European General Practice Research Network

Presentation 1: Theme paper, ongoing study with preliminary results

Friday, 9 May 2008, 09.30–10.00

Rheumatic diseases in general practice

Maria Panchovska, Nikolai Nikolov, Lyubima Despotova-Toleva, Radost Asenova, Gergana Foreva

Medical University Plovdiv – General Practice, 15a Vasil Aprilov, 4000 Plovdiv, Bulgaria. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: General practitioners (GPs) as gatekeepers meet many health problems, including the most common rheumatic diseases.

Research question: The aim of the study is to explore annual prevalence and incidence of rheumatic diseases, and to investigate GPs’ strategies to manage them.

Method: Design: general practice consultation database and structured questionnaire. Setting: eight general practices—13 666 patients on their lists. Participants: eight general practitioners. Main variables measured: prevalence and incidence; self-assessment of their qualification to diagnose and to treat rheumatic diseases. Analysis: alternative analysis. Statistics: SPSS version 15.0.

Results: The prevalence of rheumatic diseases is about 15–40%. Patients with arthrosis prevail, followed by patients with rheumatic arthritis, psoriatic arthritis, reactive arthritis, and collagenosis. A graduate scale from 0 to 6 was used for self-evaluation of GPs concerning their training in rheumatology. Only three of them evaluate their knowledge as very good, two as good, and three as average.

Conclusions: Rheumatic diseases are a common problem in general practice. GPs need additional and continual training. Specific education and training programmes focused on the above-mentioned topic and targeted at GPs need to be developed.

Presentation 2: Theme paper

Friday, 9 May 2008, 10.00–10.30

Polymyalgia rheumatica: Is PMR-AS usable in general practice?

Benjamin Lefebvre, Sébastien Cadier, Alain Sarraux, Jean Yves Le Reste

Université de Bretagne Occidentale – Med. Générale, Rue Camille Désmoulins, 29200 Brest, France. Tel: [phone]; E-mail: [email protected]

Background: The polymyalgia rheumatica activity score (PMR-AS) is used by rheumatologists for the follow-up of polymyalgia rheumatica (PR), but the follow-up of most patients is done by general practitioners (GPs).

Research question: Is the PMR-AS usable in general practice?

Method: Quantitative observational study with a postal survey form sent to a random sample of GPs in Brittany. The form was based on seven clinical cases validated by an expert group (French Society of Rheumatology) with questions concerning practice related to PR. For each case, the GP had to evaluate a visual analogue scale (VAS) of pain, diagnose (or not) a PR stroke, and prescribe (or not) steroids. Analysis was done with a receiver operating characteristic (ROC) curve association between PR stroke diagnosis, PMR-AS, and its components (CRP, VAS-GP, VAS-patient, morning stiffness, shoulder propulsion).

Results: Older GPs (age>40 years) follow more PR than younger GPs (p<0.05), and their amount of collaboration with rheumatologists is lower (p<0.01). Diagnosis of stroke is followed by steroid prescription in most cases (p<0.001). GPs correctly diagnosed no stroke (PMR-AS<7) in 99.5% of cases. Among diagnoses of a stroke (PMS-AS>7), 8.2% were false positive, but the rest were correct. PMR-AS = 7 was of good diagnostic value for PR stroke (99.4% discrimination, 93.3% uniqueness).

Conclusions: Use of PMS-AS in general practice for the follow-up of polymyalgia rheumatica seems to be useful. It could help to identify patients with specific needs and to aid better coordination between health professionals. A complementary study based on patients and not on clinical cases is necessary in order to validate these results.

Points for discussion: What type of complementary study should we design to validate our results?

Presentation 3: Theme paper

Friday, 9 May 2008, 11.00–11.30

Effect of glucosamine sulphate on hip osteoarthritis: A randomized trial

Rianne Rozendaal, Bart Koes, Gerjo van Osch, Elian Uitterlinden, Eric Garling, Sten Willemsen, Abida Ginai, Jan Verhaar, Harrie Weinans, Sita Bierma-Zeinstra

Department of General Practice, Erasmus University Medical Centre, Erasmus MC Rotterdam, the Netherlands. E-mail: [email protected]

Background: The effectiveness of glucosamine sulphate as a symptom and disease modifier for osteoarthritis is still under debate.

Objective: We conducted a long-term double-blind RCT (ISRCTN54513166) in primary care patients with hip osteoarthritis to assess effectiveness over a period of 24 months on pain, function, and joint space narrowing.

Methods: We randomly assigned 222 patients with hip osteoarthritis to either 1500 mg of oral glucosamine sulphate (GS) once daily or a placebo for 2 years. General practitioners recruited prevalent hip osteoarthritis patients. Patients were eligible when they met the clinical set of ACR criteria for hip osteoarthritis. Primary outcome measures were WOMAC pain and function subscales over 24 months and joint space narrowing after 24 months. Subgroup analyses were predefined for severity of radiographic osteoarthritis (radiological severity=1 vs = 2) and for type of osteoarthritis (localized vs generalized). For additional exploratory analyses, patients were divided into groups based on pain level, pain medication use, joint space width, and absence or presence of co-occurring knee osteoarthritis.

Results: Overall, WOMAC pain did not differ (mean difference −1.54, 95% confidence interval [CI] −5.43 to 2.36), nor did WOMAC function (−2.01, 95% CI −5.38 to 1.36). Joint space narrowing also did not differ after 24 months (−0.029, 95% CI −0.122 to 0.064). Subgroups based on radiographic severity, pain level, pain medication use, and joint space yielded similar results. Mean difference in WOMAC pain was 1.40 (95% CI −5.55 to 8.34) for the localized osteoarthritis group, and −3.45 (95% CI −8.19 to 1.28) for the generalized group. For patients with co-occurring knee osteoarthritis, the outcome for pain was −5.68 (95% CI −12.62 to 1.26) compared to −0.12 (95% CI −4.91 to 4.68) for patients without.

Conclusion: Overall, glucosamine sulphate was not better than placebo in reducing symptoms and progression of hip osteoarthritis.

Presentation 4: Theme paper, ongoing study with preliminary results

Friday, 9 May 2008, 11.30–12.00

Topic-related web-based resources: Information focused on musculoskeletal diseases

Luybima Despotova-Toleva

MU Plovdiv – General Practice, 15a Vasil Aprilov, 4000 Plovdiv, Bulgaria. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: This report presents a part of the author's research on E-medicine, which is supported by the Bulgarian Ministry of Education and Science. A part of this study was done during the author's stay at the University of Illinois at Chicago, USA.

Research question: To assess the availability and usefulness of different web-based materials focused on musculoskeletal diseases.

Method: Design: structured questionnaire and free search on the Internet. Setting: university hospital. Participants: assistant professors and professors from Medical University Plovdiv and other hospitals. Main variables measured: information based on official websites; universities, professional societies, hospitals, patient organizations, commercial, etc.; kind of information, usefulness, information update, etc. Analysis: alternative analysis. Statistics: SPSS version 15.0.

Results: We compare the information found on international and Bulgarian websites and analyse it. There is a lack of topic-related information focused on musculoskeletal diseases in the Bulgarian language and on Bulgarian websites, a lack of educational materials, and a lack of web-based materials devoted to patients with these diseases.

Conclusions: We have to develop and publish specific information on the Internet, especially for different target groups such as clinical specialists, general practitioners, and patients.

Presentation 5: Theme paper

Friday, 9 May 2008, 12.00–12.30

Persistent musculoskeletal pain at multiple sites: A sign of hyperglycaemia

Pekka Mäntyselkä, Juhani Miettola, Leo Niskanen, Esko Kumpusalo

University of Kuopio, School of Public Health and Clinical Nutrition, Unit of Family Medicine, PO Box 1627, 70211 Kuopio, Finland. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Type 2 diabetes and glucose regulation abnormalities are common in the population and are increasing rapidly. There are some reports about an excessive prevalence of chronic pain or pain syndromes among patients with diabetes.

Research question: To analyse the prevalence of elevated fasting plasma glucose (FPG) level and diabetes in subjects with chronic pain (duration at least 3 months) graded by pain frequency and number of painful sites.

Method: A population-based study including 469 adults aged 30–65 years. Elevated FPG was defined as a glucose concentration of more than 6.0 mmol/l. Diabetes diagnosis was based on self-reported diagnoses, reimbursed medication, or with a fasting plasma glucose level or an oral glucose tolerance test. The number of painful sites in the upper and lower extremities, shoulders and hips, and in the neck and back was summed. Chronic pain status was graded as 0 for no chronic pain, 1 for non-daily pain at multiple sites or daily pain at a maximum of three sites, and 2 for daily pain with at least four localizations (daily chronic widespread pain; DCWP). Logistic regression analysis was used to analyse the association of chronic pain status with elevated plasma glucose level and diabetes.

Results: Elevated FPG was found in 13% (n=33) of those having no chronic pain (n=252) and in 51% (n=21) of those having DCWP (n=41). Diabetes was found in 6% (n=16) of those subjects having no chronic pain and in 32% (n=13) of the subjects with DCWP. In the logistic regression analysis adjusted for age, gender, and body-mass index, DCWP was associated with elevated FPG (odds ratio [OR] 5.82, 95% confidence interval [CI] 2.69–12.57) and with diabetes (OR 5.65, 95% CI 2.36–13.52).

Conclusions: Persistent chronic pain at multiple sites is associated with elevated FPG and diabetes.

Points for discussion: 1) Are we able to generalize these results to other populations? 2) Comorbidity constitutes a challenge in the treatment and prevention of chronic musculoskeletal pain in general practice, especially among older patients.

Presentation 6: Freestanding paper

Friday, 9 May 2008, 14.00–14.30

Teaching a minimal intervention for stress-related mental disorders to general practitioners: effects on sick-leave duration in distressed patients

Henri E. J. H. Stoffers, on behalf of the steering and writing committee of AMUSE, Amsterdam Maastricht Utrecht Study on thromboEmbolism

Department of General Practice, School of Public Health and Primary Care (CAPHRI), Universiteit Maastricht, PO Box 616, 6200 MD Maastricht, the Netherlands. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Up to 90% of referred patients with suspected deep venous thrombosis (DVT) do not have this disease. It would be ideal to safely exclude DVT at initial presentation.

Objective: We conducted a management study in primary care to evaluate the safety and efficiency of excluding DVT using a clinical decision rule previously validated in a primary care population, including a point-of-care D-dimer assay.

Method: Prospective cohort study in primary care (300 GPs) of consecutive patients with clinically suspected DVT. Patient management was based on the result of the clinical decision rule including the D-dimer. Patients with a score of 3 were not referred for ultrasound but were invited to visit the GP's office 1 week later; patients with a score of 4 were referred for ultrasound and received care conforming to regional guidelines. The primary outcome was symptomatic, objectively confirmed venous thromboembolism (VTE: DVT, pulmonary embolism) during 3 months of follow-up.

Results: The mean age of the 1028 study patients was 58 years; 37% were male. In 26 patients (2.5%), the rule was not completed according to protocol. In 500 patients (49%), the score was 3, seven of whom developed VTE within 3 months (1.4%; 95% confidence interval [CI] 0.6–2.9%). In 502 patients (49%), the score was 4, and DVT was present in 125 (25%); in three patients, ultrasound was not performed. Of the 374 patients with a score of 4 in whom the ultrasound was normal, four developed VTE within 3 months (1.1%, 95% CI 0.3–2.7%).

Conclusions: A diagnostic management strategy for suspected DVT in primary care using a clinical decision rule including a point-of-care D-dimer assay reduces the need for referral by almost 50%, improves the yield of ultrasound in referred patients, and is associated with a low risk of VTE in patients who are not referred.

Points for discussion: 1) We tested a clinical decision rule including a point-of-care D-dimer assay. The GPs were instructed to score items from medical history and physical examination as well as perform the finger-prick test.

Presentation 7: Freestanding paper, ongoing study with preliminary results

Friday, 9 May 2008, 14.30–15.00

Effects of a series of intensified prevention consultations on the changes in the risk factors of hypertensive patients at high cardiovascular risk. ESCAPE study: Method and baseline data

Denis Pouchain, Dominique Huas, Patrick Chevallier, J. Chapman

Paris Ile de France Ouest University – General Practice, 6 bis, rue des 2 communes, 94300 Vincennes, France. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Observational studies have shown that a majority of patients with hypertension do not reach the treatment targets recommended in guidelines.

