [Table 2].
Findings: Agreement on Intervention Strategies
Descriptive analyses found that that the agreement between the authors and participants on bottom-up strategies/activities was 47.5% and on top-down strategies/activities was 77.56%. Among the 15 strategies/activities that were identified as “bottom-up” by the authors, six strategies/activities were identified as top-down by a number of participants. For example, the strategy of “Help with emptying dishwasher and folding laundry” was identified as bottom-up by the authors, but nearly 75% of the participants called it top-down. The strategy of “Recommended buying frozen or prepared breakfast items” was identified as bottom-up by the authors, but 76% of the participants called it top-down [Table 3].
No significant differences were found between authors' and participants' agreement percentages based on demographic variables, including education, years of being an OT, experience working with TBI populations, and the practice setting. The agreement across the participants was also examined. The results found a significant ICC (0.939, 95% CI [0.903, 0.966], P < 0.001) on all the items, suggesting that there was nearly perfect agreement in all item scores between the participants. The participants also had substantial agreement on the “top-down” items (0.872, 95% CI [0.756, 0.949], P < 0.001) and on the “bottom-up” items (0.954, 95% CI [0.915, 0.981], P < 0.001).
| Discussion|| |
An integrative approach to neurorehabilitation combines bottom-up and top-down approaches to exert a synergistic impact on improving overall day-to-day functioning. While bottom-up approaches are important for strengthening foundational cognitive domains, top-down approaches provide generalizable strategies that are relevant for daily life tasks. Occupational therapists incorporate top-down and bottom-up strategies/activities in functional contexts to facilitate and optimize adaptation and integration at home, community, school, and work. Consistency in use of terminology could make therapy more beneficial for the client and improve communication between therapists. The percentage agreement on bottom-up was low (<50%) indicating that either the authors' examples of bottom-up were unclear or the participants' definitions of bottom-up are inconsistent with authors' definition. Inconsistency on bottom-up approaches could be secondary to evolving definitions of top-down and bottom-up in rehabilitation context.
Historically, cognitive domains were viewed as building blocks that guided functions. Therefore, cognitive deficits were regarded as the primary cause of dysfunction, and remediation of these deficits (e.g., attention process training) would automatically help the individual return to function. Empirically, a positive relationship was demonstrated between cognition and function. However, predictability of function from cognition and vice versa was not very successful. Nonetheless, cognitive remediation was regarded as a primary target for training, and addressing these cognitive deficits was considered a top-down approach to rehabilitation. On the other hand, functional remediation or “neurofunctional approach” was considered a bottom-up intervention by some authors. This approach was predominantly employed for participants with TBI with significantly more cognitive challenges and those who were unlikely to develop self-care or community independence skills spontaneously. Intervention focused on learning by “doing” in a structured setting, strengthening existing skills and motivation, errorless learning approach and repetition, all of which are considered bottom-up approaches.,
In the last 15–20 years, definitions of bottom-up and top-town cognitive interventions have evolved. Bottom-up approaches are considered beneficial for foundational cognitive domains such as “attention” and top-down approaches target higher-order domains such as “executive functions.” While attention and other skills such as basic information processing can be remediated though repetition practice, higher order skills are trained in a top-down fashion through the application of a series of problem-solving and reasoning strategies that can be generalized across situations with practice. The bottom-up approaches are also referred to as “restorative approaches” and top-down as “compensatory approaches.” In the context of occupational therapy, top-down interventions are also referred to as “activity interventions.” That is, these interventions focus on improving the performance of everyday activities, which in turn can facilitate improvements in underlying foundational and preparatory neurocognitive function., A possibility for the low disagreement between the authors and the participants on bottom-up strategies could have been that any occupation-based intervention was considered top-down whether or not it involved higher level thinking. Thus, it is evident that the terminology, in the context of cognitive rehabilitation, has changed over time. Bottom-up is associated with a “functional approach” (1970, 1980s),, including “foundational skill training,” “restorative,” and “specific cognitive domains” (1990s–date). On the other hand, “top-down” is associated with a “cognitive approach” (1970s, 1980s),, which includes “executive function training,” “compensatory,” and “activity interventions” (1990s–date)., There may be other terms associated with bottom-up and top-down that we are unaware of and may be specific to certain practice settings, which could have contributed to the low agreement with the authors.
