Abstract Background Key recommendations for the management of patients with polypharmacy are structured medication counselling (SMC), medication lists and systematic medication reviews. Objective The aim of this study was to identify determinants (hindering and facilitating factors) for the implementation of the recommendations in general practice. Methods This study was linked to a tailored intervention aimed at improving the implementation of the recommendations in German general practice. Interviews and focus groups with different healthcare professionals were carried out in the design phase and after delivery of the intervention. The material from both data collections was analysed separately in a content analytical approach resulting in two sets of categories. For this study, the subcategories of both sets were assigned to the Tailoring Interventions for Chronic Diseases (TICD) checklist, a comprehensive framework of determinants of practice. Results Interviews with 24 general practitioners (GPs), 4 other medical specialists, 1 pharmacist, 3 nurses and 6 medical assistants as well as 2 mixed focus groups with 17 professionals were conducted. We identified 93 determinants: 29 referred to medication counselling, 21 to the use of medication lists, 20 to medications reviews and 23 to all recommendations. The determinants were related to all 7 main domains and to 25 of the 57 subdomains on the TICD checklist including guideline factors, patient factors, individual healthcare professional factors, social, political and legal factors, incentives and resources, and capacity for organizational change. Conclusion While many interventions to improve polypharmacy focus on the provision of pharmacological knowledge, a much wider range of domains need to be addressed, such as communication skills, patient involvement and practice organization. Communication, guideline adherence, general practice, medication list, medication reconciliation, polypharmacy, potentially inappropriate medication Background About one-quarter of the general population and two-thirds of patients aged over 65 years suffer from at least two chronic diseases (1) and 36% receive polypharmacy, i.e. more than five drugs on a long-term basis (2). It is well known that patients with polypharmacy have a higher risk for undesired health outcomes such as low quality of life (3), adverse drug reactions and hospitalization (4). Uncoordinated, high use of healthcare services is common among this patient group, especially in healthcare systems without formal gatekeeping such as Germany. Thus, general practitioners (GPs), who act as both central providers and coordinators of care, play an important role in drug treatment. Within the scope of the international project ‘Tailoring Interventions for Chronic Diseases (TICD)’ (5) conducted in five European countries, a tailored intervention to improve the care of multimorbid patients using polypharmacy was developed and evaluated in a cluster-randomized controlled trial (6). The first step of the intervention development consisted of a selective literature search to identify key recommendations for this patient group. We searched for scientific literature published in English and German in PubMed, for guidelines in national and international guideline portals and for ‘grey literature’ in the internet. We organized a meeting with experts in the field (employed in our department) during which the results of the literature search were presented and discussed. We chose three recommendations from a range of identified recommendations applying the following criteria: the recommendations should be evidence based and likely to improve patient outcomes or care processes but insufficiently implemented into routine care (so-called evidence-practice gap). They should be relevant for and applicable to German general practices and cover different areas of care. The following recommendations were chosen: Structured medication counselling and brown bag reviews This recommendation focuses on ‘patient-physician communication’ about medication for which deficits are well described. In a representative survey among patients with polypharmacy aged over 65 years, 48% specified that their GP had never asked them whether they take additional medications (7). Furthermore, there is little evidence for shared decision-making (8). Structured medication counselling (SMC) involves a complete inventory of the actual medication on the basis of the medication packages the patient brings to the practice (‘brown bag review’) and a subsequent counselling appointment during which the medication is discussed in a structured way (9). Consistent use of medication lists This recommendation aims at improving the ‘documentation’ of medication-related information. Discrepancies between the medication documented and actually taken by the patient appear in about 75% of the cases, while 30% are considered potentially harmful (10). In spite of ongoing attempts to establish electronic patient records, the paper-based medication list still is the most frequently used way of documenting and exchanging information about medicines in Germany. However, medication lists are frequently lacking important information, are barely readable (11) or are not available when needed. To counteract this, in Germany, minimum standards for medication lists have been defined by the Drug Commission of the German Medical Association (12) and the right to receive a medication list has recently been incorporated into German law (SGBV, §31a). Although scientific evidence about the impact of medication lists on health outcomes is lacking, it is plausible to recommend that patients with long-term medication should always take along a complete and comprehensive medication list. Medication reviews to reduce potentially inappropriate medication This recommendation focuses on the ‘quality of prescribing’ or ‘potentially