Risk factors and comorbid conditions associated with lower urinary tract symptoms: EpiLUTS

Authors


Karin Coyne, PhD, Senior Research Leader, United BioSource Corporation, 7101 Wisconsin Ave, Suite 600, Bethesda, MD 20814, USA.
e-mail: Karin.Coyne@unitedbiosource.com

Abstract

OBJECTIVE

To explore the risk factors and comorbid conditions associated with subgroups of lower urinary tract symptoms (LUTS) in men and women aged ≥40 years in three countries, using data from the EpiLUTS study, as LUTS are common amongst men and women and increase in prevalence with age.

SUBJECTS AND METHODS

This cross-sectional, population-representative survey was conducted via the Internet in the USA, the UK and Sweden. Participants were asked to rate how often they experienced individual LUTS during the past 4 weeks on a 5-point Likert scale. Eight LUTS subgroups were created. Descriptive statistics and logistic regressions within each LUTS subgroup were used to assess the data.

RESULTS

The survey response rate was 59%. The final sample was 30 000 (men and women). The voiding + storage + postmicturition (VSPM) group reported the highest rates of comorbid conditions for both men and women, and the fewest were reported in the no/minimal LUTS and the postmicturition-only groups. Increasing age was associated with increasing LUTS in men, but not in women. Comorbid conditions significantly associated with the VSPM group were arthritis, asthma, chronic anxiety, depression, diabetes (men only), heart disease, irritable bowel syndrome, neurological conditions, recurrent urinary tract infection, and sleep disorders. Risk factors, such as body mass index, exercise level and smoking, played less of a role, except for childhood nocturnal enuresis, which was significantly associated with most LUTS subgroups.

CONCLUSION

In this large population study, many comorbid conditions and risk factors were significantly associated with LUTS among both men and women. Further longitudinal investigations of the associations noted here would help physicians to understand the pathophysiology of LUTS and comorbid conditions, and provide clinical guidelines for patient management of comorbid conditions sharing common pathophysiological pathways.

Abbreviations
BMI

body mass index

SUI

stress urinary incontinence

PM

postmicturition

VSPM

voiding + storage + PM

NHANES

National Health and Nutrition Examination Survey

OAB

overactive bladder

EpiLUTS

Epidemiology of LUTS

IC

interstitial cystitis

IBS

irritable bowel syndrome

HRT

hormone-replacement therapy

PBIS

painful bladder syndrome

INTRODUCTION

Increasingly many reports have examined the comorbid conditions and risk factors for LUTS in men and women; the most noted risk factor for LUTS in both genders is increasing age. Many comorbid conditions, such as diabetes, hypertension, and cardiovascular disease, are also associated with LUTS [1–6]. More recent research, particularly among men, has associated the metabolic syndrome with LUTS; hyperinsulinaemia, obesity, non-insulin-dependent diabetes, dyslipidaemia, and hypertension were found to be associated with an increased risk of BPH [4,7,8]. Given that many of the comorbid conditions noted are modifiable by increased physical activity and weight loss, it is not surprising that low levels of physical activity and elevated body mass index (BMI) are well-documented risk factors for LUTS in men and women [1,3,6,9–12]. Research on smoking as a risk factor for LUTS has been inconclusive, with the finding that current smokers are at greater risk of LUTS than are former smokers or non-smokers [4,6,13], and conversely reports of either no association or a possible inverse relationship between cigarette smoking and LUTS [2,4,14].

Female-specific comorbid conditions and risk factors for LUTS have also been examined. Women who have had a vaginal delivery are more likely than those who delivered via Caesarean section to report LUTS, especially stress urinary incontinence (SUI), even a year or more postpartum [10,15–17]. Having a hysterectomy has also been shown to increase the likelihood of LUTS, including hesitancy, incomplete emptying, postmicturition (PM) dribble, frequency, urgency, and urgency incontinence [1,10,18,19]. Postmenopausal status and longer duration of peri-menopause have also been associated with storage and voiding LUTS, including SUI, nocturia, and a weak stream [20–22].

Although data from population-based samples have been used to study LUTS, the studies have lacked symptom-specific data (e.g. National Health and Nutrition Examination Survey, NHANES [4]) or were restricted to certain LUTS such as overactive bladder (OAB, e.g. EPIC) [23]. Also, most studies examining risk factors and comorbid conditions associated with LUTS have used the IPSS (for men) or the Bristol Female LUTS questionnaire for women. Because the IPSS is limited to seven LUTS questions, not all LUTS have been considered in these previous population-based studies. It has been suggested that a more expansive view of LUTS, considering the collection of symptoms in its entirety rather than focusing on individual gender- and organ-specific disorders, would improve the clinical recognition and management of LUTS [24]. The objective of the present analysis was to further explore the risk factors and comorbid conditions associated with LUTS, and specifically LUTS subgroups, using data from the Epidemiology of LUTS (EpiLUTS) study.

