Hospitalization after Oral Antibiotic Initiation in Finnish Community Dwellers with and without Alzheimer’s Disease: Retrospective Register-Based Cohort Study

Abstract

Background:

Persons with Alzheimer’s disease (AD) are frequently hospitalized from infection-related causes. There are no previous studies investigating hospitalization associated with antibiotic initiation in persons with AD.

Objective:

To investigate the frequency and risk of hospitalization associated with oral antibiotic initiation among community dwellers with and without AD.

Methods:

We performed a retrospective register-based study utilizing register-based Medication Use and Alzheimer’s disease (MEDALZ) cohort. It includes all community dwellers diagnosed with AD during 2005–2011 in Finland and their matched comparison persons without AD. Antibiotic use was initiated by 34,785 persons with and 36,428 without AD. Drug use data were collected from Prescription Register and comorbidities from Special Reimbursement and Hospital Care Registers. Infection diagnoses were collected from the Hospital Care Register. Factors associated with hospitalization were estimated utilizing logistic regression models.

Results:

Risk of hospitalization following antibiotic initiation was higher among antibiotic initiators with AD than without AD (adjusted odds ratio, aOR, 1.37, 95% Cl 1.28–1.46).

Strongest association with hospitalization was found for oral glucocorticoid use, aOR 1.41 (1.25–1.59); epilepsy, aOR 1.33 (1.10–1.63); and active cancer, aOR 1.30 (1.14–1.49). Among initiators of cephalexin, pivmecillinam, amoxicillin/amoxicillin, and enzyme inhibitor and doxycycline, persons with AD were more frequently hospitalized than persons without AD. A quarter of hospitalized antibiotic initiators had infection diagnosis in their hospital care records.

Conclusions:

Persons with AD initiating an antibiotic had a higher risk for hospitalization than antibiotic initiators without AD. Further research is needed to determine whether infection-related hospitalization could be reduced.

Keywords

Alzheimer’s disease antibiotic dementia hospitalization infection pharmacoepidemiology

INTRODUCTION

Persons with Alzheimer’s disease (AD) use more healthcare services compared to general older population [1 –3]. Infections, for example pneumonia and urinary tract infections (UTI), are common causes for hospitalization in persons with dementia [1 , 4–8]. In addition to elevated healthcare costs, hospitalization is also a risk for further cognitive and functional decline in older cognitively impaired persons [9, 10]. Moreover, infections [11] and long hospital stay [12] are associated with a higher risk of delirium which in turn may cause functional and cognitive decline and other adverse outcomes [13, 14].

In Finnish primary health care, approximately 43–51% of patients with infection-related visits to physician are prescribed antibiotics [15]. In the United States, hospitalization from infection-related causes comprised 12–19% of all hospitalizations in adults over 65 years old and main causes were infections of lower respiratory tract (46%), the urinary tract (16%), septicemia (12%), and cellulitis (8%) [16]. In a previous study, during 2 year follow-up, 35% of older community-dwelling persons were hospitalized and 44% of them because of an infection [17]. Pneumonia and UTI covered over half of all infections.

To our best knowledge, there are no previous studies investigating hospitalization associated with antibiotic use in persons with AD. Therefore, the purpose of this study was to investigate the frequency of hospitalization associated with oral antibiotic initiation among community dwellers with and without AD. Furthermore, our aim was to investigate differences in antibiotic drugs in relation to hospitalization and in specific infections in persons with and without AD.

METHODS

Cohort

This study was based on the MEDALZ (Medication use and Alzheimer’s disease) cohort which included all community dwellers who received a clinical diagnosis of AD during the years 2005–2011 in Finland (n = 70,718) [18]. We identified patients with AD from a nationwide Special Reimbursement Register, which is maintained by the Social Insurance Institution of Finland (SII). The register contains information on entitlement to special reimbursements of drugs for chronic illnesses. The SII monitors the diagnostic procedure and diagnoses of AD are based on the NINCDS-ADRDA [19] and DMS-IV [20] criteria including a computed tomography or magnetic resonance imaging scan and confirmation of the diagnosis by a neurologist or a geriatrician [21]. Persons with AD were matched with comparison persons without AD according to age, sex, and region of residence at the date of AD diagnosis and this date was assigned as index date for persons with AD and comparison persons. According to the Finnish law, ethics committee approval or informed consent is not required when the data is de-identified and study participants are not contacted.

Study design

A 3-month washout period for oral antibiotic use prior to the index date was applied and persons using any oral antibiotic during this time period were excluded (Fig. 1). Only persons initiating oral antibiotic use after the index date were included in this study and the term antibiotic in this study refers to oral antibiotics only and their use in outpatient setting. Furthermore, we excluded persons discharged from hospital within 14 days prior to antibiotic initiation. These exclusions were done to identify new antibiotic users and to concentrate on community-acquired infections. After exclusions, the study population included 34,785 antibiotic initiators with AD and 36,428 antibiotic initiators without AD.

