SEARCH

SEARCH BY CITATION

Keywords:

  • overactive bladder;
  • central nervous system;
  • cardiovascular;
  • co-morbidities;
  • medications

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Study Type – Therapy (individual cohort)

Level of Evidence 2b

What's known on the subject? and What does the study add?

It has been documented that currently used anti-muscarinics can cause adverse effects on the CNS.

This study adds knowledge that the OAB population has more CNS disorders that make them vulnerable to the potential CNS effects of OAB anti-muscarinics.

OBJECTIVE

  • • 
    To determine the proportion of patients with overactive bladder (OAB) potentially at risk for adverse events by assessing their pre-existing central nervous system (CNS), cardiovascular (CV) and other co-morbidities.

PATIENTS AND METHODS

  • • 
    The GE Centricity Electronic Medical Record database was utilized to identify patients with a diagnosis of OAB using International Classification of Diseases, Ninth Revision (ICD-9) codes or a prescription between 1 January 1996 and 30 March 2007 for an OAB anti-muscarinic agent.
  • • 
    Matched non-OAB patients were assigned the same index date as the corresponding OAB patient. Based on the presence of ≥ one pharmacy claim for an OAB anti-muscarinic agent, the OAB cohort was stratified as treated or untreated. A random sample of age- and gender-matched patients formed a non-OAB control cohort.
  • • 
    An additional and separate analysis focusing on all co-morbidities was performed examining non-OAB patients who were matched to OAB patients on 1:1 propensity score matching, based on age, body mass index (BMI) and gender at baseline.
  • • 
    Charlson Comorbidity Index (CCI), using ICD-9 codes, and the Chronic Disease Score (CDS), using prescribed drugs, were calculated.

RESULTS

  • • 
    When compared with non-OAB patients (N= 77 272; 83.2% women; median age 64 years), OAB patients (N= 41 440; 83.6% women; median age 65 years) had more overall CNS co-morbidities (45.4 vs 29.0%; P < 0.001).
  • • 
    In addition, OAB patients had a higher use of medications with anti-muscarinic effects (39.6 vs 25.4%; P < 0.001). OAB patients were also more likely to have CV co-morbidities (57.6 vs 44.6%; P < 0.001).
  • • 
    CNS co-morbidities were slightly more common in untreated (n= 8 106) than in treated (n= 33 334) OAB patients (47.2 vs 45.0%; P < 0.001). CV co-morbidities were higher in treated OAB patients (58.8 vs 53.7%; P < 0.001).
  • • 
    In the additional separate analysis, which focused on all co-morbidities, patients with OAB had higher mean CCI and CDS scores than patients without OAB (CCI: 1.17 vs 1.11 [P < 0.001]; CDS: 2.95 vs 1.74 [P < 0.001]).
  • • 
    After controlling for other covariates, the linear regressions (n= 22 544) showed that OAB patients had higher CCI and CDS than patients without OAB.

CONCLUSIONS

  • • 
    Among OAB patients, CNS, CV and all co-morbidities were more prevalent than in non-OAB patients.
  • • 
    Prior exposure to CNS medications was more prevalent in OAB patients who received anti-muscarinic treatment than in those who did not.
  • • 
    Co-morbidities and concomitant medications affecting the CNS and the CV system should be taken into account when making the decision on the most appropriate OAB treatment option for each individual patient.

Abbreviations
BMI

body mass index

CV

cardiovascular

CVS

cardiovascular system

CCI

Charlson Comorbidity Index

CDS

Chronic Disease Score

GE EMR

GE Centricity Electronic Medical Record database

HR

heart rate

OAB

overactive bladder.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Anti-muscarinic drugs are the standard treatment of the overactive bladder (OAB) syndrome [1]. This therapy has well-known side-effects, such as dry mouth and constipation. Of more concern are the potential adverse effects on the CNS and the cardiovascular system (CVS) [2]. Patients with OAB have previously been shown to have more cardiovascular (CV) morbidities than an age-matched control group [2], and this has focused interest on the importance of co-morbidities for the adverse effects of the whole group of anti-muscarinic drugs. Several investigations have documented that currently used anti-muscarinics can cause adverse effects on the CNS [3–6]. However, it is not known whether the OAB population has more (predisposing) CNS disorders that could make them more vulnerable than non-OAB patients to the potential CNS effects of OAB anti-muscarinics. Also of concern is the possibility of increasing the total anti-muscarinic burden in OAB patients, especially if they are on concomitant other drugs with anti-muscarinic effects.