SUBJECTS AND METHODS

An Internet-based, cross-sectional, population-based survey was administered between June 2007 and April 2008 in the USA, the UK and Sweden. A complete description of the EpiLUTS study rationale, design and survey details are described elsewhere [25,26]. In brief, the EpiLUTS study examined the full range of LUTS as well as potential risk factors and comorbid conditions in a large, population-based sample of men and women aged ≥40 years from the three countries.

LUTS were defined using ICS definitions [27], with language modifications to increase lay-person understanding of the questions. LUTS assessed were urinary frequency, urinary urgency, nocturia, incontinence (stress, urgency, mixed, nocturnal enuresis, leaking during sexual activity, PM, and leaking for no reason), weak stream, terminal dribble, hesitancy, straining, intermittency, split stream, incomplete emptying, bladder pain, and dysuria. The response options for most LUTS were on a 5-point Likert scale (‘never’, ‘rarely’, ‘sometimes’, ‘often’, and ‘almost always’).

In addition to sociodemographic questions (age, gender, race/ethnicity, marital status, education), participants were asked several questions to assess comorbid conditions and risk factors for LUTS. Specifically, participants were asked if they had ever been told by a doctor or other healthcare provider that they had the following conditions: arthritis, asthma, diabetes, high blood pressure, heart disease, bladder cancer, interstitial cystitis (IC) or painful bladder syndrome, recurrent UTIs, neurological conditions, depression, chronic anxiety, irritable bowel syndrome (IBS), or sleep apnoea/sleep disorder. Men were asked if they had been diagnosed with an enlarged or inflamed prostate, prostatitis, or prostate cancer; women were asked if they had a uterine prolapse or hysterectomy.

To assess risk factors associated with LUTS, participants were asked about their height and weight, history of childhood nocturnal enuresis, previous and current cigarette use, daily physical activity, and frequency of moderate and vigorous physical activity. (Activity questions from the NHANES survey were used.) Women were also asked about number of births, current menopausal status, and use of hormone-replacement therapy (HRT).

Sample matching was used to construct population-representative samples of respondents within each country’s Internet-based panel, and poststratification weights were calculated to correct small amounts of imbalance based on differences in response rates. As such, all presented data are weighted.

Eight LUTS subgroups were created: no/minimal LUTS, voiding only, storage only, PM only, voiding + storage, voiding + PM, storage + PM, and voiding + storage + PM (VSPM). Group classification was determined two ways: the proportion of participants who responded at least ‘sometimes’ (‘sometimes’, ‘often’ and ‘almost always’) for one or more of the symptoms within each LUTS subgroup and those who responded at least ‘often’ (‘often’ and ‘almost always’). Exceptions to this group classification were questions that did not use the 5-point Likert response, including the questions for nocturia (which was continuous, but counted as symptom presence when there were two or more nocturia episodes); daytime frequency (presence was a ‘yes’ response to the question ‘Do you feel that you urinate too often during the day?’) and urinary incontinence (presence was a ‘yes’ response to one or more from a list of situations in which people can leak urine and a response of at least ‘a few times a month’ to the follow-up frequency question).

Descriptive statistics were used to present demographic characteristics, comorbid conditions and risk factors. All data are presented separately for men and women and by LUTS subgroup for the more conservative at least ‘sometimes’ group (effects noted in these analyses are more pronounced in the at least ‘often’ group). Logistic regressions were used to identify associations between LUTS subgroups and comorbid conditions and risk factors, respectively. Regression models were used separately for men and women, with the dependent variable being the specific LUTS subgroup vs the no/minimal LUTS subgroup. Covariates included in each model were age, race and country, followed by each of the assessed comorbid conditions (yes/no) and presence/absence of LUTS risk factors, which included BMI, smoking status (current, ex-, non-), moderate activity, vigorous activity, and childhood nocturnal enuresis. Additional risk factors for women included uterine prolapse (yes/no), menopausal status (yes/no), and parity. Given the multiple analyses and large sample size, P < 0.01 was considered to indicate significance in the regression models.

RESULTS

In all, 30 000 men and women aged ≥40 years participated in this epidemiological study. Among men, the mean age by subgroup ranged from 53.1 (PM only) to 60.3 (voiding + storage) years (Table 1). The mean age among women did not vary widely (55.0–57.8 years) among the LUTS subgroups, with the oldest mean age in the storage-only group. Because of the representative sampling design in each country, most of the men and women were white.

Table 1.  Demographics of participants by LUTS occurring at least sometimes *
n (%) variableNo LUTSV onlyS onlyPM onlyV + SV + PMS + PMV + S + PMP
  • *

    All numbers presented are weighted; subgroups might or might not equal total n because of rounding or weighted values. V, voiding; S, storage; PM, postmicturition.

