Pharmacist participation in the infectious disease clinical team at a Spanish hospital increased the number of drug-related problems identified and the number of interventions with positive outcomes
N VÃctor JimÃ©nez-Torres*â€
Head of Service
Dr Peset University Hospital
â€ Department of Pharmacy and Pharmaceutical Technology
University of Valencia
Traditionally, hospital pharmacists have developed their own functions centred on the pharmacy department. While this system lets pharmacists target a large number of patients and optimise available resources, it does have disadvantages: there may be limited knowledge about patientsâ€™ conditions, lower participation in pharmacotherapeutic and clinical decision-making, and communication difficulties. In a bid to overcome these barriers, pharmacists have become frontline healthcare providers with more patient contact. They now work within the interdisciplinary team focusing on disease management rather than drug therapy alone.
In our institution ï¿½” a 600-bed academic, public and general hospital ï¿½” pharmaceutical care is nowadays implemented at different quality levels based on patient needs.
At the basic level, care is provided through checking manual or electronic prescriptions. At the next level, more patients are monitored using predefined alerts based on risk of drug-related morbidity. At the third level, patient follow-up includes the entire pharmacotherapeutic treatment.
The topmost level requires a higher degree of integration of pharmacists in the healthcare team, using other tools in addition to prescription checking to identify patients with actual or potential drug-related problems (DRPs).
Such tools include clinical chart review, interview and voluntary reporting by patients or health professionals. Care at this level is provided to patients admitted to the internal medicine, haematology, oncology, orthopaedics and infectious diseases units.
The resource implications for the different levels of service vary widely so it is important to know whether the outcomes differ.
We undertook a study to evaluate the impact of the quality of pharmaceutical care offered to patients admitted to the infectious diseases unit (IDU).
We compared two service models: centralised pharmacotherapeutic monitoring (model C) versus decentralised pharmacotherapeutic monitoring (model D), with participation of the clinical pharmacist in daily rounds and patient follow-up.
This prospective cohort study was conducted in the IDU at our hospital between February 2005 and January 2006. It was divided into two periods of five (model C) and seven (model D) months, respectively. All DRPs identified and pharmacist interventions (PhIs) occurring per period were recorded in AtefarmÂ® 2006.0.1.2. (IMF, Spain, 2006). Data collected comprised number of patients and patient-days, number of identified DRPs and their categories and potential severity (on a scale of 1ï¿½”5), number of medication errors (MEs) intercepted and stage in drug-use process, number of PhIs, acceptance, type and patient outcomes (also on a scale of 1ï¿½”5). Incidence or risk ratios with 95% confidence intervals (CI) were calculated using SPSS for Windows v.15.0 (SPSS Inc, 1989ï¿½”2006) to compare the two models.
Results and discussion
The number of patients admitted to the IDU was 413 for model C and 467 for model D. The numbers of patient-days were 3,467 and 3,502, respectively. Table 1 shows the results obtained for the indicators specified.
Previous studies have demonstrated the benefits of pharmacists working as full members of healthcare teams in different clinical settings.[2ï¿½”9] Despite some limitations due to methodological deficiencies, these studies commonly concluded that active pharmacist participation in the interdisciplinary team was associated with a reduction in the rate and duration of MEs and preventable drug-related morbidity. In our study, the incidence of DRPs identified was doubled in the decentralised model (compared to the centralised model). There were, therefore, many more patients in this group whose pharmacotherapy could be optimised and in whom drug-related morbidity could be prevented. A similar increase was seen in the rate of DRPs identified with potential severity equal to or greater than three, which indicates that they could have caused reversible harm to patients, required additional treatment or increased length of stay. These data suggest that pharmacists working actively in a healthcare team should not only focus on drug therapy problems with major clinical relevance, as they usually do in a centralised system, but also that they should address patientsâ€™ other drug-related needs ï¿½” such as reconciling chronic medication, and mild adverse drug reactions ï¿½” that could have been missed in the centralised model. The largest increases were seen in the rates of DRPs concerned with patientsâ€™ safety and drug therapy indications. The incidence of problems related to effectiveness of therapy was similar in the two models. Differences in definitions and categorisation of DRPs made it difficult to compare our data with those of other studies.
One of the most interesting findings of our study is that the rate of MEs intercepted before they affected patients was significantly increased ï¿½” by 90% ï¿½” when there was pharmacist participation in the healthcare team. Regarding sequential stages in the drug-use process when MEs took place, prescribing, transcription and follow-up errors increased with the decentralised model. This is consistent with the findings of other studies that also reported an increase in MEs intercepted, especially those that occurred during prescribing, probably due to pharmacistsâ€™ assessment of medication use with a more extensive knowledge of the patientâ€™s medical history, drug contraindications, organ function, and the reason for prescribing a drug.[2ï¿½”7]
Changing drug dose or interval and stopping drugs are the PhIs that are more frequently performed in the decentralised model, compared to initiating drugs or drug-level monitoring, which are most common in the centralised model. Changes in the distribution of types of PhI correlated with the increase in the numbers of DRPs in safety and indication categories identified. Similar studies also showed that pharmacists conducting pharmacotherapeutic monitoring from the pharmacy department made recommendations regarding sequential therapy or pharmacokinetic monitoring, since a detailed knowledge of patientsâ€™ characteristics and conditions provided more opportunities to participate in designing pharmacotherapeutic plans and follow-up of patients.[3ï¿½”8]
Surprisingly, no statistical differences were noted in percentage of acceptance of PhIs between the two periods studied. In fact, a lower acceptance was obtained in the decentralised model. This was probably due to the higher rate of DRPs related to therapy indication in which pharmacists were involved and their potentially lower clinical relevance. Finally, positive patient outcomes rose by 20%, indicating that the integration of pharmacists in the clinical team allowed improvement in opportunities to optimise drug use and contribute to reducing the risk of drug-related morbidity in patients.
Integrating the pharmacist into the IDU clinical team boosts ability to intercept MEs before they can harm patients, and to identify DRPs, especially those concerning patient safety and drug therapy. Similar gains have been obtained for total and most severe DRPs, probably due to the appropriateness of the centralised system based on pharmacotherapeutic alerts to detect patients at high risk of drug therapy problems. In the decentralised model, interventions with positive outcomes reducing or preventing drug-related morbidity rose 20%. â–
1. Climente M, JimÃ©nez NV. Manual para la atenciÃ³n farmacÃ©utica. Valencia: AFAHPE; 2005.
2. Leape LL, et al. JAMA 1999;281:267-70.
3. Scarsi KK, et al. Am J Health-Syst Pharm 2002;59:2089-92.
4. KÃ¼Ã§Ã¼karslan SN, et al. Arch Intern Med 2003;163:2014-8.
5. Izco N, et al. Farm Hosp 2002;26:18-27.
6. Arroyo C, et al. Farm Hosp 2006;30:284-90.
7. Castillo I, et al. Farm Hosp 2000;24:27-31.
8. GonzÃ¡lez MA, et al. Aten Farm 2002;61:384-96.
9. Yee DK, et al. Am J Health-Syst Pharm 1997;54:670-3.