inappropriate medication (PIM)’, which can be defined by explicit and implicit criteria. Explicit criteria are lists of substances which should be avoided or prescribed, respectively, in elderly, multimorbid patients. A range of explicit criteria is available, e.g. the Beer’s criteria (13), the PRISCUS list (14), the FORTA criteria (15) and the START-STOPP criteria (16). Implicit criteria, such as the Medication Appropriateness Index (MAI) (17), are checklists to assess medication regimes for certain key questions. The PIM prescription rate according to explicit criteria among elderly adults in Europe is about 23% (18). The prevalence of prescribing and monitoring errors according to implicit criteria in general practice varies between 1% and 40% (19). About 41% of the prescriptions at hospital discharge contain therapeutic errors (20). Dosing errors, timing errors and prescriptions of unnecessary drugs are among the most frequent causes (19,21). Thus, the majority of prescribing errors are potentially avoidable. There is evidence that systematic medication reviews may improve the appropriateness of prescribing and reduce accident and emergency visits (22,23). As Figure 1 shows, the recommendations are interlinked and could be regarded as a circle illustrating an optimal medication management process. Enhancing the implementation of these recommendations in general practice is likely to improve medical care and patient outcomes. Knowing the determinants helping or hindering successful implementation would be useful for both researchers designing implementation programs and for GPs caring for multimorbid patients. While internationally several qualitative studies about de-prescribing (24–28) and communication (8) have been conducted, few studies focus on medication lists and Brown Bag Reviews, and few were undertaken in Germany (29). Figure 1. View largeDownload slide Recommendations for an optimal medication management process based on (9). Figure 1. View largeDownload slide Recommendations for an optimal medication management process based on (9). This study aimed to identify determinants for the implementation of the three recommendations from the perspective of the healthcare professionals. Methods Study design This study is linked to the development and evaluation of a tailored intervention to implement recommendations about polypharmacy in general practice (5,6). It focuses on the qualitative approaches (semi-structured interviews and focus groups), which were used to identify determinants (both barriers and enablers) for the implementation of the recommendations before and after delivery of the intervention. Participants, recruitment and data collection Data collection took place at two different time points and among different target groups: (i) During the design phase of the intervention from February to April 2012, we conducted focus groups and interviews with a purposive sample of different healthcare professionals (GPs, pharmacists, specialists working in inpatient and outpatient care and nurses) to identify a broad spectrum of determinants for the implementation of the recommendations. GPs were recruited within a network of research practices established by the University Hospital Heidelberg. To recruit specialists, pharmacists and nurses working in outpatient care, we contacted all practices, nursing services and pharmacies registered in the telephone directory in the district of Heidelberg by mail. For the recruitment of specialists working in inpatient care, we contacted the senior physicians of three hospitals near Heidelberg by mail. Their contact details were obtained from the hospital websites. All respondents were given the opportunity to be interviewed or to participate in a focus group. Focus groups took place in the research department in Heidelberg and were moderated by the first author of this article. Interviews were conducted by a trained female doctoral candidate (medical student), first, in person, at a place of the participants’ choice and later on, when data saturation was felt to be reached, by telephone. (ii) After the delivery of the intervention from August to September 2014, we conducted semi-structured interviews with the GPs and medical assistants who participated in the intervention group of the randomized-controlled trial for the purpose of a process evaluation (30). All practices who had participated in the intervention were offered the chance to take part in an interview. In group practices, we asked for the participation of at least one GP and one medical assistant. The interviews took place in the practices and were conducted by a trained female doctoral candidate (medical student). During all the interviews, only the researcher and the interviewee were present. Each time the researcher who had not previously met the participants introduced herself stating her professional background (medical student) and the reason for the study. None of the participants interviewed in 2012 took part in the implementation study conducted in 2014. All interviewees gave written informed consent to participate. All interviews and focus groups followed a semi-structured interview guideline (see Supplementary material 1) were audio-taped and fully transcribed. In 2012, the guideline for the focus groups was developed on the basis of the TICD checklist (31), and the guideline for the interviews was informed by the results of the focus groups. The guidelines for the interviews in 2014 followed a previously described ‘logic model’ specifying the assumed mechanisms of the intervention (30) and a previously published protocol coordinating the process evaluations conducted in all five trials of the TICD project (32). Since we interviewed participants with and without an experience of the implementation study, the content of the interview guides differed. For example, the interview guide used after the delivery of the