Medications That Cause Dry Mouth As an Adverse Effect in Older People: A Systematic Review and Metaanalysis † ‡§ §¶ Edwin C. K. Tan, PhD,* Duangjai Lexomboon, PhD, Gunilla Sandborgh-Englund, PhD, † †§ Ylva Haasum, PhD, and Kristina Johnell, PhD

CONCLUSION: Medication use was significantly associ- OBJECTIVES: To assess and quantify the risk of drug- ated with xerostomia and salivary gland hypofunction in induced dry mouth as a side effect in older people. older adults. The risk of dry mouth was greatest for drugs DESIGN: Systematic review and metaanalysis. used for urinary incontinence. Future research should SETTING: A search of the literature was undertaken develop a risk score for medication-induced dry mouth to using Medline, Embase, Cochrane, Web of Science, and assist with prescribing and medication management. JAm PubMed from 1990 to 2016. Geriatr Soc 66:76–84, 2018. PARTICIPANTS: Older people (aged ≥60) who partici- pated in intervention or observational studies investigating Key words: xerostomia; dry mouth; medication; drug use as an exposure and xerostomia or salivary gland systematic review; metaanalysis hypofunction as adverse drug outcomes. MEASUREMENTS: Two pairs of authors screened titles and abstracts of studies for relevance. Two authors inde- pendently extracted data, including study characteristics, definitions of exposure and outcome, and methodological quality. For the metaanalyses, random-effects models were lder adults are high users of medications. It is esti- 2 used for pooling the data and I statistics for exploring Omated that 40% of community-dwelling and 75% of heterogeneity. institutionalized older adults take 5 or more medications, RESULTS: Of 1,544 potentially relevant studies, 52 were with approximately 10% of older adults taking 10 or deemed eligible for inclusion in the final review and 26 in more.1 This high burden of comorbidity and polyphar- metaanalyses. The majority of studies were of moderate macy put older adults at risk of adverse drug events. methodological quality. In the intervention studies, urolog- Dry mouth is one of the most common adverse effects ical medications (odds ratio (OR) = 5.91, 95% confidence of medication use in older adults.2,3 This includes salivary 2 interval (CI) = 4.04–8.63; I = 62%), antidepressants gland hypofunction (objectively measured decrease in sali- 2 (OR = 4.74, 95% CI = 2.69–8.32, I = 21%), and psyc- vation) and xerostomia (subjective feeling of dry mouth). It 2 holeptics (OR = 2.59, 95% CI = 1.79–3.95, I = 0%) has been shown that the prevalence of hyposalivation were significantly associated with dry mouth. In the obser- increases with the number of medications used,2 but few vational studies, numbers of medications and several medi- studies have investigated the severity of medication-induced cation classes were significantly associated with xerostomia dry mouth and associated sequelae.3 Adverse effects of sali- and salivary gland hypofunction. vary gland hypofunction include dental caries, dysgeusia, oral mucosal soreness, and oral candidiasis.4 The process of chewing and swallowing may also be affected, and this can affect the nutritional status of older people. From the *Centre for Medicine Use and Safety, Faculty of Pharmacy and Drug-induced dry mouth has been attributed to a Pharmaceutical Sciences, Monash University, Melbourne, Australia; † Aging Research Center, Center for Alzheimer Research, Department of range of mechanisms. Although anticholinergic effects Neurobiology, Care Sciences and Society, Karolinska Institutet and underlie the majority of cases, other mechanisms include ‡ Stockholm University, Stockholm; Department of Health Science, § sympathomimetic effects, topical effects of inhaled medica- Karlstad University, Karlstad; Academic Center for Geriatric Dentistry; ¶ tions, dehydration, vasoconstriction in salivary glands, and Department of Dental Medicine, Karolinska Institutet, Stockholm, Sweden. alterations in electrolyte and fluid balance, and changes in saliva composition.5 Address correspondence to Edwin Tan, Aging Research Center, Karolinska Institutet, Gavlegatan€ 16, SE-113 30 Stockholm, Sweden. There is strong evidence that certain drug classes cause E-mail: [email protected] dry mouth. A previous systematic review identified particu- DOI: 10.1111/jgs.15151 lar drug groups that induce salivary gland hypofunction

