Introduction

Drug information centers (DICs) are defined as “institution[s] dedicated to provide objective, independent and current information on drugs and their use, and communicate to the different categories of users for better understanding and benefit of patients”1. DICs play a key role in the dissemination of drug information (DI) to healthcare professionals all over the globe, thus promoting the rational use of drugs1,2. The first DIC was inaugurated at the University of Kentucky in the United States in 19623,4, and numerous DICs followed in the United States and Europe throughout the 1960s to the 1980s1,4,5,6. Today, DICs are present in most countries worldwide1, also in emerging countries such as India7,8,9,10,11, Nepal12, Brazil13, Uganda14, or Ethiopia15,16,17.

Staff composition and modus operandi of DICs differ markedly between countries. In the United States, for example, DICs are typically led by pharmacists and responses to DI queries are generally provided by telephone18,19. European DICs are frequently headed by a clinical pharmacologist or similarly qualified physician, and are run jointly by physicians (usually clinical pharmacologists) and pharmacists5,6,19,20. However, there also exist numerous pharmacist-led DICs in Europe. Owing to their affiliation with university hospitals, European DICs are often assisted by visiting physicians and medical students21. European DICs generally provide written responses in addition to verbal communication with inquirers21.

While several reports from emerging countries7,8,9,10,11,13,14,15,16,17 and Scandinavia5,19,22,23,24,25 have been published in recent years, there is a considerable lack of studies on DICs from central Europe, in particular from Germany. To the best of our knowledge, the most recent data from Germany on clinical pharmacologist-led DICs which provide DI to other healthcare professionals stem from Tröger and Meyer from 200026 and from Schwarz and colleagues from 199927. To close this information gap of more than 20 years, we analyzed all queries submitted to the DIC operated by the Institute for Clinical Pharmacology of Hannover Medical School between October 2018 and April 2022. We compared the frequency and characteristics of DI queries with previous studies from Germany and other countries and investigated potential differences between patient-specific and general queries. In addition, we present and comment on an illustrative and educational selection of clinically complex queries answered by our clinical-pharmacological DIC during the study period.

Methods

Ethics approval

This study adheres to the Declaration of Helsinki (1964) and its later amendments (latest version 2013). Due to anonymous data analysis, the Ethics Committee of Hannover Medical School waived a formal ethics vote.

Modus operandi of the clinical-pharmacological drug information center of Hannover Medical School

The clinical-pharmacological DIC of Hannover Medical School was established at the Institute for Clinical Pharmacology in 1994. As a specialist pharmacotherapeutic consultation service, the clinical-pharmacological DIC is available to physicians and other healthcare professionals (e.g., dentists, nurses) working at Hannover Medical School or affiliated academic teaching hospitals free of charge. Occasionally, the clinical-pharmacological DIC also answers queries from physicians working in private practice, for example, if they are the ambulatory physicians of patients who were treated at Hannover Medical School. Of note, the hospital pharmacy of Hannover Medical School provides its own drug information service, which operates independently from the clinical-pharmacological DIC and which is specialized on pharmaceutical questions. A Computerized Provider Order Entry and Clinical Decision Support System (CPOE-CDSS) is not routinely available at Hannover Medical School.

The modus operandi of the clinical-pharmacological DIC is laid down in standard operating procedures (SOPs). The clinical-pharmacological DIC is operated by two senior physicians (one clinical pharmacologist (DOS) and one internist (CS)), one resident in clinical pharmacology (JH), one clinical pharmacist with specialty training in drug information (according to the curriculum of the German Federal Chamber of Pharmacists (Fachapothekerin für Arzneimittelinformation gemäß Curriculum der Bundesapothekerkammer); RS), and one secretary. Queries are submitted to the clinical-pharmacological DIC by healthcare professionals in written form via the hospital information system SAP (Systems, Applications, and Products in Data Processing, Walldorf, Germany) or via e-mail. Queries are attended to by JH or RS, with occasional support from visiting physicians or medical students. If relevant clinical information is missing (e.g., information about the patient’s medication or comorbidities), the inquiring healthcare professional is contacted via telephone or e-mail. Subsequently, a comprehensive literature review is conducted, taking advantage of the following references (non-exhaustive list): summaries of product characteristics (SmPCs), medical databases (e.g., PubMed, Cochrane Library), current editions of standard textbooks of (clinical) pharmacology (e.g., Stockley’s Drug Interactions, Goodman & Gilman’s The Pharmacological Basis of Therapeutics, The Renal Drug Handbook), and drug information and interaction programs (e.g., UpToDate (Wolters Kluwer N.V., Alphen aan den Rijn, The Netherlands), AiDKlinik (Dosing GmbH, Heidelberg, Germany), mediQ (Psychiatrische Dienste Aargau AG, mediQ Kompetenzzentrum für Medikamentensicherheit, Windisch, Switzerland)). A written response is drafted, and it is discussed with and countersigned by a senior physician (DOS or CS). The finalized response is sent to the inquiring healthcare professional via SAP or e-mail within one to three workdays. In urgent cases, a preliminary answer is communicated by telephone.