Men, n409717041280427144914632883433 
Mean (sd) age, years  53.9 (9.9)  55.5 (10.3)  57.2 (11.4) 53.1 (8.6)  60.3 (11.6)  55.0 (10.2) 55.7 (10.7)  59.4 (11.2)<0.001
Race, n (%)         
 White3377 (82.4)1465 (86.0) 943 (73.7)359 (84.0)1201 (82.9)1282 (87.7)214 (74.2)2854 (83.1)<0.001
 Black 252 (6.2)  87 (5.1) 142 (11.1) 18 (4.1) 102 (7.1)  48 (3.3) 28 (9.9) 222 (6.5) 
 Asian 124 (3.0)  45 (2.7)  41 (3.2)  4 (1.0)  42 (2.9)  38 (2.6)  8 (2.8)  80 (2.3) 
 Hispanic 270 (6.6)  77 (4.5)  118 (9.2) 37 (8.7)  77 (5.3)  73 (5.0) 28 (9.8) 206 (6.0) 
 Other  73 (1.8)  30 (1.7)  35 (2.7)  9 (2.1)  26 (1.8)  22 (1.5)  9 (3.2)  71 (2.1) 
Women, n4006 82235451432340  3115744120 
Mean (sd) age, years  55.8 (10.5)  55.9 (10.4)  57.8 (10.5) 55.0 (8.9)  57.3 (10.7)  55.4 (10.5) 56.5 (10.0)  56.7 (10.6)<0.001
Race, n (%)         
 White3277 (81.8) 680 (82.7)2944 (83.0) 116 (81.4)1956 (83.6) 271 (87.1)479 (83.4)3446 (83.6) 0.002
 Black 282 (7.0)  62 (7.5) 252 (7.1)  9 (6.2) 178 (7.6)  14 (4.4) 43 (7.4) 281 (6.8) 
 Asian 130 (3.2)  25 (3.0)  83 (2.3)  8 (5.7)  40 (1.7)   9 (3.1)  9 (1.6)  69 (1.7) 
 Hispanic 247 (6.2)  35 (4.3) 209 (5.9)  8 (5.4) 125 (5.3)  14 (4.5) 35 (6.1) 251 (6.1) 
 Other  69 (1.7)  20 (2.4)  58 (1.6)  2 (1.2)  41 (1.7)   3 (1.0)  8 (1.5)  74 (1.8) 

The presence of comorbid conditions is presented by LUTS subgroup (Table 2). In general, the VSPM LUTS subgroup had the highest rates of comorbid conditions for both men and women, whereas the no/minimal LUTS and PM-only subgroups had the lowest rates of comorbid conditions. Among men, there were significantly higher rates of chronic anxiety, depression, enlarged prostate, neurological conditions and sleep apnoea in the VSPM subgroup than in all other subgroups. Among women, there were significantly higher rates of asthma, heart disease, IBS, neurological conditions, recurrent UTIs, sleep apnoea, and uterine prolapse in the VSPM subgroup than in all other LUTS subgroups.

Table 2.  Frequencies of comorbid conditions by LUTS occurring at least sometimes in men and women
n (%) variableNo LUTSV onlyS onlyPM onlyV + SV + PMS + PMV + S + PMP
  • *

    All numbers presented are weighted; subgroups might or might not equal total n because of rounding or weighted values. V, voiding; S, storage; PM, postmicturition. PBlS, painful bladder syndrome.

Men, n409717041280427144914632883433 
Arthritis 650 (15.9) 335 (19.6) 287 (22.4) 82 (19.3) 397 (27.4) 361 (24.7) 65 (22.7)1242 (36.2)<0.001
Asthma 264 (6.4) 137 (8.1)  112 (8.8) 38 (8.9) 123 (8.5) 147 (10.0) 21 (7.2) 435 (12.7)<0.001
Bladder cancer  10 (0.2)   4 (0.2)   6 (0.4)  3 (0.8)  14 (1.0)   7 (0.5)  0 (0.0)  33 (1.0)  0.001
Chronic anxiety  45 (1.1)  29 (1.7)  31 (2.4) 10 (2.4)  35 (2.4)  61 (4.2)  7 (2.5) 241 (7.0)<0.001
Depression 313 (7.6) 194 (11.4) 148 (11.6) 54 (12.7) 199 (13.7) 226 (15.4) 45 (15.5) 906 (26.4)<0.001
Diabetes 405 (9.9) 184 (10.8) 235 (18.4) 44 (10.2) 247 (17.1) 142 (9.7) 46 (16.0) 703 (20.5)<0.001
Heart disease 207 (5.0) 135 (7.9)  107 (8.4) 24 (5.6) 209 (14.5) 156 (10.7) 24 (8.3) 551 (16.0)<0.001
High blood pressure1208 (29.5) 622 (36.5) 566 (44.2)137 (32.2) 694 (47.9) 566 (38.7)129 (44.8)1736 (50.6)<0.001
IC or PBlS   1 (0.0)   1 (0.0)   2 (0.1)  0 (0.0)   3 (0.2)   6 (0.4)  0 (0.0)  22 (0.6)<0.001
IBS 121 (3.0)  70 (4.1)  38 (3.0) 24 (5.7)  63 (4.3) 102 (7.0) 16 (5.7) 288 (8.4)<0.001
Neurological  58 (1.4)  52 (3.0)  33 (2.6)  6 (1.4)  40 (2.8)  44 (3.0)  5 (1.6) 217 (6.3)<0.001
Recurrent UTIs  10 (0.3)  15 (0.9)   6 (0.5)  6 (1.4)  15 (1.0)  16 (1.1)  2 (0.6) 117 (3.4)<0.001
Sleep apnoea or sleep disorder 324 (7.9) 169 (9.9) 177 (13.8) 54 (12.6) 178 (12.3)  211 (14.4) 44 (15.2) 770 (22.4)<0.001
Enlarged prostate/prostatitis 132 (3.2)  115 (6.8)  102 (7.9) 21 (5.0) 247 (17.0) 170 (11.6) 31 (10.8) 1014 (29.5)<0.001
Prostate cancer  47 (1.1)  21 (1.2)  49 (3.9)  4 (0.9)  67 (4.6)  12 (0.8)  8 (3.0) 161 (4.7)<0.001
Women, n4006 82235451432340  3115744120 
Arthritis 954 (23.8) 238 (29.0)1231 (34.7) 30 (21.2) 897 (38.3) 100 (32.2)247 (43.0)1923 (46.7)<0.001
Asthma 371 (9.3)  91 (11.1) 471 (13.3) 14 (9.6) 349 (14.9)  51 (16.5) 82 (14.2) 904 (21.9)<0.001
Bladder cancer   3 (0.1)   0 (0.0)   3 (0.1)  1 (0.5)   0 (0.0)   0 (0.0)  1 (0.1)   11 (0.3)  0.029
Chronic anxiety  95 (2.4)  35 (4.3) 131 (3.7)  5 (3.5) 162 (6.9)  16 (5.2) 30 (5.1) 481 (11.7)<0.001
Depression 479 (12.0) 145 (17.6) 719 (20.3) 27 (18.7) 652 (27.9)  92 (29.7)166 (28.8)1486 (36.1)<0.001
Diabetes 235 (5.9)  73 (8.9) 362 (10.2)  6 (4.5) 300 (12.8)  24 (7.8) 62 (10.8) 611 (14.8)<0.001
Heart disease  112 (2.8)  23 (2.8) 159 (4.5)  3 (2.1) 150 (6.4)  10 (3.2) 36 (6.3) 370 (9.0)<0.001
High blood pressure1008 (25.2) 213 (26.0)1293 (36.5) 26 (18.3) 913 (39.0)  82 (26.4)247 (43.1)1797 (43.6)<0.001
IC or PBlS  43 (1.1)  12 (1.4)  47 (1.3)  1 (0.4)  72 (3.1)  13 (4.0) 18 (3.1) 172 (4.2)<0.001
IBS 251 (6.3)  85 (10.4) 336 (9.5) 13 (9.3) 307 (13.1)  46 (14.7) 71 (12.3) 831 (20.2)<0.001
Neurological  51 (1.3)  24 (3.0)  70 (2.0)  2 (1.6)  81 (3.5)  17 (5.5) 10 (1.8) 277 (6.7)<0.001
Recurrent UTIs  69 (1.7)  37 (4.5) 147 (4.2) 12 (8.1) 173 (7.4)  23 (7.5) 46 (8.0) 588 (14.3)<0.001
Sleep apnoea or sleep disorder 166 (4.1)  54 (6.6) 274 (7.7)  9 (6.5) 285 (12.2)  33 (10.5) 73 (12.7) 756 (18.4)<0.001
Uterine prolapse  36 (0.9)  10 (1.3)  54 (1.5)  3 (1.8)  69 (2.9)  13 (4.1) 10 (1.8) 195 (4.7)<0.001

When examining risk factors among men and women, all LUTS subgroups reported higher rates of childhood nocturnal enuresis than the no/minimal LUTS group (Table 3). Also, subjects in the no/minimal LUTS group were more likely to be non-smokers than those in the VSPM group, and tended to engage in more vigorous and moderate physical activities than the more symptomatic LUTS subgroups. Among women only, the no/minimal LUTS group tended to have a lower BMI than the more symptomatic LUTS subgroups. The no/minimal LUTS group also reported the lowest rate of hysterectomy, postmenopausal status, and HRT use than the other LUTS subgroups.

Table 3.  Risk factors of participants by LUTS occurring at least sometimes *
n (%) or mean (sd) variableNo LUTSV onlyS onlyPM onlyV + SV + PMS + PMV + S + PMP
  • *

    All n’s presented are weighted; subgroups may or may not equal total n because of rounding or weighted values.