Fig.1

Flowchart of the formation of the study sample. AD, Alzheimer’s disease.

Antibiotics

We obtained data on drug use from the Prescription Register (1995–2015) which includes information on all purchased and reimbursed drugs from Finnish pharmacies. Drugs were classified according to the WHO Anatomical Therapeutic Chemical (ATC) classification system [22]. Antibiotics were defined as drugs from ATC class J01, excluding amphenicols (J01BA), aminoglycoside antibacterials (J01G), and combinations of antibacterials (J01RA). These exclusions were applied as these antibiotics are not used orally or are not prescribed for community dwellers. Over-the-counter (OTC) drugs and drugs used during stays in hospitals or public nursing homes are not included in these data; however, antibiotics are not available OTC in Finland. During the study period, certain antibacterials were not reimbursed and therefore could not be investigated. These antibacterials were nitrofurantoin (J01XE01) and methenamine (J01XX05). Only the first purchase of any antibiotic after the index date was considered for each person.

We utilized a mathematical method “From prescription drug purchases to drug use periods” (PRE2DUP) for modelling drug use periods for each person and each ATC code, i.e., when continuous drug use started and ended [23]. This method was based on individual purchase histories and takes into account individual variation in regularity of purchases, hospital stays and stockpiling of drugs. PRE2DUP has been previously validated [24] and utilized in similar study designs [25, 26]. Modelled drug use was utilized when defining use during washout period (i.e., whether persons had ongoing antibiotic use or not), whereas the first antibiotic initiation during the study period was defined based on the first dispensing date.

Hospitalization and infection diagnoses

The main outcome in this study was any hospitalization within 14 days after initiation of oral antibiotic. We obtained data on hospitalization and discharge diagnoses from the Hospital Discharge Register (1972–2015). Diagnoses were based on the International Statistical Classification of Diseases and Related Health Problems 10th revision (ICD-10). Infection-related diagnoses were collected from all the main and additional diagnoses and grouped to following categories: intestinal infectious diseases (A00-A09), tuberculosis (A15-A19), certain zoonotic bacterial diseases (A20-A28), other bacterial diseases (A30-A49), infections of the eye, adnexa, ear and mastoid process (H00-H01, H60, H66 and H70), infections of the heart and pericardium (I30 and I40.9), acute upper respiratory infections (J00-J06), influenza and pneumonia (J10-J18), other acute lower respiratory infections (J20-J22), suppurative and necrotic conditions of the lower respiratory tract (J85-J86), infections of the mouth (K04, K10.2-K10.3, K11.23, K12.2 and K14), infections of the gastrointestinal tract (K35, K61, K65 and K81), infections of the skin and subcutaneous tissue (L00-L08), infections of the musculoskeletal system and connective tissue (M00-M01, M46 and M86), infectious tubulo-interstitial nephritis (N10-N12), and genital infections (N30, N41, N45.0, N48.1-N48.2, N70-N74, N75.1 and N76.4).

Covariates

We obtained comorbidities at the time of antibiotic initiation from the Special Reimbursement Register and the Hospital Discharge Register. Comorbidities retrieved from the Special Reimbursement Register (1972–2015) included diabetes, asthma or chronic obstructive pulmonary disease (COPD), rheumatoid arthritis or other connective tissue diseases (rheumatoid arthritis), cardiovascular disease and epilepsy. Cardiovascular disease was defined as one or more of the following: chronic heart failure, arterial hypertension, coronary artery disease, or chronic arrhythmia. Data on active cancer was combined from the Prescription and Hospital Discharge Registers as any cancer diagnoses (recorded main or additional diagnoses) or use of antineoplastic drugs within 12 months before the antibiotic initiation, as described previously [27].

We obtained data on the history of psychiatric diseases from the Hospital Discharge Register including one of the following: schizophrenia, schizotypal or delusional disorders (F20-29), bipolar disorder (F30-31), or depression (F32-39). Diagnoses were based on ICD-10 criteria. History of psychiatric diseases was defined at least five years before the index date to exclude prodromal symptoms of AD.

Data on the use of the following drugs at the time of antibiotic initiation was retrieved from the Prescription Register and modelled with PRE2DUP method: immunosuppressants (L04A), oral corticosteroids (H02AB), antipsychotics (N05A, excluding lithium N05AN01), benzodiazepines (N05BA, N05CD, N05CF), and antidepressants (N06A).