All queries and corresponding answers are saved in a password-protected Microsoft Access 2016 database (Redmond, Washington, USA) which is only accessible to registered users, i.e., staff of the clinical-pharmacological DIC. In addition, paper-based documents, such as printed scientific articles which were used to answer queries, handwritten notes, protocols of telephone communications, answer drafts, and so forth, are stored in folders which are kept in locked rooms in locked cupboards (separate sets of keys for rooms and cupboards). With the exception of age and sex, any patient-related data is deleted prior to saving the queries and corresponding answers in the electronic database and in the paper-based documentation.

All queries and corresponding answers are presented and discussed at weekly pharmacotherapeutic colloquia. These colloquia serve as additional internal quality assurance and are accredited by the Medical Association of Lower Saxony (Ärztekammer Niedersachsen) as continuing medical education (CME) seminars for physicians. Besides, medical students in the practical year (Praktisches Jahr—i.e., the final year of the medical curriculum in Germany) are encouraged to participate in pharmacotherapeutic colloquia as part of their medical training at Hannover Medical School.

Of note, the clinical-pharmacological DIC is not the only service provided by the Institute for Clinical Pharmacology. Other tasks comprise pharmacovigilance in clinical trials, teaching of medical students, interdisciplinary ward rounds, clinical research, and the office of the Drug Commissioner of Hannover Medical School.

Data acquisition

For the purpose of this study, all queries submitted to the clinical-pharmacological DIC between October 2018 and April 2022 were analyzed. For each query, the following parameters were assessed:

  • origin of the query: Hannover Medical School; academic teaching hospital; private practice; other

  • type of query: patient-specific (i.e., queries referring to individual patients) or general (i.e., queries of broader pharmacological interest or queries referring to larger patient populations)

  • age and sex of the patient (only for patient-specific queries)

  • the medical specialty of the inquiring healthcare professional

  • query categories (assignment of more than one category per query was possible): adverse drug reaction (ADR); indication/contraindication; posology/dose adjustment (e.g., due to renal or hepatic insufficiency); therapeutic drug monitoring (TDM); pharmacogenetics; pharmacodynamic interaction (PDI); pharmacokinetic interaction (PKI); pregnancy and breastfeeding; pharmacotherapy in the elderly (i.e., patients ≥ 65 years of age); other.

Of note, the category “pharmacotherapy in the elderly” was not automatically assigned to all queries concerning patients of chronological age of ≥ 65 years, but only to queries explicitly inquiring about drug specifics in advanced age.

Statistical analysis

Descriptive statistical techniques were used to summarize the data. Quantitative variables are depicted as means ± standard deviations (SDs) and additionally as medians with interquartile ranges (IQRs) for data that were not normally distributed. For categorical variables, absolute and relative frequencies are presented. Differences between general and patient-specific queries regarding involved query categories were analyzed with Pearson’s Chi-squared test or Fisher’s exact test (the latter was applied if any of the four cells of a 2 × 2 table had less than five observations). P values < 0.05 were considered statistically significant. All statistical analyses were conducted with Microsoft Excel 2019 (Redmond, Washington, USA) and IBM SPSS Statistics for Windows, version 28 (Armonk, New York, USA).

