Beyond accuracy: hidden motives in diagnostic testing

BACKGROUND

Diagnostic decision-making is usually disease-focussed and intended to examine the patient's medical condition accurately. But diagnostic interventions may serve further purposes that are not yet fully understood.

OBJECTIVE

To explore GPs' diagnostic behaviour not related to confirming or refuting a specific disease.

METHODS

We recorded 295 primary care consultations in 12 practices. One hundred thirty-four consultations comprised at least one diagnostic episode. GPs were asked to reflect on their own diagnostic thinking in interviews for every single case. Qualitative and quantitative analyses were applied with focus on the GPs' cognitive processes during diagnostic decision-making.

RESULTS

Primary care physicians clearly stated that they requested some tests for other reasons than diagnosing disease. A feeling of uncertainty stimulated diagnostic procedures aiming to regulate the anticipation of regret. We identified patients' reassurance, patients' requests and strategic issues as further motives for diagnostic actions.

CONCLUSION

Besides focussing on disease in the diagnostic process, emotional and strategic goals are hidden motives that play a critical role in clinical decision-making. They might even represent an initial factor in a cascade of interventions leading to overdiagnosis. How GPs might control these influences provides an important aspect for further research, practice and teaching.

Factors associated with prescribing costs: analysis of a nationwide administrative database

Objective

All health care systems in the world struggle with rising costs for drugs. We sought to explore factors impacting on prescribing costs in a nationwide database of ambulatory care in Germany. Factors identified by this research can be used for adjustment in future profiling efforts.

Methods

We analysed nationwide prescription data of physicians having contractual relationships with statutory health insurance funds in 2014. Predictor and outcome variables were aggregated at the practice level. We performed analyses separately for primary care and specialties of cardiology, gastroenterology, neurology and psychiatry, pulmology as well as oncology and haematology. Bivariate robust regressions and Spearman rank correlations were computed in order to find meaningful predictors for our outcome variable prescription costs per patient.

Results

Median age of patients and proportion of DDD issued were substantial predictors for prescription costs per patient in Primary Care, Cardiology, and Pulmology with explained variances between 41 and 61%. In Neurology and Psychiatry only proportion of patients with polypharmacy ≥ 2 quarters was a significant predictor for prescription costs per patient, explaining 20% of the variance. For gastroenterologists, oncologists and haematologists no stable models could be established.

Conclusions

Any analysis of prescribing behaviour must take the degree into account to which an individual physician or practice is responsible for prescribing patients' medication. Proportion of prescriptions/DDDs is an essential confounder for future studies of drug prescribing.

[Deliberations of the expert advisory council on innovation fund applications]

The "Innovation Fund" provides incentives for the development and testing of healthcare innovations in the area of the statutory health insurance with the aim to improve the quality of care in Germany. Over a period of initially four years (2016-19), 300 million Euro will be allocated annually to projects on "innovative forms of healthcare provision" and "health services research". Using a formalized procedure, the ten-member expert advisory board appointed by the German Federal Ministry of Health (BMG) assess all applications on the basis of various criteria for scientific quality, potential of innovation, relevance for health service delivery, and implementability. The present discussion paper sets out important considerations for submission and assessment and puts them up for discussion. (As supplied by the authors).

The long way of implementing patient-centered care and shared decision making in Germany

The main focus of the paper is on the description of the development and current state of research and implementation of patient-centered care (PCC) and shared decision making (SDM) after fifteen years of substantial advances in health policy and health services research. What is the current state of SDM in health policy? The "Patients' Rights Act" from 2013 standardizes all rights and responsibilities within the framework of medical treatment for German citizens and legal residents. This comprises the right to informed decisions, comprehensive and comprehensible information for patients, and decisions based on a clinician-patient-partnership. What is the current state of SDM interventions and patient decision support tools? SDM training programs for healthcare professionals have been developed. Their implementation in medical schools has been successful. Several decision support tools - primarily with support from health insurance funds and other public agencies - are to be implemented in routine care, specifically for national cancer screening programs. What is the current state of research and routine implementation? The German government and other public institutions are constantly funding research programs in which patient-centered care and shared decision-making are important topics. The development and implementation of decision tools for patients and professionals as well as the implementation of CME trainings for healthcare professionals require future efforts. What does the future look like? With the support of health policy and scientific evidence, transfer of PCC and SDM to practice is regarded as meaningful. Research can help to assess barriers, facilitators, and needs, and subsequently to develop and evaluate corresponding strategies to successfully implement PCC and SDM in routine care, which remains challenging.