Research question: Can a two-step intervention approach, firstly at the level of general practitioners (GPs) and secondly from GPs to patients, improve patients’ healthcare outcomes?

Method: National pragmatic multicentre cluster randomized controlled trial, involving 278 GPs. Inclusion criteria: patients aged from 40 to 75 years in primary prevention, with hypertension treated for at least 6 months, plus at least two other cardiovascular risk factors. Intervention: a 1-day training course in addition to one specific structured consultation centred on cardiovascular disease prevention every 6 months for 2 years and feedback on the results.

Results: Of a total of 278 GPs, 255 (93.8%) included 1828 patients (mean age 61 years) who had hypertension for 10 years or more. Of these, 57.8% of patients had type 2 diabetes (T2D). All clinical and paraclinical characteristics were comparable between the two groups with the exception of blood pressure (BP); indeed, mean systolic BP and diastolic BP (7 and 3 mmHg, respectively) were superior in the intervention group. With respect to guidelines, 25.2% of patients displayed controlled hypertension, low-density lipoprotein cholesterol (LDL-C) was at target in 40.5%, 54.7% exhibited HbA_1c<7%, and 78.5% no longer smoked. Of all of the included patients, 7.2% reached all treatment targets. For patients with hypertension in the absence of T2D, 11% reached three targets. For hypertensive patients with T2D, 1.7% reached the five targets proposed in the guidelines.

Conclusions: The very poor level of patients reaching the targets warrants interrogation as to the relevance and applicability of the treatment targets proposed in the guidelines. Differences in BP between the two groups are probably due to the systematic use of an electronic device in the intervention group versus more usual measurements in the control group.

Points for discussion: Is it relevant to give an electronic device to GPs in the control group for the last consultation in 2009? Are targets in the guidelines really achievable in daily practice?

Presentation 8: Freestanding paper, ongoing study with preliminary results

Friday, 9 May 2008, 15.00–15.30

Should we systematically screen with ECG young athletes in primary care?

Alain Mercier, P. Trochu, S. Cerqueira

Université de Rouen, General practice, Faculté de médecine, 22 Bd Gambetta, 76000 Rouen, France. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background:

The sudden unexpected death of young athletes is rare, but has great impact for the community and the physician. Nevertheless, these events are probably more common than previously thought. While allowing practising sports, the general practitioner's (GP's) aim is to reduce this risk by identifying the potential risk of cardiovascular diseases. In France, this is currently based on an investigation into personal and family history, and a careful physical cardiovascular examination. Performing an ECG to detect abnormalities potentially linked with tachyarrhythmia during exercise, in primary care, to prevent such events, is still a matter of debate: could it be relevant to increase early detection of cardiac diseases that can predispose to lethal events?

Research question: Is ECG effective in detecting anomalies in primary care among a population of young athletes?

Method: We re-read retrospectively 735 ECGs performed by GPs between 2000 and 2006 in a primary care sports centre. The screened population was every outpatient, aged from 14 to 35 years, asking for a non-contraindication sport certificate. An adapted algorithm for primary care, prioritizing feasibility, was used to assess the ECGs. All abnormalities detected were sent for cardiologic assessment. Results were compared to patients’ history and physical examination results.

Results: This study is ongoing. Among the 735 patients (average age 20 years), a few problems potentially leading to dangerous tachyarrhythmia were detected: four “accessory pathways”, two “long QT space”, and 65 electric cardiac hypertrophies. Among those hypertrophies, one was associated with deep inverted T waves (arrhythmogenic right ventricular dysplasia?). Many minor non-significant abnormalities were also detected (27% of all patients), such as incomplete right bundle branch block.

Conclusions: Developing a relevant tool, with high specificity and sensitivity, in primary care still remains a challenge.

Points for discussion: 1) Are you aware of the European Society of Cardiology consensus statements advising the screening of young athletes with a standard 12-lead electrocardiogram? 2) What about the Italian experience, as it seems to be different in this country?

Presentation 9: Freestanding paper

Friday, 9 May 2008, 14.00–14.30

The ambiguous effect of social support in preventing burnout

Yvonne Winants, M. Twellaar, I. Houkes

Maastricht University, General Practice – Gender Studies FHM L, PO Box 616, 6200 MD Maastricht, the Netherlands. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Nowadays, societal changes and recent changes in the Dutch healthcare system are challenging the resilience and inner strength of Dutch general practitioners (GPs). Our longitudinal burnout study revealed that the level of burnout among GPs fluctuated from 19% in 2002 to 8.5% in 2004, and then increased to 12.8%, in 2006 especially among female GPs. To prevent burnout and loss of human capital, factors must be found that moderate the negative health effects of work-related stressors among GPs.

Research question: Can social support buffer the negative effects of work-related stressor/burnout outcome parameters such as emotional exhaustion and depersonalization among male and female GPs? How can gender differences be explained?

Method: The study population consisted of a random sample of 700 working Dutch GPs (350 male, 350 female). The study had a full panel design in three waves (2002, 2004, 2006) using self-reported questionnaires. Social support was measured by means of the VOS-D (Dutch Questionnaire on Organizational Stress), and burnout was assessed with the Maslach Burnout Inventory. Buffer effects were tested with the hierarchical moderated regression approach.

Results: Among female GPs, we found a positive buffer effect of both support of spouse and colleagues on burnout, meaning that the detrimental health effects of high workload and dissatisfaction among women are lessened by organizational or private social support. In male GPs, however, we only found a buffer effect due to private support of the spouse, and remarkably this effect appeared to be reversed. Subanalyses revealed that support of the spouse worsened emotional exhaustion and depersonalization of male GPs in the case of high workload, lack of work control, and dissatisfaction.

Conclusions: Gender differences were found in the moderating effect of social support. The gender congruence hypothesis, stating that seeking emotional support is not congruent with the masculine gender role of autonomy and instrumentality, could be a possible explanation.

Points for discussion: Various explanations for the empirical findings of gender-specific patterns can be discussed. Reflection on the implications of these findings on the development of preventive interventions for burnout among GPs is warranted.

Presentation 10: Freestanding paper

Friday, 9 May 2008, 14.30–15.00

The use of APGAR in the assessment of family functioning

Janko Kersnik, Katja Pesjak

Department for Family Medicine, Medical Faculty, Ljubljana Hrase 17, 4248 Lesce, Slovenia. Tel: [phone]; E-mail: [email protected]

Background: Assessment of family functioning is an important part of working with families in family practice. Smilkstein has developed an instrument for the assessment of family functioning, i.e., APGAR. With this questionnaire, family physicians collect basic information about families and family members they treat in a relatively uncomplicated and efficient way. Our study will prove that use of APGAR can effectively replace a variety of other questions about the patient and his/her family and social background.

Research question: Due to lack of time, a problem that physicians cope with every day, we decided to pose one important research question. Is the already-existing instrument for assessment of family functioning valid for use in everyday practice?

Method: We analysed 101 student reports on families, and extracted 30 variables that were common to all reports. APGAR questions were also included. Each year medical students are given an assignment to interview families that they visit as part of the family medicine curriculum. Beside APGAR, they use other guidelines to obtain enough data to adapt medical care to individual needs and contexts. All students ask questions about the number of persons in the family, the family form, understanding in the family, questions about diseases across generations, about marriages and divorces, causes of death, etc.

Results: We found a good correlation of answers on 23 interview variables (questions) with the APGAR scale. In particular, different questions about understanding in the family show how adequate Smilkstein's instrument is. APGAR reflects family functioning. We have proven that APGAR is a reliable instrument and that family physicians can use it.

Conclusions: APGAR can serve as a surrogate to the family interview or can provide important additional better understanding of patients and their families.

Points for discussion: 1) Is the already-existing instrument for assessment of family functioning valid for use in everyday practice? 2) Can physicians trust the obtained results?

Presentation 11: Freestanding paper, ongoing study with preliminary results

Friday, 9 May 2008, 15.00–15.30

Domestic violence: Trying to reach European consensus on registration

Kristof Hillemans, Leo Pas, K. Ampe, L. De Deken, P. Dhauwe, A. M. Offermans, M. Van Halewyn

Research Institute Domus Medica, Kerkplein 4, 2547 Lint, Belgium. Tel: [phone]; E-mail: [email protected]

Background: Domestic violence (DV) has a high prevalence, and 80% of all victims are women. It often remains undetected and untreated in general practice. A European network has been created to improve this. One of its goals is to develop a registration tool for DV in primary healthcare to describe actual practice, identify problems, and evaluate care pathways.

Research question: What items of information should a general practitioner (GP) best register in the medical record of (suspected) victims or patients at risk of DV in order to deliver and evaluate care adequately?

Method: After literature study and consensus meetings, 113 items were subjected through an online questionnaire (Formsite) to be scored on importance (1–4) by Belgian and European experts. Opportunity to comment was given. Scores were analysed using SPSS. In this way, items were categorized as: necessary (score – SD>3.5), to be recommended (3.5>score – SD>3), or useful in more extensive documentation (3>score – SD>2.5).

Results: Sixteen experts completed the questionnaire. Six items achieved a maximal score from all participants: relationship client-declared perpetrator, mechanism of sustained injuries, nature of violence, exact location of injuries, psychological impression of the patient by the physician, and seriousness/urgency/safety assessment. In total, 18 items were categorized as always important to register and 37 items as recommendable.

Conclusions: Based on this survey, and after renewed consultation with experts, a prototype for a structured registration form will be submitted in a qualitative research protocol to a representative sample of GPs in order to study feasibility and acceptability.

Points for discussion: What is your experience with this kind of study design? Do you have any suggestions to improve response rate and validity? Are you interested in using this prototype for a structured registration form in your own practice and sharing your opinions with us?

Presentation 12: Freestanding paper, ongoing study, no results yet

Friday, 9 May 2008, 16.00–16.30

Can the level of activity and quality of research in primary care be significantly improved by the development of a national managed research network?

Paul Wallace

Primary Care and Population Sciences, University College London, Rowland Hill Street, NW3 2PF London, United Kingdom. Tel: [phone]; E-mail: [email protected]

Background: The UK has recently undergone a major review of its national health service research strategy. As a result, the National Institute for Health Research (NIHR) has been established, with an annual budget of more than £100m p.a., in order to promote the development and delivery of high-quality clinical trials and other well-designed studies. The UK Clinical Research Network has been modelled on the positive experience with the UK National Cancer Research Network, which saw an increase in patient recruitment from 4% to more than 15% of eligible patients over a period of 3 years. The Primary Care Research Network (PCRN) is designed to provide the infrastructure necessary to promote effective involvement of general practices in recruitment and retention of patients to an approved portfolio of research studies.

Research question: Can the level of activity and quality of research in primary care be significantly improved by the development of a national managed research network?

Method: The PCRN was launched in March 2007, and consists of eight Local Research Networks. Each has a budget of £250k p.a. to provide for a full-time network manager, a set of 4–6 research nurses, sessional time for a general practitioner (GP) clinical research leader, and administrative support. Mechanisms are being developed to provide financial support through mainstream GP contracting mechanisms.

Results: The PCRN has been fully established, and, at the time of writing, the UK primary care research portfolio consists of more than 85 studies. At the end of 2007, recruitment of patients in primary care accounted for more than 25% of the entire UK clinical research activity.

Conclusions: Initial findings from the PCRN indicate that a managed network may be an effective mechanism to promote high-quality research in general practice. Significant questions remain about the sustainability of this approach in the UK, and about its generalizability in international settings.

Points for discussion: What is the experience of other countries with managed primary care research networks? Is there an interest in developing this initiative on an international basis? If so, what should be the next steps?