In addition to the evolving terminology, the lower agreement rate could be secondary to the analyses of the data. The rating of the intervention strategy, which was initially on a 4-point scale, was later collapsed into two groups. Although we gave a definition, the dichotomy of “somewhat top-down” and “somewhat bottom-up” may appear similar, which could lead to erroneous finding.
Cognition does not occur in isolation, and neither do real-life activities. Therefore, cognitive rehabilitation/training requires the interplay of both bottom-up and top-down strategies. A bottom-up approach could assist in improving specific fundamental processes of attention and arousal through repetitive drill-based exercises, which are prerequisites for learning, memory, and other higher functions. However, bottom-up approaches alone are unlikely to generalize to untrained tasks. A top-down approach could complement that bottom-up skills to optimize higher level functional skills such as problem-solving, flexibility, and decision-making are generalizable to several daily life tasks. This evidence of generalized benefits to top-down training to daily-life relevant skills comes from several randomized control trials of cognitive training programs in adults with TBI. Some examples of the programs include goal management training, executive plus program, problem-solving therapy,, goal oriented attention self-regulation, strategic memory advanced reasoning training, and cognitive orientation to daily occupational performance.
Limitations and Future Directions
The current findings, while informative, require further validation by addressing at least five limitations. First, the case study that was part of the questionnaire may not be generalized to all mTBI. Future studies need to include more case studies with broader representation of cognitive and functional impairments. Second, the definition of bottom-up and top-down strategies/activities is based on authors' and their team's understanding of these terms. Use of comparable terminology and their definitions should be studied more as the definition of the terms may vary between therapists/centers. Third, the intervention strategies/activities that were listed in the questionnaire were described in one sentence, which may not have captured the comprehensive nature of an activity. Future studies should elaborate on the intervention strategies to assist participants accurately rate bottom-up and top-down activities. Fourth, the findings are based on the responses of a survey. The Likert-scale ratings do not allow the participants to justify their responses. Adding a focus group or an interview could validate the findings. At last, the small sample size did not lend itself to further analyses of reliability and validity of responses. Increased sample size could address this problem for future studies.
| Conclusion|| |
The findings of the pilot study helped us learn about occupational therapists' interpretation and practice of bottom-up and top-down cognitive rehabilitation activities for adults with mTBI. The majority of the therapists are familiar with bottom-up and top-down treatment strategies irrespective of the varying levels of education, experience, and geographical locations. Concerted effort to engage in reasonably challenging functional activities that offer opportunities to practice cognitive flexibility, identify and limit interferences, and most importantly, strategies/activities applicable to different situations and contexts, can improve long-term outcomes and improve quality of life in adults with mTBI.
We appreciate the support from Graduate Research Associate Award, Texas Woman's University.
Financial Support and Sponsorship
Conflicts of Interest
There are no conflicts of interest.
| Appendix|| |
Appendix A: Traumatic Brain Injury Questionnaire
Demographics Please circle the answer that best describes you
Years working as an OT: 0–3 3–6 6–10 10+
Years of experience working with TBI populations: 0–3 3–6 6–10 10+
Current work setting: Rehab Center Outpatient Clinic Hospital SNF Other
Highest Level of Education: Bachelor's MOT Advanced Masters OTD PhD
Part 1 –Definitions
1. Are you familiar with bottom-up and top-down approaches to TBI cognitive rehabilitation? Yes __ No__
If yes, please provide your definition and/or an example of bottom-up and top-down approaches to cognitive rehabilitation:
Bottom-up approaches in cognitive rehabilitation target domains such as (a) memory, (b) processing speed, and (c) attention. Therapeutic/functional activities may involve “repetition/recall exercises,” “drill and practice,” and “attention training” exercises.
