Nursing home care is an important component of the care services provided for the elderly population. In many societies, as the proportion of elderly people continues to increase, the nursing home resident population also increases and ages [1
]. Compared with the community-dwelling elderly, nursing home residents generally require more medication and this puts them at greater risk of inappropriate medication use [2
]. Indeed, in nursing homes multiple medication use (polypharmacy) and inappropriate prescription are frequently reported problems [4
]. While the rate of potentially inappropriate medication use is about 20% in community settings [7
], almost half of nursing home residents are exposed to potentially inappropriate medication [9
]. For these reasons, there has been a focus on the quality of medication management in this vulnerable population [10
Inadequate care of the elderly is not only a safety concern but also a factor behind excessive healthcare expenditure [12
]. Indeed, drug-related morbidity and mortality in nursing homes poses a great economic burden [13
]. A previous study reported that the average expenditure per elderly person for nursing home residents was five times higher than for community residents, and that the overall national expenditure was three times higher [14
]. As a possible response to this issue, incorporating pharmacist input into medication management has been suggested and implemented. It is well known that pharmacist involvement can improve medication management for the elderly through reducing inappropriate prescriptions [15
]. Furthermore, in both U.S. and Europe ensuring the effective use of medication therapy and reducing the unnecessary costs caused by drug-related problems has generated huge cost-savings.
Incorporating pharmacist activities into medication management, however, demands a significant time commitment from the pharmacist and utilizes healthcare resources. Regardless of the efficacy gains, healthcare policy-makers or other stakeholders frequently require evidence supporting the cost-effectiveness of this approach [17
]. Thus, given the rising healthcare costs in this era, it is important to explore the cost effectiveness of pharmacist-participated medication management in nursing homes. Yet, compared to the extensive studies carried out measuring the safety or economic impact of pharmacist participation with the community-dwelling elderly, limited effort has been applied to studying the economic outcomes in nursing homes. This leaves the economic impact of mandating pharmacist activities of various participation types in nursing homes questionable, despite their promising potential. Thus, the main purpose of this review was to examine the studies undertaken of the economic outcomes of various types of pharmacist medication management participation for nursing home residents.
2. Materials and Methods
2.1. Search Strategy
The search strategy of this study was designed to retrieve and produce an unbiased analysis of the research into the economic effects of pharmacist-participated medication management in nursing homes. Databases (PubMed, EMBASE and Ovid Medline) covering articles published in English between January 1990 and July 2017 were searched using appropriate search terms adapted for each bibliographic source. These databases were searched for the MeSH or text keywords “nursing home”, “long term care facility”, “pharmacist”, “pharmacy”, “pharmaceutical care”, “interdisciplinary”, “costs”, “cost analysis”, “economics”, and related words in different combinations. The subject headings used as search terms were split into four categories: those related to pharmacy (“pharmacy” OR “pharmacist” OR “pharmaceutical care”); AND those related to pharmacist participation (“interdisciplinary” OR “multidisciplinary”); AND those related to cost (“economics” OR “cost” OR “cost analysis”); AND those related to nursing home (“nursing home” OR “long term care” OR “aged facility”). The search terms were used in title, abstract, keywords searches. There were no restrictions regarding the study design or method of economic analysis, reflecting the need to be over-inclusive in an attempt to be complete.
2.2. Study Selection
All the references retrieved by the literature search were screened by two independent reviewers (the first and second authors) based on the title and abstract of each citation. The articles retrieved could be eliminated if their titles clearly indicated that they did not focus on the economic outcomes of pharmacist activities in nursing homes. Next, the remaining abstracts were examined to identify papers that measured the cost or cost-effectiveness of pharmacist-participated medication management for elderly in nursing homes. The articles to be included in the study were finalized after a full review of each paper. The inclusion criterion was whether the full-text of the articles was primarily focused on economic analysis of pharmacist participation in nursing homes. To access the full-text of the articles, the current authors utilized their institutions’ electronic resources and tried to contact the original authors via e-mail. The following types of articles were excluded from the review: those featuring home or residential nursing settings lacking care services provided by registered healthcare professionals, such as nurses on duty at all times, those that had residents under 65 years old, those that were about the nursing home system itself rather than the residents within, or those that included interventions not carried out by pharmacists (e.g., by nurses or other healthcare professionals or multidisciplinary teams without a pharmacist). The full-text review also excluded articles measuring only clinical outcomes or those based on the same study. Throughout the selection process, any disagreement was resolved either by discussion among the reviewers or by considering the opinion of additional reviewers (the corresponding author and the other two co-authors) to reach a consensus.