Presentation 13: Freestanding paper

Friday, 9 May 2008, 16.30–17.00

Composing an international research agenda on gut feelings in general practice using the nominal group technique

Yvonne van Leeuwen, Paul van Royen, M. van de Wiel, M. A. van Bokhoven, P. H. H. Houben, S. Hobma, T. van der Weijden, G. J. Dinant, Erik Stolper

School for Public Health and Primary Care (Caphri), Department of General Practice, Maastricht University, PO Box 616, 6200 MD Maastricht, the Netherlands. Tel: [phone]; E-mail: [email protected]

Background: Although gut feelings play a substantial role in the diagnostic reasoning of general practitioners (GPs), there is little evidence concerning their diagnostic and prognostic values. Consensus on the two types of gut feelings, a sense of alarm and a sense of reassurance, enabled us to operationalize the concept. Now, we need to know what aspects of gut feelings are most relevant to daily practice and medical education, and thus need further study.

Research question: How can we formulate research questions enabling us to validate the concept of gut feelings and estimate its usefulness for daily practice and medical education?

Method: The nominal group technique (NGT) is a qualitative research method of judgmental decision-making, which includes four phases: generating ideas, recording, evaluation, and group decision. We used NGT because gut feelings are conceptually complex and intricate. Dutch and Belgian university teachers and researchers (n=18) specializing in general practice attended one of three scheduled meetings.

Results: The three groups produced 20 research questions and appropriate designs mostly regarding diagnostic value, the validation of determinants, and opportunities for integrating gut feelings into medical education. Examples of proposed questions: What is the prevalence and the diagnostic relevance of gut feelings? Are there differences in significance between surgery hours and out-of-office hours? What does the GP's work experience contribute? What is the significance of the contextual information? How should we integrate gut feelings into medical education? Examples of proposed designs:

prospective recording of gut feelings and their determinants and expected diagnoses, with follow-up after some months; observational studies using case vignettes or stimulated recall interviews; experimental study with an educational intervention.

Conclusions: NGT helped us to compose an international research agenda on gut feelings in general practice, which we assume can be used in collaborative research in other countries.

Points for discussion: Are there other experiences with the development of an international research agenda? How can this agenda facilitate international research on gut feelings?

Presentation 14: Freestanding paper

Friday, 9 May 2008, 17.00–17.30

Medical students’ patient centeredness during hospital internships: Another argument for communication skills training?

Katrien Bombeke, Luc Debaene, Sandrina Schol, Benedicte De Winter, Paul Van Royen

Centre for General Practice, University of Antwerp, Universiteitsplein 1, B-2610 Wilrijk, Belgium. E-mail: [email protected]

Background: Research on patient centeredness (PtC) has exploded in the past two decades. Medical schools worldwide have included PtC as a core competence in their learning goals. In spite of these efforts, there is still a decline in PtC as students progress through medical school. Several factors have been suggested to contribute: student factors (e.g., gender, speciality preference) as well as educational factors (e.g., hospital internships).

Research question: 1) How does the PtC of medical students evolve during hospital internships? 2) Is the change in PtC during hospital internships different in a curriculum with communication skills training?

Methods: A comparative prospective cohort study in pre-/post-design: two cohorts of medical students were measured before and after hospital internships (6th year). Due to a curriculum change, the first cohort did not have communication skills training, while the second cohort did during 5 years of medical education. We used a combination of four validated measurement scales available in the literature: the Doctor-Patient Scale (attitude towards PtC; DeMonchy & Batenburg), the Communication Skills Attitude Scale (Rees), the Leeds Attitudes Towards Concordance Scale (Thistlethwaite), and the Jefferson Scale of Physician Empathy (Hojat). We also collected personal data such as gender and speciality preference. Results were analysed with SPSS 14.0.

Results: The pre-test/post-test comparison of participants of both cohorts together with a complete data set (n=85) showed a small but significant decline in the scores of the Doctor-Patient Scale: the mean score declined from 200.6 to 195.9/290, by 4.7 points (95% confidence interval [CI] 2.5–7.0). This decline was especially, and unexpectedly, seen in the second cohort (with communication skills training) and not in the cohort without communication skills training.

Conclusion: The PtC of medical students declines significantly during hospital internships. More results about the influence of communication skills training will be presented at the conference.

Points for discussion: 1) Any suggestions for further analysis? 2) What are the experiences of other researchers with the difficulty of measuring effects of intervention on students’ skills and attitudes during a medical curriculum? 3) How is education in PtC implemented in the medical curriculum in other universities?

Presentation 15: One slide/five minutes

Friday, 9 May 2008, 16.00–16.10

PREVACA: An evaluation of preventive hepatitis A vaccination in risk groups

Pascale Santana, Lise Abravanel, Anne Gervais, Jean-Pierre Aubert

Department of General Practice, University Paris 7, 16 rue Henri Huchard, 75018 Paris, France. Tel: [phone]; E-mail: [email protected]

Background: The hepatitis A vaccination is recommended in several risk groups. It is not known whether French general practitioners (GPs) identify people at risk.

Objective: To evaluate the knowledge and the implementation of the hepatitis A vaccination recommendations in risk groups in general practice.

Method: We conducted an audit via the Internet among general practitioners teaching at the family medicine department of Paris VII University. Nineteen practitioners participated in the survey and prospectively recruited 108 patients with a chronic liver disease and/or men who have sex with men and/or injection drug users, for whom hepatitis A vaccination is recommended.

Results: The risk groups studied in this sample visited their general practitioner and risk factors were cumulated. HAV serology was only prescribed to 18% of recruited patients, hepatitis A vaccination was proposed to 11% of them and was carried out in only 14% of the cases. General practitioners do not have thorough knowledge of the recommendations: their average score at the final knowledge evaluation was 9.90/20. Half of them were reluctant to vaccinate their patients due to a lack of interest or financial reasons. The recruited patients who were vaccinated (14%) were globally well vaccinated but mainly by general practitioners who had an interest in viral infections. Eighty per cent of those vaccinated were French and had obtained a university degree.

Conclusions: Hepatitis A vaccination recommendations in risk groups were globally not followed by the general practitioners, but 73% of them were convinced of the necessity to vaccinate these risk groups by the end of our study. The use of the Internet in continuing medical education can improve prevention in general medicine.

Points for discussion: How can we improve the ability of GPs to identify patients at risk of hepatitis A? What is the situation in other European countries (discussion with EGPRN members)?

Presentation 16: One slide/five minutes, ongoing study, no results yet

Friday, 9 May 2008, 16.10–16.20

Model record design and results in a check-up centre of a university hospital

Fatih Yuksel, Ilhami Unluoglu, Ayse Cavusoglu, Kismet Bulbul

Department of Family Medicine, Eskisehir Osmangazi University, Buyukdere Mah. Cevahir Sk. No:6/6, 26040 Eskisehir, Turkey. Tel: [phone]; E-mail: [email protected]

Background: We designed this model in a university hospital check-up centre that belongs to a family medicine department to screen most common diseases and lifestyle risks in healthy participants. The model was primarily designed to computerize records and to analyze them. It was also desirable, if needed, to include new questions and test results for new research and questionnaires.

Research question: What simple record design is most suitable in patient screening centres, and what problems might be experienced during its implementation?

Method: In this model, we used SPSS (version 16) for records. A form for history taking and physical examination as well as a cardiovascular risk questionnaire were prepared. We routinely took blood samples to evaluate complete blood count, conducted biochemical studies, lipid profiles, urine analysis, chest X-ray, abdominal ultrasound, and electrocardiography, and identified certain patient-specific risk markers. Moreover, we investigated cardiological risk factors using diet content and physical activity indices. A nurse at the centre transposed all written forms into a digital medium.

Results: We collected 129 relevant participant records. We evaluated the design during this period, and made certain changes according to needs. Data on eating habits and physical activity were obtained from only 65 (50.3%) participants. We are planning to cross-analyse lipid profiles with each of the eating habit properties and with physical activity index when we complete the study design.

Conclusions: This was initially an incomplete database, but improvements have been made, and further adjustments are planned in order to improve the model's design. Moreover, questionnaires and our simple physical activity indices need to be standardized and tested in a sample study of primary care clinics.

Points for discussion: 1) Is it possible to create record designs like this that are suited to our clinic experience? 2) What standardizations for physical activity index and dietary index would be advisable?

Presentation 17: One slide/five minutes

Friday, 9 May 2008, 16.20–16.30

Fibromyalgia: Are GPs able to identify patients’ expectations?

Florent Verfaillie

Department of General Practice, University of Picardie Jules Verne, French Association of Young Researchers in General Practice, 3 rue des Louvels, 80036 Amiens Cedex 1, 80036 Amiens, France. Tel: [phone]; E-mail: [email protected]

Background: Fibromyalgia has been defined by the American College of Rheumatology. Patients feel musculoskeletal pains but no organic lesion is noticed. This disorder may be considered as a model of chronic pain. In this disorder, the relationship between general practice and patients is often considered a difficult one by physicians.

Research question: What do patients expect from their general practitioners (GPs)? Can GPs identify patients’ expectations?

Method: The study investigated 72 patients attending a symposium during the World Fibromyalgia Day and 54 GPs attending a continuing medical education session. Inclusion criteria were, for patients, diagnosis by a physician as suffering from fibromyalgia and, for GPs, taking care of at least one patient suffering from fibromyalgia. Patients had to answer in no more than three words what they expected from their GP. Similarly, GPs had to express in no more than three words what, according to them, patients suffering from fibromyalgia expected from their GPs. Answers were analysed by the qualitative logico-semantic method, by classifying and counting responses.

Results: 68 patients and 51 GPs were included in the study according to the inclusion criteria. Patients declared that they expected from their GP “understanding and listening” (50%), “information and advice” (14.71%), “acknowledgement” (13.24%), and “medical education for GPs about fibromyalgia” (13.24%). GPs thought that patients expected “fewer or no more symptoms” (80.39%), “some listening” (60.78%), “understanding, empathy, or compassion” (45.10%), and “acknowledgement” (31.37%).

Conclusions: Patients’ expectations are not sufficiently identified by GPs. It is important for GPs in association with their patients to elaborate on common and measurable objectives.

Points for discussion: Do patients know consciously what they expect? With fibromyalgia, patients expectations are not different from the general population, but GPs’ perceptions are different.

Presentation 18: One slide/five minutes, study proposal/idea

Friday, 9 May 2008, 16.30–16.40

Ankylosing spondylitis

Dan Baruch, Laurence Coblentz-Baumann

Dept. de Médecine Generale, Université Paris 7, 4 rue Th. Ribot, 75017 Paris, France. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Back pain represents 1.4% of patient requests in general practice in France. Ankylosing spondylitis (AS) is difficult to identify, even though there are consensual criteria for its diagnosis (Amor or ESSG). The prevalence in France of AS is reported to be approximately 0.5%. The pain and risk of ankylosis are variable but always present. Various treatments are available.

Objective: The aim of the present research is to assess the prevalence of AS, which appears to be underdiagnosed. The most important study thus far was conducted by telephone, in which participants were asked if diagnosis had been established, which was verified at a later date. There has been no study of prospective prevalence in the general population. In France, most studies present AS as a predominantly male disease (ratio 4/1) (3). However, in daily practice as well as in a recent study, prevalence according to sex seems to be equal (ratio 1/1).

Research question: The prevalence of AS is higher than indicated by official statistics. How many patients visit a GP for back pain or fulfil the criteria of AS? What is the sex ratio among patients with AS?

Method: Quantitative study, multicentric, conducted in GP offices, over a period of 3 months. Among patients with back pain, the clinical criteria of AS will be assessed and, if positive, a test will be run in order to confirm this. The sex of each patient will be recorded. Prevalence of diagnosed AS will be calculated among back pain patients, and among global consultations. The sex ratio of diagnosed AS will be calculated.

Results: Not yet obtained

Presentation 19: One slide/five minutes, study proposal/idea

Friday, 9 May 2008, 16.40–16.50

Improving general practitioners’ care for COPD patients through computerized checklists and feedback

Mette Koefoed

Research Unit for General practice, University of Southern Denmark, J. B. Winsløws Vej 9A, 5000 Odense C, Denmark. Tel: [phone] 6699; E-mail: [email protected]

Background: There is a gap between evidence-based recommendations and clinical practice. Several interventions to improve professional performance are available, but none have proved to be consistently effective. Systematic registration of patients’ diagnoses is increasingly being used in general practice in Denmark. Combining this with reminders and feedback may improve implementation of evidence-based recommendations in general practice. Chronic obstructive pulmonary disease (COPD) will be used for testing the intervention.