Individual utilisation thresholds and exploring how GPs' knowledge of their patients affects diagnosis: a qualitative study in primary care

BACKGROUND

One of the tenets of general practice is that continuity of care has a beneficial effect on patient care. However, little is known about how continuity can have an impact on the diagnostic reasoning of GPs.

AIM

To explore GPs' diagnostic strategies by examining GPs' reflections on their patients' individual thresholds for seeking medical attention, how they arrive at their estimations, and which conclusions they draw.

DESIGN AND SETTING

Qualitative study with 12 GPs in urban and rural practices in Germany.

METHOD

After each patient consultation GPs were asked to reflect on their diagnostic reasoning for that particular case. Qualitative and quantitative analyses of consultations and interview content were undertaken.

RESULTS

A total of 295 primary care consultations were recorded, 134 of which contained at least one diagnostic episode. When elaborating on known patients, GPs frequently commented on how 'early' or 'late' in an illness progression a patient tended to consult. The probability of serious disease was accordingly regarded as high or low. This influenced GPs' behaviour regarding further investigations or referrals, as well as reassurance and watchful waiting. GPs' explanations for a patient's utilisation threshold comprised medical history, the patient's characteristics, family background, the media, and external circumstances.

CONCLUSION

The concept of an individual threshold for the utilisation of primary care would explain how GPs use their knowledge of individual patients and their previous help-seeking behaviour for their diagnostic decision making. Whether the assumption behind this concept is valid, and whether its use improves diagnostic accuracy, remains to be investigated.

Time-to-event versus ten-year-absolute-risk in cardiovascular risk prevention - does it make a difference? Results from the Optimizing-Risk-Communication (OptRisk) randomized-controlled trial

Background

The concept of shared-decision-making is a well-established approach to increase the participation of patients in medical decisions. Using lifetime risk or time-to-event (TTE) formats has been increasingly suggested as they might have advantages, e.g. in younger patients, to better show consequences of unhealthy behaviour. In this study, the most-popular ten-year risk illustration in the decision-aid-software arriba^TM (emoticons), is compared within a randomised trial to a new-developed TTE illustration, which is based on a Markov model.

Methods

Thirty-two General Practitioners (GPs) took part in the study. A total of 304 patients were recruited and counseled by their GPs with arriba^TM, and randomized to either the emoticons or the TTE illustration, followed by a patient questionnaire to figure out the degree of shared-decision-making (PEF-FB9, German questionnaire to measure the participation in the shared decision-making process, primary outcome), as well as the decisional conflict, perceived risk, accessibility and the degree of information, which are all secondary outcomes.

Results

Regarding our primary outcome PEF-FB9 the new TTE illustration is not inferior compared to the well-established emoticons taking the whole study population into account. Furthermore, the non-inferiority of the innovative TTE could be confirmed for all secondary outcome variables. The explorative analysis indicates even advantages in younger patients (below 46 years of age).

Conclusion

The TTE format seems to be as useful as the well-established emoticons. For certain patient populations, especially younger patients, the TTE may be even superior to demonstrate a cardiovascular risk at early stages. Our results suggest that time-to-event illustrations should be considered for current decision support tools covering cardiovascular prevention.

Trial registration

The study was registered at the German Clinical Trials Register and at the WHO International Clinical Trials Register Platform (ICTRP, ID DRKS00004933); registered 2 February 2016 (retrospectively registered).