JAGS 66:76–84, 2018 © 2017, Copyright the Authors Journal compilation © 2017, The American Geriatrics Society 0002-8614/18/$15.00 JAGS JANUARY 2018–VOL. 66, NO. 1 MEDICATIONS THAT CAUSE DRY MOUTH IN OLDER PEOPLE 77 and xerostomia, including those affecting the alimentary, DL and KJ). Full-text copies were obtained if a study cardiovascular, genitourinary, nervous, and respiratory appeared to meet the inclusion criteria or it was unclear systems.4 Examples of drug classes that induce dry mouth whether it would meet the criteria. Two authors (ET and include antidepressants, antipsychotics, anticholinergics, DL) independently reviewed the full texts to assess each antihypertensives, antihistamines, and sedatives,4,6 but study’s suitability for inclusion. Disagreements or uncer- there is a lack of reviews focusing on elderly adults, even tainties about study inclusion were resolved by discussion though they are at greater risk of drug-induced dry mouth. in the presence of the two other authors (GS and KJ). Additionally, although certain drugs have been found to be associated with dry mouth, few studies have tried to Data Extraction and Validity Assessment quantify the risk. Two authors (ET and DL) independently extracted data using a standardized data abstraction form. Data extracted Objective included year of publication; study design; country; study The aim of this study was to systematically review the lit- setting; follow-up duration (for dry mouth outcome); study erature to assess and quantify the risk of drug-induced dry sample characteristics (age and relevant comorbidities); mouth as a side effect in individuals aged 60 and older. sample size; analytical sample size; year of data collection; definition of drug exposure, including name, dose, dose form, duration of use, and method used to determine drug METHODS exposure; definition of xerostomia or salivary gland hypo- function outcome; method used to determine presence of Protocol and Registration the outcome; and proportions in exposed and unexposed The protocol for this review was registered with the Inter- groups. national Prospective Register of Systematic Reviews Methodological quality of studies was assessed using (PROSPERO) (registration number CRD42016040033). the Jadad score for randomized controlled trials (RCTs)7 and an adapted version of the Newcastle-Ottawa Scale modified for cross-sectional studies.8 For studies reporting Search Strategy and Information Sources pooled analyses, the original studies were assessed for A literature search was undertaken using Medline, Embase, quality. If at least one study did not meet a particular Cochrane, Web of Science, and PubMed from 1990 to quality criterion, the pooled study was assigned a 0 for 2016. Medical Subject Headings (MeSH), Emtree subject that quality criterion. headings, topics and key words related to dry mouth (xeros- tomia, asialia, hyposalivation, salivary dysfunction, salivary Metaanalysis flow, salivary secretion, dry mouth), medication use (drugs, polypharmacy, specific drug classes), older people (aged, If 2 or more studies reported a similar drug exposure and elder, senior, older), and study design (controlled trials, primary outcome measure with appropriate extractable observational studies) were used. The full search strategy is data, a metaanalysis was undertaken. Metaanalysis was included in the supplementary material (Supplementary performed using Review Manager version 5.3 (Nordic Table S1). Searches were limited to English-language arti- Cochrane Centre, Cochrane Collaboration, Copenhagen, cles and excluded conference abstracts. Denmark). Random-effects models were used for pooling the data, and I2 statistics were used to explore heterogene- ity. The effect size for the metaanalysis was calculated as Eligibility Criteria odds ratios (ORs) and risk differences (RDs). Significance Studies were included in the review if they were original was tested using Z-statistics. research published as a scientific peer-reviewed article from 1990 onwards; were intervention studies or observational RESULTS studies (case-control studies, cohort studies, before-after studies and cross-sectional studies, conducted in humans); Study Selection included persons aged 60 and older (if studies included subjects <60, data had to have been extractable for those The search of electronic databases retrieved 2,591 articles. ≥60; included a group exposed to drugs, including specific After removing duplicates, the titles and abstracts of 1,544 drugs and drug classes or number of drugs; included a studies were reviewed, of which 867 were excluded comparator or control group that did not receive drugs or because they clearly did not meet the inclusion criteria. the drug of interest (studies that involved active compara- The full-text versions of 677 articles were obtained and tors or persons who acted as their own controls were eligi- scrutinized, and 624 were excluded after independent ble); and reported on xerostomia, hyposalivation, or review by at least two authors (Figure 1). A total of 52 salivary gland hypofunction as adverse drug effects, mea- studies were included in the final review and are summa- sured in exposed and unexposed groups. rized below and in Supplementary Tables S2 and S3.

Study Selection Study Characteristics The titles and abstracts of studies were screened indepen- Of the included studies, 33 were experimental studies, – dently for relevance by two pairs of authors (ET and GS, including 25 placebo controlled9 33 and 8 with active 78 TAN ET AL. JANUARY 2018–VOL. 66, NO. 1 JAGS

Records identified through Additional records identified database searching through other sources (n = 2591) (n = 0) Identification

Records after duplicates removed (n = 1544)

Screening Records screened Records excluded (n = 1544) (n = 867)

Full-text articles assessed Full-text articles excluded, for eligibility with reasons (n=625) (n = 677) Not elderly (n = 562)

Eligibility Not drug exposure (n=14)

Studies included in Not dry mouth outcome qualitative synthesis (n = 18) (n = 52) No comparator (n=12) Study design (n = 3) No extractable data (n = 8) Studies included in

Included quantitative synthesis Pooled data, original study (meta-analysis) already included (n = 2) (n = 26) Review article (n = 2) Conference abstract (n=4)

Figure 1. PRISMA flow diagram. [Color figure can be viewed at wileyonlinelibrary.com]