Research question: To assess the effectiveness of systematically registering patients, using computer-based checklists as decision support and reminders, and feedback with clinical data at the patient level.

Method: Unblinded randomized controlled trial. All general practices in the region of southern Denmark will be invited to participate. Each practice will be randomized to one of three management strategies: 1) registering patients, 2) registering patients and filling out checklists, and 3) registering patients, filling out checklists, and receiving feedback. Follow-up time 2 years. Primary outcome measure will be hospital admissions in COPD patients. Secondary outcomes will be practitioners’ use of spirometry, their prescription pattern with COPD-related medications, and influenza vaccinations. This project will be part of a PhD study.

Points for discussion: 1) Study design; 2) outcome; 3) generalizability (especially to other European countries)

Presentation 20: One slide/five minutes, study proposal/idea

Friday, 9 May 2008, 16.50–17.00

A plea for a common language in general medicine

Waltraud Fink, Gustav Kamenski, Dietmar Kleinbichler

Department of Public Health, Family Medicine University Vienna, Karl-Landsteiner Institute, Straning 142, 3722 Straning, Austria. Tel: [phone]; E-mail: [email protected]

Background: There is an ongoing debate regarding the labelling and coding of diseases. The musculoskeletal system can serve as an example of how a consensus on concepts could be reached.

Research question: Can R. N. Braun's labelling and classification system lead to a multi-language agreement?

Method: We reviewed the “casugraphic” concepts developed by R. N. Braun along with his long-term “case” (i.e., episode of care) recordings, later described by him and others with inclusion criteria as well as with the usual course of illness. The casugraphic concepts are designed to be mutually exclusive. Each one has a list of concurrent labels together with potentially dangerous conditions for the specific common disorder. Of the 300 concepts for illnesses which occur with regular frequency, those applied to disorders of the musculoskeletal system were chosen, looking for accordance in ICD and ICPC.

Results: Twenty-six casugraphic labels for the musculoskeletal system were identified. They will be displayed on a slide according to their anatomical distribution. Half of them are localized at a specific anatomic site (e.g., epicondylitis humeri); the location of the other half varies (e.g., myalgia, acute arthritis, etc.). The discussion in Antalya will provide information on whether the concepts could be used in practices in other countries with other classification systems.

Conclusions: The 26 different musculoskeletal disorders are of specific interest for the GP as they occur with regular (and some with high) annual frequency. As the casugraphic concepts represent the usual range of complaints seen in an average general practice, any symptom or sign which does not fit should raise the GP's attention and be labelled with a “red flag”. The casugraphic concepts are validated by long-term practice morbidity registration, but only in few practices—mostly because they are not known. A broader consideration could show their applicability and usefulness.

Points for discussion: As we continue the process of reviewing all of Braun's 300 casugraphic concepts, published 15 years ago in German and French, we are interested in 1) whether our international colleagues find them interesting and applicable in their practices.

Presentation 21: One slide/five minutes, ongoing study with preliminary results

Friday, 9 May 2008, 17.00–17.10

Time spent by doctors on administrative tasks

Jean Paul Canevet, Ariane Richard

Dépt. de Médecine Générale, Faculté de Médecine de Nantes, 1 rue Gaston Veil, 44000 Nantes, France. Tel: [phone]; E-mail: [email protected]

Background: The medical demography of general practitioners is in relative decline in France, which leads us to think about refocusing medical practice on medical tasks. Indeed, general practitioners have the impression that they devote a lot of time to tasks that are not merely care.

Objective: In this study, the objective was to measure time spent by doctors on administrative tasks.

Method: An investigation was carried out by direct observation in the office of three general practitioners. These practices were in three different environments: rural, peri-urban, and urban. The investigator was a trainee. She attended the consultations, and timed the amount of time spent on administrative work, according to predefined criteria. Data were reported in a grid detailing the various types of administrative tasks. One hundred consultations were analysed for each general practitioner (i.e., 300 in total).

Results: The analysis of a part of the results shows the wide range of administrative tasks performed by a general practitioner. The average time spent on these tasks was about 3 min for each consultation (one-fifth of the time allotted), which is important in terms of consulting time. Moreover, this is responsible for many interruptions during the workday. These results need to be compared with data from opinion polls conducted by surveys on the same subject.

Conclusions: A part of the time of general practitioners could be optimized, refocused on care and on medical decision-making, by the questioning of some administrative procedures and the delegation of tasks.

Points for discussion: Do you have the same questions/issues in your countries?

Presentation 22: Theme paper, study proposal/idea

Saturday, 10 May 2008, 09.30–10.00

Occupational risk factors for musculoskeletal symptoms among artists

Serap Cifcili, Mehmet Akman, A. Uzuner, Cigdem Apaydin Kaya, P. Unalan

Department of Family Medicine, Marmara University, Tophanelioglu Cad Marmara University Hospital Altunizade, Uskudar, 34662 Istanbul, Turkey. Tel: [phone]; E-mail: [email protected]

Background: Quite a number of academic staff from the Faculty of Fine Arts have been admitted to our family medicine outpatient clinic during recent years with complaints of musculoskeletal symptoms. Occupational risk factors for musculoskeletal symptoms include repetition of micro-traumas, standing in non-ergonomic postures, fixed body positions, forces concentrated on small parts of the body, and lack of sufficient rest between tasks such as working in a fine-art studio. Fine-art faculty members have occupational risk factors. The purpose of this study is to find the prevalence, incidence, and possible risk factors of low back pain and other musculoskeletal disorders among the members of a fine-art faculty.

Research question: What are the prevalence, incidence, and possible risk factors of low back pain and other musculoskeletal disorders among the members of a fine-art faculty?

Method: Cross-sectional survey will be carried out on a representative sample of 150 academic staff and 1500 students. Data related to musculoskeletal symptoms and work-related risk factors will be collected by questionnaire and an appropriate assessment tool. Physical examination (PE) including GALS (gait-arms-legs-spine) locomotor system inspection will be performed on each subject. Subjects with a suspicious clinical finding will be referred to hospital for further diagnosis and treatment. Four years’ follow-up with 6-month intervals is planned to determine the incidence of musculoskeletal symptoms and to establish a causal relationship between symptoms and related risk factors. Each follow-up will include PE, GALS inspection, and a self-administered questionnaire.

Conclusions: By determining common musculoskeletal problems and related risks among artists, we can develop better management and prevention strategies specific to this group.

Points for discussion: 1) What assessment tools could be used to identify musculoskeletal disorders and risk factors specific to the research population? 2) What should be the critical methodological points regarding musculoskeletal research on this specific group?

Presentation 23: Theme paper

Saturday, 10 May 2008, 10.00–10.30

Dentists' spine problems and physical activity

Nuri Topsakal, Pemra C. Ünalan, Mustafa Karahan, Serap Çifçili, Arzu Uzuner

Department of Family Medicine, Marmara University Medical Faculty, Çiçekli Bostan, Sok. MESA Koruevleri C4/D2 Altunizade, 34662 Istanbul, Turkey. Tel: [phone]; E-mail: [email protected]

Background: The dental profession exposes dentists during their work to many harmful factors. One of those factors is the irrational posture adopted during working hours, which causes discomfort and disorders of the musculoskeletal system and the peripheral nervous and venous system.

Objective: Dentists are health professionals who can easily understand the health benefits of physical activity. This study aimed to evaluate the frequency of both the musculoskeletal discomfort and physical activity that they report.

Method: This study was of a cross-sectional design, using a self-reported questionnaire that was completed by 503 dentists, at their offices, who were randomly selected from the lists of the Turkish dentist association.

Results: 63.6% of the study population was male, 50.1% were 35 years of age, 48.1% had been in the profession for 10 years, 87.5% worked 5 days/week, 44.5% worked 9 hours/day, 19.1% reported discomfort in at least two parts of the spine (cervical, dorsal, and/or lumbar), and 92.6% mentioned that regular physical activity might improve their discomfort, but only 55.3% did regular physical exercise. More male dentists did regular physical exercise than females (p=0.007), and, as much as they believe in the health benefits of exercise, they exercise regularly (p=0.028). Interestingly, those whose total spine discomfort score was higher exercised regularly (p=0.024).

Conclusion: Having a busy working schedule or having musculoskeletal disorders is not a barrier for physical exercise. In particular, belief in the effectiveness of health is the major point that influenced dentists’ change in behaviour. Some physical exercise programmes may be presented to dentists as a form of prophylaxis concerning the musculoskeletal system.

Points for discussion: What analysis might be added to this study? Might a behavioural intervention be useful?

Presentation 24: Theme paper

Saturday, 10 May 2008, 10.30–11.00

Musculoskeletal complaints facilitate recognition of somatization

Berend Terluin, Harm W. J. van Marwijk, Petra Jellema

Department of General Practice, EMGO Institute, VU University Medical Centre, Rotterdamweg 4, 1324 LN Almere, the Netherlands. Tel: [phone]; E-mail: [email protected]

Background: Most general practice patients with psychosocial problems present with physical symptoms. In most cases, doctors fail to establish psychosocial diagnoses, even though they may recognize the psychosocial background of the symptoms. Little is known about the role of musculoskeletal illness.

Research question: 1) Are psychosocial problems associated with musculoskeletal illness? 2) Do general practitioners (GPs) recognize psychosocial problems and is this mediated by musculoskeletal illness?

Method: Cross-sectional cohort study in general practice. Thirty-seven GPs included 2127 consecutive adult patients (15–64 years of age). Psychosocial problems were assessed with the Four-Dimensional Symptom Questionnaire (4DSQ). GPs recorded their diagnoses and assessment of a possible psychosocial background. Patients with musculoskeletal diagnoses were compared with patients presenting other physical diagnoses regarding proportions with elevated 4DSQ scores. Using logistic regression analysis, GPs' recognition of psychosocial problems was related to elevated distress, depression, anxiety, somatization, and type of physical complaints (musculoskeletal or other).

Results: 516 patients had musculoskeletal diagnoses, and 1338 patients had other physical diagnoses. These groups differed with respect to gender (men 40% vs 29%) and elevated anxiety (7% vs 10%), but did not differ in age, elevated distress, depression, somatization, or recognition of psychosocial problems (22%

vs 24%). Recognition of psychosocial problems was associated with elevated distress and somatization, the interaction between distress and somatization, and the interaction between musculoskeletal illness and somatization. In patients with elevated distress and low or moderate somatization, GPs recognized psychosocial problems in 26–50%. Severe somatization led to the recognition of psychosocial problems in the presence of musculoskeletal illness in 58–85%, but in its absence in 17–46%.

Conclusions: Musculoskeletal illness is in itself not associated with psychosocial problems, but it does facilitate the recognition of such problems in general practice patients presenting with physical illness.

Points for discussion: Any idea why musculoskeletal illness seems to sensitize GPs to severe somatization in their patients while other physical illness tends to make GPs less sensitive?

Presentation 25: Poster

Saturday, 10 May 2008, 11.30–13.00

Are PPI correctly used by GPs in patients with osteoarthritis or rheumatoid arthritis at high risk of gastroduodenal events?

Nicola Buono, Ferdinando Petrazzuoli, Filippo D. Addio, Carmine Farinaro, Baldassarre Mirra, Jean Karl Soler

Department of Family Medicine, SNAMID Caserta, Via Tartari 5, 81010 Prata Sannita, Italy. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Chronic treatment with non-steroidal antiinflammatory drugs (NSAIDs) increases the risk of peptic ulcer (PU) and serious gastroduodenal complications more than COX-2 inhibitors. In Italy, proton-pump inhibitor (PPI) prescription is free of charge when patients are chronically treated with NSAIDs or low dose of aspirin in the presence of the following conditions: 1) positive anamnesis for past PU or gastrointestinal bleeding; 2) concomitant use of anticoagulant and steroid drugs, and 3) old age.

Research question: Do general practitioners (GPs) usually follow guidelines when they prescribe NSAID and/or COX-2 inhibitors in patients with osteoarthritis (OA) or rheumatoid arthritis (RA) and high risk of gastroduodenal damage?

Method: A cross-sectional study of a representative sample of adults (13 724), 11 870 of whom had at least one prescription of antiinflammatory drugs between 1 January 2005 and 30 June 2007, and were selected from the electronic clinical patient records of 10 GPs.