intervention contained questions about the implementation strategies and intervention fidelity, which were not applicable to the interviews conducted during the design phase of the intervention. However, both guides used in 2012 and 2014 contained a detailed description of the recommendations and the following key aspects: (i) What was the participants’ view of the recommendations on polypharmacy? (ii) Which factors facilitate or hinder the implementation of the recommendation into practice, in their opinion? Data analysis Interviews and focus groups were analysed according to the principles of qualitative content analysis (33), meaning that first main categories reflecting the research questions were created in a deductive manner. Subsequently, two researchers analysed the material independently using the software Atlas.ti (34) and added subcategories by inductive coding. After every 1–3 interviews, the evolving sets of categories were discussed, and a consensus version was created. After the final set of categories was determined, the assignment of quotations to subcategories was checked for consistency. The data collected in 2012 and 2014 were analysed separately so that two independent sets of categories were created. For this study, the determinants resulting from the analysis in 2012 and 2014 were assigned to the TICD checklist—a comprehensive framework of determinants of practice (31). This step was carried out by the first author of this article. Results Sample Table 1 shows the sociodemographic characteristics of the participants. In 2012, we contacted 84 GPs (response rate 33.3%, n = 28), 69 specialists (response rate 10.1 %, n = 7), 138 pharmacists (response rate 2.2%, n = 3) and 57 nurses (response rate 8.8%, n = 5) who participated in either an interview or a focus group. In 2014, we interviewed 7 of the 11 GPs and 6 of the 8 medical assistants of the intervention group. In total, 38 individual interviews with 24 GPs, 4 specialists, 1 pharmacist, 3 nurses and 6 medical assistants as well as 2 mixed focus groups with 17 professionals were conducted. The interviews lasted on average 43 min (16–93), the focus groups, 110 min. Table 1. Characteristics of the study participants interviewed pre- and post-intervention Design phase of the intervention (2012) After delivery of the intervention (2014) Interviews with HCP Focus group 1 with HCP Focus group 2 with HCP Interviews with HCP Number of participants 26 8 9 12 Professions GPs (18) Specialists (4) Pharmacists (1) Nurses (3) GPs (4) Specialists (2) Pharmacists (1) Nurses (1) GPs (6) Specialists (1) Pharmacists (1) Nurses (1) GPs (6) MA (6) Age in years average (range) 51.5 (27–63) 50.5 (27–64) 52.3 (32–69) 41.8 (25–63) Female% (n) 42.3 (11) 25 (2) 33 (3) 75 (9) Duration in minutes average (range) 50 (29–93) 107 112 36 (16–58) Design phase of the intervention (2012) After delivery of the intervention (2014) Interviews with HCP Focus group 1 with HCP Focus group 2 with HCP Interviews with HCP Number of participants 26 8 9 12 Professions GPs (18) Specialists (4) Pharmacists (1) Nurses (3) GPs (4) Specialists (2) Pharmacists (1) Nurses (1) GPs (6) Specialists (1) Pharmacists (1) Nurses (1) GPs (6) MA (6) Age in years average (range) 51.5 (27–63) 50.5 (27–64) 52.3 (32–69) 41.8 (25–63) Female% (n) 42.3 (11) 25 (2) 33 (3) 75 (9) Duration in minutes average (range) 50 (29–93) 107 112 36 (16–58) HCP, healthcare professional; GP, general practitioner; MA, medical assistant. View Large Table 1. Characteristics of the study participants interviewed pre- and post-intervention Design phase of the intervention (2012) After delivery of the intervention (2014) Interviews with HCP Focus group 1 with HCP Focus group 2 with HCP Interviews with HCP Number of participants 26 8 9 12 Professions GPs (18) Specialists (4) Pharmacists (1) Nurses (3) GPs (4) Specialists (2) Pharmacists (1) Nurses (1) GPs (6) Specialists (1) Pharmacists (1) Nurses (1) GPs (6) MA (6) Age in years average (range) 51.5 (27–63) 50.5 (27–64) 52.3 (32–69) 41.8 (25–63) Female% (n) 42.3 (11) 25 (2) 33 (3) 75 (9) Duration in minutes average (range) 50 (29–93) 107 112 36 (16–58) Design phase of the intervention (2012) After delivery of the intervention (2014) Interviews with HCP Focus group 1 with HCP Focus group 2 with HCP Interviews with HCP Number of participants 26 8 9 12 Professions GPs (18) Specialists (4) Pharmacists (1) Nurses (3) GPs (4) Specialists (2) Pharmacists (1) Nurses (1) GPs (6) Specialists (1) Pharmacists (1) Nurses (1) GPs (6) MA (6) Age in years average (range) 51.5 (27–63) 50.5 (27–64) 52.3 (32–69) 41.8 (25–63) Female% (n) 42.3 (11) 25 (2) 33 (3) 75 (9) Duration in minutes average (range) 50 (29–93) 107 112 36 (16–58) HCP, healthcare professional; GP, general practitioner; MA, medical assistant. View Large Results of the content analysis We identified 93 determinants of which 29 referred to SMC, 21 to the use of medication lists and 20 to medications reviews. Twenty-three ‘general determinants’ could not be clearly assigned to a specific recommendation. The set of categories is provided in table form in Supplementary material 2. The determinants mentioned after the delivery of the intervention were largely identical with those prioritized in the design phase. Yet, healthcare professional factors were mentioned less frequently by the participants of the intervention group. General determinants Besides ‘lack of time and reimbursement’, a range of ‘patient factors’ was mentioned as determinants for the implementation of all recommendations. Patients used a high number of over-the-counter drugs but were at the same time not willing or unable to report their medication completely. Continuity of care, i.e. whether a patient was closely connected to one pharmacy and one GP, was seen as a facilitator but difficult to establish in large practices, in nursing homes with a high staff turnover and in patients with migration