Polypharmacy and saliva volumes in the northeast of Germany – The Study of Health in Pomerania

Objectives: Associations between saliva volumes or salivary flow rates and poten tially xerogenic medication are rarely evaluated in cohorts with a wide


| INTRODUC TI ON
Salivary flow substantially differs between and within individuals. 1 Besides the individual's degree of hydration, other important factors affecting salivary flow include olfactory stimuli, 2 body position 3 and chewing. 4 Unstimulated salivary flow of ≤0.1 mL/min 5,6 and/or stimulated salivary flow of ≤0.5 up to 0.7 mL/min 5,7 is defined as hyposalivation. Among older people, its prevalence ranges from 15% to 23%; among hospitalized older people, its prevalence ranges from 17% to 50%. 8 Hyposalivation is often associated with xerostomia (subjective feeling of dry mouth). 9 Thus, xerostomia well predict hyposalivation. 10,11 This widespread symptom significantly reduces an individual's quality of life, 12 with a prevalence of 5.5 to 39% in the general population, and up to 72% in older persons living in residential care homes 8 but only 10% among adults in their early 30s. 13 The main reasons for hyposalivation and xerostomia among older people are the increasing number of systemic diseases (eg, diabetes mellitus 14 and renal insufficiency) and the resulting medication intake, herein referred to as xerogenic medications. 15 Potentially xerogenic medication seems to be one of the most common causes of xerostomia, 16,17 predominantly diuretics, psychotropics and cardiovascular medications, but also analgesics and antihistamines. 10,18 Xerogenic medications seem to either reduce salivary flow rates or alter individual thresholds for perceived dry mouth. However, a meta-analysis has shown that whole salivary flow rates were significantly lower in older than in younger participants, independent of medication. 19 Additionally, a systematic review reported that at the first level of the Anatomical Therapeutic Chemical (ATC) Classification System, anatomical groups of the alimentary, cardiovascular, nervous and respiratory system might affect salivary gland function and induce xerostomia. 20 In this context, polypharmacy (5-9 medications) and hyperpolypharmacy (>10 medications) 21

| Study design
Analyses are based on data from two independent cohorts of the  Figure S1).
A detailed description of the study design can be found in the Appendix S1.

| Covariates and medical history
A computer-based questionnaire was used to collect information on age, sex and participant's medical history in face-to-face interviews conducted by specially trained interviewers. Prescription and nonprescription medications taken within the 7 days prior to the interview were recorded, including homeopathic substances and dietary supplements by asking: "During the last 7 days: have you taken any medications such as tablets, drops, suppositories, or have you had any injection?" and every compound was recorded.
Participants had to bring the packages or an intake plan from a physician of all medications for identification by linking the German central pharmaceutics number with the German drug databases and the code of the ATC Classification System. Data on the administration interval (regularly consumed/as required) were gathered for each medication.

| Statistical analysis
Continuous data were summarized as means and standard deviations (SD). For categorical variables, numbers and percentages were presented.
Mann-Whitney U tests were applied to detect differences in numbers of medications or stimulated saliva volumes between two groups.
Mixed linear models (including a random intercept for study, as

| RE SULTS
Participant's characteristics are provided in Table 1. The average age of the participants was 53.8 years, just under half (48%) were male, and 65% and 45% regularly took medication or potentially xerogenic medications, respectively. Stimulated saliva volume was on average 968.4 µL/min (SD 432.5). Table 2 presents proportions of participants taking at least one potentially xerogenic medication classified according to ATC main therapeutic groups. Highest prevalences were observed among medications of the main therapeutic group C (cardiovascular system); beta-blocking agents and agents acting on the renin-angiotensin system were found in 26.0% and 30.1% of participants, respectively. Only 0.4% of participants used functional gastrointestinal disorder medications (A03).  Table S1. Average numbers of all medications, potentially xerogenic medications and potentially xerogenic cardiovascular medications were significantly higher in males than in females, whereas females were prescribed a higher number of potentially xerogenic medications impacting the nervous system than were males. Mean stimulated saliva volumes by age and sex are shown in Table 3. Unadjusted mean stimulated saliva volumes varied widely TA B L E 1 Characteristic of participants of the pooled sample (SHIP-2 and SHIP-Trend-0)

| D ISCUSS I ON
Using data from two cohorts of the prospective SHIP (including 6753 Caucasian subjects in total), we showed that, firstly, average saliva volumes were lower with more medications. Secondly, potentially xerogenic medications, especially those with effects on the cardiovascular or the nervous system, were associated with lower saliva volumes.
Thirdly, males more frequently took potentially xerogenic medications affecting the cardiovascular system, while females more often took potentially xerogenic medication affecting the nervous system.
A strength of this study is the use of strict protocols to collect data. Medical history data were collected with high accuracy. A computer-based questionnaire was used to determine medication, and in addition, participants had to bring the packages or an intake plan from a physician of all medications to gain further information.
Also, saliva sampling and preparation were conducted according to a strict protocol, thereby minimizing any methodological biases. The spitting method of saliva collection, as used in this study, is the most reproducible and effective method even when chewing length (stimulation) is as much as 2 minutes. 25 This study also has some limitations. Fourthly, medication consumption was classified as regularly or irregular (as needed) using self-reports, which is a further limitation.
Finally, in addition to multiple medications, a range of medical con- Advanced age is associated with disease and consequently with increased medication intake. 32,34,35 This finding was also observed in the current study: prevalence of regularly consumed medication was   medications regularly. However, our study did not distinguish between polypharmacy and hyperpolypharmacy. 21,22 In the Swedish registry, 32 the number of dispensed medications was significantly related to age, with 2 medications (median) among the 20-to 29-year-olds and 7 medications (median) among the 70-to 79-year-olds, which are slightly higher than those in SHIP. In a Dutch study, 37% of the older adults (≥65 years) took 2 to 5 medications, and 4% were prescribed >5 medications in the year 1997. 35 In SHIP, estimates were somewhat lower. Differences might be explained by period effects (data presented were the most recent). Furthermore, a meta-analysis on stimulated salivary flow rates in adults 19 revealed significantly lower stimulated flow rates in the older group; however, decreased flow rates might have been partly caused by a higher medication intake. 19 Given potentially xerogenic medications might promote hyposalivation, it is important to consider the association between ageing and multimorbidity and polypharmacy and hyperpolypharmacy 21,22 given the rising numbers of older people and ageing populations.
The evidence to date is unclear about whether medication, either in total or specific therapeutic groups, is sex specific. In SHIP, total medication consumption was significantly higher among men than women, but the actual difference was only marginal (Δ0.03 medications). This was in line with findings from a previous Japanese study. 36 In contrast, the Swedish group 32 reported that females received on average 0.7 more medications than men (5.0 versus 4.3). Regarding TA B L E 4 Mean stimulated saliva volume by the number of regularly consumed medication and by regularly consumed potential xerogenic medication (no potentially xerogenic medication versus at least one medication of the specified type) as classified by their main therapeutic group Note: Linear mixed models evaluated associations of the number of regularly consumed medications and stimulated saliva volume.
Data are presented as numbers (percentages) or means ± standard deviations. P values were derived from linear mixed models, adjusted for age, sex, and time point of saliva collection.
Abbreviations: B, linear regression coefficient; CI, confidence interval. a P for differences compared to the references group.; b P for trend across categories.
sex-specific distributions of certain main therapeutic groups, our study showed that men more frequently were prescribed potentially xerogenic medication affecting the cardiovascular system than were women, while women more frequently took xerogenic medication affecting the nervous system. Interestingly, a Danish study reported comparable results. 23 Sex-specific differences in medication use might indeed be restricted to certain main therapeutic groups. In Germany, cardiovascular diseases in women are diagnosed only half as often as in men, 37 while women suffer from depression (eg, due to double burden of family/work and violence) nearly twice as often as men. 38 Current literature suggests that polypharmacy might be associated with lower salivary flow rates, 22,39-43 even when distinguishing between poly-and hyperpolymedication. 22 Although the values in the present study were within the range of normal stimulated saliva volumes, a high number of regularly consumed medications, potentially xerogenic medications, and those affecting the cardiovascular or the nervous system was significantly associated with lower stimulated saliva volumes. Only for beta-blocking agents (C07) medication intake was statistically significantly associated with lower saliva volumes (Δ70.7 μL/min). For agents acting on the renin-angiotensin system (C09) and psychoanaleptics (N06), the sample size was too small to detect an association with saliva volumes, although regression coefficients indicated lower saliva volumes on average for participants with medication intake (Δ12.1 μL/min and Δ62.7 μL/min, respectively).
These results are in agreement with other studies on cardiovascular medication, 39 with special focus on beta-blocking agents (C07). 40 Neurologic diseases and their pharmacologic treatment could result in hyposalivation and xerostomia, as they centrally block the inner-brain salivation centre. 41,42 Psycholeptics (N05) and psychoanaleptics (N06) block certain receptors 43 and inhibit salivation. However mental pressure, depression and anxiety disorders are physiologically associated with hyposalivation, even without medication. 29 Smidt et al (2010) 23 found that stimulated salivary flow rates were lower in participants taking antidepressants (N05; primarily citalopram), cardiac agents (C01; mostly digitalis glycosides, digoxin) and calcium-channel blockers (C08; especially verapamil). Interestingly, unstimulated saliva flow was only reduced in psychoanaleptics (N06).
Although certain ATC groups have been identified as being associated with lower saliva volumes, it remains to be clarified whether certain combinations of medication potentiate the effects on saliva volumes and therefore the limit clinical relevance of the present study. For example, using data from long-term hospital stays, stimulated parotid flow was lower among men taking combinations of tricyclic antidepressants (N06) and diuretics (C03). 44 In this context, long-term effects of certain combinations of xerogenic medications on saliva volumes might be evaluated using prospective follow-up data from this cohort.
To conclude, polypharmacy is common in the older population.
Higher consumption of medications, in general, but also intake of potentially 'xerogenic' medications from specific main therapeutic groups (eg, affecting the cardiovascular or the nervous system) are potentially associated with lower stimulated saliva volumes, thereby constituting a relevant oral health problem.

ACK N OWLED G EM ENTS
The authors acknowledge the participation of the study partici- GABA, Therwil, Switzerland, supported this analysis with an unrestricted educational grant.

CO N FLI C T O F I NTE R E S T
The authors do not report any known competing financial interests or personal relationships that could have appeared to influence the work of this reported study.

AUTH O R CO NTR I B UTI O N S
OL contributed to interpretation of data and drafting the article. BH contributed to conception and design, analysis and interpretation of data, and drafting the article. CP contributed to analysis of data and critically revised the article for important intellectual content. SS contributed to acquisition of data and critically revised the article for important intellectual content. HV contributed to acquisition of data and critically revised the article for important intellectual content.
TK contributed to conception and design and acquisition of data and critically revised the article for important intellectual content. All authors gave their final approval of the version to be published.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data from SHIP are available after data application and signature of a data transfer agreement. The data dictionary and the online application form are available at: fvcm.med.uni-greifswald.de/dd_service/data_use_intro.php. Involving a local collaborative partner to facilitate the application process is recommended.