Top-down approaches in cognitive rehabilitation target domains such as (a) abstract thinking, (b) problem-solving, and (c) inhibition. Therapeutic/functional activities may include “strategy training,” “metacognition” exercises.
Part 2 – Case Study
John, a 39-year-old male is diagnosed with mild TBI. He is referred to OT for cognitive rehabilitation 3/wk. X 6–8 weeks.
- Sustained two TBIs in the last 2 years following separate motor vehicle crashes. The last TBI was 6 months ago. Symptoms included dizziness, headaches, and feeling “foggy.” His cognitive test scores were within normal range
- Was treated for headaches and dizziness. Recommended to get back to routine 'as tolerated'.
Prior functional history:
- Married, has two children ages 6 and 9, worked fulltime as a computer programmer.
- Roles and responsibilities: Household chores (e.g. vacuuming), child care (e.g. helping make breakfast for children, homework), taking the dogs for daily runs. Weekends were occupied with baseball coaching for his son's team, spending time with friends, attending church, helping with household chores, and managing finances. He considered himself a multitasker.
- Overwhelmed at work, especially when problem-solving new programming related issues.
- Difficulty completing one chore (e.g. folding laundry) without getting distracted (e.g. pending task of vacuuming); and consequently the 'to-do' list does not get done without help from his wife.
- Feels 'less energetic' to spend time with his children, dogs, and coaching responsibilities; and has thus cut his participation by nearly a third.
- Processing information (e.g. during Church service) is difficult, especially remembering details.
- His wife reports that John often feels frustrated with his level of performance at work and home, which has begun to impact their relationship.
Your OT treatment for John: Include 3-5 specific examples:
Part 3: How would you categorize the following treatment examples for John?
| References|| |
Prince C, Bruhns ME. Evaluation and treatment of mild traumatic brain injury: The role of neuropsychology. Brain Sci 2017;7:105.
McInnes K, Friesen CL, MacKenzie DE, Westwood DA, Boe SG. Mild traumatic brain injury (mTBI) and chronic cognitive impairment: A scoping review. PLoS One 2017;12:e0174847.
Rabinowitz AR, Levin HS. Cognitive sequelae of traumatic brain injury. Psychiatr Clin North Am 2014;37:1-11.
Williams WH, Potter S, Ryland H. Mild traumatic brain injury and postconcussion syndrome: A neuropsychological perspective. J Neurol Neurosurg Psychiatry 2010;81:1116-1122.
Babikian T, McArthur D, Asarnow RF. Predictors of 1-month and 1-year neurocognitive functioning from the UCLA longitudinal mild, uncomplicated, pediatric traumatic brain injury study. J Int Neuropsychol Soc 2013;19:145-154.
Cicerone KD, Mott T, Azulay J, Friel JC. Community integration and satisfaction with functioning after intensive cognitive rehabilitation for traumatic brain injury. Arch Phys Med Rehabil 2004;85:943-950.
Giles GM. Cognitive versus functional approaches to rehabilitation after traumatic brain injury: Commentary on a randomized controlled trial. Am J Occup Ther 2010;64:182-185.
Gordon WA, Cantor J, Ashman T, Brown M. Treatment of post-TBI executive dysfunction: application of theory to clinical practice. J Head Trauma Rehabil 2006;21:156-167.
Malec JF. Impact of comprehensive day treatment on societal participation for persons with acquired brain injury. Arch Phys Med Rehabil 2001;82:885-895.
Fleming JM, Ownsworth T. A review of awareness interventions in brain injury rehabilitation. Neuropsychol Rehabil 2006;16:474-500.
