Clinical Articles Cigarette smoking and urinary incontinence in women Richard C. Bump, MD: and Donna K. McClish, PhD b Richmond, Virginia OBJECTIVE: The goal of this case control study was to evaluate the relationship between smoking and female urinary incontinence. STUDY DESIGN: The study included 606 women whose smoking histories were known; 322 were incontinent and 284 were continent. The condition(s) causing each subject's incontinence was determined by urodynamic testing; 40% of the continent subjects had the same testing. RESULTS: There were highly significant overall differences (p = 0.000009) in the distribution of current, former, and never smokers between incontinent (35%, 16%,49%) and continent (24%, 8%, 68%) groups. The odds ratio for genuine stress incontinence was 2.20 for former (95% confidence interval 1.18 to 4.11) and 2.48 for current smokers (95% confidence interval 1.60 to 3.84); for motor incontinence it was 2.92 for former (95% confidence interval 1.58 to 5.39) and 1.89 (95% confidence interval 1.19 to 3.02) for current smokers. Increasing daily and lifetime cigarette consumption was associated with an increasing odds ratio for genuine stress incontinence but not for motor incontinence. The increased risk for incontinence was not due to differences in age, parity, weight, or hypoestrogenic status. CONCLUSION: The data establish a strong statistical relationship between current and former cigarette smoking and both stress and motor urinary incontinence in women. (AM J OBSTET GVNECOL 1992;167: 1213-8.)
Key words: Urinary incontinence, cigarette smoking, stress incontinence, motor incontinence Cigarette smoking and urinary incontinence have been recognized as two critical but independent health problems. Since the United States Surgeon General's first report in 1964, cigarette smoking has been identified as the single most important source of preventable morbidity and premature mortality in the United States. The direct health care costs attributable to smoking were estimated to exceed $16 billion in 1985. 1 Urinary incontinence afflicts at least 10 million adult Americans, two thirds of whom are women. With an estimated prevalence of 15% to 30% for older, community-dwelling persons and 50% for nursing home residents, the annual cost of managing urinary incontinence was conservatively estimated at $10.3 billion in 1988! Although clinicians involved in the care of women with urinary incontinence often attest to the importance of smoking as a contributing factor, mainly to From the Departments of Obstetrics and Gynecology" and Biostatistics, b Medical College of Virginia, Virginia Commonwealth University. Received for publication March 30,1992; revised April 14, 1992; accepted April 15, 1992. Reprint requests: Richard Bump, MD, Department of Obstetrics and Gynecology, Box 34, MCV Station, Richmond, VA 23298. 611138741
stress incontinence and its recurrence after successful continence surgery,' a relationship between smoking and incontinence is not recognized in the general medicalliterature or in various governmental reports concerning the two conditions. The 1988 National Institutes of Health Consensus Development Conference on adult urinary incontinence noted that defining risk factors for incontinence would be extremely useful for prevention or remediation of the disorder, but smoking was not incuded in the list of established or suspected risk factors. 2 Likewise, incontinence is not cited as one of the many health consequences of cigarette smoking in recent authoritative reviews on the subject. I, 4, 5 Using a computerized literature search we were unable to locate a single study that documented a link between smoking and incontinence. The primary goal of this retrospective case-control study was to examine the relationship between cigarette smoking and urinary incontinence in adult women.