To assess the overall co-morbidity burden, the Charlson Comorbidity Index (CCI), using International Classification of Diseases, Ninth Revision (ICD-9) codes, and the Chronic Disease Score (CDS) are often used. The CCI is the most commonly employed index to measure and summarize health conditions. The CDS uses pharmacy dispensation information for specific drug classes to estimate the burden of co-morbidities. The CDS is the sum of all chronic diseases identified from drug therapies over the full follow-up period.

The objective of the present study was to determine, based on data from a large electronic medical record database (the GE Centricity Electronic Medical Record database [GE EMR]), the proportion of OAB patients potentially at risk for adverse events. CNS and CV co-morbidities were assessed using CNS and CV diagnoses and concomitant drug use among treated and untreated OAB patients. Overall pre-existing co-morbidities were assessed by CCI and CDS.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

The GE EMR was adopted for the present study using ICD-9 codes or a prescription between 1 January 1996 and 30 March 2007 for an OAB anti-muscarinic agent. A graphical representation of the inclusion/exclusion criteria is shown in Fig. 1. From the GE EMR, all patients ≥18 years of age with a diagnosis of OAB (International Statistical Classification of Diseases and Related Health Problems [ICD]-9: 596.51, 596.54, 596.55, 596.59, 788.31, 788.33, 788.34, 788.39) or at least one prescription order for an OAB drug (tolterodine, oxybutynin, trospium, solifenacin or darifenacin) between 1 January 1996 and 30 March 2007 were identified. The date of the first prescription order in GE EMR history or the date of first diagnosis (whichever occurred first if patients had both a diagnosis of OAB and drug treatment) was defined as the index date. Among these patients, continuously eligible patients from 395 days before the index date to 395 days after the index date were selected. The 395-day pre-index and post-index period was selected to allow for a 1-month gap between when an annual prescription expired and a new prescription order was documented in the database. To avoid confounding by rare and severe diseases, individuals with the diagnoses of HIV/AIDS during the same time period were excluded. To assess whether the proportion of pre-existing CNS and CV conditions differed from what would be expected, a non-OAB cohort was created with the same inclusion/exclusion criteria. This non-OAB cohort was matched to the OAB cohort using the propensity score matching technique, a statistic tool of matching to reduce selection bias by finding the balance between OAB and non-OAB groups and, consequently, increase robustness [7,8].

image

Figure 1. Identification of OAB and non-OAB patients in the GE EMR study.

Download figure to PowerPoint

For each OAB patient identified, three patients without OAB (1:3 matching between OAB and non-OAB) were matched by age and gender (n= 41 440 with OAB vs n= 124 320 without OAB). The non-OAB patients had no diagnosis of OAB, any indication of urinary bladder dysfunction, or pharmacy claim for an OAB anti-muscarinic agent. The matched non-OAB patients were assigned the same index date as the corresponding OAB patient. After an index date was assigned, non-OAB patients with <395 days continuous eligibility before and after the index date were excluded. After applying the inclusion/exclusion criteria, 37.84% of non-OAB patients were excluded in the analysis (Fig. 1).

From the population of OAB patients, all users with greater than one prescription order for an OAB drug were identified as treated patients. The earliest prescription of the OAB drug during the study period was defined as the index date. The patients were required to have ≥395 days of continuous eligibility before the index date. During this 395-day continuous eligibility period, a patient could not have any prescriptions for OAB drugs, which ensured that all treated patients identified in the study were newly initiated on treatment (n= 33 334). Patients with a diagnosis of OAB, but without a prescription of an OAB drug during the study period were identified as untreated patients (n= 8106).

