Poster-Abstracts

Background and Purpose: Great efforts have been made to find effective treatments for back pain. However, the effectiveness of a treatment may be dependent on patient compliance. This study investigated prospectively whether demographic factors, clinical factors, external barriers, satisfaction with care, patientâs perception of disability, quality of life, depression, and health beliefs are predictive of compliance with a physical therapy program.

Subjects and Methods: All participants with low back pain answered a questionnaire at an initial physical therapy evaluation and were followed during the treatment. This was a observational prospective cohort study, that used three indicators of compliance: attendance at scheduled physical therapy sessions, following a home exercise program, and watching back education videotapes. The study employed a battery of standardized instruments such as: Oswestry scale, The Spitzer Quality of Life Index, The Center for Epidemiologic Studies Depression Scale, and Health Locus of Control Scales. Student t-tests for independent samples and Chi-square tests were used to determine differences between 2 groups of subjects: non/low compliant or high compliant. Logistic regression was used to estimate adjusted odds ratios.Ê

Results: The results demonstrated that 51% of patients were non/low compliant with low back pain treatment. The rate of drop-out with physical therapy appointments was 50% and there were differences in compliance behavior according to each specific treatment regimen. Poor compliance was positively associated with unsatisfaction with care (p = < 0.01) and the presence of possible problems in following the treatment proposed (p = < 0.01).

Conclusion: The findings of this study indicate that compliance with physical therapy treatment for low back pain is a serious and complex problem, and confirm the need for continued research in this area. Assessing the degree of compliance with treatment for low back pain is a problem because there is a lack of standard measures or methods for assessing compliance.ÊÊ

Key Words: Low back pain, Compliance, Physical therapy treatment.

This study was approved by the Institutional Review Board of New York University Medical Center (IRB-NYUMC).

This research was funded by a grant from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS/ 5 RO1 AR 44288 04)), and by Funda‹o de Amparo ˆ Pesquisa do Estado de S‹o Paulo (FAPESP).

Neusa Maria Costa Alexandre, RN, Ph.D., Margareta Nordin, Dr.Sci., Hiebert, M.A., Marco Campello, P.T., M.A.

Department of Nursing, Faculty of Medical Sciences, State University of Campinas, Campinas, S.P., Brazil¹/ Occupational and Industrial Orthopaedic Center, Hospital for Joint Disease, Mt Sinai New York University Health, New York, NY, USA².

Corresponding address:

Neusa Maria Costa Alexandre, RN, PhD
Department of Nursing
Faculty of Medical Sciences
States University of Campinas ö UNICAMP
CidadeÊ Universit‡ria ãZeferino Vazä- Distrito de Bar‹o Geraldo
CEP 13083-970 ö Campinas ö S‹o Paulo
Brazil

Tel:Ê +55 19 788 8844
Fax: +55 19 788 8822
Email: Neusalex@headfcm.unicamp.br

Return top of page

A Comparison of Gender-specific Profiles Using the Personality Assessment Inventory in Patients with Chronic Back Pain

Marilee D. Allerdings

Poster #7, Fri. Dec 1, 2000, 9:45 - 10:00 am,
Sat. Dec. 2, 2000, 2:25 pm - 2:45 pm (Presenter at Poster 15)

Problem and Objectives: Assessments that focus on personality and psychopathology are fundamental to establishing appropriate diagnosis and treatment in multidisciplinary rehabilitation settings. To date, the Personality Assessment Inventory (PAI), a measure that includes both of these components, has been normed on community and psychiatric populations. The objectives of this study were to: a) derive a representative profile for chronic back pain, and b) examine if profiles differed for men and women.

Research Design: The research model was a descriptive study.

Methods: Research participants were 254 individuals (149 women, 105 men) with chronic back pain due to injury who were assessed in a multidisciplinary rehabilitation center. Injuries resulted from motor vehicle (n = 182) or work-related (n = 70) accidents (2 cases fit neither category). The mean age of participants was 37.4 (SD = 11.9), the average time since injury was 11.3 months (SD = 12.4), and the average level of education was 12.6 years (SD = 2.5). The PAI was administered as part of the psychological component of the multidisciplinary assessment. Only valid profiles were included.

Analysis: Raw scores for all scales were compiled and analyzed using the Statistical Package for the Social Sciences 10 for Windows. Following the generation of descriptive statistics, one-way Analyses of Variance were undertaken to detect any gender differences. T-scores for the PAI scales were also computed for both men and women using normative data from community and psychiatric samples.

Results: No significant differences were found on age, education, time since injury, or any of the validity scales of the PAI. Significant differences were found between the men and women on the raw scores for 9 of the 18 scales. Men were found to have significantly higher levels on the paranoia, antisocial features, alcohol problems, drug problems, aggression, and non support scales. Women reported significantly higher levels on the anxiety-related, treatment rejection, and interpersonal warmth scales. The t-score profiles based on the psychiatric normative data ranged between 41 and 56. With profiles based on community normative data all but one of the scales fell within 1 standard deviation of the mean. The somatization scale was mildly elevated (t = 61).

Conclusion: Our representative profile revealed no significant psychopathology. However, it did reflect notable concern with physical symptoms and slight elevations suggesting modest emotional distress. This finding is in keeping with other studies of the psychological concomitants of chronic pain. Statistically and clinically significant gender differences were evident in the areas of interpersonal style and treatment-related factors. The results from this study also provide support for using the PAI as a psychological assessment instrument in individuals with chronic pain. T-score profiles based on community normative data were more sensitive to psychological distress in our sample. Therefore, they may be a more appropriate scoring standard for this group than the psychiatric normative data.

Marilee D. Allerdings, Jason E. Peebles and Jody Burnett

Corresponding address:

Marilee D. Allerdings,
Functional Rehabilitation Program, Regina Health District,
Regina, Saskatchewan
Canada

Tel.: +1 306-766-5785
Fax: +1 306-766-5359
Email: jpeebles@reginahealth.sk.ca

Return top of page

The Sherbrooke Model: Implementation and Evaluation in the Netherlands

JR Anema

Poster #27, Fri. Dec 1, 2000, 4:00 - 4:150 pm, Sat. Dec. 2, 2000, 2:25Ê - 2:45 pm

Background: The impact of chronic back pain on society is enormous in terms of cost and disability. Chronic back pain is related to a complex mix of medical, psychosocial and occupational factors. Intervention on only one of these factors can not be expected to have a large impact in secondary prevention. An occupational & clinical back pain management program was evaluated in a randomised population based trial in Sherbrooke Canada (Loisel et al 1997). This back pain management program was developed on the basis of the report of the Quebec Taskforce on Spinal disorders (Spitzer et al 1987). In that report a schedule of desirable interventions was proposed to be implemented at specific times in the course of back pain.

We adjusted the Sherbrooke back pain management program & its interventions to the DutchÊ Health Care and Social Security System. In the next years it will be implemented and evaluated in a number of Occupational Health and Safety Services (OHS) and companies.