Statistical analyses

χ² test, independent samples T-test, and Mann-Whitney test were utilized to report the p-values for categorical and continuous variables. Logistic regression was used to estimate factors associated with hospitalization and reported as odds ratios (ORs) with 95% confidence intervals (CIs). Multivariable models were adjusted for: AD, age, gender, diabetes, asthma or COPD, rheumatoid arthritis, cardiovascular disease, epilepsy, antipsychotic use, benzodiazepine use, antidepressant use, history of psychiatric diseases, active cancer, immunosuppressant use, and oral corticosteroid use. Confidence intervals for percentages (hospitalization following initiation of specific antibiotic drugs) were calculated using Wilson method. Statistical analysis was performed using IBM SPSS Statistics for Mac (version 24, IBM Corporation, Armonk, NY, USA).

RESULTS

Study population

Altogether 7.9% (n = 2750) of antibiotic initiators with AD, and 5.6% (n = 2032) without AD were hospitalized within 14 days after antibiotic initiation (Table 1). The mean time from AD diagnosis or index date until first antibiotic initiation was 50 days shorter for antibiotic initiators with AD compared to non-AD initiators. Persons with AD initiating an antibiotic had diabetes (14.0% versus 12.6%, p < 0.0001), epilepsy (2.3 versus 1.4%, p < 0.0001), and history of psychiatric diseases (4.7% versus 3.6%, p < 0.0001) more often and were more likely to use antipsychotics (20.9% versus 3.6%, p < 0.0001), benzodiazepines (23.0% versus 21.6%, p < 0.0001) and antidepressants (25.8% versus 9.5%, p < 0.0001) than antibiotic initiators without AD. Antibiotic initiators with AD has less frequently asthma or COPD (8.5% versus 9.4%, p < 0.0001), rheumatoid arthritis (4.4 versus 4.9%, p = 0.001), cardiovascular disease (50.5% versus 51.6%, p = 0.003) and active cancer (3.9% versus 4.3%, p = 0.002) and they used oral glucocorticoids (4.5% versus 5.7%, p < 0.0001), and immunosuppressants (0.9% versus 1.3%, p < 0.0001) less frequently than antibiotic initiators without AD.

Table 1

Characteristics of antibiotic initiators with and without Alzheimer’s disease (AD)

	Antibiotic initiators with AD n = 34,785	Antibiotic initiators without AD n = 36,428	p
Age, mean (years)	79.6	79.3	<0.0001
SD	7.2	7.2	<0.0001
Gender, n (%)
Female	24,029 (69.1)	24,389 (67.0)	<0.0001
Male	10,756 (30.9)	12,039 (33.0)	<0.0001
Mean time from AD diagnosis/index date until antibiotic initiation (d, CI)	667 (660–674)	717 (710–724)	<0.0001
Hospitalized within 14 days after antibiotic initiation, n (%)	2750 (7.9)	2032 (5.6)	<0.0001
Comorbidities
Diabetes, n (%)	4883 (14.0)	4579 (12.6)	<0.0001
Asthma/COPD, n (%)	2941 (8.5)	3415 (9.4)	<0.0001
Rheumatoid arthritis, n (%)	1517 (4.4)	1785 (4.9)	0.001
Cardiovascular disease, n (%)	17,550 (50.5)	18,786 (51.6)	0.003
Epilepsy, n (%)	805 (2.3)	516 (1.4)	<0.0001
Active cancer, n (%)	1340 (3.9)	1571 (4.3)	0.002
History of psychiatric diseases, n (%)	1648 (4.7)	1325 (3.6)	<0.0001
Medication use
Antipsychotic use, n (%)	7273 (20.9)	1295 (3.6)	<0.0001
Benzodiazepine use, n (%)	7999 (23.0)	7851 (21.6)	<0.0001
Antidepressant use, n (%)	8973 (25.8)	3451 (9.5)	<0.0001
Immunosuppressant, n (%)	320 (0.9)	481 (1.3)	<0.0001
Oral glucocorticoid, n (%)	1553 (4.5)	2078 (5.7)	<0.0001

SD, standard deviation; COPD, chronic obstructive pulmonary disease.

AD was associated with higher risk of hospitalization (adjusted odds ratio 1.37, 95% Cl: 1.28–1.46) (Table 2). Other factors associated with hospitalization were older age, male gender, epilepsy, active cancer, rheumatoid arthritis, diabetes, asthma/COPD, cardiovascular disease, use of oral glucocorticoids, antipsychotics, benzodiazepines, and antidepressants.