Haskins EC, Cicerone KD, Trexler LE. Cognitive Rehabilitation Manual: Translating Evidence-Based Recommendations into Practice. Reston, VA: ACRM Publishing; 2012.
Jaber AF, Hartwell J, Radel JD. Interventions to address the needs of adults with postconcussion syndrome: A systematic review. Am J Occup Ther 2019;73:7301205020p1-7301205020p12.
Stephens JA, Williamson KN, Berryhill ME. Cognitive rehabilitation after traumatic brain injury: A reference for occupational therapists. OTJR (Thorofare N J) 2015;35:5-22.
Skidmore ER. Activity interventions for cognitive problems. Pediatr Blood Cancer 2014;61:1743-1746.
Sohlberg MM, Mateer CA. Effectiveness of an attention-training program. J Clin Exp Neuropsychol 1987;9:117-130.
Smith-Knapp K, Corrigan JD, Arnett JA. Predicting functional independence from neuropsychological tests following traumatic brain injury. Brain Inj 1996;10:651-661.
Scherzer BP. Rehabilitation following severe head trauma: Results of a three-year program. Arch Phys Med Rehabil 1986;67:366-374.
Giles GM, Ridley JE, Dill A, Frye S. A consecutive series of adults with brain injury treated with a washing and dressing retraining program. Am J Occup Ther 1997;51:256-266.
Parish L, Oddy M. Efficacy of rehabilitation for functional skills more than 10 years after extremely severe brain injury. Neuropsychol Rehabil 2007;17:230-243.
Cicerone KD, Langenbahn DM, Braden C, Malec JF, Kalmar K, Fraas M, et al
. Evidence-based cognitive rehabilitation: Updated review of the literature from 2003 through 2008. Arch Phys Med Rehabil 2011;92:519-530.
Willard HS, Spackman CS, Crepeau EB, Cohn ES, Schell BA. Willard and Spackman's Occupational Therapy. Wolters Kluwer Health/Lippincott Williams and Wilkins; 2009.
Boake C. A History of cognitive rehabilitation of head-injured patients, 1915 to 1980. J Head Trauma Rehabil 1989;4:1-8.
Wilson BA. Cognitive rehabilitation: How it is and how it might be. J Int Neuropsychol Soc 1997;3:487-496.
Dams-O'Connor K, Gordon WA. Integrating interventions after traumatic brain injury: A synergistic approach to neurorehabilitation. Brain Impair 2013;14:51-62.
Levine B, Schweizer TA, O'Connor C, Turner G, Gillingham S, Stuss DT, et al
. Rehabilitation of executive functioning in patients with frontal lobe brain damage with goal management training. Front Hum Neurosci 2011;5:9.
D'Zurilla T, Nezu A. Social problem-solving. In: Kendall PC, editor. Advances in Cognitive-Behavioral Research and Therapy. New York, NY: Academic Press; 1982:201-74.
Rath JF, Simon D, Langenbahn DM, Sherr RL, Diller L. Group treatment of problem-solving deficits in outpatients with traumatic brain injury: A randomised outcome study. Neuropsychol Rehabil 2003;13:461-488.
Chen AJ, Novakovic-Agopian T, Nycum TJ, Song S, Turner GR, Hills NK, et al
. Training of goal-directed attention regulation enhances control over neural processing for individuals with brain injury. Brain 2011;134:1541-1554.
Vas A, Chapman S, Aslan S, Spence J, Keebler M, Rodriguez-Larrain G, et al
. Reasoning training in veteran and civilian traumatic brain injury with persistent mild impairment. Neuropsychol Rehabil 2016; 26:502-531.
Dawson DR, Gaya A, Hunt A, Levine B, Lemsky C, Polatajko HJ. Using the cognitive orientation to occupational performance (CO-OP) with adults with executive dysfunction following traumatic brain injury. Can J Occup Ther 2009;76:115-127.
[Table 1], [Table 2], [Table 3]