Objectives: To evaluate whether a back pain case-management program, adjusted to the Dutch situation, consisting of:

1. Occupational intervention;

2. Clinical intervention is effective in preventing chronic dysfunctioning.

Methods:

Inclusion: We will first recruit participating OHS, occupational physicians (OPâs), ergonomists/ occupational health nurses and companies. All companies are eligible for participation if they have more than 50 employees. Patients who consult their OP for low back pain are eligible for participation if sickleave duration is at least 2 weeks and less than 6 weeks. Additional selection criteria include: non-specific back pain. A total of 200 patients will be enrolled.ÊÊ

Interventions: The occupational intervention consists of an occupational medicine protocol by the OP and worksite intervention by an ergonomist/occupational health nurse according to the participative approach. The clinical intervention is a back school intervention based on a graded activity program including a cognitive behavioural approach.

Study Design: Randomised controlled trial. First randomisation öoccupational intervention- is performed at the level of the OP. Second randomisation ö clinical intervention- is performed at the level of the patient.

Outcome Assessments: Outcomes are assessed at 9, 26, 52 weeks after inclusion. Primary outcome measures are: duration of absence from regular/any work, Functional status Roland Disability Questionnaire (RDQ), pain intensity. Secondary:Ê Tampa Scale for Kinesiophobia (TSK), Pain Coping Questionnaire, Euroqol, OPs/patients satisfaction.Ê Also a process and a cost-effectiveness analysis is included.

Preliminary Results: 7 OHS, 49 OPâs and 25 ergonomists/occupational health nurses have agreed to participate in this study. Also 27 companies agreed to participate with a source population of 44.422 employees. The OPâs and ergonomists/occupational health nurses were recently trained in the occupational medicine and participative work adjustment protocol, respectively. Inclusion of patients will start in September 2000.

Discussion:Ê Adjustment of the interventions and research protocol to the Dutch situation and difficulties in recruting companies will be discussed during the presentation.Ê

JR Anema, IA Steenstra, PM Bongers, HCW de Vet, W van Mechelen; TNO Work & Employment/Institute for Research in Extramural Medicine, Vrije Universiteit

Corresponding address:

JR Anema MD
Occupational Physician, Researcher
Van der Boechorststraat
7 1081 BT Amsterdam
The Netherlands

Tel.: +31-2-35549599
Fax: +31-2-35549304
Email: h.anema@arbeid.tno.nl

Return top of page

Early Return to Work of Injured Workers: Nature and Role of Organizational Factors.

Raymond Baril

Poster #28, Fri. Dec 1, 2000. 3:40 - 3:55 pm, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

The present study was undertaken to remedy the scarcity of scientific knowledge about the implementation of early return to work (early-RTW) interventions by business firms. One of its aims was to identify the organizational factors which could account for the variation in early-RTW activities.

Two study populations were used:

1) Ê occupational injuries reported in four regional offices of the Commission de la santŽ et de la sŽcuritŽ du travail (Quebec WCB) between 1994 and 1997 and recorded in the WCB's administrative file created to allow follow-up of RTW (n=13 728) and

2) Ê companies from furniture, lodging, sawmill, and printing sectors in which return to work had been reported.

We identified through multifactorial analyses the associations between the presence (n=2 933) or absence (n =10 795) of early-RTW measures and the following variables: workerâs age, sex and occupation, the site and nature of injuries, duration of work absence, presence or absence of relapse, and size and sector of the company. In addition, a multiple case study design (n=16) was used to oppose two firms who used early-RTW intervention to two firms who did not in each industrial sectors. Semi-directed interviews were used to collect our data from workplace respondents and WCB professionals active in the participating companiesâ sectors.

The multifactorial analyses revealed the existence of interdependent and complex relationships between socio-demographic characteristics of the workers, type of injuries, corporate structure, and implementation of early-RTW interventions.Ê

In addition, the RTW process appear to be influenced by the structural characteristics of early-RTW interventions particularly human resources (OHS manager competency and attitude towards employees and early RTW interventions; the extent to which the teams involved in promoting early-RTW are bipartite) and the financial resources. Eight categories of organizational variables seem to be modifying factors.Ê Six of the categories are related to corporate characteristics: nature of the most prevalent injuries, structure (sector, size), corporate culture (organizational actorsâattitudes, values, attributions), rules (formal OHS procedures, content of collective agreements), work organization (multiple professional qualification, physical requirements, subcontracting, new technology, employment status), and intra-organizational relations between stakeholders (communication, collaboration, conflict). The two other categories are related to the larger organizational environment: the Quebec occupational health and safety network (WCB specialists, local OHS teams, sectoral associations, private firms), and regional characteristics (social proximity). These factors seem to either increase or decrease the association between early-RTW structure and process.

Baril R¹, Berthelette D², Ross C², Gourde D², Massicotte P¹, Pajot A²

⁽¹⁾Institut de recherche en santŽ et en sŽcuritŽ du travail, Montreal (Quebec), Canada.

⁽²⁾ Business School of Administration, UniversitŽ du QuŽbec ˆ MontrŽal, Montreal (Quebec), Canada.

Corresponding address:

R. Baril, PhD
Institut de recherche en santŽ et en sŽcuritŽ du travail
505 boul. de Maisonneuve Ouest
Montreal, PQ H3A 3C2
Canada

Tel.: +1 514-288-1551
Fax: +1 514-288-0998
Email: baril.raymond@irsst.qc.ca

Return top of page

The Effects of a New Pragmatic Approach to Low-back Pain Including Manual Therapy and Steroid Injections in Three RCTs Original Results Successfully Replicated in a New Study

Stefan Blomberg

Poster #14, Fri. Dec 1, 2000, 3:45 - 4:00 pm, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

The poster covers three different RCTs: Study 1 - Abstract

101 outpatients with acute or subacute low-back pain were randomly allocated to one of two treatment groups. One group was given standardized, conventional optimal activating treatment by primary health care teams. The other group received, according to a pragmatic approach, another treatment program including manipulation, specific mobilization, muscle stretching, autotraction and cortisone injections.

The two groups were similar in most of the pretrial variables, including age, sex, occupation, education, previous low-back pain problems, previous treatment, sick-leave, findings at the physical examination, quality-of-life score, presence of common symptoms, disability rating and pain score.

After one month in the study, the proportion of patients on sick-leave was six times larger in the conventionally treated group than in the group receiving the specific manual treatment. The difference diminished over time but was significant concerning the average number of days on sick-leave per patient during the eight months of follow-up. Significant differences in favour of manual therapy were also shown for pain score, disability rating, recovery score, drug consumption, quality of life and prevalence of common symptoms. There was also a difference in objective findings, assessed by blinded, independent and unbiased orthopedic surgeons, in favor of the group receiving manual treatment. The blinding procedure in the objective evaluation was proved to be successful. The outcome difference increased during the four monthsâ follow-up for many of the efficacy measures.

As expected, the trial treatment was, due to injections and muscle stretching, more painful than the conventional treatment but only a handful of patients experienced the manipulation and specific mobilization as painful. Rare, mild, transient side-effects and no complications were reported in the two groups.

The experimental patients also had a more positive opinion of the treatment than the patients in the conventionally treated group. Parallel therapy other than the intended treatment program in the two groups was closely supervised and there was no such treatment in the experimental group and negligible parallel treatment in the conventionally treated group.