Results

Number, type, and origin of queries submitted to the clinical-pharmacological drug information center

In total, 594 queries were submitted to the clinical-pharmacological DIC during the study period (i.e., 43 months). Of these, 82.8% (492/594) were patient-specific queries, while the remaining 17.2% (102/594) were general queries (Table 1). The vast majority of queries were submitted by physicians (96.1%; 571/594), and more than three-quarters of queries (75.6%; 449/594) came from healthcare professionals working at Hannover Medical School. On an aggregated level (i.e., without differentiation between subdisciplines of internal medicine, surgery, etc.), approximately one in three queries (31.1%; 185/594) came from internists. If subdisciplines of internal medicine, surgery, etc. were evaluated separately, psychiatrists, urologists, and trauma surgeons most frequently consulted the clinical-pharmacological DIC, submitting 10.6% (63/594), 9.3% (55/594), and 8.4% (50/594) of all queries, respectively (Supplementary Table 1).

Table 1 Characteristics of queries submitted to the clinical-pharmacological drug information center of Hannover Medical School between October 2018 and April 2022 (n = 594).
Full size table

Patient characteristics

Age and sex of patients were documented in 73.2% (360/492) and 92.3% (454/492) of patient-specific queries, respectively. The mean age of patients in patient-specific queries was 55.6 ± 22.9 years (median 60 years, IQR 37 to 72 years, range 3 days to 100 years). 51.4% (253/492) of the patients were female (male: 40.9% (201/492); sex not documented: 7.7% (38/492)). As displayed in Fig. 1, patient-specific queries were most frequently related to patients aged between 61 and 70 years (13.4% (66/492) of patient-specific queries).

Figure 1
figure 1

Age distribution in patient-specific queries (n = 492) submitted to the clinical-pharmacological drug information center of Hannover Medical School between October 2018 and April 2022. ND denotes not documented.

Full size image

Query categories

ADRs, indications/contraindications, and PDIs represented the three most frequent query categories, being involved in 44.8% (266/594), 43.3% (257/594), and 34.3% (204/594) of all queries, respectively (Table 2).

Table 2 Absolute and relative frequencies of query categories involved in queries submitted to the clinical-pharmacological drug information center of Hannover Medical School between October 2018 and April 2022. Assignment of more than one category per query was possible.
Full size table

Absolute and relative frequencies of query categories, stratified by specializations of inquiring healthcare professionals, are shown in Supplementary Table 2. ADRs were most frequently inquired by physicians working in radiology/radiotherapy (55.0% (11/20) of all queries submitted by physicians from these two disciplines), while indications/contraindications and pharmacogenetics were most frequently referred to by psychiatrists (60.0% (51/85) and 10.6% (9/85) of all queries submitted by psychiatrists, respectively). Queries concerning posology and/or dose adjustments were most commonly submitted by dermatologists (44.4% (8/18) of all queries from dermatologists), whereas TDM was most frequently inquired by pediatricians (22.2% (6/27) of all queries from pediatricians). PDIs were involved in nearly two-thirds (65.5% (36/55)) of all queries from urologists, while PKIs were implicated in almost half (45.8% (11/24)) of all queries from gynecologists. Pharmacotherapy during pregnancy and breastfeeding was most frequently addressed by dermatologists (11.1% (2/18) of all queries from dermatologists), while pharmacotherapy in older people was referred to by approximately every fifth query (20.5% (17/83)) submitted by surgeons.

Comparison between general and patient-specific queries

As compared to general queries, patient-specific queries were statistically significantly more often related to ADRs, PDIs, and PKIs (ADRs: 35.3% vs. 46.7%, P = 0.034; PDIs: 14.7% vs. 38.4%, P < 0.001; PKIs: 20.6% vs. 31.5%, P = 0.028) (Table 2). By contrast, the query category “other” was more often addressed by general queries as compared to patient-specific queries (27.5% vs. 11.4%, P < 0.001).

Examples of clinically complex queries answered by the clinical-pharmacological drug information center

To illustrate the clinical complexity of DI queries submitted to and answered by the clinical-pharmacological DIC, we present and comment on an educational selection of ten queries (Table 3). Half of these cases were related to rare diseases (i.e., cases no. 1, 2, 7, 8, and 10), while one case each referred to toxicology (case no. 3), drug–drug interactions (case no. 4), pharmacotherapy in the elderly (case no. 5), pharmacotherapy during pregnancy (case no. 6), and TDM (case no. 9).

Table 3 Examples of queries answered by the clinical-pharmacological drug information center of Hannover Medical School between October 2018 and April 2022.
Full size table

Discussion

In this study, we analyzed a total of 594 queries submitted to the clinical-pharmacological drug information center of Hannover Medical School. 82.8% of the queries were patient-specific; approximately one in three queries came from internists. However, when subdisciplines were evaluated separately, psychiatrists, urologists, and trauma surgeons represented the healthcare professionals who most frequently consulted the clinical-pharmacological DIC. ADRs, indications/contraindications, and PDIs were the three most frequently inquired query categories. As compared to general queries, patient-specific queries were statistically significantly more often related to ADRs, PDIs, and PKIs.

In their article from 1999, Schwarz and colleagues suggested that “the periodical analysis of the types and sources of enquiries may point to latent problems and needs in the medical community”, a notion we fully agree with. The most recent data from Germany on clinical pharmacologist-led DICs which provide DI to other healthcare professionals stem from the year 200026. Therefore, updating this knowledge was urgently warranted.

A strength of the present analysis is the inclusion of nearly six hundred queries, which represents a case number of similar2,12,15,16,17,25,27,28 or higher magnitude compared to other studies in the field8,9,10,11,14. Furthermore, the structure of the clinical-pharmacological DIC of Hannover Medical School complies with recommendations such as a mixed staff of clinical pharmacologists and clinical pharmacists as core team19 and the involvement of physicians from other disciplines (e.g., internists, geriatricians, and psychiatrists) in terms of close professional cooperation and regular interdisciplinary exchange19,29.

The high degree of specialization at our university hospital may account for the large proportion of patient-specific queries (82.8%) as compared to general queries (17.2%), a finding that stands in contrast with reports by Almazrou et al.30 and Tefera et al.15 from pharmacist-led Saudi Arabian and Ethiopian DICs, respectively, where patient-specific and general queries were more evenly distributed (Almazrou et al. study: 44% patient-specific queries vs. 55% general queries; Tefera et al. study: 53.1% patient-specific queries vs. 46.9% general queries). While answers to general queries usually aim at delivering factual DI, responses to patient-specific queries focus on individualized treatment approaches and provide critical evaluations of the advantages and disadvantages of specific pharmacological therapies in distinct patient contexts, taking the patient’s age, sex, severity of the disease in question, clinically relevant comorbidities, and comedication into consideration. The delivery of factual DI—e.g., information on drug availability, drug stability, drug compatibility, adverse drug reactions, drug interactions, and so forth—is a core competence of pharmacists. Benefit–risk evaluations of pharmacological therapies within the individual patient context, on the other hand, are the mainstay of physicians. These professional differences may explain the discrepancies observed between the Almazrou et al. study30 or the Tefera et al. study15 (both pharmacist-led) and our investigation (physician-led). Of particular importance, it must be highlighted that both pharmacists and physicians play decisive roles in delivering DI and in taking care of pharmacotherapy safety. Pharmacists and physicians should not be regarded as competitors in these fields but—quite the opposite—they should be perceived as complementing each other with their specific expertise and competences.

In the present study, queries submitted by physicians dominated in comparison with queries from pharmacists (96.1% vs. 0.3%), which is in line with previous reports9,14,17,22,25, but in contrast to other studies13,15,16,28,30. The comparatively low proportion of queries submitted by pharmacists in our study may be owing to the fact that pharmacists in Germany have the opportunity to consult other institutions for DI, such as the DI services provided by the Chambers of Pharmacists (Landesapothekerkammern).

In the study by Schwarz et al., the primary users of a regional DIC in Dresden, Germany, were general practitioners, internists, and pediatricians27, as opposed to psychiatrists, urologists, and trauma surgeons in our analysis. However, if the different subdisciplines of internal medicine (e.g., rheumatology, pneumology, cardiology, and so forth) were aggregated, approximately one in three queries submitted to our clinical-pharmacological DIC came from internists. Intriguingly, we received a considerable proportion of queries (i.e., 14.0%) from surgeons, which contrasts with an older report by Tröger and Meyer from Magdeburg, Germany, in which surgical disciplines only played a minor role26. The high proportion of queries submitted to our clinical-pharmacological DIC by surgeons may reflect increasing interdisciplinary cooperation between conservative and surgical disciplines that has developed at Hannover Medical School in recent years. For example, besides operating the clinical-pharmacological DIC, our institute actively participates in the orthogeriatric co-management of older patients, together with trauma surgeons, internists, and geriatricians31.

Reppe and colleagues suggested that “the physical proximity of the [drug information] centers might lower the threshold [for enquirers] to initiate contact”5. Based on the results of our study, we can draw similar conclusions: many queries were received from departments with whom close professional cooperations exist, for example through interdisciplinary ward rounds with colleagues from psychiatry29 or trauma surgery31, and through the participation of our institute in the antimicrobial stewardship (AMS) program at Hannover Medical School32. Of note, personal contacts may lower the threshold to submit queries to the clinical-pharmacological DIC. Notwithstanding, queries from cooperating departments also have to be submitted via the regular route, that is, via e-mail or SAP.

Interestingly, the query category most frequently addressed by healthcare professionals in our study—i.e., ADRs—did not change between 199633 and the current analysis. ADRs were also the query category most frequently addressed by 26.8% (22/82) of European DICs surveyed by Müllerová and Vlček in 199620. The higher proportion of queries dealing with ADRs in our study (44.8%) as compared to other investigations (between 8.7 and 32.5%2,8,9,15,16,17,21,26,27,34) may be explained by the fact that our institute also houses the pharmacovigilance unit of Hannover Medical School and therefore possesses particular expertise in the evaluation of (serious) adverse events. According to a survey conducted by Müllerová and Vlček, only approximately one-third of European DICs are additionally engaged in pharmacovigilance activities20.

Müllerová and Vlček reported that approximately half of European DICs answered between 100 and 1000 queries per year20. Hence, the number of queries answered by our clinical-pharmacological DIC during the study period (i.e., 594 queries in three years and seven months) can be considered as low average. However, it must be taken into account that 43% (35/82) of the European DICs targeted their activities at healthcare professionals and the lay public20, whereas our clinical-pharmacological DIC is restricted to healthcare professionals working at a university hospital and academic teaching hospitals. Therefore, we assume that the complexity of queries submitted to our clinical-pharmacological DIC is considerably higher—on average—compared to queries submitted to the DICs investigated by Müllerová and Vlček20. This is exemplified by a selection of queries and corresponding answers (Table 3), which relate to rare and complex diseases such as mycosis fungoides (case no. 1), insulin autoimmune syndrome (case no. 2), periodic hypothermia (case no. 7), acute porphyria (case no. 8), and visual snow syndrome (case no. 10). In addition, temporal effects may play a role as queries to DICs have become increasingly complex in recent years35. For example, in a survey among DICs in the United States, 70% (46/66) of respondents described an increase in the number of complex DI queries between 2003 and 200818.

To date, only five author groups have presented individual DI queries6,10,11,21,33, but only in abbreviated form6,21, or without provision of corresponding answers10,11. Apart from the Lumpe et al. article33 from 1998, our study is the only report to present and comment on a selection of clinically complex and intriguing DI queries so far, which can be considered a strength of our report.

PDIs (34.3% of all queries; 38.4% of patient-specific queries) and PKIs (29.6% of all queries; 31.5% of patient-specific queries) constituted frequent query categories in our study, in accordance with previous analyses by Alván et al., who stated that “questions concerning drug–drug interactions have increased steadily over the years, in parallel to a general increase in the number of drugs per patient, especially in the elderly”23. Our figures regarding drug–drug interactions were markedly higher than in a recent study by Tefera et al., who reported that only 5.9% of queries submitted to Gondar University Specialized Hospital DIC, Ethiopia, were related to drug interactions15. This discrepancy might be due to different patient populations, as patients in our study were considerably older (mean age 55.6 ± 22.9 years vs. 31.2 ± 20.1 years15) and presumably took more drugs on average, although no data are available to confirm this hypothesis.