  • Significant differences between mean BMI: no LUTS vs S only, P < 0.001; no LUTS vs V + S + PM, P < 0.001; V only vs S only, P < 0.001; V only vs V + S + PM, P < 0.001; S only vs V + PM, P < 0.01; V + PM vs V + S + PM, P < 0.001;

  • Significant differences between mean BMI: no LUTS vs S only, P < 0.001; no LUTS vs V + S, P < 0.001; no LUTS vs S + PM, P < 0.001; no LUTS vs V + S + PM, P < 0.001; V only vs S only, P < 0.001; V only vs V + S, P < 0.001; V only vs S + Pm, P < 0.001; V only vs V + S + PM, P < 0.001; S only vs PM only, P < 0.01; S only vs V + S + PM, P < 0.001; PM only vs V + S, P < 0.001; PM only vs S + PM, P < 0.001; PM only vs V + S + PM, P < 0.001; V + PM vs S + PM, P < 0.01; V + PM vs V + S + PM, P < 0.01.

  • §

    Significant differences between parity: no LUTS vs S only, P < 0.01; no LUTS vs V + S + PM, P < 0.001. V, voiding; S, storage; PM, postmicturition.

Men, n409717041280427144914632883433 
History of bed-wetting 453 (11.1) 277 (16.2) 221 (17.3) 69 (16.2) 255 (17.7) 275 (18.8) 57 (19.8) 751 (21.9)<0.001
Cigarette use         
 Current smoker 848 (20.8) 331 (19.6) 246 (19.3) 60 (14.1) 286 (19.9) 266 (18.4) 46 (16.2)  671 (19.7)<0.001
 Ex-smoker1402 (34.3) 687 (40.6) 457 (36.0)158 (37.1) 658 (45.7) 601 (41.5) 116 (40.6)1700 (49.9) 
 Non-smoker1833 (44.9) 674 (39.8) 568 (44.7)208 (48.8) 496 (34.5) 582 (40.2)124 (43.2)1036 (30.4) 
Primary daily activity         
 Sit during the day 960 (23.4) 453 (26.6)  310 (24.3) 112 (26.3) 351 (24.3) 439 (30.1) 73 (25.5) 1147 (33.4)<0.001
 Stand or walk about quite a lot during the day1955 (47.8) 797 (46.8) 618 (48.4)194 (45.6) 741 (51.2) 684 (46.8)154 (53.5)1550 (45.1) 
 Lift or carry light loads or climb stairs or hills often 851 (20.8) 359 (21.1) 272 (21.3)103 (24.2) 305 (21.1) 264 (18.1) 35 (12.3) 593 (17.3) 
 Do heavy work or carry heavy loads 326 (8.0)  92 (5.4)  77 (6.0) 17 (3.9)  50 (3.5)  73 (5.0) 25 (8.7) 143 (4.2) 
Vigorous activities1811 (44.3) 741 (43.6) 508 (39.7)206 (48.2) 513 (35.4) 597 (40.8)123 (42.7) 1137 (33.1)<0.001
 Mean (sd) times/week   3.5 (1.6)   3.3 (1.6)   3.6 (1.6)  3.4 (1.5)   3.5 (1.7)   3.3 (1.5)  3.3 (1.6)   3.4 (1.6) 0.003
Moderate activities2780 (68.4) 1167 (69.1) 792 (62.7)307 (72.4) 873 (61.3) 981 (67.6)188 (65.9)2038 (59.8)<0.001
 Mean (sd) times/week   3.6 (1.7)   3.7 (1.8)   3.7 (1.7)  3.7 (1.7)   3.8 (1.9)   3.5 (1.7)  3.6 (1.7)   3.7 (1.8) 0.091
Mean (sd) BMI  28.3 (5.7)  28.4 (5.2)  29.5 (6.8) 28.6 (5.4)  29.0 (6.0)  28.5 (4.8) 29.4 (6.4)  29.5 (6.3)<0.001
Women, n4006 82235451432340  3115744120 
History of bed-wetting 354 (8.9)  101 (12.4) 478 (13.7) 12 (8.8) 379 (16.3)  43 (13.9)108 (18.9) 723 (17.7)<0.001
Cigarette use         
 Current smoker 892 (22.3) 184 (22.4) 755 (21.4) 30 (21.3) 646 (27.7)  75 (24.1)131 (22.8)1269 (30.9)<0.001
 Ex-smoker1105 (27.7) 270 (33.0)1151 (32.7) 40 (27.8) 770 (33.0) 103 (33.2)218 (38.0)1358 (33.1) 
 Non-smoker1995 (50.0) 365 (44.6) 1619 (45.9) 72 (50.9) 916 (39.3) 132 (42.7)225 (39.2)1479 (36.0) 
Primary daily activity
 Primarily sit during the day 833 (20.8) 186 (22.7) 838 (23.7) 36 (25.1) 636 (27.2)  98 (31.5)185 (32.3)1255 (30.5)<0.001
 Stand or walk about quite a lot during the day 2171 (54.3) 440 (53.7)1941 (54.8) 80 (56.4)1238 (52.9) 153 (49.1)281 (48.9)2033 (49.4) 
 Lift or carry light loads or climb stairs or hills often 883 (22.1)  171 (20.9) 669 (18.9) 26 (18.4) 416 (17.8)  57 (18.4) 95 (16.6) 725 (17.6) 
 Do heavy work or carry heavy loads  111 (2.8)  22 (2.7)  93 (2.6)  0 (0.0)  49 (2.1)   3 (1.0) 12 (2.2) 100 (2.4) 
Vigorous activities 1514 (37.9) 281 (34.1) 1130 (31.9) 46 (32.5) 660 (28.2) 104 (33.5)155 (27.0) 1185 (28.8)<0.001
 Mean (sd) times/week   3.4 (1.5)   3.3 (1.6)   3.3 (1.6)  3.3 (1.7)   3.3 (1.6)   3.4 (1.3)  3.1 (1.5)   3.3 (1.7) 0.090
Moderate activities2499 (63.1) 512 (62.9)2021 (57.7) 86 (62.7)1291 (56.0) 193 (62.4)307 (54.2)2174 (53.5)<0.001
 Mean (sd) times/week   3.6 (1.7)   3.6 (1.7)   3.6 (1.6)  3.9 (1.8)   3.6 (1.7)   3.5 (1.6)  3.4 (1.7)   3.5 (1.7) 0.016
Mean (sd) BMI  26.8 (5.7)  27.4 (6.2)  29.1 (7.9) 26.0 (5.3)  29.6 (8.5)  28.2 (7.1) 30.5 (7.9)  30.2 (8.4)<0.001
Ever given birth3281 (81.9) 664 (80.8)2973 (83.9)106 (74.0)1979 (84.6) 259 (83.1)494 (86.2)3524 (85.6)<0.001
 Mean (sd) parity§   2.4 (1.2)   2.5 (1.3)   2.6 (1.3)  2.4 (1.1)   2.6 (1.4)   2.4 (1.1)  2.5 (1.2)   2.6 (1.4)<0.001
Hysterectomy 843 (21.1) 222 (27.1) 961 (27.2) 30 (20.8) 665 (28.4)  87 (27.9)179 (31.5)1485 (36.2)<0.001
Perimenopausal 435 (25.3) 120 (34.0) 368 (31.4) 23 (41.6) 275 (34.3)  39 (30.6) 87 (39.7) 612 (42.1)<0.001
Completed menopause2278 (57.0) 466 (56.9)2369 (66.9) 86 (60.3)1535 (65.7) 185 (59.5)351 (61.4)2660 (64.7)<0.001
Taking HRT 273 (6.8)  64 (7.8) 342 (9.7)  8 (5.4) 251 (10.7)  22 (7.1) 60 (10.4) 531 (12.9)<0.001

Results of the logistic regression analyses conducted to identify associations between LUTS subgroups and comorbid conditions and risk factors are presented in Tables 4 (men) and 5 (women). Among men, increasing age was associated with the following symptom constellations: voiding only, storage only, voiding + storage, and VSPM. Several country differences were apparent: men in Sweden were less likely to have storage-only symptoms than those in the USA; and men in the UK were more likely than those in the USA to have voiding-only, voiding + storage, voiding + PM and VSPM symptoms. There were no significant associations for race.

Table 4.  Summary of logistic regressions with LUTS subgroups as the dependent variables and comorbid conditions as the independent variables: Men
VariableV onlyS onlyPM onlyV + SV + PMS + PMV + S + PM
  1. +, positive association; −, negative association; *P < 0.01, **P < 0.001. †There were no significant relationships for Hispanic vs white, other vs white, Asian vs white or Black vs white. V, voiding; S, storage; PM, postmicturition.

Age*** **  **
Country       
 Sweden vs USA *     
 UK vs USA+**  +**+** +**
Ethnicity
Comorbid conditions       
 Arthritis    +* +**
 Asthma    +* +**
 Bladder cancer       
 Chronic anxiety    +** +**
 Depression+*  +**+**+**+**
 Diabetes +**    +**
 Heart disease+*  +**+** +**
 Hypertension +** +**+**+**+**
 IC       
 IBS    +** +**
 Neurological condition+*     +**
 Recurrent UTIs      +**
 Sleep apnoea or sleep disorder    +** +**
 Enlarged prostate or prostatitis+**+** +**+**+**+**
 Prostate cancer +** +**  +**
Risk factors       
 BMI +** +**  +**
 Moderate exercise       
 Vigorous exercise      +*
 Ex-smoker vs non-smoker +*     
Current smoker vs non-smoker  +*    
Childhood nocturnal enuresis+**+**+*+**+**+** 
Table 5.  Summary of logistic regressions with LUTS subgroups as the dependent variables and comorbid conditions as the independent variables: Women
VariableV onlyS onlyPM onlyV + SV + PMS + PMV + S + PM
  1. +, positive association; −, negative association; *P < 0.01, **P < 0.001. †There were no significant relationships for Hispanic vs white, other vs white, Asian vs white or Black vs white. V, voiding; S, storage; PM, postmicturition.

Age
Country       
 Sweden vs USA       
 UK vs USA       
Ethnicity
Comorbid conditions       
 Arthritis +** +** +**+**
 Asthma   +*  +**
 Bladder cancer       
 Chronic anxiety   +*  +**
 Depression +** +**+**+**+**
 Diabetes   +*   
 Heart disease   +**  +**
 Hypertension +** +** +**+**
 IC   +*   
 IBS+*+** +**+**+*+**
 Neurological conditions   +**+** +**
 Recurrent UTIs+**+**+**+** +**+**
 Sleep apnoea or sleep disorder   +**  +**
Risk factors       
 BMI +** +**+*+**+**
 Moderate exercise       
 Vigorous exercise   +*   
 Ex-smoker vs non-smoker       
 Current smoker vs non-smoker   +**  +**
 Hysterectomy     +*+**
 Uterine prolapse   +**+** +**
 Completed menopause vs premenopausal +*     
 Parity +* +*  +*
 Childhood nocturnal enuresis +** +**+*+**+**

The most consistently robust associations for comorbid conditions and risk factors among men with LUTS subgroups were found in the logistic regression with VSPM symptoms as the dependent variable, with the strongest associations (P < 0.001) for arthritis, asthma, chronic anxiety, depression, diabetes, heart disease, hypertension, IBS, neurological conditions, sleep apnoea or sleep disorder, enlarged prostate or prostatitis, and BMI. Recurrent UTIs were also significantly associated with VSPM (P < 0.001), whereas vigorous exercise was protective (P < 0.01).

Some of these comorbid conditions, most notably depression, heart disease, hypertension, and enlarged prostate or prostatitis, were also frequently associated with other LUTS groups. Depression and heart disease were linked to all groups involving voiding symptoms, and there was also an association for depression and storage + PM symptoms. Hypertension was linked to all groups involving storage symptoms and to voiding + PM symptoms. Finally, enlarged prostate/prostatitis was associated with all symptom constellations except PM-only. Other significant associations for comorbid conditions among men (in addition to those noted above for the VSPM group) were found for arthritis, asthma, chronic anxiety, IBS, and sleep apnoea or sleep disorder with voiding + PM symptoms; neurological conditions with voiding-only symptoms; and prostate cancer with storage-only and voiding + storage symptoms.

For men, the most prominent risk factor across all LUTS groups was childhood nocturnal enuresis, which was linked to all symptom combinations except VSPM. Higher BMI was associated with storage-only and voiding + storage symptoms (in addition to VSPM). Associations with LUTS and exercise and smoking were sparse: ex-smokers were more likely to have storage-only symptoms than non-smokers, and current smokers were more likely to have PM-only symptoms. The only link for exercise was noted above for vigorous activity in relation to VSPM symptoms.

Among women, there were no statistically significant associations for age, country, or race in any of the regression models. The greatest number of significant associations for comorbid conditions and risk factors were in the regressions for voiding + storage and VSPM symptoms, respectively. Recurrent UTIs and IBS were among the most robust predictors, with links for the former in relation to all LUTS groups except voiding + PM, and links for the latter in relation to all groups except PM-only. Other comorbid conditions that were also prominent in statistical significance across LUTS groups were arthritis, depression, and hypertension (linked to all groups with storage symptoms) and neurological conditions (linked to voiding + storage, voiding + PM, and VSPM symptoms). Asthma, chronic anxiety, heart disease, and sleep apnoea/sleep disorder were each associated with voiding + storage and VSPM symptoms, and diabetes and IC were each associated with voiding + storage symptoms.

Consistently significant associations with the LUTS subgroups were increased BMI and childhood nocturnal enuresis, with links found for each in relation to all symptom groups except voiding-only and PM-only. Being a current smoker was associated with voiding + storage and VSPM symptoms. Uterine prolapse was associated with voiding + storage, voiding + PM, and VSPM symptoms, parity was related to all storage symptom groups except the storage + PM group; and hysterectomy was associated with storage + PM and VSPM symptoms. Postmenopausal status was related to storage-only symptoms. The only significant association with physical activity and the LUTS groups was that vigorous activity was protective for voiding + storage symptoms.

Not surprisingly, the logistic regression models of the VSPM subgroup were the strongest predictive models among all of the regression analyses, with a C-Index (concordance index) of 0.81 for men and 0.80 for women. The C-Index for the voiding-only and PM-only groups tended to be the lowest, which is probably because of the smaller sample sizes in these groups, as well as fewer LUTS to characterize each subgroup, indicating less symptomatic conditions.

DISCUSSION

This population-representative, cross-sectional dataset provides the opportunity to examine the association between a multitude of risk factors and comorbid conditions assessing storage, voiding, and PM constellations of LUTS based on current ICS definitions [27].

While categorizing patients into the eight LUTS subgroups described here might not represent a routine clinical approach, understanding the relationship between distinct symptom categories, previously identified risk factors, and other disease processes offers new insights into possibly previously unrecognized LUTS associations. Furthermore, this approach provides new information into symptoms not assessed by commonly used tools such as the IPSS. Of critical importance is the potential clinical implication for there being a shared pathophysiology in several disease processes resulting in LUTS. For example, heart disease and hypertension are significantly associated with all voiding symptom groups (except voiding-only group for hypertension), providing further support for the proposed link between BOO/BPH and metabolic syndrome [2–4,7].