background who frequently spent several months of the year in their home countries. The availability of carers in patients who were not able to care for themselves was also mentioned as general determinant. I got the impression over the years, we work on ideal conceptions (…) and in the discussions it is always the same. But in the end everything fails because of the patient. Whether he is disciplined enough, whether he considers it necessary, whether he has support (Focus group 2012) Several aspects referred ‘to interactions between health professionals’. The participants described a perceived ‘hierarchy of prescribers’ meaning that GPs may have inhibitions to question prescriptions from clinicians or specialists. The professor said yesterday what you have to take in the afternoon and now the little GP comes and says: You don’t have to take all that. Then the question is, whether the GP or the professor is right (Focus group 2012) Collaboration and information exchange between physicians, pharmacies and nursing services was another issue. Especially, changes of medication after hospital discharge and specialist appointments were frequently not communicated well and therefore hindered the creation of correct medication lists and a thorough review and discussion of the medication. From the perspective of the participants, collaboration was easier to establish in rural areas where only a small number of practices and services exist. Delegating the Brown Bag Review or screening for risk drugs to medical assistants in general practice could facilitate implementation. Some participants suggested that, for instance, the medication review or instructions about drug administration could be done by pharmacists, yet also concerns about involving pharmacists who did not know the complete patient history were expressed. I think the quality of counselling by local pharmacists is frequently not good. Too uncritical, too profit-oriented, many things are not questioned (…). Very problematic! (Specialist 2012) Some determinants referred to the ‘individual health professionals’ factors’. Besides profound pharmacological knowledge, attitudes and convictions played a role. Some GPs expressed their reluctance to standardize care in multimorbid patients because they felt that this was a way of questioning their competence and impeding individual patient care. Others doubted that adherence to the recommendations would improve care at all. Several statements suggested that the participants were not aware of the care problem or the deficiencies in their own practice and therefore saw no need to implement the recommendations. In our practice all this happens automatically already and this is why there was nothing that needed changed (Medical assistant 2014) On a ‘systems level’ so-called rebate contracts were frequently mentioned as a hindering factor for successful implementation. In Germany, health insurance companies can make contracts with the pharmaceutical industry. Pharmacists are obliged to distribute only the generic drugs specified in the respective contract regardless of the specification on the prescription form. Due to the multitude and frequent changes of these contracts, the trade name and appearance of the drug may change every time the patient gets a prescription. According to the interviewees, this hindered the creation of comprehensive medication lists, caused difficulties of comprehension during medication counselling and increased the risk of PIM due to drug application errors. Determinants specific for the implementation of SMC Generally, SMC on the basis of a Brown Bag Review was perceived as an important but laborious procedure which depended largely on the ‘compatibility with existing structures’. Practices scheduling appointments and using recall systems were expected to face less difficulties in implementing SMC than practices with open consultation hours. Combining SMC with existing routines, e.g. disease management programs, check-ups or home visits, was seen as a facilitator. Many GPs felt that talking about medication was something they already did even if not in such a structured way. Due to the high effort involved, SMC could only be offered to patients really in need of it. Identifying those patients was seen as a challenge. I don’t do it in such a structured way that I say: Today is medication counselling! I do it only if there is a need (Focus group 2012) The outcomes expected from SMC might also influence the motivation of GPs to implement it. Some expected SMC to improve patients’ self-management and increase physicians’ self-reflection, others were afraid that providing extensive information about medication might unsettle patients who already received a lot of unsettling and contradictory information from many sides, e.g. from the media, friends and family, health insurance companies or package inserts. The patient’s intellectual capacities and language barriers in non-native speakers might limit the information transfer during SMC. Some interviewees doubted whether patients were really interested in talking in-depth about their medication. They expected patients not to show up to appointments or to forget to bring their medication packages for the Brown Bag Review. That some patients might feel inhibited in asking questions and that the information provided to them might quickly be forgotten were seen as hindering factors in the uptake of SMC: Part of the problem is that patients do not dare to say: I have not understood this! This applies rather to certain generations (…). Elderly patients I would say (…). Communicating on an equal level is not so well developed in many cases. The patients sits there reverently and nods” (GP 2012). GPs communication skills (e.g. the ability to use patient-centred language or to structure the conversation) and attitude towards shared decision-making were determinants on a