In conclusion, all of the applied efficacy measures indicated that manual therapy is superior to standardized, optimized conventional activating management of patients suffering from low-back pain. Assuming that the results can be reproduced in future studies, it may also be concluded that manual therapy can reduce public costs for low-back problems, since the treatment volume, drug consumption and sick-leave were considerably less in the experimental group than in the conventionally treated group.

The study has been published in a number of papers including:

I. Blomberg S, SvŠrdsudd K, Mildenberger F. A controlled, multicentre trial of manual therapy in low-back pain; initial status, sick-leave and pain score during follow-up. Scandinavian Journal of Primary Health Care 1992;10:170÷78

II. Blomberg S, SvŠrdsudd K, Tibblin G. A randomized study of manual therapy with steroid injections in low-back pain; telephone interview follow-up of pain, disability, recovery and drug consumption. European Spine Journal 1994;3: 246÷54

III. Blomberg S, Hallin G, Grann K, Berg E, Sennerby U. Manual therapy with steroid injections ÷ a new approach to treatment of low back pain; a conÐtrolled multicenter trial with an evaluation by orthopedic surgeons. Spine 1994;19: 569÷77

IV. Blomberg S, Tibblin G. A controlled, multicentre trial of manual therapy with steroid injections in low-back pain; functional variables, side effects and compliÐcations during four months follow-up. Clinical Rehabilitation 1993;7:49÷62

V. Blomberg S, SvŠrdsudd K, Tibblin G: Manual therapy with steroid injections in low-back pain; improvement of quality of life in a controlled trial with four monthsâ follow-up. Scandinavian Journal of Primary Health Care 1993;11:83÷90

and: A thesis: Blomberg S, 1993. A pragmatic approach to low-back pain including manual therapy and steroid injections: A multicentre study in primary health care. Acta Universitatis Upsaliensis, Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 394, 148 pp.

Study 1Ê Long-Term Follow-up

Study design, objectives, summary of background data, and methods. See above.

Results: Sick-leave data for the entire three-year period are presented. In addition pain scores, disability rating scores, quality-of-life, prevalence of general symptoms, including psychosomatic symptoms concerning a two-year follow-up period. The outcome of the experimental approach was consistently better than that of the conventional treatment. Many variables, including sick-leave, showed statistically significant differences while most other variables showed a consistent and similar trend.

Conclusions: This study provides some of the first indications of long-term effect of a treatment approach including manual therapy with steroid injections in low back pain.

Study II - The Partial Crossover Study (to be published):

Study Design: A pragmatic, prospective, randomized single-blinded controlled trial with a 2-yearsâ follow-up.

Objectives: To test the effect of the evaluated experimental treatment approach on chronic LBP patients, refractory to the control treatment.

Summary of Background Data: Only scant evidence supports the continued use of manual therapy on chronic LBP patients. The pragmatic treatment concept including steroid injections has to be tested in chronic patients.

Methods: Ten of the twenty-one patients, refractory to the conventional treatment, were crossed over to the experimental treatment and followed for the remainder of the three year period.

Results: A shift to the experimental treatment produced major improvements in all evaluated variables compared to continued conventional follow-up.

Conclusions: This study supports the hypothesis that experimental pragmatic approach is beneficial in chronic LBP patients, refractory to the control treatment of the original study. Long-term effects ÷ two years.

Study III - The reproducibility study, (in the publication process):

A Randomized Clinical Trial Comparing Four Different Treatment Regimens: Manual Therapy Including Steroid Injections, Manual Therapy, Muscle Stretching and Orthopedic Care

Study Design: A pragmatic, prospective, randomized, controlled trial with a 2-year follow-up of return to work rates, sick-leave volumes, x-rays and neuroradiological investigations and incidence of back surgery. A 10-week follow-up of pain, functional variables, pain drawings, quality-of-life and psychosomatic symptoms.

Objectives: To replicate the results of a previous study, to investigate whether the evaluated manual treatment program is communicable to other therapists, and to evaluate the efficiency of four different treatment regimens. The present study population is compared to the earlier study population

Summary of Background Data: Manual therapy is a subject of controversy and further studies are needed to evaluate its efficiency. A previously published study in which beneficial effects in favor of a new pragmatic manual treatment approach including steroid injections had to be replicated before more firm conclusions could be drawn.

Methods: One hundred and sixty outpatients with acute or subacute LBP were randomly allocated to one of two experimental groups (manual therapy with or without steroid injections) or two control groups (standardized orthopedic care with or without muscle stretching). Pain, disability and quality-of-life were recorded using 100 mm visual analog scales (VAS). Information on the prevalence of 27 mainly psychosomatic symptoms was obtained by questions answered by Îyesâ or Înoâ. Sick-leave information was obtained from the social insurance offices and other sources. The incidence of low-back surgery and the rates of radiological examinations were recorded. Planned subgroup analyses in patients with radiating pain were performed.

Results: With regards to regression of over time, the manual therapy, in combination with steroid injections, was consistently superior to the three other treatments. Manual therapy without steroid injections scored better than the control treatments, while the control treatment including muscle stretching scored worse than the other control group at five weeks. No differences were found between the control groups after ten weeks.

Conclusions: The complete pragmatic manual therapy concept combined with steroid injections is superior to the other three evaluated treatments, and are in agreement with the results from a similar and previously published study. The method was even more effective in patients with pain radiating to the leg(s) than in patients with no radiation. The favorable results of the experimental treatments were even more evident if the pain radiated below the knee(s). The method is communicable to other physicians and physical therapists, and a modest amount of education in the evaluated treatment model is enough to achieve beneficial effects. Many of the differences in favor of the manual treatment increased during the follow-up period, implying long-term treatment effects.

Steroid injections are particularly important in helping patients whose everyday life and/or work place great demands on lower back function. The manual treatment without steroid injections is more successful than both control treatments. Therefore, the differences in favor of the complete pragmatic treatment concept in comparison to the control treatments cannot solely be due to steroid effects but that it was also dependent on other items of the therapeutic arsenal. The addition of muscle stretching to the basic control treatment, as a manual single therapy, seems to postpone recovery during the first weeks.

S. Blomberg, MD, PhD, J. Bogefeldt, MD, M. Grunnesjš, DN, K. SvŠrdsudd, MD, PhD. Uppsala University, Department of Public Health and Caring Sciences Family Medicine Section and Stockholm Clinic of Manual Medicine, Stockholm, Sweden

Correspondence address:

Stefan Blomberg, MD, PhD
Department of Public Health and Caring Sciences
Family Medicine Section
Uppsala Science Park
SE-751 85 Uppsala
Sweden

Tel.: +46 8 54480600

Fax: +46 8 54480610
E-mail: stefan.blomberg@allmmed.uu.se

Return top of page

The Reliability and validity of the Motility and Functional Capacity Evaluation.

Jodi Warman Chardon

Poster #17, Sat. Dec. 2, 2000, 9:35 - 9:50 am, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

Problem:Ê Occupational injury expenses exceed $12 billion annually in Ontario. Functional capacity evaluations (FCE) examine physical abilities of injured workers, and are utilized to make recommendations regarding safe work return. There is little scientific evidence of the reliability and validity of FCEs.