Results: 3208 (23,4%) out of 11 870 had OA and 50 (0,34%) had RA. The prevalence of OA and RA (2006) was 14.8% and 0.2%, respectively, while the percentage of females with OA and RA was 65% (average 68.3±11.9%) and 95.8% (average 67.4±12.3%), respectively. The relative risk (RR) of gastroprotection in patients treated with drugs for OA and RA was 1.69 (95% confidence interval [CI] 1.54–1.86) for all antiinflammatories, 1.60 (95% CI 1.45–1.76) for NSAIDs, 2.48 (95% CI 1.95–3.14) for COX-2 inhibitors, and 2.73 (95% CI 2.09–3.57) for NSAIDs and COX-2 inhibitors in combination.

Conclusions: GPs do not usually follow guidelines when they prescribe NSAIDs and/or COX-2 inhibitors in patients with OA or RA. Nimesulid and ketoprofen represent the most common prescriptions for OA (70%). Gastroprotection is underperformed in subjects with high risk of gastroduodenal damage, while better gastroprotection is received for those who are on COX-2 inhibitors.

Points for discussion: 1) What is the prevalence of OA and RA in our respective countries? 2) Are PPI free of charge when prescribed to patients with high risk of gastroduodenal damage treated for OA/RA with antiinflammatory drugs? 3) Do our GPs follow the advice of guidelines?

Presentation 27: Poster, ongoing study with preliminary results

Saturday, 10 May 2008, 11.30–13.00

Young French women with hypovitaminosis D and musculoskeletal pain

Contardo Gaelle, Marie France Le Goaziou, S. Belaid, C. Dupraz

University Claude Bernard Lyon 1, 121 rue Professeur Beauvisage, 69008 Lyon, France. Tel: [phone]; E-mail: [email protected]

Background: Dr S. Belaid showed that hypovitaminosis D is frequent in the young veiled female population in the Rhône Alpes area.

Research question: Gaelle Contardo with the GP Lyon College group is conducting a new study in the general female population that addresses three research questions: What is the hypovitaminosis D prevalence in this young female population? What are the confounding factors: sunlight exposure, diet, skin colour, clothes, sport? How many vitamin D units should be given to each woman per year in order to maintain a correct blood level (75 nmol/l)?

Method: 200 non-veiled women and 100 veiled women, 19–49 years of age, will be selected by 12 GP investigators in the Rhône Alpes area, during winter 2008. Data will be extracted from a questionnaire concerning quality of life (SF12), quantity of vitamin D in their food, sunlight exposure time, and vitamin D and PTH dosages. Secondly, a deficiency group will be treated with vitamin D doses and will be followed for 18 months. Nutritional education and advice about sunlight exposure will be given. Every 3 months, vitamin D and PTH dosages will be given and, finally, the same questionnaire will be administered.

Results: 90% of veiled women are expected to be severely deficient in vitamin D and 30% of the non-veiled women. Seventy per cent of the deficient women are expected to be tired or to suffer from musculoskeletal pain. Clothes are expected to be the most important factor of deficiency, but it is possible that diet will also be important. We expect an improvement in the blood level of vitamin D, but most important is better quality of life and, in particular, less musculoskeletal pain and tiredness.

Conclusion: In May, data and results from the first part of the study will be available to be presented in Antalya.

Points for discussion: 1) Are the same problems encountered in others countries? 2) Do you have guidelines concerning this issue?

Presentation 28: Poster, study proposal/idea

Saturday, 10 May 2008, 11.30–13.00

Can a short exercise prescription for osteoarthritis patients provide better adherence?

Serap Cifcili, Mehmet Akman, P. C. Unalan, A. Uzuner, C. Apaydin

Department of Family Medicine, Marmara Medical School, Tohanelioglu c. No:15-17 Altunizade, 34662 Istanbul, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Exercise prescription is a major part of osteoarthritis management. However, poor adherence is the most common explanation for the declining impact of the benefits of exercise over time.

Research question: Can a short exercise prescription with demonstration in the general practitioner's office improve exercise adherence among patients with osteoarthritis?

Method: Patients who are admitted to our outpatient clinic with knee pain, and who fulfil the American College of Rheumatology criteria or have radiographically established knee osteoarthritis will be included. A randomized controlled trial with three groups is planned. The first group will receive a short exercise prescription primarily focusing on strengthening the quadriceps and demonstration with a leaflet; the second group will receive a more comprehensive exercise programme with a leaflet; and the third group will be given a short exercise prescription with only a leaflet. A minimum of 50 patients in each group will be included. Primary outcome measure will be exercise adherence, determined by patient diaries, weekly follow-up phone calls, and monthly face-to-face interviews for at least 6 months. A researcher who is blinded to initial randomization will perform follow-up interviews. Follow-up secondary outcome measure will be the effectiveness of exercise, assessed with Western Ontario and McMaster Universities Osteoarthritis (WOMAC) scores and thigh circumference.

Results: We expect that the “short prescription, demonstration, and leaflet” group will adhere more to exercise, giving rise to better WOMAC scores.

Conclusions: This study might have practical implications for osteoarthritis management.

Points for discussion: 1) Other methods for exercise adherence follow-up? 2) Is this an appropriate randomization design for this type of research question?

Presentation 30: Poster

Saturday, 10 May 2008, 11.30–13.00

Musculoskeletal disorders in a family medicine clinic in Turkey

Tamer Edirne, Ozcan Hiz, Mahir Baloglu

Department of Family Medicine, Erdemli Public Hospital, Erdemli Devlet Hastanesi, 33100 Mersin, Turkey. Tel: [phone]; E-mail: [email protected]

Background: The prevalence of musculoskeletal complaints is high in general practice. The aim of this study was to investigate the characteristics of patients with musculoskeletal complaints of the low back, neck, shoulders, and hand/wrist.

Research question: 1) What are the frequencies of musculoskeletal disorders in family practice? 2) What are the re-admission rates for patients with musculoskeletal disorders?

Method: 18 366 patient records from a family medicine clinic at the Public Hospital of Erdemli, Turkey, were investigated for musculoskeletal complaints and problems. Between January 2005 and July 2007, data from 1645 patients were found and evaluated for age, sex, complaint, average duration of outpatient care, and re-admission rates.

Results: Preliminary results: Mean age of patients with musculoskeletal disorders was 46.6 years. Female patients were in the majority, with 66.5% (1093). Myalgia (40.2%), sprain and strain of lumbar spine (11.9%), other soft-tissue disorders, and not classified elsewhere (6.0%) were the most diagnosed disorders. Patients spent 5 days on average in outpatient care.

Conclusions: Mean age of patients with musculoskeletal disorders is relatively low in the population of Erdemli.

Points for discussion: 1) What are the reasons for the appearance of musculoskeletal disorders among young patients in Erdemli? 2) How can musculoskeletal disorders be prevented?

Presentation 31: Poster, ongoing study with preliminary results

Saturday, 10 May 2008, 11.30–13.00

Low back pain in general practice

Maria Panchovska, Lyubima Despotova-Toleva, Nikolai Nikolov, Gergana Foreva, Radost Asenova

Department of General Practice, MU Plovdiv, 15a Vasil Aprilov, 4000 Plovdiv, Bulgaria. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Low back pain is a considerable health problem. However, in Bulgaria, guidelines for the management of low back pain in primary healthcare settings have not been published. It is known that low back pain can be determined by inflammatory, degenerative, malignant, traumatic, and other reasons. The most important symptoms of non-specific low back pain are pain and disability.

Objective: To explore how general practitioners (GPs) manage patients with low back pain.

Method: Design: structured questionnaire. Setting: 12 general practices. Participants: 12 general practitioners. Main variables measured: prevalence of low back pain in general practice and reported need for interdisciplinary approach to solve patient's problems. Analysis: alternative analysis. Statistics: SPSS version 15.0.

Results: GPs reported that about 1/3 of their patients have been consulted for low back pain. The results reveal that, in 100% of cases, the general practitioners referred their patients to a consultant neurologist and significantly fewer of them referred patients to rheumatologists, orthopaedists, or other specialists (40%).

Conclusions: Guidelines for the management of low back pain in primary healthcare settings are needed in Bulgaria.

Presentation 33: Poster

Saturday, 10 May 2008, 11.30–13.00

A school for osteoporosis: A new experience for effective health promotional activities in Bulgarian family practice

Valentina ?adjova, Svetlana Hristova, Andrej Zabounov, Paraskeva Mancheva

Department of Family Medicine, Medical University, Varna, Bulgaria. E-mail: [email protected]

Background: The widespread prevalence of osteoporosis puts the condition in third place among socially significant diseases. A basic marker for its development and related fractures is bone density. Risk assessment includes not only biological factors but also factors connected with the lifestyle of the patient. As bone density is accumulated in childhood and in late adolescence, these periods are very important in primary promotion of osteoporosis awareness.

Research question: Is the new form of health promotional activity–-a school for osteoporosis–-effective in putting into practice new behaviour models for changing eating habits and physical activities in the families of adolescents with obesity, predisposed to developing early osteoporosis?

Methods: An investigation into individual eating habits, physical activity, family predisposition to osteoporosis, and risk factors. Questionnaires are completed independently by patients and parents. Adolescents are educated together with their parents for 2 hours, twice a week. Patients: 42 adolescents, aged between 10 and 18 years (26 girls, 16 boys), and their parents–-patients of 112 general practitioners. We initially measure height, weight, waist circumference, blood glucose, lipids, physical activity, and eating regimen, which are evaluated after 3 months of education. The study will be completed after 36 months.

Results: Advice and education are given by GPs, together with paediatricians, endocrinologists, dieticians, and psychologists. The results are not shown, since there are still 2 months remaining of the study.

Conclusions: Special education should help families to improve their attitudes towards leading a healthy lifestyle. The success of the “school of osteoporosis” will support the health promotional activities of GPs in reducing osteoporosis.

Presentation 34: Poster, ongoing study with preliminary results

Saturday, 10 May 2008, 11.30–13.00

Patients’ perceptions of osteoarthritis handling

Isabelle Aubin, Laurence Baumann-Coblentz, Dan Baruch, V. Zéline, Alain Mercier

Dept. Paris 7, University Denis Diderot, 40 rue Carnot, 95230 Soisy/Montmorency, France. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Osteoarthritis is a very common pathology, affecting approximately 10 million patients in France. It is a disabling disease leading to possible important changes in a patient's life because of both pain and movement limitations. From a general practitioner's (GP's) perspective, the condition's mild pathology may be considered somewhat banal since it does not represent a threat to life. EBM recommendations propose first non-pharmacological measures such as daily activity, weight reduction, and physiotherapy. The first medication recommended is paracetamol. It appears that there could be a gap between the patient's perception of his/her pathology and his/her everyday life impairment and the medical management that offers no “heavy” treatment, relieves pain but does not treat, and involves the patient's goodwill. It is likely that some patients will turn to alternative medicine because they feel misunderstood and not cared for and are disappointed by traditional care.

Research question: How do patients feel about their osteoarthritis treatment and its apparent simplicity?

Method: Qualitative study. Semi-structured interviews will be conducted by four researchers, among patients from different regions in France, Paris and its suburbs, which will be typed, transcribed, coded, and analysed. Until saturation of data. Patients with different osteoarthritis ailments will be interviewed. Three pilot interviews will be conducted in order to be able to modify the questions.

Expected results and conclusions: This study is ongoing. We expect a better understanding of patients’ perceptions of their pathology and their expectations, which could improve management of the disease.

Points for discussion: Do you think your osteoarthritis patients seek other types of care?

Presentation 35: Poster

Saturday, 10 May 2008, 11.30–13.00

Education on reproductive health: It works!

Vildan Mevsim, Dilek Güldal, Tolga Gunvar, Özge Saygin, Emel Kuruolu

Dokuz Eylul Universitesi Aile Hekimligi AD Inciralti, 35340 Izmir, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background:

While the Turkish population continues to get younger, the proportion of the population capable of reproduction is also increasing and is expected to reach the 40% mark in 2025. Due to the rise and spread of HIV/AIDS, providing a number of services addressing the sexual and reproductive health needs of young people has become more important.

Research question: How much does the level of students’ knowledge of sexual and reproductive health and the usage rate of special reproductive health units change as a result of receiving peer education and reproductive health counselling?

