Use and efficacy of dry-mouth interventions in Sjögren’s disease patients and possible association with perceived oral dryness and patients’ discomfort

A cross-sectional study was performed among members of the Dutch Sjögren’s Patients Federation (Nationale Vereniging Sjögrenpatiënten, NVSP) who attended the annual meeting of this Federation in Breukelen, the Netherlands on October 1st, 2022. According to self-reported information from these members, they suffered from Sjögren’s disease. Volunteers could anonymously fill in a questionnaire and return it in a designated mailbox during the meeting or return the questionnaire by mail using an enclosed prepaid envelope. The local Ethics Review Committee of the Academic Center for Dentistry Amsterdam (ACTA) confirmed that the Medical Research Involving Human Subjects Act (WMO) did not apply to this study (protocol number 201930). The reporting of this study conforms to the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) statement [9].

The questionnaire, developed for this study, consisted of five parts. The first part consisted of some general questions with regard to age, sex, and a question in which year Sjögren’s disease had been diagnosed by a rheumatologist. Second, the internationally validated xerostomia inventory (XI) was used to measure the overall oral dryness of the patients [10]. The XI consists of 11 items on a 5-point Likert scale ranging from 1—“Never” to 5—“Very often”. The items concern patients’ oral dryness and feeling of the mouth, for example, “my mouth feels dry when eating a meal”. For each item, patients indicate how often they experience the feeling of dryness in the mouth and problems regarding oral dryness. The scores of the 11 items are summed to produce a total XI-score that ranges between 11 (no xerostomia) and 55 (extreme xerostomia) [10]. The third part of the questionnaire comprised of the bother index (BI). In the BI, the patient was asked to rate the severity of dry mouth on a scale from 0 to 10 [11‐15]. The patients could also indicate which moment of the day they were mostly affected by a dry mouth: morning, afternoon, evening, or at night. The fourth part of the questionnaire was the Regional Oral Dryness Inventory (RODI), which determines differences in dry mouth perception at different intra-oral regions. The RODI questionnaire contains nine schematic illustrations of different locations in the oral cavity [16, 17]. Four illustrations represent areas in the upper jaw: the upper lip, the posterior part of the palate (from the rugae up to the end of the soft palate), the anterior part of the palate (including the rugae), and the inside part of the cheeks. Four other illustrations represent areas in the lower jaw: the lower lip, the anterior part of the tongue (from the tip of the tongue up to the vallate papilla), the posterior part of the tongue (from the vallate papilla up to the end of the tongue), and the floor of the mouth. Finally, one illustration represents the pharynx. At each location, the patient can indicate the severity of the intra-oral dryness on a 5-point Likert scale ranging from 1—“No dryness” to 5—“Severe dryness” [16, 17]. The questionnaire concluded with a list of potential interventions to relief the feeling of a dry mouth [5, 18] (Table 3). “Drinking small volumes” was specifically added to highlight the conscious decision when using liquid dry-mouth interventions. This refers to the distinction between regular drinking, for example, drinking water at once, or a conscious decision to use small sips. For each intervention, a participant could report whether or not they used the intervention, and if they did, what the frequency of use was throughout the day. Besides,they could indicate how effective they perceived the intervention to be.

SjD patients who actively checked "YES” for the question on the current use of an intervention were considered as current users. Patients who used an intervention method in the past could, however, also fill in how they had perceived the efficacy of that intervention, leading to an overall assessment of the efficacy of each intervention method. The efficacy of each intervention was questioned by a 5-point Likert scale, ranging from 1—“Not effective” to 5—“Highly effective”. To reflect on the perceived efficacy by the current users, the number of current users who indicated an efficacy of ≥ 4 (effective or highly effective) for each intervention was also reported. If a participant indicated to use the intervention “throughout the day” instead of a specific number; then, the highest number reported by other participants for this intervention was chosen. For “rinsing of the mouth”, participants could also indicate which liquid was used for rinsing. For “using a mouth gel”, participants had to indicate where they apply the mouth gel, using the same regions of the RODI as described above. In addition, participants were asked whether they used any odors, which could be any smell, perfume, or scent, to relief their perceived oral dryness, and if so, which odor they used for this purpose.

The data were statistically analyzed with SPSS, version 27.0 (IBM Corp SPSS statistics, Armonk, NY, USA). The Shapiro–Wilk test was used to assess the normality of the data. As not all variables were normally distributed, the data are presented as medians and their interquartile range (IQR) and/or as the mean and standard deviation (SD). Because most of the data were not normally distributed, non-parametric tests had been used for all statistical analysis. The questionnaires were not filled in completely by all volunteers; therefore, the total number (N) could vary.

A Friedman test was conducted for the RODI scores of the study sample, followed by a Wilcoxon signed rank test as a post hoc procedure. The differences in the efficacy of the five most commonly used interventions by the participants were investigated using the Friedman test, followed by a Wilcoxon signed rank test as a post-hoc procedure.

The association between the mostly used interventions (> 50% of participants) with oral dryness (XI and RODI scores) and patients’ discomfort (BI scores) was initially investigated with a binary logistic regression. Each dry-mouth intervention was considered as a dependent variable, and the total XI, BI, and RODI scores of the nine intra-oral regions were considered as independent variables. To identify the degree of multicollinearity among the independent variables, the variance inflation factor (VIF) was calculated. The VIF for these variables was < 5, which indicates that there was no multicollinearity present among these variables [18, 19], so they do not influence each other. The backward conditional method was used to analyze these independent variables. If there was a significant association between a dry-mouth intervention and one or more independent variables; then, the odds ratio (OR) and the 95% confidence interval (95% CI) were reported. Furthermore, the last step of the Omnibus test chi-square and the Hosmer and Lemeshow (H–L) test chi-square including their degree of freedom (df) and their p values were reported. Also, the Coxx-Snell R square and the Nagelkerke R square were mentioned, if the association was significant.

All significance levels (α) were set at 0.05.

The overall dry-mouth feeling was quantified with the XI. The mean total XI score of the study sample was 44.7 ± 6.6 with a median score of 45.0 and IQR of 41.0–50.0 (N = 90). Patients’ dry-mouth discomfort as measured with BI had a mean of 7.5 ± 1.9 with a median score of 8.0 and IQR of 7.0–9.0 (N = 88). Most of the patients reported having discomfort mostly at night (82.6%) and in the morning (75.0%). Patients’ discomfort was less during the afternoon (62%) and in the evening (63%).

The perceived oral dryness at various intra-oral locations, as determined by the RODI questionnaire, is presented in Table 1. Perceived oral dryness in the study sample differed significantly among the nine intra-oral regions (Friedman test p < 0.05, followed by Wilcoxon signed-rank test). It was found that the perceived intra-oral dryness was most severe for the pharynx and the posterior palate. In contrast, the inside of the cheeks was experienced as the least dry. The pharynx had significant differences with all intra-oral regions, except for the posterior palate.

Most of the SjD patients used, or previously used, multiple interventions to relief their perceived oral dryness (Table 2). Only a limited number of patients used no or a single intervention (2.2%). The majority of the patients used 6–9 different interventions for relief of their dry mouth (64.2%). One participant even used ≥ 15 different interventions to relief oral dryness.

Table 3 gives an overview of all interventions, as used by SjD patients, and the frequency of the use per day of the applied interventions. The interventions mostly used (by > 50% of participants) for the relief of oral dryness were “drinking water”, “eating fruit”, “drinking tea”, “moistening the lips”, “drinking small volumes”, “drinking coffee”, and “rinsing of the mouth”. Other interventions such as “drinking soft drinks”, “eating cucumber with olive oil during meal”, “drinking lemonade”, “drinking beer”, “putting olive oil in the mouth”, “using pilocarpine”, “putting lemon slices in the mouth”, and “sucking ice cubes” were less prevalent among SjD patients. Over 50% of the patients indicated to rinse their mouth as an intervention. “Water” was the most reported liquid used for rinsing, while 3 participants indicated that they used a “mouth wash” for rinsing.

For patients who indicated that they used a mouth gel for relief of their oral dryness, the most frequently reported regions where the mouth gel was applied were the anterior part of the tongue (23.9%), inside cheeks (21.7%), and posterior part of the tongue (19.6%). The mouth gel was less frequently applied to all other intra-oral regions, with percentages ranging between 12% for the upper and lower lip to 16.3% for the posterior part of the palate.

The frequency of use per day was the highest for “drinking water”, “drinking small volumes”, and “concentrating on other activities”, indicating that patients used these interventions multiple times a day (≥ 14 times/day). All other interventions were used between 1–7 times/day (Table 3).

The spontaneously reported “using other interventions” included “using mint flavored lozenges”, “using mouth wash”, “using specialized toothpaste”, and “eating cucumber”. No frequencies and efficacies were reported in this subgroup.

Table 3 reports the efficacy of the interventions used to reduce perceived dry-mouth feeling. The highest efficacy (≥ 3.5) was reported for “chewing gum” and “using a mouth gel”. The efficacy for the majority of the other interventions ranged between 1.9 and 2.9, indicating that most of the interventions were not found to be effective for the relief of a dry mouth. Only a few interventions were reported to have a moderate effectivity (≥ 3.0 efficacy < 3.5); “drinking water”, “moistening the lips”, “drinking small volumes”, “rinsing of the mouth”, “concentrating on other activities”, “sucking sour candies”, and “using XyliMelts”.

The efficacy of the top five interventions differed significantly (Friedman test p < 0.01). “Drinking coffee” was a less effective intervention compared to almost all other interventions (Wilcoxon signed-rank test p < 0.05). “Moistening the lips” was reported to have the highest efficacy of the top five interventions and differed significantly from “eating fruit” and “drinking tea”. Furthermore, no significant difference in efficacy was found between “drinking water”, “moistening the lips”, and “drinking small volumes” (Wilcoxon signed-rank test p > 0.05), suggesting that these three interventions can be considered equally effective in their ability to relief perceived oral dryness.

Table 3 also reports the number of current users who indicated an efficacy of ≥ 4 (effective to highly effective). Highly effective interventions are “putting lemon slices in the mouth”, “chewing gum”, “sucking ice cubes”, “using XyliMelts”, and “using a mouth gel” (> 67% of the SjD patients using these interventions). This result is in line with the overall efficacy for “chewing gum”, “using a mouth gel”, and “using XyliMelts”, which also had the highest overall perceived efficacy. However, for “putting lemon slices in the mouth” and “sucking ice cubes”, the overall perceived efficacy ranged between 1.9 and 2.3. For the other interventions, the percentage of users reporting an efficacy of ≥ 4 ranged between 22.0 and 61.3%, indicating that these interventions were less effective for relieving perceived oral dryness.

The association between the use of dry-mouth interventions and perceived oral dryness was further investigated. Severe perceived dryness of the posterior palate and the pharynx (RODI score ≥ 4) showed no association with the use of any dry mouth intervention (Mann–Whitney U test, p > 0.05, Supplementary file S1).

A binary logistic regression was performed to investigate whether perceived oral dryness and/or patients’ discomfort was associated with the use of the more mostly used dry-mouth interventions. In Table 4, the odds ratios for the dry-mouth interventions are reported. Interventions such as “drinking water”, “drinking tea”, “drinking coffee”, and “using mouth gel” did not have any significant association with any of the included independent variables; for this reason, these interventions are not included in Table 4. The use of “drinking small volumes”, “rinsing the mouth”, and “using XyliMelts” were all significantly associated with the BI scores of the users. The odds ratio varied between 1.45 and 2.85, indicating that higher BI scores (increased patients’ discomfort) were associated with an increased likelihood of using these interventions. “Moistening of the lips” was significantly associated with the total XI score (OR = 1.1); thus, higher XI scores (severe overall mouth dryness) increase the likelihood of use of this specific intervention.

Various dry-mouth interventions were significantly associated with various regions of the RODI questionnaire (Table 4). None of the most commonly used interventions were significantly associated with the RODI scores of inside cheeks, posterior palate, floor of the mouth, and pharynx. As a result, these particular regions are not included in Table 4. Notably, even the most dry regions, the posterior palate, and the pharynx, did not have any significant association. However, “drinking small volumes” was significantly associated with the RODI score of the anterior palate (OR > 4.0), which indicates that severe dryness at the palate increased the likelihood to use this intervention. “Concentrating on other activities” was significantly associated with the RODI scores of the lower lip, while “eating fruit” had an OR = 3.6 with the RODI score for the posterior tongue. Otherwise, some interventions reported in Table 4 had an OR < 1, which indicated that SjD patients with less dry intra-oral regions (low RODI scores) are more likely to use this specific dry-mouth intervention.