Subjects and methods The study population included 606 women who were seen by R.C.B. for urogynecologic assessment. Mean age was 46 years (SE 0.5, range 13 to 87 years), and 1213
1214 Bump and McClish
mean parity was 2.7 (SE 0.08, range 0 to 12). Three hundred twenty-two subjects had urinary incontinence as the primary complaint and were consecutive referrals for urodynamic testing between July 1985 and June 1990. This incontinent group was divided by urodynamic diagnosis according to criteria discussed later. A total of 284 subjects had no complaints of urinary incontinence. Of these, 113 subjects were consecutive referrals during the same time period for urodynamic testing to evaluate other symptoms of lower urinary tract dysfunction such as urgency, frequency, hesitancy, or retention. These subjects are designated as the abnormal continent group. The other 171 continent subjects had no symptoms of lower urinary tract dysfunction and either were seen for routine gynecologic care or were referred for evaluation of gynecologic symptoms unrelated to pelvic organ prolapse. These subjects are designated as the normal continent group. These normal continent subjects were allocated consecutively from all general gynecologic patients seen by R.C.B. during the same 5-year period according to criteria that allowed this group to approximate the age and parity mixture of the incontinent subjects. These criteria mandated that the number of nulliparous continent subjects not exceed the number of nulliparous incontinent subjects and that the proportion of continent and incontinent subjects in each half decade of life be similar. These criteria eliminated primarily young nulliparous women from the normal continent group. Subjects were asked about their lifetime smoking histories, including their current packs-per-day consumption, lifetime consumption (in pack years, calculated by multiplying the average pack-per-day consumption by the number of smoking years), and (for former smokers) the time interval since quitting. A subject was classified as a current smoker if she was actively smoking, as a former smoker if she had previously smoked regularly but had stopped smoking for >2 months, and as a never smoker if she had smoked regularly for
Key words: Urinary incontinence, cigarette smoking, stress incontinence, motor incontinence Cigarette smoking and urinary incontinence have been recognized as two critical but independent health problems. Since the United States Surgeon General's first report in 1964, cigarette smoking has been identified as the single most important source of preventable morbidity and premature mortality in the United States. The direct health care costs attributable to smoking were estimated to exceed $16 billion in 1985. 1 Urinary incontinence afflicts at least 10 million adult Americans, two thirds of whom are women. With an estimated prevalence of 15% to 30% for older, community-dwelling persons and 50% for nursing home residents, the annual cost of managing urinary incontinence was conservatively estimated at $10.3 billion in 1988! Although clinicians involved in the care of women with urinary incontinence often attest to the importance of smoking as a contributing factor, mainly to From the Departments of Obstetrics and Gynecology" and Biostatistics, b Medical College of Virginia, Virginia Commonwealth University. Received for publication March 30,1992; revised April 14, 1992; accepted April 15, 1992. Reprint requests: Richard Bump, MD, Department of Obstetrics and Gynecology, Box 34, MCV Station, Richmond, VA 23298. 611138741
stress incontinence and its recurrence after successful continence surgery,' a relationship between smoking and incontinence is not recognized in the general medicalliterature or in various governmental reports concerning the two conditions. The 1988 National Institutes of Health Consensus Development Conference on adult urinary incontinence noted that defining risk factors for incontinence would be extremely useful for prevention or remediation of the disorder, but smoking was not incuded in the list of established or suspected risk factors. 2 Likewise, incontinence is not cited as one of the many health consequences of cigarette smoking in recent authoritative reviews on the subject. I, 4, 5 Using a computerized literature search we were unable to locate a single study that documented a link between smoking and incontinence. The primary goal of this retrospective case-control study was to examine the relationship between cigarette smoking and urinary incontinence in adult women.
Subjects and methods The study population included 606 women who were seen by R.C.B. for urogynecologic assessment. Mean age was 46 years (SE 0.5, range 13 to 87 years), and 1213
1214 Bump and McClish
mean parity was 2.7 (SE 0.08, range 0 to 12). Three hundred twenty-two subjects had urinary incontinence as the primary complaint and were consecutive referrals for urodynamic testing between July 1985 and June 1990. This incontinent group was divided by urodynamic diagnosis according to criteria discussed later. A total of 284 subjects had no complaints of urinary incontinence. Of these, 113 subjects were consecutive referrals during the same time period for urodynamic testing to evaluate other symptoms of lower urinary tract dysfunction such as urgency, frequency, hesitancy, or retention. These subjects are designated as the abnormal continent group. The other 171 continent subjects had no symptoms of lower urinary tract dysfunction and either were seen for routine gynecologic care or were referred for evaluation of gynecologic symptoms unrelated to pelvic organ prolapse. These subjects are designated as the normal continent group. These normal continent subjects were allocated consecutively from all general gynecologic patients seen by R.C.B. during the same 5-year period according to criteria that allowed this group to approximate the age and parity mixture of the incontinent subjects. These criteria mandated that the number of nulliparous continent subjects not exceed the number of nulliparous incontinent subjects and that the proportion of continent and incontinent subjects in each half decade of life be similar. These criteria eliminated primarily young nulliparous women from the normal continent group. Subjects were asked about their lifetime smoking histories, including their current packs-per-day consumption, lifetime consumption (in pack years, calculated by multiplying the average pack-per-day consumption by the number of smoking years), and (for former smokers) the time interval since quitting. A subject was classified as a current smoker if she was actively smoking, as a former smoker if she had previously smoked regularly but had stopped smoking for >2 months, and as a never smoker if she had smoked regularly for