Method: A “before and after” intervention study was conducted. A questionnaire was administered to first-year university students both at the beginning and at the end of the academic year. Of 5236 students, 1734 and 1345, respectively, agreed to participate in the study. Between the two surveys, a training programme was conducted, including peer education. SPSS 12.0 was used for data evaluation and the chi-square and Mann-Whitney U tests for statistical analyses.

Results: The knowledge score concerning reproductive health increased by 15.28%, that of sexually transmitted disease by 10.56%, and that of contraception methods by 9.92%. In general, the total knowledge score increased by 2.77%.

Conclusions: Sexual and reproductive health issues are difficult to discuss openly, but it is possible to change the attitude, behaviour, and knowledge levels of young people by using truthful, direct, and honest approaches.

Points for discussion: Can we design a standard questionnaire for primary care to evaluate knowledge, attitude, and behaviour concerning reproductive health that can easily be applied to Europe?

Presentation 36: Poster

Saturday, 10 May 2008, 11.30–13.00

Is the use of videotape recording superior to verbal feedback alone in the teaching of clinical skills?

Vildan Mevsim, Nilgun Ozcakar, Dilek Guldal, Tolga Gunvar, Ediz Yildirim, Zafer Sisli, lgi Semin

Dokuz Eylul Universitesi Aile Hekimligi AD Inciralti, 35340 Izmir, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Recently, medical schools have become committed to teaching good communication and good history-taking skills. However, there remains an unresolved question as to what constitutes the best educational method.

Research question: Is the use of videotape recording superior to verbal feedback alone in the teaching of clinical skills?

Methods: A randomized controlled trial was designed. The study was conducted among 52 Dokuz Eylul University Faculty of Medicine second-year students. All students’ performances of communication and history-taking skills were assessed twice. Between these two assessments, the study group received both verbal and visual feedback by watching their video recordings of patient interviews; the control group received only verbal feedback from the teacher.

Results: Feedback based on videotaped interviews is superior to the feedback given solely based on the observation of trainers.

Conclusions: Even though feedback from videotape is superior, the financial costs and the extensive length of time involved in videotaping must be taken into consideration.

Presentation 37: Poster

Saturday, 10 May 2008, 11.30–13.00

Evaluation of the teaching of humanities and social sciences and humanism after the first year of medical studies at the Rennes Faculty of Medicine

Nicolas Chauvel

Department of Family Practice, Rennes, Faculté de Médecine, Avenue du Pr Léon Bernard, 35043 Rennes, France. Tel: [phone]; E-mail: [email protected]

Background: Extraordinary scientific advances have been made in medicine during the last century, leading in some instances to a form of scientific illusion. As a result, there has been a call, in both the profession and society at large, for a re-introduction of the humanities into early medical education. It is in this context that the social sciences were introduced to the Rennes Faculty of Medicine. We found a large body of work concerning the evaluation of this type of teaching carried out in the first year of medical studies, but little regarding the years following. However, it is during these later years that future doctors will be confronted with the concept of “man's suffering”.

Objective: This study was exclusively interested in the teaching of social sciences and humanism after the first year of medical study, and more specifically in Rennes.

Method: Firstly, a quantitative approach was used by means of a questionnaire including 31 closed-answer questions and one open-ended question. The questionnaire was distributed to postgraduate students at the faculty of general medicine. Perceptions of the teachers in social sciences concerning the subject were then collected by a qualitative approach, using semi-directed interviews.

Results: It transpires inter alia that the majority of respondents favour this type of teaching, but that they find it insufficient and unsuitable in many respects. A lack of resources and various institutional barriers are frequently reported by professors.

Conclusions: According to the investigation, the teaching of social sciences and humanism appears to be essential, but also insufficient, unsuitable, and underevaluated at the institutional level. Proposals are thus put forth to try to resolve these problems: in particular, the creation of a recognized teaching structure and a new medical model for future doctors.

Points for discussion: 1) A new medical model for future doctors; 2) the doctor–patient relationship in light of these results.

Presentation 38: Poster

Saturday, 10 May 2008, 11.30–13.00

Being a role model in the early years of medical school

Nilgun Ozcakar, Mehtap Kartal, Dilek Guldal

Dokuz Eylul Universitesi Tip Fakultesi Aile Hekimligi AD, 35340 Izmir, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Medical doctors are expected to be role models for society. They are supposed to gain this position during their time at medical school. Being confronted with real patients during their clerkship years may have an important role in this process. It may therefore be necessary to find out whether there is a change towards being a role model during the pre-clerkship years of medical school.

Research question: What affects the relationship between emphasis on health and health behaviour on pre-clerkship medical students?

Method: This cross-sectional study conducted in June 2007 included 222 (83.7%) pre-clerkship students at the DEU Medical Faculty. The questionnaire consisted of the Healthy Life Style Behaviour Scale (HLSBS) developed by Pender, with various subscales including health responsibility, physical activity, and nutrition, the Health

Emphasis Scale (HES) developed by Wallston, and questions related to the socio-demographic characteristics of the students. We used SPSS 12.0 for statistical analysis, and the t test and Mann-Whitney U test.

Results: Of the students, 72.2% were male, with a mean age 21.38±1.00 years; 64.9% believed that their living environment affected their health positively, 20.3% smoked, 44.6% used alcohol, and 10.8% had a chronic disease. Mean scores of HLSBS across all sections were similar among classes (p>0.05). The students’ score for HES was as high as 7.80±2.67 out of 10; however, this was not supported by subscales of HLSBS. The mean score for health responsibility was 22.14±4.66 (min. 10, max. 40), for nutrition 15.09±3.29 (min. 6, max. 24), and for physical activity 10.57±3.14 (min. 5, max. 20). Overall, mean score for HLSBS was 122.50±18.24 (48–192).

Conclusions: Although medical students are aware of the importance of health and leading a healthy lifestyle, their behaviour does not reflect this and they seem far from being role models in their early years of education. Some changes to the medical curriculum may be necessary in order to promote more healthy behaviour among medical students.

Points for discussion: Changes in the medical curriculum aimed at health promotion and/or special programmes for general practitioners responsible for the healthcare of university students should be considered.

Presentation 39: Poster

Saturday, 10 May 2008, 11.30–13.00

Evaluation of introduction to clinical practice by third-year students at Dokuz Eylul University

Nilgun Ozcakar, Dilek Guldal, Vildan Mevsim, Zafer Sisli

Dokuz Eylul Universitesi Tip Fakultesi Aile Hekimligi AD, 35340 Izmir, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: In the third year of the Medical Faculty of Dokuz Eylul University (DEU), there is an “Introduction to Clinical Practice” programme that targets development of patient–physician interview skills. Students meet real patients before attending clinics and use all the techniques related to history-taking practices they have learned.

Research question: What was the evaluation of the students on “Introduction to Clinical Practice” for different clinical settings?

Method: This descriptive cross-sectional study included 90 (72.0%) third-year students at the Medical Faculty of DEU. In the year 2003–2004, feedback forms for “Introduction to Clinical Practice” were used to determine the evaluations of students concerning history-taking and physical examination practices in adults in primary care centres and the medical unit of DEU, as well as paediatric age groups at DEU hospital and the state hospital, as experienced with real patients under clinical skills trainer supervision. Scoring in the forms was between 1 and 5 for both trainers and practices. We analysed the data with SPSS 11.0 and used the t test for analysis.

Results: Mean scores of the students ranged between 3.46 and 4.70 for different items. There was a significant difference between DEU hospital and the state hospital for adult practice (p=0.032), and the students reported that they experienced different clinical cases more often in the state hospital (p=0.006). For paediatric practice, there was also a significant difference (p=0.011) between DEU hospital and the state hospital in terms of the evaluation of practice environment. The students stated that they felt like a doctor and felt comfortable during their interviews. They believed that the trainers were concerned and committed to the practice (4.71), were supportive (4.52), and gave appropriate feedback (4.64).

Conclusions: For medical students, it is important to meet real patients from different age groups in different environments for history taking and physical examination, and to observe the clinical decision-making process of their supervisors.

Points for discussion: The medical curriculum needs to provide students with early exposure to patients in different clinical settings, especially in primary care settings, in order for medical students to be more comfortable and confident in clinics.

Presentation 40: Poster

Saturday, 10 May 2008, 11.30–13.00

Prescription audit of anti-hypertensive drugs in diabetic patients with arterial hypertension at a general practice in central Portugal

Tiago Villanueva, Ana Maria Pisco, Tânia Pires Silva

Department of Family Medicine, USF Tornada, Estrada Nacional 8, 27, r/c, 2500-315 Caldas da Rainha, Portugal. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Around 20–60% of diabetic patients also have arterial hypertension. Patients with diabetes mellitus have an increased risk of developing renal complications and other organ damage. Angiotensin converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARB) comprise first-line therapy for hypertension in diabetic patients according to the 2007 guidelines for the management of arterial hypertension published by the European Society of Cardiology.

Research question: What is the quality of the prescription of anti-hypertensive drugs in diabetic patients with arterial hypertension at a practice in central Portugal?

Method: The electronic health record software used at the general practice enabled the creation of a statistical group that included all diabetic patients with arterial hypertension registered at the practice. Afterwards, the electronic health record of each patient was consulted in order to gauge and register the type of anti-hypertension medication prescribed. A double-entry table was then constructed using Microsoft Excel, which registered the number of patients only medicated with ACEI, the number of patients only medicated with ARB, the number of patients simultaneously medicated with ACEI and ARA, the number of patients medicated with anti-hypertensive drugs other than renin-angiotensin system blockers, and the number of patients without any anti-hypertensive medication.

Results: Out of 179 diabetic patients with arterial hypertension in the practice, 166 (92.74%) were medicated with one or more classes of anti-hypertensive drug. A total of 142 patients (79.32%) were medicated with at least one blocker of the renin-angiotensin system.

Conclusions: About 80% of diabetic patients with hypertension at a general practice in central Portugal are medicated with first-line therapy for hypertension according to the 2007 Guidelines for the Management of Arterial Hypertension by the European Society of Cardiology.

Points for discussion: 1) Other similar studies in other countries; 2) limitations of this sort of study

Presentation 41: Poster

Saturday, 10 May 2008, 11.30–13.00

A physiologic events cascade, irritable bowel syndrome, may even terminate with chronic gastritis

Mehmet Rami Helvaci, Cahit Ozer, Mustafa Cem Algin, Hasan Kaya, Guven Kuvandik

Department of Family Medicine, Mustafa Kemal University School of Medicine, 31100 Antakya, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: When specifically asked, about one-third of people report recurrent upper-abdominal discomfort, and irritable bowel syndrome (IBS) and chronic gastritis (CG) are among the most frequently diagnosed.

Research question: Is there any relationship between IBS and CG?

Method: Consecutive patients with upper-abdominal discomfort presenting to the Internal Medicine Polyclinic were included in the study. IBS is diagnosed according to Rome II criteria and CG is diagnosed histologically. All cases with IBS were put into one group, and the age- and sex-matched and randomly selected cases without IBS were put into the other group. Comparison of proportions was used as a method of statistical analysis.

Results: One hundred and fifty-six patients with IBS and 179 patients without IBS were studied. CG was detected in 72.4% (113 cases) of cases with IBS, whereas this ratio was only 36.3% (65 cases) in patients without IBS (p<0.001).

Conclusions: IBS may be a cascade of several physiological events, initiated by infection, inflammation, and psychological disturbances such as stress, and eventually terminated with dysfunctions of the gut and potentially other systems of the body via a low-grade inflammatory process. CG may therefore be one of the terminating points of the IBS physiological events cascade. In this way, the gap between the underlying aetiology and clinical onset of CG may be explained by the high prevalence of IBS in society. Keeping in mind this association will be helpful for physicians during prevention, treatment, and follow-up of these common pathologies in primary health centres and internal medicine and gastroenterology polyclinics.

Presentation 42: Poster

Saturday, 10 May 2008, 11.30–13.00

Urgent health problems and family medicine

Yesim Edirne, Dilek Kusaslan Avci, Tamer Edirne

Department of Family Medicine, University of Yuzuncu Yil, Arastirma Hastanesi, 65300 Van, Turkey. Tel: [phone]; E-mail: [email protected]

Background: The structure of healthcare services is changing in Turkey. Many pilot cities have established family medicine offices for the delivery of primary care. The preparedness of family physicians for urgent health problems outside normal office hours needs to be evaluated.