– – – comparators29,34 40 (1 pooled analysis29 included both). G04BD, n = 16),12,13,15,16,18 20,22,29 33,36,40,43 antidepres- – Six studies were pooled analyses of multiple sants (N06A, n = 10)10,11,21,25 27,34,35,38,39 and antipsy- trials.15,20,25,28,29,32 The remaining 19 studies were obser- chotics (N05A, n = 3).17,24,37 One study each assessed – vational and used a cross-sectional study design.41 59 The cardiovascular drugs (C0314 and C0754), hypnotics and majority of studies were conducted in North America sedatives (N05C),23 and antiepileptics for neuropathic pain – – (n = 19),11,18,19,23,26,34 36,39,40,42,43,46,48,52 54,59 followed (N03A).28 by Europe (n = 13)9,13,14,16,22,30,31,41,44,49,51,56,58 and Asia The cross-sectional studies mainly investigated broad (n = 7).12,38,45,50,55,57 Thirteen studies were undertaken or nonspecific groups of drugs as exposures. Eight studies across multiple continents as part of large, multinational assessed number of drugs,45,47,49,51,53,56,58 and two studies – trials.10,15,17,20,24,25,27 29,32,33,37 assessed use of any medication.44,52 Six studies assessed Study sample sizes ranged from 11 to 13,508 subjects. different types or classes of drugs,45,47,50,51,56,58 3 studies Most studies included older individuals who were gener- assessed xerogenic or hyposalivation-inducing drug ally healthy or without specific comorbidities use,46,53,59 and 2 studies assessed anticholinergic drug – – (n = 19).9,18,19,41,44 53,55 59 The most common disease use.41,48 When a duration of drug exposure was reported, states investigated were urological, including urinary this ranged from use within the previous 2 weeks44,47 to – incontinence (n = 15);12,13,15,16,20,22,27,29,31 33,36,40,43 men- 1 year.46 tal health, including major depression (n = 8),11,17,25,26,34,35,38,39 generalized anxiety disorder Outcomes (n = 2),10,24 psychosis (n = 2),37,42 cognitive impairment (n = 1),31 and insomnia (n = 2);21,23 neuralgia (n = 1);28 All experimental studies assessed xerostomia as a dichoto- and hypertension (n = 2).14,54 mous outcome, without further description on how this was assessed. Follow-up duration for the assessment of xerostomia in experimental studies ranged from 1 week19 Drug Exposure to 3 years.14 Eleven cross-sectional studies had xerostomia – A wide range of drug treatments was assessed across the as an outcome,41 51 and eight had salivary gland hypo- – included studies. Experimental studies tested specific drugs, function.52 59 One study differentiated between daytime the most common being those used for urinary frequency and nighttime xerostomia,44 2 studies assessed severity of and incontinence (Anatomical Therapeutic Chemical code xerostomia,47,50 and 2 studies objectively assessed oral JAGS JANUARY 2018–VOL. 66, NO. 1 MEDICATIONS THAT CAUSE DRY MOUTH IN OLDER PEOPLE 79 moisture.45,50 Regarding salivary gland hypofunction out- preparation,22 with the remainder assessing extended- comes, four studies assessed stimulated salivary flow release formulations. All oxybutynin studies included in – rate,52,55 57 with two studies also assessing unstimulated the metaanalysis assessed immediate-release oral prepara- salivary flow rate;52,57 one study assessed parotid and sub- tions. The metaanalysis found that urological medications mandibular gland secretion rate;54 one study assessed were significantly associated with xerostomia (OR = 5.91, hyposalivation;58 and one study assessed minor salivary 95% CI = 4.04–8.63; RD = 0.25, 95% CI = 0.13–0.36) gland output.53 (Figure 2, Supplementary Figure S1a). When examining specific urological drugs, oral immediate-release oxybu- tynin was associated with the greatest risk of dry mouth Risk of Bias Within Studies (RD = 0.56, 95% CI = 0.43–0.70), whereas the risk with The majority of experimental studies (n = 15) received a mirabegron was negligible (RD = 0.00, 95% CI = À0.01– – – – Jadad score of 3 out of 5,9 11,13 15,17 37,40 with 6 studies 0.02). Statistical heterogeneity was low across trials inves- scoring the maximum 5 out of 5.9,15,17,19,24,31,33 The main tigating antidepressants (I2 = 21%). The main drug limitations of included RCT studies were their failure to assessed in this group was duloxetine (n = 4). Antidepres- report appropriate methods of randomization sequence sant use was significantly associated with xerostomia – – – generation12 14,16,20,22,23,25 30,32,34,35,37 40 and double (OR = 4.74, 95% CI = 2.69–8.32; RD = 0.09, 95% – – – blinding10 13,16,18,20,21,23,25 29,32,34,35,37 39 (n = 21 for CI = 0.04–0.14) (Figure 3, Supplementary Figure S1b). each). The majority of cross-sectional studies scored 4 stars Statistical heterogeneity was low across trials assessing out of 10 on the adapted Newcastle-Ottawa psycholeptics (I2 = 0%). Psycholeptics were more likely Scale.41,43,46,48,50,55,59 One study scored the maximum 10 than placebo to cause xerostomia (OR = 2.59, 95% stars,44 and 3 studies scored 3 stars.52,54,57 The main limi- CI = 1.79–3.95; RD = 0.09, 95% CI = 0.05–0.12) (Fig- tation of the cross-sectional studies was related to selection ure 4, Supplementary Figure S1c). of the study sample. The majority of studies failed to describe a sample that was truly or somewhat representa- – DISCUSSION tive of the target population (n = 12),41,43,45,46,48 50,52,54,55,57,59 include a sample size that was justified and – – Summary of Evidence satisfactory (n = 13),41,43,45,46,48,50,52 55,57 59 or ade- quately establish the comparability of respondents and Our systematic review found that medication use was sig- nonrespondents and determine whether the response rate nificantly associated with xerostomia and salivary gland – – – was satisfactory (n = 15).41 43,45,46,48,50 55,57 59 Full hypofunction in older individuals. Findings of the meta- reports of quality assessment are presented in Supplemen- analyses confirm that specific drugs, including those from tary Tables S4 and S5 and Supplementary Texts S1 and therapeutic classes used for urinary incontinence, antide- S2. pressants, and psycholeptics are significantly associated with xerostomia. The risk was greatest in drugs used for urinary incontinence, for which the odds of xerostomia Results of Individual Studies were almost 6 times as great as with placebo. Methodolog- In the experimental studies, a greater proportion of those ical limitations of current studies were related to sample undergoing drug treatment reported xerostomia than of selection, randomization, and blinding. those receiving placebo. In the cross-sectional studies, med- The most common ATC categories associated with ication use, number of medications, and using a wide xerostomia were agents acting on the genitourinary and ner- range of medication classes were significantly associated vous systems, with the most cited drugs being tolterodine (4 with xerostomia, salivary flow rate, and salivary gland studies), duloxetine (4 studies), and oxybutynin (3 studies). hypofunction. In studies investigating number of drugs as This finding is reflected in a previous review.4 Most of the the exposure, all except 141 reported a significant associa- drugs used for urinary incontinence are antimuscarinics and tion between number of drugs and dry mouth in older exert dry mouth through this antimuscarinic mechanism. people. Main findings are summarized in Supplementary Oxybutynin, tolterodine, and fesoterodine are nonselective Tables S1 and S2. antimuscarinic agents. Darifenacin and solifenacin are M3- selective receptor antagonists and may be more bladder-spe- cific and less prone to causing anticholinergic side effects, Metaanalysis such as dry mouth. Mirabegron, assessed in one study, is a Twenty-two placebo-controlled RCTs were deemed eligible beta-3 adrenergic receptor agonist and an alternative to – – – – for further inclusion in metaanalyses.10 12,15 17,19 27,29 33 antimuscarinic drugs. It is less likely to cause dry mouth, Of the RCTs, included trials investigated drugs used and this is reflected in our metaanalysis in that it was not for urinary frequency and incontinence (ATC code significantly different from placebo. – G04BD) (n = 13),12,15,16,19,20,22,29 33 antidepressants Our metaanalysis of antidepressants included 3 classes – (N06A) (n = 6),10,11,21,25 27 and psycholeptics (N05) of antidepressants: duloxetine, a selective serotonin and (n = 3).17,23,24 Statistical heterogeneity was substantial noradrenaline reuptake inhibitor (SSNRI); escitalopram, a across the trials assessing drugs used for urinary frequency selective serotonin-reuptake inhibitor (SSRI); and doxepin, and incontinence (I2 = 62%). The main drugs assessed in a tricyclic antidepressant (TCA). TCAs enhance the actions this group included tolterodine (n = 4)12,22,29,33 and oxy- of noradrenaline and serotonin by blocking their reuptake butynin (n = 3).16,19,30 Of the tolterodine studies, 1 inves- at the neuronal membrane, but they also block histaminic, tigated twice-daily administration of the immediate-release a1-adrenergic, and muscarinic cholinergic receptors, 80 TAN ET AL. JANUARY 2018–VOL. 66, NO. 1 JAGS

Figure 2. Forest plots for randomized controlled trials of drugs used for urinary frequency and incontinence (G04BD) vs placebo. [Color figure can be viewed at wileyonlinelibrary.com] resulting in adverse drug reactions such as dry mouth.2 to causing dry mouth.2 Quetiapine, an atypical antipsy- Although SSRIs and SNRIs can cause dry mouth, they are chotic, works through antagonism at D2 receptors in the generally less likely to cause anticholinergic side effects mesolimbic pathway and 5HT2A receptors in the frontal than TCAs.2 cortex. Quetiapine also has a strong affinity for histamin- Several psycholeptic medications, including particular ergic and a1-adrenergic receptors, which may be responsi- antipsychotics, anxiolytics, and hypnotics, are also prone ble for dry mouth. Eszopiclone, a nonbenzodiazepine JAGS JANUARY 2018–VOL. 66, NO. 1 MEDICATIONS THAT CAUSE DRY MOUTH IN OLDER PEOPLE 81

Figure 3. Forest plots for randomized controlled trials of antidepressants (N06A) vs placebo. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4. Forest plots for randomized controlled trials of psycholeptics (N05) vs placebo. [Color figure can be viewed at wileyon- linelibrary.com] hypnotic, interacts with the gamma-aminobutyric acid-ben- for urinary frequency and incontinence (G04BD) zodiazepine receptor complex. The most common adverse (OR = 5.91), which is probably related to their antimus- effect of eszopiclone is unpleasant taste, although it may carinic properties. This was equivalent to a 25% greater also produce dry mouth. absolute risk of dry mouth and is clinically significant There were considerable differences in the magnitude given that 1 in 4 individuals treated with a urological med- of the likelihood of xerostomia between different drug ication developed dry mouth. Orally administered immedi- classes. This is probably because of variations in the mech- ate-release oxybutynin was associated with the greatest anism of actions between and within drug classes. The risk of dry mouth (56% absolute risk increase) of the greatest likelihood of xerostomia was seen with drugs used drugs in the same therapeutic class. The second largest 82 TAN ET AL. JANUARY 2018–VOL. 66, NO. 1 JAGS likelihood was seen with antidepressants (N06A) models. Some included studies were of low methodological (OR = 4.74), which mainly consisted of antidepressants quality, and thus their findings should be interpreted with with fewer anticholinergic side effects. Lastly, the psyc- caution. Our systematic review included fewer studies than holeptics (N05) in our metaanalysis had the lowest likeli- previous similar reviews,4 which probably reflects our stric- hood of causing xerostomia (OR = 1.60), because they ter inclusion criteria, including restricting the population to were newer agents with more selective receptor activity. older people and the years of publication. Antidepressants and psycholeptics were associated with a 9% greater absolute risk of dry mouth, indicating that, for Implications every 11 people treated, one would experience drug- induced dry mouth. This review has implications for practitioners treating and In our systematic review, a significant association was managing older adults using drugs. Practitioners should found between number of drugs and dry mouth in older carefully weigh the benefits of therapy against the potential people. These findings reflect the results of previous studies risk of dry mouth, oral complications, and treatment discon- conducted in the general population and highlight the link tinuation. This is particularly important in people with between polypharmacy and dry mouth.60 polypharmacy. Medications should be regularly monitored To our knowledge, this is the first review to quantify and reviewed to identify potential oral side effects of medi- the risk of medication-associated dry mouth in persons cations, and dose adjustment or change of therapy should be aged 60 and older. The odds ratio of xerostomia as a side considered when needed. Based on the quantified risk, risk effect of xerostomic drugs in our study would be expected scores can be developed that may be a useful guide for prac- to be higher than in younger individuals because salivary titioners not only in identifying individuals at risk of dry production declines with aging.61 A Swedish study also mouth, but also in evaluating the risk grade for dental reported a greater prevalence of xerostomia with older age health. This can provide information on how much the risk in nonmedicated adults,60 but there have been few other of dental complications is for elderly adults and aid in the metaanalyses assessing the risk of dry mouth as a medica- development and implementation of preventive programs tion side effect in elderly adults. and comprehensive therapy plans. Future studies should be conducted to develop risk scores for dental health based on the use of different drug groups and numbers of drugs used. Limitations Additional factors such as dose, duration, and formulation This systematic review and metaanalysis has some limita- of treatment should also be considered. The major conse- tions. Although broad search strategies were implemented quences of medication-induced xerostomia, such as the in a range of databases to ensure that all relevant studies development of dental caries and need for clinical interven- were included, unpublished and non-English-language tion, should also be explored in future research. studies were not sought. Thus, the possibility of publica- tion bias cannot be excluded. We limited the search from CONCLUSION 1990 to 2016, so currently used medications with a strong association with dry mouth that were researched before Our systematic review found that medication use was sig- this time were excluded. Common examples include older nificantly associated with xerostomia and salivary gland TCAs, potent diuretics, antihistamines, and various drugs hypofunction in older individuals. Findings of the meta- acting on the central nervous system analyses show that the risk of xerostomia was greatest for Dry mouth is usually not a primary outcome of experi- drugs used for urinary incontinence, followed by antide- mental studies; hence our search may have missed certain pressants and psycholeptics. Future research should studies that did not report dry mouth as an adverse out- develop a risk score for medication-induced dry mouth to come in the title or abstract. Heterogeneity in the defini- assist with prescribing and medication management. tions and methods used to assess dry mouth across studies may have affected the ability to compare findings and draw ACKNOWLEDGMENTS conclusions. Xerostomia is a subjective phenomenon, and different people may have experienced and reported xeros- Financial Disclosure: This study is supported by Grant tomia differently. Currently, there is no criterion standard OF14305423 from the Stockholm County Council. ECKT method of assessing xerostomia. There may have been dif- is supported by a National Health and Medical Research ferences in medication exposure between studies. Treat- Council, Australian Research Council Dementia Research ment dose, duration, and formulation were not assessed in Development Fellowship. this review. It is likely that these factors influence the sever- The authors would like to thank the Karolinska Insti- ity of medication-induced xerostomia.3 For ease of classifi- tutet Universitetsbiblioteket for their assistance in devising cation, we categorized drugs according to ATC codes, but and running the initial search. inclusion of different drugs in each group, some of which Conflict of Interest: No authors have potential con- may have different indications, may have differing effects flicts of interest with reference to this paper. on xerostomia. Most of the cross-sectional studies used Author Contributions: Study concept and design: ECKT, self-reported drug use, which may underestimate total drug DL, GSE, YH, KJ. Acquisition of data: ECKT, DL, GSE, YH, use. Additionally, the way drugs were classified into partic- KJ. Analysis and interpretation of data: ECKT, DL, GSE, KJ. ular groups may have varied between studies. We addressed Preparation of manuscript: ECKT, DL, GSE, YH, KJ. potential heterogeneity in sample characteristics and treat- Sponsor’s Role: The funding agency had no role in the ment effects in our metaanalyses by using random-effects study design, methods, and subject recruitment; in the JAGS JANUARY 2018–VOL. 66, NO. 1 MEDICATIONS THAT CAUSE DRY MOUTH IN OLDER PEOPLE 83 collection, analysis, and interpretation of data; in the writ- 23. McCall WV, Erman M, Krystal AD et al. A polysomnography study of ing of the report; and in the decision to submit the article eszopiclone in elderly patients with insomnia. Curr Med Res Opin 2006;22:1633–1642. for publication. 24. Mezhebovsky I, Magi K, She F et al. Double-blind, randomized study of extended release quetiapine fumarate (quetiapine XR) monotherapy in older patients with generalized anxiety disorder. Int J Geriatr Psychiatry REFERENCES 2013;28:615–625. 25. Oakes TM, Katona C, Liu P et al. Safety and tolerability of duloxetine in 1. Johnell K, Fastbom J. Comparison of prescription drug use between com- elderly patients with major depressive disorder: A pooled analysis of two munity-dwelling and institutionalized elderly in Sweden. Drugs Aging placebo-controlled studies. Int Clin Psychopharmacol 2013;28:1–11. 2012;29:751–758. 26. Raskin J, Wiltse CG, Dinkel JJ et al. Safety and tolerability of duloxetine 2. Scully C. Drug effects on salivary glands: Dry mouth. Oral Dis at 60 mg once daily in elderly patients with major depressive disorder. J 2003;9:165–176. Clin Psychopharmacol 2008;28:32–38. 3. Aliko A, Wolff A, Dawes C et al. World Workshop on Oral Medicine VI: 27. van Schagen Leeuwen JH, Lange RR, Jonasson AF et al. Efficacy and Clinical implications of medication-induced salivary gland dysfunction. safety of duloxetine in elderly women with stress urinary incontinence or Oral Surg Oral Med Oral Pathol Oral Radiol 2015;120:185–206. stress-predominant mixed urinary incontinence. Maturitas 2008;60:138– 4. Wolff A, Joshi RK, Ekstrom J et al. A guide to medications inducing sali- 147. vary gland dysfunction, xerostomia, and subjective sialorrhea: A systematic 28. Semel D, Murphy TK, Zlateva G et al. Evaluation of the safety and effi- review sponsored by the World Workshop on Oral Medicine VI. Drugs R cacy of pregabalin in older patients with neuropathic pain: Results from a D 2017;17:1–28. pooled analysis of 11 clinical studies. BMC Fam Pract 2010;11:85. 5. Sreebny LM, Schwartz SS. A reference guide to drugs and dry mouth, 2nd 29. Wagg A, Cardozo L, Nitti VW et al. The efficacy and tolerability of the edition. Gerodontology 1997;14:33–47. beta3-adrenoceptor agonist mirabegron for the treatment of symptoms of 6. Villa A, Wolff A, Narayana N et al. World Workshop on Oral Medicine overactive bladder in older patients. Age Ageing 2014;43:666–675. VI: A systematic review of medication-induced salivary gland dysfunction. 30. Wagg A, Dale M, Tretter R et al. Randomised, multicentre, placebo-con- Oral Dis 2015;5:365–382. trolled, double-blind crossover study investigating the effect of solifenacin 7. Jadad AR, Moore RA, Carroll D et al. Assessing the quality of reports of and oxybutynin in elderly people with mild cognitive impairment: The randomized clinical trials: Is blinding necessary? Control Clin Trials SENIOR study. Eur Urol 2013;64:74–81. 1996;17:1–12. 31. Wagg A, Khullar V, Marschall-Kehrel D et al. Flexible-dose fesoterodine in 8. Wells G, Shea B, O’Connell D et al. The Newcastle-Ottawa Scale (NOS) elderly adults with overactive bladder: Results of the randomized, double- for assessing the quality of nonrandomised studies in meta-analyses blind, placebo-controlled study of fesoterodine in an aging population trial. (online). 2016. Available at: http://www.ohri.ca/programs/clinical_epide J Am Geriatr Soc 2013;61:185–193. miology/oxford.asp. Accessed October 11, 2017. 32. Wagg A, Wyndaele J-J, Sieber P. Efficacy and tolerability of solifenacin in 9. Ahmed AIA, van den Elsen GAH, Colbers A et al. Safety and pharmacoki- elderly subjects with overactive bladder syndrome: A pooled analysis. Am J netics of oral delta-9-tetrahydrocannabinol in healthy older subjects: A ran- Geriatr Pharmacother 2006;4:14–24. domized controlled trial. Eur Neuropsychopharmacol 2014;24:1475–1482. 33. Zinner NR, Mattiasson A, Stanton SL. Efficacy, safety, and tolerability of 10. Alaka KJ, Noble W, Montejo A et al. Efficacy and safety of duloxetine in extended-release once-daily tolterodine treatment for overactive bladder in the treatment of older adult patients with generalized anxiety disorder: A older versus younger patients. J Am Geriatr Soc 2002;50:799–807. randomized, double-blind, placebo-controlled trial. Int J Geriatr Psychiatry 34. Cohn CK, Shrivastava R, Mendels J et al. Double-blind, multicenter com- 2014;29:978–986. parison of sertraline and amitriptyline in elderly depressed patients. J Clin 11. Bose A, Li D, Gandhi C. Escitalopram in the acute treatment of depressed Psychiatry 1990;51(Suppl B):28–33. patients aged 60 years or older. Am J Geriatr Psychiatry 2008;16:14–20. 35. Cohn J, Claghorn J, Ostrow D. A comparative trial of trimipramine and 12. Chung S-D, Chang H-C, Chiu B et al. The efficacy of additive toltero- doxepin in depressed elderly patients. Adv Ther 1991;8:92–102. dine extended release for 1-year in older men with storage symptoms 36. Herschorn S, Pommerville P, Stothers L et al. Tolerability of solifenacin and clinical benign prostatic hyperplasia. Neurourol Urodyn and oxybutynin immediate release in older (> 65 years) and younger (< 2011;30:568–571. 65 years) patients with overactive bladder: Sub-analysis from a Canadian, 13. Dorschner W, Stolzenburg JU, Griebenow R et al. Efficacy and cardiac randomized, double-blind study. Curr Med Res Opin 2011;27:375–382. safety of propiverine in elderly patients—a double-blind, placebo-controlled 37. Kennedy JS, Jeste D, Kaiser CJ et al. Olanzapine vs haloperidol in geriatric clinical study. Eur Urol 2000;37:702–708. schizophrenia: Analysis of data from a double-blind controlled trial. Int J 14. Fletcher A, Amery A, Birkenhager W et al. Risks and benefits in the trial Geriatr Psychiatry 2003;18:1013–1020. of the European working party on high blood pressure in the elderly. J 38. Mazeh D, Shahal B, Aviv A et al. A randomized, single-blind, comparison Hypertens 1991;9:225–230. of venlafaxine with paroxetine in elderly patients suffering from resistant 15. Foote J, Glavind K, Kralidis G et al. Treatment of overactive bladder in depression. Int Clin Psychopharmacol 2007;22:371–375. the older patient: Pooled analysis of three phase III studies of darifenacin, 39. Pollock BG, Mulsant BH, Nebes R et al. Serum anticholinergicity in elderly an M-3 selective receptor antagonist. Eur Urol 2005;48:471–477. depressed patients treated with paroxetine or nortriptyline. Am J Psychiatry 16. Iselin CE, Schmidlin F, Borst F et al. Oxybutynin in the treatment of early 1998;155:1110–1112. detrusor instability after transurethral resection of the prostate. Br J Urol 40. Sand PK, Miklos J, Ritter H et al. A comparison of extended-release oxy- 1997;79:915–919. butynin and tolterodine for treatment of overactive bladder in women. Int 17. Katila H, Mezhebovsky I, Mulroy A et al. Randomized, double-blind study Urogynecol J Pelvic Floor Dysfunct 2004;15:243–248. of the efficacy and tolerability of extended release quetiapine fumarate 41. Desoutter A, Soudain-Pineau M, Munsch F et al. Xerostomia and medica- (quetiapine XR) monotherapy in elderly patients with major depressive dis- tion: A cross-sectional study in long-term geriatric wards. J Nutr Health order. Am J Geriatr Psychiatry 2013;21:769–784. Aging 2012;16:575–579. 18. Kay GG, Maruff P, Scholfield D et al. Evaluation of cognitive function in 42. Frenchman IB. Atypical antipsychotics for nursing home patients—a retro- healthy older subjects treated with fesoterodine. Postgrad Med spective chart review. Drugs Aging 2005;22:257–264. 2012;124:7–15. 43. Grossklaus DJ, Franke JJ. Treatment of elderly women with urge inconti- 19. Kay GG, Staskin DR, MacDiarmid S et al. Cognitive effects of oxybutynin nence in middle Tennessee: A single institution practice-based study. Tenn chloride topical gel in older healthy subjects: A 1-week, randomized, dou- Med 2000;93:457–460. ble-blind, placebo- and active-controlled study. Clin Drug Investig 44. Johansson A-K, Johansson A, Unell L et al. Self-reported dry mouth in 2012;32:707–714. Swedish population samples aged 50, 65 and 75 years. Gerodontology 20. Kraus SR, Ruiz-Cerda JL, Martire D et al. Efficacy and tolerability of feso- 2012;29:e107–e115. terodine in older and younger subjects with overactive bladder. Urology 45. Kakudate N, Muramatsu T, Endoh M et al. Factors associated with dry 2010;76:1350–1357. mouth in dependent Japanese elderly. Gerodontology 2014;31:11–18. 21. Lankford A, Rogowski R, Essink B et al. Efficacy and safety of doxepin 46. Lam A, Kiyak A, Gossett AM et al. Assessment of the use of xerogenic 6 mg in a four-week outpatient trial of elderly adults with chronic primary medications for chronic medical and dental conditions among adult day insomnia. Sleep Med 2012;13:133–138. health participants. Consult Pharm 2009;24:755–764. 22. Malone-Lee JG, Walsh JB, Maugourd MF. Tolterodine: A safe and effec- 47. Murray Thomson W, Chalmers JM, John Spencer A et al. A longitudinal tive treatment for older patients with overactive bladder. J Am Geriatr Soc study of medication exposure and xerostomia among older people. 2001;49:700–705. Gerodontology 2006;23:205–213. 84 TAN ET AL. JANUARY 2018–VOL. 66, NO. 1 JAGS

48. Ness J, Hoth A, Barnett MJ et al. Anticholinergic medications in commu- and pharmacotherapy. Community Dent Oral Epidemiol 1997;25:211– nity-dwelling older veterans: Prevalence of anticholinergic symptoms, symp- 216. tom burden, and adverse drug events. Am J Geriatr Pharmacother 61. Affoo RH, Foley N, Garrick R et al. Meta-analysis of salivary flow rates in 2006;4:42–51. young and older adults. J Am Geriatr Soc 2015;63:2142–2151. 49. Norlen P, Ostberg H, Bjorn AL. Relationship between general health, social-factors and oral health in women at the age of retirement. Commu- nity Dent Oral Epidemiol 1991;19:296–301. SUPPORTING INFORMATION 50. Ohara Y, Hirano H, Yoshida H et al. Ratio and associated factors of dry mouth among community-dwelling elderly Japanese women. Geratr Geron- Additional Supporting Information may be found in the tol Int 2011;11:83–89. online version of this article: 51. Smidt D, Torpet LA, Nauntofte B et al. Associations between oral and ocu- lar dryness, labial and whole salivary flow rates, systemic diseases and medications in a sample of older people. Community Dent Oral Epidemiol Table S1. Full search strategy for each database. 2011;39:276–288. Table S2. Characteristics of included experimental 52. Billings RJ, Proskin HM, Moss ME. Xerostomia and associated factors in studies with xerostomia outcome. a community-dwelling adult population. Community Dent Oral Epidemiol 1996;24:312–316. Table S3. Characteristics of included cross-sectional 53. Bretz WA, Loesche WJ, Chen YM et al. Minor salivary gland secretion in studies with xerostomia and salivary gland hypofunction the elderly. Oral Surg Oral Med Oral Pathol Oral Radiol Endod as outcomes. 2000;89:696–701. Table S4. Quality assessment of experimental studies. 54. Cowman RA, Frisch M, Lasseter CJ et al. Effects of beta-adrenergic antag- onists on salivary secretory function in individuals of different ages. J Table S5. Quality assessment of cross-sectional studies Gerontol 1994;49:B208–B214. Text S1. Jadad scoring system. 55. Ikebe K, Matsuda K-I, Morii K et al. Relationship between bite force and Text S2. Newcastle-Ottawa Scale adapted for cross- salivary flow in older adults. Oral Surg Oral Med Oral Pathol Oral Radiol sectional studies. Endod 2007;104:510–515. 56. Johanson CN, Osterberg T, Lernfelt B et al. Salivary secretion and drug Figure S1a-c. Forest plots for RCTs with xerostomia treatment in four 70-year-old Swedish cohorts during a period of 30 years. as an outcome reported as risk differences. S1a. Drugs Gerodontology 2015;32:202–210. used for urinary frequency and incontinence (G04BD) vs 57. Kagawa R, Ikebe K, Enoki K et al. Influence of hypertension on pH of sal- placebo. S1b. Antidepressants (N06A) vs placebo. S1c. iva in older adults. Oral Dis 2013;19:525–529. 58. Narhi TO, Tenovuo J, Ainamo A et al. Antimicrobial factors, sialic acid, Psycholeptics (N05) vs placebo. and protein concentration in whole saliva of the elderly. Scand J Dent Res Please note: Wiley-Blackwell is not responsible for the 1994;102:120–125. content, accuracy, errors, or functionality of any support- 59. Persson RE, Persson GR, Kiyak HA et al. Oral health and medical status ing materials supplied by the authors. Any queries (other in dentate low-income older persons. Spec Care Dentist 1998;18:70–77. 60. Nederfors T, Isaksson R, Mornstad H et al. Prevalence of perceived symp- than missing material) should be directed to the corre- toms of dry mouth in an adult Swedish population—relation to age, sex sponding author for the article.