The multivariable models within each LUTS group provide enlightening results regarding risk factors and comorbid conditions. Of particular interest in the multivariate models, increasing age (a hallmark risk factor for LUTS) was significantly associated with voiding and storage symptoms among men but was not associated with any LUTS group in women. The increase in voiding and storage symptoms with age in men is probably due to the increase in BOO associated with prostatic enlargement, whereas women experience a multitude of LUTS across all age groups [26]. Another common link that merits further examination is that of LUTS and conditions with inflammatory aetiology. The significance of the association of arthritis, asthma and recurrent UTI with voiding and storage symptoms, in addition to the metabolic syndrome link, adds further support to there being a causal association between inflammation and LUTS. Given the association between inflammation and prostate disease [28,29], the association of two inflammatory conditions (arthritis and asthma) with several of the LUTS subgroups in both men and women suggests the possibility of an inflammatory disease with a possible immune aetiology. In this context, it is notable that the presence of recurrent UTI, which creates a chronic inflammatory response in the bladder, has also previously been noted to be associated with LUTS and OAB [3,30].

The association of depression and chronic anxiety with many of the LUTS subgroups among both men and women highlights again the possibility of neurochemical changes related to serotonin reuptake in the CNS influencing the function of the urinary tract [3]. Certainly, OAB and depression have been linked [31], and serotonin reuptake inhibitors have been used to treat OAB [32], providing further support to the assertion of Litman et al.[3] that altered serotonergic function may be associated with visceral and psychiatric symptoms. This plausible association, coupled with the high rates of reported anxiety and depression associated with multiple LUTS [33], certainly supports the need for an appropriate mental health assessment when LUTS are present.

Surprisingly, moderate exercise was not significantly associated with any LUTS subgroup, and vigorous exercise was only protective in the voiding + storage group in women and the VSPM group in men. As with previous research, increased BMI was significantly associated with all LUTS storage groups, and the impact of smoking on LUTS was under-whelming. Female current smokers were more likely to have voiding + storage and VSPM symptoms, which is understandable given that coughing could increase the likelihood of SUI. However, conversely, amongst men, the only association smoking had with LUTS was in the PM-only group. Male ex-smokers were more likely to have storage-only symptoms, and no other significant associations between smoking and LUTS were noted. The most notable risk factor associated with nearly all LUTS subgroups was childhood nocturnal enuresis. Although the recall of childhood nocturnal enuresis is naturally prone to bias, the findings here support previous research suggesting that childhood nocturnal enuresis is associated with LUTS and OAB in adulthood [30,34]. Future research needs to examine whether effective treatments of childhood nocturnal enuresis will lessen urinary symptoms in adulthood.

Although the cross-sectional design of this study is certainly a limitation that does not allow for a temporal relationship to be established, it provides strong associations that merit future research. Clearly though, the additional limitations of this research are that not all possible risk factors, comorbid conditions, or treatments were assessed (e.g. alcohol intake was not captured) and that the diagnosis of comorbid conditions is based on participant self-report of a clinical diagnosis rather than a diagnosis verified by a clinician. However, despite these potential limitations, this analysis provides the impetus for future longitudinal studies to further examine potential causal relationships between comorbid conditions, risk factors and LUTS. Longitudinal studies would not only aid in physicians’ understanding of the pathophysiology of LUTS and comorbid conditions, but also inform clinical guidelines on patient management, including diagnosing and treatment of comorbid conditions.

The importance of this research to clinicians is to evaluate the impact of modifying risk factors and treating comorbid conditions on LUTS. Prevention and early treatment are the most effective means of tackling important clinical problems such as this, and several important questions remain to be answered. Do behavioural modification programmes focusing on exercise, weight reduction, smoking cessation, etc., help in reducing the long-term burden of LUTS? Can lifestyle modification reduce the symptom burden of current LUTS? Perhaps most importantly, will treatment of LUTS in childhood affect long-term LUTS morbidity? LUTS are not organ-specific; growing evidence supports the association of LUTS with a multitude of comorbid conditions sharing common pathophysiological pathways. Clearly an increased understanding of these common pathophysiological pathways might provide integrated treatment options and reduced LUTS burden for future generations.

ACKNOWLEDGEMENTS

Additional editorial support was provided by Tracy Johnson and Janet E. Matsuura, PhD, at Complete Healthcare Communications, Inc., and was funded by Pfizer Inc. This research was supported by funding from Pfizer Inc.

CONFLICT OF INTEREST

Karin Coyne, Chris Sexton and Elizabeth Bush are employees of United BioSource Corporation, who are scientific consultants to Pfizer. Steven Kaplan is a scientific consultant and lecturer with Pfizer. Christopher Chapple is a scientific consultant and researcher with Allergan, Astellas, Novartis, and Pfizer. Zoe Kopp and Lalitha Aiyer are employees of Pfizer Inc.

Ancillary