Table 2

Factors associated with hospitalization within 14 days after oral antibiotic initiation

	Hospitalized n = 4782	Not hospitalized n = 66,431	Unadjusted OR (95% CI)	Adjusted OR (95% CI)^*–
AD, n (%)	2750 (57.5)	32,035 (48.2)	1.45 (1.37–1.54)	1.37 (1.28–1.46)
Age, mean (years)	81.1	79.3	1.04 (1.04–1.05)	1.04 (1.04–1.05)
Male gender, n (%)	1607 (33.6)	21,188 (31.9)	1.08 (1.02–1.15)	1.04 (1.04–1.05)
Comorbidites
Diabetes, n (%)	765 (16.0)	8697 (13.1)	1.26 (1.17–1.37)	1.22 (1.12–1.33)
Asthma/COPD, n (%)	491 (10.3)	5865 (8.8)	1.18 (1.07–1.30)	1.13 (1.02–1.25)
Rheumatoid arthritis, n (%)	289 (6.0)	3013 (4.5)	1.35 (1.20–1.53)	1.30 (1.13–1.49)
Cardiovascular disease, n (%)	2687 (56.2)	33,649 (50.7)	1.25 (1.18–1.33)	1.14 (1.07–1.21)
Epilepsy, n (%)	112 (2.3)	1209 (1.8)	1.29 (1.06–1.57)	1.33 (1.10–1.63)
Active cancer, n (%)	263 (5.5)	2648 (4.0)	1.40 (1.23–1.60)	1.30 (1.14–1.49)
History of psychiatric diseases, n (%)	238 (5.0)	2735 (4.1)	1.22 (1.07–1.40)	1.13 (0.99–1.31)
Medication use
Antipsychotic use, n (%)	774 (16.2)	7794 (11.7)	1.45 (1.34–1.58)	1.28 (1.18–1.39)
Benzodiazepine use, n (%)	1218 (25.5)	14,632 (22.0)	1.21 (1.13–1.29)	1.09 (1.02–1.17)
Antidepressant use, n (%)	971 (20.3)	11,453 (17.2)	1.22 (1.14–1.32)	1.08 (1.00–1.16)
Immunosuppressant, n (%)	59 (1.2)	742 (1.1)	1.11 (0.85–1.44)	0.89 (0.66–1.18)
Oral glucocorticoid, n (%)	347 (7.3)	3284 (4.9)	1.50 (1.34–1.69)	1.41 (1.25–1.59)

AD, Alzheimer’s disease; OR, odds ratio; COPD, chronic obstructive pulmonary disease; CI, confidence interval. ^*Adjusted for AD, age, gender, diabetes, asthma or COPD, rheumatoid arthritis, cardiovascular disease, epilepsy, active cancer, history of psychiatric diseases, antipsychotic use, benzodiazepine use, antidepressant use, immunosuppressant use and oral glucocorticoid use.

Equal proportion of hospitalized antibiotic initiators with and without AD had an infection diagnosis in their hospital discharge records; 689 persons with AD (25.1% of hospitalized antibiotic initiators with AD) and 537 persons without AD (26.4% of hospitalized antibiotic initiators without AD). There were no major differences in infection related diagnoses between hospitalized persons with and without AD (Table 3). Most frequent specific infection diagnoses were pneumonia (ICD-10 code J18) in 208 persons with AD (7.6%) and 148 persons without AD (7.3%), acute infectious tubulo-interstitial nephritis (ICD-10 code N10) in 105 persons with AD (3.8%) and 70 persons without AD (3.4%) and erysipelas (ICD-10 code A46) in 52 persons with AD (1.9%) and 83 persons without AD (4.1%).

Table 3

Infection-related diagnoses in hospitalized antibiotic initiators with and without Alzheimer’s disease (AD)

ICD class	ICD class name	Persons with AD N = 2750, n (%)	Persons without AD N = 2032, n (%)	p
A00-A09	Intestinal infectious diseases	64 (2.3)	45 (2.2)	0.796
A15-A19	Tuberculosis	1 (0.0)	0 (0.0)	–
A20-A28	Certain zoonotic bacterial diseases	0 (0.0)	3 (0.1)	–
A30-A49	Other bacterial diseases	97 (3.5)	108 (5.3)	0.003
H *	Infections of the eye, adnexa, ear and mastoid process	5 (0.2)	1 (0.0)	0.200
I †	Infections of the heart and pericardium	0 (0.0)	0 (0.0)	–
J00-J06	Acute upper respiratory infections	32 (1.2)	34 (1.7)	0.135
J10-J18	Influenza and pneumonia	256 (9.3)	193 (9.5)	0.825
J20-J22	Other acute lower respiratory infections	81 (2.9)	55 (2.7)	0.623
J85-J86	Suppurative and necrotic conditions of lower respiratory tract	0 (0.0)	0 (0.0)	–
K ‡	Infections of the mouth	4 (0.1)	1 (0.0)	0.309
K §	Other infections of the GI-tract	9 (0.3)	8 (0.4)	0.703
L00-L08	Infections of the skin and subcutaneous tissue	21 (0.8)	12 (0.6)	0.475
M \|\|	Infections of the musculoskeletal system and connective tissue	3 (0.1)	1 (0.0)	0.479
N10-N12	Infectious tubulo-interstitial nephritis	115 (4.2)	74 (3.6)	0.343
N ¶	Genital infections	1 (0.0)	2 (0.1)	0.397

^*H00-H01, H60, H66, H70.