Electronic supplementary material

The online version of this article (doi:10.1186/s12911-016-0393-1) contains supplementary material, which is available to authorized users.

When Decisions Should Be Shared

Background. Shared decision making (SDM) is often advocated as an ideal for making medical decisions. Until now, however, opinions regarding which treatment situations warrant SDM have not been systematically investigated. The purpose of this study was to examine social norms regarding medical decision making, using a factorial survey design. Methods. The factorial survey applied in this study consisted of 7 situational factors (e.g., the reason for consultation), each with 2 to 3 levels (e.g., prevention and severe disease). These factors were turned into various descriptions of treatment situations. A total of 101 physicians, 115 patients, and 113 members of self-help groups participated in the study. Each participant assessed 10 vignettes using a 5-point scale to indicate who they thought should make the decision in each specific situation. Results. Most assessments across the 3 groups called for a shared decision (39%). Ordered logistic regression analysis demonstrated that, according to study participants, all 7 situational factors (reason for consultation, time frame of negative outcomes, time pressure, number of therapeutic options, side effects, scientific evidence of efficacy, and desire to participate) significantly affected how decisions regarding treatment should be made. The strongest factor was the patient’s desire to participate in decision making (odds ratio = 1.84; P ≤ 0.001), followed by the reason for consultation (odds ratio = 0.69; P ≤ 0.001). Conclusions. This study reveals that there is a general desire for SDM in a variety of treatment situations. Furthermore, based on the responses of our participants, our findings also lay the framework in determining which treatment situations warrant SDM.

The differential diagnosis of tiredness: a systematic review

Background

Tiredness is one of the most frequent complaints in primary care. Although often self-limiting and frequently associated with psychosocial stress, patients but also their physicians are often uncertain regarding a serious cause and appropriate diagnostic work-up. We conducted a systematic review and meta-analysis of studies reporting on differential diagnosis of fatigue in primary care.

Methods

MEDLINE, EMBASE and conference abstracts were searched for primary care based studies of patients presenting with tiredness. Twenty-six studies were included. We report on anaemia, malignancy, serious organic disease, depression and the chronic fatigue syndrome (CFS) as causes of tiredness as presenting complaint.

Results

We found considerable heterogeneity of estimates which was reduced by limiting our analysis to high quality studies. Prevalences were as follows-anaemia: 2.8 % (CI (confidence interval) 1.6–4.8 %); malignancy: 0.6 % (CI 0.3–1.3 %); serious somatic disease: 4.3 % (CI 2.7–6.7 %); depression 18.5 % (CI 16.2–21.0 %). Pooling was not appropriate for CFS.

In studies with control groups of patients without the symptom of tiredness, prevalence of somatic disease was identical to those complaining of tiredness. Depression, however, was more frequent among those with tiredness.

Conclusions

Serious somatic disease is rare in patients complaining of tiredness. Since prevalence is similar in patients without tiredness, the association may not be causal. Extensive investigations are only warranted in case of specific findings from the history or clinical examination. Instead, attention should focus on depression and psychosocial problems.

Electronic supplementary material

The online version of this article (doi:10.1186/s12875-016-0545-5) contains supplementary material, which is available to authorized users.

Pooled individual patient data from five countries were used to derive a clinical prediction rule for coronary artery disease in primary care

OBJECTIVE

To construct a clinical prediction rule for coronary artery disease (CAD) presenting with chest pain in primary care.

STUDY DESIGN AND SETTING

Meta-Analysis using 3,099 patients from five studies. To identify candidate predictors, we used random forest trees, multiple imputation of missing values, and logistic regression within individual studies. To generate a prediction rule on the pooled data, we applied a regression model that took account of the differing standard data sets collected by the five studies.