Research Design: A pre-experimental design was used to examine the reliability and validity of the English version of the Motility and Functional Capacity Evaluation (MFCE).

Methods:Ê Participants included 15 injured workers (18 to 64 years of age), referred to a private physiotherapy clinic for a FCE. Exclusion criteria included secondary medical conditions, inadequate English comprehension, and pending litigation. The MFCE was used to assess 19 functional activities, in three Categories (Manual Materials Handling, Endurance, and Balance) to obtain Category scores, as well as a Security score. A Total Functional score (total of Category and Security scores) and a Global Rating score (evaluator's classification of the subjectâs predicted work level) were determined from the MFCE. Work duties were determined one month post MFCE using a Current Work Status Questionnaire, which was completed by the subject and the employer.

Analysis: Three evaluators scored ten subjects to determine inter-rater reliability of Category, Total Functional and Global Rating scores. Construct validity was assessed by determining the level of agreement between the measures. Predictive validity was determined by comparing MFCE Total Functional and Global Rating scores to actual work status at one month post evaluation, as reported by subject and employer.Ê

Results:Ê Inter-rater reliability for Category and Total Functional Scores was excellent (intra class correlation coefficients=.99). Reliability of the Global Rating score was good (Spearmanâs rho, .83 to .93; p<.01). Convergent validity between the Total Functional score and the Global Rating score was also good (r=.84; p=.00). The correlations between the Global Rating and the Category/Security scores ranged from weak to moderate (r=.38 to .69; p=.00 to .16)). The MFCE measures predicted work status as reported by the subject one month post evaluation (r=.81; p<.00,Ê r=.66; p<.01, Global Rating and Total Functional scores, respectively). The level of agreement between MFCE measures and work status, as determined by the employer was lower (r=.67; p<.01, r=.50; p=.06, Global Rating and Total Functional scores, respectively).Ê

Conclusion:Ê The MFCE had excellent inter-rater reliability, and demonstrated construct validity. The work status one month post evaluation could be predicted using the MFCE. Predictive validity was higher using the Global Rating score and when the work status information was provided by the subject, rather than the employer.Ê A limitation of this study is the small number of subjects in the validity portion, and currently, the study is continuing in order to confirm the present results.

Jodi Warman Chardon, M.Sc. Candidate, B.Sc.PT (presenting author) School of Rehabilitation Therapy, Queenâs University, Kingston, Ontario

Elsie Culham, Ph.D. Affiliation: Queen's University,
Kingston Ontario, K7L 3N6

Corresponding address:

Jodi P. Warman Chardon, MSc., BSc.PT
Clinic Co-ordinator
Sante Montfort Health
5 - 1967 Portobello Blvd.
Ottawa, Ontario K4A 4E7
Canada

Tel: +1 613 837-0188
Fax: +1 613 725-1855
Email: jchardon@cyberus.ca

Return top of page

Determinants of Occupational Disability Following a Low Back Injury: A Critical Review of the Literature

Joan Crook

Poster #11, Thurs. Nov. 30, 2000, 2:45 - 3:00 pm, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

Problem: Musculoskeletal low back injuries pose a formidable health care problem for the injured worker, industry and compensation.Ê Several studies have confirmed that it is a small number of claimants who acrue most of the back injury costs.Ê It would be useful, therefore, to be able to predict, shortly after injury, those workers who are at high risk for continued disability.

The purpose of the investigation was to determine prognostic indicators of work disability in occupational back pain as reported in the literature, by systematically searching the research literature, assessing the methodological quality of the research, and finally synthesizing the findings into a concise summary.

Methods: A rigorous process for locating original research for review was done using three electronic data bases (MEDLINE, PsychINFO and EMBASE), review articles and bibliographic sources.Ê Two research assistants screened each abstract using pre-established criteria, that is, cases were: pain/injury of the back, based on original research, published in English, inception cohort under six months of episode with at least one follow-up assessment.ÊÊ Each eligible article was independently reviewed by two trained ãblinded reviewersä using 19 appraisal criteria for methodological quality of prognostic studies.

Results: The number of abstracts screened was 2169; 66 studies met the screening criteria. Methodological criteria were applied.Ê Disagreements were resolved by consensus and arbitrated by a third ãblindedä reviewer. Twenty-one studies met the methodological standards.

The use of an inception cohort and of multivariate statistical analysis to control for confounding variables were the key criteria for assessing methodological quality.Ê

The direction of the relationship between the prognostic variable and each outcome was assessed.Ê Time since onset, functional disability, psychological distress, pain reports, previous episodes and work environment were important indicators.Ê Older age was a prognostic factor while the results for gender were mixed.

Conclusions: It is difficult to generalize the prognostic factors of back pain from this literature.Ê There is a lack of a conceptual framework for the study of back pain prognosis.Ê Most studies have compartmentalized their assessment of factors, e.g. administrative database, medical factors primarily, or psychological factors primarily.Ê More recent studies have begun to assess the impact of workplace, economic and social milieu factors.Ê What is needed is a comprehensive multivariate biopsychosocial job-related model of work disability.

Acknowledgments: WCB-BC and WCB Alberta

J. Crook, I. Schultz, R. Milner and B. Stringer,
Workersâ Compensation Board of British Columbia
Vancouver, BC
Canada.

Corresponding address:

J. Crook, RN, PhD, Professor
McMaster University
1200 Main Street West, Room 3N28
Hamilton, ON L8N 3Z5 Canada

Tel.: +1 905-525-9140 Ext. 22298
Fax: +1 905-521-8834
Email: crook@fhs.csu.mcmaster.ca

Return top of page

Back Related Functional Limitations Among Homemakers

Clermont Dionne

Poster #8, Fri. Dec. 1, 2000, 9:50 - 10:05 am, Sat. Dec. 2, 2000, 2:25 - 2:35 pm

Background: Back pain is clearly not limited to the working population, yet there is little specific information on the consequences and evolution of this health problem in the non-working population.

Objectives: This study aimed at describing functional limitations among homemakers affected by back pain and at comparing them to those of full-time working women.

Methods: Subjects were members of a large Health Maintenance Organization of Washington state who consulted a physician for back pain. Telephone interviews were conductedÊ four to six weeks after the medical consultation and one and two years later. The analyses were conducted on 392 women; 104 of them were homemakers.

Results: The mean modified 16-item Roland-Morris scores were similar in both groups of women and decreased over two years. There were no statistically significant differences between homemakers and full-time working women on back-related functional limitations measured at each of the three interviews, even after controlling for confounding variables.

Conclusion: The nature and evolution over two years of back-related functional limitations are not different among homemakers and full-time working women.ÊÊ

C.E. Dionne and M. ChŽnard

Corresponding address:

Clermont Dionne, PhD
Professor, Epidemiology Research Group
Research Centre of the Laval University Affiliated Hospital
Hopital du St-Sacrement
1050 Chemin Ste-Foy
Quebec City, PQ G1S 4L8 Canada
Tel: +1 418 682-7511 Ext. 4675
Fax: +1 418 682-7949
E-mail: clermont.dionne@erg.ulaval.ca

Return top of page

Job Strain and Back-Related Functional Limitations

Clermont Dionne

Poster #9, Fri. Dec. 1, 2000, 9:55 - 10:10 am,
Sat. Dec. 2, 2000, 2:35 - 2:45 pm (Presenter at poster 8)

Background: Job strain, a combination of high psychological demands and low decision latitude at work, is known to be associated with the frequency of several health problems.