healthcare- professional level. I take the decision and we don’t know if there will be a side effect, but they have to trust me that the medication is right for them. They are not able to understand all this, I don’t even know if they understand me (GP 2014). Determinants specific to the implementation of medication lists Generally, the participants believed that keeping medication lists up-to-date would take a great deal of effort because of frequent medication changes, which were not actively reported by patients or other healthcare professionals. However, a comprehensive medication list was also regarded as a good advertisement for the practice. The clinicians will be grateful. They will say, this patient comes from a good practice (Focus group 2012) Many interviewees preferred different formats and contents for the medication list than those defined by the template. While some preferred less information to keep the list ‘simple and clear’, others wanted to add elements such as allergies, important diagnoses or the kidney function. Some interviewees doubted that the recommendation to always take along a complete medication list would meet patients’ preferences. They objected that patients already had to carry a multitude of documents (e.g. passports for pace-makers, phenprocoumon or vaccinations) with them and that not all patients want to make their complete medication transparent to all healthcare professionals. Many patients also preferred their own, hand-written medication lists or were convinced that they knew their medication from memory and therefore did not see the need to carry a medication list with them. Furthermore, patients frequently confused their current medication list with older versions. A medication list was not a useful tool for illiterate patients, patients with severe visual impairments or those with language barriers. I notice this again and again: I print a new medication list at every appointment and when I go for a home visit I see that the patient uses a medication list that is half a year old and cannot find the current one (GP 2012) Few [patients] are interested in their medication list (…) they do not feel responsible for it (Nurse 2012) Data protection directives and the familiarity of patients with digital media were raised as determinants for the use of electronic medication lists, e.g. stored on the health insurance card or on smartphones. The functions of the practice software were another issue because not all systems allowed easy adaption of the template for medication lists, and compatibility with the systems in hospitals or other practices was usually not given. Typing errors or accidently switching a line in the drug information system could lead to incorrect medication lists. Determinants specific to the implementation of medication reviews Performing systematic medication reviews with instruments such as the MAI or the PRISCUS list was regarded critically because they were not seen as feasible for daily use. Tools should be integrated into the practice software and provide extensive information and functions, such as full access to the Summaries of Product characteristics, interaction checks or automatic import of diagnoses and laboratory results. Regardless of technical aids, profound pharmacological knowledge was regarded as a prerequisite to reduce PIM. The interviewees expressed discontent with existing possibilities for pharmacological training, which were often influenced by the pharmaceutical industry. While conflicting guidelines and the quick change of recommendations made training more difficult, the limited number of medications used in general practice was regarded as facilitator. The culture of prescribing a drug for every symptom instead of considering alternative, non-pharmacological treatment options needed to be changed. In some patients, the lack of alternatives or the fear that stopping a medication might have negative consequences played a role. Sometimes you don’t have a choice, if the patient has complaints. You want to help the patient, you must help the patient. So what can you do? (GP 2012) Sometimes you hope: Hopefully nothing will happen! Because if I change something and then something goes wrong and (…) then they say “This [medication] was not given at all”! (GP 2012) Also, patient preferences and patient behavior contributed to PIM, e.g. if patients insisted on prescriptions of certain drugs, were unwilling to change a medication regime, which they had used for years or if they were addicted. Some risky drugs (e.g. NSAR) were even available as over-the-counter drugs. Also, non-adherence and application errors (e.g. changing the dose, confusing medications etc.) could lead to a PIM. Patients have their own wishes and somehow they are also customers. If I tell them three times “You will not get this” they will go to someone else who gives it to them (GP 2012) The process of issuing prescriptions was seen as a frequent source of error with high potential to improve the use of medication lists and to reduce PIM prescriptions. In many practices, patients could request repeat prescriptions at any time. The prescriptions were prepared by the medical assistant and signed by the GP at the next opportunity, usually between two patient appointments. There was a risk of issuing repeat prescriptions for medications that were not on the medication list, not indicated any more, had become contraindicated or required a change of dose. Discussion This qualitative study provided a broad compilation of hindering and enabling factors to improve the implementation of structured medication counselling, Brown Bag Reviews, systematic medication reviews and the use of medication lists in general practice. Most qualitative studies in this field focus on reasons for prescribing PIM (24,26,28,29), or polypharmacy in general (25,27), whereas this study examined concrete, evidence based and interlinked recommendations for medication management in general practice. While many determinants we identified for medication reviews and reduction of PIM are identical with those described in the international literature, this study adds insights about other important processes, namely medication-related communication and documentation. Interventions to improve the appropriateness of polypharmacy usually focus on the provision of expert knowledge and education. For instance, in a Cochrane review, 11 of the 12 studies involved pharmacists in medication reviews. Education of patients and healthcare professionals and computerized decision support systems were other strategies used (23). In our study, only a small proportion of the identified determinants was connected to pharmacological knowledge. This finding has implications for endeavours to improve the management of polypharmacy in the fields of education, practice, politics and research. Implications for education Education of health professionals should address their attitudes and cognitions, as well as their knowledge and skills. For example, the extent to which healthcare professionals agree with the recommendations or their individual attitudes towards shared decision- making, extensive prescribing or non-pharmaceutical treatment options are likely to influence adherence to the recommendations. Provision of evidence and exchange with colleagues and other professionals are useful for opinion shaping and could be widely implemented by ‘quality circles’. Quality circles are peer groups of healthcare professionals who meet on a regular basis to improve the quality of care in their own practices. In Southern Germany, some quality circles receive additionally evidence based information and repeated written feedback on individual prescribing patterns (35). Intersectoral and multi-disciplinary quality circles, involving representatives of hospitals, pharmacies, specialists and GPs of one area, could help to improve professional interactions and team processes, which have also been mentioned as determinants for the implementation of the recommendations. In our study, the co-operation of the patient and/or the informal carer was regarded essential for the implementation of all three recommendations. Thus, these determinants should be targeted by interventions and enforced in the patient-physician relationship. This requires among other things, advanced communication skills. In the Netherlands, training on communication skills by means of a video feedback instrument has been an obligatory part of postgraduate training for GPs for more than 20 years (36). This concept could be adapted to communication about medicine and implemented in other European countries (37). Provision of information, reminders or incentives for patients could be other measures to improve their involvement in structured medication management. Implications for practice Practice organization, e.g. how the process of issuing prescriptions is organized or the extent to which a practice has established structures and routines for related care processes such as disease management programs, may influence the ability to adhere to the recommendations. In Germany, participation in care programs is usually voluntary and standards for medication management do not exist, resulting in a great variety concerning the organizational processes in general practices. In particular, the role of practice assistants in patient care varies from reasonably high to absent. So, change management interventions considering the particularities of the individual practice could be promising in improving polypharmacy. GPs could use the compilation of determinants provided by this study to analyse the situation in their own practice and find individual solutions. Implications for research and politics Barriers related to the healthcare system, e.g. that some PIM are available as over-the-counter drugs, lack of time and reimbursement or intersectoral communication, cannot be influenced by individual healthcare professionals and quality improvement programs, yet they might be decisive for successful implementation. Thus, political solutions should be sought, e.g. by the professional associations of GPs. As the study provides a broad spectrum of barriers and enablers, future research should address the prioritization of determinants more explicitly. Since the determinants cover different areas of care, it is likely that multi-faceted interventions will be needed to achieve meaningful improvement. The evaluation of such complex programs poses challenges, which implies that well-designed process evaluations and methodological debates about appropriate research designs are crucial (38). Strengths and limitations This study focuses on three interlinked areas of care (communication, documentation and prescribing), which are essential to improve the management of patients with polypharmacy. It is based on a comprehensive data collection among different stakeholders before and after delivery of an implementation intervention to improve the management of polypharmacy. Thus, it gathers a broad spectrum of perspectives and determinants. However, there is an imbalance concerning the type and number of the targeted participants. The majority of the respondents was GPs while comparatively few pharmacists, nurses and medical assistants participated. Therefore, it has to be assumed that the results are dominated by the GP perspective and that the views of the other healthcare professionals are not fully represented. This is supported by the observation that the determinants collected among a broad spectrum of target groups in the design phase of the intervention were largely identical with those identified after the delivery of the intervention among GPs and medical assistants of the intervention group only—with the exception of healthcare professional factors, which were less