^†I30, I40.9.

^‡K04, K10.2-K10.3, K11.23, K12.2, K14.

^§K35, K61, K65, K81.

^||M00-M01, M46, M86.

^¶N30, N41, N45.0, N48.1-N48.2, N70-N74, N75.1, N76.4.

Antibiotic use and hospitalization

Among all antibiotic initiators, cephalexin, pivmecillinam, and amoxicillin/amoxicillin and enzyme inhibitor composed 62.8% (n = 44,735) of all antibiotics (Table 4). Almost two thirds of all hospitalizations occurred following initiation of these antibiotics in both persons with AD (n = 1753, 63.7%) and without AD (n = 1238, 60.9%). The most frequently used antibiotic among hospitalized antibiotic initiators with AD was pivmecillinam (28.4%) and cephalexin (26.4%) in persons without AD.

Table 4

Specific oral antibiotics in the study population and among those hospitalized during 14 days after the initiation, in persons with and without Alzheimer’s disease (AD)

Drug (ATC code)	All antibiotic initiatorsn = 71,213, %	Antibiotic initiators with AD			Antibiotic initiators without AD
		Antibiotic initiatorsn = 34,785,n (%)	Hospitalizedn = 2750,n (%)	% of initiatorshospitalized(95% CI)	Antibioticinitiatorsn = 36,428n (%)	Hospitalizedn = 2032 n (%)	% of initiatorshospitalized(95% CI)
Cephalexin (J01DB01)	18,772 (26.4)	9225 (26.5)	683 (24.8)	7.4 (6.9–8.0)	9547 (26.2)	537 (26.4)	5.6 (5.2–6.1)
Pivmecillinam (J01CA08)	15,842 (22.2)	9565 (27.8)	782 (28.4)	8.2 (7.6–8.7)	6186 (17.0)	399 (19.6)	6.5 (5.9–7.1)
Amoxicillin or amoxicillin and enzyme inhibitor (J01CA04,J01CR02)	10,121 (14.2)	4168 (12.0)	288 (10.5)	6.9 (6.2–7.7)	5953 (16.3)	302 (14.9)	5.1 (4.5–5.7)
Doxycycline (J01AA02)	6783 (9.5)	2489 (7.2)	191 (7.0)	7.7 (6.7–8.8)	4294 (11.8)	190 (9.4)	4.4 (3.9–5.1)
Phenoxymethylpenicillin (J01CE02)	5531 (7.8)	2225 (6.4)	78 (2.8)	3.5 (2.8–4.4)	3306 (9.1)	88 (4.3)	2.7 (2.2–3.3)
Trimethoprim (J01EA01)	3613 (5.1)	2192 (6.3)	217 (7.9)	9.9 (8.7–11.2)	1421 (3.9)	107 (5.3)	7.5 (6.3–9.0)
Ciprofloxacin (J01MA02)	3373 (4.7)	1733 (5.0)	204 (7.4)	11.8 (10.3–13.4)	1640 (4.5)	165 (8.1)	10.1 (8.7–11.6)
Sulfadiazine and trimethoprim (J01EE02)	1016 (1.4)	532 (1.5)	66 (2.4)	12.4 (9.9–15.5)	484 (1.3)	35 (1.7)	7.2 (5.3–9.9)
Levofloxacin (J01MA12)	746 (1.0)	460 (1.3)	63 (2.3)	13.7 (10.9–17.1)	286 (7.9)	40 (2.0)	14.0 (10.4–18.5)

For antibiotic initiators with and without AD, the risk of hospitalization was highest following levofloxacin (13.7% and 14.0% of users were hospitalized, respectively) and ciprofloxacin (11.8% and 10.1% of users were hospitalized, respectively) initiation. In addition, antibiotic initiators with AD were more frequently hospitalized following cephalexin, pivmecillinam, amoxicillin/amoxicillin, and enzyme inhibitor and doxycycline than persons without AD treated with the same antibiotic.

DISCUSSION

To our best knowledge there are no previous studies investigating antibiotic use associated with hospitalization in persons with AD. Our new finding is that persons with AD had about 40% increased relative risk of hospitalization following antibiotic initiation compared with persons without AD. Other factors strongly associated with hospitalization were several pre-existing somatic comorbidities, oral glucocorticoid use, and psychotropic use. Persons with AD initiating cephalexin, pivmecillinam, amoxicillin/amoxicillin, and enzyme inhibitor and doxycycline were more frequently hospitalized than persons without AD initiating the same antibiotics.