To identify the number and percentage of patients with pre-existing CNS conditions at any time before or on the index date, ICD-9 codes and medications identified using Generic Product Identifier (GPI) related to CNS medications were selected. Patients were required to have ≥ one CNS-related medication or ICD-9 codes for CNS diagnoses at any time before or on the index date to be identified as having a pre-existing CNS condition. The specific CNS conditions included are described in Table 1.

Table 1.  Description of the conditions
CNS
 Psychosis, dementia, neurotic disorders, personality disorders and other non-psychotic mental disorders
 Inflammatory diseases of the CNS
 Hereditary and degenerative diseases of the CNS
 Alzheimer's disease, senility without mention of psychosis
 Syncope and dizziness
 Specific CNS-related medications
 Propoxyphene (mono or combination), indomethacin, pentazocine, flurazepam, meprobamate, lorazepam, oxazepam, alprazolam, temazepam, triazolam, chlordiazepoxide, diazepam, quazepam, halazepam, clorazepate/chlorazepate, methyldopa, reserpine, barbiturates, meperidine, amphetamines, fluoxetine, methyltestosterone, clonidine, and cimetidine
CV
 Beta-blockers
 Calcium channel blockers
 Angiotensin-converting enzyme (ACE) inhibitors
 Angiotensin II receptor blockers (ARBs), other anti-hypertensives
 Nitrates, digoxin, platelet aggregators
 Anti-arrhythmics and anti-hyperlipidaemics

The same method as for CNS conditions was used to identify pre-existing CV conditions. The specific CV conditions included are described in Table 1.

The database was used to identify the number and percentage of patients exposed to medications that could add to CNS and CV risk through increased overall anti-muscarinic burden during the pre-index period. The lists of the medication codes were identified from an updated version of the Beer's criteria [9]. The Beer's criteria are a list of individual medications or classes of medications that need to be avoided in older adults.

A separate analysis was conducted with additional, more current data available in the GE dataset to assess overall co-morbidity burden in OAB and non-OAB patients examined between 1 January 1996 and 30 March 2007. The overall co-morbidity burden was assessed using the CCI and CDS. The CCI is the most commonly used index to measure and summarize health conditions. The CCI contains 19 categories of co-morbidity, defined primarily using ICD-9 diagnosis codes [10,11]. Each category has an associated weight based on the adjusted risk of 1-year mortality [10]. The overall co-morbidity score reflects the cumulative increased likelihood of 1-year mortality. The CCI has been popularly used and validated in research [12–16].

Another commonly used co-morbidity index is the CDS. The CDS uses pharmacy dispensation information for specific drug classes to estimate the burden of co-morbidities [17]. The CDS is a risk-adjusted metric based on age, gender and history of dispensed drugs. This score is an aggregate co-morbidity measure based on medication use. It has been proven to be valid in predicting hospitalization, health resource utilization and mortality [18–23]. The CDS is the sum of all chronic diseases identified from drug therapies over the full follow-up period.

The index date for OAB patients in the present study was defined as the date of their first prescription (activity date) for an anti-muscarinic agent or a diagnosis for OAB identified by ICD-9 codes. The index date of non-OAB patients without diagnosis or pharmacy claim was defined as 1 year after the first activity date in the GE EMR. Patients >18 years old were included and they had 395 days of continuous enrolment before and after the index date. The study excluded those with a diagnosis of HIV/AIDS. Non-OAB patients were matched to OAB patients on 1:1 propensity score matching based on age, body mass index (BMI) and gender at baseline. The difference between the groups (OAB vs non-OAB) was checked to confirm that all variables were not statistically different. Two linear regressions were constructed using the outcome variables of the CCI, using ICD-9 codes, and CDS, using prescribed drugs, respectfully.

For statistical analysis, all OAB patients were compared with the non-OAB patients to assess the prevalence of CNS and CV co-morbidities and prior exposure to drugs with anti-cholinergic (anti-muscarinic) effects in the OAB patients. A similar comparison was made between the treated and the untreated OAB patients to assess if their baseline CNS and CV co-morbid conditions and anti-muscarinic medication use were considered by the prescribing physician when selecting the OAB medications for their patient. Statistical comparisons were conducted using the t-test for continuous variables and the chi-squared test for categorical variables. Linear regression analyses were constructed, with CCI being the dependent variable in one regression model and CDS serving as the dependent variable in the other regression model.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

A total of 41 440 OAB patients from the GE EMR database were eligible for analysis, along with a total of 77 272 matched non-OAB patients (Fig. 1). The 41 440 OAB patients comprised subsets of 33 334 treated vs 8106 matched, untreated OAB patients. Among OAB patients, a total of 83.6% were women and 53.6% were ≥65 years at the index date. Similarly, 83.2% of non-OAB patients were women and 49.7% were ≥65 years (see Tables 2,3)

Table 2.  Baseline characteristics of OAB and non-OAB patients
CharacteristicsOAB patientsNon-OAB patientsP
  1. DBP, diastolic blood pressure; SBP, systolic blood pressure.

Total number of patients41 44077 272 
Women, %83.6%83.2%0.040
Age, years (mean ±sd[median])62.47 ± 13.46 (65)61.46 ± 13.58 (64)0.002
Age ≥ 65 years, %53.5749.74<0.001
Clinical characteristics   
 Baseline HR, beats/min (mean ±sd[median])75.67 ± 11.15 (76)75.11 ± 11.50 (74)<0.001
Patients' HR, %   
 61–70 beats/min22.9224.27<0.001
 71–80 beats/min43.5442.160.001
 81–90 beats/min16.3115.11<0.001
 91–100 beats/min6.255.870.028
Higher HR (≥80 beats/min)31.3619.62<0.001
Baseline SBP (mean ±sd[median])130.65 ± 18.25 (130)129.57 ± 18.07 (130)0.004
Baseline DBP (mean ±sd[median])75.74 ± 10.34 (76)75.33 ± 10.25 (76)<0.001
BMI (mean ±sd[median]), kg/m230.15 ± 7.31 (28.86)29.07 ± 6.73 (27.93)<0.001
Framingham risk score9.939.62<0.001
Table 3.  Prevalence of pre-existing CV and CNS co-morbidities in OAB and non-OAB patients
ConditionOAB patients, % (N= 41 440)Non-OAB patients, % (N= 77 272)P
CV co-morbidities   
 Any CV co-morbidity57.6044.62<0.001
 Congenital heart disease0.300.270.010
 Ischaemic heart disease10.808.92<0.001
 Other forms of heart disease6.305.22<0.001
 Conduction disorders8.507.28<0.001
 Heart failure3.602.71<0.001
 CV symptoms5.604.56<0.001
 Hypertension44.2032.00<0.001
 Hypotension1.200.62<0.001
 Pulmonary heart disease1.000.69<0.001
 Cerebrovascular disease7.504.39<0.001
 Disease of blood vessels5.504.04<0.001
 Diabetes15.5010.96<0.001
 Renal diseases3.302.33<0.001
CNS co-morbidities   
 Any CNS co-morbidities45.4029.00<0.001
 Psychosis (includes dementia)8.604.62<0.001
 Dementia0.900.50<0.001
 Neurotic disorders, personality disorders and other non-psychotic mental disorders35.3022.47<0.001
 Inflammatory diseases of the CNS0.200.06<0.001
 Hereditary and degenerative diseases of the CNS (including Alzheimer's disease and senility without psychosis)5.502.82<0.001
 Alzheimer's disease0.700.47<0.001
 Senility without mention of psychosis0.000.000.093
 Syncope and dizziness8.305.57<0.001

The characteristics of treated and untreated OAB patients are shown in Tables 4 and 5. Among treated OAB patients, a total of 83.0% were women and 53.6% were 65 years of age or older at index date. On the other hand, 86.4% of untreated OAB patients were women, and 40.3% were ≥65 years at index date.

Table 4.  Baseline characteristics of treated and untreated OAB patients
CharacteristicsTreated OAB patientsUntreated OAB patientsP
  1. DBP, diastolic blood pressure.

Total patients33 3348106 
Women, %83.0086.40<0.001
Age, years (mean ±sd[median])62.47 ± 13.46 (66)58.61 ± 14.03 (59)<0.001
Age ≥ 65 years, %53.5740.29<0.001
Clinical characteristics   
 Baseline HR, beats/min (mean ±sd[median])75.72 ± 11.19 (76)75.41 ± 10.98 (76)0.046
Patients' HR, %   
 61–70 beats/min22.8123.460.272
 71–80 beats/min43.4943.720.743
 81–90 beats/min16.3716.040.532
 91–100 beats/min6.345.80.112
Baseline DBP (mean ±sd[median])75.83 ± 10.32 (77)75.37 ± 10.41 (76)0.001
BMI (mean ±sd[median]), kg/m230.22 ± 7.42 (28.89)29.77 ± 7.10 (28.49)<0.001
Framingham risk score10.218.57<0.001
Table 5.  Prevalence of pre-existing CV and CNS co-morbidities in treated and untreated OAB patients
ConditionTreated OAB patients, % (N= 33 334)Untreated OAB patients, % (N= 8106)P
CV co-morbidities   
 Any CV co-morbidity58.8053.66<0.001
 Congenital heart disease0.300.360.894
 Ischaemic heart disease11.109.45<0.001
 Other forms of heart disease6.206.590.191
 Conduction disorders8.807.44<0.001
 Heart failure3.603.520.668
 CV symptoms5.505.970.099
 Hypertension45.3039.61<0.001
 Hypotension1.201.060.384
 Pulmonary heart disease1.000.940.406
 Cerebrovascular disease8.005.66<0.001
 Disease of blood vessels5.505.270.322
 Diabetes16.0013.45<0.001
 Renal diseases3.303.200.713
CNS co-morbidities   
 Any CNS co-morbidities45.0047.190.000
 Psychosis (includes dementia)8.508.930.187
 Dementia1.000.730.027
 Neurotic disorders, personality disorders and other non-psychotic mental disorders34.8037.790.000
 Inflammatory diseases of the CNS0.200.200.447
 Hereditary and degenerative diseases of the CNS (including Alzheimer's disease and senility without psychosis)5.505.480.939
 Alzheimer's disease0.700.850.176
 Senility without mention of psychosis0.000.000.393
 Syncope and dizziness8.407.660.029

The baseline clinical characteristics of patients with and without OAB are shown in Table 2. On the index date or closest to index date, the mean (±sd) baseline heart rate (HR) in all OAB patients was 75.7 ± 11.2 beats/min compared with 75.1 ± 11.5 beats/min in non-OAB patients (P≤ 0.001). A total of 31.4% of OAB patients had HR ≥80 beats/min on the index date or closest to index date, compared with 19.6% among non-OAB patients (P≤ 0.001). Overall, OAB patients had a higher Framingham risk score than non-OAB patients (9.9 vs 9.6; P≤ 0.001) and a higher baseline BMI (30.2 ± 7.3 vs 29.1 ± 6.7 kg/m2, P < 0.001). The baseline clinical characteristics of treated vs untreated OAB patients are shown in Table 4. The mean (±SM) baseline HRs in treated and untreated OAB patients were 75.7 ± 11.2 and 75.4 ± 11.0 beats/min, respectively (P= 0.046).

When compared with non-OAB patients, all patients with OAB had a higher proportion of CNS disorders at baseline overall (45.4 vs 29.0%; P≤ 0.001) (Table 3). The percentage of patients with OAB and pre-existing psychosis, including dementia, was higher than in those without OAB (8.6 vs 4.6%, P≤ 0.001). Similarly, pre-existing neurotic disorders, personality disorders and other nonpsychotic mental disorders were more prevalent in OAB patients than in non-OAB patients (35.3 vs 22.5%; P≤ 0.001). These disorders were the most common baseline CNS pre-existing co-morbidities in both OAB and non-OAB patients. Other statistically significant pre-existing CNS co-morbidities that were more prevalent in OAB than in non-OAB patients included inflammatory diseases of the CNS, hereditary and degenerative diseases of the CNS (including Alzheimer's disease and senility without psychosis), and syncope and dizziness (P≤ 0.001). The only baseline CNS co-morbidity that did not differ significantly between OAB patients and non-OAB patients was senility without mention of psychosis (0.0 vs 0.0%; P= 0.093).

When compared with non-OAB patients, all OAB patients had a higher proportion of pre-existing CV disorders at baseline overall (57.6 vs 44.6%; P < 0.001) as well as for each of the individual pre-existing CV conditions. Hypertension was the most common pre-existing condition (44.2% for OAB patients vs 32.0% for non-OAB patients; P < 0.001) (Table 3).

The use of drugs with known anti-muscarinic effects was higher in the OAB patients than in the non-OAB patients (39.6 vs 25.4%; P < 0.001) The most commonly prescribed drugs with anti-muscarinic effects were anti-histamines in both OAB patients (24.6%) and non-OAB patients (15.9%), followed by cyclobenzaprine (9.4% in OAB patients and 6.0% in non-OAB patients) and amitriptyline (6.6% in OAB patients and 3.6% in non-OAB patients). Surprisingly, a greater proportion of treated OAB patients had a history of using drugs with known anti-muscarinic effects than was the case among the untreated OAB patients (41.5 vs 31.8%; P < 0.001). Again, anti-histamines were the most commonly prescribed among drugs with anti-muscarinic effects (25.8% among treated OAB patients and 19.6% among untreated OAB patients).

The baseline prevalence of pre-existing CNS co-morbidities in treated and untreated OAB patients is shown in Table 5. Overall, untreated patients had a higher proportion of any CNS co-morbidity than did treated patients (47.2 vs 45.0%; P≤ 0.001). A higher percentage of untreated OAB patients were found to have pre-existing neurotic disorders, personality disorders and other nonpsychotic mental disorders than was the case among treated OAB patients (37.8 vs 34.8%, P≤ 0.001). Conversely, a higher percentage of treated OAB patients had dementia (1.0 vs 0.7%; P= 0.027), syncope and dizziness (8.4 vs 7.7%; P= 0.029) than was the case among untreated OAB patients. There were no statistically significant differences between the treated and untreated OAB patients with psychosis, inflammatory diseases of the CNS, and hereditary and degenerative diseases of the CNS (including Alzheimer's disease and senility without psychosis).

As mentioned in the ‘Patients and Methods’ section, an additional study of other pre-existing conditions was performed. There were 38 739 OAB patients (mean [±sd] age 61.2 ± 13.3 years; 85.7% women) and 38 739 matched non-OAB patients (mean [±sd] age 61.2 ± 13.2 years; 85.7% women). The age, gender and baseline BMI variables were checked using two-sample t-tests with equal variances. The distribution reflected that good matching had occurred.

Table 6 shows a comparison of the characteristics of OAB and non-OAB patients. Statistical significance was noted among race/ethnicity, insurance type and smoking status. Certain baseline clinical variables were also determined to be significant, with the exception of BMI (what the population was matched on). No significant differences were determined for age and gender, which was to be expected as they were used in matching criteria.

Table 6.  Characteristics of patients with and without OAB
VariableOAB (n= 38 739)Non-OAB (n= 38 739)P (OAB vs non-OAB)
N/mean%/sdN/mean%/SD
  1. DBP, diastolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Age (as continuous), years61.1813.2661.1713.240.906
 18–301.441.44 
 31–407.147.13 
 41–5015.5415.52 
 51–6426.6126.59 
 ≥6549.2749.32 
Gender  0.894
 Female85.6785.70 
 Male14.3314.30 
Race/ethnicity  0.000
 White34.8929.78 
 Black3.444.07 
 Hispanic1.071.10 
 Other0.981.20 
 Unknown59.6263.85 
Insurance type  0.000
 Commercial30.6730.92 
 Medicare43.9740.38 
 Medicaid1.711.41 
 Self pay0.490.89 
 Unknown23.1626.39 
Smoking status  0.000
 Never36.3529.46 
 Ever23.0218.77 
 Current9.708.08 
 Unknown30.9343.69 
Baseline clinical variables     
 Heart rate75.6111.2675.4211.660.030
 Total cholesterol198.0941.35198.9542.240.050
 HDL55.2016.2354.0815.740.000
 LDL114.8535.01117.3536.380.000
 SBP130.2318.02130.7918.710.000
 DBP75.8010.2276.3610.450.000
 BMI30.087.3030.087.230.992
 CCI1.171.151.111.090.000
 CDS2.953.191.742.630.000

As seen in Tables 6–8, patients with OAB had higher mean CCI and CDS scores than patients without OAB (CCI: 1.17 vs 1.11 [P < 0.001]; CDS: 2.95 vs 1.74 [P < 0.001]). After controlling for other covariates, the linear regressions (n= 22 544) showed that OAB patients had higher CCI and CDS scores than patients without OAB by 0.037 (P < 0.001) and by 0.881 (P < 0.001), respectively.

Table 7.  Linear regression results (dependent variable = CCI) (number of patients is 22 544)
CovariateCoefficientseP95% CI
  1. DBP, diastolic blood pressure; SDP, systolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; se, standard error; TC, triglycerides.

OAB status     
 Non-OABReference    
 OAB0.0370.0100.0000.0160.057
Age status     
 <65 yearsReference    
 ≥65 years1.4780.0130.0001.4521.505
Gender     
 MaleReference    
 Female−0.1250.0150.000−0.153−0.096
Race/ethnicity     
 CaucasianReference    
 African-American0.1550.0290.0000.0990.212
 Hispanic0.1660.0410.0000.0870.246
 Other0.0880.0470.060−0.0040.180
 Unknown−0.0110.0110.310−0.0330.010
Payment status     
 CommercialReference    
 Medicare0.2070.0140.0000.1790.235
 Medicaid0.0550.0450.214−0.0320.143
 Self-pay−0.0830.0630.182−0.2060.039
 Unknown0.0760.0150.0000.0470.105
Smoking status     
 Non-smokerReference    
 Past smoker0.0540.0130.0000.0270.080
 Current smoker−0.0080.0200.677−0.0470.030
 Unknown−0.0370.0130.003−0.062−0.013
Clinical variables     
 HR0.0000.0000.578−0.0010.001
 BMI−0.0020.0010.042−0.0030.000
 TC0.0010.0000.0140.0000.001
 HDL−0.0010.0000.003−0.0020.000
 LDL−0.0020.0000.000−0.003−0.002
 SBP0.0050.0000.0000.0040.006
 DBP−0.0080.0010.000−0.009−0.007
 Constant0.5860.0630.0000.4620.709
Table 8.  Linear regression results (dependent variable = CDS) – number of patients is 22 544
CovariateCoefficientseP95% CI
  1. DBP, diastolic blood pressure; SDP, systolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SE, standard error; TC, triglycerides.

OAB status     
 Non-OABReference    
 OAB0.8810.0400.0000.8020.959
Age status     
 <65 yearsReference    
 ≥65 years0.7670.0520.0000.6660.868
Gender     
 MaleReference    
 Female0.0310.0560.581−0.0790.140
Race/ethnicity     
 CaucasianReference    
 African-American0.6930.1110.0000.4760.911
 Hispanic−0.2740.1560.078−0.5790.030
 Other0.1300.1800.469−0.2230.484
 Unknown−0.3930.0420.000−0.476−0.310
Payment status     
 CommercialReference    
 Medicare0.4390.0550.0000.3310.547
 Medicaid1.2810.1710.0000.9451.617
 Self-pay0.6140.2400.0110.1431.084
 Unknown0.1910.0560.0010.0810.302
Smoking status     
 Non-smokerReference    
 Past smoker0.1040.0510.0430.0030.204
 Current smoker0.5960.0760.0000.4480.745
 Unknown−0.3070.0480.000−0.402−0.212
Clinical variables     
 HR0.0160.0020.0000.0130.020
 BMI0.0910.0030.0000.0860.097
 TC0.0100.0010.0000.0080.012
 HDL−0.0230.0020.000−0.026−0.020
 LDL−0.0230.0010.000−0.025−0.020
 SBP0.0180.0010.0000.0150.020
 DBP−0.0280.0020.000−0.033−0.024
 Constant−0.1720.2420.478−0.6460.303