2.3. Quality Assessment
The two independent reviewers (the first and second authors) assessed the quality of the selected articles using the “Quality Assessment Tool For Quantitative Studies” developed by the Effective Public Health Practice Project (EPHPP), which, given its content and construct validity, has been judged suitable for use in systematic reviews of effectiveness [18
]. The scoring was based on the objective guidelines provided by the EPHPP. Eight domains were assessed: selection bias, study design, confounders, blinding, the data collection method, withdrawals/dropouts, intervention integrity, and analysis. Following the guidelines of the tool, for each article, every domain except intervention integrity and analysis was rated as either strong (three points), moderate (two points), or weak (one point), and then these scores were summed to provide the total score. Based on their total score, the studies were assigned a quality rating of weak (two or more weak ratings), moderate (one weak rating), or strong (no weak ratings) for their overall methodology.
2.4. Data Extraction and Analysis
The two independent reviewers (the first and second authors) extracted data from each article including the study design, characteristics of the enrolled patients, type of activity, type of economic analysis, and the outcome measures of interest. Any disagreement was resolved either by discussion among the reviewers or by considering the opinion of additional reviewers (the corresponding author and other two co-authors) to reach consensus. Because of the high degree of heterogeneity in terms of study design, activity type, and economic outcome measures, a meta-analysis could not be performed. Instead, the results are presented as a narrative review. To synthesize the various participation types, the studies were categorized according to the main activity type based on the level of pharmacist participation: interprofessional networks, interprofessional coordination, or interprofessional teamwork. These terms are based on the glossaries published in previous studies [20
]. That is, the activities were categorized as interprofessional networks if a pharmacist’s main activities were limited to attending multidisciplinary case conferencing on a periodic basis. They were categorized as interprofessional coordination if they included mainly pharmacist-led medication reviews in which pharmacists provided advice to prescribers, yet left the implementation of such advice to other practitioners. Finally, they were considered interprofessional teamwork if they were cases of interventions through regular team care in which activities to solve problems and deliver services were carried out collaboratively and in an integrated and interdependent manner rather than being solely pharmacist-driven. Note that the monetary values in the tables are presented as they were in the original papers.
As the elderly population increases globally, in many developed countries the number of elderly who require long-term care services in nursing homes as well as the prices of their medicines have been steadily increasing. The rising health care expenditure of this age group is placing upward pressure not only on per capita expenditures but also on total national medical expenses [35
]. This review summarizes the studies that measured the economic value of pharmacist participation in the medication management of the elderly living in nursing homes, and identifies a body of evidence indicating positive results in terms of cost-related outcomes. Recent systematic reviews have evaluated the economic outcomes of pharmacist interventions for community-dwelling elderly [36
]. Others have evaluated care facilities for the aged including not only nursing homes but also the residential care or mixed care homes that do not provide round-the-clock care services from registered healthcare professionals such as nurses [38
]. However, to our best knowledge, this study is the first systematic review of the economic impact of pharmacist participation focused on nursing homes, that is, facilities licensed to provide personal care and skilled nursing care to residents on a 24-h-a-day basis.