According to our study, AD was the main factor associated with hospitalization and also previous studies have found that persons with AD and/or dementia have higher risk of hospitalization from any cause compared to general older population [1, 3, 5, 6]. It has been shown that infections exacerbate existing chronic illnesses in persons with heart failure [28], type 2 diabetes [29], and asthma [30] leading to more frequent hospitalizations. Similarly, we found that immunosuppressive conditions and drug use, e.g., rheumatoid arthritis and oral glucocorticoid use, were associated with higher risk of hospitalization.

Infections are a common precipitating factor for delirium in older persons [11, 31] and cognitive disorders are a well-known predisposing factor for delirium. Especially, UTI is a risk factor for delirium in persons with AD [32]. Infections may cause other adverse symptoms like dehydration [33] and falls associated with hospitalization. In addition, persons with cognitive impairment and dementia are at a higher risk of falls [34, 35]. The use of benzodiazepines [36], antipsychotics [37], and antidementia drugs [38] are known to be a risk factor for pneumonia and hospitalization. In addition, psychotropic drug use is also known to be a risk factor for falls [39, 40]. During infection, persons with cognitive impairment need comprehensive good care to avoid injurious falls and other adverse events like delirium.

Previous studies have found that persons with dementia have a higher risk for hospitalization due to UTI than cognitively intact persons [4, 6]. In our study there was a similar trend seen in the use of pivmecillinam as it is exclusively indicated only for UTI. Therefore, the higher chance of being hospitalized following UTI might partly be due to delirium symptoms. In our study, antibiotics used in the treatment of UTI, for example pivmecillinam, and trimethoprim, were more commonly used by antibiotic initiators with AD. Differences in hospitalization rates in antibiotic initiators with and without AD for cephalexin, pivmecillinam, amoxicillin/amoxicillin, and enzyme inhibitor and doxycycline might be explained by delirium symptoms triggered by infection especially in the case of pivmecillinam [32, 41]. In addition, as these antibiotics were the four most commonly initiated antibiotics among persons with and without AD it is possible that there was less statistical power for other, less frequently used antibiotics.

In our study, levofloxacin and ciprofloxacin users had highest risk of hospitalization in both groups. Explanation for that might be that both antibiotics are broad-spectrum antibiotics and indicated for more severe infections. The official indications of levofloxacin include sinusitis, community-acquired pneumonia, complicated skin and soft tissue infections, infectious exacerbation of chronic bronchitis and complicated UTI, as per on label instructions [42]. Ciprofloxacin is indicated for exacerbation of COPD, UTI, and infections of the gastro-intestinal tract.

Strengths and limitations

This study was based on nationwide registers which cover all residents regardless of their socioeconomic status and thus, selection bias was minimal. The data in these registers have been collected for decades and provide good coverage and valid information on AD diagnoses [43, 44]. Consequently, these results can be directly applied to community dwellers with AD elsewhere. Furthermore, antibiotics require a prescription in Finland and are not available as OTC drugs. A limitation in our study is that the Special Reimbursement Register does not provide information on the severity of AD and thus, we were not able to analyze antibiotic use and hospitalization in different stages of the disease. In addition, the Prescription Register contains data on reimbursed drug purchases only and therefore purchases of, for example, nitrofurantoin were missing from our data. This is a major limitation in our study since nitrofurantoin is considered as one of the first line treatment options of UTI along with pivmecillinam and trimethoprim. The prevalence of antibiotic use is underestimated in the register-based data. Similarly, the registers do not contain indications for drug use and diagnosis of delirium in the Hospital Discharge Register is not thorough and therefore could not be investigated.

Conclusions

Among oral antibiotic initiators, persons with AD were more likely hospitalized than persons without AD. In both groups, comorbidities like diabetes, rheumatoid arthritis, and epilepsy and use of antipsychotics, benzodiazepines, and oral glucocorticoids were associated with an increased risk of hospitalization. Further research is needed to find practices to prevent infections in older persons and especially in persons with cognitive disorders and so reduce need for infection-related hospitalization.

Footnotes

ACKNOWLEDGMENTS

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors’ disclosures available online ().

References

Albert

, Costa

, Merchant

, Small

, Jenders

, Stern

(1999) Hospitalization and Alzheimer’s disease: Results from a community-based study. J Gerontol A Biol Sci Med Sci 54, 267.

Tolppanen

, Taipale

, Purmonen

, Koponen

, Soininen

, Hartikainen

(2015) Hospital admissions, outpatient visits and healthcare costs of community-dwellers with Alzheimer’s disease. Alzheimers Dement 11, 955-963.

