Pharmacists' documentation of interventions in seamless care : PharmDISC

Patient transition across care settings represent a high-risk period for the occurrence of drug-related problems (DRP), such as discrepancies. These DRPs often result in patient readmission, resulting in higher costs of care in public health. A DRP is commonly defined as an event or circumstance involving drug therapy that actually, or potentially, interferes with the desired health outcomes. In both community and hospital settings, there is evidence that interventions initiated by pharmacists can reduce the occurrence of DRPs. For this thesis, we defined a ‘‘pharmaceutical intervention’’ as a recommendation initiated by a pharmacist in response to a DRP occurring in an individual patient in any phase of the medication process. The pharmaceutical intervention aims at optimising pharmacotherapy, in terms of efficacy, safety, economic, and humanistic aspects.
The exchange of information (e.g. on pharmaceutical interventions) between primary and secondary care remains, however, a major challenge. The access to complete and accurate patient medical information and good communication is essential for the healthcare professionals to ensure safe and efficient care to the patients. In the current practice, the medication management at the time of admission and discharge from hospital is not seamlessly guaranteed through complete documentation and communication of clinical pharmaceutical interventions between inpatient and outpatient care. Seamless care is defined as any process which optimises efficiency, quality, and safety of medication management at transitions to establish a continuum of care.
These pharmaceutical interventions should no longer be loose fragments, but should be brought together like a mosaic in an overall concept and documented in a form that enables the most seamlessly possible exchange of information at the hospital discharge of patient in the outpatient situation. To accomplish this task, the first step is to document the pharmaceutical interventions in their respective care setting by developing valid structured instruments in order to depict the practice. Such documentation of care represents the evidence of practice. It is therefore essential to be recorded in a standardised and structured manner. Classification, as essential part of documentation, enables a precise representation of what has been done (e.g. pharmaceutical interventions) by categorising key elements in a standardised manner, and as a consequence, facilitates the transfer of information.
Once the pharmaceutical interventions are documented in the respective healthcare setting, the information exchange between the hospital and the community pharmacy and vice versa still remains challenging. Improving information exchange regarding pharmaceutical interventions could enable a more efficient and safer transfer of patients between inpatient and outpatient care. Thus as a second step, an aligned classification system in both settings would facilitate a standardised documentation of pharmaceutical interventions.
Hence, validated, structured, and standardised classification systems for pharmaceutical interventions, which fulfil both requirements of comprehensiveness and easy application with little time expenditure in daily clinical practice, are rare. Furthermore, there was no national consensus in Switzerland on how to record pharmaceutical interventions in a standardised manner to obtain data allowing epidemiological studies for research and political purposes. Therefore, we recognised the need of proper instruments able to depict the practice in their representative setting.
The goal of this thesis was to create structured instruments for daily practice to improve the continuity of documentation and communication of pharmaceutical interventions during transitions of care. We approached this goal through
a) developing and validating classification systems for pharmaceutical interventions (one for hospital and one for community setting)
b) testing their feasibility in daily practice in observational studies
c) exploring pharmacists’ satisfaction and opinions on documentation of pharmaceutical interventions
d) analysing the documented pharmaceutical interventions
e) investigating with an observation study the dispensing process of prescribed medicines in daily practice of community pharmacies, focusing on counselling activities
f) assessing pharmacist’s opinions on patient counselling and on transfer of documented pharmaceutical interventions
In the first project of this thesis (Project A), we developed together with the working group in clinical pharmacy of the Swiss Society of Public Health Administration and Hospital pharmacists (GSASA) an intervention oriented classification system for the hospital setting, the GSASA system. Study A1 aimed at validating the GSASA system. The GSASA system includes 5 categories (problem, type of problem, cause of intervention, intervention, and outcome of the intervention) and 41 subcategories. Total interrater reliability was moderate (Fleiss’ Kappa coefficient K=0.52). Interrater reliability and acceptability of the GSASA system were comparable to those of the well-established Pharmaceutical Care Network Europe (PCNE) system V6.2.
In 2011, GSASA proposed the GSASA classification system of pharmaceutical interventions to all Swiss hospitals that are members of this society, and encouraged its application. One and a half years later, the implementation of the GSASA system was evaluated to assess implementation outcome such as the number of hospital pharmacies using the system, to analyse the pooled data retrieved from Swiss hospitals, and to explore the user satisfaction (A2). Forty-four chief hospitals pharmacists responded by online questionnaire about the use and satisfaction with the classification system. Eleven of 12 hospitals using the GSASA system provided us voluntary all classification data, covering an observation period of 121.5 months. Of a total of 9’543 recorded pharmaceutical interventions, 8.8% were not fully classifiable (n=840). In general, users were satisfied (3.8±0.9, Likert-scale 1-5) with the GSASA system, especially with its adequate time expenditure (4.1±1.0). Ten users (83.3%) reported to need less than two minutes and two (16.7%) up to four minutes to classify one intervention. The extent to which the system is used and the good acceptance within a short time after implementation are promising results to use it as basis for a further development.