frequently mentioned by the latter. We chose to mix different professions in the focus group to provoke an interdisciplinary discussion, but it is also possible that this prevented the participants in talking openly. Furthermore, the focus groups were moderated by the first author of this study and not by an independent researcher. Due to time constraints, the participants did not provide feedback on the findings. Data collection was done in 2012 and 2014. Meanwhile, new developments have occurred in the German healthcare system. For example, the E-health law (SGBV, §31a), which has been enacted in 2015, regularizes the patients’ right to receive a medication list. It has to be assumed that such new developments may help to overcome some of the identified barriers and bring about new determinants not identified by this study. Yet, to date it is not possible to judge the actual impact of this recent change. Future studies will be needed to evaluate the impact of the E-Health law and the implementation of the federal medication list, which has begun fairly recently, in October 2016. Conclusion Implementing recommendations about polypharmacy in practice is facilitated or hindered, respectively, by a range of distinct determinants. Healthcare professional factors, patient factors, organizational factors and systems factors play a role. It is likely that multi-faceted interventions targeting barriers in different areas are needed to achieve meaningful improvement in the management of polypharmacy, although they are frequently more difficult to implement. Besides the provision of pharmacological expertise, improving patient involvement, practice organization and communication training might be relevant key elements. Supplementary material Supplementary material is available at Family Practice online. Acknowledgements We are grateful to Sarah Kuse and Johanna Katharina Reichel for collecting the interview data. Declaration Funding: The study was funded by the European Union Framework 7 program within the theme HEALTH.2013.3.1-1 under grant agreement no 258837. Ethical approval: This study is linked to a cluster-randomized controlled trial, which was registered under the registration number ISRCTN34664024. It was approved by the ethics committee in Heidelberg (reference number S252/2013) and carried out in compliance with the Declaration of Helsinki (2013 version, Fortaleza, Brazil). All participants provided their written informed consent prior to participation in the study. Authors’ contributions: CS and JS designed the study and conducted the focus groups. CS analysed the data and wrote the first draft of the manuscript. JS, SZ, TF and MW provided substantial intellectual input. All authors critically revised the manuscript and agreed on the final version. Conflict of interest. All authors declare that they have no competing interests. References 1. Barnett K , Mercer S-W , Norbury M , et al. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study . Lancet 2012 ; 380 : 37 – 43 . Google Scholar CrossRef Search ADS PubMed 2. Grandt D , Schubert I , Barmer GEK . Arzneimittelreport 2016, page 55, table 28 . https://www.barmer.de/blob/36730/5d1b2964c4fe2dc9de815c357fda7dc8/data/pdf-arzneimittelreport-2016.pdf (accessed on 4 October 2017 ). 3. Montiel-Luque A , Núñez-Montenegro AJ , Martín-Aurioles E , et al. ; Polipresact Research Group . Medication-related factors associated with health-related quality of life in patients older than 65 years with polypharmacy . PLoS One 2017 ; 12 : e0171320 . Google Scholar CrossRef Search ADS PubMed 4. Patel NS , Patel TK , Patel PB , et al. Hospitalizations due to preventable adverse reactions-a systematic review . Eur J Clin Pharmacol 2017 ; 73 : 385 – 98 . Google Scholar CrossRef Search ADS PubMed 5. Wensing M , Oxman A , Baker R , et al. Tailored implementation for chronic diseases (TICD): a project protocol . Implement Sci 2011 ; 6 : 103 . Google Scholar CrossRef Search ADS PubMed 6. Jäger C , Freund T , Steinhäuser J , et al. Impact of a tailored program on the implementation of evidence-based recommendations for multimorbid patients with polypharmacy in primary care practices-results of a cluster-randomized controlled trial . Implement Sci 2017 ; 12 : 8 . Google Scholar CrossRef Search ADS PubMed 7. Zok K . Einstellungen älterer Menschen zur Arzneimitteltherapie. Ergebnisse einer Umfrage unter 1.000 GKV-Versicherten ab 65 Jahren . WIdO-monitor 2012 ; 9 : 1 – 8 . 8. Stevenson FA , Cox K , Britten N , et al. A systematic review of the research on communication between patients and health care professionals about medicines: the consequences for concordance . Health Expect 2004 ; 7 : 235 – 45 . Google Scholar CrossRef Search ADS PubMed 9. Mahler C , Freund T , Baldauf A , et al. [Structured medication management in primary care - a tool to promote medication safety] . Z Evid Fortbild Qual Gesundhwes 2014 ; 108 : 258 – 69 . Google Scholar CrossRef Search ADS PubMed 10. Tulner LR , Kuper IM , Frankfort SV , et al. Discrepancies in reported drug use in geriatric outpatients: relevance to adverse events and drug-drug interactions . Am J Geriatr Pharmacother 2009 ; 7 : 93 – 104 . Google Scholar CrossRef Search ADS PubMed 11. Freigofas J , Seidling HM , Quinzler R , et al. . Characteristics of medication schedules used by elderly ambulatory patients . Eur J Clin Pharmacol 2015 ; 71 : 1109 – 20 . Google Scholar CrossRef Search ADS 12. Drug Commission of the German Medical Association (AKDAE). Specifications for a Patient-related Medication List [Spezifikation für einen patientenbezogenen Medikationsplan] . http://www.akdae.de/AMTS/Medikationsplan/docs/Medikationsplan.pdf (accessed on 4 October 2017 ). 