Pharmacotherapy in older people is particularly complex as pharmacokinetics (e.g., reduced liver and kidney function) and pharmacodynamics (e.g., increased sensitivity to benzodiazepines) are significantly altered compared to younger individuals36. Approximately 6% of queries in our study were related to pharmacotherapy in older people. Previous studies, by contrast, did not investigate this topic as separate category9,15,16,28,30,34,35. The significant proportion of queries concerning pharmacotherapy in the elderly in our study may reflect the demographic development in Germany, which can be considered an aging society: to date, more than one-fifth of the German population is 65 years of age or older37, and this proportion is projected to rise over the coming decades38. Similar demographic trends are expected in emerging countries such as Brazil, China, and India39. Therefore, DICs in emerging countries might soon be facing a similar increase in the number of queries related to pharmacotherapy in the elderly. In addition, the comparatively high age of hospital patients as well as the participation of our institute in the orthogeriatric co-management must be taken into account as factors that might explain the sizable proportion of queries related to pharmacotherapy in the elderly in our study.

The importance of clinical pharmacologists in counseling on drug use during pregnancy was investigated by Erdeljić and co-workers, who demonstrated that clinical pharmacologists’ assessments were superior to the United States Food and Drug Administration (FDA) classification system in predicting pregnancy outcomes following pregnancy-related drug exposures40. Moreover, in a survey among physicians who consulted a Norwegian DIC on drug use during pregnancy, 9% of respondents stated that termination of a wanted pregnancy was avoided owing to the information provided by the DIC24. Overall, 92% of the respondents in this survey claimed that the answer from the DIC had a patient-specific clinical impact24. Surprisingly, only 3.5% of queries submitted to our clinical-pharmacological DIC during the study period referred to pregnancy or breastfeeding, which is significantly lower compared to older studies2,27, but in line with more recent reports15. The relatively low number of queries concerning pregnancy or lactation issued by healthcare professionals (88.6% (526/594) of whom were hospital-based) in our study is in accordance with previous investigations, which showed that hospital-based physicians less frequently inquire about these topics compared to primary healthcare physicians27,33,41. Moreover, the advent of freely accessible online databases such as Embryotox42, Bumps (Best Use of Medicines in Pregnancy)43, and Drugs and Lactation Database44 may explain the decline in the number of queries related to pregnancy or lactation compared to previous analyses2,27.

Of note, ADRs and drug–drug interactions (both PDIs and PKIs) were significantly more often addressed by patient-specific queries as compared to general queries in our study. This might be due to the fact that general information about ADRs and drug–drug interactions can easily be retrieved by physicians and other healthcare professionals themselves in SmPCs and drug interaction compendia (for example, via the drug information program AiDKlinik), without consultation of the clinical-pharmacological DIC.

Limitations of the present study mainly arise from its monocentric and retrospective design. Besides, we did not investigate the time spent answering queries and the references used in doing so. These topics have been covered comprehensively by other authors2,9,11,12,15,16,17,22,25,27,34,45. However, the most important limitation of the present study is that no information on patient-related outcomes was available. Hence, it was not possible to determine if consultation of the clinical-pharmacological DIC by treating physicians benefited their patients. To investigate this aspect, a control group would have been necessary, which was not feasible within the constraints of this study—a limitation our analysis shares with other reports7,8,9,10,11,12,14,15,16,17,21,25,26,27,30,34. Due to the exploratory character of our study, the statistical test results were not adjusted for multiple testing and must therefore be interpreted with circumspection.

Healthcare professionals usually do not have time resources to address time-consuming DI queries themselves22. Although the accessibility of scientific literature has been facilitated considerably over the past decades, primarily due to the advent of online search capabilities, the amount of available literature on a given pharmacological topic is substantially greater nowadays22. In addition, the complexity of DI queries has increased in recent years22. Therefore, the time spent on answering DI queries has not decreased significantly compared to former times22, suggesting an ongoing need for DICs to unburden clinicians in their daily routine.

Taken together, our study suggests that DI query characteristics depend on country-specific demographic factors (e.g., emerging vs. developed countries), the affiliation and catchment area of DICs, as well as on the composition of the DIC staff (clinical pharmacologist-led vs. pharmacist-led). We hypothesize that clinical pharmacologists and pharmacists can contribute effectively to a high quality of patient care and a high level of pharmacotherapy safety through DICs. Further studies are required to elucidate whether the utilization of DICs actually improves clinically relevant patient outcomes.