Objectives: The current study aimed at assessing the relationship between job strain and the functional consequences of back pain and at examining the effect of social support at work on this relationship.

Methods: Subjects were workers aged 18-65 years who consulted for back pain in emergency rooms of three major hospitals and missed at least one day of work because of back problems. They were interviewed over the phone two weeks after their consultation. Job strain was assessed with the Job Content Questionnaire while functional limitations were measured with the 24-item Roland-Morris scale. Analyses were conducted on 542 subjects.

Results: The majority of subjects were males with recurrent back pain. Job strain was statistically associated with back-related functional limitations. Social support at work did not modify this relationship. The statistical association persisted in multivariate analyses, after adjustment for several confounding variables.

Conclusion: Job strain may be one determinant of back-related functional limitations among workers.

C.E. Dionne, R. Bourbonnais, P. FrŽmont, M. Rossignol, S.R. Stock and M. Mondor

Epidemiology Research Group, Laval University, Quebec City (QUE) Canada

Corresponding address:

Return top of page

The Inter-rater Reliability of a Functional Capacity Evaluation: the Physical Work Performance Evaluation

Marie-JoseÊ Durand

Poster #18, Sat. Dec. 2, 2000, 9:40 - 9:55 am, Sat. Dec 2, 2000, 2:25 - 2:45 pm

In occupational rehabilitation, a Functional Capacity Evaluations (FCE) is a systematic assessment of an individualâs ability to perform a series of tasks safely.Ê Despite their recent popularity, several literature reviews have noted the lack of reliability or validity studies concerning the majority of FCEs.Ê The Physical Work Performance Evaluation (PWPE) (Lechner et al., 1994) is the only FCE that has been submitted to a reliability study for all its sections (dynamic strength, position tolerance, mobility).Ê However, the characteristics of the subjects included in the reliability study were heterogeneous and included few back pain subjects.Ê The present study evaluated the inter-rater reliability of the PWPE in a population of workers disabled from back pain.Ê Five trained raters (three OTs and two PTs) completed the evaluations.Ê For each subjectâs assessment, one rater administered the PWPE while another was used as a silent rater.Ê Each rater was matched twice with the four other raters.Ê Forty workers were evaluated (31 men, 9 women, mean age: 40.9, mean Oswestry score: 49.1)Ê Kappa coefficients were calculated to determine the inter-rater reliability.Ê Eighteen of the 21 tasks demonstrated a substantial (k=0.61 ö 0.80) to almost perfect agreement (k>0.80), two tasks of the mobility section a moderate (k=0.41 ö 0.60) agreement and one task of the mobility section a fair (k=0.21 ö 0.40) agreement.Ê Overall, the PWPE appears to be a Functional Capacity Evaluation with acceptable inter-rater reliability.

M.J. Durand, P. Loisel, S. Poitras, R. Mercier, S. Stock & J. Lemaire, PREVICAP Occupational Rehabilitation Centre, Charles-Lemoyne Hospital and Sherbrooke University, Longueuil, PQ, Canada.

Corresponding address:

M-J. Durand, PhD
PREVICAP Occpational Rehabilitation Centre
Charles Lemoyne Hospital and
Sherbrooke University
1111 St. Charles W, Office 101
Longueuil, PQ J4K 5G4 Canada
Tel: +1 450 674 5908 Ext. 234
E-mail: +1 450 674 5237
E-mail: marie-jose.durand@courrier.usherb.ca

Return top of page

Work is a Risk Factor for Musculoskeletal Pain in Adolescents

D. Ehrmann Feldman

Poster #12, Thurs. Nov. 30, 2000, 2:55 - 3:10 pm, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

Work has been associated with adult musculoskeletal pain but has not received much attention among adolescents even though more teens are working. The prevalence of musculoskeletal pain in adolescents is reported to be high (>30%). The objectives of this study were to determine the incidence of musculoskeletal pain in a cohort of adolescents and to investigate whether work is a risk factor for its development.Ê A cohort of 502 seventh to ninth graders in Montreal, Canada was followed three times, six months apart over a 12 month period. Students responded to a questionnaire addressing musculoskeletal health and lifestyle factors (including work). Musculoskeletal pain occurring at a frequency of at least once a week in the past six months was defined as the outcome. Work was measured by frequency level and was also classified by type of job. Multivariate methods (generalized estimating equations analysis) were used to model the repeated measures outcome as a function of work and other covariates.Ê The cumulative annual incidence of musculoskeletal pain was 38%. Adolescents who worked developed pain more than those who did not work (adjusted odds ratio: 1.23; 95% confidence interval: 1.04-1.43). Those who worked in white-collar jobs were at a higher risk of developing pain (adjusted OR: 2.92, 95% CI: 0.97-8.83) than those in blue-collar jobs (adjusted OR: 1.78, 95% CI: 1.00-3.18) or childcare (adjusted OR: 1.51, 95% CI: 0.78-2.91). Work was a risk factor for development of low back pain (adjusted OR: 2.19 95% CI: 1.12-4.28), neck and upper limb pain (adjusted OR: 1.89, 95% CI: 1.11-3.21), and for lower limb pain (adjusted OR: 2.08 95% CI: 1.45-2.98).Ê We conclude that work is a risk factor for musculoskeletal pain development in adolescents. Prevention strategies against injuries in the workplace should include adolescents who work.

D. Ehrmann Feldman, I. Shrier, M. Rossignol, L. Abenhaim, GRIS UniversitŽ de MontrŽal and SMBD Jewish General Hospital, Centre for Clinical Epidemiology, MontrŽal, Canada.

Corresponding address:

D. Ehrmann Feldman
GRIS UniversitŽ de MontrŽal and
SMBD Jewish General Hospital
Centre for Clinical Epidemiology
MontrŽal, PQ
Canada.

Return top of page

Reliability of Safe Maximum Lifting Determinations of a Functional Capacity Evaluation

Douglas P. Gross

Poster #19, Sat. Dec. 2, 2000, 9:45 - 10:00 am, Sat. Dec 2, 2000, 2:25 - 2:45 pm

Problem: Functional Capacity Evaluations (FCE) are measurement tools designed to assist in determining an individualâs safe, tolerable levels of function and readiness to return to work. However, if FCE is to be considered a valid tool, reliability and validity must be demonstrated. Inter-rater and test-retest reliability have been deemed the most important forms of reliability in FCE, and were the focus of this study of the reliability of determinations of maximal safe work levels for manual handling tasks using kinesiophysical FCE in a Low Back Pain (LBP) population receiving workerâs compensation.

Methodology: A repeated measure design was used to assess inter-rater and test-retest reliability. Raters assessed simultaneously in a primary, secondary format over two test sessions.Ê The primary therapist guided the subject through testing and progressed weight levels based on cues from two secondary therapists who observed and rated performance.Ê Secondary therapists were blinded to each otherâs cues and determinations.Ê Five occupational therapists from the Workersâ Compensation Boardâs (WCB) rehabilitation centre in Edmonton, Alberta, Canada, were enrolled in the study.Ê Each had completed training (by representatives of Isernhagen Work Systems) and was experienced in conducting kinesiophysical FCEs.Ê Rehabilitation Centre clients completing an occupational rehabilitation program following low back injuries were recruited and then rated on two separate test sessions, conducted two to four days apart.Ê The same therapists were used on the second test session as the first, and primary and secondary therapists remained the same, as did the location, and time of day of assessment.

Analysis: Intraclass Correlation Coefficients (ICC) and 95% confidence intervals were calculated using the secondary therapistsâ determinations of safe maximal lifting levels in pounds.

Results: Twenty-eight subjects initially volunteered, but only 25 attended both test sessions and three of these (12%) elected not to participate in test session two due to pain. ICC values for inter-rater reliability ranged from 0.95 for left side carrying to 0.98 for floor to waist lifting. ICC values for test-retest reliability for subjects participating in both days ranged from 0.81 for waist to overhead lifting to a high of 0.95 for right side carrying.Ê When all subjects were included in the analysis (including scores of zero for those feeling incapable of performing on the second day) test-retest ICC values ranged from 0.58 to 0.79.

Conclusions: Inter-rater reliability of kinesiophysical lifting and carrying determinations performed on a sample of low back injured WCB claimants was excellent.Ê Test-retest reliability was slightly lower but still very acceptable when subjects willing to participate in both days of testing were included in analysis. It fell to an unacceptable level, however, when values of zero were included for subjects declining to participate in a second test session because of increased pain.Ê The 12% of subjects who deemed themselves incapable of repeat testing is probably inflated in this study due to the ease with which subjects could decline. Yet their presence and the marked variation in their FCE results, suggest that assessment of maximal manual handling ability should take place over a two-day period.

DP Gross, PT, MC BattiŽ, PhD

Corresponding address:

Douglas P Gross, PT
Faculty of Rehabilitation Medicine
3 -48 Corbett Hall, University of Alberta
Edmonton, Alberta T6G 2G4
Canada

E-mail:dgross@ualberta.ca

Return top of page

The Web of the Back

M.T. Gil del Real

Poster #10, Fri. Dec 1, 2000, 10:00 - 10:15 am, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

Background: Common back pain constitutes a major health problem in the majority of industrialized countries due to its high prevalence and its health and labor costs. The patient suffering from back pain, whether acute or chronic, often wants to know more about the ailment and the nature of the pathology that causes the pain. On the other hand, the health professional who manages the patient is often unable to keep up with the latest literature on this subject, or to discriminate between quality studies and those that do not meet criteria for scientific quality. It is to meet this need for information - on the part of both patient and clinician - that The Web of the Back has been created.

Objectives: To provide general information to the patient on the causes, physiology, prevention, and treatment of common (or unspecific) back pain. To provide specific information to the health professional on the latest published research on common back pain, selecting only those studies that meet scientific standards for quality.

Methodology: The Web of the Back consists of two bilingual sites (Spanish and English): a General Information Site and a Professional Site. The first is designed for the lay person, and provides information, stated in a non-technical manner, on the different causes and physiology of common back pain, as well as prevention mechanisms, discouraging bed rest and promoting early return to activity. The different diagnostic tests and treatments are also explained. The second site has been designed for the health professional. A team of scientific reviewers evaluates all research on the back published in scientific journals since 1990, retrieved through a search of the MEDLINE, EMBASE, HEALTH STAR, PSYCHOINFO, PASCAL, LILACS, AND INDICE MEDICO ESPA„OL databases. Articles are evaluated according to a set of scientific criteria, and either accepted for inclusion in the Web or rejected. An article that is accepted is summarized and translated before inclusion in the site.Ê Twenty percent of all articles are independently evaluated by two reviewers as a quality control mechanism.

Results: The General Information Site began in November 1999 and the Professional Site has its scientific structure in place since that date. The first summarized articles were available in September 2000. Both sites of the Web are updated on a regular basis to include new information andÊ research.

M.T. Gil del Real, F.M. Kovacs, V. Abraira, P. L‡zaro, F. Pozo, J. L—pez
Scientific Department, Kovacs Foundation
Madrid and Palma de Mallorca, Spain

Corresponding address:

M.T. Gil del Real
Kovacs Foundation
Plaza Valparaiso, 8
28016 Madrid
Spain

Tel.: +34-91-344-0244
Fax: +34-91-344-1950
Email: mtgildelreal@kovacs.org

Return top of page

Perspectives of Primary Care Physicians on Return to Work After a Soft Tissue Injury

J. Guzm‡n

Poster #29, Fri. Dec 1, 2000, 3:50 - 4:05 pm, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

Objective: We surveyed primary care physicians in the province of Manitoba, Canada to assess their knowledge and attitudes about return to work (RTW) after a soft tissue injury, and to explore their views on facilitators and barriers for their effective participation in RTW.

Methods: A 60-item survey developed with input from occupational health stakeholders and pre-tested with family physicians was faxed in November 1999 to all practicing primary care physicians who cared for at least 10 work injuries in 1998.Ê Reminders were faxed 3, 6, and 16 weeks later.Ê Analysis was done by frequency distributions, chi square tests to compare proportions, t-tests to compare subgroupsâ means and multiple regression to assess the influence of physician and practice characteristics on their response profile.

Results:Ê Fifty one percent of eligible physicians provided usable answers (n=232).Ê Respondents did not differ from non-respondents in demographics (except that 60.8% of respondents practiced in the metropolitan area vs 70.9% of non-respondents, p=0.03).Ê Early respondents did not differ from late respondents in their response profile.Ê Most physicians were aware of appropriate management strategies and the main expectations of their role in the RTW process.Ê Nevertheless, a substantial minority failed to identify:Ê 1. The need to recommend activity instead of a brief period of bed rest for acute low back pain,Ê 2. The lack of evidence for early intensive medical and physiotherapy interventions,Ê 3. The importance of considering the well-being of co-workers before recommending RTW,Ê 4. The importance of making RTW plans at the first visit, andÊ 5. The confidentiality of medical information.Ê The highest ranked facilitator was the physicianâs ability to explain the nature and prognosis of the injury to the worker (69.0%). The highest ranked barrier was workerâs misunderstanding and fears about the injury (70.7%).Ê Respondents felt the following changes would help them most in treating injured workers:Ê increase/facilitate workplace accommodation (48%), educate the worker/public about nature and prognosis of soft tissue injuries (24.2%), and increase the availability of physiotherapy and occupational therapy (20.6%).Ê Physicians year of graduation, country of graduation and kind of practice (group/solo) were weakly associated with their response profile.

Conclusion: This survey of Manitoba primary care physicians: 1. Identifies five areas in need of training/education, 2. Shows most physicians believe main facilitators and barriers relate to patient-physician communication, 3. Shows physicians would like to see more workplace accommodation, education for workers/public, and availability of physiotherapy and occupational therapy.ÊÊÊÊÊÊ

J. Guzm‡n, J. E. Cooper, J. Khokhar, A. Yassi.Ê University of Manitoba, Canada R3E 0W3.Ê Study funded through a research grant from the Workers Compensation Board of Manitoba

Corresponding address:

J. Guzm‡n, MD
Research Associate
Occupational & Environmental Health Unit
S112-750 Bannatyne Avenue
Winnipeg, MBÊ R3E 0W3
Canada

Tel.: +1 204-789-3457
Fax: +1 204-789-3905
Email: guzmanj0@cc.umanitoba.ca

Return top of page

The Prevalence of Standardized Outcome Measures in the Physiotherapy Management of Low Back Pain.

Carmen S. Kirkness

Poster #2, Fri. Dec 1, 2000, 10:25 - 10:40 am, Sat. Dec. 2, 2000 2:25 - 2:45 pm

Rationale:Ê Because the treatment of individuals with low back pain is especially complex and multi-faceted, it would be expected that the use of assessment tools including those focused on impairment, disability and handicap would be commonplace in clinical practice.

Objectives:Ê To determine the prevalence of standardized outcome measure use and the frequency and timing of use in the physiotherapy management of low back pain and to determine the factors associated with use of standardized outcome measures by physiotherapists.

Research Design: This was a multi-centered, retrospective cross-sectional study that investigated 265 charts of clients who had received treatment for low back pain from 53 randomly selected physiotherapists working in two Canadian cities, Montreal, Quebec, and Ottawa, Ontario.Ê

Methods: Five randomly selected charts were reviewed for each physiotherapist to identify patterns of outcome measure use at three defined assessment points: initial, interim and discharge, in the treatment of individuals with low back pain.Ê

Analysis:Ê Analysis was based mainly on descriptive statistics.Ê Univariate statistics were used to identify variables associated with the use of both standardized and non-standardized measures.

Results:Ê Review of the 265 charts identified seven standardized outcome measures being used in actual clinical practice.Ê While 34% of the physical therapists were identified as users of standardized measures, only 17% indicated they were using standardized measures and 26% indicated present use of standardized measures.Ê Measures of impairment were much more common than measures of disability or impairment.Ê The three most commonly used measures included the Numeric Pain Rating Scale (53%), the Visual Analog Scale (7%), and the Modified Schober (2%).Ê Only one therapist was found to be using a measure of disability.Ê Measures were almost never used during the treatment process, nor at the discharge assessment times. Interestingly, the services used by clients who were evaluated on standardized tools at initial evaluation were different from the services used by those who were not formally evaluated.Ê

Conclusion:Ê Two-thirds of physical therapists have yet to adopt standardized outcome measures into clinical practice.Ê When tools were used, they focused exclusively on impairment.Ê It was rare for standardized tools to be used to monitor the clientâs progress over time.Ê On a more positive note, this research has led us to uncover a wide variety of tools that could be extremely useful in measuring disability associated with low back pain in daily practice.Ê

Authors C.S. Kirkness, N. Korner-Bitensky
School of Physical and Occupational Therapy, McGill University, Montreal, PQ, Canada.

Corresponding address:

Carmen S. Kirkness
E-mail: casche@hotmail.com

Return top of page

Evaluation of Measurement Strategies to Increase the Reliability of EMG Spectal Indices to Assess Back Muscle Fatigue

Christian Larivire

Poster #20, Sat. Dec. 2, 2000, 9:50 - 10:05 am, Sat. Dec 2, 2000, 2:25 - 2:45 pm

Objective: The purpose of the study was to evaluate measurement strategies to maximize the reliability of electromyographic (EMG) indices to assess back muscle fatigue. An experimental study using a repeated measures design was performed.Ê

Methods: Forty male volunteers [20 controls and 20 patients with chronic low back pain (CLBP)] were assessed on 3 sessions at least 2 days apart within 2 weeks. Surface EMG was collected for 8 back muscles (4 muscle pairs) while the subjects performed, in a static dynamometer, 2 static trunk extension tasks at 75% MVC separated by a 60 s rest period: (1) a 30 s fatigue task and (2) a 5 s recovery task. Two EMG spectral indices were computed to assess fatigue and recovery from fatigue. Each index was computed for each muscle, for the mean of each bilateral muscle pair and for the mean of all muscles.Ê Absolute differences between bilateral muscles was also computed for each index to obtainÊ additional EMG indices sensitive to left-right differences.Ê Intra-class correlation coefficient and standard error of measurement were calculated for each EMG index. Generalizability theory was applied to predict the reliability of EMG indices corresponding to theÊ mean of 2 fatigue trials performed in the same session.Ê

Results: Reliable EMG indices were obtained for both groups of subjects whenÊ (1) electrodes were positioned on medial muscles (contrary to lateral muscles) of the erector spinae and (2) measures were averaged across bilateral muscles and across electrodes positioned at different vertebral levels.Ê The average of two fatigue tests in the same session was also preferable.Ê Reliability of left-right EMG indices was poor.

Conclusions: The reliability of EMG indices based on individual muscles was too low to show muscle specific impairments or left-right differences.Ê However, reliable EMG indices was achievable by averaging measures of bilateral muscles (averaging across all muscles was not necessary).

C. Larivire, A.B. Arsenault, D. Gravel, D. Gagnon, P. Loisel
Research Center, Montreal Rehabilitation Institute, Montreal, PQ, Canada

Corresponding address:

Christian Larivire
Researcher
PREVICAP
1111, rue St-Charles Ouest, bureau 101
Longueuil, PQÊ J4K 5G4
Canada

Tel.: +1 450-674-5908 Ext. 239
Fax:Ê +1 450-674-5237
Email: christian.lariviere@courrier.usherb.ca

Return top of page

Utility of the Spinal Function Sort to Assess Patients With Chronic Back Pain

Dianne MacAusland-Berg

Poster #3, Fri. Dec 1, 2000, 10:30 - 10:45 am,
Sat. Dec. 2, 2000, 2:25 - 2:45 pm (Presenter at poster 15)

Problem and Objectives: The purpose of this study was to assess the Spinal Function Sort (SFS, A 50 item questionnaire designed with the purpose of quantifying a patientâs perception of disability regarding work activities which involve use of the spine) against objective functional testing of patients with chronic back pain.Ê This was done before and after a course of treatment based upon the functional restoration model with patients who had chronic back pain.Ê As the SFS purports to assess patientsâ perception of their capabilities cognitions which may limit actual levels of functioning, its ability to predict return to work status was also tested.

Research Design: A pre-post measurement design was utilized in this study.

Methods:Ê A sample of 53 consecutive discharges (30 women) from an interdisciplinary treatment program was utilized.Ê These participants had sustained back injuries either in worksite (n=25) or motor vehicle (n=28) accidents.Ê The mean age of participants was 38 and they had been off work for an average of 10.6 months.Ê At the time of admission, the SFS was administered by an occupational therapist in addition to objective functional testing.Ê This process was repeated during participants last week on program (an average of 3.4 months later).Ê Return to work status (RTW) was coded using the following serial categorization: returned to work; returned with modified duties; took alternate work; did not return to work.

Analysis: A paired-samples t-test was used to determine if significant change had occurred in the SFS scores following treatment.Ê Pearson product moment correlations were used to determine the relationships between SFS scores and objective functional ratings.