Research question: 1) What are the types and characteristics of urgent health problems in this region? 2) Are all urgent admissions to the emergency department (ED) appropriate?

Method: We designed this study prospectively to record the characteristics, problems, outcomes, and behaviour of patients admitted with urgent health problems to the ED of the University Hospital in Van, Turkey. Presented problems and diagnoses recorded based on the International Classification of Diseases 10 (ICD-10) coding system were compared with the International Classification of Primary Care-2-Revised (ICPC-2-R) coding system. The data were analysed in SPSS, version 12.0. Pearson's chi-square test was used to test for differences in two-by-two tables, using a level of significance of p<0.05.

Results: Data were collected on 123 patients with a mean age of 22.6 years. According to ICPC-2-R, a total of 109 complaints (codes) were used by patients. Only 39% of admissions were hospitalized, of which 23.6% were managed ambulatory, and 34.1% of patients were discharged from the ED after a short observation period.

Conclusions: Many admissions to the ED could be handled in the primary care setting. Data from the ED may reflect the profiles of urgent health problems in a community. Practices can initiate a preparedness programme by learning the characteristics of ED patients that reflect the spectrum of anticipated emergencies in their patient populations.

Points for discussion: 1) How can Turkish family practitioners (FPs) prepare themselves for out-of-hours emergencies? 2) How can FPs contribute to reduced hospital admission rates?

Presentation 43: Poster, ongoing study, no results yet

Saturday, 10 May 2008, 11.30–13.00

The POPIB study

Jean-Pierre Lebeau, C. Renoux Jr, A. M. Lehr, D. Huas

Department of General Practice, University of Tours, 2 Boulevard Tonnelle, 37000 Tours, France. Tel: [phone]; E-mail: [email protected]

Background: In 2000, one out of ten 10-year-old children suffered from obesity. Obesity prevalence was 4% at 4 years of age, and 12% between 6 and 12 years. Average normal evolution of child corpulence involves an adiposity peak between 6 and 12 months, after which body-mass index (BMI) decreases, until 5 years of age. Adiposity rebound starts at 5 years. Early adiposity rebound is predictive of future obesity. The earlier and the higher the adiposity rebound, the higher the risk of future obesity. A brief intervention targeted at parents about their infant food intakes is likely to modify adiposity rebound, and therefore lead to a reduction in the number of overweight or obese children. POPIB's aim is the design of a tool allowing these brief interventions.

Research question: What are the characteristics and content of a brief intervention concerning food intake targeted at parents of children, newborn to 2 years of age, in general practice?

Method: Three-step qualitative study: 1) literature synthesis and elaboration of a brief intervention tool, using the Delphi method; 2) qualitative study involving general practitioners: validation of content and relevance of the tool, using focus groups; 3) qualitative study involving parents: validation of form and understanding of the intervention, using focus groups.

Conclusion: This qualitative study should lead to a subsequent quantitative study: a prospective cluster-randomized controlled intervention study will be designed to validate intervention effectiveness. Primary endpoint will be the number of early adiposity rebounds 3 years after the intervention.

Points for discussion: 1) Inclusion criteria for the forthcoming quantitative study. 2) What should the parent focus really test?

Presentation 44: Poster

Saturday, 10 May 2008, 11.30–13.00

Smoking habits and knowledge of family physicians and their attitude towards smoking cessation

Joseph Azuri, Shlomit Peled, Eli Kitai, Shlomo Vinker

Department of Family Medicine, Maccabi Healthcare Services, 27 Hamered St., 61000 Tel Aviv, Israel. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Smoking is considered the most important preventable risk factor for morbidity and early mortality. Various methods are proven as aids in smoking cessation, and the primary care physician (PCP) plays an important role. However, the PCP's activities may vary because of different physician and patient characteristics.

Research question: What is the relationship between attitudes towards smoking cessation and physician–patient characteristics in Israel?

Method: A questionnaire was built with case descriptions, knowledge, and attitudes regarding smoking cessation. Data analysis was done according to the PCP characteristics.

Results: 314 PCPs participated in the study; 11.5% reported being smokers, and 58% of them reported their wish to quit; 90% reported advising their patients regarding smoking cessation. Smoking PCPs used significantly less printed materials, whereas ex-smokers reported higher referrals to smoking cessation groups. All PCPs reported greater efforts for adult patients with other risk factors or existing complications. Ex-smokers reported greater efforts to promote smoking prevention among teenagers and pregnant women. The primary complaint of the patient was the most influencing factor in addressing the subject. Although HMO guidelines and maintaining good relationships with patients do not prevent smoking cessation advice, high workload is very influential. Some PCPs supported prescribing smoking cessation drugs beyond their prescribing indications.

Conclusions: Israeli primary care physicians consider smoking cessation as very important in their daily work. Differences between PCPs are common according to their smoking habits and their patients’ characteristics. It is highly important to establish smoking cessation and prevention guidelines in combination with specific training for PCPs, including pharmacologic treatments.

Points for discussion: Are differences in attitudes according to the physician's personal habits more greatly emphasized in smoking cessation counselling than in other health promotion issues?

Presentation 45: Poster, ongoing study with preliminary results

Saturday, 10 May 2008, 11.30–13.00

From research to practice: A model approach to the mushroom poisoning problem

Aysegul Uludag, Ahmet Uludag, A. Alper Cevik, Ilhami Unluoglu

Department of Family Medicine, Eskisehir Osmangazi University, 26480 Eskisehir, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]/[email protected]

Background: Mushroom poisoning is an important problem in Turkey, especially in the spring and autumn. We decided to address the problem by educating people and by investigating the issue in Mihaliccik, a district approximately 100 km from Eskisehir, as a pilot study for the whole region.

Research question: What are the characteristics and traditions of gathering, cooking, and eating wild mushrooms, and what are the attitudes of people when mushroom poisoning occurs? What is the response to education concerning mushroom poisoning?

Method: We performed the questionnaire face to face, and then provided information about wild mushrooms and mushroom poisoning to participants from 10 villages in Mihaliccik in 2005. We asked mushroom poisoning-related questions such as whether participants had eaten wild mushrooms before, and, if so, from where; how did they know the mushrooms were not poisonous; had anyone close to them been poisoned by wild mushrooms; had they received any education/information about wild mushrooms; and was there anything specific they wanted to know.

Results: Initially, 788 participants enrolled in the study and were educated about mushroom poisoning. Six hundred and ninety-eight (88.6%) participants ate mushrooms regularly, of whom 605 (76.8%) had gathered the mushrooms themselves, 48 (6.1%) had got them from their neighbours, and 45 (5.7%) had bought them from the local bazaar. Six hundred and eighty-four (86.8%) participants depended on those who gathered wild mushrooms as a source of whether or not they were poisonous, 396 (50.3%) participants had been witness to a previous mushroom poisoning event, and 112 (14.2%) participants wanted to learn how to select non-poisonous wild mushrooms. The total number of poisoned patients from the Mihaliccik district was 19 in 2003 and 2004, and in 2006–2007 (after the education programme) there were only three events of mushroom poisoning admitted to the university hospital.

Conclusions: The health problems of the community need to be investigated in detail, and education programmes must be planned after this stage.

Points for discussion: 1) What are your suggestions for education and further studies? 2) How can the limitations of such a study in a rural area be ignored/overcome? 3) Similarities and differences for other countries?

Presentation 46: Poster

Saturday, 10 May 2008, 11.30–13.00

General practitioners’ continuous care and home visits to Statutory Health Care patients in Germany, 1996–2006

Johannes Hauswaldt, Eva Hummers-Pradier, Gudrun Theile

Department of General and Family Medicine, Hannover Medical School, Carl-Neuberg-Strasse 1, 30625 Hannover, Germany. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Continuous care and home visits are core activities specific to general and family practices.

Objective: To describe in Germany the workload from continuous care patients and home visits, and its change from 1996 until 2006, and to relate this to patients’ sex, age group, and disease burden.

Method: Electronic patient records from 144 general practices in Lower Saxony and elsewhere in Germany, from 1996 until 2006, serving a total of 331 801 Statutory Health Care (SHC) patients, were analysed per year and per quarter-year cross-sectionally for the number of patients who needed continuous and coordinated care (NoCoPs), at home or in foster homes, and for the number of home visits (NoViPs). Results were then related to total number of patients (NoPs) and number of contacts (NoCs) by calculating rates (NoCoPs/NoPs and NoViPs/NoCs). For longitudinal analysis over the 5-year period beginning 1999 until end of 2003, a complete sample of 16 practices with 66 595 SHC patients was extracted from the above total and analysed accordingly. Patients’ need for continuous care and for home visits was modelled according to sex, age group, and disease burden (ICD-10 codes).

Results: Numbers and rates indicating need for continuous care were higher for patients with female sex, old age, and higher number of diagnoses. Numbers and rates of home visits were higher for teenagers and rapidly increasing for those over 70 years. The insignificant decrease from 1999 until 2003 in patients needing continuous care or home visits vanished behind a large inter-practice variability. Age group and number of diagnoses, but not patient's sex, were significant predictors for need of continuous care or home visits.

Conclusions: Older patients in Germany, and those with high burden of disease, profit most from the specific activities of general and family practitioners, i.e., continuous care and home visits.

Points for discussion: 1) Proportion of continuous care patients and home visits in other European countries. 2) Who cares for the elderly in your country? 3) Foster home and institutionalizing–-an acceptable perspective for European elderly citizens?

Presentation 47: Poster

Saturday, 10 May 2008, 11.30–13.00

Cross-cultural adaptation and validation of the Turkish Patient Satisfaction Questionnaire short form (PSQ 18) in primary care patients

Aysegul Yildirim, Makbule Soyer

Faculty of Health Sciences, Kartal-Cevizli-Istanbul, 34000 Istanbul, Turkey. Tel: [phone]; E-mail: [email protected]

Background: Patient satisfaction in primary care has emerged as an important component of the quality of medical care. Primary healthcare organizations must use patient satisfaction data to improve the quality of their services.

Objective: To adapt and validate the RAND's Patient Satisfaction Questionnaire short form (PSQ-18) into the Turkish languages and culture in primary care practice.

Method: All patients of the study were attending the Maltepe primary healthcare centre outpatient department every month on regular follow-up visits for hypertension therapy. We approached 98 patients during 1 month. Of these patients, 90 fully completed the questionnaire. For reliability analyses, we repeated the same questionnaire 1 month later with the same 90 patients. The PSQ 18's acceptability was high (<8.16% of non-responders). These patients comprised of 52 women and 38 men, their ages ranging from 37 to 58 years (mean 47.59, SD 10.13 years).

Results: Cronbach's alpha for test and re-test 1 month later was found to be 0.87 and 0.89 for reliability, respectively. Construct validity was evaluated by factor loading. The PSQ 18 has 18 items that are collected into seven factors (general, technical quality, interpersonal manner, communication, financial aspects, time spent with provider, and access/convenience) with factor loadings of 0.55–0.89. Revealed loadings of the factors were above 0.55.

Conclusions: The results obtained from the development and validation of the questionnaire provide evidence of its psychometric properties. The short form (PSQ 18) questionnaire showed satisfactory reliability and satisfactory validity. Therefore, it could become a useful instrument in quality-of-care assessment. A questionnaire which addresses primary care satisfaction is now available for research purposes as well as for daily practice.

Points for discussion: Patient satisfaction; primary health care; Short Form Patient Satisfaction Questionnaire, PSQ 18

Presentation 48: Poster

Saturday, 10 May 2008, 11.30–13.00

The incidence and related factors of urinary incontinence in patients attending Ankara University IBN-I Sina Hospital

S. Kocaman, Ayse Gulsen Ceyhun, A. S. Tekiner, Ak F. Parlak

Department of Family Medicine, Ankara University, IBN-I Sina Hastanesi 4.Kat Sihhiye, 06100 Ankara, Turkey. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Although urinary incontinence is not directly related to death, it causes major physical, social, and psychological problems. It is also an important public health problem due to its high incidence.

Research question: What is the incidence of urinary incontinence in patients 50 years and older who attend outpatients clinics of the IBN-I Sina Hospital, and what are the factors related to incontinence?

Method: This cross-sectional study was conducted between January and July 2007 in the outpatient clinics of the IBN-I Sina Hospital. The study population comprised 507 patients aged 50 years and older. We obtained written informed consent and volunteer participation forms from each participant. Questionnaire forms were completed through face-to-face interviews with the study group. SPSS 11.5 was used for statistical analysis. The ethics committee of Ankara University Medical School approved this study.

Results: The study population comprised 142 male and 365 female patients. The incidence of urinary incontinence (UI) for the whole group was 54.8%, whereas 68.6% of them did not seek professional help for this problem. Ninety-five patients did not know the initiating or aggravating factors for their UI, 61 of them considered it natural in older age, 47 of them said that it began with their co-existing illness (DM, HT, etc.). The most frequent reasons for avoiding professional help were considering this condition normal in older age, not feeling discomfort, and reluctance to talk about the subject.

Conclusions: Although urinary incontinence has a high prevalence and is a longstanding condition, we found that patients had little insight to seek help. On the other hand, healthcare professionals did not provide sufficient guidance on this topic. Urinary incontinence must be considered thoroughly in primary care by taking preventive measures in order to meet the requirements of patients.

Points for discussion: 1) How can we raise the awareness of patients to seek help with this condition? 2) How can we imply or create guidelines for urinary incontinence that can be used in outpatient clinics?

Presentation 49: Poster

Saturday, 10 May 2008, 11.30–13.00

Acceptability of a computer-based counselling system for promotion of physical activity in primary care for patients with chronic diseases

Annette Becker, D. Herzberg, N. Marsden, Erika Baum, H. Jung, S. Thomanek, C. Leonhardt

Department of General Practice, Preventive and Rehabilitative Medicine, Philipps University Marburg, Robert-Koch-Str. 5, 35032 Marburg, Germany. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: There is evidence of the effectiveness of physical activity in primary and secondary prevention for patients with chronic diseases. Computer-based counselling systems (CBCS) in healthcare are an important tool in supporting doctors to inform and motivate their patients with respect to therapeutic activity.

Objective: Our aim was to develop a CBCS for promotion of physical activity, and to study its acceptability for primary care patients with diabetes and/or coronary heart disease.

Method: In an interdisciplinary team, we developed an interactive CBCS tailored to the patient's motivational level and disease (transtheoretical model of behaviour change). We used TabletPCs with a touch-sensitive screen. The information was given by a professional speaker, including videos, pictures, and interactive dialogues. A pilot study was performed within five general practices. Seventy-nine patients with diabetes or coronary heart disease tested the CBCS while waiting or during a scheduled home visit. Immediately after the session, semi-structured interviews were held on a subgroup of 26 patients (following the Dynamic Acceptance Model for the Reevaluation of Technologies [DART]). All interviews were subject to content analysis.

Results: The study sample included 34 women and 42 men (mean age 64 years, range 18–87 years). Patients of all ages and with varying computer knowledge reported that the CBCS was easy to use; some commented on the usefulness of the multimedia approach and its advantages over written information. Even though most patients reported the information gained to be useful, opinions differed concerning the applicability of the message to them.

Conclusions: CBCS may be useful in general practice, even in higher age groups and for patients without any prior computer knowledge or experience. However, its effectiveness has to be studied further.

Presentation 50: Poster

Saturday 10^th May, 2008, 11.30–13.00

Administrative and reporting tasks of family physicians in Europe

Imre Rurik, Laszló Kalabay

Department of Family Medicine, Semmelweis University, Kútvölgyi út 4., Budapest, H-1125, Hungary. E-mail: [email protected]/[email protected]

Background: Administrative tasks in primary and further care levels are increasing continuously. Electronic health records and hard copies are stored in parallel. The former are more useful and precise for data management. The administrative and reporting tasks of family physicians are regulated by rules and contracts.

Research question: What are the differences in the administrative tasks of European primary care physicians?

Methods: Family physicians from 23 countries of the European General Practice Research Network were asked to complete a questionnaire. Twenty-one questions were constructed on the main domains of administration and reporting duties for insurance companies or health authorities.

Results: General practitioners (GPs) are considered employees in 13 countries; in the others, they are contracted with 1–250 insurance companies. The activity report should contain: the name of patients (in five countries); coding of identity, insurance, and birth date (in five countries); and referral to specialist or hospitals (in nine countries). Sick leave does not have to be reported in two countries. In six countries, the ICD code should be indicated on the prescription form. Receipts are handwritten or printed in only six countries but in Latin; in the others, native languages are used. An option agreement on the price of drugs is needed in three countries. Free medical systems (without co-payment of the patient) exist in 11 countries. GPs have a financial budget in five countries. The results of their answers regarding the other domains are presented and analyzed, comparing countries and insurance systems.

Conclusions: Different systems exist in Europe. No clear relationships between administrative workload and the number of insurance companies were found. Financial data and epidemiological data are rarely consistent. State-operated primary care systems need less administration. Primary care systems work in very different administrative circumstances. A data recording system serving both purposes should be the ideal solution.

Points for discussion: 1) Are you satisfied with the required amount of administration in your country? 2) What could you suggest or advise in order to improve the situation, besides decreasing administration?

Presentation 51: Freestanding paper

Saturday, 10 May 2008, 16.00–16.30

Is idiopathic facial paralysis an infectious disease?

Karen Barr, Fergus Daly, Frank Sullivan

Department of Community Health Sciences, University of Dundee, Mackenzie Building, DD2 4BF Dundee, United Kingdom. Tel: [phone]; E-mail: [email protected]

Background: Since Will Pickles’ epidemiology in a country practice first postulated an infectious aetiology for Bell's palsy, other observations have been made. In particular, paired serology and studies of the cerebral ganglia have tended to support that hypothesis. One consequence has been the increasing prescription of antiviral agents by general practitioners (GPs).

Research question: Does the spatiotemporal incidence pattern of Bell's palsy in Scotland support or refute an infectious aetiology?

Method: The subjects in this study were participants of the Scottish Bell's Palsy Trial (SBPT). The diagnosis was made by a GP and confirmed by an otorhinolaryngology specialist within 72 hours of onset. Information required for each subject included: date of onset, postcode sector, and age. Postcode sectors were used to protect the anonymity of the subjects. Software that analyzes spatial, temporal, and space-time data using the spatial, temporal, or space-time scan statistics (SaTScan) software was used to analyse the date of onset and postcode sector for 548 cases to determine the presence of geographical clusters. A Poisson model was used for analysis.

Results: There was marked seasonal variation in incidence: 18 cases occurred in June 2005 and 11 cases in June 2006 compared to 35 cases in December 2004 and 31 cases in January 2006. Twelve spatial and 12 spatiotemporal clusters were found, with a p value<0.05. Further analysis was focused upon two major clustering events on the west and east coast. For each of these separate events, the clustering is likely to be part of the same phenomenon. The time period between consecutive cases within a cluster may represent the incubation period of an infectious organism, which may play a role within that cluster.

Conclusions: These results, which found significant spatial and spatiotemporal clusters, are in keeping with previous work supporting an infectious aetiology.

Points for discussion: 1) What other conditions are best studied in GP research networks? 2) Why did the SBPT not demonstrate any benefit from aciclovir?

Presentation 52: Freestanding paper

Saturday, 10 May 2008, 16.30–17.00

Influenza vaccination in Lower Saxony during 1995–1996, 2002–2003, and 2005–2006: A secondary analysis of recorded data from Statutory Health Insurance (SHI) practices

Johannes Hauswaldt

Department of General and Family Medicine, Hannover Medical School, Carl-Neuberg-Strasse 1, 30625 Hannover, Germany. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Seasonal influenza vaccination is highly efficient in primary prevention of additional morbidity or death, and consecutive indirect costs from influenza infection, especially in subgroups of those over 60 years or at risk due to impaired health condition. Comprehensive, reliable, or longitudinal data on the influenza vaccination coverage rate (VCR) in Germany are not available.

Objective: To extract and analyse information about influenza vaccination from routine patient record data of ambulatory practices serving the SHI population of Lower Saxony, Germany

Method: Routine data of the seasons 1995–1996, 2002–2003, and 2005–2006 from two primary sources were analysed: 1) reimbursement data of all ambulatory practices in Lower Saxony, and 2) comprehensive patient records from 79 general practices serving more than 100 000 patients.

Results: Number of influenza vaccinations rose from 204 146 (1995–1996) to 637 016 (2002–2003), and to 1 016 554 (2005–2006), of which 91.8% (2002–2003) and 90.4% (2005–2006) were done by general practitioners and family doctors. Small and medium-sized practices showed higher influenza VCR, compared to those in bigger practices (p<0.01). Being aged over 60 years or being of impaired health condition were reliable predictors for influenza vaccination, with odds ratios of 4.96 (95% confidence interval [CI] 4.73–5.20) and 4.03 (95% CI 4.03–4.58), respectively, in a combined model.

Conclusions: Secondary analysis of routine SHI practice data in Lower Saxony, Germany, from two independent sources shows that vaccination against influenza is given to increasingly more patients, more than 90% by general and family practices. Influenza VCRs in small and medium-sized practices are significantly higher than in bigger practices; VCRs in subgroups aged over 60 years and/or those at risk due to impaired health condition are up to sevenfold higher, compared to the alternative population; VCRs found in Lower Saxony are far below those needed for effective protection of the total population or vulnerable subgroups, positioning this German state at a low rank in international comparison.

Points for discussion: 1) How to increase influenza vaccination rates in Germany: within or outside general and family practices; 2) secondary data analysis in primary care setting: gains and pitfalls

Presentation 53: Freestanding paper

Saturday, 10 May 2008, 17.00–17.30

Effect of C-reactive protein point-of-care testing and clinician communication skills training in lower respiratory tract infections on antibiotic use and patient recovery: a cluster randomized trial

Jochen W. L. Cals, Christopher C. Butler, Rogier M. Hopstaken, Kerenza Hood, Geert-Jan Dinant

Department of General Practice, Maastricht University, PO Box 616, 6200 MD Maastricht, the Netherlands. Tel: [phone]; Fax: [phone]; E-mail: [email protected]

Background: Lower respiratory tract infection (LRTI) accounts for 28 million general practice consultations in the EU and the US annually. Overprescribing antibiotics for LRTI is an important driver of antimicrobial resistance. We aimed to assess the effect of a C-reactive protein (CRP) point-of-care test (a disease-focused approach) and enhanced communication skills training for general practitioners (GPs) (an illness-focused approach) singly and combined on reducing antibiotic prescribing for LRTI.

Research question: Can CRP point-of-care testing and enhanced communication skills training for GPs, either separately or combined, reduce antibiotic prescribing for LRTI without compromising clinical recovery and patient satisfaction?

Method: Pragmatic, 2x2 factorial, cluster randomized controlled trial: 40 GPs recruited 431 LRTI patients. The main outcome measure was antibiotic prescribing at the index consultation. The primary analysis was intention to treat and assessed the predefined marginal effects of the two interventions in a three-level logistic regression model.

Results: GPs in the CRP group prescribed antibiotics to 31% of patients versus 53% in the control group (adjusted odds ratio [OR] 0.21, 95% confidence interval [CI] 0.06–0.78, p=0.02). GPs trained in enhanced communication skills prescribed antibiotics to 27.4% of patients versus 53.5% (adjusted OR 0.12, 95% CI 0.03–0.47, p<0.01). There was a statistically significant effect of both interventions on antibiotic prescribing at any point within 28-day follow-up. Clinicians allocated to both interventions prescribed antibiotics to 23% of included patients (adjusted OR 0.05, 95% CI 0.01–0.21). Patient recovery and satisfaction was similar in all study groups.

Conclusions: Both CRP point-of-care testing and enhanced communication skills training significantly reduced antibiotic prescribing for LRTI without compromising patient recovery and satisfaction. A combined illness- and disease-focused approach was superior to either approach alone in achieving evidence-based management of this common condition in general practice.

Points for discussion: 1) If we had to choose between the two interventions (illness or disease focused) for implementation on a larger scale, which would be preferred? 2) Could the interventions have implications for other conditions in general practice?