Zhao

, Kuo

, Weir

, Kramer

, Ash

(2008) Healthcare costs and utilization for Medicare beneficiaries with Alzheimer’s. BMC Health Serv Res 8, 108.

Natalwala

, Potluri

, Uppal

, Heun

(2008) Reasons for hospital admissions in dementia patients in Birmingham, UK, during 2002-2007. Dement Geriatr Cogn Disord 26, 499-505.

Malone

, McLaughlin

, Wahl

, Leibman

, Arrighi

, Cziraky

, Mucha

(2009) Burden of Alzheimer’s disease and association with negative health outcomes. Am J Manag Care 15, 481-488.

Phelan

, Borson

, Grothaus

, Balch

, Larson

(2012) Association of incident dementia with hospitalizations. JAMA 307, 165-172.

Guijarro

, Roman

CMS

, Gomez-Huelgas

, Villalobos

, Martin

, Guil

, Martinez-Gonzalez

, Toledo

(2010) Impact of dementia on hospitalization. Neuroepidemiology 35, 101-108.

Sampson

, Blanchard

, Jones

, Tookman

, King

(2009) Dementia in the acute hospital: Prospective cohort study of prevalence and mortality. Br J Psychiatry 195, 61-66.

Pedone

, Ercolani

, Catani

, Maggio

, Ruggiero

, Quartesan

, Senin

, Mecocci

, Cherubini

, Group

(2005) Elderly patients with cognitive impairment have a high risk for functional decline during hospitalization: The GIFA Study. J Gerontol A Biol Sci Med Sci 60, 1576-1580.

10.

Wilson

, Hebert

, Scherr

, Dong

, Leurgens

, Evans

(2012) Cognitive decline after hospitalization in a community population of older persons. Neurology 78, 950-956.

11.

George

, Bleasdale

, Singleton

(1997) Causes and prognosis of delirium in elderly patients admitted to a district general hospital. Age Ageing 26, 423-427.

12.

Ahmed

, Leurent

, Sampson

(2014) Risk factors for incident delirium among older people in acute hospital medical units: A systematic review and meta-analysis. Age Ageing 43, 326-333.

13.

Fick

, Agostini

, Inouye

(2002) Delirium superimposed on dementia: A systematic review. J Am Geriatr Soc 50, 1723-1732.

14.

Fong

, Jones

, Marcantonio

, Tommet

, Gross

, Habtemariam

, Schmitt

, Yap

, Inouye

(2012) Adverse outcomes after hospitalization and delirium in persons with Alzheimer disease. Ann Intern Med 156, 848-856, W296.

15.

Rautakorpi

, Huikko

, Honkanen

, Klaukka

, Makela

, Palva

, Roine

, Sarkkinen

, Varonen

, Huovinen

, MIKSTRA Collaborative Study Group (2006) The Antimicrobial Treatment Strategies (MIKSTRA) program: A 5-year follow-up of infection-specific antibiotic use in primary health care and the effect of implementation of treatment guidelines. Clin Infect Dis 42, 1221-1230.

16.

Curns

, Holman

, Sejvar

, Owings

, Schonberger

(2005) Infectious disease hospitalizations among older adults in the United States from 1990 through 2002. Arch Intern Med 165, 2514-2520.

17.

Ruben

, Dearwater

, Norden

, Kuller

, Gartner

, Shalley

, Warshafsky

, Kelsey

, O’Donnell

, Means

(1995) Clinical infections in the noninstitutionalized geriatric age group: Methods utilized and incidence of infections. The Pittsburgh Good Health Study. Am J Epidemiol 141, 145-157.

18.

Tolppanen

, Taipale

, Koponen

, Lavikainen

, Tanskanen

, Tiihonen

, Hartikainen

(2016) Cohort profile: The Finnish Medication and Alzheimer’s disease (MEDALZ) study. BMJ Open 6, 12100.

19.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939-944.

20.

American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders, American Psychiatric Association, Washington, DC.

21.

Finnish Medical Society Duodecim (2017) Current care: Memory Disorders.

22.

(2017) WHO Collaborating Centre for Drug Statistics Methodology: The Anatomical Therapeutic Chemical (ATC) classification system. Oslo, Norway.

23.

Tanskanen

, Taipale

, Koponen

, Tolppanen

, Hartikainen

, Ahonen

, Tiihonen

(2015) From prescription drug purchases to drug use periods - a second generation method (PRE2DUP). BMC Med Inform Decis Mak 15, 21.

24.

Taipale

, Tanskanen

, Koponen

, Tolppanen

, Tiihonen

, Hartikainen

(2016) Agreement between PRE2DUP register data modeling method and comprehensive drug use interview among older persons. Clin Epidemiol 8, 363-371.

25.

