Medication use in patients with chronic kidney disease

The incidence and prevalence of chronic kidney disease (CKD) has escalated in the past decade due to an ageing population, increased prevalence of diabetes and cardiovascular disease. This presents diverse challenges to the healthcare system. One such challenge relates to medication prescribing practices for patients with CKD. The process of selecting, prescribing and maintaining the correct drug dosing is challenging, partly because of CKD-induced changes in drug pharmacokinetics and pharmacodynamics. Further, there is high occurrence of chronic comorbidities and resultant multiple drug prescribing in patients with CKD. Providing optimal care to patients with CKD is an area of concern for health care professionals as there is evidence of suboptimal prescribing, particularly in older people. Age-related heterogeneity coupled with overall decline in bodily function puts older people at higher risk of drug toxicity. Identifying ways for optimising prescribing and minimising harm in this vulnerable population is increasingly a priority for health care providers and policy makers. The overall aim of this thesis was to determine how to improve use of high-risk medications in patients with CKD in community settings. Five connected studies were conducted to address the overall aim and quantify the extent of prevalence of inappropriate prescribing of renally-cleared drugs and explore the factors associated with it. We also aimed to identify the system-level and practice-level confounding factors that serve as barriers to implementing evidence-based guidelines in CKD care and to determine the intervention strategies most likely to have an effect in overcoming the observed barriers. The findings of this thesis provide a basis for designing intervention programs to optimise prescribing in patients with CKD in community settings. The first part of this thesis quantified the extent of the prevalence of inappropriate prescribing of renally-cleared drugs in elderly patients living in the Australian community and aged care settings, and determined the factors associated with inappropriate prescribing. A key observation from this study was that, despite the existence of published guidelines for dosage adjustments based on renal function, aged care residents and community-dwelling older people are often prescribed renally-cleared medicines that are either contraindicated or above the recommended dosage. Over one-quarter (n=1135 out of 4035; 28.1%) of the patients prescribed the renally-cleared drugs examined had evidence of inappropriate prescribing of at least one of the drugs, based on renal function. The drugs/drug class most commonly prescribed inappropriately were perindopril, fenofibrate, glibenclamide, gliptins, metformin, olmesartan, bisphosphonates and strontium. The factors independently associated with patients being prescribed one or more potentially inappropriate renally-cleared drugs were; advancing age (odds ratio (OR)=1.06 per year increase, 95% confidence interval [CI] 1.05-1.07, P<0.001), the total number of renally-cleared drugs prescribed (OR=1.44 per unit increase, 95% CI 1.29-1.61, P<0.001), presence of diabetes (OR=1.51, 95% CI 1.30-1.76, P<0.001), presence of heart failure (OR=1.38, 95% CI 1.13-1.69, P<0.005) and living in aged care facilities (OR=1.28, 95% CI 1.06-1.55, P<0.05). The second part of this thesis explored the factors contributing to inappropriate prescribing in detail. The asymptomatic nature and opportunistic diagnosis of CKD are considered to be some of the major reasons for the higher prevalence of inappropriate prescribing. Other contributing factors reported include prescribers' lack of knowledge of medications requiring dosage adjustment, the presence of renal impairment being overlooked by prescribers, underestimation of potential adverse events, and the lack of evidence-based data to guide prescribers on precautions and dosage adjustments. Moreover, a lack of quantitative data in the available drug information sources, and contradiction and inconsistency in dosing information may augment the problem of dosing error. Based on this background, we evaluated five standard drug information sources (Australian Medicines Handbook, British National Formulary, American Hospital Formulary System, Monthly Index of Medical Specialties, Drug Prescribing in Renal Failure) for the availability and concordance of renal dosing recommendations for 61 drugs recommended to be used with caution in renal impairment. We observed a lack of consistency among the sources regarding the definition and categorisation of CKD. Only a slight agreement was observed in the renal dosing recommendations among the sources (Fleiss Kappa: 0.3). Qualitative data were not well defined, and there was a lack of consistency in quantitative values. Some drugs marked as contraindicated in one source were not mentioned as such in others. Also, drugs considered as not requiring dosage adjustment in one source had explicit recommendations in other sources. We concluded that a lack of data to guide the prescribers on dosage adjustments and lack of consistency among the standard information sources could potentially contribute to inappropriate prescribing. In the third part of the thesis, we investigated the extent to which renal dosing information was available in the manufacturer's product information (PI) and determined the concordance in renal dosing recommendations across the 155 brands of 27 drugs. For each brand, the PI was consulted and data referring to renal impairment was collated. The renal dosing recommendations varied significantly among different brands of the same drugs. Also, there was inconsistency in use of CKD terminology and classification of renal impairment. There was generally a lack of detailed information in the PI regarding the use of drugs in patients with renal impairment. The majority of PI documents (88 of 155 PI; 57%) provided quantitative recommendations regarding dosage adjustments for renal impairment, but this was often not detailed enough to help users make an informed decision. For 37 PI documents (24%), an altered dosage regimen was proposed without a quantifiable measure of renal function reported in the dose recommendation. The results of pharmacokinetic studies in patients with renal impairment were not presented in 59 PI documents (38%). Twenty brands did not have full PI and advice such as contact the manufacturer‚ÄövÑvp was given. Four PIs mentioned the lack of data regarding use in patients with renal impairment and thus recommended avoiding the drug in renally impaired patients. We concluded that a lack of data to guide the prescribers on precautions and dosage adjustments and lack of consistency among the PIs could potentially contribute to inappropriate prescribing. In the fourth part of the thesis, we determined the agreement among the Cockcroft-Gault, the Modification of Diet in Renal Disease (MDRD) and the CKD-Epidemiology Collaboration (CKD-EPI) equations, if hypothetically used in dosing of renally-cleared drugs in primary care settings. There was a significant difference in the kidney function values estimated from the three equations. There was a good overall agreement among doses rendered using the equations. Dosing based on either CrCl or an eGFR with body surface area normalisation removed appeared acceptable and practicable for the purpose of dosing of non-critical drugs in the primary care settings. However, it is worth noting that in some instances there were potentially important discrepancies among the doses rendered from the equations so caution should be exercised. In the fifth part of the thesis, we conducted a survey of general practitioners (GPs) to determine what sort of implementable strategies are most likely to have an effect on GPs to improve renal dosing in patients with CKD. There was a low familiarity with CKD management guidelines and a general lack of confidence in identifying and performing dosage adjustment of renally-cleared drugs among the GPs. There was a general scepticism concerning the usefulness, reliability and applicability of the information sources for renal dosing. The major barriers to using guidelines were lack of easy access during consultations and ambiguous recommendations on renal dosing between the different sources. The factors responsible for inappropriate prescribing from the GPs' perspective were a lack of awareness on the availability of information sources and a lack of practice of routine monitoring of renal function. The most favoured interventions were decision support systems, online education training and medication reviews by pharmacists. In the last part of the thesis, we also conducted a systematic review to explore the nature and types of interventions conducted by pharmacists for patients with CKD and the outcomes. Pharmacist-mediated drug use evaluation and monitoring appeared promising in decreasing the rate of over dosing, usage of unnecessary drugs and improving physician adherence to dosing guidelines. The high level of acceptance of pharmacists' recommendations by the physicians indicates the greater opportunity for pharmacists to be involved in CKD care. In conclusion, optimising prescribing in patients with CKD requires accurate identification of CKD in clinical settings and individualisation of medication prescribing based on the patient's renal function, pharmacokinetic parameters and goals of care. Updating the information sources to present the key elements in an unambiguous format, in conjunction with efforts to build consensus among the standard information sources, seems necessary. As a result, GPs can easily incorporate the recommendations into daily practice. Regular updating of the content of drug information sources is also warranted. An emphasis should be placed on conducting and disseminating large population-based studies focused on determining the correct drug dosage based on renal function....