13. The American Geriatrics Society 2015 Beers Criteria Update Expert Panel . American Geriatrics Society 2015 updated beers criteria for potentially inappropriate medication use in older adults . J Am Geriatr Soc 2015 ; 63 : 2227 – 46 . CrossRef Search ADS PubMed 14. Holt S , Schmiedl S , Thürmann PA . Potentially inappropriate medications in the elderly: the PRISCUS list . Dtsch Arztebl Int 2010 ; 107 : 543 – 51 . Google Scholar PubMed 15. Pazan F , Weiss C , Wehling M ; FORTA . The FORTA (Fit fOR The Aged) list 2015: update of a validated clinical tool for improved pharmacotherapy in the elderly . Drugs Aging 2016 ; 33 : 447 – 9 . Google Scholar CrossRef Search ADS PubMed 16. Knol W , Verduijn MM , Lelie-van der Zande AC , et al. [Detecting inappropriate medication in older people: the revised STOPP/START criteria] . Ned Tijdschr Geneeskd 2015 ; 159 : A8904 . Google Scholar PubMed 17. Samsa GP , Hanlon JT , Schmader KE , et al. A summated score for the medication appropriateness index: development and assessment of clinimetric properties including content validity . J Clin Epidemiol 1994 ; 47 : 891 – 6 . Google Scholar CrossRef Search ADS PubMed 18. Tommelein E , Mehuys E , Petrovic M , et al. Potentially inappropriate prescribing in community-dwelling older people across Europe: a systematic literature review . Eur J Clin Pharmacol 2015 ; 71 : 1415 – 27 . Google Scholar CrossRef Search ADS PubMed 19. Avery AJ , Ghaleb M , Barber N , et al. The prevalence and nature of prescribing and monitoring errors in English general practice: a retrospective case note review . Br J Gen Pract 2013 ; 63 : e543 – 53 . Google Scholar CrossRef Search ADS PubMed 20. Michaelson M , Walsh E , Bradley CP , et al. Prescribing error at hospital discharge: a retrospective review of medication information in an Irish hospital . Ir J Med Sci 2017 . 21. Koper D , Kamenski G , Flamm M , et al. Frequency of medication errors in primary care patients with polypharmacy . Fam Pract 2013 ; 30 : 313 – 9 . Google Scholar CrossRef Search ADS PubMed 22. Christensen M , Lundh A . Medication review in hospitalised patients to reduce morbidity and mortality . Cochrane Database Syst Rev 2016 ; 2 : CD008986 . Google Scholar PubMed 23. Cooper JA , Cadogan CA , Patterson SM , et al. Interventions to improve the appropriate use of polypharmacy in older people: a Cochrane systematic review . BMJ Open 2015 ; 5 : e009235 . Google Scholar CrossRef Search ADS PubMed 24. Ailabouni NJ , Nishtala PS , Mangin D , et al. Challenges and enablers of deprescribing: a general practitioner perspective . PLoS One 2016 ; 11 : e0151066 . Google Scholar CrossRef Search ADS PubMed 25. Anthierens S , Tansens A , Petrovic M , et al. Qualitative insights into general practitioners views on polypharmacy . BMC Fam Pract 2010 ; 11 : 65 . Google Scholar CrossRef Search ADS PubMed 26. Bolmsjö BB , Palagyi A , Keay L , et al. Factors influencing deprescribing for residents in Advanced Care Facilities: insights from General Practitioners in Australia and Sweden . BMC Fam Pract 2016 ; 17 : 152 . Google Scholar CrossRef Search ADS PubMed 27. Mc Namara KP , Breken BD , Alzubaidi HT , et al. Health professional perspectives on the management of multimorbidity and polypharmacy for older patients in Australia . Age Ageing 2017 ; 46 : 291 – 9 . Google Scholar PubMed 28. Palagyi A , Keay L , Harper J , et al. Barricades and brickwalls–a qualitative study exploring perceptions of medication use and deprescribing in long-term care . BMC Geriatr 2016 ; 16 : 15 . Google Scholar CrossRef Search ADS PubMed 29. Voigt K , Gottschall M , Köberlein-Neu J , et al. Why do family doctors prescribe potentially inappropriate medication to elderly patients ? BMC Fam Pract 2016 ; 17 : 93 . Google Scholar CrossRef Search ADS PubMed 30. Jäger C , Steinhäuser J , Freund T , et al. A tailored programme to implement recommendations for multimorbid patients with polypharmacy in primary care practices-process evaluation of a cluster randomized trial . Implement Sci 2017 ; 12 : 31 . Google Scholar CrossRef Search ADS PubMed 31. Flottorp SA , Oxman AD , Krause J , et al. A checklist for identifying determinants of practice: a systematic review and synthesis of frameworks and taxonomies of factors that prevent or enable improvements in healthcare professional practice . Implement Sci 2013 ; 8 : 35 . Google Scholar CrossRef Search ADS PubMed 32. Jäger C , Freund T , Steinhäuser J , et al. Tailored implementation for chronic diseases (TICD): a protocol for process evaluation in cluster randomized controlled trials in five European countries . Trials 2014 ; 15 : 87 . Google Scholar CrossRef Search ADS PubMed 33. Mayring P . Qualitative content analysis . Forum Qualitative Social Research 2000 ; 1 . http://www.qualitative-research.net/index.php/fqs/article/view/1089/2385 (accessed on 4 October 2017 ). 34. ATLAS. Ti . ATLAS. ti: The Qualitative Data Analysis & Research Software . http://atlasti.com/ (accessed on 4 October 2017 ). 35. Wensing M , Broge B , Riens B , et al. Quality circles to improve prescribing of primary care physicians. Three comparative studies . Pharmacoepidemiol Drug Saf 2009 ; 18 : 763 – 9 . Google Scholar CrossRef Search ADS PubMed 36. Van Dalen J , Bartholomeus P , Kerkhofs E , et al. Teaching and assessing communication skills in Maastricht: the first twenty years . Med Teach 2001 ; 23 : 245 – 51 . Google Scholar CrossRef Search ADS PubMed 37. Hammersen F , Goetz K , Soennichsen A , et al. Effects of communication training with the MAAS-Global-D instrument on the antibiotic prescribing for respiratory infections in primary care: study protocol of a randomised controlled trial . Trials 2016 ; 17 : 180 . Google Scholar CrossRef Search ADS PubMed 38. Van Belle S , Rifkin S , Marchal B . The challenge of complexity in evaluating health policies and programs: the case of women’s participatory groups to improve antenatal outcomes . BMC Health Serv Res 2017 ; 17 : 687 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Family Practice – Oxford University Press
Published: Dec 16, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera