University of Southern Denmark Characteristics of complaints resulting in disciplinary actions against Danish GPs

Objective . The risk of being disciplined in connection with a complaint case causes distress to most general practitioners. The present study examined the characteristics of complaint cases resulting in disciplinary action. Material and methods. The Danish Patients ’ Complaints Board ’ s decisions concerning general practice in 2007 were examined. Information on the motives for complaining, as well as patient and general practitioner characteristics, was extracted and the association with case outcome (disciplinary or no disciplinary action) was analysed. Variables included complaint motives, patient gender and age, urgency of illness, cancer diagnosis, healthcare settings (daytime or out-of-hours services), and general practitioner gender and professional seniority. Results. Cases where the complaint motives involved a wish for placement of responsibility (OR 2.35, p 0.01) or a wish for a review of the general practitioner ’ s competence (OR 1.95, p 0.02) were associated with increased odds of the general practitioner being disciplined. The odds of discipline decreased when the complaint was motivated by a feeling of being devalued (OR 0.39, p 0.02) or a request for an explanation (OR 0.46, p 0.01). With regard to patient and general practitioner characteristics, higher general practitioner professional seniority was associated with increased odds of discipline (OR 1.97 per 20 additional years of professional seniority, p 0.01). None of the other characteristics was statistically signifi cantly associated with discipline in the multiple logistic regression model. Conclusion. Complaint motives and professional seniority were associated with decision outcomes. Further research is needed on the impact of professional seniority on performance.


Introduction
The risk of receiving a patient complaint case seriously impacts on the work of medical doctors [1]. Professional self-esteem comes into play and especially in general practice the continuous patientdoctor relationship is at stake. For those general practitioners (GPs) receiving a complaint, the risk of being disciplined becomes a major concern. The characteristics of the complaints most likely to result in disciplinary action have received limited attention in the research literature. A Norwegian study suggested that male GPs and male patients are associated with complaint cases resulting in discipline. In that small study, 55 of the 108 cases (51%) concerned the out-of-hours service [2]. No larger studies Sanctions may be a major concern for general practitioners involved in a patient complaint case, but little is known about the characteristics of cases leading to criticism.
Complaint cases motivated by a wish for • placement of responsibility or a wish for review of the general practitioner ' s competence were associated with increased odds of the general practitioner being disciplined. When the patient ' s feeling of being devalued • or a request for an explanation motivated the complaint, odds of discipline decreased. High professional seniority of the general • practitioner was associated with increased odds of discipline.
suggested to play an important role in malpractice litigations [5,6]. This study aimed to analyse what characteristics (complaint motives, patients and GPs) were associated with being disciplined in connection with complaint cases against GPs.

Material and methods
Setting: Primary health care in Denmark and the disciplinary system In 2006, more than 99% of the Danish population was listed with one of 3765 GPs working in approximately 2200 single-handed or partnership practices. Danish general practice is based on a contract with the tax-fi nanced Danish National Health Insurance. The GPs act as gatekeepers with regard to the secondary healthcare system. Patients choose their GP and it is possible to change GP according to preferences (for a small fee). GPs are responsible for the care at all hours and GPs in a region collaborate in an organized out-of-hours service.
Patients who are dissatisfi ed with their GP may decide to fi le a written complaint. As in other countries, a national disciplinary system has been established to handle complaints about authorized healthcare professionals, including GPs. The board makes judgements only about professional conduct. Complaints not disputing professional conduct but only expressing dissatisfaction with the level of service (e.g. long waiting time or bad manners) are directed to the healthcare providers. Any compensation claims included in a complaint are handled in a separate Patient Insurance, while the Complaints Board makes decision about professional conduct. Alternatively, patients may fi le a complaint with the ' Patientombuddet ' system, which handles complaints about courses of health care without intending named health professionals to be disciplined. Comparable complaints-handling authorities have been established in other countries [2,7]. At the initial stage, the board ' s secretariat clarifi es the issues of the complaint. In this process the complainant and the defendant health professional will be heard. Like in other countries, the decision may be based on evaluations made by appointed experts. The fi nal decision, however, is made by a fi ve-person committee consisting of two public representatives, two representatives of the health profession concerned (e.g. GP specialists), and a chairperson who is a judge. The board has the authority to impose sanctions in the form of a discipline, the most commonly used being " criticism " oruntil 1 January 2011 -disputing professional conduct (a milder comment). The other possible sanctions are " discipline with injunction " , or bringing the health professional before the prosecuting authority.

Methods
All complaint case decisions concerning GPs completed in 2007 were reviewed. Information was extracted from case fi les (including the letter of complaint and all documents gathered in the handling process) and registered in a structured database.
The evaluation of motives behind the complaint was based on a review of the complaint letter. Based on the model described by Bismark and colleagues [8,9], the complainant motives were categorized in accordance with the patients ' expressed wish for: explanation , placement of responsibility; quality improvement for future patients , review of the GP ' s competence ; economic compensation, better level of general service ; professional discipline ; and other sanction . A complaint may have more than one motive [8,9] and motives were, therefore, treated as eight separate variables in the statistical analysis. According to Bismark et al., the above eight motives cover the following four categories: Communication, correction, restoration, and sanction (see Table I). Additionally, it was noted whether the complaint was due to feeling devalued by the GP. Information was gathered on patient gender , patient age , and the illness concerned. With regard to patient illnesses, ICPC-2 coding was used. A cancer variable was constructed. Also, based on the ICPC-2 codes, a serious urgent illness variable was constructed after consensus between the authors (SB, JK, and ND). Deciding what illnesses to categorize as a serious urgent illness might imply diffi culties. We chose only to consider diagnoses commonly resulting in death if untreated as serious urgent illness . Additionally, it was registered if the patient concerned, according to case management documentation, had died ( death of patient ). Other independent variables considered to be possible factors associated with discipline or potential confounders were healthcare settings ( daytime care or out-of-hours ), general practitioner gender , and professional seniority (years from graduation until event concerned). Information on GP professional seniority was gathered through manual look-up in a publicly available list covering Danish medical doctors [10].
Finally, information on the decision outcome (discipline or no discipline) was noted. To analyse characteristics associated with discipline (the dependent variable), odds ratios (ORs) were estimated by means of a multiple logistic regression model including all the other mentioned variables. In some complaint cases decisions on professional conduct were made about two or more GPs. The unit used for the statistical analysis was the decision about individual GPs. All analyses were performed using STATA ® , release 11.1 (StataCorp, College Station, TX, USA). P-values Ͻ 0.05 were considered statistically signifi cant.

Results
In 2007, the Danish Patient Complaints Board completed 571 decisions against individual GPs. Sample characteristics are given in Table I. In 22% of decisions, the GP was disciplined. Criticism was expressed in 96 decisions (17%) and the professional conduct was disputed in 30 decisions (5%). The 96 decisions resulting in the GP being criticized included eight GPs being disciplined with injunction. One of these GPs was brought before the prosecuting authority, but the charge was later dropped. The average patient age was 45.3 years (range 0 -91 years) and the average professional seniority of GPs was 22.2 years (range 0 -47 years). The motives for complaining most often involved the categories of Communication and Correction , whilst the Sanction motive was encountered much less often. Table II presents the analysis of variables predictive of discipline. One case was omitted from the analysis because patient age was unknown.
When including variables concerning complaint motives, patient gender and patient age , patient serious urgent illness , cancer , death of patient , healthcare settings ( daytime care or out-of-hours ), and general practitioner gender and professional seniority in a multiple logistic regression model, odds of discipline were halved when the complaint was motivated by feeling devalued (OR ϭ 0.39, p ϭ 0.02) or a request for an explanation (OR ϭ 0.46, p ϭ 0.01). However, when complaints involved a wish for placement of responsibility (OR ϭ 2.35, p ϭ 0.01) or a request for a review of the GP ' s competence (OR ϭ 1.95, p ϭ 0.02) the odds of discipline doubled, just as with professional seniority : a GP with 20 years more seniority had doubled odds of being disciplined in connection with a complaint case (OR ϭ 1.97, p ϭ 0.01).

Discussion
The key fi ndings of this study are decreased odds of being disciplined when the complaint was motivated by feeling devalued or a request for an explanation . Increased odds of discipline were observed in complaint cases where the complaint was based on a wish for placement of responsibility or a wish for a review of the GP ' s competence . In addition, higher professional seniority was associated with increased odds of discipline. No statistical signifi cance of general practitioner gender could be demonstrated.
The present study represents all complaint cases concerning GPs in Denmark completed during one should be noted that any misclassifi cation is likely to reduce the estimated associations between motives and decision outcomes.
As in the study from New Zealand by Bismark et al. [9], most complaints were based on the Communication and Correction motives. Bismark et al. did not investigate the impact of complaint motives on decision outcome, and we believe that our study is the fi rst to do that. Feeling deserted or humiliated (in the sense of feeling objectifi ed, insulted, ignored, or ridiculed) has been suggested to be important when patients lose trust in the doctor -patient relationship [11]. Accordingly, Beckman et al. [12] identifi ed Devaluing patient and/or family views as the most important issue in almost one-third of malpractice suits and poor communication skills have previously been suggested to be predictive of receiving patient complaints [3]. In such situations our study suggests a decreased likelihood of being disciplined, perhaps because those feeling devalued or seeking punitive measures are overwhelmed by resentment whilst complaints motivated by a wish for expression of responsibility or a review of the GP ' s competence might mirror a relatively higher degree of matter-of-factness.
Two previous case-control studies (including hospital doctors) have demonstrated an association between increased professional seniority and complaint case discipline [13,14], but the fi ndings were contradicted in a third study demonstrating decreased odds [15]. The suggested association between professional seniority and discipline may refl ect seniority-dependent job contents: senior GPs might be those handling the most complex patient encounters. Anyhow, the study fi ndings could not verify that patient gender , patient age , and serious urgent illness had any impact. Alternatively, the signifi cance of professional seniority might refl ect an unspecifi c burnout phenomenon. A comprehensive European cross-sectional questionnaire survey speaks in favour of this conception: this analysis of self-reports demonstrated a positive connection between professional seniority and burnout in terms of emotional exhaustion [16]. Hence, the question arises whether any association exists between complaint cases, GPs ' communication skills, and for example manifestations of burnout. Future studies should focus on the impact of GP professional seniority on performance.  3.26 year and is based on reliable register data and case fi les. The motives for complaining were recorded through review of the complaint letters and categorized according to the instrument produced by Bismark and colleagues [8]. The instrument has been validated previously [9]; among independent reviewers, the coding of 157 complaint letters matched in 83% (131) of cases. In the present study, no measure of inter-rater reliability was calculated even though, based on independent review and rating of a small test sample of complaint letters by two raters (authors SB and JK), it was concluded that the instrument could be transferred to the Danish complaint material. Nevertheless, when interpreting study results, it case decisions for the study. Thanks are also offered to Professor Mette Hartlev for input and ideas.