Abstract

background

First-generation antihistamines with sedative effect, such as doxylamine, are discussed as a possible risk factor for increased risk of falls, based on limited and conflicting data.

objectives

This study was conducted to collect data on the use and tolerability of doxylamine under real-life conditions in a large, unselected group of patients. Of particular interest was a possible risk of an increased fall rate in patients ≥ 65 years (and for defined age subgroups) when taking doxylamine. A further objective was to determine the causes of falls in patients ≥ 65 years.

methods

This was a prospective, multicenter, non-interventional post-authorisation safety study conducted in Germany over a two-year period (July 2020 to July 2022). Patients were recruited in pharmacies after purchasing doxylamine (Hoggar® Night tablets or orodispersible tablets, STADA Arzneimittel AG, Germany). The patient survey was performed using a paper diary for 21 days and analysed with descriptive statistics. results: Of 4,883 pharmacies contacted, 59 (1.2%) pharmacies were willing to participate. 84 patient diaries were evaluable. No falls or evidence of falls were reported throughout the study. Adverse events were reported by 8.3% of study patients, of which no adverse event was serious. 95.2% of study patients reported resuming or continuing use of Hoggar® Night if needed.

Conclusions

The study found no evidence for a possible causal relationship between falls in patients of any age group and the use of doxylamine. A major limitation of the study is that despite considerable recruitment efforts, only 5.4% of the targeted number of patients could be recruited.

Full Text

introduction

Doxylamine (doxylamine succinate), is a first-generation antihistamine with sedative effect that is approved in Germany for the short-term treatment of sleep disorders and is available as over-the-counter (OTC) product¹. As for many other drugs, the use of first-generation antihistamines is being discussed as a potential risk factor for an increased fall risk in older patients^2-4. However, there are only few and conflicting data on a possible association between first-generation antihistamines, in particular doxylamine, and an increased risk of falls.

A meta-analysis conducted in 2016³ identified and evaluated five studies on that topic^2,4-7. In their analysis, the authors identified an increased risk of falls with the use of first-generation antihistamines among the elderly. However, three of the five studies did not distinguish between antihistamines of different generations, which differ in terms of indication, duration of use, frequency of use, and dosage. In the remaining two studies, a substance-specific evaluation was not possible on the basis of the published data. Overall, the evidence from these studies does not allow a conclusive answer regarding a possible association between first-generation antihistamines, in particular doxylamine, and a potentially increased risk of falls.

A retrospective non-interventional safety study published in 2020 examined a possible causal relationship between falls in patients ≥ 65 years of age and use of first-generation antihistamines⁸. The evaluation included verified data sets of 169 physicians who had treated a total of 313,046 patients within the previous six months. Less than 10% of the 9,922 patients who had fallen (940 patients, 9.5%) had taken a first-generation antihistamine. Detailed information was available on 505 falls in 379 patients who had taken a first-generation antihistamine. In almost three-quarters of falls (72.5%), a causal relationship between the intake of first-generation antihistamines and the fall was not to be assumed. For 16 falls (3.2%), the causal relationship could not be assessed due to missing information. In almost 95% of the remaining 123 falls, the physicians considered between one and seven possible alternative causes for the falls (mainly underlying and concomitant diseases). The authors concluded that first-generation antihistamines do not play a relevant role in the fall incidents.

Many drugs are suspected of increasing the risk of falls. The so-called fall risk-increasing drugs (FRIDs) include a broad range of drugs, including central nervous system (CNS)-acting agents, cough preparations, nonsteroidal anti-inflammatory drugs (NSAIDs), anti-Alzheimer’s agents, antiplatelet agents, calcium antagonists, diuretics, alpha-blockers, digoxin, hypoglycaemic drugs, neurotoxic chemotherapeutic agents, nasal preparations, and antiglaucoma ophthalmic preparations⁹. Due to its mechanism of action (non-selective binding to H1 receptors peripherally and centrally), doxylamine falls into the FRID category of CNS-acting agents. However, drugs are only one possible risk factor, and the risk of falling depends on a variety of factors. These include physical changes associated with age such as frailty, multimorbidity, increased inactivity in old age, as well as depression, polypharmacy, sleep disorders and many environmental factors^10-13.

This study was conducted to collect data on the use and tolerability of doxylamine under real-life conditions in a large, unselected group of patients. Of particular interest was a possible risk of an increased rate of falls in patients ≥ 65 years.

Materials and Method

Study design
This study was a prospective, multicenter, non-interventional post-authorisation safety study (PASS) conducted in Germany according to § 63f of the German Medicines Act (AMG) and the recommendations of Good Pharmacovigilance Practices (GVP) Module VIII as well as the Good Pharmacoepidemiology Practices (GPP) published by the International Society for Pharmacoepidemiology (ISPE). The study was approved by the ethics committee of the Ärztekammer Westfalen-Lippe und der Westfälischen Wilhelms-Universität. The study was registered in the European Union electronic Register of Post-Authorisation Studies (EU PAS Register) with the number EUPAS35589 and was notified to the German Federal Institute for Drugs and Medical Devices (BfArM).

As the study aimed to collect unselected data on the use of an OTC product, conducting the study in general practitioner’s offices was excluded and it was decided to take on the challenge of a pharmacy-based study. The target study size was 2,000 patients in 200 to 400 pharmacies.

Data collection and analysis
The participating pharmacies were asked to provide their patients with product-independent counselling as usual. After buying doxylamine (Hoggar® Night tablets or orodispersible tablets, STADA Arzneimittel AG, Germany), patients were invited to participate in the survey. All participating patients provided written informed consent. In order to have the possibility to follow up patients, it was necessary to indicate the general practitioner at the beginning of the study.

Patients were asked to complete a paper diary for 21 days starting with the next doxylamine intake, which included general questions about the person and the intake of doxylamine; existing medical conditions; details of doxylamine intake; concomitant medication; adverse events (primary data source). In the occurrence of a fall or an event that might indicate a fall, details of the fall were requested using a separate fall questionnaire in order to perform a root cause analysis (secondary data source). To achieve the highest possible diary return rate, intensive follow-up of non-received diaries was conducted via pharmacists by contacting study patients.

Data collection was performed from July 2020 until July 2022. The analysis of study results was descriptive.

results

Pharmacy and patient recruitment
In total, 4,883 pharmacies were invited to participate in the study. Of these, 59 (1.2%) could be recruited of which 25 pharmacies recruited at least one evaluable patient (Table 1). The study target of recruiting 200 to 400 pharmacies was not achieved, although numerous measures were implemented to strengthen the recruitment. To improve pharmacy recruitment, pharmacies were approached through various channels (e-mails, phone calls, and company field staff). Pharmacies received an expense allowance. In addition, a pilot phase had been carried out to optimise the study processes. Most of the contacted pharmacies did not respond or declined without giving reasons. Reasons given: study does not fit into pharmacy operations; no time; no capacity; too much bureaucracy; short-time work, lack of staff and few customers because of corona pandemic; expected low willingness of patients to participate; generally no participation in studies.

Table 1: Overview of pharmacy recruitment

108 patients signed informed consent and received the diary, of whom 86 (79.6%) completed and returned the diary. 84 diaries (77.8%) were evaluable. The study target of recruiting 2,000 patients was not achieved, although numerous measures were implemented to support pharmacies in patient recruitment (Table 2).

Table 2: List of implemented measures to assist pharmacies with patient recruitment

Evaluation of motivational calls to the 58 participating pharmacies revealed that 40% of pharmacies indicated that patient recruitment was challenging and 50% of pharmacies indicated that study implementation was difficult or not feasible for a variety of reasons (corona pandemic, lack of staff/time, other priorities, etc.). Reasons given by patients: no interest; no time; only sporadic use; product bought for another person; not willing to provide information on general practitioner or frequency of intake; general disinterest; only short time spent in pharmacy due to corona pandemic.

The following evaluation refers to the 84 patients with evaluable diaries. Since no falls were reported in any age group, the subgroups are not discussed in more detail.

Description of study patients
Of the 84 patients, 59 (70.2%) were female, 22 (26.2%) were male; no gender information for 3 (3.6%) patients. The median patient age was 63 years (range: 18 – 96 years, age group distribution see Table 3).

Table 3: Age group distribution of the 84 evaluable patients

A total of 102 concomitant diseases were documented in 45 (53.6%) patients (Table 4); the proportion in the subgroup of patients ≥ 65 years (n=38) was even higher: 62 concomitant diseases in 26 (68.4%) patients. Of the 45 patients with concomitant diseases, 17 (20.2%) patients had 1, 25 (29.8%) patients had 2 to 5, and 3 (3.6%) patients had 6 to 10 concomitant diseases.