RESULTS

The most parsimonious rule included six equally weighted predictors: age ≥55 (males) or ≥65 (females) (+1); attending physician suspected a serious diagnosis (+1); history of CAD (+1); pain brought on by exertion (+1); pain feels like "pressure" (+1); pain reproducible by palpation (-1). CAD was considered absent if the prediction score is <2. The area under the ROC curve was 0.84. We applied this rule to a study setting with a CAD prevalence of 13.2% using a prediction score cutoff of <2 (i.e., -1, 0, or +1). When the score was <2, the probability of CAD was 2.1% (95% CI: 1.1-3.9%); when the score was ≥ 2, it was 43.0% (95% CI: 35.8-50.4%).

CONCLUSIONS

Clinical prediction rules are a key strategy for individualizing care. Large data sets based on electronic health records from diverse sites create opportunities for improving their internal and external validity. Our patient-level meta-analysis from five primary care sites should improve external validity. Our strategy for addressing site-to-site systematic variation in missing data should improve internal validity. Using principles derived from decision theory, we also discuss the problem of setting the cutoff prediction score for taking action.

A simple heuristic for Internet-based evidence search in primary care: a randomized controlled trial

BACKGROUND

General practitioners (GPs) are confronted with a wide variety of clinical questions, many of which remain unanswered.

METHODS

In order to assist GPs in finding quick, evidence-based answers, we developed a learning program (LP) with a short interactive workshop based on a simple three-step-heuristic to improve their search and appraisal competence (SAC). We evaluated the LP effectiveness with a randomized controlled trial (RCT). Participants (intervention group [IG] n=20; control group [CG] n=31) rated acceptance and satisfaction and also answered 39 knowledge questions to assess their SAC. We controlled for previous knowledge in content areas covered by the test.

RESULTS

Main outcome - SAC: within both groups, the pre-post test shows significant (P=0.00) improvements in correctness (IG 15% vs CG 11%) and confidence (32% vs 26%) to find evidence-based answers. However, the SAC difference was not significant in the RCT.

OTHER MEASURES

Most workshop participants rated "learning atmosphere" (90%), "skills acquired" (90%), and "relevancy to my practice" (86%) as good or very good. The LP-recommendations were implemented by 67% of the IG, whereas 15% of the CG already conformed to LP recommendations spontaneously (odds ratio 9.6, P=0.00). After literature search, the IG showed a (not significantly) higher satisfaction regarding "time spent" (IG 80% vs CG 65%), "quality of information" (65% vs 54%), and "amount of information" (53% vs 47%).

CONCLUSION

Long-standing established GPs have a good SAC. Despite high acceptance, strong learning effects, positive search experience, and significant increase of SAC in the pre-post test, the RCT of our LP showed no significant difference in SAC between IG and CG. However, we suggest that our simple decision heuristic merits further investigation.

The Phenomenology of the Diagnostic Process

Background. While dichotomous tasks and related cognitive strategies have been extensively researched in cognitive psychology, little is known about how primary care practitioners (general practitioners [GPs]) approach ill-defined or polychotomous tasks and how valid or useful their strategies are. Objective. To investigate cognitive strategies used by GPs for making a diagnosis. Methods. In a cross-sectional study, we videotaped 282 consultations, irrespective of presenting complaint or final diagnosis. Reflective interviews were performed with GPs after each consultation. Recordings of consultations and GP interviews were transcribed verbatim and analyzed using a coding system that was based on published literature and systematically checked for reliability. Results. In total, 134 consultations included 163 diagnostic episodes. Inductive foraging (i.e., the initial, patient-guided search) could be identified in 91% of consultations. It contributed an average 31% of cues obtained by the GP in 1 consultation. Triggered routines and descriptive questions occurred in 38% and 84% of consultations, respectively. GPs resorted to hypothesis testing, the hallmark of the hypothetico-deductive method, in only 39% of consultations. Limitations. Video recordings and interviews presumably interfered with GPs’ behavior and accounts. GPs might have pursued more hypotheses and collected more information than usual. Conclusions. The testing of specific disease hypotheses seems to play a lesser role than previously thought. Our data from real consultations suggest that GPs organize their search for information in a skillfully adapted way. Inductive foraging, triggered routines, descriptive questions, and hypotheses testing are essential building blocks to make a diagnosis in the generalist setting.