The aim of next study (A3) of the project was to design an innovative seamless concept of classification of pharmaceutical interventions in patient care. The basic structure of the GSASA classification system, currently used in hospitals, should be adopted as far as possible. As a first exploratory trial to test the suitability of the GSASA system in ambulatory settings, we analysed 65 protocols of medication reviews (Polymedication-Check, PMC) performed by community pharmacists, and all 190 interventions could be classified using the GSASA system (median of 3 per PMC). However, the system does not provide detailed information about certain interventions. We identified the need for a new classification system which allows high flexibility in documenting pharmaceutical interventions. According to the complexity of the case, the available information, the type of medication review, and the need for follow-up, different levels of classification may be indicated. This classification system should be suitable for both, community and hospital pharmacy practices to facilitate continuity of care.
The second project of the thesis (Project B) reports the development process of the Pharmacists’ Documentation of Interventions in Seamless Care (PharmDISC) system which was split into two parts and four stages: Part 1 covered the development and piloting stages (B1), while Part 2 covered the evaluation and implementation stages (B2).
The aim of Part 1 (B1) was to develop an intervention oriented classification system for community setting, the PharmDISC system, based on the GSASA system for the hospital setting (development stage), and to validate (interrater reliability, appropriateness, interpretability) it in an academic environment (piloting stage). In a prospective observational study in community pharmacies, 77 master students in pharmacy consecutively collected each 10 first prescriptions requiring a pharmaceutical intervention and classified these interventions with the PharmDISC system. The classification system includes 5 categories and 52 subcategories. Most of the 725 pharmaceutical interventions (n=686, 94.6%) were completely classified. The PharmDISC system reached an overall substantial users agreement (K=0.61). Additionally, with a focus group of nine pharmacists (six community and three hospital pharmacists), we assessed their opinions on the documentation of pharmaceutical interventions, and assessed face and content validity of the PharmDISC system. Despite some arising points for optimisation, the pharmacists were satisfied with the PharmDISC system. They recognised the importance of documentation of pharmaceutical interventions and believed that this may allow traceability, facilitate communication within the team and other healthcare professionals, and increase quality of care. Refinement based on the pharmacists’ suggestions resulted in a final version to be tested in an observational study with community pharmacists.
In Part 2 of the PharmDISC development process (B2), the PharmDISC was tested on interrater reliability, appropriateness, interpretability, acceptability, feasibility, and validity in the daily life environment of community pharmacies (evaluation stage) and first implementation aspects were explored (implementation stage). In an observational study, 21 pharmacists each classified 30 prescriptions requiring a pharmaceutical intervention with the PharmDISC system on 5 selected days within a 5-weeks period. The participating pharmacists were trained with an online training and could use a descriptive manual of the PharmDISC system to support them in the classification of pharmaceutical interventions. The PharmDISC system reached an average substantial user agreement (K=0.66). Of 519 documented pharmaceutical interventions, 430 (82.9%) were completely classified. Most users found the system comprehensive and practical. The PharmDISC system raised the awareness regarding drug-related problems for most users. To facilitate its implementation, an electronic version that automatically connects to the prescription together with a task manager for pharmaceutical interventions needing follow-up was suggested. Barriers could be time expenditure and lack of understanding the benefits.
A subanalysis (B3) based on the data obtained from the validation results (B2) allowed characterising the pharmaceutical interventions performed during dispensing of prescribed medicines in community pharmacies, and identifying the frequent problems with the prescribed medicines. Pharmacists performed individualised pharmaceutical interventions to solve or prevent DRPs concerning prescribed medicines. Pharmacists mainly intervened to substitute a drug (n=132, 30.7%), adjust a dose (n=57, 13.3%), and clarify/complete information (n=48, 11.2%). In 138 (32.1%) cases, the pharmacists contacted the prescriber whereas in 292 cases (67.9%), only the pharmacist was involved (alone n=59, with the patient n=222, with the caregiver n=11). Direct patient-pharmacist interaction during the dispensing was essential to detect patient-reported problems with prescribed medicines.
In the third project of the thesis (Project C), we observed on site the whole dispensing process of prescribed medicines, focusing on counselling activities, in order to depict the current practice in community pharmacies (C1). One master student in pharmacy performed non-participant observations during one day at each of the 18 included community pharmacy. Within 556 prescription encounters, counselling was provided to 367 (66.0%) customer on 2.9 ±3.1 themes per prescription encounter (first 4.9±3.0; refill 1.0±1.7, p<0.001), predominantly about drug administration, use and dose. We identified factors influencing counselling provision at patient, prescription and pharmacy level. Significantly more counselling was provided by pharmacists, to customers with a first prescription, with a prescription requiring a pharmaceutical intervention, to carers who filled the prescription for a patient, to new customers, and to customers who did not refuse counselling. While pharmacists intervened frequently, only few additional activities and no further services were offered.
Additionally, at the end of the observation day (C1), an interview (C2) was conducted with one pharmacist of each participating pharmacy to assess pharmacists’ opinions on patient counselling during dispensing of prescribed medicines in daily community pharmacy practice, and on documentation and transfer of pharmaceutical interventions. For the eighteen interviewed pharmacists, most important themes to be discussed at first prescription dispensing were indication, administration, and anamnesis and at refill prescription dispensing, adherence, therapy benefits, and adverse effects. The most frequently counselling triggers that pharmacists expressed were patient’s knowledge gap, patient’s motivation/interest, drug-drug interaction, polypharmacy/polymorbidity and special patient population. Barriers were refusal by patients, communication problems, lack of medical data, and lack of time. Pharmacists occasionally documented their pharmaceutical interventions, however almost always not in a standardised way. Pharmacists found important to transfer the performed pharmaceutical interventions to the other involved healthcare providers, but some barriers (e.g. too time-consuming, overwork) could hinder it. Therefore, a simple and fast in use computerised documentation system, with an additional intervention history option, could be a promising approach.
In summary, this thesis showed the following:
The GSASA system
• The GSASA classification system appeared to be reliable and promising for the documentation of pharmaceutical interventions in daily practice (practical and less time-consuming). Its validation was successful in terms of appropriateness, interpretability, validity, acceptability, feasibility, and reliability (A1).
• After 18 months of introduction (2013), the GSASA classification system is already widely accepted in Swiss hospitals, suggesting to be suitable also to daily life settings. Most pharmaceutical interventions can be classified with adequate time effort and overall users’ satisfaction is good. The extent to which the system is used and the good acceptance within a short time after implementation are promising results to use it as basis for a further development (A2).
• The GSASA classification system was tested in primary care and proved to be suitable also to classify interventions of medication reviews performed by community pharmacists in primary care; however, further refinements were necessary to improve the precision of the system. Thus, the development of one classification system suitable for both, primary and secondary care, flexible for addressing different levels of complexity, and easily integrable in daily practice and in electronic patient file was recognised as a promising approach (A3).
The PharmDISC system
• In a focus group interview, pharmacists recognised the importance of the documentation of pharmaceutical interventions and were convinced that this may allow traceability, facilitate communication within the team and other healthcare professionals, and eventually would increase quality of care (B1).
• Substantial interrater reliability and high rating of acceptability and feasibility indicates that the new PharmDISC system is a valid system for the documentation of pharmaceutical interventions in daily practice of community pharmacies. The pharmacists were satisfied with the system and considered it helpful, easy to use, and practical for daily work. They appraised the fact that by using an intervention oriented classification system, their awareness of DRPs and concurrently the intervention rate increased (B2).
• The developed descriptive manual of the PharmDISC system and the online training were helpful elements for an accurate use of the PharmDISC system and are promising utilities to enhance its implementation (B2).
Depicting real-life daily practice
• The high number of pharmaceutical interventions following DRPs and patient-reported problems highlights the importance of a direct patient-pharmacist interaction when dispensing prescribed medicines (B3).
• The observation of the dispensing process of prescribed medicines allowed to depict the community pharmacy practice from the customers’ perspective (at the counter). However, counselling was not equally provided, indicating that prescription encounters need different degrees of counselling. A more transparent practice and patient-centered counselling is necessary to better meet the patients’ needs on information. While pharmacists intervened frequently, only few additional activities and no further services were offered (C1).
• Factors influencing counselling provision were identified at patient, prescription and pharmacy level. Significantly more counselling was provided by pharmacists, to customers with a first prescription, to customers with a prescription requiring a pharmaceutical intervention, to carers who filled the prescription for a patient, to new customers, and to customers who did not refuse counselling (C1).
• A discrepancy in counselling content by observation compared to pharmacists’ opinions was revealed. Observations show a focus on product-centered themes (e.g. drug administration, dose), whereas pharmacists’ interviews highlight the importance of patient-centered themes (e.g. benefit, adherence). This might indicate that pharmacists are aware but hindered by barriers to practice according to good pharmacy practice guidelines (C2).
• Pharmacists recognised the importance of the documentation of pharmaceutical interventions and their transfer to others healthcare providers, but reported also possible reasons of non-transfer (e.g. minor relevant of pharmaceutical interventions, overwork) (C2).
• A simple and fast in use computerised documentation system, with an additional intervention history option, could be a promising approach according to the positive reactions and the needs of the pharmacists. As stated by the pharmacists, its implementation should increase the appreciation and visibility of pharmacists’ work, facilitate data handling by saving time and costs, ensure seamless care by improving collaboration among healthcare providers, and ultimately improve the therapy outcomes (C2).