In this review, the included studies were categorized into three groups according to the level of pharmacist participation. All the articles, regardless of the level of pharmacist participation studied, reported that the pharmacist participation in medication management led to positive outcomes for medication or treatment cost per resident. Of the eight studies performing statistical comparison analyses [24
], one study of interprofessional coordination [24
] and three of interprofessional teamwork [27
] reported their statistical significance. It is notable that the two studies of interprofessional networks showed only non-significant outcomes despite both being rated of strong methodologic quality according to the EPHPP checklist while the studies of interprofessional coordination or interprofessional teamwork did show significant outcomes in economic terms. Yet, the studies of interprofessional networks in our review had limitations; particularly, the study durations were too short for the teams to reach peak performance and prove maximal effects. In addition, the limited sample size and risk of selection bias arising from how the cases were selected by facility staff may have affected the non-significant outcome. Thus, it would be inappropriate to conclude that the interprofessional network level of pharmacist participation is ineffective, and more correct to say that this level has the potential to prove cost-effective if the elements of appropriate study design are fulfilled.
The results of this review deserve attention because previous studies have shown that pharmacists working in isolation and in a fragmented manner have limited impact on cost effectiveness [39
]. Previous studies have reported that the essential elements for effective multidisciplinary care are regular conferences [40
], the participation of all professionals with increased interaction [41
], and the active engagement and incorporation of different views when making patient care decisions [42
]. In this regard, from this review, we can consider the essential prerequisites for pharmacist participation in medication management leading to economic benefit to be activities with structured models, communication, and sufficient time for building working relationships.
The key findings of the review follow. First, team care is facilitated by a structured model with shared goals and clear roles. The studies with significant positive economic outcomes in this review were based on models such as a polypharmacy initiative as a collaborative program [27
] and a wound treatment program [30
]. Despite including different content and personnel, the models specified the role of each healthcare provider involved in service delivery as well as guidelines for the shared goals and missions of the overall team care service. Other studies have similarly suggested the importance of sharing common goals and clarifying roles to promote successful collaboration [43
]. Second, the team members involved in the collaboration should have high levels of communication. Through this review, we found that cost-effective interventions offered interdisciplinary training and education that aimed to improve communication. Direct communication such as face-to-face and phone interaction is useful for identifying drug-related problems and improving physicians’ acceptance rates of pharmacist recommendations [43
]. Effective communication can also improve working and personal relationships [46
]. By contrast, the impact of pharmacist activities may be minimal when communication and/or good working relationships are absent [47
]. Third, team members need sufficient time to get to know each other to build trust and professional respect [43
]. This finding is further supported by McDonough and Doucette’s five-stage collaborative working relationships model, which is a framework that helps pharmacists develop working relationships with physicians through a series of stages [48
]. In this model, the pre-relationship stage (stage 0) starts with professional awareness, after which pharmacists progress to professional recognition (stage 1). Stage 2 is exploration and trial, stage 3 is the expansion of professional relationships, and stage 4 is commitment to the collaboration. Since this model is composed of communication opportunities, sufficient time to take a step-wise approach, and the establishment of clearly defined roles and shared responsibilities, it can be applied practically to reduce costs.
Nevertheless, the findings of this study should be interpreted cautiously due to limitations. When the economic components of the studies were reviewed, few articles met the ideal requirements of the cost analysis presented in Zermansky and Silcock [49
]: a cost-benefit or cost-utility analysis over a period of at least one year from a societal perspective. With respect to the economic components, only a few studies met the ideal type of economic evaluation, namely a cost-benefit or a cost-utility analysis. Cost-benefit analyses place monetary values on both inputs (costs) and outcomes, thereby allowing a comparison of interventions across the entire economy. Unlike alternative economic evaluation models, this analytic method can indicate the desirability of an intervention independent of a comparison to alternatives. Cost-utility analyses use a non-financial common metric that allows comparisons across the health sector, such as QALY, and thus can compare different drugs or technologies. Instead, most studies presented cost savings by comparing the costs incurred with those saved, which might inadequately show the quality of patient care or fail to highlight better resource use by pharmacists. With respect to the timeline of the intervention, the durations of six studies in this review were shorter than the one year requirement suggested by Zermansky and Silcock [49
], and therefore may have underestimated the economic outcomes.