Influence of thyroid hormone therapy on the fracture rate - A claims data cohort study

INTRODUCTION

It has been debated for years whether long-term thyroid hormone intake causes fractures. Not only have previous studies suffered from design limitations, they also reached contradictory conclusions. We investigated thyroid hormones (thyroxine) as a possible risk factor for fractures in a cohort of 6.7 million persons based on administrative data.

METHODS

The database consists of anonymized settlement data of approximately 70 German statutory health insurances covering a time period of six years. All subjects aged 60 and above were included in the study; subjects with repeated thyroxine prescriptions were assigned to the exposure group; members without thyroxine prescriptions to the control group. Outcome was any incident fracture during a declared time period. In order to calculate fracture risk, we performed multivariate cox regression analyses to adjust for confounders.

RESULTS

Of 798 770 subjects fulfilling the inclusion criteria, 11.7% took thyroxine regularly and belong to the exposure group. The final cox regression showed that subjects taking thyroxine have a 6.3% higher risk (HR 1.063; CI 1.046-1.080, p=<.0001) than members of the control group.

DISCUSSION

The study supports the assumption that long term thyroxine intake leads to an increase in fracture risk among patients older than 60years. The findings have implications for long term thyroxine treatment.

Causes of chest pain in primary care--a systematic review and meta-analysis

Aim

To investigate the frequencies of different and relevant underlying etiologies of chest pain in general practice.

Methods

We systematically searched PubMed and EMBASE. Two reviewers independently rated the eligibility of publications and assessed the risk of bias of included studies. We extracted data to calculate the relative frequencies of different underlying conditions and investigated the variation across studies using forest plots, I², tau², and prediction intervals. With respect to unexplained heterogeneity, we provided qualitative syntheses instead of pooled estimates.

Results

We identified 11 eligible studies comprising about 6500 patients. The overall risk of bias was rated as low in 6 studies comprising about 3900 patients. The relative frequencies of different conditions as the underlying etiologies of chest pain reported by these studies ranged from 24.5 to 49.8% (chest wall syndrome), 13.8 to 16.1% (cardiovascular diseases), 6.6 to 11.2% (stable coronary heart disease), 1.5 to 3.6% (acute coronary syndrome/myocardial infarction), 10.3 to 18.2% (respiratory diseases), 9.5 to 18.2% (psychogenic etiologies), 5.6 to 9.7% (gastrointestinal disorders), and 6.0 to 7.1% (esophageal disorders).

Conclusion

This information may be of practical value for general practitioners as it provides the pre-test probabilities for a range of underlying diseases and may be suitable to guide the diagnostic process.

The Interdisciplinary Management of Acute Chest Pain

BACKGROUND

Acute chest pain of non-traumatic origin is a common reason for presentation to physician's offices and emergency rooms. Coronary heart disease is the cause in up to 25% of cases. Because acute chest pain, depending on its etiology, may be associated with a high risk of death, rapid, goal-oriented management is mandatory.

METHODS

This review is based on pertinent articles and guidelines retrieved by a selective search in PubMed.

RESULTS

History-taking, physical examination, and a 12-lead electrocardiogram (ECG) are the first steps in the differential diagnostic process and generally allow the identification of features signifying a high risk of lifethreatening illness. If the ECG reveals ST-segment elevation, cardiac catheterization is indicated. The timedependent measurement of highly sensitive troponin values is a reliable test for the diagnosis or exclusion of acute myocardial infarction. A wide variety of other potential causes (e.g., vascular, musculoskeletal, gastroenterologic, or psychosomatic) must be identified from the history if they are to be treated appropriately. Elderly patients need special attention.

CONCLUSION

Acute chest pain is a major diagnostic challenge for the physician. Common errors are traceable to non-recognition of important causes and to an inadequate diagnostic work-up. Future studies should be designed to help optimize the interdisciplinary management of patients with chest pain.

The experiences of Russian-speaking migrants in primary care consultations

BACKGROUND

Some three million Russian-speaking immigrants from the former Soviet Union live in Germany today. Many of them underwent a different kind of medical socialization than the indigenous population, but the experiences and expectations of this group of patients have hardly been studied to date.

METHODS

In a qualitative study, 24 chronically ill native Germans and 25 chronically ill Russian-speaking immigrants were recruited via notices, through their primary care physicians, and by word of mouth and underwent a semistructured interview in their mother tongue (German or Russian) about their experiences with their primary care physicians. The interviews were recorded using an audio device, translated into German if necessary, and transcribed, and their content was analyzed with the MAXQDA software package.

RESULTS

The immigrants were less satisfied with their primary care physicians than the native Germans. This manifested itself in a weaker patient-physician connection and frequent changes of physician due to dissatisfaction with treatment. Both groups considered themselves inadequately informed about matters of health, but they gave differing reasons for this. On the other hand, the participants in both groups had practically the same general expectations from their primary care physicians. However, detailed analysis revealed cultural differences.

CONCLUSION

Physicians in Germany should be more aware of the culturally based expectations of immigrant patients in order to understand their needs better, improve the physician-patient relationship, and ensure equal opportuities in health care. For example, many immigrants would prefer their doctors to communicate with them in a manner that non-immigrants would consider paternalistic.

Studies of the symptom abdominal pain--a systematic review and meta-analysis

BACKGROUND

Diagnostic reasoning in primary care patients with abdominal pain is a complex challenge for GPs. To ensure evidence-based decision making for this symptom, GPs need setting-specific knowledge about the prevalence, potential risks for diseases and chance of recovery or risk of undesirable courses of disease.

AIM

We conducted a systematic review of symptom-evaluating studies on prevalence, aetiology or prognosis of abdominal pain.

METHODS

We included all studies evaluating the symptom 'abdominal pain' as a reason for consultation in primary care. We included all types of study designs except for qualitative studies. Studies focussed solely on children or settings other than primary care were also excluded.

RESULTS

We identified 14 studies. Mean consultation prevalence is 2.8% for abdominal pain. In about one-third of patients the underlying cause of abdominal pain cannot be specified. The most common aetiologies are gastroenteritis (7.2-18.7%), irritable bowel disease (2.6-13.2%), urological cause (5.3%) and gastritis (5.2%). About 1 in 10 abdominal pain patients suffers from an acute disease like appendicitis (1.9%), diverticulitis (3.0%), biliary/pancreatic (4.0%) or neoplastic (1.0%) diseases needing immediate therapy.

CONCLUSION

There is a high prevalence of patients consulting GPs for abdominal pain. The review identified a comparably high rate of acute underlying diseases in need of further investigation or therapy. At the same time, the underlying cause of the complaints often remains unexplained. Further symptom-evaluating studies are necessary, ideally using standardized methodology in order to gain sufficient evidence for developing much-needed guidelines and decision support tools.

Interventions for improving the adoption of shared decision making by healthcare professionals

BACKGROUND

Shared decision making (SDM) can reduce overuse of options not associated with benefits for all and respects patient rights, but has not yet been widely adopted in practice.

OBJECTIVES

To determine the effectiveness of interventions to improve healthcare professionals' adoption of SDM.

SEARCH METHODS

For this update we searched for primary studies in The Cochrane Library, MEDLINE, EMBASE, CINAHL, the Cochrane Effective Practice and Organisation of Care (EPOC) Specialsied Register and PsycINFO for the period March 2009 to August 2012. We searched the Clinical Trials.gov registry and the proceedings of the International Shared Decision Making Conference. We scanned the bibliographies of relevant papers and studies. We contacted experts in the field to identify papers published after August 2012.

SELECTION CRITERIA

Randomised and non-randomised controlled trials, controlled before-and-after studies and interrupted time series studies evaluating interventions to improve healthcare professionals' adoption of SDM where the primary outcomes were evaluated using observer-based outcome measures (OBOM) or patient-reported outcome measures (PROM).

DATA COLLECTION AND ANALYSIS

The three overall categories of intervention were: interventions targeting patients, interventions targeting healthcare professionals, and interventions targeting both. Studies in each category were compared to studies in the same category, to studies in the other two categories, and to usual care, resulting in nine comparison groups. Statistical analysis considered categorical and continuous primary outcomes separately. We calculated the median of the standardized mean difference (SMD), or risk difference, and range of effect across studies and categories of intervention. We assessed risk of bias.

MAIN RESULTS

Thirty-nine studies were included, 38 randomised and one non-randomised controlled trial. Categorical measures did not show any effect for any of the interventions. In OBOM studies, interventions targeting both patients and healthcare professionals had a positive effect compared to usual care (SMD of 2.83) and compared to interventions targeting patients alone (SMD of 1.42). Studies comparing interventions targeting patients with other interventions targeting patients had a positive effect, as did studies comparing interventions targeting healthcare professionals with usual care (SDM of 1.13 and 1.08 respectively). In PROM studies, only three comparisons showed any effect, patient compared to usual care (SMD of 0.21), patient compared to another patient (SDM of 0.29) and healthcare professional compared to another healthcare professional (SDM of 0.20). For all comparisons, interpretation of the results needs to consider the small number of studies, the heterogeneity, and some methodological issues. Overall quality of the evidence for the outcomes, assessed with the GRADE tool, ranged from low to very low.

AUTHORS' CONCLUSIONS

It is uncertain whether interventions to improve adoption of SDM are effective given the low quality of the evidence. However, any intervention that actively targets patients, healthcare professionals, or both, is better than none. Also, interventions targeting patients and healthcare professionals together show more promise than those targeting only one or the other.

Development and validation of a generic questionnaire for the implementation of complex medical interventions

Introduction: The implementation of complex medical interventions in daily practice is often fraught with difficulties. According to the iterative phase model proposed by the British Medical Research Council (MRC), the development, implementation and evaluation of complex interventions should be theory-driven. A conceptual model that seems to be a promising framework is the Theory of planned behaviour (TPB). In our study we aimed to develop and validate a generic and multifaceted questionnaire based on the TPB to detect physicians’ willingness to implement complex medical interventions and the factors influencing this willingness.

Methods: The questionnaire was developed according to the literature and was informed by previous qualitative research of our department. It was validated on the example of an electronic library of decision aids, arriba-lib. The sample consisted of 181 General Practitioners (GPs) who received a training regarding arriba-lib and subsequently filled in the questionnaire, assessing the TPB variables attitude, subjective norm, perceived behaviour control and intention. Follow-up assessments were conducted after two (assessing retest reliability) and eight weeks (assessing target behaviour). We performed a confirmatory factor analysis investigating the factorial structure of our questionnaire according to the TPB. Beside the calculation of the questionnaire’s psychometric properties we conducted a structural equation model and an ordinal regression to predict actual behaviour regarding the installation and application of arriba-lib.

Results: The postulated three factorial model (attitude, subjective norm, perceived behaviour control) of our questionnaire based on the TPB was rejected. A two factorial model with a combined factor subjective norm/perceived behaviour control was accepted. The explained variance in the ordinal regression was low (Nagelkerke’s R²=.12). Neither attitude nor intention were able to predict the use or non-use of arriba-lib (attitude: p=.68, intention: p=.44). For the combined factor subjective norm/perceived behaviour control a significant, but small effect (p=.03) was shown.

Conclusions: The TPB is not an adequate theoretical framework to guide the development of a generic questionnaire in the context of the implementation of complex interventions. To enable the successful implementation of complex medical interventions evaluators have to go through the whole development and evaluation process according to the MRC-model, without short cuts. Further, it has to be discussed if a generic instrument can be valid and useful. Regarding the TPB a publication bias regarding the theory’s applicability might have to be considered.