Hamina

, Taipale

, Tanskanen

, Tolppanen

, Tiihonen

, Hartikainen

(2017) Differences in analgesic use in community-dwelling persons with and without Alzheimer’s disease. Eur J Pain 21, 658-667.

26.

Saarelainen

, Tolppanen

, Koponen

, Tanskanen

, Sund

, Tiihonen

, Hartikainen

, Taipale

(2017) Risk of hip fracture in benzodiazepine users with and without Alzheimer disease.. J Am Med Dir Assoc 18, 87.e21.

27.

Hamina

, Taipale

, Tanskanen

, Tolppanen

, Karttunen

, Pylkkanen

, Tiihonen

, Hartikainen

(2017) Long-term use of opioids for nonmalignant pain among community-dwelling persons with and without Alzheimer disease in Finland: A nationwide register-based study. Pain 158, 252-260.

28.

Alon

, Stein

, Korenfeld

, Fuchs

(2013) Predictors and outcomes of infection-related hospital admissions of heart failure patients. PLoS One 8, e72476.

29.

Hamilton

, Martin

, Makepeace

, Sillars

, Davis

(2013) Incidence and predictors of hospitalization for bacterial infection in community-based patients with type 2 diabetes: The fremantle diabetes study. PLoS One 8, e60502.

30.

Teichtahl

, Buckmaster

, Pertnikovs

(1997) The incidence of respiratory tract infection in adults requiring hospitalization for asthma. Chest 112, 591-596.

31.

Holtta

, Laurila

, Laakkonen

, Strandberg

, Tilvis

, Pitkala

(2014) Precipitating factors of delirium: Stress response to multiple triggers among patients with and without dementia. Exp Gerontol 59, 42-46.

32.

Lerner

, Hedera

, Koss

, Stuckey

, Friedland

(1997) Delirium in Alzheimer disease. Alzheimer Dis Assoc Disord 11, 16-20.

33.

Lavizzo-Mourey

(1987) Dehydration in the elderly: A short review. J Natl Med Assoc 79, 1033-1038.

34.

Tinetti

, Speechley

, Ginter

(1988) Risk factors for falls among elderly persons living in the community. N Engl J Med 319, 1701-1707.

35.

van Dijk

, Meulenberg

, van de

Sande HJ

, Habbema

(1993) Falls in dementia patients. Gerontologist 33, 200-204.

36.

Taipale

, Tolppanen

A-M

, Koponen

, Tanskanen

, Lavikainen

, Sund

, Tiihonen

, Hartikainen

(2017) Risk of pneumonia associated with incident benzodiazepine use among community-dwelling adults with Alzheimer disease.. CMAJ 189, E519-E529.

37.

Tolppanen

A-M

, Koponen

, Tanskanen

, Lavikainen

, Sund

, Tiihonen

, Hartikainen

, Taipale

(2016) Antipsychotic use and risk of hospitalization or death due to pneumonia in persons with and those without Alzheimer disease. Chest 150, 1233-1241.

38.

Lampela

, Tolppanen

A-M

, Tanskanen

, Tiihonen

, Lavikainen

, Hartikainen

, Taipale

(2017) Use of antidementia drugs and risk of pneumonia in older persons with Alzheimer’s disease. Ann Med 49, 230-239.

39.

Leipzig

, Cumming

, Tinetti

(1999) Drugs and falls in older people: A systematic review and meta-analysis: I. Psychotropic drugs. J Am Geriatr Soc 47, 30-39.

40.

Perttila

, Ohman

, Strandberg

, Kautiainen

, Raivio

, Laakkonen

, Savikko

, Tilvis

, Pitkala

(2017) How do community-dwelling persons with Alzheimer disease fall? Falls in the FINALEX Study. Dement Geriatr Cogn Dis Extra 7, 195-203.

41.

Laurila

, Laakkonen

, Tilvis

, Pitkala

(2008) Predisposing and precipitating factors for delirium in a frail geriatric population. J Psychosom Res 65, 249-254.

42.

Finnish Medicines Agency Summary of Product Characteristics (SPC). 2017.

43.

Solomon

, Ngandu

, Soininen

, Hallikainen

, Kivipelto

, Laatikainen

(2014) Validity of dementia and Alzheimer’s disease diagnoses in Finnish national registers. Alzheimers Dement 10, 303-309.

44.

Wettermark

, Zoega

, Furu

, Korhonen

, Hallas

, Norgaard

, Almarsdottir

, Andersen

, Sundell

, Bergman

, Helin-Salmivaara

, Hoffmann

, Kieler

, Martikainen

, Mortensen

, Petzold

, Wallach-Kildemoes

, Wallin

, Sorensen

(2013) The Nordic prescription databases as a resource for pharmacoepidemiological research–a literature review. Pharmacoepidemiol Drug Saf 22, 691-699.