Results: A significant difference was found between the SFS score on admission and at discharge (t = 4.95(41) p < .0005).Ê The correlation matrix revealed a weak relationship between participantsâ SFS scores at admission and their objective performance on the primary physical task required for their job (r = .172).Ê At discharge, however, the relationship was much stronger (r = .344, p = .018). Similarily, change in objective functional testing was significantly correlated with change in SFS score (r = .366, p = .016).Ê Return to work status was best predicted by performance on the primary physical job task (r = .271, p = .038) and less so by the SFS score at discharge (r = .229, p = .077).Ê Participants were also categorized based upon the physical demands characteristics of their work (PDC).Ê In 57% of cases, participants successfully returned to jobs which required physical abilities 1 or more PDC categories higher than estimated by their SFS scores.Ê Subsequent discussions with the participantsâ occupational therapists revealed that they did not see the participants as symptom magnifiers.Ê Rather, the therapists thought that the SFS-based PDC categories underestimated patients' perceived and actual abilities.

Conclusion: The SFS was sensitive to changes in participantsâ functional status and was found to be a useful predictor of return to work status.Ê Using the SFS-based PDC categories, however, led to underestimating participantsâ functional abilities.

Dianne MacAusland-Berg, Jason E. Peebles, Marilee Allerdings and Jody Burnett, Functional Rehabilitation Program, Regina Health District
Regina, Saskatchewan, Canada

Corresponding address:

Dianne MacAusland-Berg
Functional Rehabilitation Program
Regina Health District
Regina, Saskatchewan
Canada

Return top of page

Testing the Validity of a Non-Exercise Method of Estimating Aerobic Capacity in Patients with Chronic Back Pain

Randy McPeek

Poster #21, Sat. Dec. 2, 2000, 9:55 - 10:10 am,
Sat. Dec 2, 2000, 2:25 - 2:45 pm (Presenter at poster 15)

Problem and Objectives: Aerobic exercise is an important component in the rehabilitation of individuals with chronic back pain. Determination of aerobic capacity is necessary to establish baseline fitness level, prescribe appropriate exercise intensity, determine functional limitations of cardiorespiratory fitness in terms of Metabolic Equivalent Units (MET), and to measure progress during treatment. The purpose of this study was to assess the utility and validity of a non-exercise method, the Physical Activity Rating(PAR; Jackson, Blair, et al., 1990), to determine maximum aerobic capacity (MVO₂) in individuals with chronic back pain.

Research Design:Ê The research design was correlational.Ê Two methods of estimating aerobic capacity (MVO₂) were compared at two points in time.

Methods:. The sample consisted of 34 males and 16 females with back injuries from work-site (N=25) orÊ motor vehicle (N=25) accidents. The mean age was 40.14 years (SD=10.71). The mean time from injury to admission was 11.5 months (SD=10.9). Participantsâ maximum aerobic capacity (MVO₂) was predicted with a submaximal aerobic exercise test. Each participant also completed the NASA/JSC Physical Activity Scale (which queries regular activity level) and age, gender, and body mass index (BMI) were recorded on admission.Ê A previously developed regression equation (PAR) combines this information to predict MVO₂.These procedures were repeated at discharge from treatment.

Analysis:. Following the generation of descriptive statistics, paired t-tests were conducted to determine the statistical significance between MVO₂ values from the submaximal aerobic exercise test and the non-exercise method. Pearson Product Moment Correlations were also conducted comparing MVO₂ estimates.

Results: At admission, the PAR and submaximal aerobic testing estimates were significantly different among females (t = 4.11(12), p < .01) but not for males (t = .87(24), p> .05).Ê The reverse was found at discharge. Strong correlations were found between the questionnaire and submaximal based estimates of MVO₂ at admission (rÊ = .66, p < .001) and at discharge (rÊ = .58, p < .001).Ê When the sample was broken down by sex, the correlations at admission were stronger, especially for women (rÊ = .84, p < .001)(for men, rÊ = .64, p < .01).Ê The correlations were still statistically significant when measured at discharge, but they were not as robust (rÊ = .67, p < .01 for women)(rÊ = .56, p < .01 for men).

Conclusion: The PAR was found to be a relatively accurate method of determining the aerobic capacity in patients with chronic back pain. Using regression weights established in a healthy sample of adults, the PAR-based MVO₂ estimates were highly correlated with those from submaximal testing.Ê The PAR slightly underestimated womenâs aerobic capacity at admission and slightly overestimated menâs aerobic capacity at discharge in comparison to submaximal testing (which has a 10-20% margin of error).Ê Clinically speaking, this might lead to a patient being misclassified by less than 1 MET, for example, rating aerobic capacity as average rather than above average.Ê A future study will utilize a larger sample and maximal aerobic capacity testing in order to develop a regression equation tailored to patients with chronic back pain.

Randy McPeek and Jason E. Peebles
Functional Rehabilitation Program, Regina Health District
Regina, Saskatchewan, Canada

Corresponding address:

Randy McPeek
Functional Rehabilitation Program
Regina Health District
2180 23^rd Avenue
Regina, Saskatchewan, S4S 0A5
Canada

Return top of page

PREDICTIVE TESTS FOR RETURN TO WORK IN PATIENTS WITH CHRONIC LOW BACK PAIN.

P. R. Oesch

Poster #13, Thurs. Nov. 30, 2000, 3:05- 3:20 pm, Sat. Dec. 2, 2000, 2:25 - 2:45 pm

Introduction: Return To Work (RTW) is the primary goal in the rehabilitation of patients with chronic low back pain (CLBP). To achieve this goal, an interdisciplinary evaluation and an intensive active treatment is recommended. The intensity of any treatment is strongly depending on the patientâs conscious motivation to perform optimally during the evaluation and rehabilitation process. Patients, whose efforts are not sincere during evaluation may overuse treatment, may have prolonged recovery or may have an increased cost of care. To increase cost effectiveness, predictive tests for RTW are needed to select patients for rehabilitation.

Purpose of this Study: To determine the predictive validity of an insincere effort as measured with a modified version of the Step Test and a Pseudo Strength Test as well as a Momentary Pain Intensity of 9 or 10 (NRS) for RTW in patients with CLBP. The second aim is to determine whether the prediction of RTW can be improved with a combination of these tests and the Waddell test. The third aim is to compare the predictive value of this 4 clinical tests with known psycho-social predictors such as; Work load, nationality, time off-work and employment status.

Methods: Prospective cohort study in 99 patients with chronic low back pain. Upon entry, physical work load, time off work, unemployment and nationality were recorded and the 4 investigated clinical tests with an anticipated prognostic value for RTW were performed. The average length of stay in the rehabilitation centre was 28.2 days (SD 7.4). The treatment duration was 16.2 hrs/wk with an emphasis on strength and endurance training, physiotherapy, back school and swimming with fins to increase endurance(11.2 hrs/wk). Psychological treatment, relaxation techniques and passive treatment completed the interdisciplinary rehabilitation. After 12 months the RTW rate was obtained from the referring physicians by postal survey.

Results: Response rate at 12 months was 91%. The RTW rate at 1 year was 20%. Unimproved working activity could be reliably predicted. The positive predictive v