Table 4: Concomitant diseases of the 84 evaluable patients according to MedDRA SOC sorted by frequency

Concomitant medications were documented in a total of 57 (67.9%) patients (78.9% in the subgroup of patients ≥ 65 years). Of the 57 patients with concomitant medications, 24 (28.6%) patients had 1, 29 (34.5%) patients had 2 to 5, and 4 (4.8%) patients had 6 to 10 concomitant medications. The most common Anatomical Therapeutic Chemical (ATC) level 2 of concomitant medications were agents acting on the renin-angiotensin system (C09) in 19 (22.6%) patients, beta blocking agents (C07) in 13 (15.5%) patients, and thyroid therapy (H03) in 12 (14.3%) patients. According to the FRID definition by Chen et al.⁹, 38 different concomitant medications taken by 35 (41.7%) patients can be classified as FRIDs (Table 5).

Table 5: FRIDs (other than doxylamine) taken by the 84 evaluable patients according to ATC level 5

Use of doxylamine
For the majority of 66 (78.6%) patients, the reason for purchasing doxylamine was their own decision (84.2% in the subgroup of patients ≥ 65 years). 18 (21.4%) patients had received a recommendation from their physician without a prescription. No patient had a prescription from a physician. With 70.2% (59 patients), the majority of patients were already taking Hoggar® Night prior to the study (78.9% in the subgroup of patients ≥ 65 years).

The most common current reason for taking doxylamine was initial insomnia in 73 (86.9%) patients, followed by middle insomnia in 54 (64.3%) patients (multiple responses were possible).

The dose of one doxylamine tablet or orodispersible tablet is equivalent to 25 mg doxylamine hydrogen succinate. The median dose was 25 mg doxylamine per patient per night (range: 6.25-50 mg per patient per night).

In total, 79 patients (94%) took doxylamine for more than seven days during the observational period so that the risk of falls was mainly studied in patients who had not only a single or sporadic use of doxylamine, but an intake over several days. This also applies to the subgroup of patients ≥ 65 years (94.7%).

With 80 patients (95.2%), the majority stated in the questionnaire that they would continue or resume taking Hoggar® Night if needed (97.4% in the subgroup of patients ≥ 65 years). This was denied by 3 (3.6%) patients without giving a reason. One patient (1.2%) did not provide any information on that question.

Reported falls and other adverse events
The main objective of the study was to investigate a possible causal relationship between falls and doxylamine use, particularly in patients ≥ 65 years. A further objective was to determine the causes of falls in patients ≥ 65 years. No falls were reported throughout the study (95% confidence interval 0.0%-4.3%; large confidence interval due to small number of patients). Since there were no falls, the fall rate in any age group is 0 and causes of falls could not be analysed.

The reported adverse events (AEs) are listed in Table 6. 77 patients (91.7%) did not report any AEs. None of the reported AEs were serious.

Table 6: Reported AEs according to MedDRA SOC and PT sorted by frequency

discussion

The major challenge during study set-up was to identify an adequate study design to collect unselected data on the use of an OTC product. Conducting the study in general practitioner’s offices would have led to a preselection of patients which was confirmed by the fact that 78.6% of patients bought doxylamine without consulting a physician. Pharmacies were assessed as the only unselected source to users of an OTC product.

In order to recruit a large number of study patients, a total of 4,883 pharmacies were invited to participate in the study. However, after a study period of two years, only 59 pharmacies could be recruited and the diaries of only 84 patients from 25 pharmacies could be evaluated. Thus, the study goal of recruiting 2,000 patients in 200 to 400 pharmacies was not met. Even a preceding pilot phase to optimise the processes, several contact efforts to strengthen the pharmacy recruitment as well as numerous measures to support pharmacies in patient recruitment could not change this. This resulted in the small number of study patients, which is the main limitation of the study. Consequently, the authors are aware of a low statistical power and the high risk of a type 2 error. Nevertheless, it is highly unlikely that the planned sample size of 2,000 patients could be achieved if the study were repeated. Non‐interventional PASS are conducted to gather valuable data on the routine clinical use in all patient groups14-16. Unfortunately, low interest and poor recruitment are common problems in such studies¹⁷. Creating incentives to participate in such studies without causing bias is a key challenge in the context of PASS. In addition, the expense allowances paid to patients is restricted by compliance and ethical requirements. The permitted amount does not appear to be a sufficient incentive for patients to participate in such studies. Another obstacle to recruitment was that patients feel embarrassed when they disclose the intake of sleeping pills. Sleep disorders are a sensitive topic in society, and patients are reluctant to give detailed information about this – especially if they are aware that they are not adhering to all (dosage) information in the package leaflet. Another reason for the patients’ reluctance could be their concern to become suspected of substance abuse. An approach to overcome these indication- and substance-related challenges could be the development of an improved patient communication strategy. Through a well-considered and empathic communication with the patient, pharmacy staff might be able to mitigate patients’ fears, clear up misinformation and eliminate data privacy concerns. In addition to the above-mentioned recruitment barriers, the corona virus pandemic coincided with the start of the study, which had a non-negligible impact on the unwillingness of pharmacists and patients to participate in the study.

Falls are a general health problem. Particularly among older people and with increasing frailty, the risk of falls increases¹⁸. Per year, an estimated 28% to 35% of persons aged 65 years and older fall^19-21. Besides drugs as a possible factor, there are a number of other factors that can influence the risk of falls^10-13. Identifying specific factors that increase the risk of falls and determining their magnitude is difficult due to the multitude of factors^22-24.

In a cross-sectional analysis by Elliott et al.²⁵ using data from the National Ambulatory Medical Care Survey in the United States, the prescription of new FRIDs in patients aged 65 years and older between those with and without an injurious fall was evaluated. The sample included 239,016,482 ambulatory care visits, of which 5,095,734 (2.1%) were related to an injurious fall. An injurious fall was associated with a non-statistically significant increase in odds of at least one new FRID prescription: adjusted odds ratio = 1.6 (95% confidence interval 0.6-4.0). Interestingly, it was observed that adults aged 65 to 74 years showed a non-statistically significant increase, whereas adults aged 75 years and older showed a non-statistically significant decrease in odds of being prescribed a new FRID than their counterparts²⁵.

Discontinuation of FRIDs is a common strategy for fall prevention but according to a systematic review and meta-analysis by Lee et al.²², this has no consequence on fall rate and incidence. Analysis of five trials with 1,305 participants found that the FRIDs deprescribing over a follow-up period of 6 to 12 months did not reduce the rate of falls, the frequency of falls or the rate of fall-related injuries. Specifically for the incidence of falls, a total of 971 FRID users aged 65 years and older out of 4 studies were analysed. Of these, 499 patients discontinued or reduced the use of FRIDs, and 472 patients continued the FRID use unchanged. In the follow-up period, there were 190 fallers in the first group (38%) and 170 fallers in the second group (36%; risk ratio 1.04, 95% confidence interval 0.86-1.26). The authors concluded that deprescribing of FRIDs as the sole fall prevention strategy made little to no difference in the rate or risk of falls²². As an individual intervention, only exercise has robust evidence demonstrating reductions in the incidence of fallers and rate of injurious falls^23,26.

The main focus of the study was to identify a potential risk for increased fall rates in patients ≥ 65 years of age and for specific age subgroups when taking doxylamine. No falls were reported by the 84 study patients. Hence, the study found no evidence relevant to a possible causal relationship between falls in patients of any age group and the intake of doxylamine. Despite limitations due to the small number of patients, the current study results are consistent with the findings of the study by Gomez Perez et al.⁸, a non-interventional, retrospective, cross-sectional study with general practitioners and resident neurologists which found no evidence of an increased risk of falls and concluded that first-generation antihistamines do not play a relevant role in the fall incidents.

conclusions

Doxylamine was well tolerated when used as directed and the majority of study patients reported resuming or continuing use of Hoggar® Night if needed. The study found no fall events in the patients recruited, and therefore no evidence for a possible causal relationship between falls in patients of any age group and the use of doxylamine. In January 2022, the German Federal Ministry of Health decided not to change the OTC status of doxylamine.

Statement of Personal Interest: Sonja Gomez Perez is an employee of STADA Arzneimittel AG. Iris Hassel is an employee of STADA Arzneimittel AG. Markus Torben Schweimer is an employee of STADA Arzneimittel AG. Sandra Gilbert is an employee of STADA Arzneimittel AG. Andreas Iwanowitsch is an employee of STADA Arzneimittel AG.

Declaration of funding interests: This study was funded in full by STADA Arzneimittel AG. The preparation of this paper was funded by STADA Arzneimittel AG. Initial data analyses were undertaken by AMS Advanced Medical Services GmbH funded by STADA Arzneimittel AG.

Acknowledgement: The authors would like to thank all pharmacists and patients who participated in the study and AMS Advanced Medical Services GmbH for conducting the study. The study was registered in the European Union electronic Register of Post-Authorisation Studies (EU PAS Register) with the number EUPAS35589.

Correspondence to: Sonja Gomez Perez, Stadastr. 2 – 18, 61118 Bad Vilbel, sonja.gomez-perez@stada.de