Prescription Drugs

1
Prescription Drugs Lower Urinary Tract
SymptomsA pharmacoepidemiologic analysis among
men

• Susan A. Hall, Ph.D.
• New England Research Institutes, Watertown, MA
  02472 USA
• shall_at_neriscience.com NIDDK DK-56842

2
Background

• Prescription drugs a common exposure
• 52 of U.S. adults use 1 prescription drug
  weekly
• 83 of men 65, 30 use 5 weekly
• Overall drug use and polypharmacy (5 meds) ?
  over time Raofi 2006, Kaufman 2005
• Potential intersection with LUTS through age
• Urologic symptoms as an iatrogenic consequence of
  drug use?

3
Lower urinary tract symptoms (LUTS)

• Age-related, occur commonly have large impact
  on QOL
• Commonly measured using the American Urologic
  Association Symptom Index scale
• Storage (irritative) Voiding (obstructive)
• Frequency -Incomplete emptying
• Urgency -Difficulty starting/stopping
• Nocturia -Weak stream
• -Straining

4
Prior Epidemiology Studies

• SSRIs nocturia in Swedish community study
  Asplund 2005
• Other cardiac drugs nocturia in Tierp, Sweden
  community study (65) Rembratt 2003
• Antidepressants, antihistamines LUTS in Olmsted
  County, MN USA study (men 40-79) Su, 1996
• Generally, data are sparse

5
Study Objective

• To understand the association between 25
  commonly-used
• prescription medications
• urologic health
• (voiding and storage LUTS)
• among community-dwelling men

6
Boston Area Community Health (BACH) Survey Design

• Epidemiologic study of urologic symptoms
• Community-dwelling men women
• Randomly selected, cluster-stratified sample of
  Boston, MA, U.S.A.
• N2,301 men
• Broad age range (30-79y)
• Diversity of race/ethnicity/socioeconomic
• Approx. equal Ns of black/white/Hispanic
• Data collected 2002-05, in-home interview

7
BACH Measurements

• Outcome AUA Symptom Index
• Voiding disorders Score of 5 (range 0 to 20)
• Storage disorders Score of 4 (range 0 to 15)
• Exposure Medication collection methodology
• Drug inventory bottle collection/label
  transcription
• Query with prompts by indication
• Use in the past 4 weeks captured
• Risk factors
• Comorbidities-self report of physician diagnosis
• Depression (abridged CES-D scale)
• Body mass index (derived from interviewer
  measures)

8
Analysis Sample Methods

• Analysis sample
• Exclusions Missing data on outcomes (18)
• Taking outcome-related medications (152)
• alpha blockers or 5-alpha reductase inhibitors,
  urinary incontinence or overactive bladder meds,
  anticholinergics
• 2,131 men in analysis sample
• 2,077 for models (no missing data on covariates)
• Estimating prevalence of exposure
• Prevalence of drug use for 25 popular
  prescription medications
• Weighted by the inverse of probability of sampling

9
Multivariate Analysis Methods

• Preliminary models
• Preliminary logistic regression model 25 drugs
  2 outcomes
• Adjusted only for age and race
• Odds ratio (OR) and 95 confidence interval
• Large significant associations ( 2.0 OR) further
  investigated
• Fully-adjusted logistic regression models
• Included risk factors comorbidities
• Fully-adjusted linear regression models
• Using full scale for outcome
• Included risk factors comorbidities

Surviving associations
10
Results
11
Characteristics of analysis sample
12
Prevalence of exposure, Use of 25 prescription
drugs, N2,131 men
13
Outcomes defined using American Urologic
Association scale Voiding Score of 5 for 4
symptoms (incomplete emptying, starting/stopping,
weak stream, straining, range 0-20)
Storage Score of 4 for 3 symptoms (frequency,
urgency, nocturia, range 0-15) LUTS Score of 8
for all symptoms above (range 0-35)
14
Multivariate Model ResultsVoiding Disorders
No relationship 20 drugs
1st logistic model Age, race adjusted
statins, narcotics, benzodiazepines, SSRIs, oral
anti-diabetics
2nd logistic model Fully-adjusted for risk
factors comorbidities
Adjusted for age, race, depression, history of
UTI or kidney infection, and other
comorbidities.
statins, narcotics
No relationships
3rd model-Linear Fully-adjusted for risk factors
comorbidities
15
Final associations-Voiding disorders
16
Multivariate Model ResultsStorage Disorders
No relationship 22 drugs
1st logistic model Age, race adjusted
SSRIs, corticosteroids, loop diuretics
statins, narcotics
2nd logistic model Fully-adjusted for risk
factors comorbidities
Adjusted for age, race, BMI, depression,
history of UTI or kidney infection, diabetes
and other comorbidities.
statins, corticosteroids
statins steroids
3rd model-Linear Fully-adjusted for risk factors
comorbidities
17
Final associations-Voiding and Storage
18
Discussion-Statins

• Statins protective against storage disorders
• Urologic adverse events in atorvastatin clinical
  trials
• UTI reported 2 of participants
• lt2 urinary frequency, retention, urgency,
  dysuria, nocturia incontinence
• Recent clinical trial of statins to improve LUTS
  non-efficacious Mills IW, Euro Urol, 2007
• More severe LUTS/BPH population (score of 13)
• No separate treatment of voiding vs. storage
• Mechanism? Effects on smooth muscle, improvement
  in bladder fibrosis
• Novel finding that should be replicated

19
Discussion-Corticosteroids

• Positively associated with storage in both models
  (nearly 3-fold increase)
• Rare reported adverse events of the drug class
• Urinary frequency, urgency enuresis
• Asthma meds included
• Relaxation of smooth muscle of bladder?

20
Strengths Limitations

• Broad range of medications considered
• Duration of use, compliance unknown
• Drugs used only in doctors office not captured
• Compliance unknown
• Large diverse population of community-dwelling-men
  
• Ability to understand influence of other risk
  factors

21
Implications and Future Research

• Some reassurance that most popular prescription
  drugs were not associated with voiding storage
  disorders
• Corticosteroids appear to aggravate storage
  disorders among men
• Statins result should be examined in future
  studies
• Clinical trial of storage disorders