gms | German Medical Science

Background: Drug therapy is increasingly complex and challenging due to polypharmacy, complexity of drug therapy, demanding drug application techniques and schedules [1], [2], [3], [4]. Furthermore, verbal information is poorly remembered by patients [5], [6], thus increasing the need for written medication lists.

We aimed to investigate the prevalence, characteristics, and quality of medication lists of elderly ambulatory patients, with a particular emphasis on the origin. Moreover, a brochure with information on correct drug application along with an empty medication list template was randomly sent to half of the study participants in advance to test the effect on the number of medication lists by comparing intervention and non-intervention group.

Materials and Methods: Data originate from the fourth follow-up of the ESTHER study (2011-2013) [7]. During a home visit, trained study physicians scanned all existing medication lists, recorded the patients´ medication electronically, and performed thorough geriatric assessments (e.g. Mini Mental State Examination, Barthel Index, and frailty evaluation). Information on diagnoses and self-rated adherence was collected. Medication lists were evaluated according to predefined parameters using dichotomous variables, thus providing information on allocation to the patient, date of issue, specification of the drug (name, dosage, indication), further information (abbreviations, Over-the-counter medication, usage of the Pharmazentralnummer (PZN)), general comments and corrections. Complexity was determined in terms of frequency, amount of drug intake and demanding drug application techniques [3]. Medication lists were considered to be either issued by a health care professional (HCP; physicians, hospitals, or pharmacies) or the patients themselves or relatives.

We compared the characteristics of the participants with or without a medication list, and evaluated which patients were more likely to receive a medication list from a HCP. The effect of the brochure was evaluated by comparing the number of actually found medication lists in intervention and non-intervention group.

Categorical data were compared using chi-square test and metric data were compared by Wilcoxon-Rank sum test. Forward selection logistic regression (entry significance level p<0.2) was used to identify factors with impact on possessing a medication list as well as self-issuing a medication list. The statistical analysis was conducted with the SAS statistical software package, version 9.3 (SAS Institute Inc., Cary, USA). A p-value <0.05 was considered significant.

Results: In total, 2,470 of 2,761 participants with a home visit were taking at least one medication and were therefore analyzed (89.5%). Altogether, after exclusion of 15 medication lists due to illegibility or no recorded medication data at the home visit, 553 medication lists were found (22.4% of the included patients). Of them, 353 were issued by HCPs (63.8%). Medication lists originating from HCPs had a higher quality (e.g. containing information on the patient, date of issue, and dosages), but still had some shortcomings (e.g. less OTC-medication listed, more corrections). Factors associated with possession of a medication list were male sex, the number of medicines taken, the complexity of drug regimen, and a diagnosis of cancer. Younger age, multimorbidity, a diagnosis of Parkinson’s disease, heart failure, coronary heart disease, pre-frailty, or frailty were factors associated with self-issuing a medication list. Patients with a self-written medication list had a higher self-rated adherence than patients with a medication list from a HCP (p<0.05). The intervention did not have an effect on the number of medication lists found.

Conclusion: Only a minority of elderly ambulatory patients had a medication list, leaving space for improvement. Patients at particular risk were female, took more medications, had a more complex drug regimen and a diagnosis of cancer. Sending a brochure with an empty medication list template did not increase the proportion of medication lists.