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

The GE EMR contains a broad, clinically rich and geographically diverse spectrum of longitudinal medical and pharmacy data from health plans in the south-eastern, mid-Atlantic, central and western regions of the United States. The database represents over 100 consortium member institutions located in over 35 states, and a variety of practice types ranging from solo practitioners to community clinics, academic medical centres and large integrated delivery networks.

The present study found that 45.4% of the patients with OAB had pre-existing CNS conditions at baseline, a significantly higher proportion than the 29.0% of non-OAB patients with the same CNS conditions. When these CNS conditions were categorized, the relationship held, with OAB patients having a higher CNS co-morbidity rate than non-OAB patients for all categories (except senility without mention of psychosis). These data suggest a possible association between CNS conditions and OAB. A study by Hubeaux et al.[24] suggests that autonomic nervous system dysfunction might be a component of the pathophysiology of OAB. Therefore, it is possible that the mechanisms underlying the CNS co-morbidities included in the present study also affect components of the autonomic nervous system involved in the voiding process. Further studies are necessary to define the exact nature of this relationship.

The patients included in the present study represent a diverse population. Over half (57.6%) of the OAB patients had a higher proportion of pre-existing CV disorders at baseline overall as well as for each individual pre-existing CV condition, including 44.2% of OAB patients having hypertension at baseline. These findings are consistent with and substantiate the results from a previous study [2], which found that the prevalence of CV co-morbidities was significantly higher in patients with OAB than in those without OAB. The present study also found an increased exposure of OAB patients to medications with an anti-muscarinic effect, further supporting the idea that the presence of CV co-morbidity is not being considered before treating OAB patients with anti-muscarinic medications [25].

One finding of potential concern is that a higher proportion of the treated OAB patients had pre-existing syncope and dizziness than was the case among untreated OAB patients (8.4 vs 7.7%, P= 0.029). This is significant as one of the major treatment options for OAB is anti-muscarinic medications, which have side-effect profiles that can potentially worsen these pre-existing co-morbidities. A study by Herschorn et al.[25] found that 9% of patients treated with oxybutynin for overactive bladder symptoms experienced dizziness. In addition, we found that treated OAB patients are exposed to more medications with anti-muscarinic effects, including anti-histamines and cyclobenzaprine, than untreated patients (41.5 vs 31.9%, P≤ 0.001), potentially further worsening these patients' pre-existing CNS conditions. Since worsening of these conditions can lead to further morbidity (e.g. falls), careful consideration should occur when prescribing these medications.

The present study has some limitations. It was based on a retrospective analysis of administrative claims data, and thus inherent limitations of the claims data, such as coding errors and systematic factors upcoding, could have an impact on the results of the study. OAB often remains under-diagnosed, resulting in the potential inclusion of OAB patients within the non-OAB control cohort.

In conclusion, the results of this analysis indicate that CNS, CV and all co-morbidities are more prevalent among OAB patients than among non-OAB patients. Among the OAB patients, CNS co-morbidity and prior exposure to CNS medications were more prevalent in those who received anti-muscarinic treatment than in patients who did not receive treatment. CNS/CV co-morbidities and concomitant CNS/CV medication use should to be taken into account when making the decision as to the most appropriate OAB treatment option for each individual patient.

CONFLICT OF INTEREST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Funding for the study was provided by Novartis Pharmaceuticals Corporation.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES